Synthesis and structural characterization of zirconium complexes containing aminodiolate ligands and their use as Lewis acid catalysts

Article abstract of DOI:10.1139/v00-001

A series of aminodiols RN(CH₂CH₂C(OH)R′₂)₂ (R, R′ = Me, Me 4; Me, Ph 5; tert-butyl, Me 6; tert-Bu, Ph 7; (S)-PhCH(Me), Me 8) were prepared by the Michael addition of a primary amine to methyl acrylate followed by reaction of the resulting aminodiester with excess methyl or phenyl lithium. Reaction of two equivalents of the aminodiols 4-8 with tetrabenzylzirconium afforded the zirconium bis(aminodiolates) 10-14 in excellent yield. Complex 11 (R, R′ = Me, Ph) adopts a cis, fac-octahedral geometry in solution and in the solid state. Complexes 10-14 are fluxional in solution by NMR spectroscopy: small substituents at nitrogen and large substituents at the alkoxide carbons slow the rate of exchange. The chiral complex 14 functions as a Lewis acid catalyst in the nitroaldol (Henry) reaction and the oxidation of geraniol by tert-butyl hydroperoxide with modest enantioselectivities (30 and 46% enantiomeric excess (ee), respectively).

Full text of DOI:10.1139/v00-001

255
Synthesis and structural characterization of
zirconium complexes containing aminodiolate
ligands and their use as Lewis acid catalysts
Pengcheng Shao, Roland A.L. Gendron, and David J. Berg
Abstract: A series of aminodiols RN(CH2CH2C(OH)R′2)2 (R, R′ = Me, Me 4; Me, Ph 5; tert-butyl, Me 6; tert-Bu, Ph
7; (S)-PhCH(Me), Me 8) were prepared by the Michael addition of a primary amine to methyl acrylate followed by
reaction of the resulting aminodiester with excess methyl or phenyl lithium. Reaction of two equivalents of the
aminodiols 4–8 with tetrabenzylzirconium afforded the zirconium bis(aminodiolates) 10–14 in excellent yield. Complex
11 (R, R′ = Me, Ph) adopts a cis, fac-octahedral geometry in solution and in the solid state. Complexes 10–14 are
fluxional in solution by NMR spectroscopy: small substituents at nitrogen and large substituents at the alkoxide
carbons slow the rate of exchange. The chiral complex 14 functions as a Lewis acid catalyst in the nitroaldol (Henry)
reaction and the oxidation of geraniol by tert-butyl hydroperoxide with modest enantioselectivities (30 and 46%
enantiomeric excess (ee), respectively).
Key words: diol, Michael addition, zirconium, synthesis, structure, fluxional, alkoxide, Lewis acid, catalysis.
Résumé : On a préparé une série d’aminodiols, RN(CH2CH2C(OH)R2)2 (R, R′ = Me, Me 4; Me, Ph 5; tert-butyl, Me
6; tert-butyl, Ph 7; (S)-PhCH(Me), Me 8), par le biais d’une addition de Michael d’une amine primaire sur l’acrylate
de méthyle, suivie par une réaction de l’aminodiester qui en résulte avec un excès de méthyl ou de phényllithium. La
réaction de deux équivalents des aminodiols 4–8 avec du tetrabenzylzirconium conduit aux bis(aminodiolates) de zirco-
nium 10–14 avec d’excellents rendements. En solution ainsi qu’à l’état solide, le complexe 11 (R, R′ = Me, Ph) adopte
une géométrie octaédrique cis, fac. En solution, la spectroscopie RMN indique que les complexes 10–14 sont en état
de flux: des substituants de faibles tailles au niveau de l’azote et de taille importante au niveau des carbones des
alcoolates ralentissent l’échange. Le complexe chiral 14 agit comme acide de Lewis pour catalyser la réaction de
nitroaldol (Henry) et l’oxydation du géraniol par l’hydroperoxyde de tert-butyle avec des énantiosélectivités modestes
(30 et 46% respectivement).
Mots clés : diol, addition de Michael, zirconium, synthèse, structure, fluxion, alcoolate, acide de Lewis, catalyse.
[Traduit par la Rédaction] Shao et al. 264
Introduction
Alkoxides (1), aryloxides (2), and siloxides (3) have been
extensively used as ancillary ligands in group
4
organometallic chemistry. In addition, chelating biphenolates
(4) and binaphtholates (5) have been employed, the latter be-
ing especially important in asymmetric catalysis due to its
inherent C2 symmetry. Chelating tripodal alkoxides based on
triethanolamine have been extensively investigated by
Verkade et al. (6) who demonstrated the hydrolytic stability
of the chelate array in complexes such as I. Nugent has de-
veloped a C3 chiral analog of I, using the homochiral triol
II, which shows high catalytic enantioselectivity in the ring
opening of meso epoxides by azidotrialkylsilanes (7).
Ligands such as those derived from triethanolamine or
triol II are best suited to forming complexes with smaller
metals such as titanium. In order to form monomeric zirco-
nium complexes, additional steric bulk is required and this is
most easily achieved by the introduction of two substituents
at the alkoxide carbon. Replacement of one chelate arm by
an alkyl group allows for further steric tuning by variation of
the substituent at nitrogen and lowers the charge to two. This
latter point is important because an aminodiolate ligand
Received October 26, 1999.
P. Shao, R.A.L. Gendron, and D.J. Berg.1 Department of
Chemistry, University of Victoria, P.O. Box 3065, Victoria,
BC V8W 3V6, Canada.
1Author to whom correspondence should be addressed.
Telephone: (250) 721-7161. Fax: (250) 721-7147.
e-mail: DJBERG@UVIC.CA
Can. J. Chem. 78: 255–264 (2000)
© 2000 NRC Canada

256
Can. J. Chem. Vol. 78, 2000
Table 1. H NMR spectra for aminodiesters 1–3 and aminodiols 4–8.a
MeN t-BuN PhCHN PhCHN MeCHN NCH2
Aminodiester
1
CH2CO
CO2Me COH
C(O)Me2 C(O)Ph2
1
2
3
2.19
2.65 (t, 7.2) 2.41 (t, 7.2) 3.61
2.78 (t, 7.6) 2.40 (t, 7.6) 3.62
1.33 (d, 6.8) 2.82 (dt) 2.40 (t, 7.0) 3.61
2.70 (dt)
1.01
3.81 (q, 6.8) 7.27 (m)
7.20 (m)
Aminodiol
4
5
2.22
2.22
2.56 (t, 6.9) 1.61 (t, 6.9)
2.36 (m)b 2.36 (m)b
4.05 (br) 1.19
5.22 (br)
7.38 (m)
7.27 (m)
7.18 (m)
6
7
0.95
0.69
2.58 (t, 6.9) 1.67 (t, 6.9)
4.54
5.23
1.22
2.51 (m)b
2.47 (m)b
7.58 (ortho)
7.13 (meta)
7.00 (para)
8
3.93 (q)
7.3 (m)
1.36 (d)
2.68 (t)
2.57 (t)
1.60 (m)
4.19
1.10
1.00
aRecorded at 360 MHz in CDCl3; chemical shifts (δ) in ppm relative to TMS; multiplicity is not given for singlets.
bOverlapping resonances.
