Carbazole-bis(oxazolines) as monoanionic, tridentate chelates in lanthanide chemistry: Synthesis and structural studies of thermally robust and kinetically stable dialkyl and dichloride complexes.

Article abstract of DOI:10.1021/om200588y

The carbazole-bis(oxazoline) ligand 1,8-bis(4′,4′- dimethyloxazolin-2′-yl)-3,6-di-tert-butylcarbazole (H-Czx, 14) was prepared and reacted with Ln(CH₂SiMe₃)₃(THF) ₂ to afford the five-coordinate dialkyl complexes (Czx)Ln(CH ₂SiMe₃)₂ (Ln = Y, 15; Er, 16; Yb, 17) in high yield. Deprotonation of 14 with NaN(SiMe₃)₂, followed by reaction with LnCl₃(THF)_x, affords the six-coordinate dichloride complexes cis,mer-(Czx)Ln(Cl)₂(THF) (Ln = Y, 18; Er, 19; Yb, 20). Dialkyl complexes 15-17 were also prepared by reaction of the dichloride complexes 18-20, respectively, with 2 equiv of LiCH ₂SiMe₃. Complexes 16-20 were characterized by X-ray crystallography. The dialkyl complexes 16 and 17 display a five-coordinate trigonal-bipyramidal geometry with the Czx ligand spanning both axial and one equatorial site with the CH₂SiMe₃ rotated such that the SiMe₃ groups sit over the carbazole ring system. The solution structure for 15-20 was found to be consistent with that in the solid state by NMR spectroscopy. However, paramagnetic dichloride 20 was shown by variable-temperature ¹H NMR studies to undergo rapid THF exchange by a dissociative process in toluene-d₈ solution (ΔH = 98 ± 10 kJ mol^-1, ΔS = +185 ± 25 J mol^-1 K ^-1). Treatment of dialkyl 18 with [Ph₃C ⁺[B(C₆F₅)₄^- cleanly produces the alkyl cation [(Czx)Y(CH₂SiMe₃) ⁺[B(C₆F₅)₄^- (21) by ¹H NMR spectroscopy, but this species does not insert 2,3-dimethylbutadiene.

Full text of DOI:10.1021/om200588y

ARTICLE
pubs.acs.org/Organometallics
Carbazole-bis(oxazolines) as Monoanionic, Tridentate Chelates in
Lanthanide Chemistry: Synthesis and Structural Studies of Thermally
Robust and Kinetically Stable Dialkyl and Dichloride Complexes.
Jin Zou,† David J. Berg,*,† David Stuart,† Robert McDonald,‡ and Brendan Twamley§
†Department of Chemistry, University of Victoria, P.O. Box 3065 Victoria, British Columbia V8W 3V6, Canada
‡X-ray Crystallography Laboratory, Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2, Canada
§Department of Chemistry, University of Idaho, P.O. Box 442343, Moscow, Idaho 83844-2343, United States
S Supporting Information
b
ABSTRACT: The carbazole-bis(oxazoline) ligand 1,8-bis(40,40-dimethyloxazolin-20-yl)-
3,6-di-tert-butylcarbazole (H-Czx, 14) was prepared and reacted with Ln(CH2SiMe3)3-
(THF)2 to afford the five-coordinate dialkyl complexes (Czx)Ln(CH2SiMe3)2 (Ln = Y, 15;
Er, 16; Yb, 17) in high yield. Deprotonation of 14 with NaN(SiMe3)2, followed by reaction
with LnCl3(THF)x, affords the six-coordinate dichloride complexes cis,mer-(Czx)Ln(Cl)2-
(THF) (Ln = Y, 18; Er, 19; Yb, 20). Dialkyl complexes 15ꢀ17 were also prepared by
reaction of the dichloride complexes 18ꢀ20, respectively, with 2 equiv of LiCH2SiMe3.
Complexes 16ꢀ20 were characterized by X-ray crystallography. The dialkyl complexes 16
and 17 display a five-coordinate trigonal-bipyramidal geometry with the Czx ligand
spanning both axial and one equatorial site with the CH2SiMe3 rotated such that the
SiMe3 groups sit over the carbazole ring system. The solution structure for 15ꢀ20 was
found to be consistent with that in the solid state by NMR spectroscopy. However,
paramagnetic dichloride 20 was shown by variable-temperature 1H NMR studies to undergo
rapid THF exchange by a dissociative process in toluene-d8 solution (ΔH‡ = 98 ( 10 kJ molꢀ1, ΔS‡ = +185 ( 25 J molꢀ1 Kꢀ1). Treatment
1
of dialkyl 18 with [Ph3C]+[B(C6F5)4]ꢀ cleanly produces the alkyl cation [(Czx)Y(CH2SiMe3)]+[B(C6F5)4]ꢀ (21) by H NMR
spectroscopy, but this species does not insert 2,3-dimethylbutadiene.
’ INTRODUCTION
in lanthanide chemistry. Despite its obvious attractions, the oxazo-
line ring system is neutral and therefore likely to suffer from high
lability when coordinated to a lanthanide metal center. However,
attaching two oxazolines to a carbazole-based core gives a triden-
tate, monoanionic carbazole-bis(oxazoline) (Czx) chelate that
should be well suited to organolanthanide chemistry. In addition
to being strongly anchored through the carbazole amido nitrogen,
the rigid, planar nature of the chelate array should enforce a well-
defined mer coordination geometry at the metal ion.
Chiral carbazole-bis(oxazoline) ligands (8, Chart 3) were orig-
inally developed by Nakada and Inoue for use in the asymmetric
NozakiꢀHiyama allylation, allenylation, and propargylation of
aldehydes using Cr2+ complexes of these ligands formed in situ as
catalysts.5 More recently, Connell has extended this chemistry to
the synthesis of butadien-2-ylcarbinols, also using Cr2+ Czx
complexes as catalysts.6 It should also be noted that transition
metal complexes of related carbazole-bis(imino)7 and carbazole-
bis(pyridine)8 ligands have been previously reported (Chart 3, 9
and 10, respectively). To our knowledge, Czx ligands have not
been previously used in lanthanide chemistry, but a Y3+ dialkyl
complex of a carbazole-bis(phosphine), 11, and a Lu3+ dialkyl
Oxazolines ligands have been extensively used in transition metal
coordination chemistry for many years.1 Easy access to these ligands
by condensation of amino alcohols with carboxylic acid derivatives
or nitriles allows straightforward modification of the ligand archi-
tecture including the incorporation of chirality when chiral amino
alcohols, derived from readily available amino acids, are used. Enantio-
selective synthesis based on pybox ligands (Chart 1, 1) is one
particularly well-established example of oxazoline-derived ligand
systems in current use.2 In contrast to their extensive use in
transition metal chemistry, relatively few lanthanide complexes
containing coordinated oxazoline rings have been reported, and
most of these are neutral ligand systems (e.g., 1ꢀ4, Chart 1) that
are not especially suited to these metals. Despite this, lanthanide
and group 3 complexes of 1ꢀ4 have proven successful as asym-
metric catalysts in many reactions.3 A few lanthanide complexes
of anionic ligands containing oxazolines as part of the architec-
ture (5ꢀ7, Chart 2) have been reported,4 but only ligand 74a has
been applied to lanthanide organometallic chemistry.
Our interest in these ligands stems from the fact that oxazoline
rings coordinate almost exclusively as hard σ-donors through the
imino nitrogen, but the CdN linkage remains inert to attack by
strong nucleophiles such as carbanions. The fact that the steric
bulk of the oxazoline is readily tunable is an important advantage
Received: July 4, 2011
Published: August 24, 2011
r
2011 American Chemical Society
4958
dx.doi.org/10.1021/om200588y Organometallics 2011, 30, 4958–4967
|

Organometallics
ARTICLE
Chart 1
Chart 3
Chart 2
where specified, were confirmed by 2D 1Hꢀ C COSY (HMQC, HMBC)
experiments on diamagnetic compounds. Mass spectra were recorded
on a Kratos Concept H spectrometer using electron impact (70 eV)
ionization. Elemental analyses were conducted by Canadian Microana-
lytical, Delta, BC, or the Department of Chemistry, University of British
Columbia. Although repeated attempts were tried, satisfactory analyses
of 17, 18, and 20 were not obtained. Melting points were carried out in
sealed capillary tubes using a B€uchi melting point apparatus and are not
corrected. Infrared spectra were obtained using KBr discs on a Perkin-
Elmer Spectrum One FTIR spectrometer.
