Organometallic reactivity on a calix[4]arene oxo surface. The stepwise migratory insertion of carbon monoxide and isocyanides into zirconium-carbon bonds anchored to a calix[4]arene moiety

Article abstract of DOI:10.1021/ja9708225

The present report deals with the multistep migratory insertion of carbon monoxide and isocyanides into the metal-carbon bonds of the ZrR₂ fragment anchored to a tetroxo matrix, defined by the dimethoxycalix[4]arene dianion [p-Bu(t)-calix[4]-(O

Full text of DOI:10.1021/ja9708225

J. Am. Chem. Soc. 1997, 119, 9709-9719
9709
Organometallic Reactivity on a Calix[4]arene Oxo Surface.
The Stepwise Migratory Insertion of Carbon Monoxide and
Isocyanides into Zirconium-Carbon Bonds Anchored to a
Calix[4]arene Moiety
Luca Giannini,† Alessandro Caselli,† Euro Solari,† Carlo Floriani,*,†
Angiola Chiesi-Villa,‡ Corrado Rizzoli,‡ Nazzareno Re,§ and Antonio Sgamellotti§
Contribution from the Institut de Chimie Mine´rale et Analytique, BCH, UniVersite´ de Lausanne,
CH-1015 Lausanne, Switzerland, Dipartimento di Chimica, UniVersita` di Parma,
I-43100 Parma, Italy, and Dipartimento di Chimica, UniVersita` di Perugia,
I-06100 Perugia, Italy
ReceiVed March 13, 1997X
Abstract: The present report deals with the multistep migratory insertion of carbon monoxide and isocyanides into
the metal-carbon bonds of the ZrR2 fragment anchored to a tetroxo matrix, defined by the dimethoxycalix[4]arene
dianion [p-But-calix[4]-(OMe)2(O)2]2-. The relevant differences between the two fragments [p-But-calix[4]-(OMe)2-
(O)2Zr]2+ and [Cp2Zr]2+ are experimentally proven and theoretically interpreted. Unlike those for [Cp2Zr]2+ deriv-
atives, the reaction of CO with [p-But-calix[4]-(OMe)2(O)2ZrR2] [R ) Me, 1; CH2Ph, 2; p-MeC6H4, 3] proceeds Via
a two-step migration directly to the corresponding η2-metal-bonded-ketones [p-But-calix[4]-(OMe)2(O)2Zr{η2-C(R2)O]
(R ) Me, 4; CH2Ph, 5; p-MeC6H4, 6). The Zr-C functionality in 4-6 maintains its inserting capability in the
reaction with ketones, i.e., Ph2CO, leading to the diolate derivative [p-But-calix[4]-(OMe)2(O)2Zr{OC(R2)C(Ph2)O}]
(8, R ) p-MeC6H4), or with ButNC, forming [p-But-calix[4]-(OMe)2(O)2Zr{OC(Me2)C(CNBut)N(But)}] (7). The
reaction of 1-3 with ButNC proceeds through two different pathways, depending on the temperature and the R
substituent at the metal. The first one leads to the η2-metal-bonded imine [p-But-calix[4]-(OMe)2(O)2Zr{η2-
N(But)C(R2)}] [ R ) Me, 12; CH2Ph, 13], which inserts two additional molecules of ButNC to give a
metalladiazacyclopentane, [p-But-calix[4]-(OMe)2(O)2Zr{N(But)C(Me2)C(CNBut)N(But)}] (14). The second pathway
led, at low temperature, to the bis-η2-iminoacyls [p-But-calix[4]-(OMe)2(O)2Zr{η2-N(But)dC(R)}2] [R ) Me, 15;
p-MeC6H4, 16], which at 60 °C couple to the corresponding enediamido ligand in [p-But-calix[4]-(OMe)2(O)2Zr-
{N(But)C(R)dC(R)N(But)}] [R ) Me, 17; p-MeC6H4, 18]. The coupling of 15 and 16 showed first order kinetics
in both cases and allowed the calculation of the corresponding activation parameters. The most significant differences
between [p-But-calix[4]-(OMe)2(O)2Zr]2+ and [Cp2Zr]2+ have been interpreted on the basis of the frontier orbital
sets of the two fragments.
Introduction
thus creating the possibility of having functionalizable sites at
the metal. An additional advantage is the potential of using
the calix[4]arene bridging methylene and the But groups as
spectroscopic probes for revealing the symmetry of the fragment
attached to the transition metal.
Calix[4]arenes exemplify,1 at the molecular level, the chemi-
cal environment that an organometallic fragment is supposed
to experience when anchored to solid state supports, such as
oxides or zeolites.2 Although the anchoring groups in calix-
[n]arene are methylene-bridged phenoxo anions, the role of
calix[n]arene as an ancillary ligand is different from that played
by the same number of monomeric phenoxo moieties. This is
easily understood for the following reasons: (i) the calix[4]-
arene provides a nearly planar oxo surface for anchoring the
metal; (ii) the charge provided by the O4 moiety can be adjusted
from -4 to -2 Via partial methylation of the phenoxo groups,
In order to emphasize the peculiar role of calix[4]arene as a
supporting ligand3-5 compared to cyclopentadienyl and phenoxo
anions, we chose the Zr-C bond for its versatility and poten-
tial in organic synthesis6,7 and focused on one of the most
useful, and as it turned out peculiar, reactions, namely the mi-
gratory insertion of carbon monoxide and isocyanides.8 Herein,
we report the kinetically and structurally defined stepwise
* To whom correspondence should be addressed.
† Universite´ de Lausanne.
(3) Giannini, L.; Solari, E.; Zanotti-Gerosa, A.; Floriani, C.; Chiesi-Villa,
A.; Rizzoli, C. Angew. Chem., Int. Ed. Engl. 1996, 35, 85-87.
(4) Castellano, B.; Zanotti-Gerosa, A.; Solari, E.; Floriani, C.; Chiesi-
Villa, A.; Rizzoli, C. Organometallics 1996, 15, 4894-4896.
(5) (a) Giannini, L.; Solari, E.; Zanotti-Gerosa, A.; Floriani, C.; Chiesi-
Villa, A.; Rizzoli, C. Angew. Chem., Int. Ed. Engl. 1996, 35, 2825-2827.
(b) Giannini, L.; Solari, E.; Zanotti-Gerosa, A.; Floriani, C.; Chiesi-Villa,
A.; Rizzoli, C. Angew. Chem., Int. Ed. Engl. In press. (c) Gardiner, M. G.;
Lawrence, S. M.; Raston, C. L.; Skelton, B. W.; White, A. H. Chem.
Commun. 1996, 2491. (d) Gardiner, M. G.; Koutsantonis, G. A.; Lawrence,
S. M.; Nichols, P. J.; Raston, C. L. Chem. Commun. 1996, 2035. (e) Gibson,
V. C.; Redshaw, C.; Clegg, W.; Elsegood, M. R. J. J. Chem. Soc., Chem.
Commun. 1995, 2371. (f) Acho, J. A.; Doerrer, L. H.; Lippard, S. J. Inorg.