1
Table 2. H NMR spectra for zirconium aminodiolates 10–14.a
MeN t-BuN PhCHN
PhCHN
MeCHN
NCH2
CH2CO
C(O)Me2 C(O)Ph2
10
2.14
3.46 (br t)
3.40 (br t)
1.20 (br)b
1.3 (br)b
2.50 (br)
3.14 (t, 12.4)
2.94 (t, 12.3)
2.00 (br t)
1.72b
1.52
1.48
1.28
1.22
1.68b
1.43b
1.60 (br)
2.53 (dd, 15.3, 5.3)
2.29 (dd, 15.2, 11.9)
10c 2.20
11d 1.56
1.33
8.25 (dd)
8.19 (dd)
6.7–7.7 (m)
1.38 (dd, 17.7, 5.5) 2.11 (dd, 14.0, 5.2)
1.00 (dd, 12.7, 5.5) 1.93 (dd, 13.8, 5.1)
12
13
1.11
2.90 (t, 6.7)
2.51 (br)
1.75 (t, 6.7)
2.51 (br)
1.25
0.75
7.41 (d, 7.3)
7.11 (t, 7.4)
7.02 (d, 7.3)
1.31
14e
4.15 (q, 6.7) 7.0–7.3 (m) 1.54 (d, 6.7) 2.98 (m)
2.57 (m)
1.75 (m)
1.68 (m)
1.23
aRecorded at 360 MHz in C6D6; chemical shifts (δ) in ppm relative to TMS; multiplicity and coupling constants (where measured) are given in
brackets; singlets have not been denoted.
bPartially obscured.
cRecorded at 80°C.
d1H–13C COSY shows the following pairs of resonances are due to protons attached to the same carbon: 3.14, 1.00; 2.94, 1.38; 2.53, 2.29; 2.11, 1.93.
eRecorded at 70°C.
allows access to dialkyls, alkyl cations, imides, and many
other important classes of compounds not possible for sys-
tems such as I. In this contribution, we describe the synthe-
sis of several aminodiolate ligands and their zirconium
coordination complexes. The nature of the alkoxide carbon
and nitrogen substituents will be shown to control the flux-
ional behaviour of these complexes.
ing standard glovebox (Braun MB150–GII) or Schlenk tech-
niques. Tetrahydrofuran (THF), hexane, and toluene were
dried by distillation from sodium benzophenone ketyl under
argon immediately prior to use. Tetrabenzylzirconium was
prepared according to the literature (8).
1H, 13C, and all variable temperature NMR spectra were
recorded on a Bruker AC–300 or AMX–360 MHz spectrom-
eter. Spectra were recorded in C6D6 or C7D8 solvent, previ-
ously distilled from sodium under argon, using 5 mm tubes
fitted with a teflon valve (Brunfeldt). 1H and 13C NMR spec-
Experimental section
1
tra were referenced to residual solvent resonances. H NMR
General procedures
All manipulations were carried out under an argon atmo-
sphere, with the rigorous exclusion of oxygen and water, us-
data for the ligands and their precursors are given in Table 1
while data for their zirconium complexes are listed in
© 2000 NRC Canada

Shao et al.
257
Table 2; 13C NMR data is given in the following section.
Melting points were recorded using a Reichert hot stage and
are not corrected. Elemental analyses were performed by
Canadian Microanalytical, Delta, B.C. or Atlantic
Microanalytical, Atlanta, Ga. Mass spectra were recorded on
a Finnegan 3300 or a Kratos Concept H spectrometer using
chemical ionization or electron impact (70 eV) sources, re-
spectively.
fractional distillation. Further drying over molecular sieves
(4 Å) in toluene solution was carried out prior to reaction
with tetrabenzylzirconium.
MeN(CH2CH2C(OH)Me2)2 (4): Colourless oil. Yield: 81%;
13C(1H) NMR (CDCl3) δ: 70.6 (COH), 53.9 (NCH2), 41.9
(NMe), 38.8 (CH2CO), 29.7 (C(OH)Me). MS(CI) m/z: 204
(M+ +1, 100%), 186 (M+ – OH, 11%), 130 (M+ – (CH2C(OH)Me2),
71%). Anal. calcd. for C11H25NO2: C, 64.98, H 12.39, N
6.89; found: C 65.31, H 11.90, N 7.14.
Aminodiester preparation
The esters 1–3 were prepared by Michael addition of the
appropriate amine to methyl acrylate. In a representative
procedure, the amine (0.1 mol) was dissolved in 200 mL
methanol in a 1 L round bottom flask fitted with a dropping
funnel and reflux condenser. Dropwise addition of methyl
acrylate (0.2 mol) over a period of 1 h followed by stirring
at room temperature for 5 h was sufficient to complete the
formation of ester 1. In contrast, overnight reflux was neces-
sary to complete the addition in the case of 3 and prolonged
reflux (1 week) still resulted in a mixture of mono and
diester for 2. After cooling, methanol and any low boiling
materials were removed by rotary evaporation using a water
aspirator. The remaining liquids were then distilled under re-
duced pressure (0.1 Torr (1 Torr = 133.3 Pa)). In the case of
2, the monoester was easily removed by fractional distilla-
tion leaving the pure diester as the higher boiling fraction.
MeN(CH2CH2C(OH)Ph2)2 (5): White crystals. Yield: 48%;
mp 170°C; 13C(1H) NMR (CDCl3) δ: 147.2 (quat-arylC),
128.1 (arylC), 126.6 (arylC), 125.8 (arylC), 78.2 (COH),
53.7 (NCH2), 42.3 (NMe), 37.5 (CH2CO); MS(CI) m/z: 480
(M+ +29, 20%), 452 (M+ +1, 100%), 374 (M+ – Ph, 12%),
– (CH2C(OH)Ph2), 5%). Anal. calcd. for
C31H33NO2: C 82.45, H 7.36, N 3.10; found: C 82.48, H
7.43, N 3.54.
254 (M+
tert-BuN(CH2CH2C(OH)Me2)2 (6): White crystals. Yield:
65%; mp 52°C; 13C(1H) NMR (CDCl3) δ: 70.6 (COH), 55.8
(NCMe3), 46.5 (NCH2), 42.4 (CH2COH), 30.2 (C(OH)Me),
26.9 (NCMe3); MS(CI) m/z: 274 (M+ +29, 13%), 246 (M+
+1, 100%), 230 (M+
–
Me, 21%), 172 (M+
–
(CH2C(OH)Me2), 78%). Anal. calcd. for C14H31NO2: C
68.52, H 12.73, N 5.71; found: C 68.43, H 12.47, N 6.18.