13
complex of the closely related carbazole-bis(phosphinimine), 12,
have very recently been reported.9 In this contribution, we show
that the carbazole-bis(oxazoline) ligand allows easy access to
five-coordinate lanthanide dialkyl and six-coordinate lanthanide
dichloride complexes that exhibit remarkable stability to thermal
degradation and ligand redistribution.
1,8-Dicyano-3,6-di-tert-butylcarbazole, 13.
’ EXPERIMENTAL SECTION
General Procedures. All reactions were done under an argon or
nitrogen atmosphere using glovebox (Braun MB150-GII) or Schlenk
methods. Tetrahydrofuran (THF) was dried by distillation from sodium
benzophenone ketyl under argon immediately before use; other solvents
were used directly from an MBraun solvent purification system. Toluene
and hexane were stored in the glovebox for prolonged periods over
activated 4 Å sieves. Carbazole, CuCN, and N-methyl-2-pyrrolidone
were purchased from Aldrich and used as received. Anhydrous LnCl3
salts (Ln = Y, Er, Yb) were purchased from Aldrich and used as received
or converted to their THF solvates, LnCl3(THF)x, by extraction with
anhydrous THF. 3,6-Di-tert-butylcarbazole,7 1,8-dibromo-3,6-di-tert-
butylcarbazole,7 Ln(CH2SiMe3)3(THF)2 (Y, Er, Yb),10aꢀc and NaN-
(SiMe3)210d were prepared according to literature procedures.
1,8-Dibromo-3,6-di-tert-butylcarbazole6 (10.02 g, 22.92 mmol)
was dissolved in 180 mL of N-methyl-2-pyrrolidone in a foil-
covered flask. CuCN (5.33 g, 60 mmol) was added, and the
solution was refluxed at 180 °C overnight. After cooling, the
solution was poured into a beaker containing 400 mL of ammo-
nium hydroxide and 200 mL of ethyl acetate, and the resulting
suspension was stirred for 30 min. The precipitate was filtered
through Celite, and the filtrate was extracted with ethyl acetate
(5 ꢁ 300 mL). The combined organic extracts were washed with
concentrated brine (2 ꢁ 300 mL) and dried over anhydrous
MgSO4. Filtration, removal of the solvent from the filtrate, and
recrystallization of the crude product from ethyl acetate afforded
NMR spectra were recorded on a Bruker AMX-500 (1H, 500 MHz;
13C, 125 MHz) or Bruker AMX-360 (1H, 360 MHz) spectrometer.
Benzene-d6 and toluene-d8 were dried over activated 4 Å sieves, and
THF-d8 was distilled from sodium benzophenone and stored in the
glovebox. Air-sensitive samples for NMR analysis were sealed using a
5 mm tube fitted with a Teflon valve (Brunfeldt). NMR assignments,
1
pure 13. Yield: 6.04 g (80%). Mp: 287ꢀ290 °C. H NMR
(500 MHz, CDCl3, 298 K): δ 9.92 (s, 1H, NH), 8.29 (d, J = 1.8
Hz, 2H, 4,5-arylH), 7.82 (d, J = 1.8 Hz, 2H, 2,7-arylH), 1.43 (s,
18H, C(CH3)3). 13C{1H} NMR (125 MHz, CDCl3): δ 144.02
(3,6-arylC), 139.36 (9a,9d-arylC), 128.70 (2,7-arylC), 123.99
4959
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
(9b,9c-arylC), 122.33 (4,5-arylC), 117.24 (CN), 94.33 (1,8-arylC),
35.12 (C(CH3)3), 31.92 (C(CH3)3). EI-MS: m/z 329 (M+).
1,8-Bis(40,40-dimethyloxazolin-20-yl)-3,6-di-tert-butyl-
carbazole (H-Czx), 14.
Yield: 0.77 g (92%). Mp: 206 °C (dec). 1H NMR (360 MHz, C6D6, 298 K):
δ81.2 (s, 12H, ν1/2 =567Hz, 40-(CH3)2), 76.5 (s, 4H, ν1/2 =158Hz, CH2),
42.9 (s, 2H, ν1/2 = 75 Hz, 2,7-arylH), 7.26 (s, 2H, ν1/2 = 20 Hz, 4,5-arylH),
6.45 (s, 18H, ν1/2 = 17 Hz, C(CH3)3), ꢀ62.2 (s, 18H, ν1/2 = 154 Hz,
Si(CH3)3), ꢀ185.0 (s, 4H, ν1/2 = 2360 Hz, CH2Si(CH3)3). IR (KBr):
ν 2955, 1635, 1576, 1476, 1363, 1176, 852, 730, 668 cmꢀ1. Anal. Calcd for
C38H60N3O2Si2Er: C 56.05, H 7.43, N 5.16. Found: C 56.52, H 7.38, N 5.29.
(Czx)Yb(CH2SiMe3)2, 17. This complex was prepared on the same
scale and using the same procedure reported for 15 above. Orange-red
cubes of 17 were isolated by recrystallization from a tolueneꢀhexane
mixture. Yield: 0.75 g (91%). Mp: 250 °C (dec). 1H NMR (360 MHz,
C6D6, 298 K): δ 69.2 (s, 16H, ν1/2 = 59 Hz, 40-(CH3)2 and CH2), 33.5
(s, 2H, ν1/2 = 18 Hz, 2,7-arylH), 2.61 (s, 18H, ν1/2 = 6 Hz, C(CH3)3),
ꢀ1.60 (s, 2H, ν1/2 = 9 Hz, 4,5-arylH), ꢀ39.9 (s, 18H, ν1/2 = 26 Hz,
Si(CH3)3), ꢀ174.8 (s, 4H, ν1/2 = 54 Hz, CH2Si(CH3)3). IR (KBr):
ν 2954, 1634, 1575, 1476, 1363, 1176, 852, 730, 668 cmꢀ1. Anal. Calcd
for C38H60N3O2Si2Yb: C 55.65, H 7.37, N 5.12. Found: C 52.60, H 6.28,
N 5.60. The crystals changed color from orange-red to yellow when
exposed to high vacuum.
(Czx)Y(Cl)2(THF), 18. YCl3(THF)x (0.45 g, 1.0 mmol) was par-
tially dissolved in 10 mL of THF in a 50 mL Erlenmeyer flask in the
glovebox. A solution of NaCzx (0.50 g, 1.0 mmol) in 5 mL of THF was
prepared in situ by deprotonation of H-Czx, 14 (0.48 g, 1.0 mmol), with
1 equiv of Na[N(SiMe3)2] (0.18 g, 1.0 mmol). The NaCzx solution was
added to the YCl3(THF)x suspension dropwise by Pasteur pipet,
resulting in an immediate color change to neon yellow-green. After 1
h at room temperature, the reaction mixture was filtered through Celite
to remove NaCl, and the solvent was removed from the filtrate under
reduced pressure using a rotary evaporator. Crude 18 was recrystallized
from benzene to afford yellow-green crystals of pure 18.Yield: 0.63 g
(90%). Mp: 272 °C (dec). 1H NMR (500 MHz, C6D6, 298 K): δ 8.54
(d, J = 2.2 Hz, 2H, 4,5-arylH), 8.48 (d, J = 2.1 Hz, 2H, 2,7-arylH), 3.53 (s,
4H, CH2), 3.38 (t, 4H, R-CH2 THF), 1.76 (s, 12H, 40-(CH3)2), 1.45 (s,
18H, C(CH3)3), 0.72 (t, 4H, β-CH2 THF). 13C{1H} NMR (125 MHz,
C6D6): δ 169.04 (CdN), 147.58 (9a,9d-arylC), 140.57 (3,6-arylC), 127.57
(9b,9c-arylC), 126.77 (2,7-arylCH), 123.31 (4,5-arylCH), 110.23 (1,
8-arylC), 79.85 (CH2), 69.34 (40-C(CH3)2)), 35.05 (C(CH3)3), 32.43
(C(CH3)3), 28.45 (40-C(CH3)2). Anal. Calcd for C46H58Cl2N3O3Y: C
64.25, H 6.75, N 4.89. Found: C 64.45, H 6.87, N 4.98.