Chem. 1995, 34, 2542.
‡ Universita` di Parma.
§ Universita` di Perugia.
X Abstract published in AdVance ACS Abstracts, September 15, 1997.
(1) Gutsche, C. D. Calixarenes; The Royal Society of Chemistry:
Cambridge, U.K., 1989. (b) Calixarenes, A Versatile Class of Macrocyclic
Compounds; Vicens, J., Bo¨hmer, V., Eds.; Kluwer: Dordrecht, The
Netherlands, 1991.
(2) (a) Corker, J.; Lefebvre, F.; Lecuyer, C.; Dufaud, V.; Quignard, F.;
Choplin, A.; Evans, J.; Basset, J.-M. Science 1996, 271, 966-969. (b)
Niccolai, G. P.; Basset, J.-M. Appl. Catal. A 1996, 146, 145-156. (c) Vidal,
V.; Theolier, A.; Thivolle-Cazat, J.; Basset, J.-M.; Corker, J. J. Am. Chem.
Soc. 1996, 118, 4595-4662.
S0002-7863(97)00822-6 CCC: $14.00 © 1997 American Chemical Society

9710 J. Am. Chem. Soc., Vol. 119, No. 41, 1997
Giannini et al.
exo-CH2), 1.68 (s, 6H, Me), 1.27 (s,18H, But), 1.11 (s, 9H, But),
overlapping with 1.13 (s, 9H, But). Anal. Calcd for 4, C49H64O5Zr:
C, 71.40; H, 7.83. Found: C, 71.15; H, 8.16. The product is thermally
stable in C6D6 (14 h, 65 °C) but slowly decomposes when exposed to
solar light.
insertion processes occurring on the dialkyl derivatives shown
below:
Synthesis of 5. Compound 2 (5.56 g, 5.86 mmol) was dissolved in
benzene (200 mL), and the N2 atmosphere was replaced with CO. The
mixture was stirred at room temperature for 30 min and became
colorless. Volatiles were then removed in Vacuo, and hexane (50 mL)
was added to the residue. The flask was kept at -25 °C for 14 h, and
the white 5‚(C6H6)1.5 was collected and dried in Vacuo (4.2 g, 65.5%).
1H NMR (C6D6, 298 K): δ 7.63 (m, 4H, ArH(PhCH2)), 7.24-7.15
(m, ArH), 6.90 (m, 2H, ArH(PhCH2)), 6.82 (s, 2H, ArH), 6.79 (s, 2H,
ArH), 4.45 (s, 3H, MeO), 4.22 (d, J ) 12.4 Hz, 4H, endo-CH2), 4.03
(s, 3H, MeO), 3.65 (d, J ) 13 Hz, 2H, PhCH2), 3.47 (d, J ) 13 Hz,
2H, PhCH2), 3.19 (d, J ) 12.4 Hz, 2H, exo-CH2), overlapping with
3.18 (d, J ) 12.4 Hz, 2H, exo-CH2), 1.44 (s, 18H, But), 0.74 (s, 9H,
But), overlapping with 0.72 (s, 9H, But). 1H NMR (CD2Cl2, 298 K):
δ 7.51 (m, 4H, ArH(PhCH2)), 7.40 (s, 9H, C6H6), 7.20 (m, 4H, ArH-
(PhCH2)), 7.05 (s, 4H, ArH), 6.98 (s, 4H, ArH), 6.88 (m, 2H, ArH-
(PhCH2)), 4.70 (s, 3H, MeO), 4.36 (s, 3H, MeO), 4.15 (d, J ) 12.4
Hz, 2H, endo-CH2), overlapping with 4.13 (d, J ) 12.4 Hz, 2H, endo-
CH2), 3.31 (s, 4H, PhCH2), 3.28 (d, J ) 12.4 Hz, 2H, exo-CH2), 3.22
(d, J ) 12.4 Hz, 2H, exo-CH2), 1.27 (s,18H, But), 1.11 (s, 9H, But),
These reactions proceed Via metal-assisted C-C bond forma-
tion and lead to a variety of carbon-metal-bonded fragments
supported by the unusual tetraoxo [calix[4]-(OMe)2(O)2]2-
dianion in a quasi-planar arrangement. A preliminary report
on this subject appeared recently.3 Extended Hu¨ckel calcula-
tions have been performed with the purpose of rationalizing
the major differences between the [Cp2Zr]2+ and the [calix[4]-
(OMe)2(O)2Zr]2+ fragments in driving migratory insertion
reactions.
overlapping with 1.10 (s, 9H, But). Anal. Calcd for 5‚(C6H6)1.5
,
Experimental Section
C70H81O5Zr: C, 76.88; H, 7.47. Found: C, 76.86; H, 7.66. Crystals
suitable for X-ray analysis were obtained by cooling a saturated solution
of 5 in hexane/benzene (ca. 7:1) to -25 °C. The product is thermally
stable in C6D6 (14 h, 65 °C) but slowly decomposes when exposed to
solar light. When a proton source is present, [p-But-calix[4]-(OMe)2-
(O)2Zr{OCH(CH2Ph)2}2] is easily formed.
All operations were carried out under an atmosphere of purified
nitrogen. All solvents were purified by standard methods and freshly
distilled prior to use. NMR spectra were recorded on 200-AC and 400-
DPX Bruker instruments. The syntheses of 1-3 are described in detail
elsewhere.9 The content of solvent of crystallization has been checked
by thermal treatment of the sample followed by a gas chromatography
analysis.
Synthesis of 4. Complex 1‚C6H6 (5.29 g, 6.06 mmol) was suspended
in benzene (150 mL), and the N2 was replaced with CO at room
temperature. The mixture became colorless in a few minutes. After 1
h, volatiles were removed in Vacuo and hexane (50 mL) was added to
the residue. The flask was kept at -25 °C for 14 h, and then the white
4 was collected at -20 °C and dried in Vacuo (4.18 g, 83.7%). 1H
NMR (C6D6, 298 K): δ ) 7.25 (d, J ) 2.5 Hz, 2H, ArH), 7.21 (d, J
) 2.5 Hz, 2H, ArH), 6.89 (s, 2H, ArH), 6.81 (s, 2H, ArH), 4.45 (d, J
) 12.5 Hz, 2H, endo-CH2), overlapping with 4.43 (s, 3H, MeO),
overlapping with 4.42 (d, J ) 12.5 Hz, 2H, endo-CH2), 4.20 (s, 3H,
MeO), 3.27 (d, J ) 12.5 Hz, 2H, exo-CH2), 3.15 (d, J ) 12.5 Hz, 2H,
exo-CH2), 2.11 (s, 6H, Me), 1.45 (s, 18H, But), 0.77 (s, 9H, But),
overlapping with 0.74 (s, 9H, But). 1H NMR (CD2Cl2, 298 K): δ 7.14
(d, J ) 2.5 Hz, 2H, ArH), 7.12 (d, J ) 2.5 Hz, 2H, ArH), 7.08 (s, 2H,
ArH), 7.04 (s, 2H, ArH), 4.66 (s, 3H, MeO), 4.62 (s, 3H, MeO), 4.41
(d, J ) 12.4 Hz, 2H, endo-CH2), 4.35 (d, J ) 12.4 Hz, 2H, endo-
CH2), 3.41 (d, J ) 12.4 Hz, 2H, exo-CH2), 3.35 (d, J ) 12.4 Hz, 2H,
Synthesis of 6. An ampule containing 3 (0.955 g, 1.01 mmol) was
suspended in toluene (160 mL). The flask was saturated with CO.
