Linear trimer analogues of calixarene as chiral coordinating ligands: X-ray crystallographic and NMR spectroscopic characterization of chiral and achiral trisphenolates complexed to titanium(IV) and aluminum(III)

Article abstract of DOI:10.1021/ic010685j

Achiral and chiral linear trisphenol analogues of calixarene (HOArCH₂Ar′(OH)C(R)HArOH, Ar = 4,6-di-tert-butylphenyl; Ar′ = 4-tert-butylphenyl; R = H (achiral), Me (chiral)) were prepared in anticipation of their adoption of a chiral conformation upon coordination to Lewis acidic metal centers. The trisphenols react with 1 equiv of Ti(OR′)4 (R′ = i-Pr or t-Bu) to yield complexes with molecular formula Ti₂(OArCH₂Ar′(O)C(R)HArO)₂ (OR′)₂ (R = H, Me; R′ = i-Pr or t-Bu). An X-ray crystal structure of the titanium complex of the achiral trisphenol (R = H; R′ = t-Bu) reveals that the trisphenolate ligand adopts an unsymmetrical (and therefore chiral) conformation, with η²-coordination to one metal center and η¹-coordination to the second metal center. The chiral trisphenol, which contains a stereogenic center (indicated as C in the shorthand notation used above), coordinates titanium in an analogous fashion to produce only one diastereomer (out of four possible); therefore, the configuration of the stereogenic center controls the conformation adopted by the bound ligand. The reaction of achiral trisphenol with AlMe₃ produces a compound with molecular formula Al₂(OArCH₂Ar′(O)CH₂ArO)₂. ¹H NMR spectroscopy and X-ray crystallography reveal that the trisphenolate ligand adopts an asymmetric, C₂ conformation in this complex, where the central phenolate oxygen bridges the aluminum centers and the terminal phenolate oxygens each coordinate a separate aluminum center. Because these trisphenolate ligands adopt chiral conformations when coordinated to metal centers, they may be useful for developing diastereo- or enantioselective catalysts and reagents.

Full text of DOI:10.1021/ic010685j

Inorg. Chem. 2002, 41, 3656−3667
Linear Trimer Analogues of Calixarene as Chiral Coordinating Ligands:
X-ray Crystallographic and NMR Spectroscopic Characterization of
Chiral and Achiral Trisphenolates Complexed to Titanium(IV) and
Aluminum(III)
William O. Appiah,† Angela D. DeGreeff,† Gina L. Razidlo,† Sarah J. Spessard,† Maren Pink,‡
,†
Victor G. Young, Jr.,‡ and Gretchen E. Hofmeister*
Department of Chemistry, GustaVus Adolphus College, 800 West College AVenue,
St. Peter, Minnesota 56082, and X-Ray Crystallographic Laboratory, Chemistry Department,
The UniVersity of Minnesota, 207 Pleasant Street, S.E., Minneapolis, Minnesota 55455
Received June 29, 2001
Achiral and chiral linear trisphenol analogues of calixarene (HOArCH Ar′(OH)C(R)HArOH, Ar ) 4,6-di-tert-butylphenyl;
2
Ar′ ) 4-tert-butylphenyl; R ) H (achiral), Me (chiral)) were prepared in anticipation of their adoption of a chiral
conformation upon coordination to Lewis acidic metal centers. The trisphenols react with 1 equiv of Ti(OR′)4 (R′
) i-Pr or t-Bu) to yield complexes with molecular formula Ti2(OArCH Ar′(O)C(R)HArO)2(OR′)2 (R ) H, Me; R′ )
2
i-Pr or t-Bu). An X-ray crystal structure of the titanium complex of the achiral trisphenol (R ) H; R′ ) t-Bu) reveals
that the trisphenolate ligand adopts an unsymmetrical (and therefore chiral) conformation, with η2-coordination to
one metal center and η1-coordination to the second metal center. The chiral trisphenol, which contains a stereogenic
center (indicated as C in the shorthand notation used above), coordinates titanium in an analogous fashion to
produce only one diastereomer (out of four possible); therefore, the configuration of the stereogenic center controls
the conformation adopted by the bound ligand. The reaction of achiral trisphenol with AlMe3 produces a compound
1
with molecular formula Al2(OArCH Ar′(O)CH ArO)2. H NMR spectroscopy and X-ray crystallography reveal that
2
2
the trisphenolate ligand adopts an asymmetric, C conformation in this complex, where the central phenolate oxygen
2
bridges the aluminum centers and the terminal phenolate oxygens each coordinate a separate aluminum center.
Because these trisphenolate ligands adopt chiral conformations when coordinated to metal centers, they may be
useful for developing diastereo- or enantioselective catalysts and reagents.
One of the central challenges in chemistry is to understand
and control the selectivity of organic transformations, such
as carbon-carbon bond forming or polymerization reactions.
Much effort has focused on developing new alkoxide and
aryloxide ligands for titanium or aluminum, in order to
generate more selective reagents that can be used to promote
carbon-carbon bond forming reactions.1,2 Highly enantio-
selective reagents have been prepared from metal complexes
of tartrate,3 binaphthol,4 or phenolate-Schiff base5 derivatives.
Excellent chemo- and stereoselectivities using sterically
hindered aluminum phenolates for a variety of carbonyl
addition reactions have been demonstrated.2 In addition,
(1) Reviews of titanium-based reagents: (a) Duthaler, R. O.; Hafner, A.
Chem. ReV. 1992, 92, 807-832. (b) Narasaka, K. In Stereocontrolled
Organic Synthesis; Trost, B. M., Ed.; Blackwell: Oxford, U.K./Boston,
MA, 1994; pp 17-36. (c) Noyori, R. In Asymmetric Catalysis in
Organic Synthesis; John Wiley & Sons: New York, 1995; pp 212-
228. (d) Gregory, R. J. H. Chem. ReV. 1999, 99, 3649-3682. (e)
Mikami, K.; Terada, M. In Lewis Acid Reagents A Practical Approach;
Yamamoto, H., Ed.; Oxford University Press: New York, 1999; pp
93-136.
(2) Reviews of aluminum-based reagents: (a) Saito, S.; Yamamoto, H.
Chem. Commun. 1997, 1585-1592. (b) Maruoka, K. In Lewis Acid
Reagents A Practical Approach; Yamamoto, H., Ed.; Oxford Univer-
sity Press: New York, 1999; pp 5-29.
* Corresponding author. E-mail: ghofmeis@gustavus.edu. Tel: (507)
933-7039. Fax: (507) 933-7041.
† Gustavus Adolphus College.
‡ University of Minnesota.
3656 Inorganic Chemistry, Vol. 41, No. 14, 2002
10.1021/ic010685j CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/12/2002