1,8-Dicyano-3,6-di-tert-butylcarbazole, 13 (2.00 g, 6.07 mmol),
was dissolved in 100 mL of chlorobenzene along with 2-methyl-
2-aminopropanol (1.62 g, 18.2 mmol) and anhydrous ZnCl2
(2.08 g, 15.3 mmol). The reaction mixture was refluxed for 72 h
and then cooled to room temperature. The reaction mixture was
poured into a solution of ethylenediamine (10 mL) and water
(100 mL) and stirred for 1 h. The resulting biphasic solution was
extracted with CH2Cl2 (3 ꢁ 100 mL), and the combined organic
layers were washed with brine (2 ꢁ 100 mL) and dried over
anhydrous MgSO4. After the filtration, the organic layer evapo-
rated to dryness and the residue was purified by column chroma-
tography on silica using CH2Cl2 as eluant. Pure H-Czx, 13, was
isolated as a white powder (1.15 g, 40%). Mp: 278ꢀ280 °C. 1H
NMR (500 MHz, CDCl3, 298 K): δ 11.87 (s, 1H, NH), 8.19 (d,
J = 1.8 Hz, 2H, 4,5-arylH), 7.89 (d, J = 1.8 Hz, 2H, 2,7-arylH), 4.16
(s, 4H, CH2), 1.47 (s, 12H, 40-(CH3)2), 1.45 (s, 18H, C(CH3)3).
13C{1H} NMR (125 MHz, CDCl3): δ 161.27 (CdN), 141.34
(3,6-arylC), 137.61 (9a,9d-arylC), 123.18 (9b,9c-arylC), 122.65
(2,7-arylCH), 119.56 (4,5-arylCH), 109.66 (1,8-arylC),78.32(CH2),
67.55 (40-C(CH3)2)), 34.68 (C(CH3)3), 32.26 (C(CH3)3), 28.84
(40-C(CH3)2). IR (KBr): ν 3336 (br s), 2961, 1648, 1490, 1363,
1286, 1152, 727, 668 cmꢀ1. EI-MS: m/z 473 (M+). Anal. Calcd
for C30H39N3O2: C 76.08, H 8.30, N 8.87. Found: C 75.66,
H 8.28, N 8.77.
(Czx)Y(CH2SiMe3)2, 15. In the glovebox, Y(CH2SiMe3)3(THF)2
(0.49 g, 1.0 mmol) was dissolved in 10 mL of toluene in a 50 mL
Erlenmeyer flask. A solution of H-Czx (0.47 g, 1.0 mmol) in 5 mL of
toluene was added by pipet, resulting in an immediate color change to
neon yellow. After 2 h at room temperature, the solvent was removed
under reduced pressure using a rotary evaporator, and the solid residue
obtained was recrystallized from a mixture of toluene and hexane to
afford 15 as pale yellow crystals (0.66 g, 90%). Mp: 196 °C (dec).
1H NMR (500 MHz, C6D6, 298 K): δ 8.65 (d, J = 2.2 Hz, 2H, 4,5-arylH),
8.51 (d, J = 1.8 Hz, 2H, 2,7-arylH), 3.72 (s, 4H, CH2), 1.56 (s, 12H,
40-(CH3)2), 1.39 (s, 18H, C(CH3)3), 0.04 (s, 18H, Si(CH3)3), 0.00 (s,
JY,H = 2.6 Hz, 4H, CH2Si(CH3)3). 13C{1H} NMR (125 MHz, CDCl3):
δ 170.28 (CdN), 148.05 (9a,9d-arylC), 140.78 (3,6-arylC), 127.84
(9b,9c-arylC), 126.46 (2,7-arylCH), 123.31 (4,5-arylCH), 109.76 (1,
8-arylC), 79.09 (CH2), 69.14 (40-C(CH3)2)), 37.46 (JY,C = 38.5 Hz,
CH2Si(CH3)3)), 35.01 (C(CH3)3), 32.30 (C(CH3)3), 27.89 (40-
C(CH3)2), 4.76 (Si(CH3)3). IR (KBr): ν 2956, 1634, 1596, 1472, 1362,
1176, 861, 776, 668 cmꢀ1. This compound decomposed after several
hours at room temperature in the glovebox so an elemental analysis was
not obtained.
(Czx)Er(Cl)2(THF), 19. This complex was prepared on the same
scale and using the same procedure reported for 18 above. Yellow
crystals of 19 were isolated by recrystallization from a tolueneꢀhexane
mixture. Yield: 0.74 g (95%). Mp: 270 °C (dec). 1H NMR (360 MHz,
C6D6, 298 K): δ 41.9 (s, 16H, ν1/2 = 2730 Hz, 40-(CH3)2 and CH2), 20.2
(s, 2H, ν1/2 = 1080 Hz, 2,7-arylH), 0.80 (s, 18H, ν1/2 = 60 Hz,
C(CH3)3), ꢀ2.44 (s, 2H, ν1/2 = 34 Hz, 4,5-arylH), ꢀ14.1 (s, 4H, ν1/
2 = 114 Hz, β-THF), ꢀ30.6 (s, 4H, ν1/2 = 578 Hz, R-THF). Anal. Calcd
for C46H58Cl2N3O3Er: C, 58.96, H 6.20, N 4.49. Found: C 57.80, H
5.86, N 4.66.
(Czx)Yb(Cl)2(THF), 20. This complex was prepared on the same
scale and using the same procedure reported for 18 above. Yellow-green
crystals of 20 were isolated by recrystallization from a tolueneꢀhexane
mixture. Yield: 0.73 g (93%). Mp: 270 °C (dec). 1H NMR (360 MHz,
C6D6, 298 K): δ 49.3 (s, 4H, ν1/2 = 670 Hz, CH2), 43.8 (s, 12H, ν1/2
=
680 Hz, 40-(CH3)2), 20.0 (s, 2H, ν1/2 = 102 Hz, 2,7-arylH), ꢀ0.97 (s,
18H, ν1/2 = 6 Hz, C(CH3)3), ꢀ8.63 (s, 2H, ν1/2 = 26 Hz, 4,5-arylH),
ꢀ9.6 (s, 4H, ν1/2 = 42 Hz, β-CH2 THF), ꢀ26.0 (s, 4H, ν1/2 = 266 Hz,
R-CH2 THF). Anal. Calcd for C46H58Cl2N3O3Yb: C 58.46, H 6.14, N
4.45. Found: C 58.23, H 6.39, N 4.41.
[(Czx)Y(CH2SiMe3)]+[B(C6F5)4]ꢀ, 21. Treatment of a solution of
15 (0.012 g, 0.016 mmol) dissolved in toluene-d8 (0.6 mL) with 1 equiv
of Ph3C+ [B(C6F5)4]ꢀ (0.015 g, 0.016 mmol) resulted in an orange-brown
solution. This complex decomposed during attempted isolation. The prod-
uct was characterized in situ by NMR spectroscopy in toluene-d8. 1H NMR
(Czx)Er(CH2SiMe3)2, 16. This complex was prepared on the same
scale and using the same procedure reported for 15 above. Yellow cubes
of 16 were isolated by recrystallization from a tolueneꢀhexane mixture.
4960
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
(360 MHz, 298 K): δ 8.53 (d, J = 1.9 Hz, 2H, 4,5-arylH), 8.48 (d, J =
1.9 Hz, 2H, 2,7-arylH), 7.25ꢀ7.28* (m, 6H, o-arylH Ph3CCH2Si-
(CH3)3), 7.00ꢀ7.11* (m, 9H, m,p-arylH Ph3CCH2Si(CH3)3), 3.65 (s,
4H, CH2), 2.04* (s, 2H, Ph3CCH2Si(CH3)3), 1.42 (s, 18H, C(CH3)3),
1.10 (s, 12H, 40-(CH3)2), 0.07 (d, 2JY,H = 2.8 Hz, 2H, YCH2Si(CH3)3),
ꢀ0.23 (s, 9H, Si(CH3)3), ꢀ0.25 (s, 9H, Si(CH3)3). 13C{1H} NMR
(90.6 MHz): δ 172.32 (CdN), 149.57 (9a,9d-arylC), 142.96 (3,
THF-free bis(alkyls) (Czx)Ln(CH2SiMe3)2 (Ln = Y (15),
Er (16), Yb (17)) in good yields by direct protonolysis
(Scheme 1). While this reaction works efficiently for the smaller
lanthanides, it does not allow access to the larger members of the
series because the tris(alkyl) precursors are unstable for metals
larger than samarium. However, the salt metathesis reaction
between 1 equiv of Na+Czxꢀ and LnCl3(THF)x cleanly affords
the six-coordinate THF solvates (Czx)LnCl2(THF) (Ln =
Y (18), Er (19), Yb (20)), and these can in turn be converted
to the corresponding bis(alkyl) complexes 15ꢀ17 in good yield
(Scheme 1).