MeN(CH2CH2CO2Me)2 (1): Yield: 95%; bp 120°C (0.1
Torr); 13C(1H) NMR (CDCl3) δ: 172.8 (C=O), 52.4 (NCH2),
51.5 (OMe), 41.7 (NMe), 32.3 (CH2CO); MS(CI) m/z: 204
(M+ +1, 100%), 130 (M+ – (CH2CO2Me), 88%).
tert-BuN(CH2CH2C(OH)Ph2)2 (7): White crystals. Yield:
75%; mp 120°C; 13C(1H) NMR (CDCl3) δ: 147.5 (quat-
arylC), 128.1 (arylC), 126.5 (arylC), 125.9 (arylC), 78.6
(COH), 56.1 (NCMe3), 47.2 (NCH2), 40.0 (CH2CO), 26.6
(NCMe3); MS(CI) m/z: 512 (M+ +29, 15%), 494 (M+ +1,
100%), 296 (M+ – (CH2C(OH)Ph2), 15%). A satisfactory el-
emental analysis was not obtained for this compound.
tert-BuN(CH2CH2CO2Me)2 (2): Yield: 52%; bp 110°C (0.1
Torr); 13C(1H) NMR (CDCl3) δ: 173.1 (C=O), 55.1
(NCMe3), 51.4 (OMe), 46.1 (NCH2), 36.5 (CH2CO), 27.1
(NCMe3); MS(CI) m/z: 246 (M+ +1, 92%), 230 (M+ – Me,
15%), 188 (M+ – tert-Bu, 80%), 172 (M+ – (CH2CO2Me),
100%).
(S)-PhMeC(H)N(CH2CH2C(OH)Me2)2 (8): This compound
was isolated as a sticky, colourless oil after column chroma-
tography on silica gel (eluant: 95% ethyl acetate, 5% metha-
nol). Yield: 62%; 13C(1H) NMR (CDCl3) δ: 141.5 (quat-
arylC), 128.4 (arylC), 128.1 (arylC), 127.2 (arylC), 70.5
(COH), 58.2 (PhCHN), 45.0 (NCH2), 38.6 (CH2CO), 29.6
(C(OH)Me), 29.3 (C(OH)Me), 14.7 (PhCMeN); MS(CI) m/z:
322 (M+ +29, 12%), 294 (M+ +1, 100%), 278 (M+ – Me,
8%), 220 (M+ – (CH2C(OH)Me2), 71%). Anal. calcd. for
C18H31NO2: C 73.68, H 10.65, N 4.77; found: C 72.58, H
10.40, N 5.22.
(S)-PhMeC(H)N(CH2CH2CO2Me)2 (3): Yield: 85%; bp
162–166°C (0.05 Torr); 13C(1H) NMR (CDCl3) δ: 172.9
(C=O), 148.4 (quat-arylC), 128.1 (arylC), 127.6 (arylC),
126.8 (arylC), 58.9 (PhCHN), 51.4 (OMe), 45.8 (NCH2),
33.4 (CH2CO), 16.2 (MeCHN).
Aminodiol preparation
Treatment of aminodiesters (1–3) with an excess of
methyl or phenyllithium allowed isolation of a series of
aminodiols (4–8). In a typical preparation, a 10-fold excess
of methyl or phenyllithium was generated in situ from
iodomethane and lithium or bromobenzene and n-
butyllithium, respectively (diethyl ether solvent). The lith-
ium reagent was then added by dropping funnel to a solution
of the aminodiester in diethyl ether cooled to 0°C. The re-
sulting turbid solution was stirred for 1 h at 0°C and allowed
to warm to room temperature overnight. The reaction mix-
ture was subsequently quenched with ice water, extracted
with dichloromethane and the organic extract dried over an-
hydrous MgSO4. Filtration and removal of the solvent under
reduced pressure yielded the aminodiols as white crystals or
colourless oils. The secondary amino alcohol 9 was obtained
as a byproduct in the preparation of 8 and was separated by
(S)-PhMeC(H)NH(CH2CH2C(OH)Me2) (9): Colourless oil.
1H NMR (C6D6) δ: 7.22–7.34 (5H, m, arylH), 3.71 (1H, q,
PhC(H)N), 3.6 (2H, br s, OH and NH), 2.73 (1H, dt, NCH2),
2.69 (1H, dt, NCH2), 1.62 (1H, dt, CH2C(OH)), 1.48 (1H, dt,
CH2C(OH)), 1.39 (3H, d, PhC(Me)), 1.19 (3H, s, C(OH)Me),
1.07 (3H, s, C(OH)Me). The resonance at 3.6 ppm disap-
pears after treatment with D2O.
Zirconium complexes
Complexes 10–14 were prepared by adding a solution of
two equivalents of aminodiol 4–8 in toluene to a stirred tolu-
ene solution of tetrabenzylzirconium. After stirring for 1 h,
the volatiles were removed under reduced pressure and the
crude solid product was recrystallized from a mixture of to-
luene and hexane.
© 2000 NRC Canada

258
Can. J. Chem. Vol. 78, 2000
Zr{MeN(CH2CH2C(O)Me2)2}2 (10): Yellow microcystalline
powder. Yield: 96%; mp 206–212°C; 13C(1H) NMR (C6D6)
23°C δ: 73.7, 72.6 (CO); 58.0, 52.7 (NCH2); 45.8 (NMe);
38.9, 38.8 (CH2CO); 33.7, 33.1, 29.5 (br, two overlapping
resonances), (C(O)Me); 80°C δ 73.3 (CO), 56.0 (br, NCH2),
46.1 (NMe), 39.3 (CH2CO), 31.6 (C(O)Me); MS(EI) m/z:
493 (M+, 31%), 407 (M+ – CH2CH2C(O)Me2, 81%), 114
(H2CNCH2CH2C(O)Me2, 100%). Anal. calcd. for C22H46N2O4Zr:
C 53.51, H 9.39, N 5.67; found: C 53.71, H 9.28, N 5.94.
(9) δ: 6.20 (1H, dd, =CHCHMe, J = 6, 2 Hz), 4.87 (1H, dd,
OCH(OEt), J = 9, 2 Hz), 4.53 (1H, m, =CH(O)). The re-
maining resonances of this product were obscured by
unreacted starting material and free ligand signals.