Linear Calixarene Analogues as Chiral Coordinating Ligands
titanium and zirconium complexes of sterically congested
bidentate aryloxides are stereoselective catalysts for R-olefin
polymerization.6 These examples demonstrate that it is
important to develop new chiral ligands that are sterically
congested at the metal center, so that they both increase the
reactivity (by forming monomeric, unsaturated metal com-
plexes) and increase the selectivity (by controlling substrate
approach).
We have begun to investigate the synthesis and coordina-
tion properties of trisphenols (1a,b, Figure 1A), linear
analogues of the macrocyclic calixarenes,7 which we pre-
dicted would coordinate metal centers in an asymmetric
fashion to yield stereoselective reagents. Calixarene com-
plexes of metals such as titanium,8 zirconium,8c,9 and
aluminum10 possess cavities or pockets that are bordered by
(3) For representative examples, see: (a) Narasaka, K.; Iwasawa, N.; Inoue,
M.; Yamada, T.; Nakashima, M.; Sugimori, J. J. Am. Chem. Soc. 1989,
111, 5340-5345. (b) Hayashi, M.; Matsuda, T.; Oguni, N. J. Chem.
Soc., Perkin Trans. 1 1992, 3135-3140. (c) Narasaka, K.; Hayashi,
Y.; Shimadzu, H.; Nihata, S. J. Am. Chem. Soc. 1992, 114, 8869-
8885. (d) Seebach, D.; Plattner, D. A.; Beck, A. K.; Wang, Y. M.;
Hunziker, D.; Petter, W. HelV. Chim. Acta 1992, 75, 2171-2209. (e)
Ito, Y. N.; Ariza, X.; Beck, A. K.; Bohac, A.; Ganter, C.; Gawley, R.
E.; Kuehnle, F. N. M.; Tuleja, J.; Wang, Y. M.; Seebach, D. HelV.
Chim. Acta 1994, 77, 2071-2110. (f) Dahinden, R.; Beck, A. K.;
Seebach, D. In Encyclopedia of Reagents for Organic Synthesis;
Paquette, L. A., Ed.; Wiley: Chichester, U.K., 1995; Vol. 3, pp 2167-
2170. (g) Schafer, H.; Seebach, D. Tetrahedron 1995, 51, 2305-2324.
(h) Seebach, D.; Dahinden, R.; Marti, R. E.; Beck, A. K.; Plattner, D.
A.; Kuehnle, F. N. M. J. Org. Chem. 1995, 60, 1788-1799. (i)
Seebach, D.; Lyapkalo, I. M.; Dahinden, R. HelV. Chim. Acta 1999,
82, 1829-1842.
(4) For representative examples, see: (a) Noyori, R.; Tomino, I.; Tanimoto,
Y. J. Am. Chem. Soc. 1979, 101, 2129-3131. (b) Mikami, K.; Terada,
M.; Nakai, T. J. Am. Chem. Soc. 1990, 112, 3949-3954. (c) Keck,
G. E.; Krishnamurthy, D.; Chen, X. Tetrahedron Lett. 1994, 35, 8323-
8324. (d) Keck, G. E.; Krishnamurthy, D.; Chen, X. Tetrahedron Lett.
1994, 35, 8323-8324. (e) Maruoka, K.; Saito, S.; Yamamoto, H. J.
Am. Chem. Soc. 1995, 117, 1165-1166. (f) Matsukawa, S.; Mikami,
K. Tetrahedron: Asymmetry 1995, 6, 2571-2574. (g) Posner, G. H.;
Dai, H.; Bull, D. S.; Lee, J. K.; Eydoux, F.; Ishihara, Y.; Welsh, W.;
Pryor, N.; Petr, S. J. J. Org. Chem. 1996, 61, 671-676. (h) Mori, M.;
Imma, H.; Nakai, T. Tetrahedron Lett. 1997, 38, 6229-6232. (i)
Zhang, F.-Y.; Yip, C.-W.; Cao, R.; Chan, A. S. C. Tetrahedron:
Asymmetry 1997, 8, 585-589. (j) Hamashima, Y.; Sawada, D.; Kanai,
M.; Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 2641-2642. (k) Ishii,
A.; Soloshonok, V. A.; Mikami, K. J. Org. Chem. 2000, 65, 1597-
1599.
Figure 1. (A) Structures of linear trisphenols. (B) Proposed asymmetric
coordination of titanium(IV) by trisphenol ligand (X ) O-t-Bu, O-i-Pr). If
R ) H, isomers i and ii are enantiomers; if R ) CH3, i and ii are
diastereomers. In this case, i would be preferred, due to the steric interaction
of the methyl group with the titanium in ii.
phenol substituents and are capable of binding small
molecules or ligands. In particular, 4-tert-butylcalix[8]arene
wraps around two titanium or zirconium centers to form two
asymmetric cavities, and generates preferentially one dia-
stereomeric complex when chiral alcoholate ligands are
bound in the cavities (10:1 ratio of diastereomers).8d We
envisioned that trisphenols 1a,b would also bind metals in
an asymmetric fashion because a steric interaction between
the terminal ortho substituents would prevent the ligand from
adopting a symmetrical partial-basket conformation (Figure
1B). We also anticipated that steric congestion from the tert-
butyl groups would prevent oligomerization and subsequent
coordinative saturation at the metal center. Another feature
of the trisphenol that we wanted to exploit was its potential
for preparation in chiral form (e.g., 1b, Figure 1A). Chiral
trisphenols are promising ligands for preparing enantio- or
stereoselective Lewis acid reagents for carbonyl addition or
polymerization reactions. In addition, we expect that they
(5) For representative examples, see: (a) Hayashi, M.; Miyamoto, Y.;
Inoue, T.; Oguni, N. J. Org. Chem. 1993, 58, 1515-1522. (b) Carreira,
E. M.; Singer, R. A.; Lee, W. J. Am. Chem. Soc. 1994, 116, 8837-
8838. (c) Carreira, E. M.; Lee, W.; Singer, R. A. J. Am. Chem. Soc.
1995, 117, 3649-3650. (d) Jiang, Y.; Zhou, X.; Hu, W.; Li, Z.; Mi,
A. Tetrahedron: Asymmetry 1995, 6, 2915-2916. (e) Sigman, M.
S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120, 5315-5316. (f)
Belokon’, Y. N.; Caveda-Cepas, S.; Green, B.; Ikonnikov, N. S.;
Khrustalev, V. N.; Larichev, V. S.; Moscalenko, M. A.; North, M.;
Orizu, C.; Tararov, V. I.; Tasinazzo, M.; Timofeeva, G. I.; Yashkina,
L. V. J. Am. Chem. Soc. 1999, 121, 3968-3973.
(6) (a) van der Linden, A.; Schaverien, C. J.; Meijboom, N.; Ganter, C.;
Orpen, A. G. J. Am. Chem. Soc. 1995, 117, 3008-3021. (b) Fokken,
S.; Spaniol, T. P.; Kang, H.-C.; Massa, W.; Okuda, J. Organometallics
1996, 15, 5069-5072. (c) Okuda, J. Masoud, E. Macromol. Chem.
Phys. 1998, 199, 543-545. (d) Tshuva, E. Y.; Goldberg, I.; Kol, M.
J. Am. Chem. Soc. 2000, 122, 10706-10707. (e) Tshuva, E.; Goldberg,
I.; Kol, M.; Goldschmidt, Z. Chem. Commun. 2001, 2120-2121. (f)
Matsui, S.; Mitani, M.; Saito, J.; Tohi, Y.; Makio, H.; Matsukawa,
N.; Takagi, Y.; Tsuru, K.; Nitabaru, M.; Nakano, T.; Tanaka, H.;
Kashiwa, N.; Fujita, T. J. Am. Chem. Soc. 2001, 123, 6847-6856.
(7) The literature in this area is extensive. For leading references, see:
(a) Gutsche, C. D. Calixarenes; The Royal Society of Chemistry:
Cambridge, England, 1989. (b) Boehmer, V. Angew. Chem., Intl. Ed.
Engl. 1995, 34, 713-745. (c) Wieser, C.; Dieleman, C. B.; Matt, D.
Coord. Chem. ReV. 1997, 165, 93-161. (d) Gutsche, C. D. Calixarenes
ReVisited; Texas Christian University: Fort Worth, TX, 1998.
(8) (a) Olmstead, M. M.; Sigel, G.; Hope, H.; Xu, X.; Power, P. P. J.
Am. Chem. Soc. 1985, 107, 8087-8091. (b) Bott, S. G.; Coleman, A.
W.; Atwood, J. L. J. Chem. Soc., Chem. Commun. 1986, 610-611.
(c) Andreetti, G. D.; Calestani, G.; Ugozzoli, F. J. Inclusion Phenom.
1987, 5, 123-126. (d) Hofmeister, G. E.; Hahn, F. E.; Pedersen, S.
F. J. Am. Chem. Soc. 1989, 111, 2318-2319. (e) Clegg, W.; Elsegood,
M. R. J.; Teat, S. J.; Redshaw, C.; Gibson, V. C. J. Chem. Soc., Dalton
Trans. 1998, 3037-3039. (f) Zanotti-Gerosa, A.; Solari, E.; Giannini,
L.; Floriani, C.; Re, N.; Chiesi-Villa, A.; Rizzoli, C. Inorg. Chim.
Acta 1998, 270, 298-311. (g) Ozerov, O. V.; Ladipo, F. T.; Patrick,
B. O. J. Am. Chem. Soc. 1999, 121, 7941-7942. (h) Radius, U.;
Friedrich, A. Z. Anorg. Allg. Chem. 1999, 625, 2154-2159.
(9) Giannini, L.; Caselli, A.; Solari, E.; Floriani, C.; Chiesi-Villa, A.;
Rizzoli, C.; Re, N.; Sgamellotti, A. J. Am. Chem. Soc. 1997, 119,
9198-9210.
(10) (a) Bott, S. G.; Coleman, A. W.; Atwood, J. L. J. Inclusion Phenom.
1987, 5, 747-758. (b) Coleman, A. W.; Bott, S. G.; Atwood, J. L. J.
Inclusion Phenom. 1987, 5, 581-590. (c) Atwood, J. L.; Bott, S. G.;
Jones, C.; Raston, C. L. J. Chem. Soc., Chem. Commun. 1992, 1349-
1351. (d) Atwood, J. L.; Gardiner, M. G.; Jones, C.; Raston, C. L.;
Skelton, B. W.; White, A. H. Chem. Commun. 1996, 2487-2488. (e)
Gardiner, M. G.; Koutsantonis, G. A.; Lawrence, S. M.; Nichols, P.
J.; Raston, C. L. Chem. Commun. 1996, 2035-2036. (f) Smith, J.
M.; Bott, S. G. Chem. Commun. 1996, 377-378. (g) Gibson, V. C.;
Redshaw, C.; Clegg, W.; Elsegood, M. R. J. Polyhedron 1997, 16,
4385-4387.
Inorganic Chemistry, Vol. 41, No. 14, 2002 3657