6-arylC), 148.8 (dm, 1JCF = 238 Hz, o-arylC C6F5), 138.5 (dm, 1JCF
=
1
246 Hz, p-arylC C6F5), 137.0 (dm, JCF = 247 Hz, m-arylC C6F5),
129.01* (o-arylCH, Ph3C), 127.52* (m-arylCH, Ph3C), 127.39 (9b,9c-
arylC), 126.50 (2,7-arylCH), 125.86* (p-arylCH, Ph3C), 124.58 (4,
5-arylCH), 108.46 (1,8-arylC), 78.86 (CH2), 67.67 (40-C(CH3)2)),
47.80 (d, 1JY,C = 48.8 Hz, YCH2Si(CH3)3)), 35.04 (C(CH3)3), 32.11*
(Ph3CCH2), 31.49 (C(CH3)3), 27.13 (40-C(CH3)2), 3.01* (Ph3CCH2Si-
(CH3)3), 0.47 (YCH2Si(CH3)3). 19F NMR (282.4 MHz): δ ꢀ131.6 (br s,
Scheme 1
3
o-arylF), ꢀ160.4 (t, JFF = 21 Hz, p-arylF), ꢀ164.8 (br s, m-arylF). *
indicates resonances due to the reaction byproduct, Ph3CCH2Si(CH3)3.
X-ray Crystallographic Studies. Crystals for analysis were at-
tached to a glass fiber, and data were collected at low temperature (90 K
for 16; 193 K for 17ꢀ20) using a Bruker/Siemens SMART APEX (16) or
SMART 1000 (17ꢀ20) instrument (Mo KR radiation, λ = 0.71073 Å).
Data were measured using omega scans 0.3° per frame for 20 s, and a full
sphere of data was collected. Data reduction and correction for Lorentz
polarization and decay were performed using SAINTPlus software.11
Absorption corrections were applied using SADABS.12 The structures
were solved by direct methods and refined using least-squares on F2 as
implemented in the SHELXTL program package.13 All non-hydrogen
atoms were refined anisotropically. No significant decomposition was
observed during data collection. A summary of the data collection and
refinement parameters is given in Table 1. Further details are provided in
the Supporting Information.
’ RESULTS AND DISCUSSION
Synthesis. Reaction of 1 equiv of H-Czx, 14, with the lanthanide
tris(alkyls) Ln(CH2SiMe3)3(THF)2 afforded the corresponding
Table 1. Summary of Crystallographic Data
16a
17b
18b
19b
20b
formula
fw
C38H60ErN302Si2
814.33
C38H60YbN302Si2
820.11
C46H58Cl2N3O3Y
860.76
C46H58Cl2ErN3O3
939.11
C46H58Cl2N3O3Yb
944.89
cryst syst
space group
a (Å)
orthorhombic
Pna21 (No. 33)
10.983(4)
18.0056(6)
20.1636(7)
90
triclinic
orthorhombic
Pbca (No. 61)
13.5915(12)
17.7957(15)
37.215(3)
90
orthorhombic
Pbca (No. 61)
13.5749(7)
17.7904(9)
37.247(2)
90
orthorhombic
Pbca (No. 61)
13.5448(7)
17.7920(9)
37.249(2)
90
P1 (No. 2)
14.6863(9)
15.5101(10)
18.6074(12)
80.2013(8)
89.0033(9)
89.1795(9)
4175.8(5)
4
b (Å)
c (Å)
R (deg)
β (deg)
γ (deg)
V (Å3)
Z
90
90
90
90
90
90
90
90
3987.5(2)
4
9001.2(13)
8
8995.4(8)
8
8976.7(8)
8
F
calc (g cmꢀ3
μ (mmꢀ1
)
1.356
1.305
1.270
1.387
1.398
)
2.199
2.329
1.456
2.026
2.244
2θmax (deg)
measd reflns
unique reflns
55
55
51
55
55
52 288
36 696
61 378
75 551
75 251
9149
18 922
8438
10 321
10 282
R,c Rw
0.024, 0.049
0.033, 0.090
0.050, 0.118
0.033, 0.080
0.034, 0.084
d
GOF
1.03
1.05
1.03
1.15
1.15
a Collected using a Bruker/Siemens Smart APEX CCD system (graphite-monochromated Mo KR γ = 0.71073 Å) at 90 K. b Collected using a Bruker
Smart 1000 CCD system (graphite-monochromated Mo KR γ = 0.71073 Å) at 193 K. c R = ∑(|Fo| ꢀ |Fc|)/∑|Fo|. d Rw = [∑w(|Fo| ꢀ |Fc|2)/∑w(|Fo|2)1/2
.
4961
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
Table 2. Selected Bond Lengths and Angles for
(Czx)Ln(CH2SiMe3)2 (Ln = Er, 16; Yb, 17)a
17 (Yb)
16 (Er)
molecule A
molecule B
LnꢀNOx1
2.428(2)
2.423(2)
2.312(2)
2.398(3)
2.404(3)
160.55(8)
80.64(8)
95.93(9)
94.60(9)
80.23(8)
96.11(9)
92.35(9)
117.9(1)
121.49(9)
120.6(1)
113.5(2)
115.9(1)
2.409(3)
2.404(3)
LnꢀNOx2
2.416(3)
2.410(3)
LnꢀNCbz
2.308(2)
2.278(2)
LnꢀC1
2.343(3)
2.353(4)
LnꢀC2
2.355(3)
2.364(4)
NOx1ꢀLnꢀNOx2
NOx1ꢀLnꢀNCbz
NOx1ꢀLnꢀC1
NOx1ꢀLnꢀC2
NOx2ꢀLnꢀNCbz
NOx2ꢀLnꢀC1
NOx2ꢀLnꢀC2
NCbzꢀLnꢀC1
NCbzꢀLnꢀC2
C1ꢀLnꢀC2
LnꢀC1ꢀSi1
LnꢀC2ꢀSi2
160.91(9)
80.73(9)
160.93(10)
80.81(9)
95.49(10)
95.40(12)
80.29(8)
94.47(14)
94.80(14)
80.13(9)
Figure 1. ORTEP314 plot (40% probability ellipsoids) of (Czx)Er-
(CH2SiMe3)2 (16).
93.64(10)
97.34(11)
124.78(10)
125.87(11)
109.35(12)
120.11(16)
118.72(17)
95.29(13)
95.08(14)
118.54(12)
123.70(13)
117.77(15)
117.36(19)
118.35(19)
a NOx1, NOx2, NCbz, C1, and C2 refer to the following atom labels
for 16: N(1), N(32), N(34), C(36), and C(40); for 17 these labels
refer to atoms labeled N(10), N(20), N(1), C(1), and C(2),
respectively. Si1 and Si2 refer to atoms labeled Si(1) and Si(2) in
both structures.
Figure 2. ORTEP314 plot (25% probability ellipsoids) of (Czx)Yb-
(Cl)2(THF) (20).
Table 3. Selected Bond Lengths and Angles for
(Czx)LnCl2(THF) 2C6H6 (Ln = Y, 18; Er, 19; Yb, 20)a
3
18 (Y)
19 (Er)
20 (Yb)
Structural Studies. The solid-state structures of dialkyl com-
plexes 16 and 17, as well as the dichloride complexes 18, 19, and
20, were determined by X-ray crystallography. Although erbium
dialkyl 16 and ytterbium dialkyl 17 are not isostructural, the
coordination geometry is essentially identical in both cases. All
three dichloride complexes are isostructural. Table 1 summarizes
the crystallographic data for all five structures, while representa-
tive ORTEP plots for the dialkyl and dichloride complexes are
shown in Figures 1 and 2, respectively. Selected bond lengths and
angles for the dialkyl and dichloride complexes are collected in
Tables 2 and 3, respectively.