At 15 °C and at a pressure of 1.002 atm, the ampule was broken;
24.7 mL (1.04 mmol) of CO were absorbed during the reaction
(complete in 5 min). The glass was filtered off, and the volatiles were
removed in Vacuo. n-Hexane was added to the residue, and the resulting
solution was kept at -24 °C overnight. Compound 6 precipitated as
a light yellow powder, which was collected and dried in Vacuo (0.44
g, 45%). 1H NMR (C6D6, 298 K): δ ) 7.96 (d, J ) 8.2 Hz, 4H,
ArH), 7.18 (d, 4H, ArH), overlapping with 7.15 (s, 4H, ArH), 6.79 (s,
4H, ArH), 4.47 (s, 3H, MeO), overlapping with 4.43 (d, 4H, endo-
CH2), 4.32 (s, 3H, MeO), 3.15 (d, J ) 12.4 Hz, 4H, exo-CH2), 2.17 (s,
6H, CH3), 1.39 (s, 18H, But), 0.75 (s, 9H, But), 0.68 (s, 9H, But). Anal.
Calcd for 6, C61H72O5Zr: C, 75.03; H, 7.43. Found: C, 74.99; H,
7.78.
Synthesis of 7. ButNC (0.61 g, 7.34 mmol) was added at room
temperature to a solution of 4 (2.95 g, 3.57 mmol) in toluene (100
mL), and the resulting yellow solution was stirred for 7 h. Volatiles
were removed in Vacuo until near dryness, and hexane (40 mL) was
added to the residue. The flask was kept for 2 days at -24 °C; the
pale yellow 7‚C7H8‚(C6H14)0.5 was then collected and dried in Vacuo
(1.2 g, 30%). 1H NMR (CD2Cl2, 298 K): δ ) 7.35-7.0 (m, 13H,
ArH, C7H8), 4.46 (s, 6H, MeO), 4.37 (d, J ) 12.4 Hz, 1H, CH2),
overlapping with 4.34 (d, J ) 12.4 Hz, 1H, CH2), 4.25 (d, J ) 12.8
Hz, 1H, CH2), overlapping with 4.23 (d, J ) 12.8 Hz, 1H, CH2), 3.42-
3.3 (m, 4H, CH2), 2.38 (s, 3H, C7H8), 1.56 (s, 3H, Me), 1.54 (s, 9H,
But), 1.46 (s, 3H, Me), 1.32 (s, 9H, But), overlapping with 1.31 (s,
13H, But), 1.29 (s, 9H, But), 1.18 (s, 18H, But), 0.93 (m, 3H, hexane).
IR: (Nujol mull) 1973, 1952 cm-1 (s); (toluene solution) 1958 cm-1
(s). Anal. Calcd for 7‚C7H8‚(C6H14)0.5, C69H97O5N2Zr: C, 73.62; H,
8.68; N, 2.49. Found: C, 73.37; H, 8.83; N, 2.30.
(6) (a) Negishi, E.-I.; Takahashi, T. Acc. Chem. Res. 1994, 27, 124-
130 and references therein. (b) Erker, G.; Pfaff, R. Organometallics 1993,
12, 1921-1926. (c) Reetz, M. T. In Organometallics in Synthesis; Schlosser,
M., Ed.; Wiley: New York, 1994; Chapter 3. (d) Negishi, E.-I. In
ComprehensiVe Organic Synthesis; Paquette, L. A., Ed.; Pergamon: Oxford,
1991; Vol. 5, p 1163. (e) Schore, N. E. In ComprehensiVe Organic Synthesis;
Paquette, L. A., Ed.; Pergamon: Oxford, 1991; Vol. 5, p 1037. (f) Buchwald,
S. L.; Nielsen, R. B. Chem. ReV. 1988, 88, 1047-1058. (g) Cardin, D. J.;
Lappert, M. F.; Raston, C. L. Chemistry of Organozirconium and Hafnium
Compounds; Wiley: New York, 1986.
(7) Coates, G. W.; Waymouth, R. M. In ComprehensiVe Organometallic
Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.;
Pergamon: Oxford, 1995; Vol. 12, Chapter 12.1. Ryan, E. J. In Compre-
hensiVe Organometallic Chemistry II; Abel, E. W., Stone, F. G. A.,
Wilkinson, G., Eds.; Pergamon: Oxford, 1995; Vol. 4, Chapter 9. Jordan,
R. F.; Lapointe, R. E.; Bradley, P. K.; Baenzinger, N. Organometallics 1989,
8, 2892 and references therein. Buchwald, S. L.; La Maire, S. J.; Nielsen,
R. B.; Watson, B. T.; King, S. M. Tetrahedron Lett. 1987, 28, 3895.
Neghishi, E.; Takahashi, T. Aldrichim. Acta 1985, 18, 31. Neghishi, E.;
Miller, J. A.; Yoshida, T. Tetrahedron Lett. 1984, 25, 3407. Neghishi, E.;
Yoshida, T. Tetrahedron Lett. 1980, 21, 1501. Schwartz, J. Pure Appl. Chem.
1980, 52, 733. Carr, D. B.; Yoshifuji, M.; Shoer, L. I.; Gell, K. I.; Schwartz,
J. Ann. N.Y. Acad. Sci. 1977, 295, 127. Schwartz, J.; Labinger, J. A. Angew.
Chem., Int. Ed. Engl. 1976, 15, 333.
Synthesis of 8. Compound 3 (5.46 g, 5.76 mmol) was dissolved in
benzene (150 mL), and the N2 atmosphere was replaced with CO. The
mixture was stirred overnight in the dark to give a light yellow solution.
CO was then replaced with N2, and a solution of Ph2CO (1.04 g, 5.70
mmol) in benzene (150 mL) added dropwise at room temperature, under
stirring, to the reaction mixture. Volatiles were removed in Vacuo,
n-hexane was added to the residue, and the resulting solution was kept
at -24 °C overnight. It was then frozen (liquid N2) and allowed to
warm to room temperature. After 4 h, 8‚(C6H14)0.5‚(C6H6)0.5 precipitated
as a white powder, which was collected and dried in Vacuo (4.11 g,
58%). 1H NMR (CD2Cl2, 298 K): δ 7.64 (m, 4H, ArH), 7.43 (d, J )
9.1 Hz, 4H, ArH), 7.36 (s, 3H, benzene), 7.17 (m, 6H, ArH), 7.09 (s,
(8) Durfee, L. D.; Rothwell, I. P. Chem. ReV. 1988, 88, 1059-1079 and
references therein.