Appiah et al.
will have broader applications as coordinating ligands for a
variety of other metals or chiral organic species.
Our immediate goals were to prepare linear trisphenols
(1a,b) and their complexes with titanium and aluminum, and
to determine the three-dimensional structure adopted by the
trisphenolate ligands when coordinated to these metals. It
was important to us to prepare complexes that were asym-
metric and coordinatively unsaturated at the metal center,
in order to exploit their reactivity in organic synthesis.
Furthermore, we wanted to determine if the conformation
of the chiral trisphenol (1b), upon coordinating a metal
center, was controlled by the configuration at the stereogenic
center. We postulated that the sterically larger methyl group
introduced at the benzylic position would preferentially orient
away from the metal center, thereby providing the means
for achieving this control (Figure 1B). In this paper, we
describe the synthesis of achiral and chiral trisphenols (1a,b
(as the racemate)) and their complexes with titanium(IV) and
aluminum(III). We show that the trisphenols adopt chiral
conformations upon coordinating titanium or aluminum, that
the titanium remains coordinatively unsaturated in complexes
with trisphenol, and that the conformation adopted by the
chiral trisphenol when coordinated to titanium is controlled
by the configuration at the stereogenic center, according to
our hypothesis.
by 3 or another species such as 4 to yield the tetrakisphenol
or higher polyphenols (eq 2). Alternative preparations of
trisphenol 1a12a and analogues of 1a12b have been reported;
our method is useful because it can be modified for preparing
chiral versions of this ligand (e.g., 1b, vide infra). Compound
3 has also been prepared previously via a two-step proce-
dure.13
Titanium Complexes of Achiral Trisphenol
Complexes of 1a with titanium were prepared by addition
of Ti(OR′)4 (R′ ) t-Bu, i-Pr) to 1 equiv of trisphenol 1a,
yielding Ti2(trisphenolate)2(OR′)2 (6a, 7a, R′ ) t-Bu, i-Pr,
eq 3). The titanium complexes appear to have identical
Synthesis of Achiral Trisphenol
Achiral trisphenol 1a is easily prepared in two steps from
commercially available starting materials, using a procedure
modified from the literature (eq 1).11 We found that use of
structures, with the exception of the R′ group, according to
1H NMR spectroscopy. Both titanium complexes form as a
mixture of two predominant isomers, one of which is favored
at room temperature in a solution of benzene, chloroform,
1
THF, or pentane (the major isomer). H NMR spectra of
these compounds indicate that the trisphenolate ligand is in
an unsymmetrical conformation because the benzylic hy-
drogens resonate as four separate doublets. Recrystallization
of compound 6a results in isomer mixtures of varying
composition, with the minor isomer present in higher
proportion than is observed in the solution equilibrium. We
believe that the isomers interconvert readily in solution, and
that the minor isomer crystallizes preferentially. When a pure
sample of the minor isomer of 6a (obtained after several
recrystallization steps) is dissolved in benzene-d6 and
monitored over time by 1H NMR spectroscopy, it eventually
converts to a mixture in which the major isomer is preferred
in a 10:1 ratio (see Figure 2). The major isomer of compound
7a can be precipitated with pentane from the crude reaction
product residue; however, we have been unsuccessful in
poly(4-vinylpyridine) (PVP) as a scavenger for the HBr
produced in the second step prevented the formation of minor
quantities of higher oligomers, thereby facilitating the
purification and improving the yield of 1a. Presumably, the
higher oligomers form via an acid-catalyzed retro-Friedel-
Crafts reaction that breaks apart the trisphenol in the opposite
fashion to generate a resonance-stabilized benzylic carbo-
cation on the central phenol (4, eq 2). This species could
then couple with another 4-tert-butyl phenol to generate the
unsymmetrical trisphenol (5), which could then be capped
(12) (a) Weber, D.; Habicher, W. D.; Nifantev, E. E.; Teleshev, A. T.;
Zhdanov, A. A.; Belsky, V. K. Phosphorus, Sulfur Silicon Relat. Elem.
1999, 149, 143-165 and references therein. (b) Gordon, B. W. F.;
Scott, M. J. Inorg. Chim. Acta 1999, 297, 206-216 and references
therein.
(11) (a) Bright, W. M.; Cammarata, P. J. Am. Chem. Soc. 1952, 74, 3690.
(b) Bo¨hmer, V.; Marschollek, F.; Zetta, L. J. Org. Chem. 1987, 52,
3200-3205.
(13) Sokolowski, A.; Mu¨ller, J.; Weyhermu¨ller, T.; Schnepf, R.; Hilde-
brandt, P.; Hildenbrand, K.; Bothe, E.; Wieghardt, K. J. Am. Chem.
Soc. 1997, 119, 8889-8900.
3658 Inorganic Chemistry, Vol. 41, No. 14, 2002

Linear Calixarene Analogues as Chiral Coordinating Ligands
Figure 2. 1H NMR spectra of 6a (C6D6). Bottom: Initial sample of pure minor isomer. Middle: Sample after 4 h at room temperature. Top: Sample after
24 h at room temperature (primarily major isomer). Labels are according to diagram, determined by two-dimensional NMR spectroscopy. Signals labeled
with an asterisk (/) are due to Ti2(trisphenoxide)2O, and those labeled with x are due to toluene.
obtaining single crystals suitable for X-ray diffraction of
either isomer of 7a by recrystallization. In contrast, we were
able to obtain single crystals of the minor isomer of 6a,
suitable for X-ray diffraction studies; therefore, we focused
on compound 6a for further study.
substantial π-donation from the oxygens to the metal,15 and
this appears to prevent a single trisphenolate from simulta-
neously coordinating one titanium center by all three
oxygens. The crystal structure makes clear why the four
benzylic hydrogens are magnetically inequivalent, since the
two methylene bridges are of two different types. The
“straddle” methylene group links rings 2 and 3 and spans
the two metal centers, and the “nonstraddle” methylene group
links phenolates 1 and 2, which are coordinated to the same
titanium center (Figure 3B). In addition, the two hydrogens
at each methylene carbon are inequivalent, since one
hydrogen points toward the metal center (the “downfield”
hydrogen, vide infra), and one hydrogen points toward the
aromatic hydrogens (the “upfield” hydrogen, vide infra,
Figure 3B).
A crystal structure of the minor isomer of 6a shows that
the complex is a dimer, with inversion symmetry relating
the trisphenolate ligands, the tert-butoxide ligands, and the
titanium centers with one another, respectively (Figure 3A;
crystal data and selected bond distances and angles are in
Tables 1 and 2). Two phenolate oxygens of the tridentate
ligand are coordinated to one titanium center, and the third
phenolate oxygen coordinates the second titanium center,
resulting in an overall asymmetric conformation adopted by
the ligand. Surprisingly, the central phenolate oxygen does
not bridge the two metal centers and the titanium center
remains tetracoordinate, with O-Ti-O bond angles ranging
from 107° to 112°. Thus, the electron deficient metal does
not gain electron density by expanding its coordination
number upon dimerization. This is unusual for titanium
alkoxide or aryloxide dimer structures, which commonly
have bridging oxygens that serve to increase the coordination
number at titanium.14 However, the Ti-O-C(phenolate)
bond angles (143-154°) in the complex are indicative of
The solution structure of 6a has been investigated using
two-dimensional 1H NMR spectroscopy. Both isomers of 6a
have very similar 1H NMR spectra (C6D6), including features
such as an upfield-shifted aromatic proton resonance at either
6.5 (major isomer) or 6.6 (minor isomer) ppm, four benzylic
proton resonances from 3.3 to 5.4 ppm (vide supra), and six
tert-butyl proton resonances from 0.9 to 1.5 ppm (Figure
2). Two-dimensional COSY and NOESY experiments were
performed with the major isomer of 6a at room temperature,
enabling us to assign each of the resonances in the 1H NMR
spectrum. The COSY spectrum indicated which aromatic
resonances were paired in the same ring (cross peaks were
observed for the four-bond coupling between aromatic
(14) (a) Bradley, D. C.; Mehrotra, R. C.; Gaur, D. P. In Metal Alkoxides;
Academic Press: London, 1978; pp 74-104. For representative
examples, see: (b) Williams, I. D.; Pedersen, S. F.; Sharpless, K. B.;
Lippard, S. L. J. Am. Chem. Soc. 1984, 106, 6430-6431. (c) Boyle,
T. J.; Eilerts, N. W.; Heppert, J. A.; Takusagawa, F. Organometallics
1994, 13, 2218-2229. (d) Corey, E. J.; Letavic, M. A.; Noe, M. C.;
Sarshar, S. Tetrahedron Lett. 1994, 35, 7553-7556 (both bridging
and nonbridging titanium aryloxide structures are presented). (e) Damo,
S. M.; Lam, K.-C.; Rheingold, A.; Walters, M. A. Inorg. Chem. 2000,
39, 1635-1638.
(15) For comparison, Ti-O-C bond angles in a related titanium bis-
phenolate complex range from 135° to 151°: Chisholm, M. H.;
Huang, J.-H.; Huffman, J. C.; Streib, W. E.; Tiedtke, D. Polyhedron
1997, 16, 2941-2949.
Inorganic Chemistry, Vol. 41, No. 14, 2002 3659