LnꢀNOx1
2.418(3)
2.431(3)
2.331(2)
2.6031(10)
2.5592(10)
2.391(2)
159.59(10)
80.14(9)
99.32(7)
92.20(7)
82.78(9)
80.03(10)
98.75(7)
96.01(8)
89.87(9)
166.13(7)
100.10(7)
82.52(9)
93.76(4)
83.67(6)
173.89(7)
2.404(3)
2.418(3)
2.321(3)
2.5911(8)
2.5442(9)
2.390(2)
160.25(9)
80.60(9)
99.03(6)
92.11(6)
82.86(8)
80.21(9)
98.47(7)
95.78(7)
90.05(9)
166.04(7)
100.09(7)
82.49(8)
93.87(3)
83.62(6)
173.93(6)
2.391(3)
2.405(3)
2.289(3)
2.5635(9)
2.5233(10)
2.363(2)
161.53(10)
81.18(10)
98.55(7)
91.91(7)
83.23(9)
80.89(10)
97.91(7)
95.40(8)
90.22(10)
166.92(7)
99.57(7)
82.98(9)
93.50(3)
84.00(7)
174.13(7)
LnꢀNOx2
LnꢀNCbz
LnꢀCl1
LnꢀCl2
LnꢀOTHF
NOx1ꢀLnꢀNOx2
NOx1ꢀLnꢀNCbz
NOx1ꢀLnꢀCl1
NOx1ꢀLnꢀCl2
NOx1ꢀLnꢀOTHF
NOx2ꢀLnꢀNCbz
NOx2ꢀLnꢀCl1
NOx2ꢀLnꢀCl2
NOx2ꢀLnꢀOTHF
NCbzꢀLnꢀCl1
NCbzꢀLnꢀCl2
NCbzꢀLnꢀOTHF
Cl1ꢀLnꢀCl2
Cl1ꢀLnꢀOTHF
Cl2ꢀLnꢀOTHF
Dialkyl complexes 16 and 17 adopt a monomeric, five-coordi-
nate geometry at the lanthanide center. The geometry at the metal
is best described as a distorted trigonal bipyramid with the anionic
carbazole N and the alkyl C atoms in the equatorial plane (NCbz
ꢀ
LnꢀC: 117.9(1)°, 121.49(9)° for 16 and 118.54(12)ꢀ125.87-
(11)° for 17; CꢀLnꢀC: 120.6(1)° for 16 and 109.35(12)°,
117.8(2)° for the two independent molecules in 17) and the
oxazoline N donors in the axial positions (NOxꢀLnꢀNOx: 160.55(8)°
for 16; 160.91(9), 160.93(10)° for 17). Somewhat surprisingly,
both alkyl groups in each complex are rotated away from the
oxazoline methyl substituents so that they sit over the carbazole
ring plane in a “folded wing” geometry. The bend angles at
the alkyl C of 16 are among the smallest ever observed in
a lanthanide trimethylsilylmethyl complex, and those in 17, while
not quite as acute, are also among the most bent known (LnꢀCꢀSi:
16, 113.5(2)°, 115.9(1)°; 17: 117.4(2)°, 118.4(2)°, 118.7(2)°,
a NOx1, NOx2, NCbz, and OTHF refer to the following atom labels
for 18, 19, and 20: N10, N20, N1, and O30; Cl1 and Cl2 refer
to the chlorine atoms trans and cis to the carbazole nitrogen,
respectively.
4962
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
Figure 3. Schematic view showing the disposition of the 40,40-dimethyl
groups of the Czx ligand in the solid-state structure of complexes 16
and 17.
Figure 4. 1H NMR spectrum (360 MHz, toluene-d8) of (Czx)Yb-
(CH2SiMe3)2, 17, at 250 K (*residual solvent resonances; ꢁ denotes an
impurity).
120.1(2)°; literature:15 mean = 127°, range 114.4ꢀ139.7°).
Close inspection of the orientation of the methyl groups on
the two oxazoline rings shows that the 40-carbon is rotated so that
one methyl group is about 30° out of the LnꢀNꢀC40 plane
(31.9ꢀ35.9°), while the other is nearly perpendicular to this
plane (88ꢀ93°). This allows a staggered arrangement of methyl
groups on the two oxazoline rings, but it also forces the
trimethylsilylmethyl groups to rotate back over the carbazole
plane to avoid strong steric repulsions with the perpendicular
methyl groups. The relative arrangement of the oxazoline methyl
groups and trimethylsilylmethyl ligands is shown schematically
in the cartoon in Figure 3. The small LnꢀCꢀSi angle observed
in 16 and 17 places a hydrogen of one methyl group on each
trimethylsilyl unit close to the Ln center (Ln H distance: 3.3
3 3 3
Å). While this is a relatively close contact, it is still outside the
sum of the van der Waals radii (3.2 Å),16 and the 1JCH coupling
constant for this CH3 is identical to that in Si(CH3)4 (1JCH
=
Figure 5. Variable-temperature 1H NMR (360 MHz, toluene-d8) δvs Tꢀ1
plot over the temperature range 200ꢀ340 K for (Czx)Yb(CH2SiMe3)2, 17.
118 Hz), so there is no evidence to support a γ-agostic interac-
tion. The LnꢀC bond distances fall well within the normal
range for five-coordinate lanthanide trimethylsilyl complexes
(distances arbitrarily corrected to reflect Y3+ in five-coordination:
16: 2.408, 2.414 Å; 17: 2.383, 2.393 Å; literature:17,18 mean =
2.39 ( 0.03 Å).
supports an electronic effect. Krogh-Jespersen and Brennan20
have examined a number of mixed ligand, mer-octahedral lantha-
nide complexes and shown conclusively that anions exert a
stronger trans influence than neutral donors in all cases.
Solution Behavior. The variable-temperature 1H NMR spec-
tra of the paramagnetic (Czx)Ln(CH2SiMe3)2 complexes, 16
(Er) and 17 (Yb), show seven resonances in toluene-d8 between
210 and 380 K, as illustrated for 17 at 250 K in Figure 4. The
corresponding δ versus 1/T plot for 17 is shown in Figure 5. The
observation of only seven resonances and reasonable straight line
behavior in the δ versus 1/T plots is consistent with the five-
coordinate, monomeric structure observed in the solid-state
X-ray structures.
There is some minor curvature evident in the lines for the
more shifted resonances at low temperature in Figure 5. This may
be due to hindered rotation about the LnꢀC bonds as the
temperature is lowered since slow rotation on the NMR time
scale would not result in decoalescence, assuming that the
approximately C2v geometry found in the X-ray structure repre-
sents the thermodynamically most favored conformation. This
may also explain why the alkyl CH2 resonance for the diamag-
netic complex (Czx)Y(CH2SiMe3)2, 15, shows a small, but
noticeable, temperature-dependent chemical shift (Figure 6).
The dichloride complexes 18ꢀ20 also show seven resonances
at room temperature (cf. Figure 7). This is inconsistent with the cis,
mer-geometry observed in the X-ray structures of 18ꢀ20, suggesting
that THF dissociation/reassociation is fast on the NMR time scale at
In contrast to the five-coordinate dialkyl complexes 16 and 17,
dichlorides 18ꢀ20 all coordinate one THF molecule and adopt a
six-coordinate, distorted octahedral geometry (Figure 2). Not
surprisingly given its rigid, planar structure, the Czx ligand is
coordinated in mer fashion with the two oxazoline N atoms in
trans, axial sites (NOxꢀLnꢀNOx: 159.59ꢀ161.53°) and the
anionic carbazole N in an equatorial position. In all three
dichloride structures (18ꢀ20), only the cis-dichloride isomer is
observed with one chloride trans to the carbazole N (NCbzꢀLnꢀCl:
166.04ꢀ166.92°) and one trans to the THF O (OTHFꢀLnꢀCl:
173.89ꢀ174.13°). The LnꢀN distances for the Cbz ligand are
very similar to those observed in the five-coordinate dialkyls 16
and 17. Given the fact that the average decrease in effective ionic
radii across the lanthanide series on going from six- to five-
coordination is 0.06 Å,18 we conclude that the dialkyl complexes
are more crowded than the dichlorides, an observation also
supported by the absence of coordinated THF in the dialkyls.