Nitroaldol reaction
A solution of 14 (5 mol%) in 10 mL dry nitromethane
was prepared under argon in a Kontes valve flask. Approxi-
mately 0.5 g of ArC(O)H (Ar = phenyl or β-naphthyl) was
added, the flask sealed and the reaction mixture stirred at
room temperature for up to 7 days. Samples were periodi-
Zr{MeN(CH2CH2C(O)Ph2)2}2 (11): Colourless crystals. Yield:
90%; mp > 285°C; 13C(1H) NMR (C6D6) δ: 152.9, 151.2,
149.6, 149.4 (quat-arylC); 126.3, 126.1, 126.0, 125.9 (para-
arylC); 129.3, 128.6, 128.5, 127.6, 127.3, 127.0, 126.7,
125.5 (ortho- and meta-arylC); 83.7, 82.1 (CO); 58.3, 53.7
(NCH2); 46.6 (NMe); 37.3, 36.5 (CH2CO); MS(EI) m/z: 989
(M+, 40%), 254 (H2CNCH2CH2C(O)Ph2, 100%). Anal.
calcd. for C62H62N2O4Zr: C 75.19, H 6.31, N 2.83; found: C
75.51, H 6.37, N 2.78. Although X-ray analysis of this com-
plex indicated (vide infra) that 1 equivalent of toluene was
present in the crystals, grinding the product and exposure of
the resulting powder to vacuum for 18 h was sufficient to re-
move this solvent.
1
cally removed and examined by H NMR to monitor the
progress of the reaction. At the end of each run, the solvent
was removed under vacuum and the residue was
chromatographed on silica gel using 75:25 CH2Cl2–hexane
as eluant. The dinitro products eluted first along with any
unreacted aldehyde followed by the β-nitroalcohol.
1
PhCH(OH)CH2NO2: H NMR (CDCl3) δ: 7.37 (5H, m, PhH),
5.45 (1H, m, CH(OH)), 4.54 (2H, m, CH2NO2); 13C(1H) δ:
138.08 (quat-arylC), 129.05, 129.00, 125.93 (arylC), 81.21
(CH2NO2), 71.01 (CH(OH)). The 1H and 13C data agree with
that reported in the literature (10, 11, 12).
1
PhCH(CH2NO2)2: H NMR (CDCl3) δ: 8.09 (2H, d, PhH),
Zr{tert-BuN(CH2CH2C(O)Me2)2}2 (12): Colourless crystals.
Yield: 95%; 13C(1H) NMR (C6D6) δ: 76.7 (CO), 54.7
(NCMe3), 43.6 (NCH2), 42.6 (CH2CO), 31.0 (NCMe3), 28.8
(C(O)Me2). Neither a satisfactory elemental analysis nor a
mass spectrum could be obtained on this complex.
7.63 (1H, m, PhH), 7.35 (2H, m, PhH), 4.76 (4H, dd,
CH2NO2, J = 1.8, 7.0 Hz), 4.30 (1H, pentet, CH(CH2NO2)2,
J = 7.0 Hz); 13C(1H) δ: 133.88, 129.61, 128.49, 127.35
(arylC), 76.73 (CH2NO2), 41.77 (CH(CH2NO2)2). The 1H
and 13C NMR data are very similar to those reported in the
literature (13).
Zr{tert-BuN(CH2CH2C(O)Ph2)2}2 (13): Yellow crystals. Yield:
93%; mp 275–280°C; 13C(1H) NMR (C6D6) δ: 148.5 (quat-
arylC), 128.2 (arylC), 126.9 (arylC), 126.7 (arylC), 85.9
(CO), 55.9 (NCMe3), 45.4 (NCH2), 42.0 (CH2CO), 28.0
(NCMe3); MS(EI) m/z: 1072 (M+, 55%), 1057 (M+ – Me,
100%), 1015 (M+ – tert-Bu, 56%), 995 (M+ – Ph, 12%), 877
(M+ – CH2CH2CPh2, 15%). A satisfactory elemental analy-
sis was not obtained for this complex.
1
β-NaphCH(OH)CH2NO2: H NMR (CDCl3) δ: 7.85 (4H, m,
β-naphH), 7.49 (2H, m, β-naphH), 7.42 (1H, d, β-naphH),
5.60 (1H, br d, CH(OH)), 4.61 (2H, m, CH2NO2); 13C(1H) δ:
135.40, 133.42, 133.19 (quat-β-naphC), 129.03, 128.07,
127.80, 126.74, 126.71, 125.35, 123.20 (β-naphC), 81.19
(CH2NO2), 71.16 (CH(OH)). Exact Mass MS, anal. calcd.
for C12H11NO3: 217.0739; found: 217.0743 amu.
1
Zr{(S)-PhMeC(H)N(CH2CH2C(O)Me2)2}2 (14): Yellow
microcrystalline powder. Yield: 92%; 13C(1H) NMR (C6D6)
70°C δ: 144.1 (quat-arylC), 128.7 (ortho-arylC), 128.4
(meta-arylC), 127.2 (para-arylC), 75.4 (CO), 62.0 (PhCHN),
46.5 (NCH2), 40.5 (CH2CO), 32.1 (C(O)Me), 30.6
(C(O)Me), 22.7 (PhC(Me)N); MS(EI) m/z: 673 (M+, 100%),
658 (M+ – Me, 82%). Anal. calcd. for C36H58N2O4Zr: C
64.15, H 8.67, N 4.15; found: C 64.28, H 8.31, N 4.13.
β-NaphCH(CH2NO2)2: H NMR (CDCl3) δ: 7.78–8.00
(4H, m, β-naphH), 7.49–7.68 (2H, m, β-naphH), 7.36 (1H, d,
3
β-naphH), 4.85 (4H, d, CH2NO2, JHH = 7.2 Hz), 4.48 (1H,
3
pentet, CH(CH2NO2)2, JHH = 7.2 Hz); 13C(1H) δ: 133.31,
133.24, 131.37 (quat-β-naphC), 129.73, 127.92, 127.77,
127.04, 126.98, 124.22 (β-naphC), 76.73 (CH2NO2), 41.92
(CH(CH2NO2)2). Exact Mass MS, anal. calcd. for
C13H12N2O4: 260.0797; found: 260.0797 amu.
Geraniol oxidation
Lewis acid catalysis
Geraniol (0.309 g, 2.0 mmol) was dissolved in 10 mL dry,
distilled CH2Cl2 and a solution of 14 (0.050 g, 3.7 mol%) in
5 mL CH2Cl2 was added under argon in a Kontes flask. Af-
ter stirring 10 min, 0.5 mL of a 5 M solution of tert-butyl
hydroperoxide in decane was added via syringe (2.5 mmol).
The flask was sealed and the reaction mixture stirred at
room. Aliquots (1 mL) were withdrawn daily and examined
Hetero Diels–Alder reaction
Ethyl vinyl ether was dried for several days over molecu-
lar sieves and then vacuum transferred (5 mL) to a Schlenk
flask fitted with a Kontes valve. Under an argon atmosphere,
0.2 mL trans-crotonaldehyde was added to the ethyl vinyl
ether solution followed by a solution of 14 (5 mol%) in
0.5 mL toluene. The flask was sealed and the reaction mix-
ture was stirred at room temperature for 4 days. At the end
of this period, the reaction vessel was opened to air and the
solvent was removed by rotary evaporation (water aspirator,
1
by H NMR to monitor the progress of the reaction. The re-
action appeared to reach a plateau at 65% completion after 3
days so an additional 2.5 mL tert-butyl hydroperoxide was
added. The reaction was stopped after 6 days at ca. 85%
completion since no further conversion was evident. The re-
action mixture was extracted with 10 mL water and the
1
35°C). Examination of the residue by H NMR revealed the
presence of trace cis-2-ethoxy-4-methyl-3,4-dihydro-2-pyran
© 2000 NRC Canada

Shao et al.