Appiah et al.
Table 1. Summary of X-ray Crystallographic Data
[Ti(trisphenolate)(O-t-Bu)]2
(6a)
[Al(trisphenolate)]2
(8a)
a
b
formula
C116H184O15Ti2
C54H71AlO3
cryst color, habit
colorless block
colorless multifaceted
block
0.55 × 0.35 × 0.18
orthorhombic
Pbcn
cryst size, mm3
cryst syst
space group
unit cell
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
0.34 × 0.27 × 0.22
triclinic
P1h
14.688(2)
15.470(2)
15.872(2)
117.321(1)
108.001(2)
99.821(2)
2831.5 (4)
1
23.311(2)
20.250(2)
23.093(2)
90
90
90
10901 (2)
8
Z
fw
1914.43
1.123
0.200
1044
795.09
0.969
0.073
D
calcd, Mg/m3
µ, mm-1
F(000)
3456
diffractometer
Bruker SMART
Platform CCD
0.71073
Siemens SMART
Platform CCD
0.71073
λ(Mo KR
(graphite)), Å
T, K
173(2)
173(2)
θ range, deg
1.53-27.51
-18 e h e 18
-20 e k e 20
-20 e l e 19
25238
1.33-25.04
0 e h e 27
0 e k e 24
0 e l e 27
53253
index ranges
total no. of data
indep reflns
syst used
12507 (Rint ) 0.0548)c
SHELXTL-V5.0
a ) 0.1204,
b ) 0
9628 (R int ) 0.0440)c
SHELXTL-V5.0
a ) 0.1052,
b ) 2.7351
SADAB
weighting schemed
(a, b)
abs corrn
SADABS
(Sheldrick, 1999)
6526
(Sheldrick, 1996)
6828
no. of obsd data
(I > 2σ(I))
no. of restraints
no. of params
Re (I > 2σ(I))
wR2 f (I > 2σ(I))
GOF on F2 g
144
686
0.0705
0.1785
0.936
182
634
0.0623
0.1775
1.100
largest diff peak
0.671, -0.769
0.338, -0.255
and hole, e Å-3
Figure 3. (A) An ORTEP representation (50% probability) of the crystal
structure of the minor isomer of 6a. Carbons are unshaded. Hydrogen and
disordered tert-butyl carbon atoms are removed for clarity. Labeling is
according to the distances and angles listed in Table 2. The inversion center
is located between the two titanium centers. (B) Chem3D representation of
the crystal structure of the minor isomer of 6a with tert-butyl substituents
on the aromatic rings removed for clarity. Benzylic and some aromatic
hydrogens are shown in half of the structure. Labeling of the aromatic rings
(1-3) and hydrogens (D-M) is according to the discussion in the text.
See Figure 2 for NMR chemical shift assignments of these hydrogens.
Double-headed arrows connect hydrogens that show NOE cross peaks in
the NOESY spectrum.
a Includes 7 molecules of THF that crystallized with the compound.
c
b Includes 2 molecules of toluene that crystallized with the compound. Rint
2
2
2
) ∑|Fo
-
Fo |/∑|Fo |. d w ) [σ2(Fo ) + (AP)2 + (BP)]-1, where P )
2
2
2
2
(Fo + 2Fc )/3. e R ) ∑||Fo| - |Fc||/∑|Fo|. f wR2 ) [∑[w(Fo - Fc )2]/
2
2
2
∑[w(Fo )2]]1/2, where w ) q/σ2(Fo ) ) (aP)2 + bP. g GOF ) S ) ∑[w(Fo
2
- Fc )2]/(n - p)1/2
.
Table 2. Selected Bond Distances (Å) and Angles (deg) for
[Ti(trisphenolate)(O-t-Bu)]2 (6a)
Ti(1)-O(1)
Ti(1)-O(2)
1.820(2)
1.798(2)
Ti(1)-O(3′)
Ti(1)-O(4)
1.805(2)
1.763(2)
O(1)-Ti(1)-O(2) 106.96(10)
O(1)-Ti(1)-O(3′) 109.49(10)
O(1)-Ti(1)-O(4) 107.29(10)
O(2)-Ti(1)-O(3′) 112.16(10)
O(2)-Ti(1)-O(4) 109.52(10)
O(3′)-Ti(1)-O(4)
C(2)-O(1)-Ti(1)
C(13)-O(2)-Ti(1)
111.21(10)
142.9(2)
148.8(2)
hydrogens) and which benzylic resonances were due to
protons attached to the same carbon. The results showed that
each methylene carbon has an upfield and a downfield
hydrogen associated with it. This is consistent with 1H NMR
data from dititanacalix[8]arenes.8d
The NOESY spectrum enabled us to link the tert-butyl
resonances with the aromatic rings, by revealing cross peaks
between the tert-butyl groups and the aromatic protons
(Figure 4A). We were also able to connect the aromatic rings
in order, by analyzing cross peaks between the benzylic
methylene groups and the aromatic rings (Figure 4B). Only
the upfield methylene resonances from each pair show cross
C(16′)-O(3′)-Ti(1) 169.18(19)
C(41)-O(4)-Ti(1) 154.5(2)
peaks with the aromatic protons. Therefore, the upfield
resonances are the ones that direct away from the metal
center and toward the aromatic hydrogens. The tert-butyl
resonance that has no NOE cross peaks with resonances from
the trisphenolate ligand was assigned as that due to the tert-
butoxide group. The combined information from both two-
dimensional experiments provided a means for assigning each
proton resonance relative to its neighbors in the trisphenolate
3660 Inorganic Chemistry, Vol. 41, No. 14, 2002

Linear Calixarene Analogues as Chiral Coordinating Ligands
Figure 4. Expanded plots of NOESY data for 6a. Assignments are according to the sketch above and Figure 2. (A) Expanded plot of cross peaks between
tert-butyl (F1) and aromatic (F2) hydrogens, with representative correlations indicated for hydrogens H, J, and G. (B) Expanded plot of cross peaks between
benzylic (F1) and aromatic (F2) hydrogens, with a representative correlation indicated for hydrogens K and M. Note the absence of a correlation between
hydrogens K and J.
ligand. This led to the labeling scheme for the major isomer,
shown at the top of Figure 2. However, we required the
crystal structure of the minor isomer to correlate the NMR
data with a three-dimensional understanding of the trisphe-
nolate conformation.
further support of this conclusion, the two-dimensional NMR
data acquired for the minor isomer (collected on the chiral
trisphenolate complex, described below) reveal NOE cross
peaks between the pairs of protons analogous to those that
produce NOE cross peaks in the major isomer. The only
significant differences between the NMR spectra of the
isomers are chemical shift differences for several resonances
(vide infra).
Although the crystal structure is of the minor isomer and
the solution structure is of the major isomer, the similarity
of the NMR spectra of the two different isomers indicates
that the trisphenolate ligand is in a very similar conformation
in the two isomers (vide supra). Therefore, it is reasonable
to compare the NMR data of the major isomer with the
crystal structure, in order to evaluate whether the crystal
structure conformation of the minor isomer is consistent with
the NMR data from the major isomer. First, the upfield-
shifted aromatic proton (observed for both isomers), located
on the central phenol ring, is most likely that indicated in
Figure 3B as proton J. This proton directs toward one face
of the adjacent aromatic ring (distance of HJ to ring 3 is ca.
2.5 Å), the ring current from which would cause an upfield
shift. Second, benzylic proton K shows an NOE cross peak
with the aromatic proton of only one adjacent ring (proton
M), whereas proton E shows cross peaks with the aromatic
protons on both neighboring rings (protons D and G, see
arrows in Figure 3B). Inspection of the crystal structure
suggests why these are reasonable observations: proton J is
attached to the “straddle” methylene carbon, and the resultant
geometry places aromatic proton K too distant (J-K distance
is ca. 3.7 Å) to observe an NOE cross peak between them.
All of these assignments from the crystal structure are also
consistent with the ordering of the resonances determined
solely by NMR spectroscopy. Together, the NMR data and
crystal structure suggest that the trisphenolate conformation
in the major isomer is very similar to that in the minor
isomer, at least with respect to phenolate rings 2 and 3. In
Synthesis of Chiral Trisphenol
We have prepared the chiral trisphenol (1b) as a racemate,
according to the procedure outlined in eq 4. Excess 4-tert-
butyl phenol was used to prepare the unsymmetrical bisphe-
nol (9) shown in the first step. Separation of the excess
phenol from the desired product according to the literature
procedure for preparing an analogous compound11 was
problematic for us: it was very difficult to separate the final
2 equiv of 4-tert-butyl phenol from the unsymmetrical
bisphenol. We have developed a simple alternative method
in which we remove the remaining 4-tert-butyl phenol using
a Kugelrohr bulb-to-bulb distillation.
Titanium Complexes of Chiral Trisphenol
The titanium complex of the chiral trisphenol (6b) was
prepared according to eq 3. The two isomers that result
Inorganic Chemistry, Vol. 41, No. 14, 2002 3661

Appiah et al.
neighboring phenol groups, which confirm that the methyl
group is oriented away from the titanium center and toward
the aromatic protons of the phenol groups. In addition, these
cross peaks are consistent with the methyl group being
located on the “nonstraddle” methylene carbon (where proton
E is indicated in Figure 3B), because only one cross peak
with the aromatic region would be observed if the methyl
group were on the straddle methylene carbon (vide supra).
Evaluation of the crystal structure suggests that the methyl
group is not introduced into the straddle group because that
location is sterically more congested: the methyl group
would interfere with the titanium-oxygen core if it replaced
the downfield proton, or an aromatic hydrogen if it replaced
the upfield proton. Therefore, the steric size of the methyl
group controls both the location of this group in the
nonstraddle position and the orientation (and hence, chirality)
of the trisphenolate ligand when it coordinates the metal
center. This explains why only two isomers are observed
for 6b, and implies that enantiomerically pure trisphenol
would produce a specific enantiomer of trisphenolate con-
formation in the metal complex.16
Once we had assignments of the proton resonances for
the minor isomer of 6b, we translated those assignments to
the 1H NMR spectrum of the minor isomer of 6a (Figure 2,
bottom spectrum). We then compared the assigned NMR
spectra of the major and minor isomers, in order to evaluate
the spectroscopic differences in the isomers. There were
many consistencies between the two isomers, including the
upfield shift of aromatic proton J, and the fact that proton E
shows NOE cross peaks with two aromatic protons (D and
G), while proton K shows an NOE cross peak with only
one aromatic proton (M). Altogether, there are only two
significant spectroscopic differences between the isomers.
The first is that the tert-butoxide resonance is shifted slightly
upfield in the minor isomer, from 1.5 to 1.2 ppm (resonance
Q, Figure 2). An examination of the crystal structure shows
a relatively close interaction (ca. 3 Å) between the tert-
butoxide hydrogens and the central aromatic ring of one of
the trisphenolate ligands, suggesting that a ring current effect
could contribute to the slight upfield shift. The second
major difference is that one benzylic hydrogen resonance is
shifted upfield in the major isomer, from 4.8 to 3.8 ppm
(resonance F, Figure 2). Proton NMR spectra of dititana-
calix[8]arene complexes suggest that a “normal” chemical
shift for this proton is closer to 4.8 ppm.8d Therefore, the
major isomer appears to have one benzylic hydrogen in an
unusual location, either with respect to a titanium center or
with respect to an aromatic ring. At this time, we have
no other structural information regarding the major isomer;
FAB (fast atom bombardment) mass spectra of both isomers
of 6a and 7a produce a high-mass ion that corresponds
to loss of R′OR′ from compounds 6a and 7a, [(Ti-
(trisphenoxide))2O].
Figure 5. Benzylic region of the 1H NMR (C6D6) spectra of 6a (bottom)
and 6b (top), comparing major (left) and minor (right) isomers. The methyl
group in 6b replaces the benzylic hydrogen with the most upfield chemical
shift in each isomer of 6a. The singlet at 4.3 ppm is due to CH2Cl2 impurity
in the C6D6 solvent.
appear to be identical to the isomers of 6a, according to 1H
NMR spectroscopy. Our evidence for this is the strong
similarity between the benzylic proton regions of the NMR
spectra of the two isomers (Figure 5). The resonances due
to one set of methylene protons are virtually identical for
both isomers of 6a and 6b: those located at 5.1 and 4.2 ppm
for the major isomer and 5.4 and 4.2 for the minor isomer.
The only difference arises at the benzylic position that has
become substituted with a methyl group in the chiral
complex. In this case, one benzylic resonance is a quartet
(because of coupling with the methyl protons) and is shifted
slightly downfield relative to its achiral counterpart (3.8
shifted to 4.2 for the major isomer and 4.7 shifted to 5.2 for
the minor isomer). The change in chemical shift is reason-
able, since the carbon center to which this proton is attached
is now tertiary, rather than secondary. The fourth benzylic
resonance in the chiral complex is missing, since that proton
has been replaced with a methyl group, the protons from
which resonate at 1.7 ppm. Therefore, the methyl group in
the chiral complex has replaced only one hydrogen in the
achiral complex: that which has the most upfield chemical
shift. We were expecting to produce two additional isomers
of 6b, which could arise from introducing the methyl group
into each of the two different benzylic positions in the
titanium complex. We also could have produced two
additional diastereomers, if the methyl group were able to
replace either hydrogen at a given benzylic position in the
titanium complex. Our spectroscopic data indicate that one
diastereomer is produced overwhelmingly for each of the
two isomers of 6b, and that the benzylic methyl group in
these diastereomers is oriented away from the metal center
(vide infra).
The minor isomer of 6b isomerizes much more slowly
than the minor isomer of 6a; in fact, isomerization is not
apparent after 24 h at room temperature. This enabled us to
1
acquire two-dimensional H NMR spectroscopic data from
1
this isomer and completely assign the resonances in the H
NMR spectrum. We used a procedure analogous to that
described for the major isomer of 6a to assign all of the
proton resonances relative to their neighbors in the trisphenol
ring and assign the tert-butoxide resonance. The methyl
group introduced at the benzylic position shows cross peaks
that correlate with the aromatic protons (D and G) from two
(16) Because the titanium complexes reported in this paper are dimers with
inversion symmetry, enantiomerically pure trisphenol could not yield
an enantiomerically pure chiral version of these titanium complexes.
We are currently working to resolve the chiral trisphenol into
enantiomerically pure form, in order to determine what titanium
complexes would be produced in this case.
3662 Inorganic Chemistry, Vol. 41, No. 14, 2002