(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.

ReactiVity on a Calix[4]arene Oxo Surface
J. Am. Chem. Soc., Vol. 119, No. 41, 1997 9711
4H, ArH), 6.97 (s, 4H, ArH), overlapping with 6.95 (d, J ) 9.1 Hz,
4H, ArH), 4.29 (d, J ) 12,6 Hz, 4H, endo-CH2), 3.90 (s, 6H, MeO),
3.25 (d, J ) 12.6 Hz, 4H, exo-CH2), 2.25 (s, 6H, CH3), 1.27 (brd,
22H, But, hexane), 1.10 (s, 18H, But), 0.89 (m, 3H, hexane). Anal.
Calcd for 8‚(C6H14)0.5‚(C6H6)0.5, C80H92O6Zr: C, 77.44; H, 7.47.
Found: C, 77.39; H, 7.47.
Found: C, 75.82; H, 8.20; N, 1.13. The symmetry of the spectrum
does not change on lowering the temperature to 183 K (broadening
only was observed). Crystals suitable for X-ray analysis were obtained
at 9 °C from a toluene solution saturated at room temperature. The
product is thermally stable (12 h, 55 °C), is decomposed in solution
by the action of solar light, and reacts further with ButNC to give a
complex mixture of products.
Synthesis of 9. (Ph)2CO (0.40 g, 2.2 mmol) was added at room
temperature to a solution of 13 (1.11g, 1.08 mmol) in toluene (100
mL). A sample of the solution (ca. 1 mL) was evaporated to dryness
in Vacuo, and the residue was taken up in C6D6. The 1H NMR spectrum
of the resulting solution showed the presence of a few extra peaks with
respect to that of the isolated product (δ ) 7.7 (m, 4H), 7.3-7.05 (m),
overlapping with toluene and product peaks, 3.51 (s, 2H), 3.40 (s, 2H),
1.34 (s, 9H)). The same signals were observed for the reaction of 13
with PhCOCOPh, leading to 11. Volatiles were removed in Vacuo,
hexane was added to the residue, and the white 9 collected and dried
in Vacuo (0.40 g, 33%). 1H NMR (C6D6, 298 K): δ 7.96 (m, 8H,
ArH, Ph), 7.26 (s, 4H, ArH), 7.09 (m, 8H, ArH, Ph), 6.98 (m, 4H,
ArH, Ph), 6.84 (s, 4H, ArH), 4.50 (d, J ) 12.4 Hz, 4H, endo-CH2),
3.95 (s, 6H, MeO), 3.22 (d, J ) 12.4 Hz, 4H, exo-CH2), 1.45 (s, 18H,
But), 0.80 (s, 18H, But). Anal. Calcd for 9, C72H78O6Zr: C, 76.49;
H, 6.95. Found: C, 76.32; H, 7.24.
Synthesis of 14. A solution of ButNC (1.45 g, 17.5 mmol) in
benzene (65 mL) was added dropwise at room temperature to a
suspension of 1‚C6H6 (4.57 g, 5.23 mmol) in benzene (80 mL), and
the resulting yellow solution was stirred for 2 h. Volatiles were
removed in Vacuo, and hexane (50 mL) was added to the residue.
Yellow 14‚C6H6 was then collected and dried in Vacuo (3.5 g, 59%).
1H NMR (C6D6, 298 K): δ 7.37-7.30 (m, 4H, ArH), 6.98-6.89 (m,
4H, ArH), 5.30 (d, J ) 12.4 Hz, 1H, CH2), 4.93 (d, J ) 12.0 Hz, 1H,
CH2), 4.74 (d, J ) 12.4 Hz, 1H, CH2), 4.64 (d, J ) 12.0 Hz, 1H, CH2),
4.37 (s, 3H, MeO), 4.01 (s, 3H, MeO), 3.46-3.26 (m, 4H, CH2), 1.71
(s, 3H, Me), 1.63 (s, 3H, Me), 1.60 (s, 9H, But), 1.52 (s, 9H, But), 1.48
(s, 9H, But), 1.43 (s, 9H, But), 1.24 (s, 9H, But), 0.78 (s, 18H, But). 1H
NMR (CD2Cl2, 298 K): δ 7.37 (s, 6H, C6H6), 7.08-6.98 (m, 8H,
ArH), 4.98 (d, J ) 12.4 Hz, 1H, CH2), 4.64 (d, J ) 12.0 Hz, 1H,
CH2), 4.53 (s, 3H, MeO), overlapping with 4.48 (d, J ) 12.4 Hz, 1H,
CH2), 4.42 (d, J ) 12.0 Hz, 1H, CH2), 4.35 (s, 3H, MeO), 3.32 (d, J
) 12.0 Hz, 1H, CH2), 3.25 (d, 1H, CH2), overlapping with 3.23 (d,
1H, CH2), 3.05 (d, J ) 12.0 Hz, 1H, CH2), 1.70 (s, 3H, Me), 1.64 (s,
3H, Me), 1.50 (s, 9H, But), 1.47 (s, 9H, But), 1.39 (s, 9H, But), 1.28 (s,
9H, But), 1.26 (s, 9H, But), 1.19 (s, 9H, But), 1.17 (s, 9H, But). IR:
(Nujol mull) 1965 cm-1 (s). Anal. Calcd for 14‚C6H6, C69H97O4N3-
Zr: C, 73.75; H, 8.70; N, 3.74. Found: C, 73.10; H, 8.55; N, 3.39.
Crystals suitable for X-ray analysis were obtained from CH2Cl2/
hexane solutions kept at -20 °C. The same product can be obtained
by adding 2 equiv (or an excess) of ButNC to a solution of 12 in toluene
or C6D6. It is thermally labile (decomposition visible after 8 h at 60
°C in C6D6).