259
CH2Cl2 phase was taken to dryness. Chromatography on
silica gel using 50:50 Et2O–hexane as eluant removed all
traces of free ligand from the product but did not separate
the geraniol from its epoxide. Geraniol epoxide was identi-
Table 3. Crystallographic data for 11.
Formula
Fw
Cryst syst
Space group
a (Å)
b (Å)
c (Å)
α (°)
C69H70N2O4Zr
1082.54
triclinic
P1 (No. 2)
15.773(4)
10.889(2)
19.817(6)
89.91(2)
112.56(2)
87.78(2)
3140.2(14)
2
1.145
2.21
Mo Kα, 0.7107
ambient
44.9
1
fied by NMR (14): H (CDCl3) δ: 5.00 (1H, m, =CH), 3.75–
3.80 and 3.60–3.65 (2H, m, CH2OH), 2.88 (1H, m, CH(O)),
1.98–2.08 (2H, m, =CHCH2), 1.63 and 1.56 (3H, s and 3H,
=CMe2), 1.39–1.49 (2H, m, CH2C(O)), 1.28 (3H, s,
CH2CMe(O)); 13C(1H) δ 132.10 (=CMe2), 123.31 (=CHCH2),
63.13 (CH(O)), 61.33 (CH2OH), 59.27 (CH2CMe(O)), 38.48
(=CHCH2), 25.64 (CH2CMe(O)), 23.64 (CH2CH(O)), 17.61,
and 16.71 (=CMe2).
β (°)
γ (deg)
V (Å3)
Z
Epoxide cleavage with azidotrimethylsilane
ρ (calcd.) (g cm–3)
µ (cm–1)
Dry, distilled cyclohexene oxide (0.1 mL, 1 mmol) was
added to a solution of 14 (0.037g, 5 mol%) dissolved in
5 mL toluene under argon in a Kontes flask. Neat
azidotrimethylsilane (Aldrich, 0.13 mL, 1.0 mmol) was
added to the stirred solution via syringe and the vessel was
sealed. The reaction was allowed to continue until no
cyclohexene oxide remained (1H NMR, 4 days). Removal of
the solvent under reduced pressure afforded a colourless,
viscous oil which possessed an extremely complex NMR
spectrum. This material was shown to contain cyclohexanol,
cyclohexanone, and at least eight other compounds by GC
analysis.
Radiation, λ (C)
T
2θmax (deg)
No. of observed reflections
No. of unique reflections
Ra, Rw
8190
6568
0.099, 0.169
b
aR = Σ(|Fo|-|Fc|)/Σ|Fo|.
bRw = [Σw(|Fo|-|Fc|2)/Σw(|Fo|)2]1/2
.
surements were collected over one half of the sphere. After
the usual data reduction procedures, including an absorption
correction (max and min transmission: 0.9939, 0.8507) ac-
cording to a measured PSI scan, the structure was solved by
Patterson methods.2 The refinements minimised Σw((|Fo| –
|Fc|)2) and proceeded normally using teXsan98.3 Criteria for
inclusion of reflections were I > 3.0σ(I). The weighting
scheme was determined by counting statistics using w =
1/(σ2(F) + 0.001F2). Convergence was satisfactory: max
shift/esd <0.01; the largest positive and negative peaks in
the final difference map were 3.26 and –0.86 e/Å3, respec-
tively. A total of 630 parameters were refined; the toluene of
solvation was refined as a rigid group. Hydrogen atoms were
placed in calculated positions. No intermolecular contacts
shorter than 3.5 Å were observed. The structural perspective
plots were drawn using ZORTEP.4 Complete tables of bond
distances, angles, and anisotropic temperature parameters
have been deposited as supplementary material.5
Determination of enantiomeric excess
Enantiomeric excesses were determined using Yb(hfc)3
(hfc = 3-(heptafluoropropylhydroxymethylene)-(+)-camphorate),
Aldrich). For the β-nitroalcohols, splitting of the resonances
due to the α-hydrogens CH(OH)) was monitored as a func-
tion of lanthanide shift reagent (LSR) concentration. The
ee’s reported for these products are very crude estimates
since the α-hydrogen resonances of the two enantiomers
were only partially resolved. In the case of geraniol oxida-
tion, the resonances due to the methyl groups of the two
enantiomeric epoxides were fairly well resolved and better
estimates of the ee was obtainable in this case (See Fig. S2
of the supporting information).
X-ray crystallographic studies
Crystallographic data for 11 is summarized in Table 3.
Crystals of 11 (0.4 × 0.4 × 0.4 mm) were loaded into glass
capillaries in the glove box and subsequently flame sealed.
Results and discussion
The crystals were transferred to
a Nonius CAD4F
diffractometer equipped with Mo Kα radiation. The unit cell
of 11 was refined using 24 reflections in the 2θ range of
30–35°. An experimental density was not obtained because
of the air sensitivity of the compound. Five standard reflec-
tions, measured periodically during data collection (4, 2,
–12; 2, –6, 2; 3, 1, 11; 7, 0, 0; 0, 7, 0), showed less than a
3% decline in intensity during data collection. Intensity mea-
The aminodiol ligands 4–8 were prepared in good yield
by the two step procedure shown in Scheme 1. As might be
expected, the rate of Michael addition to methyl acrylate is
strongly dependent on the steric bulk of the amine. Thus,
tert-butylamine fails to give complete conversion to the
aminodiester even after 1 week at methanol reflux. Never-
theless, pure aminodiester 2 is easily obtained by fractional
2 P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, R. De Gelder, R. Israel, and J.M.M. Smits. DIRDIF94: The DIRDIF94 Program
System. Technical Report of the Crystallography Laboratory, University of Nijmegen, The Netherlands. 1994.
3 Molecular Structure Corporation. teXsan for Windows Version 1.05. The Woodlands, Tex. 1998.
4 L. Zsolnai. ZORTEP. University of Heidelberg, Germany. 1995.
5 Complete atom labelling scheme, atomic positional parameters and Beq, intramolecular distances and angles involving the nonhydrogen at-
oms, and anisotropic displacement parameters for 11; as well as geraniol oxidation catalyzed by 14: methyl region of the 1H NMR with
added LSR (Yb(hfc)3), have been deposited as supplementary material and may be purchased from: The Depository of Unpublished Data,
Document Delivery, CISTI, National Research Council Canada, Ottawa, Ontario, Canada, K1A 0S2.
© 2000 NRC Canada

260
Can. J. Chem. Vol. 78, 2000
Scheme 1.
Scheme 2.