The terminal LnꢀCl distances fall within the range normally
observed for six-coordinate lanthanide dichlorides (2.52ꢀ2.62 Å).19
It is noteworthy, however, that the LnꢀCl bond trans to the
carbazole N is significantly longer than that trans to the THF O in
all three cases (Δ in Å: 18, 0.044; 19, 0.047; 20, 0.040). This
apparent trans influence of the carbazole N could be attributed
to steric effects, but the observation that the difference Δ is
essentially the same for all three metal ions argues against this and
4963
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
Figure 8. δ vs 1/T plot for the oxazoline CMe2 1H NMR resonances of
20 (360 MHz, toluene-d8, 2 = coalesced resonance).
Figure 6. Variable-temperature 1H NMR spectra (360 MHz, toluene-
d8) showing the upfield region for (Czx)Y(CH2SiMe3)2, 15 (*residual
solvent resonance; ꢁ denotes impurity growing in above room temperature).
Figure 9. δ vs 1/T plot for the oxazoline CH2 1H NMR resonances of
20 (360 MHz, toluene-d8, O = coalesced resonance).
Figure 7. 1H NMR spectrum (360 MHz, benzene-d6) of (Czx)Yb-
(Cl)2(THF), 20, at 293 K (*residual solvent resonance; ꢁ denotes an
impurity).
combined THF dissociation energy and the solvent and complex
reorganization energies. The previously reported LnꢀTHF bond
strengths from experimental and theoretical studies are smaller
than the ΔH‡ value observed here, although they do span a very
this temperature for all of these complexes. However, on cooling
20, the resonances due to the oxazoline CMe2 and CH2 groups
each decoalesce into two new resonances with a coalescence
temperature (Tc) of 272 K, as shown in Figures 8 and 9,
respectively. The remaining carbazole (Figure 10) and THF
(Figure 11) signals show the expected linear δ versus 1/T
behavior down to 190 K but do not undergo decoalescence.
These observations are entirely consistent with slow THF
exchange from a thermodynamically favored cis,mer geometry
(Cs symmetry) as observed in the solid-state X-ray structure. The
Eyring plot for the THF exchange process, shown in Figure 12,21
yielded values of ΔH‡ = 98 ( 10 kJ molꢀ1 and ΔS‡ = +185 ( 25 J
large range from 50 to 80 kJ molꢀ1 24,25
.
Reactivity. The dialkyl complexes 15ꢀ17 were initially exam-
ined as possible catalysts for the polymerization of ethylene,
ꢀ
isoprene, and 2,3-dimethylbutadiene using Ph3C+B(C6F5)4
(1 equiv) as activator. No polymer was observed with ethylene
or isoprene, but a small amount of polymer was observed in all
three cases with 2,3-dimethylbutadiene. However, since
Ph3C+B(C6F5)4ꢀ is capable of catalyzing the polymerization
of 2,3-dimethylbutadiene in its own right,26,27 the polymeri-
zation experiments on this monomer were repeated using just
slightly less than 1 equiv of Ph3C+ B(C6F5)4ꢀ as initiator. In
these experiments, no poly-1,3-dimethylbutadiene was pro-
duced, confirming our suspicion that it was the trityl cation
that was polymerizing 2,3-dimethylbutadiene and not any
cationic alkyls derived from 15ꢀ17.
molꢀ1
K
ꢀ1, while a value of ΔG‡ = 48 ( 3 kJ molꢀ1 was
calculated directly from Tc.22,23 The observed ΔS‡ is so large and
positive that even recognizing the large error associated with its
determination, there is little doubt the THF exchange process is
dissociative. This is not entirely surprising, given the fact that we
have isolated the stable five-coordinate Yb dialkyl 17. Assuming
this is a dissociative process, the measured ΔH‡ value reflects the
Two possibilities for the lack of polymerization activity suggest
themselves: either 15ꢀ17 do not form cationic alkyls on treatment
4964
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
Figure 12. Eyring plot for THF exchange in complex 20.
1
Figure 10. δ vs 1/T plot for the carbazole H NMR resonances of
20 (360 MHz, toluene-d8, 2 = 2,7-carbazole H, b = t-Bu, [ = 4,
5-carbazole H).
Figure 13. Crowding in the four-center σ-bond metathesis transition
state for alkene insertion (R = CH2SiMe3; R0 = C(Me)dCH2).
[B(C6F5)4]ꢀ complexes since Δδm,p values of ca. 4 ppm has been
observed both in systems where this anion is, and is not, coor-
dinated.29ꢀ32 Despite the ambiguity of the Δδm,p values, the
observation of some broadening in the 19F resonances for the
ortho- and meta-F of the C6F5 groups suggests that the anion is
interacting with the cation in 21.32 Complex 21 is stable for at
least a week at room temperature in toluene-d8, although it
decomposes on attempted isolation. We have also observed that
21 rapidly polymerizes THF when dissolved in that solvent,
suggesting that the “free” alkyl cation is highly reactive.33
The observations above unambiguously establish that alkyl cation
formation occurs when 15 is treated with Ph3C+B(C6F5)4ꢀ, thus
implicating slow alkene insertion as the reason 21 is inactive as a
diene polymerization catalyst. The extent to which strong anion
coordination is responsible for this slow insertion has not been
established, but it is reasonable to assume that this is part of the
problem. The failure of alkenes to insert into LnꢀC bonds in these
complexes may also be due to the coordination geometry of the Czx
ligand. Evidence presented earlier suggests that the methyl substit-
uents on the oxazoline ring force the alkyl ligands in complexes
15ꢀ17 to orient themselves over the carbazole group. It seems
reasonable to assume that these substituents also limit access of an
alkene substrate to the metal and interfere with the four-center,
σ-bond metathesis transition state since the inserting alkene must
straddle the site directly opposite the carbazole N in the transition
state and this region is the most crowded by the methyl groups
(Figure 13). Finally, it should be noted that the NNN-donor set
of the anionic carbazole-bis(oxazoline) ligand framework is strongly
electron donating, and this could reduce the Lewis acidity of the
1
Figure 11. δ vs 1/T plot for the THF H NMR resonances of 20
(360 MHz, toluene-d8, ` = β-THF, f = R-THF).
with Ph3C+B(C6F5)4ꢀ or the alkyl cations fail to insert 2,3-dimethyl-
butadieneat a reasonable rate. Examination of the reaction mixture
ꢀ
formed when 1 equiv of Ph3C+B(C6F5)4 is added to 15 in
toluene-d8 by NMR shows very clean generation of the alkyl
cation 21. The 1H NMR spectrum is surprisingly uninformative,
but the resonance for the remaining CH2SiMe3 group on Y now
integrates to two protons (2JYH = 2.8 Hz vs 2.6 Hz in 15) and the
oxazoline CMe2 group has shifted upfield to δ 1.10 ppm from
1.56 ppm in 15. The 13C NMR is more definitive however, as
the CH2 carbon directly bonded to Y shifts downfield by almost
10 ppm and the coupling constant increases by ca. 10 Hz (21:
δ 47.80, 1JYC = 48.8 Hz; 15: δ 37.46, 1JYC = 38.5 Hz). The change
in 1JYC is consistent with a stronger and shorter YꢀC bond in the
alkyl cation as might be expected. Examination of the 19F NMR
spectrum for the B(C6F5)4ꢀ anion shows a difference in chemical
shift between the meta- and para-F resonances of Δδm,p = 4.4
ppm. Differences in Δδm,p of more than 3.5 ppm have been taken
as evidence of cationꢀanion association for [(PhCH2)B(C6F5)3]ꢀ
anions,28 but this criterion does not appear to hold very well for
4965
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
metal center and inhibit monomer coordination for electro-
nic reasons.34
Sweeney, Z. K.; Rovis, T.; Tedrow, J. S. J. Am. Chem. Soc. 2001, 123,
12095. (k) Desimoni, G.; Faita, G.; Filippone, M.; Mella, M.; Zampori,
M. G.; Zema, M. Tetrahedron 2001, 57, 10203. (l) Qian, C.; Wang, L.
Tetrahedron: Asymmetry 2000, 11, 2347. (m) Qian, C.; Wang, L. Tetra-
hedron Lett. 2000, 43, 2203.