Linear Calixarene Analogues as Chiral Coordinating Ligands
These spectral changes and the fact that the isomerization
from the minor to the major isomer of 6b is very slow (i.e.,
1
not apparent by H NMR spectroscopy in 24 h at room
temperature) suggest that the major isomer differs from the
minor isomer by conformational “inversion” at the non-
straddle benzylic methylene group.17 This type of inversion
is proposed to occur in analogous titanium bisphenolate
complexes,18 but is slow on the NMR time scale (hence the
methylene protons resonate as two separate doublets). The
inversion, which occurs by rotation around the carbon-
carbon bonds connecting the methylene group to the two
neighboring aromatic groups, changes the orientation of the
aromatic rings with respect to the other ligands on the metal
center and causes the interchange of the two benzylic
hydrogens; the upfield hydrogen (directed toward the aro-
matic rings) becomes the downfield hydrogen, and vice versa.
We hypothesize that when the upfield hydrogen is replaced
by a methyl group, as in the minor isomer of 6b, the strain
that would result if the methyl group oriented toward the
metal center prevents the conformational inversion (and thus
the isomerization) from taking place. The change in chemical
shift of proton F when converting from the minor to the
major isomer further supports this hypothesis, since that
proton is attached to the center that would undergo inversion.
Similarly, the change in chemical shift of the tert-butoxide
group could arise from a change in its orientation with respect
to the aromatic rings upon inversion. Altogether, the data
support the hypothesis that the difference between the major
and minor isomers is due to conformational inversion of the
nonstraddle benzylic methylene group.
Figure 6. An ORTEP representation (50% probability) of the crystal
structure of 8a. Carbon atoms are unshaded. Hydrogen and disordered tert-
butyl carbon atoms are removed for ease of viewing. Labeling is according
to the distances and angles listed in Table 3.
Table 3. Selected Bond Distances (Å) and Angles (deg) for
[Al(trisphenolate)]2 (8a)
Al(1)-O(1)
Al(1)-O(2)
Al(1)-O(1′)
Al(1)-O(2′)
Al(2)-O(2)
1.6984(14)
1.8336(14)
1.6985(14)
1.8336(14)
1.8285(14)
Al(2)-O(3)
Al(2)-O(2′)
Al(2)-O(3′)
Al(1)-Al(2)
1.7023(14)
1.8284(14)
1.7022(14)
2.7548(12)
O(1)-Al(1)-O(2) 116.69(6)
O(1)-Al(1)-O(2′) 108.31(7)
O(1)-Al(1)-O(1′) 119.10(11)
O(2)-Al(2)-O(2′)
82.57(9)
O(3)-Al(2)-O(3′) 119.16(10)
Reactivity of 6a and 7a
C(2)-O(1)-Al(1)
C(13)-O(2)-Al(1) 131.98(12)
C(16)-O(3)-Al(2) 130.09(12)
130.29(13)
O(2)-Al(1)-O(2′)
82.29(9)
Compounds 6a and 7a decompose, apparently by reaction
with 1 equiv of adventitious water, to yield a compound with
the formula Ti2(trisphenolate)2O (10), as determined by FAB
mass spectrometry. Both compounds are unstable with
respect to 10, since 10 is the major high-mass ion observed
in the positive FAB mass spectrum of either 6a or 7a. The
trisphenolate ligand in compound 10 also adopts an unsym-
O(2)-Al(2)-O(3) 115.41(6)
O(2)-Al(2)-O(3′) 109.38(6)
hydrogens from the two methylene groups on the ligand
1
resonate as two doublets in the H NMR spectrum. Mass
spectrometry indicates that the complex is a dimer of Al-
(trisphenolate) units. The crystal structure shows that the
trisphenolate ligands coordinate the two aluminum centers
so that the central phenolate group bridges the two metal
centers and each terminal phenolate group coordinates a
different aluminum center (Figure 6). Together, the two
trisphenolate ligands provide tetrahedral coordination of
the aluminum atoms, expected for an Al2O6 core.19 The tris-
phenolate in this complex adopts a C2-symmetric S-shaped
conformation, where the C2 axis lies in the plane of the
central phenolate ring, oriented along the O- to tert-butyl
axis. Although the trisphenolate conformation in the alumi-
num complex is more symmetrical than in the titanium
complex, on account of the bridging central phenoxide group,
the ligand conformation is chiral. However, the complex is
not chiral overall, due to an inversion center that interconverts
the two trisphenolate ligands. We are exploring ways to
1
metrical conformation according to H NMR spectroscopy.
We are currently preparing this compound directly and
investigating details of its structure. In addition, we are
pursuing applications of all of the titanium trisphenolate
compounds in promoting ring-opening polymerization of
lactones.
Aluminum Complex of Achiral Trisphenol
Aluminum complexes of achiral trisphenol were prepared
by addition of Al(CH3)3 to 1 equiv of trisphenol 1a, yielding
[Al(trisphenolate)]2 (8a) (eq 3). In contrast to the titanium
complexes of 1a, the trisphenolate ligand in the aluminum
complex (8a) is in a symmetrical conformation; the benzylic
(17) The authors appreciate useful suggestions provided by a reviewer.
(18) (a) Floriani, C.; Corazza, F.; Lesueur, W.; Chiesi-Villa, A.; Guastini,
C. Angew. Chem., Int. Ed. Engl. 1989, 28, 66-67. (b) Corazza, F.;
Floriani, C.; Chiesi-Villa, A.; Guastini, C. Inorg. Chem. 1991, 30,
145-148. (c) Okuda, J. Fokken, S.; Kang, H.-C.; Massa, W. Chem.
Ber. 1995, 128, 221-227.
(19) (a) Chisholm, M. H.; Huffman, J. C.; Wesemann, J. L. Polyhedron
1991, 10, 1367-1372. (b) Cotton, F. A.; Wilkinson, G.; Murillo, C.
A.; Bochmann, M. In AdVanced Inorganic Chemistry, 6th ed.; John
Wiley & Sons: New York, 1999; pp 188-189.
Inorganic Chemistry, Vol. 41, No. 14, 2002 3663