[1]
2 H2L + Zr(CH2Ph)4 → ZrL2 + 4 PhCH3
H2L = 4–8
10–14
Molecular ions showing the expected isotopic distribution
were observed for 10, 11, 13, and 14. Despite the observa-
tion of molecular ions by mass spectroscopy, dimeric and
oligomeric solution structures are very common for zirco-
nium alkoxides (15). However, cryoscopic molecular weight
measurements on 10–14 indicated that these complexes are
also monomeric in solution. Thus it appears that two
aminodiolate ligands are sufficient to saturate the zirconium
coordination sphere.
distillation of the reaction mixture. Treatment of
aminodiesters 1–3 with a large excess of methyl or phenyl
lithium afforded the aminodiols 4–8 after aqueous work up.
In the case of 8, a secondary amine byproduct (9) was also
observed. Presumably, this product results from competing
deprotonation of the protons α to the ester carbonyl by
methyl lithium (Scheme 2). Attempts to prepare the R′ = Ph
analog of 8 failed, possibly due to even more extensive α-
deprotonation with phenyl lithium. The aminodiols bearing
methyl alkoxide substituents were obtained as oils (4, 8) or
low melting solids (6) while those bearing phenyl substitu-
ents (5, 7) were isolated as colourless crystals.
Addition of two equivalents of aminodiol 4–8 to tetra-
benzylzirconium resulted in formation of the corresponding
bis(ligand) complexes 10–14 (eq. 1). The complexes bearing
methyl alkoxide substituents (10, 12, and 14) were extremely
soluble in hexane and could not be purified by recrystalli-
zation. However, both 10 and 14 were isolated as analyti-
cally pure solids after removal of solvent. The phenyl
substituted complexes, 11 and 13, could be recrystallized
from toluene–hexane mixtures.
The monomeric solution structures of 10–14 greatly facili-
tate structural studies by NMR spectroscopy. Proton NMR
evidence points to a cis, fac-octahedral (C2) geometry
(Fig. 1(a)) for both 10 and 11 since eight resonances are ob-
served in the room temperature spectrum for the backbone
CH2 protons (Table 2). A trans, fac geometry (Fig. 1(b)) can
be excluded since it would have C2h symmetry and must
therefore exhibit only four CH2 resonances. Although a
trans, mer ligand arrangement is also theoretically possible,
the geometry of the aminodiolate ligand itself appears pre-
disposed to form a fac donor arrangement. The cis, fac struc-
ture was confirmed in the solid state for 11 by an X-ray
crystallographic study (vide infra).
1
The H NMR resonances of 10 are not as sharp as those
of 11 at room temperature suggesting that a dynamic process
may be operative for the former. Indeed, when a NMR sam-
ple of 10 is heated, further broadening of the backbone
© 2000 NRC Canada

Shao et al.
261
Fig. 3. ZORTEP plot (30% probability) of 11 (toluene of solva-
Fig. 1. Possible structures for zirconium bis(aminodiolate) com-
tion omitted).
plexes.
Fig. 2. A possible fluxional process for 10.
been proposed to account for the equivalence of the chelate
arm resonances in the trigonal bipyramidal complex
Zr{(Me3Si)N(CH2CH2N(SiMe3))2}{CH2Ph}2 (19).
1
The H NMR resonances of 11 also broaden on heating
but in this case coalescence is not reached even at 110°C.
The much higher barrier to exchange for 11 deserves some
comment. Of the two processes involved, amine dissociation
will be enhanced by steric congestion at the metal centre
while nitrogen inversion may be hindered if it requires bulky
substituents to pass by one another. Since the alkyl
substituent at nitrogen is the same (methyl) in 10 and 11, it
seems reasonable to assume that amine dissociation will be
favoured in 11 due to the phenyl substituents at the alkoxide
carbon creating a more crowded coordination environment.
Therefore the observation of a higher barrier to exchange
can reasonably be ascribed to the second process, inversion
at nitrogen. It is possible that the phenyl groups at the
alkoxide carbon hinder the necessary rearrangement of the
ligand required during inversion since they must pass by one
another in order for the amine to recoordinate to the metal
centre.
resonances occurs leading to coalescence at 42°C. At 80°C,
a simple spectrum is obtained which shows a single reso-
nance integrating to eight protons for each chemically dis-
tinct CH2 (Table 2). From the coalescence temperature, the
free energy of activation for this process is calculated to be
∆G‡ = 66.7 ± 0.5 kJ mol–1.6 A process that accounts for this
behaviour is shown in Fig. 2. Amine dissociation to produce
a five coordinate complex, followed by inversion at ni-
trogen and re-coordination exchanges all backbone CH2
resonances of a given type. It should be noted that nitrogen
inversion is required to exchange inequivalent protons on
each backbone carbon; simple pseudorotation of the five co-
ordinate intermediate can be ruled out since this will only
exchange protons on different arms. Inversion at nitrogen
is normally a low energy process (17) and dissociation of a
single amine is also expected to require relatively little en-
ergy (18), so the low activation energy required to exchange
the backbone protons is not surprising. A similar process has
The simplicity of the NMR spectra for 12–14 (Table 2)
indicates that these complexes are already undergoing a
rapid fluxional process at room temperature. The presence
of bulky groups at nitrogen (12 and 13, tert-butyl; 14,
α-methylbenzyl) likely leads to weaker amine binding which
should enhance the rate of exchange. In fact, the observation
6 The free energy of activation was calculated from the coalescence temperature (Tc) using the equal population, two site exchange equation:
∆G‡ = (1.912 × 10–2)(Tc)[9.972 + log(Tc/δν )] in kJ mol–1 where δν is the separation of the resonances in Hz at coalescence and Tc is in K
(16). The error in ∆G‡ was estimated to be ±0.5 kJ mol–1 assuming a liberal error of ±2°C in estimation of Tc.
© 2000 NRC Canada

262
Can. J. Chem. Vol. 78, 2000
Table 4. Selected bond distances and angles for 11.a
Distances (Å)
Zr(1)—O(1)
Zr(1)—O(2)
Zr(1)—O(3)
Zr(1)—O(4)
Zr(1)—N(1)
1.996(5)
1.971(4)
1.976(5)
1.961(4)
2.545(5)
Zr(1)—N(2)
O(1)—C(1)
O(2)—C(6)
O(3)—C(32)
O(4)—C(37)
2.562(5)
1.399(9)
1.412(8)
1.415(8)
1.406(7)
Angles (°)
O(1)-Zr(1)-O(2)
O(1)-Zr(1)-O(3)
O(1)-Zr(1)-O(4)
O(1)-Zr(1)-N(1)
O(1)-Zr(1)-N(2)
O(2)-Zr(1)-O(3)
O(2)-Zr(1)-O(4)
O(2)-Zr(1)-N(1)
O(2)-Zr(1)-N(2)
O(3)-Zr(1)-O(4)
101.9(2)
147.0(2)
100.0(2)
78.2(2)
82.8(2)
100.9(2)
92.7(2)
79.6(2)
171.5(2)
102.4(2)
O(3)-Zr(1)-N(1)
O(3)-Zr(1)-N(2)
O(4)-Zr(1)-N(1)
O(4)-Zr(1)-N(2)
N(1)-Zr(1)-N(2)
Z(1)-O(1)-C(1)
Zr(1)-O(2)-C(6)
Zr(1)-O(3)-C(32)
Zr(1)-O(4)-C(37)
82.8(2)
77.9(2)
171.5(2)
79.5(2)
108.4(2)
152.3(4)
145.8(4)
153.5(4)
147.6(4)
aEstimated standard deviations are given in parentheses.