(4) (a) Hong, S.; Tian, S.; Metz, M. V.; Marks, T. J. J. Am. Chem. Soc.
2003, 125, 14768. (b) Kodama, H.; Ito, J.; Hori, K.; Ohta, T.; Furakawa,
I. J. Organomet. Chem. 2000, 603, 6. (c) Guillon, H.; Daniele, S.; Hubert-
Pfalzgraf, L. G.; Letoffe, J. M. Polyhedron 2004, 23, 1467. (d) Alaaeddine,
A.; Amgoune, A.; Thomas, C. M.; Dagorne, S.; Bellemin-Laponnaz, S.;
Carpentier, J.-F. Eur. J. Inorg. Chem. 2006, 3652. (e) Gorlitzer, H. W.;
Spiegler, M.; Anwander, R. J. Chem. Soc., Dalton Trans. 1999, 4287.
(5) (a) Inoue, M.; Nakada, M. Heterocycles 2007, 72, 133. (b) Inoue,
M.; Nakada, M. Angew, Chem., Int. Ed. 2006, 45, 252. (c) Inoue, M.;
Nakada, M. Org. Lett. 2004, 6, 2977. (d) Inoue, M.; Suzuki, T.; Nakada,
M. J. Am. Chem. Soc. 2003, 125, 1140. (e) Suzuki, T.; Kinoshita, A.;
Kawada, H.; Nakada, M. Syn. Lett. 2003, 570.
(6) (a) Durꢀan-Galvꢀan, M.; Worlikar, S. A.; Connell, B. T. Tetrahe-
dron 2010, 66, 7707. (b) Durꢀan-Galvꢀan, M.; Hemmer, J. R.; Connell,
B. T. Tetrahedron Lett. 2010, 51, 5080. (c) Durꢀan-Galvꢀan, M.; Connell,
B. T. Eur. J. Org. Chem. 2010, 2445.
(7) (a) Gibson, V. C.; Spitzmesser, S. K.; White, A. J. P.; Williams,
D. J. J. Chem. Soc., Dalton Trans. 2003, 2718. (b) Hollas, A. M.; Gu, W.;
Bhuvanesh, N.; Ozerov, O. V. Inorg. Chem. 2011, 50, 3673.
(8) Mudadu, M. S.; Singh, A. J.; Thummel, R. P. J. Org. Chem. 2008,
73, 6513.
Preliminary experiments have also shown that 15ꢀ17 under-
go hydrogenolysis with H2 under pressure to afford complex
hydrides, which may also serve as polymerization or hydrogena-
tion catalysts. Despite the low-reactivity catalytic activity
observed so far, we have established that the carbazole-bis-
(oxazoline) ligand system is an attractive ancillary for orga-
nolanthanide chemistry, providing a well-defined mer geom-
etry, appropriate steric saturation, and low susceptibility to
participate in chemical reactions with either strong nucleo-
philes or electrophiles.
’ ASSOCIATED CONTENT
S
Supporting Information. Tables of atomic coordinates,
b
bond distances and angles, and anisotropic thermal param-
eters for 16ꢀ20 as well as tables summarizing comparative
X-ray structural data are available free of charge via the Inter-
net at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: djberg@uvic.ca.
(9) (a) Wang, L.; Cui, D.; Hou, Z.; Li, W.; Li, Y. Organometallics
2011, 30, 760. (b) Johnson, K. R. D.; Hayes, P. G. Organometallics 2009,
28, 6352.
(10) (a) Lappert, M. F.; Pearce, R. J. Chem. Soc., Chem. Commun.
1973, 126. (b) Barker, F. K.; Lappert, M. F. J. Organomet. Chem. 1974,
76, C45. (c) Hultzch, K. C.; Voth, P.; Beckerle, K.; Spaniol, T. P.; Okuda,
J. Organometallics 2000, 19, 228. (d) Clark, D. L.; Sattelberger, A. P.
Inorg. Synth. 1997, 31, 307.
’ ACKNOWLEDGMENT
D.J.B. (Discovery Grant) gratefully acknowledges the support
of the Natural Sciences and Engineering Research Council of
Canada.
(11) SAINTPlus, v6.22, Data Reduction and Correction Program;
Bruker AXS: Madison, WI, 2001.
’ REFERENCES
(12) Sheldrick, G. M. SADABS, v2.01, Empirical Absorption Correc-
tion Program; Bruker AXS: Madison, WI, 2001.
(13) Sheldrick, G. M. SHELXTL, v6.10, Structure Determination
Software Suite; Bruker AXS: Madison, WI, 2001.
(14) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(15) Mean value and range for the LnꢀCꢀSi bond angles in 36
reported five-coordinate (L)Ln(CH2SiMe3)2 complexes. Details of the
complexes and literature references are included in the Supporting
Information.
(16) (a) Bondi, A. J. Phys. Chem. 1964, 68, 441. (b) Batsanov, S. S.
J. Mol. Struct. 1999, 468, 151. (c) Mantina, M.; Chamberlin, A. C.;
Valero, R.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. A 2009, 113, 5806.
(17) Mean value for the LnꢀC bond length in 36 reported five-
coordinate (L)Ln(CH2SiMe3)2 complexes, arbitrarily corrected to the
ionic radius of Y3+ in five-coordination (0.84 Å) as given in ref 18. Details
of the complexes and literature references are included in the Supporting
Information.
(1) Many reviews are available covering this area; some of the more
recent reviews: (a) Xie, J.-H.; Zhu, S.-F.; Zhou, Q.-L. Chem. Rev. 2011,
111, 1713. (b) Singh, P. K.; Singh, V. K. Pure Appl. Chem. 2010, 82, 1845.
(c) Framery, E.; Andrioletti, B.; Lemaire, M. Tetrahedron: Asymmetry
2010, 21, 1110. (d) Hargaden, G. C.; Guiry, P. J. Chem. Rev. 2009,
109, 2505.(e) Pfaltz, A. Bisoxazolinesꢀa privileged ligand class for
asymmetric catalysis. In Asymmetric Synthesis, 2nd ed.; Christian, M.,
Braese, S., Eds.; Wiley, VCH Verlag GmbH & Co.: Weinheim, Germany,
2008; pp 139ꢀ143. (f) Rasapan, R.; Laventine, D.; Reiser, O. Coord.
Chem. Rev. 2008, 252, 702. (g) Nishiyama, H.; Ito, J.; Shiomi, T.;
Hashimoto, T.; Miyakawa, T.; Kitase, M. Pure Appl. Chem. 2008, 80, 743.
(h) Desmoni, G.; Faita, G.; J€orgensen, K. A. Chem. Rev. 2006, 106, 3561.
(2) (a) Kobayashi, S.; Yamashita, Y. Acc. Chem. Res. 2011, 44, 58.
(b) Matsunaga, S.; Shibasaki, M. Bull. Chem. Soc. Jpn. 2008, 81, 60.
(c) Aspinall, H. C.; Greeves, N. J. Organomet. Chem. 2002, 647, 151.
(d) Johnson, J. S.; Evans, D. A. Acc. Chem. Res. 2000, 33, 325. (e) Nishiyama,
H. Enantiomer 1999, 4, 569.
(18) Shannon, R. D. Acta Crystallogr. A 1976, 32A, 751.
(19) Mean value for the LnꢀCl bond length in 27 reported six-
coordinate (L)2(X)Ln(Cl)2(THF) complexes, arbitrarily corrected to
the ionic radius of Y3+ in six-coordination (0.90 Å) as given in ref 18.
Details of the complexes and literature references are included in the
Supporting Information.
(3) (a) Sato, H.; Suzuki, Y.; Takai, Y.; Kawasaki, H.; Arakawa, R.;
Shizuma, M. Chem. Lett. 2010, 39, 564. (b) Suga, H.; Higuschi, S.;
Ohtsuka, M.; Ishimoto, D.; Arikawa, T.; Hashimoto, Y.; Misawa, S.;
Tsuchida, T.; Kakehi, A.; Baba, T. Tetrahedron 2010, 66, 3070. (c)
Matsumoto, K.; Suzuki, K.; Tsukuda, T.; Tsubomura, T. Inorg. Chem.