Appiah et al.
butoxide was prepared by treatment of titanium(IV) isopropoxide
with 4 equiv of tert-butyl acetate, followed by fractional distilla-
tion.21
prepare a heterodimer of this complex, with one trisphenolate
and one other multidentate, perhaps chiral ligand, in order
to evaluate whether the trisphenolate ligand in these com-
plexes would be capable of chiral recognition.
NMR spectra were recorded at room temperature on a Varian
Gemini 2000 spectrometer operating at 300.1 MHz (1H) and 75.5
MHz (13C), equipped with Vnmr software version 5.3B. All
chemical shift data are reported in parts per million downfield (+)
or upfield (-) from tetramethylsilane (TMS), based on the chemical
Comparison of Trisphenolate Conformation in
Titanium and Aluminum Complexes
1
shift of CDCl3(7.24 ppm, residual H; 77.00 ppm, 13C) or C6D6
The difference in trisphenolate conformation between the
titanium and aluminum structures probably results from the
difference in steric congestion at the metal centers. We
propose that the additional alkoxide ligand in the titanium
complexes and the large terminal tert-butyl groups on the
trisphenolate ligand combine to prevent the central phenolate
oxygen from bridging the two titanium centers. This causes
the Ti-O-C bond angles to open considerably (143-169°),
in order to donate more electron density to the metal center.
This open angle, in turn, prevents the trisphenolate ligand
from coordinating to a single metal center, resulting in the
observed dimeric structures. In contrast, the aluminum
complexes have no additional steric bulk at the metal center,
so the trisphenolate can bridge the two metal centers with
the central phenolate oxygen. The Al-O-C bond angles are
smaller (∼130°) because the aluminum center gains electron
density via the bridging phenolate oxygen.
(7.15 ppm, residual 1H; 128.00 ppm, 13C). Standard two-dimensional
COSY experiments and two-dimensional phase-sensitive NOESY
experiments using the States-Haberkorn method22 (mixing time
) 500 ms) were performed using the Vnmr software package. Mass
spectra were obtained on a VG ZAB2-EQ mass spectrometer,
operated by the Mass Spectrometry Laboratory in the Chemistry
Department at the University of California, Berkeley, or on a VG
7070-HF or Finnegan MAT 95 mass spectrometer, operated by the
Mass Spectrometry Service Laboratory in the Chemistry Department
at the University of Minnesota, Minneapolis. Compounds were
analyzed by positive fast atom bombardment (+FAB) using
2-nitrophenyl octyl ether (NPOE) or 3-nitrobenzyl alcohol (NBA)
as a matrix solvent. Mass spectral data are reported as m/e (percent
of base peak). Elemental analyses were performed by Atlantic
Microlab, Inc., Norcross, GA. X-ray crystallographic studies were
performed by the X-Ray Crystallographic Laboratory in the
Chemistry Department at the University of Minnesota, Minneapolis.
2-(Bromomethyl)-4,6-bis(1,1-dimethylethyl)phenol (3). In a
500 mL three-neck round-bottomed flask, fitted with a gas inlet
tube, a thermometer, a bubbler containing water, and a magnetic
stirring bar, was dissolved 2,4-di(tert-butyl)phenol (20.0 g, 96.9
mmol) in glacial acetic acid (35 mL). Paraformaldehyde (3.49 g,
116 mmol) was added with stirring until it completely dissolved,
and hydrogen bromide gas was bubbled into the reaction mixture
for 15 min, at a rate that caused the reaction temperature to increase
to between 60 and 80 °C. After the paraformaldehyde had dissolved
and the reaction mixture formed a two-phase system, the gas flow
was turned off and the reaction mixture was stirred under ambient
pressure for 0.5 h. The reaction mixture was cooled to room
temperature, and the organic phase was separated, seeded with a
crystal from a previous preparation of 3, and cooled in an ice bath
for 1 h. The resultant colorless crystals were collected by suction
filtration, washed with water, dissolved in a minimum of warm
hexane, dried (MgSO4), filtered, and cooled to -30 °C. The first
crop of crystals was collected by vacuum filtration, and a second
crop was obtained by concentration of the mother liquors by rotary
evaporation, cooling to -30 °C, and vacuum filtration. Total yield
of both crops was 18.07 g (62%). 1H NMR (CDCl3): δ 7.33 (2H,
d, J ) 2.47 Hz), 7.10 (2H, d, J ) 2.47 Hz), 5.28 (1H, s), 4.58 (2H,
s), 1.43 (9H, s), 1.29 (9H, s). 13C NMR (CDCl3): δ 151.78, 143.08,
137.28, 125.72, 124.78, 123.35, 34.82, 34.23, 32.55, 31.42, 29.79.
2,6-Bis(3,5-bis(1,1-dimethylethyl)-2-hydroxyphenylmethyl)-4-
(1,1-dimethylethyl)phenol (Trisphenol 1a). In a 500 mL three-
necked round-bottomed flask, equipped with a thermometer, con-
denser, and magnetic stirring bar, was dissolved 4-tert-butylphenol
(2.41 g, 16.1 mmol) in 20 mL of toluene. 2-(Bromomethyl)-4,6-
bis(1,1-dimethylethyl)phenol (3, 9.65 g, 32.2 mmol) was added with
stirring. Once both reagents had completely dissolved, poly(4-
vinylpyridine) (PVP, 3.40 g, 32.2 mmol, based on pyridine) was
added to the reaction mixture. The reaction mixture was heated to
between 80 and 90 °C under dry nitrogen. After 24 h, the reaction
Conclusion
In conclusion, we have prepared linear trisphenol ana-
logues of 4-tert-butylcalixarene in both a chiral (racemic)
and an achiral form. We have characterized trisphenolate
complexes with titanium(IV) and aluminum(III) and found
that in both cases the trisphenolate ligand adopts a chiral
conformation when coordinated to the metals. In addition,
we have shown that the trisphenolate conformation can be
controlled by incorporating a methyl group at one of the
bridging methylene carbons. These results suggest that the
metal-ligand complexes have potential applications in
stereo- or enantioselective transformations, either directly or
by amplifying the chirality of a different ligand that is
coordinated to the metal center.20 Along these lines, we are
currently developing applications of the coordinatively
unsaturated titanium complexes as stereoselective reagents
in organic synthesis and stereoselective initiators in ring-
opening polymerization reactions. We are also developing a
method for resolving the chiral trisphenol into enantiomeri-
cally pure form.
Experimental Section
General. All experiments involving metal-based reagents were
run under nitrogen at 25 °C in a Vacuum Atmospheres glovebox,
unless specified otherwise. Solvents were purchased in anhydrous
form under nitrogen in Sure/Seal bottles from Aldrich Chemical
Co. and transferred into the glovebox without exposure to air.
Starting materials were obtained from commercial suppliers and
used as received, except as specifically noted. Titanium(IV)
isopropoxide was vacuum distilled before use. Titanium(IV) tert-
(20) (a) Chavarot, M.; Byrne, J. J.; Chavant, P. Y.; Pardillos-Guindet, J.;
Valle´e, Y. Tetrahedron: Asymmetry 1998, 9, 3889-3894. (b) Balsells,
J.; Walsh, P. J. J. Am. Chem. Soc. 2000, 122, 1802-1803.
(21) Mehrotra, R. C. J. Am. Chem. Soc. 1954, 76, 2266-2267.
(22) States, D. J.; Haberkorn, R. A.; Ruben, D. J. J. Magn. Reson. 1982,
48, 286-292.
3664 Inorganic Chemistry, Vol. 41, No. 14, 2002