Synthesis of 15. A solution of ButNC (0.38 g, 4.6 mmol) in toluene
(50 mL) was added dropwise to a solution of 1‚C6H6 (2.0 g, 2.3 mmol)
in toluene (100 mL) at -80 °C. The colorless solution was stirred
overnight, while the temperature slowly increased to -5 °C. Volatiles
were then removed in Vacuo, and n-hexane (40 mL) was added to the
residue. The flask was kept for 1 day at -24 °C; the white 15 was
then collected and dried in Vacuo (1.80 g, 81.8%). 1H NMR (C6D6,
298 K): δ 7.31 (d, J ) 2.4 Hz, 2H, ArH), 7.29 (d, J ) 2.4 Hz, 2H,
ArH), 6.97 (d, J ) 2.4 Hz, 2H, ArH), 6.93 (d, J ) 2.4 Hz, 2H, ArH),
4.72 (d, J ) 11.8 Hz, 2H, endo-CH2), 4.40 (d, J ) 11.8 Hz, 2H, endo-
CH2), 4.12 (s, 6H, MeO), 3.42 (d, 2H, exo-CH2), overlapping with 3.35
(d, 2H, exo-CH2), 2.49 (s, 6H, Me), 1.48 (s, 18H, But), 1.20 (s, 18H,
But), 0.93 (s, 18H, But). 1H NMR (CD2Cl2, 298 K): δ 6.94-6.89 (m,
8H, ArH), 4.41 (d, J ) 11.8 Hz, 2H, endo-CH2), overlapping with 4.34
(s, 6H, MeO), 4.21 (d, J ) 11.8 Hz, 2H, endo-CH2), 3.12 (d, 2H, exo-
CH2), overlapping with 3.10 (d, 2H, exo-CH2), 2.77 (s, 6H, Me), 1.42
(s, 18H, But), 1.21 (s, 18H, But), 1.10 (s, 18H, But). IR: (Nujol mull)
1585.1 cm-1 (m). Anal. Calcd for 15, C58H82N2O4Zr: C, 72.38; H,
8.59; N, 2.91. Found: C, 71.90; H, 8.84; N, 2.78. Crystals suitable
for X-ray analysis were obtained at -25 °C from a saturated solution
in toluene/THF (ca. 3:1). The product does not react with excess But-
NC; it is thermally labile both in solution and in the solid state, evolving
to the coupled product (Vide infra).
Synthesis of 11. A solution of PhCOCOPh (0.37 g, 1.8 mmol) in
toluene (35 mL) was added at room temperature to a solution of
5‚(C6H6)1.5 (1.90 g, 1.74 mmol) in toluene (100 mL), yielding a greenish
solution. Volatiles were removed in Vacuo, n-hexane (40 mL) was
added to the residue, and the resulting solution was kept at -24
°C for 2 days. Complex 11‚(C6H14)1.5 precipitated as an orange
powder, which was collected and dried in Vacuo (1.5 g, 78%). 1H
NMR (C6D6, 298 K): δ 8.02 (m, 4H, ArH, Ph), 7.26-7.0 (m, 10H,
ArH), 6.78 (s, 4H, ArH), 4.40 (d, J ) 12.5 Hz, 4H, endo-CH2), 3.96
(s, 6H, MeO), 3.14 (d, 4H, J ) 12.5 Hz, exo-CH2), 1.45 (s, 18H,
But), 1.23 (m, 12H, hexane), 0.88 (m, 9H, hexane), 0.72 (s, 18H, But).
Anal. Calcd for 11‚(C6H14)1.5, C69H89O6Zr: C, 74.95; H, 8.11.
Found: C, 75.35; H, 7.73. Complex 11 is also obtained through the
reaction of 4 and 13 with PhCOCOPh, as shown by 1H NMR
spectroscopy.
Synthesis of 12. A solution of ButNC (0.74 g, 8.9 mmol) in toluene
(50 mL) was added dropwise, at room temperature, to a solution of
1‚C6H6 (7.85 g, 8.98 mmol) in toluene (200 mL), to give a yellow
solution. After 2 h, volatiles were removed in Vacuo and hexane (50
mL) was added to the residue. The flask was kept 2 days at -24 °C;
the pale yellow 12 was then collected and dried in Vacuo (5.06 g,
64.0%). 1H NMR (C6D6, 298 K): δ 7.21 (s, 4H, ArH), 6.79 (s, 4H,
ArH), 4.43 (d, J ) 12.4 Hz, 4H, endo-CH2), 4.08 (s, 6H, MeO), 3.15
(d, J ) 12.4 Hz, 4H, exo-CH2), 2.14 (s, 6H, Me), 1.74 (s, 9H, NBut),
1.45 (s, 18H, But), 0.86 (s, 18H, But). Anal. Calcd for 12, C53H73O4-
NZr: C, 72.39; H, 8.37; N, 1.59. Found: C, 72.37; H, 8.58; N, 1.50.
The product is thermally stable (12 h, 55 °C) but is decomposed in
solution by solar light.
Synthesis of 13. ButNC (0.38 g, 4.6 mmol) was added at room
temperature to a solution of 2 (4.51 g, 4.75 mmol) in toluene (250
mL). After 2 h, a sample of the solution (ca. 1 mL) was evaporated
1
to dryness in Vacuo and the residue was taken up in C6D6. The H
NMR of the resulting solution showed the presence of the starting
material together with a small amount of the final product. The IR
spectrum of the toluene reaction solution (taken at the same time)
coincided with the sum of the IR spectra in toluene of 2 and free But-
NC. After 3 days, volatiles were removed in Vacuo and n-hexane (50
mL) was added to the residue. The flask was kept 14 h at -24 °C; the
Synthesis of 16. A solution of ButNC (0.49 g, 5.9 mmol) in toluene
(25 mL) was added dropwise to a solution of 3 (2.76 g, 2.91 mmol) in
toluene (100 mL) at -80 °C. The colorless solution was stirred
overnight, while the temperature slowly increased to room temperature.
Volatiles were then removed in Vacuo, and n-hexane (50 mL) was added
to the residue. The flask was kept 1 day at -24 °C; the white 16‚C7H8
was then collected and dried in Vacuo (1.50 g, 42.7%). 1H NMR (C6D6,
298 K): δ 7.42 (d, J ) 2.3 Hz, 2H, ArH), 7.37 (d, J ) 2.3 Hz, 2H,
ArH), 7.15-6.94 (m, ArH), 4.89 (d, J ) 11.9 Hz, 2H, endo-CH2), 4.72
(d, J ) 11.9 Hz, 2H, endo-CH2), 3.98 (s, 6H, MeO), 3.42 (d, 4H, J )
11.9 Hz, exo-CH2), 2.11 (s, 9H, Me, C7H8), 1.49 (s, 18H, But), 1.35 (s,
18H, But), 0.89 (s, 18H, But). 1H NMR (C6D6, 298 K): δ 7.60 (d, J
) 8.0 Hz, 4H, p-MeC6H4), 7.36 (s, 4H, ArH), 6.86 (s, 4H, ArH), 6.79
(d, J ) 8.0 Hz, 4H, p-MeC6H4), 4.84 (d, J ) 12.4 Hz, 4H, endo-CH2),
3.64 (s, 6H, MeO), 3.37 (d, 4H, J ) 12.4 Hz, exo-CH2), 1.97 (s, 6H,
Me), 1.48 (s, 36H, But), 0.94 (s, 18H, But). Anal. Calcd for 16‚C7H8,
1
yellow 13 was then collected and dried in Vacuo (2.63 g, 53.5%). H
NMR (C6D6, 298 K): δ 7.77 (m, 4H, ArH(PhCH2)), 7.24 (s, 4H, ArH),
7.14 (m, 4H, ArH(PhCH2)), 7.00 (m, 2H, ArH(PhCH2)), 6.75 (s, 4H,
ArH), 4.26 (d, J ) 12.4 Hz, 4H, endo-CH2), 3.82 (d, J ) 15.2 Hz, 2H,
PhCH2), 3.73 (s, 6H, MeO), 3.68 (d, J ) 15.2 Hz, 2H, PhCH2), 3.08
(d, J ) 12.4 Hz, 4H, exo-CH2), 1.77 (s, 9H, NBut), 1.45 (s, 18H, But),
0.86 (s, 18H, But).1H NMR (CD2Cl2, 298 K): δ 7.64 (m, 4H, ArH-
(PhCH2)), 7.14-7.04 (m, 10H, ArH), 6.77 (s, 4H, ArH), 4.20 (d, J )
12.4 Hz, 4H, endo-CH2), 4.02 (s, 6H, MeO), 3.67 (d, J ) 15.3 Hz, 2H,
PhCH2), 3.51 (d, J ) 15.3 Hz, 2H, PhCH2), 3.16 (d, J ) 12.4 Hz, 4H,
exo-CH2), 1.70 (s, 9H, NBut), 1.35 (s, 18H, But), 0.96 (s, 18H, But).