2010, 49, 4717. (d) Barros, M. T.; Phillips, A. M. F. Tetrahedron:
Asymmetry 2010, 21, 2746. (e) de Bettancourt-Dias, A.; Viswanathan, S.;
Rollett, A. J. Am. Chem. Soc. 2007, 129, 15436. (f) Lu, G.; Morimoto, H.;
Matsunaga, S.; Shibasaki, M. Angew. Chem., Int. Ed. 2008, 47, 6847. (g)
Morimoto, H.; Lu, G.; Aoyama, N.; Matsunaga, S.; Shibasaki, M. J. Am.
Chem. Soc. 2007, 129, 9588. (h) Comelles, J.; Pericas, A.; Moreno-
Moreno-Manas, M.; Vallribera, A.; Drudis-Solꢀe, G.; Lledos, A.; Parella,
T.; Roglans, A.; Garcia-Granda, S.; Roces-Fernꢀandez, L. J. Org. Chem.
2007, 72, 2077. (i) Aspinall, H. C.; Bickley, J. F.; Greeves, N.; Kelly,
R. V.; Smith, P. M. Organometallics 2005, 24, 3458. (j) Evans, D. A.;
(20) Krogh-Jespersen, K.; Romanelli, M. D.; Melman, J. H.; Emge,
T. J.; Brennan, J. G. Inorg. Chem. 2010, 49, 552.
(21) The rate constants were determined by line-shape analysis
using WINDNMR: Reich, H. J. WINDNMR (Vers. 7.1.13); J. Chem.
Educ. Software 3D2: Madison, WI, 2008. Paramagnetic line broadening
results in larger than normal errors in the estimation of the rate constant
k by this method, especially at temperatures more than 10 K above and
below coalescence. However, line broadening due to exchange was still
clearly observable within this window.
4966
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967

Organometallics
ARTICLE
(22) Calculated at the coalescence temperature using the two-site,
equal-population exchange formula (ref 23): ΔG‡ = 1.912 ꢁ 10ꢀ2(Tc)-
[9.972 + log(Tc/δυ)] in kJ molꢀ1. A value of δυ = 1840 Hz was
calculated by extrapolation of the straight lines for the two inequivalent
CMe2 resonances of 20, shown in Figure 8, to Tc = 272 K. The error
estimated for ΔG‡ assumes a liberal error of 80 Hz in δυ.
(23) Sandstrom, J. Dynamic NMR Spectroscopy; Pergamon: London,
1982.
(24) LnꢀTHF bond dissociation energies (kJ molꢀ1): (a)
(C5Me5)2LuCl(THF), 54 (expt) Gong, L.; Streitwiesser, A., Jr.; Zalkin,
A. J. Chem. Soc., Chem. Commun. 1987, 460. (b) (C5H5)2LnX(THF), 50
(La), 65 (Gd), 80 (Lu) (calc): Luo, Y.; Selvam, P.; Ito, Y.; Endou, A.;
Kubo, M.; Miyamoto, A. J. Orgnomet. Chem. 2003, 679, 84. (c) Yb(BH4)3-
(THF)2, 52 (expt), 54 (calc): Gafurov, B. A.; Khakerov, I. Z.; Badalov, A. B.;
Mirsaidov, I. V. Int. J. Hydrogen Energy 2011, 36, 1309.
(25) Closely related bond dissociation energies for comparison
(kJ molꢀ1): (a) (C5H4SiMe3)3U(THF), 41 (expt): Schock, L. E.;
Seyam, A. M.; Sabat, M.; Marks, T. J. Polyhedron 1988, 7, 1517. (b)
(C5Me5)2Sm(THF), 31 (expt): Nolan, S. P.; Stern, D. L.; Marks, T. J.
J. Am. Chem. Soc. 1989, 111, 7844. (c) η6-Arene dissociation from Yb
(S-2,6-η6-C6H3Ar*2)2 (Ar* = 2,4,6-C6H2Pri3), 54 (expt); Yb(SH)2-
(C6H6)2, 32 (calc, B3LYP) and 74 (calc, MP2): Niemeyer, M. Eur.
J. Inorg. Chem. 2001, 1969. (d) (TpMe2)UCl3(THF), 21.5 (expt): Leal,
J. P.; Martinho Sim~oes, J. A. J. Chem. Soc., Dalton Trans. 1994, 2687. (e)
LaI3(2-pyrazine)3 (av), 78 (calc): Mazzanti, M.; Wietzke, R.; Pꢀecaut, J.;
Latour, J.-M.; Maldivi, P.; Remy, M. Inorg. Chem. 2002, 41, 2389.
(26) In a typical experiment, a 150:1 molar ratio of [2,3-butadiene]/
[Ph3C+B(C6F5)4ꢀ] at 25 °C in toluene resulted in complete conversion
to poly-2,3-butadiene within 5 min; Mn = 1.0 ꢁ 105 Da and Mw/Mn =
2.57 (GPC using THF as solvent calibrated against polystyrene standards).
Currently, we do not know whether polymerization is initiated by
electrophilic addition of trityl cation or H+ (from trace adventitious
water). Several examples of polymerization by trityl cation (or H+
derived from trace water in its presence) have been reported, ref 27.
(27) (a) Abu-Abdoun, I. I.; Ledwith, A. J. Polym. Res. 2007, 14, 99.
(b) Ihara, E.; Yoshida, N.; Ikeda, J.-I.; Itoh, T.; Inoue, K. J. Polym. Sci.,
Part A 2006, 2636. (c) Tasdelen, M.; A.; Degirmenci, M.; Yagci, Y.;
Nuyken, O. Polym. Bull. 2003, 50, 131. (d) Acar, M. H.; Yagci, Y.;
Schnabel, W. Polym. Int. 1998, 48, 331. (e) Acar, M.; K€u0€uk€oner, M.
Polymer 1997, 38, 2829. (f) Gonzꢀalez, M.; Rodriguez, M.; Leꢀon, L. M.;
Gonzꢀalez, M. C.; Zamora, F. Polym. Bull. 1989, 22, 163. (g) Asami, R.;
Hasegawa, K. Polym. J. 1976, 8, 53. (h) Asami, H.; Hasegawa, K.; Onoe,
T. Polym. J. 1976, 8, 43. (i) Slomkovski, S.; Kubisa, P.; Penczek, S. Nuova
Chim. 1973, 49, 51.
(28) Horton, A. D.; de With, J. Organometallics 1997, 16, 5424.
(29) Coordinated [B(C6F5)4]ꢀ anion: Zuccaccia, C.; Stahl, N. G.;
Macchioni, A.; Chen, M.-C.; Roberts, J. A.; Marks, T. J. J. Am. Chem. Soc.
2004, 126, 1448 ((C5Me5)2Th(CH3)(B(C6F5)4), Δδm,p = 3.89 ppm).
(30) Noncoordinated [B(C6F5)4]ꢀ anion: (a) Ref 27 ([C5Me4-
SiMe2NBut)Zr(CH3)(THF)]+[B(C6F5)4]ꢀ, Δδm,p = 4.00 ppm). (b)
Li, D.; Li, S.; Cui, D.; Zhang, X.; Trifonov, A. J. Chem. Soc., Dalton Trans.
2011, 40, 2151. ([(NacNac)Sc(CH2SiMe3)]+[B(C6F5)4]ꢀ, Δδm,p
3.83 ppm and directly compared with [Ph3C]+[B(C6F5)4]ꢀ, Δδm,p
3.96 ppm, both in C6D6 solution.)
=
=
(31) (a) Alonso-Moreno, C.; Lancaster, S. J.; Wright, J. A.; Hughes,
D. L.; Zuccaccia, C.; Correa, A.; Macchioni, A.; Cavallo, L.; Bochmann,
M. Organometallics 2008, 27, 5474. (b) Bryliakov, K. P.; Talsi, E. P.;
Voskoboynikov, A. Z.; Lancaster, S. J.; Bochmann, M. Organometallics
2008, 27, 6333.
(32) Jia, L.; Yang, X.; Stern, C. L.; Marks, T. J. Organometallics 1997,
16, 842.
(33) Although[Ph3C]+[B(C6F5)4]ꢀ can also polymerize THF, the same
reactivity was observed when less than 1 equiv of the trityl borate was used, so
we believe it is actually 21 that initiates the polymerization of THF in this case.
(34) The authors wish to thank a referee for pointing this possi-
bility out.
4967
dx.doi.org/10.1021/om200588y |Organometallics 2011, 30, 4958–4967