Linear Calixarene Analogues as Chiral Coordinating Ligands
mixture was cooled to room temperature and the poly(vinylpyri-
dinium bromide) was removed by vacuum filtration. The solution
was concentrated first by rotary evaporation and then under high
vacuum to a red foam. The foam was transferred into the glovebox,
where it was dissolved in a minimum amount of pentane at room
temperature and cooled to -30 °C. Colorless crystals were isolated
by vacuum filtration in the glovebox. Subsequent crops of crystals
were obtained by concentrating the mother liquors and cooling them
to -30 °C. Two crops of crystals were isolated, yielding a total of
dissolved in pentane (10 mL) and concentrated to an orange foam,
which was triturated with pentane (5 mL) to produce a yellow solid
in an orange solution. The slurry was cooled to -30 °C for 48 h,
and then the yellow solid was isolated by vacuum filtration, washed
with cold pentane (ca. 5 mL), and dried under vacuum (0.754 g,
64%). NMR spectroscopy indicates that the product is predomi-
nantly (g90%) the major isomer, with a small amount of contami-
nation by the minor isomer and 10. 1H NMR (major isomer,
C6D6): δ 7.53 (2H, d, J ) 2.5 Hz), 7.51 (2H, d, J ) 2.5 Hz), 7.36
(2H, d, J ) 2.2 Hz), 7.31 (2H, d, J ) 2.2 Hz), 7.20 (2H, d, J ) 2.2
Hz), 6.57 (2H, d, J ) 1.9 Hz), 5.11 (2H, d, J ) 17.0 Hz), 4.57
(2H, septet, J ) 6.0 Hz), 4.16 (2H, d, J ) 17.0 Hz), 3.57 (2H, d,
J ) 14.0 Hz), 3.27 (2H, d, J ) 14.0 Hz), 1.77 (18H, s), 1.59 (18H,
s), 1.32 (18H, s), 1.28 (18H, s), 1.18 (12H, d, J ) 6.0), 1.15 (18H,
s). 13C{1H} NMR (major isomer, C6D6): δ 162.59, 160.13, 159.35,
144.50, 144.22, 137.30, 136.21, 133.66, 132.31, 130.02, 127.20,
126.91, 125.04, 123.68, 122.96, 122.07, 121.95, 81.18, 35.44, 35.38,
34.64, 34.41, 34.30, 34.12, 33.93, 31.62, 31.54, 31.47, 30.81, 30.62,
26.06, 26.03. 1H NMR (minor isomer, selected resonances, C6D6):
δ 6.75 (2H, br), 5.16 (2H, d, J ) 17.4 Hz), 4.68 (2H, d, J ) 15.9
Hz), 4.36 (2H, septet, J ) 6.0 Hz), 4.24 (2H, d, J ) 17.4 Hz), 3.49
(2H, d, J ) 12.6 Hz). Mass spectrum (major isomer, (+)-FAB,
NPOE): 1278.5 (100%, [M - (i-Pr)2O]+•), 486.2(74%).
2-(1-Bromoethyl)-4,6-bis(1,1-dimethylethyl)phenol (11). In a
procedure similar to that for preparing the bromomethyl derivative
(3), 2,4-bis(tert-butyl)phenol (16.59 g, 0.0804 mole) was dissolved
in concentrated acetic acid (20 mL) in a 250 mL three-neck flask,
equipped with a gas inlet, a bubbler containing water as a trap for
excess HBr, a thermometer, and a magnetic stirring bar. Acetal-
dehyde was added (4.25 g, 0.969 mole), and then anhydrous
hydrogen bromide was bubbled through the reaction mixture at a
rate that caused the temperature to rise to between 60 and 80 °C.
Once the solution was saturated with HBr, the addition of HBr
was ceased and the reaction mixture was stirred under ambient
pressure for 0.5 h. The resultant two-phase system was separated,
and the organic layer was purged with dry nitrogen gas (15 min)
and then concentrated by rotary evaporation at 30 °C under reduced
pressure (0.1 Torr), yielding 21.54 g of viscous oil. 1H NMR
spectroscopy of the crude product showed that it was contaminated
by ca. 33% acetic acid. The crude yield, accounting for the impurity,
was approximately 57%. Attempts to further purify this product
resulted in its decomposition. 1H NMR (CDCl3): δ 7.33 (1H, d, J
) 2.2 Hz), 7.18 (1H, d, J ) 2.5), 5.46 (1H, q, J ) 7.1 Hz), 2.14
(3H, d, J ) 7.1 Hz), 2.09 (CH3CO2H impurity), 1.43, (9H, s), 1.30
(9H, s). 13C{1H} NMR (CDCl3): δ 151.09, 142.75, 137.52, 128.15,
125.01, 120.92, 49.01, 34.88, 34.38, 31.45, 29.82, 24.83.
1
6.81 g (72%). H NMR (C6D6): δ 7.82 (1H, s), 7.39 (2H, d, J )
2.47 Hz), 7.34 (2H, d, J ) 2.47 Hz), 7.28 (2H, s), 6.43 (2H, s),
3.89 (4H, s), 1.45 (18H, s), 1.39 (18H, s), 1.28 (9H, s). 13C NMR
(CDCl3): δ 149.15, 147.83, 144.19, 143.2, 135.49, 127.37, 127.01,
125.95, 125.51, 122.38, 34.43, 34.14, 33.90, 31.88, 31.44, 30.09.
Mass spectrum ((+)-FAB, NBA): 586.5 (43, M+•), 381.3 (41),
219.2 (90), 175.2 (55), 56.9 (100). Anal. Calcd for C40H58O3: C,
81.86; H, 9.96. Found: C, 81.71; H, 10.66.
Ti2(trisphenolate)2(O-t-Bu)2 (6a). In a 20 mL vial, 1a (1.5 g,
2.6 mol) was dissolved in toluene (5 mL) and stirred. In another
vial, titanium(IV) tert-butoxide (0.87 g, 2.6 mmol) was dissolved
in toluene (5 mL) and the solution of titanium(IV) tert-butoxide
was added dropwise to the solution of 1a and stirred for 24 h. The
deep orange solution was concentrated by rotary evaporation and
recrystallized from a minimum amount of pentane at -30 °C. The
resultant yellow crystals were isolated by vacuum filtration, and
subsequent crops were obtained by successive concentration and
cooling of the mother liquors. The combined yield of five crops of
crystals was 9.39 g (51%). These crops varied in composition, from
100% minor isomer to a 50:50 mixture of major:minor isomers.
After the last crop was isolated, the mother liquors were concen-
trated by rotary evaporation, and the 1H NMR spectrum indicated
that the residue was ca. 90% major isomer. 1H NMR (major isomer,
C6D6): δ 7.51 (2H, d, J ) 2.5 Hz), 7.47 (2H, d, J ) 2.5 Hz), 7.31
(2H, d, J ) 2.2 Hz), 7.28 (2H, d, J ) 2.5 Hz), 7.14 (2H, d, J ) 2.5
Hz), 6.48 (2H, d, J ) 2.2 Hz), 5.06 (2H, d, J ) 16.8 Hz), 4.20
(2H, d, J ) 17.3 Hz), 3.73 (2H, d, J ) 14.0 Hz), 3.29 (2H, d, J )
14.3 Hz), 1.75 (18H, s), 1.59 (18H, s), 1.38 (18H, s), 1.29 (18H,
s), 1.26 (18H, s), 1.17 (18H, s). 13C{1H} NMR (major isomer,
C6D6): δ 162.50, 160.18, 159.29, 144.32, 144.19, 144.10, 137.25,
136.14, 134.17, 132.88, 129.82, 127.33, 126.80, 124.94, 123.59,
123.02, 122.06, 121.92, 86.76, 35.44, 35.36, 34.79, 34.48, 34.37,
1
34.27, 34.10, 31.95, 31.58, 31.51, 30.84, 30.56). H NMR (minor
isomer, C6D6): δ 7.42 (2H, d, J ) 2.5 Hz), 7.38 (2H, d, J ) 2.2
Hz), 7.37, (2H, d, J ) 2.5 Hz), 7.28 (2H, d, J ) 2.5 Hz), 7.05 (2H,
d, J ) 2.5 Hz), 6.64 (2H, d, J ) 2.2 Hz), 5.37 (2H, d, J ) 17.6
Hz), 4.69 (2H, d, J ) 14.0 Hz), 4.12 (2H, d, J ) 17.6 Hz), 3.53
(2H, d, J ) 14.3 Hz), 1.62 (18H, s), 1.55 (18H, s), 1.29 (18H, s),
1.26 (18H, s), 1.18 (18H, s), 1.16 (18H, s). 13C{1H} NMR (minor
isomer, C6D6) 162.11, 160.29, 159.52, 144.40, 144.05, 143.92,
137.81, 136.81, 135.33, 133.46, 130.23, 127.54, 124.95, 124.57,
123.91, 122.04, 121.74, 87.02, 35.36, 34.87, 34.38, 34.33, 34.18,
31.72, 31.60, 31.55, 31.45, 30.57, 30.52. Mass spectrum (major
isomer, (+)-FAB, NPOE): 1353 (35, [M - (CH3)2CCH2]+•), 1279
(100, [M - (t-Bu)2O]+•), 568 (23). Mass spectrum (minor isomer,
(+)-FAB, NPOE): 1353 (27, [M - (CH3)2CCH2]+•), 1279 (100,
[M - (t-Bu)2O]+•), 568 (22). Anal. Calcd for C43H62O4Ti (minor
isomer): C, 74.98; H, 9.15. Found: C, 74.29; H, 8.83.
2-(3,5-Bis(1,1-dimethylethyl)-2-hydroxy-1-phenylethyl)- 4-(1,1-
dimethylethyl)phenol (9). This procedure was adapted from an
analogous literature procedure.11b In a 250 mL round-bottomed flask
fitted with a condenser, 4-tert-butylphenol (17.52 g, 0.117 mol)
was dissolved with stirring in 50 mL of toluene. After the phenol
had dissolved, PVP (2.45 g, 0.0233 mol of pyridine) was added,
followed by 2-(1-bromomethyl)-4,6-bis(1,1-dimethylethyl)phenol
(6.98 g, 0.0233 mol), as a solution in 20 mL of toluene. The reaction
mixture was heated at 60 °C for 18 h under dry nitrogen. The
resultant dark orange reaction mixture was cooled to room
temperature and filtered, and the filtrate was concentrated by rotary
evaporation to an orange foam. The residue was dissolved in
petroleum ether at room temperature (20 mL) and cooled to -30
°C to precipitate excess 4-tert-butylphenol, which was separated
by filtration. The filtrate was concentrated to an orange oil by rotary
evaporation, and the remaining 2 equiv of 4-tert-butylphenol was
removed by a bulb-to-bulb Kugelrohr distillation (90 °C/6 mmHg).
Ti2(trisphenolate)2(O-i-Pr)2 (7a). In a 100 mL round-bottomed
flask, 1a (1.00 g, 1.70 mmol) was dissolved in 20 mL of toluene.
Titanium(IV) isopropoxide (0.484 g, 1.70 mmol), dissolved in 5
mL of toluene, was added dropwise with stirring. The resultant
orange solution was stirred for 3 h at room temperature and
concentrated by rotary evaporation to an orange film. The film was
Inorganic Chemistry, Vol. 41, No. 14, 2002 3665