Anal. Calcd for 13, C65H81O4NZr: C, 75.68; H, 7.91; N, 1.36.

9712 J. Am. Chem. Soc., Vol. 119, No. 41, 1997
Giannini et al.
Table 1. Experimental Data for the X-ray Diffraction Studies on Crystalline Complexes 5, 13-15, and 17
compound
5
13
14
15
17
formula
cryst color
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
C61H72O5Zr‚2.5C6H6 C65H81NO4Zr‚C7H8 C63H91N3O4Zr‚2CH2Cl2‚0.5C6H14 C58H82N2O4Zr‚2C7H8 C58H92N2O4Zr‚C6H6
colorless
14.292(3)
19.206(4)
12.927(4)
107.61(2)
105.00(2)
88.25(2)
3262.4(15)
2
colorless
18.835(3)
24.950(4)
27.151(4)
90
96.37(1)
90
pale yellow
13.691(3)
20.022(4)
13.384(3)
93.62(2)
105.08(2)
104.44(2)
3398.3(14)
2
colorless
22.126(3)
18.562(2)
17.847(2)
90
114.99(2)
90
yellow
13.887(4)
15.317(5)
14.973(2)
90
110.69(2)
90
12680.4(34)
8
6643.6(18)
4
2979.5(14)
2
fw
1171.7
P1h (No. 2)
22
1.541 78
1.193
17.55
1123.7
1258.6
1146.8
C2/c (No. 15)
-140
1048.7
P21/m (No. 11)
22
0.710 69
1.169
2.24
0.609-1.000
5469
space group
T (°C)
λ (Å)
P21/c (No. 14)
-140
1.541 78
1.177
17.77
0.535-1.000
22129
10381 (NO)
0.066
0.161
P1h (No. 2)
-140
1.541 78
1.230
1.541 78
1.146
F
calcd (g cm-3
)
µ (cm-1
trans coeff
)
31.71
0.726-1.000
12693
5412 (NO)
0.070
0.167
17.06
0.647-1.000
5327
3713 (NO)
0.058
0.139
0.812-1.000
unique total data 8526 (NO)
unique obsd data 3495
4605 (NO)
0.060
0.166
Ra
0.069
0.203
1.106
wR2b
GOFc
1.082
1.118
1.063
1.001
2 2
a R ) ∑|∆F|/∑|Fo| calculated on the unique observed data [I > 2σ(I)]. b wR2 ) [∑w|∆F2|2/∑w|Fo | ]1/2 calculated on the unique observed data
for 13-15 and 17 and on the unique total data for 5. c GOF ) [∑w|∆F2|2/(NO - NV)]1/2 where NO ) number of observed and NV ) number of
variables.
C77H98N2O4Zr: C, 76.63; H, 8.18; N, 2.32. Found: C, 76.31; H, 8.49;
N, 1.93. The product is thermally labile and evolves to the coupled
product 18.
difference Fourier maps and introduced in the subsequent refinements
as fixed atom contributions with isotropic U values fixed at 0.08 Å2.
For complex 14, refinement was done by full-matrix least-squares first
isotropically and then anisotropically for all non-H atoms except for
some disordered carbon atoms. Some troubles were encountered in
refining the structure because the Uij parameters for some But groups
and for the CH2Cl2 molecule of crystallization reached rather high
values. The best fit was obtained considering the C38-C40 methyl
carbons of one calixarene But group distributed over two positions A
(75%) and B (25%) corresponding to different conformations of the
butyl group. For complex 15, refinement was done by full-matrix least-
squares first isotropically and then anisotropically for all non-H atoms
of the complex molecule. The X-ray analysis revealed the presence
of two toluene molecules of crystallization C61‚‚‚C66 and C71‚‚‚C74,
the former as a guest. Both of them were affected by severe disorder,
the guest molecule being disordered about the 2-fold axis running
through the complex molecule, the other one being disordered around
a crystallographic center of symmetry. For complex 17, the number
of molecules in the unit cell (Z ) 2) required the molecules to possess
an imposed crystallographic m symmetry in the centrosymmetric P21/m
space group. The structure was then solved in this space group with
the heavy-atom method. The Fourier map calculated with the contribu-
tion of the zirconium atom showed the calixarene ligand to be consistent
with the presence of the crystallographic mirror implying, by contrast,
a statistical distribution of the N-C-C-N fragment about the mirror
plane running through the N1 and N2 atoms.
Synthesis of 17. Solid 15 (1.18 g) was kept at 60 °C in a sealed
ampule for 6 days. The solid turned ocre. The reaction was then
complete, as shown by 1H NMR. 1H NMR (C6D6, 298 K): δ 7.35 (s,
4H, ArH), 6.92 (s, 4H, ArH), 4.58 (d, J ) 12.2 Hz, 4H, endo-CH2),
3.52 (s, 6H, MeO), 3.36 (d, 4H, J ) 12.2 Hz, exo-CH2), 2.10 (s, 6H,
Me), 1.48 (s, 18H, But), 1.46 (s, 18H, But), 0.81 (s, 18H, But). 1H
NMR (CD2,Cl2, 298 K): δ 7.02 (s, 4H, ArH), 6.98 (s, 4H, ArH),
4.31 (d, J ) 12.2 Hz, 4H, endo-CH2), 3.87 (s, 6H, MeO), 3.17 (d,
J ) 12.2 Hz, 4H, exo-CH2), 2.34 (s, 6H, Me), 1.41 (s, 18H, But), 1.25
(s, 18H, But), 1.12 (s, 18H, But). Anal. Calcd for 16, C58H92O4N2Zr:
C, 72.38; H, 8.59; N, 2.91. Found: C, 72.27; H, 8.80; N, 2.69. In
the low-temperature limit, a C2 symmetric spectrum, although very
broad, was obtained. 1H NMR (CD2Cl2, 183 K): δ 6.97 (ArH), 6.87
(ArH), 4.50 (d, CH2), 3.97 (s, MeO), 3.81 (d, CH2), 3.56 (s, MeO),
3.19 (d, CH2), 3.05 (d, CH2), 2.26 (s, Me), 1.30 (s, But), 1.15 (s, But),
1.05 (s, But), 1.00 (s, But). Crystals suitable for X-ray analysis
were obtained at room temperature from a supersaturated benzene
solution.