1
that the H NMR spectra for 12 and 14 remain sharp to
–80°C may indicate that these complexes exist as ground
state five coordinate structures similar to that shown in
Fig. 2. At present, there is no evidence to distinguish this
possibility from a fluxional six coordinate structure such as
those shown in Fig. 1. The observation of fluxional behav-
iour for 13 even at room temperature appears to contradict
the explanation given for the static behaviour of 11 at the
same temperature. However, it is noteworthy that 13 is con-
siderably more crowded than 11 and it is possible that ex-
change proceeds for 13 by dissociation of both amine
nitrogens to give a four coordinate intermediate. In any case,
further attempts to study the fluxional behaviour of this
complex were hindered by its rather low solubility.
coordination of the three pendant alkoxide groups. Since the
aminodiolate ligand in 11 possesses only two pendant
alkoxide groups, the amine nitrogen is not constrained to a
particular distance and can adopt a distance reflective of the
strength of the amine binding.
The zirconium-alkoxide oxygen distances in 11
(1.961(4)–1.996(5) Å) are very similar to those observed in
the related six coordinate complexes, Zr[O-2,6-tert-
Bu2C6H3]2[OCH(Me)NC5H3CH(Me)O][py] (Zr-Oaryloxide
=
2.006(5), 2.015(6); Zr-Oalkoxide = 1.988(5), 2.034(5) Å) (22)
and Cp2Zr(m-OCH2(C6H4)CH2O)ZrCp2 (1.946(6) Å) (1b).
The Zr-O-C angles in 11 (145.8(4)°–153.5(4)°) are similar to
those in the alkoxide-bridged zirconocene above (155.1(5)°).
Interestingly, the Zr-O-C bond angles for the tridentate ligand
in Zr[O-2,6-tert-Bu2C6H3]2[OCH(Me)NC5H3CH(Me)O][py]
are only 126.7(6)° and 129.5(6)° while the Zr-O-Caryl angles
in the same complex are nearly linear (172.8(5)°, 178.8(5)°).
Presumably, the intermediate angles in 11 do not reflect any
significant angular strain.
The solid state structure of 11 was investigated by X-ray
crystallography. The structure is shown in Fig. 3 and signifi-
cant bond lengths and angles are collected in Table 4. In
agreement with the solution structure, the complex consists
of a distorted cis, fac-octahedral coordination environment.
The zirconium-amine nitrogen bond lengths in 11 (2.545(5),
2.562(5) Å) are shorter than might be predicted from
the
closely
related
five
coordinate
complexes
Zr{(Me3Si)N(CH2CH2N(SiMe3))2}{Cl}{CH(SiMe3)2} and
Zr{(Me3Si)N(CH2CH2N(SiMe3))2}{BH4}2 (2.770(5) and 2.561(4),
respectively) (20).7 This is not surprising because the steric
and electronic effects of the trimethylsilyl substituent at ni-
trogen combine to make this a much less basic amine than
the methyl substituted amine in 11. Obviously, the presence
of a bulky alkyl group contributes to the very long Zr—N
distances in Zr{(Me3Si)N(CH2CH2N(SiMe3))2}{Cl}{CH(SiMe3)2}.
In comparison, the titanium-amine nitrogen bond in
{Ti[NMe2][N(CH2CH2O)3]}2 is only 2.270(2) Å, much shorter
than that observed in 11, even allowing for the difference in
metal ionic radii (ca. 0.12 Å in six coordination) (6a, 21).
However, the apical nitrogen of an atrane ligand is forced
into close proximity with the metal in order to accommodate
7 Typically, five–coordinate zirconium complexes display bond lengths 0.06 C shorter than their six-coordinate analogues (24). The predicted
Zr—N bond lengths for 11 based on these two complexes would therefore be 2.83 and 2.62 Å.
© 2000 NRC Canada

Shao et al.
263
Table 5. Lewis acid catalyst screen for 14.
Reactiona
Yield (%)b
ee (%)c
Hetero-Diels Alder (eq. 2)
Henry (eq. 3) Ar = Ph
Ar = β-Naph
Geraniol oxidation (eq. 4)
Epoxide cleavage (eq. 5)
5d
—
25
30
46
—
70e
75f
80g
0
a25°C.
bIsolated yield.
cDetermined using Yb(hfc)3.
d96 h, ethyl viny ether.
e96 h, CH2NO2 solvent, 8% PhCH(CH2NO2)2.
f70 h, CH2NO2 solvent, 15% β-NaphCH(CH2NO2)2.
g7 days, CH2Cl2.
amounts of free ligand were observed in the reaction mix-
1
ture by H NMR after one day.
Attempts to carry out a hetero Diels–Alder reaction
(eq. 2) or the addition of azidotrimethylsilane to cyclohexene
oxide (7a) (eq. 4) using 14 as a catalyst failed. The hetero
Diels–Alder reaction is particularly challenging because it
requires a fine balance of Lewis acidity: catalysts that are
too acidic (eq. Yb(OTf)3) result in ethyl vinyl ether
(dienophile) polymerization, while those that are too mild do
not effect the reaction (23). Less than a 5% yield of the de-
sired dihydropyran product was obtained after 4 days at
room temperature using 14 as catalyst. No polymerization of
ethyl vinyl ether was observed indicating that 14 is indeed a
mild Lewis acid. This result is not surprising because the
alkoxide groups are good donors and relatively low zirco-
nium Lewis acidity is to be expected. In contrast, reaction of
azidotrimethylsilane with cyclohexene oxide in the presence
of 14 resulted in the slow consumption of the starting mate-
rials and formation of a complex mixture of products. GC
analysis showed that at least 10 compounds were formed in
significant amounts during this reaction. The reason for the
lack of selectivity is not clear.
In summary, the aminodiolate ligands developed here
form simple, monomeric zirconium complexes of good hy-
drocarbon solubility. The tendency towards amine dissocia-
tion and fluxional behaviour is dependent on the nature of
the substituents at nitrogen and the alkoxide carbon. The
most stable six coordinate geometry is obtained with small
nitrogen substituents (e.g., methyl) and large alkoxide sub-
stituents (e.g., phenyl). The chiral complex 14 is a mild
Lewis acid catalyst for the nitroaldol reaction and the oxida-
tion of geraniol with low to moderate stereocontrol.