Appiah et al.
The desired product (9) remained in the first distillation bulb as an
orange, glassy solid and was used without further purification (4.2
g, 48%). 1H NMR (CDCl3): δ 7.29 (1H, d, J ) 2.5 Hz), 7.15 (2H,
s), 7.08 (1H, dd, J ) 8.2 Hz, J ′ ) 2.5 Hz), 6.67 (1H, d, J ) 8.2
Hz), 6.49 (1H, s), 5.54 (1H, s), 3.89 (2H, s), 1.38 (9H, s), 1.26
(9H, s), 1.25 (9H, s).
J ) 17.0 Hz), 4.33 (2H, q, J ) 7.3 Hz), 4.28 (2H, d, J ) 16.8 Hz),
1.81 (18H, s), 1.62 (18H, s), 1.41 (18H, s), 1.40 (18H, s), 1.38
(18H, s) (chiral trisphenolate methyl group not identified, due to
impure sample). 1H NMR (C6D6, minor isomer): δ 7.66 (2H, d, J
) 2.5 Hz), 7.58 (2H, d, J ) 1.9 Hz), 7.47 (2H, d, J ) 2.2 Hz),
7.33 (2H, d, J ) 2.5 Hz), 7.08 (2H, d, J ) 2.2 Hz), 6.70 (2H, d,
J ) 2.2 Hz), 5.41 (2H, d, J ) 17.0 Hz), 5.13 (2H, q, J ) 6.9 Hz),
4.16 (2H, d, J ) 17.6 Hz), 1.69 (6H, d, J ) 7.0 Hz), 1.64 (18H, s),
1.55 (18H, s), 1.32 (18H, s), 1.24 (18H, s), 1.19 (18H, s), 1.14
(18H, s).
2-(3,5-Bis(1,1-dimethylethyl)-2-hydroxy-1-phenylethyl)-6-(3,5-
bis(1,1-dimethylethyl)-2-hydroxyphenylmethyl)-4-(1,1-dimeth-
ylethyl)phenol (Chiral Trisphenol, 1b). A 250 mL round-bottomed
flask was equipped with a thermometer and a stir bar and charged
with bisphenol 9 (0.96 g, 2.60 mmol) and toluene (20 mL). Once
the bisphenol had dissolved, PVP (0.273 g, 2.60 mmol of pyridine)
was added, followed by 11 (0.816 g, 2.60 mmol) as a solution in
toluene (10 mL). The reaction flask was equipped with a condenser
and heated to 58 °C for 72 h. After this time, 1H NMR spectroscopy
of an aliquot indicated that the reaction was incomplete, but that
all of the 2-(1-bromoethyl)-4,6-bis(1,1-dimethylethyl)phenol had
reacted. Thus, an additional 1 equiv of both 2-(1-bromoethyl)-4,6-
bis(1,1-dimethylethyl)phenol and PVP (2.60 mmol of each) was
added and the reaction mixture was heated to 58 °C and stirred for
an additional 24 h. 1H NMR spectroscopy of an aliquot still
indicated that the reaction was incomplete, but the reaction mixture
was filtered and the filtrate was concentrated by rotary evaporation
to a glassy solid. This solid was transferred into the glovebox and
dissolved in 5 mL of pentane and concentrated by rotary evaporation
to a colorless foam. The foam was dissolved in 5 mL of pentane
and the solution cooled to -30 °C for 48 h. A colorless powder
was isolated by filtration (0.410 g), and the mother liquors were
concentrated to 2 mL and cooled to -30 °C. A second crop was
isolated, weighing 0.091 g (32% combined yield, not optimized).
1H NMR (C6D6): δ 7.45 (1H, s), 7.33 (1H, d, J ) 2.5 Hz), 7.29
(1H, d, J ) 2.2 Hz), 7.15 (3H, m), 7.10 (1H, d, J ) 2.5 Hz), 6.37
(1H, s), 6.04 (1H, s), 4.61 (1H, q, J ) 7.1 Hz), 1.53 (3H, d, J )
7.1 Hz), 1.31 (9H, s), 1.29 (9H, s), 1.24 (9H, s), 1.23 (9H, s), 1.17
(9H, s). 13C{1H} NMR (C6D6): δ 150.10, 149.84, 147.91, 144.57,
143.33, 143.18, 136.04, 135.95, 132.26, 131.96, 125.68, 122.43,
122.35, 122.01, 121.76, 34.65, 34.62, 34.43, 34.18, 34.08, 31.95,
31.59, 31.31, 31.12, 30.02, 29.99, 20.24. Mass spectrum ((+)-FAB,
NBA): 601 (9%, MH+), 600 (8%, M+), 586 (7%), 394 (72%), 233
(99%), 217 (68%), 203 (40%), 189 (100%), 133 (68%). Anal. Calcd
for C41H60O3: C, 81.95; H, 10.06. Found: C, 81.27; H, 10.06.
Al2(trisphenolate)2 (8a). In a 100 mL round-bottomed flask in
the drybox was dissolved 1a (1.00 g, 0.170 mmol) in toluene (15
mL). The colorless solution was cooled to -30 °C for 1 h, removed
from the drybox freezer, and stirred, and trimethylaluminum (0.163
mL, 0.170 mmol) was added dropwise by syringe. Gas evolution
was immediately apparent and the reaction mixture was allowed
to warm to room temperature, during which time the solution turned
pale yellow. After 1 h, the reaction mixture was concentrated by
rotary evaporation to a pale yellow foam. The residue was dissolved
in a mixture of pentane (5 mL) and THF (1 mL) and cooled to
-30 °C. The first crop was isolated by filtration (0.202 g), and the
filtrate was concentrated to a pale yellow film and dissolved in
pentane (5 mL) to yield a second crop (27 mg, 22% combined
yield). The liquors were concentrated to a pale yellow residue that
1
was impure by NMR (0.145 g). H NMR (C6D6): δ 7.14 (4H, s),
7.13 (8H, m), 4.65 (4H, d, J ) 16.8), 3.64 (4H, d, J ) 16.8), 1.23
(36H, s), 1.13 (18H, s), 0.85 (36H, s). 13C{1H} NMR (C6D6): δ
151.43, 147.88, 146.62, 141.70, 138.25, 129.02, 128.65, 126.30,
124.20, 122.85, 34.77, 34.27, 34.09, 33.09, 31.78, 31.27, 29.19.
Mass spectrum ((+)-FAB, NBA): 1221 (26, M+•), 587 (23), 381
(40), 325 (26), 219 (100), 203 (34), 175 (65), 163 (31), 147 (25),
119 (33).
X-ray Crystallography. X-ray crystallographic studies were
carried out using a Bruker SMART system (6a) or Siemens SMART
system (8a) at 173(2) K using Mo KR radiation (graphite mono-
chromatized, λ ) 0.71073 Å). A preliminary set of cell constants
was calculated from reflections harvested from three sets of 20
frames. These initial sets of frames were oriented such that
orthogonal wedges of reciprocal space were surveyed. This
produced initial orientation matrices determined from 39 (6a) or
348 (8a) reflections. Final cell constants were calculated from a
set of 5804 (6a) or 5858 (8a) strong reflections from the actual
data collection.23
The data collection technique used for these crystals is generally
known as a hemisphere collection. A randomly oriented region of
reciprocal space was surveyed to the extent of 1.3 (6a) or 1.5 (8a)
hemispheres and to a resolution of 0.77 Å (6a) or 0.84 Å (8a).
Three major sections of frames were collected with 0.30° steps in
ω, and additionally at three different φ settings and a detector
position of -28° in 2θ for compound 6a. The intensity data were
corrected for absorption.24
The space groups P1h (6a) and Pbcn (8a) were determined by
the lack of (for 6a) or presence of (for 8a) systematic absences
and intensity statistics.25 A direct-methods solution was calculated,
which provided most non-hydrogen atoms from the E-map. Full-
matrix least squares and difference Fourier cycles were performed,
which located the remaining non-hydrogen atoms. All non-hydrogen
Ti2(chiral trisphenolate)2(O-t-Bu)2 (6b). In a 20 mL vial, 1b
(150 mg, 0.25 mmol) was dissolved in toluene (1 mL). Titanium-
(IV) tert-butoxide (85 mg, 0.25 mol) was added dropwise as a
solution in toluene (2 mL), and the resultant orange solution was
stirred at room temperature for 24 h. The solution was concentrated
by rotary evaporation to an orange glassy residue. The residue was
dissolved in pentane (5 mL) and concentrated by rotary evaporation
to an orange solid, and these dissolution and concentration steps
were repeated two additional times. The residue was concentrated
under vacuum for an additional 1 h, and then it was dissolved in
pentane (2 mL) and cooled to -30 °C for 72 h. A yellow powder
was isolated by vacuum filtration (20 mg), and 1H NMR spectros-
copy indicated that it was the minor isomer. An additional
crystalline solid was isolated after concentration of the mother
1
liquors, crystallization at -30 °C, and filtration (14 mg), and H
NMR spectroscopy again indicated that it was the minor isomer
(19% combined yield). The mother liquors were concentrated to
an orange residue (12 mg), the 1H NMR spectrum of which
(23) Bruker. SAINT Version 6.01; Bruker Analytical X-Ray Systems, Inc.:
Madison, WI, 1999.
(24) (a) Blessing, R. H. Acta Crystallogr. 1995, A51, 33-38. (b) Sheldrick,
G. M. SADABS; University of Go¨ttingen: Go¨ttingen, Germany, 1999.
(25) Bruker. SHELXTL-Plus Version 5.4; Bruker Analytical X-Ray Systems,
Inc.: Madison, WI, 1997.
1
indicated that it was predominantly the major isomer. H NMR
(C6D6, major isomer): δ 7.64 (2H, d, J ) 2.7 Hz), 7.59 (2H, d,
J ) 2.5 Hz), 7.55 (2H, d, J ) 3.0 Hz), 7.37 (2H, d, J ) 2.5 Hz),
7.28 (2H, d, J ) 2.5 Hz), 6.47 (2H, d, J ) 2.2 Hz), 5.48 (2H, d,
3666 Inorganic Chemistry, Vol. 41, No. 14, 2002

Linear Calixarene Analogues as Chiral Coordinating Ligands
atoms were refined with anisotropic displacement parameters. All
hydrogen atoms were placed in ideal positions and refined as riding
atoms with individual (or group if appropriate) isotropic displace-
ment parameters. The final full-matrix least squares refinements
converged to the R values reported in Table 1. The high R values
can be attributed to the presence of large voids in the crystals, which
allowed for disorder of the incorporated solvents.
The structure of 6a was found to be a dimer with seven solvent
molecules (THF) per unit cell (per dimer). Disorder was modeled
for two tert-butyl groups of the trisphenolate ligand. SADI
instructions were applied to restrain C-C distances to be similar.
One THF molecule is disordered over two positions (53:47). SAME
instructions were applied to restrain the geometry of the two sites
to be similar. One THF molecule is disordered over a special
position.
The structure of 8a was found with several toluene solvent
molecules. Half of the molecule of interest is unique, being related
by a 2-fold rotation axis parallel to b. Both Al atoms are located
on the 2-fold rotation axis. Two solvent molecules were found with
confidence. One is disordered head-to-tail in a cavity. A region
containing some additional badly disordered solvent was identified
with PLATON/SQUEEZE.26 This void could potentially hold up
to 8-12 more molecules of toluene per unit cell. The potential
solvent volume was determined to be 1537.3 Å3 in the unit cell
volume of 10901.2 Å3. The total electron count was determined to
be 377 in the cell (this number corresponds to about 8 extra
disordered toluene molecules per unit cell). The R value improved
about 6% after the data were corrected for the disordered solvent.
One hundred eighty-two restraints were applied to assist in a
sensible refinement of the head-to-tail disordered toluene.
Acknowledgment. This research was supported by
awards from the Camille and Henry Dreyfus Foundation
(SU-95-051) and Research Corporation (CC4888). The
funding for the 300 MHz NMR spectrometer at Gustavus
Adolphus College was provided by awards from the National
Science Foundation (DUE-9352027) and the Camille and
Henry Dreyfus Foundation (SG-95-098).
Supporting Information Available: X-ray crystal structure data
for 6a and 8a, including tables of crystal data, positional and thermal
parameters, and bond distances and angles (CIF file), two-
dimensional NMR data for 6a and 6b, and 1H NMR spectra of all
reported compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
(26) Spek, A. L. Acta Crystallogr. 1990, A46, C34.
IC010685J
Inorganic Chemistry, Vol. 41, No. 14, 2002 3667