X-ray Crystallography for 5, 13-15, and 17. Crystal data and
details associated with data collection are listed in Tables 1 and S1
(Supporting Information). Data for all complexes were collected on a
single crystal (Rigaku AFC6S) at 295 K for 5 and 17 and at 133 K for
13-15. Structure solutions were based on the observed reflections [I
> 2σ(I)]. The refinements were carried out using the observed
reflections for 13-15 and 17 and on the unique total reflections for 5.
The structures were solved by the heavy-atom method starting from a
three-dimensional Patterson map.10 Refinements were done by full-
matrix minimization of the function ∑w(∆F2)2.11 In the last stage of
The final difference maps showed no unusual feature, with residual
peaks of 0.34, 1.50, 0.77, 0.66, and 0.59 eÅ-3 for 5, 13-15, and 17,
respectively, located along the direction of Zr-O bonds. See the
Supporting Information for additional details.
2
2
refinement, the weighting scheme w ) 1/[σ2(Fo ) + (aP)2] (P ) (Fo
Results
+ 2Fc2)/3] was applied with a resulting in the value of 0.0781, 0.0954,
0.0779, 0.0561, and 0.1263 for 5, 13, 14, 15, and 17, respectively. For
complex 5, refinement was carried out first isotropically and then
anisotropically for all non-H atoms. The O5, C48, and C49 atoms of
the dibenzyl ketone moiety and the C81-C86 benzene solvent molecule
of crystallization were found to be statistically distributed over two
positions (A and B), anisotropically refined with site occupation factors
of 0.6 and 0.4, respectively. For complex 13, refinement was done by
full-matrix least-squares first isotropically and then anisotropically for
all of the non-H atoms. All hydrogen atoms were located from
Synthesis and Reactivity. This paper concerns a variety of
C-C bond formation reactions assisted by zirconium bonded
to a dianionic tetraoxo macrocyclic ligand, i.e., [p-But-calix[4]-
(OMe)2(O)2Zr]2+, hereafter abbreviated as [calix[4]-(OMe)2-
(O)2Zr]2+
.
(10) Sheldrick, G. M. SHELX76: Program for Crystal Structure De-
termination; University of Cambridge: Cambridge, U.K., 1976.
(11) Sheldrick, G. SHELXL92: Program for Crystal Structure Refine-
ment; University of Go¨ttingen: Go¨ttingen, Germany, 1992.

ReactiVity on a Calix[4]arene Oxo Surface
J. Am. Chem. Soc., Vol. 119, No. 41, 1997 9713
Scheme 1
explained by the sequence shown in Scheme 1. The insertion
of ButNC on 4 produces a carbenoid metallacycle which then
adds the pseudocarbene ButNC.14 The proposed structure for
7 matches all of the analytical and spectroscopic data and is, in
addition, supported by the X-ray analysis on the analogous
compound 14 (Vide infra) derived from the insertion of ButNC
into the η2-imine derivative 12. The reaction of η2-ketone
species with a free ketone to form a C-C-coupled diolate,
although it is an easy reaction, usually produces a mixture
containing the three possible diolate forms, 8-10, derived from
the following: (i) the expected cross-coupling reaction, 8; (ii)
the coupling between two incoming ketones, with the contem-
porary displacement of the previously η2-coordinated ketone
9; (iii) the coupling between two originally η2-metal-bonded
ketones 10. The obtention of such a mixture requires that the
η2-metal-bonded ketone is replaced at an early stage by an
incoming ketone, thus both reacting species [Zr(η2-R2CO)] and
[Zr(η2-R′2CO)] can produce the observed mixture. In a single
case, we were able to isolate the mixed diolato form 8 (R )
p-MeC6H4) as the unique compound. Significant proof of the
displacement of the η2-metal-bonded ketone is found in the
reaction of 4 and 5 with PhCOCOPh to give 11. In this case,
the displacement is followed by a two-electron internal reductive
coupling leading to a CdC double bond.
We should emphasize at this stage two results having rare
precedents in the literature: (i) the reactivity of the metal-
carbon bond in the η2-metal-bonded ketones 4-6 toward
inserting groups,13 a reaction which can be of synthetic utility;
(ii) the potential use of 4-6 as a source of zirconium(II) in
displacement reactions with appropriate substrates.
In the absence of an excessive driving force derived, as in
the case of CO, from the oxophilicity of the metal, the reaction
with ButNC allowed us to better single out almost all of the
steps of the migratory insertion of the isocyanide into the Zr-C
bonds, as shown in Scheme 2.
The reaction of 1-3 with ButNC exemplifies the migratory
aptitude of the alkyl and aryl groups bonded to zirconium. A
complete range of migratory insertion pathways have been
observed as a function of the temperature and the nature of R,
though only for 1 were we able to select a specific pathway
using exclusively temperature as the controlling factor.
The reaction of 1 and 2 with ButNC at room temperature
using a 1:1 Zr/ButNC molar ratio led to the η2-imine complexes
12 and 13 Via the intermediacy of the corresponding η2-
iminoacyl.4,8,15 The migration of the second alkyl group to the
η2-iminoacyl carbon was inferred for 1 and 2 Via the isolation
of the corresponding η2-imine, while for 3 the corresponding
imine occurs with other products (see below). The metal-
carbon bond in 12 is still reactive in migration insertion
reactions, similar to those observed for the η2-metal-bonded
ketones in 4-6. The reaction of 12 with an excess of ButNC
or, more simply, the reaction of 1 with 3 mol of ButNC at room
temperature led to 14. This reaction implies the preliminary
insertion of an isocyanide into the Zr-C bond of the Zr imine
The starting materials of this study are the dialkyl and diaryl
derivatives, [calix[4]-(OMe)2(O)2ZrR2], whose syntheses and
main properties have been reported elsewhere.9 Such dialkyl
and diaryl derivatives undergo a sequential series of insertion
reactions occurring with the formation of C-C bonds and giving
organic fragments that remain bonded to the calix[4]arene-
zirconium moiety. Scheme 1 displays the migratory insertion
of carbon monoxide into the Zr-C bonds in 1-3 and the further
migratory aptitude of the resulting Zr-C functionalities to
ketones and isocyanides.
The reaction of 1-3 with CO was carried out in benzene at
room temperature under 1 atm pressure and was completed in
a few minutes. Regardless of the R substituent and the reaction
conditions, the η2-ketones were the only products detected in
solution and then isolated as solids. They have been character-
ized by the conventional analytical methods, including the X-ray
analysis on 5 (Vide infra). The formation of the η2-ketones 4-6
probably occurs Via the intermediacy of an η2-acyl, followed
by the migration of the second alkyl group to the η2-acyl
carbon.4 Such a migration is greatly enhanced by the carbenium
ion nature of the η2-acyl.12 This corresponds to a low-energy
vacant π*CO-accepting orbital. In the present case, unlike when
the [ZrR2] fragment is bonded to the cyclopentadienyl ligand,
we were unable to detect the η2-acyl intermediate.