Verkade rationalized the cis arrangement of the amine and
amide groups in III in terms of electronic properties of the
ligands: strongly electron donating amide and amine groups
preferring to occupy sites trans to the more electron with-
drawing alkoxide ligands and cis with respect to one another
(6). It is possible similar electronic factors are responsible
for the cis disposition of the amine donors in 11 although
steric effects cannot be ruled out.
Chiral complex 14 was also briefly screened for Lewis
acid catalytic activity using four typical test reactions (eq. 2–5).
The complex proved to be an effective catalyst for the
nitroaldol (Henry) reaction (eq. 3)8 and for the oxidation of
geraniol with tert-butyl hydroperoxide (eq. 5). The nitroaldol
reaction proceeded smoothly to give the β-nitroalcohols in
good yield but low enantiomeric excess (Table 5). At longer
reaction times (>48 h), in nitromethane as solvent, increas-
ing amounts of the dinitro products ArCH(CH2NO2)2 were
obtained.
The oxidation of geraniol slowly proceeded over a period
of 4 days and it was necessary to add excess tert-butyl
hydroperoxide to achieve good yields. Nevertheless, the re-
action does not proceed in the absence of 14 and the product
is formed with very high selectivity (no other products de-
rived from geraniol were detected at any stage of the reac-
tion). That even moderate enantioselectivity is obtained
(46% ee, Table 5) is somewhat surprising because ligand
loss might be expected in the presence of a large excess of
geraniol and tert-butyl hydroperoxide. Indeed, small
Acknowledgements
We thank Mrs. C. Greenwood for assistance in recording
the NMR spectra and Dr. David McGillivray for obtaining
the mass spectra. This research was supported by the
NSERC (Canada) and a University of Victoria Internal Re-
search Grant (to D.J.B.)
8 Strictly speaking, the nitroaldol reaction is not an example of simple Lewis acid catalysis because the alkoxide groups must act as bases in
the deprotonation of nitromethane. However, it is likely that coordination of the aldehyde also takes place and that this plays some role in
rate acceleration.
© 2000 NRC Canada

264
Can. J. Chem. Vol. 78, 2000
10. P.B. Kisanga and J.G. Verkade. J. Org. Chem. 64, 4298 (1999).
11. R. Ballini and G. Bosica. J. Org. Chem. 62, 425 (1997).
12. S. Kiyooka, T. Tsutsui, H. Maeda, Y. Kaneko, and K. Isobe.
Tetrahedron Lett. 36, 6531 (1995).
13. M.D. Alcántara, F.C. Escribano, A. Gómez-Sánchez, M.J.
Diánez, M.D. Estrada, A. López-Castro, and S. Pérez-Garrido.
Synthesis, 64 (1996).
14. H. Lütjens, G. Wahl, F. Möller, P. Knochel, and J.
Sundermeyer. Organometallics, 16, 5869 (1997).
15. D.C. Bradley, R.C. Mehrotra, and D.P. Gaur. Metal alkoxides.
Academic Press, London. 1978.
16. J. Sandstrom. Dynamic NMR spectroscopy. Pergamon, Lon-
don. 1982.
References
1. See for example: (a) T.V. Lubben, P.T. Wolczanski, and G.D.
Van Duyne. Organometallics, 3, 977 (1984); (b) D.W. Stephan.
Organometallics, 9, 2718 (1990).
2. (a) S.L. Latesky, A.K. McMullen, G.P. Niccolai, and I.P. Roth-
well. Organometallics, 4, 902 (1985); (b) P.E. Fanwick, A.E.
Ogilvy, and I.P. Rothwell. Organometallics, 6, 73 (1987);
(c) L.M. Kobriger, A.K. McMullen, P.E. Fanwick, and I.P.
Rothwell. Polyhedron, 8, 77 (1989).
3. (a) P.T. Wolczanski. Polyhedron, 14, 3335 (1995); (b) R.E.
LaPointe, P.T. Wolczanski, and J.F. Mitchell. J. Am. Chem.
Soc. 108, 6582 (1986).
4. A. van der Linden, C.J. Schaverien, N. Miejboom, C. Ganter,
and A.G. Orpen. J. Am. Chem. Soc. 117, 3008 (1995).
5. (a) M.T. Reetz, S.H. Kyung, C. Bolm, and T. Zierke. Chem.
Ind. (London), 824 (1986); (b) D. Seebach, A.K. Beck, R.
Imwinkelreid, S. Roggo, and A. Wonnacott. Helv. Chim. Acta,
70, 954 (1987); (c) K. Mikami, Y. Motoyama, and M. Terada.
J. Am. Chem. Soc. 116, 2812 (1994).
6. (a) A.A. Naiini, W.M.P.B. Menge, and J.G. Verkade. Inorg.
Chem. 30, 5009 (1991); (b) A.A. Naiini, S.L. Ringrose, Y. Su,
R.A. Jacobson, and J.G. Verkade. Inorg. Chem. 32, 1290 (1993).
7. (a) W.A. Nugent. J. Am. Chem. Soc. 114, 2768 (1992);
(b) W.A. Nugent and R.L. Harlow. J. Am. Chem. Soc. 116,
6142 (1994); (c) M. Bonchio, S. Calloni, F. Di Furia, G.
Licini, G. Modena, S. Moro, and W.A. Nugent. J. Am. Chem.
Soc. 119, 6935 (1997).
17. J. March. Advanced organic chemistry. 3rd ed. Wiley and
Sons, New York. 1985. p. 86.
18. (a) P. Shao, D.J. Berg, and G. Bushnell. Inorg. Chem. 33, 3452
(1994); (b) L. DeCola, D.L. Smailes, and L.M. Vallarino.
Inorg. Chem. 25, 1729 (1986), and refs. therein.
19. (a) H.C.S. Clark, F.G.N. Cloke, P.B. Hitchcock, J.B. Love, and
A.P. Wainwright. J. Organomet. Chem. 501, 333 (1995);
(b) A.D. Horton, J. de With, A.J. van der Linden, and H. van
de Weg. Organometallics, 15, 2672 (1996).
20. F.G.N. Cloke, P.B. Hitchcock, and J.B. Love. J. Chem. Soc.
Dalton Trans. 1995, 25.
21. R.D. Shannon. Acta Crystallogr. Sect. A: Cryst Phys. Diffr.
Theor. Gen. Crystallogr. A32, 751 (1976).
22. C.H. Zambrano, A.K. McMullen, L.M. Kobriger, P.E. Fanwick,
and I.P. Rothwell. J. Am. Chem. Soc. 112, 6565 (1990).
23. C. Spino, L. Clouston, and D.J. Berg. Can. J. Chem. 75, 1047
(1997).
8. U. Zucchini, E. Albizatti, and U. Giannini. J. Organomet.
Chem. 26, 357 (1971).
9. W.G. Dauben and H.O. Krabbenhoft. J. Org. Chem. 42, 282
(1977).
© 2000 NRC Canada