The higher lability of the Zr-C bond can be associated to
the unprecedented, in this kind of complex, six-coordination of
the metal (usually it is tetrahedral) and on the topology of the
[ZrO4] fragment. The metal, in such a hemisandwich structure,
has, in fact, an open face where it assists the organic transfor-
mations of the incoming substrates.
The η2-metal-bonded ketones 4-6 maintain some of the
migratory insertion properties typical of the metal-carbon
σ-bond.13 Although no reaction with CO was observed, 4 reacts
smoothly with 2 mol of ButNC. The formation of 7 can be
(13) (a) Smuck, S.; Erker, G.; Kotila, S. J. Organomet. Chem. 1995,
502, 75-86. (b) Bendix, M.; Grehl, M.; Fro¨hlich, K.; Erker, G. Organo-
metallics 1994, 13, 3366-3369. (c) Erker, G.; Mena, M.; Kru¨ger, C.; Noe,
R. Organometallics 1991, 10, 1201. (d) Erker, G.; Dorf, U.; Czisch, P.;
Petersen, J. L. Organometallics 1986, 5, 668. Erker, G.; Czisch, P.; Schlund,
R.; Angermund, K.; Kru¨ger, C. Angew. Chem., Int. Ed. Engl. 1986, 15,
364.
(14) (a) Valero, C.; Grehl, M.; Wingbermu¨hle, D.; Kloppenburg, L.;
Carpenetti, D.; Erker, G.; Petersen, J. Organometallics 1994, 13, 415. (b)
Scott, M. J.; Lippard, S. J. J. Am. Chem. Soc. 1997, 119, 3411. (c) Ruiz, J.;
Vivanco, M.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Organometallics
1993, 12, 1811-1822 and references therein.
(15) Chamberlain, L. R.; Durfee, L. D.; Fanwick, P. E.; Kobriger, L.;
Latesky, S. L.; McMullen, A. K.; Rothwell, I. P.; Folting, K.; Huffman, J.
C.; Streib, W. E.; Wang, R. J. Am. Chem. Soc. 1987, 109, 390-402.
(12) (a) Fanwick, P. E.; Kobriger, L. M.; McMullen, A. K.; Rothwell, I.
P. J. Am. Chem. Soc. 1986, 108, 8095. (b) Arnold, J.; Tilley, T. D.;
Rheingold, A. L. J. Am. Chem. Soc. 1986, 108, 5355. (c) Tatsumi, K.;
Nakamura, A.; Hofmann, P.; Stauffert, P.; Hoffmann, R. J. Am. Chem. Soc.
1985, 107, 4440. (d) Martin, B. D.; Matchett, S. A.; Norton, J. R.; Anderson,
O. P. J. Am. Chem. Soc. 1985, 107, 7952.

9714 J. Am. Chem. Soc., Vol. 119, No. 41, 1997
Giannini et al.
Scheme 2
reported in the Experimental Section. The X-ray analysis has
been performed on the prototype of each class, namely, 5, 13-
15, and 17.
(a) NMR Analysis. The NMR structural studies have mainly
focused on the possibility of stressing a solid state-solution
structural relationship and on the attempt to use parts of the
calix[4]arene, namely, the methylenes, the methyls from the
methoxy, or the p-But groups, as spectroscopic probes for
inspecting the symmetry and the bonding mode of the organo-
metallic fragments bound to the lower rim of the calix. A
guideline in this context will be discovering that we can shape
the calix[4]arene cavity using the changes in the coordination
number of the metal. In Table 2 the chemical shifts of the
various fragments of the calix[4]arene skeleton are listed
according to the organic moiety bonded to the metal on the
open face of the molecule, dedicated to the metal-assisted
reactivity. The chemical shift patterns for But, MeO, and
bridging methylenes reveal the symmetry of the metal-calix-
[4]arene moiety as a function of the fragment attached to the
metal: (i) 2 singlets (But), 1 pair of doublets (>CH2), 1 singlet
(MeO) for a C2V symmetry; (ii) 2 singlets (But), 2 pairs of
doublets (>CH2), 1 singlet (MeO) for a C2 symmetry; (iii) 3
singlets (But), 2 pairs of doublets (>CH2), 1 or 2 singlets (MeO)
for a Cs symmetry; (iv) 4 singlets (But), 4 pairs of doublets
(>CH2), 2 singlets (MeO) for a C1 symmetry. For 7 and 14-
16, the symmetry of the spectra coincides with that revealed
by the X-ray analysis. For complexes 1-6, 8-11, 17, and 18,
the symmetry observed in solution is slightly higher than the
crystallographic one and corresponds to an “idealized structure”.
The discrepancy can be ascribed to the absence in solution of
solid state effects and to a modest degree of fluxionality. On
the other hand, complexes 12 and 13 show an apparent C2V
symmetry (which is impossible for any conformation reasonably
accessible by these complexes), revealing an extreme fluxion-
ality that cannot be “frozen” even at low temperature (193 K).
(b) X-ray Analysis. Complex 5 (Figure 1) crystallizes with
2.5 molecules of benzene, one of them hosted in the calix[4]-
arene cavity. Complex 13 (Figure 2) crystallizes with one
molecule of toluene out of the cavity. In the unit cell of 13,
there are two crystallographically independent molecules (A and
B), which have very close parameters. The values referred to
molecule B are given in Table 3, but not mentioned in the text.
Complex 14 (Figure 3) crystallizes with CH2Cl2 and n-hexane
in the 1:2:0.5 molar ratio, and one of the CH2Cl2 molecules is
hosted in the calix[4]arene cavity. Complex 15 (Figure 4)
crystallizes with 2 molecules of toluene, one of them hosted in
the calix[4]arene cavity. Complex 17 (Figure 5) crystallizes
with 1 molecule of benzene as a guest. For reasons of
conciseness and clarity, the overall structures have been
examined according to common and noncommon fragments.
12 to form the carbenoid intermediate, which then adds an
additional mole of isocyanide to give 14.14 Attempts to insert
a ketone into the zirconium-carbon σ bond of the imine 12
resulted, instead, in the displacement of the imine by the ketone,
as a consequence of the high oxophilicity of zirconium. The
resulting η2-metal-bonded ketone readily inserts an additional
molecule of ketone, thus forming the C-C-coupled product 9.
The proposed displacement of the η2-metal-bonded imine16 in
12 and 13 is supported by the identification of the free imine
and by the reaction of 13 with PhCOCOPh, leading, as for 4
and 5, to 11. These results suggest the possible synthetic utility
of 12 and 13 as a source of the [calix[4]-(OMe)2(O)2Zr]0
fragment, formally containing a zirconium(II), more available
than 4 and 5, not containing strongly bonded oxygen donor
atoms.
The competitive pathway that has been singled out at low
temperature for 1 and 3 (see the Experimental Section) shows
the preferential migration of the second alkyl group to an
incoming ButNC,15 to give the bis-η2-iminoacyls 15 and 16.
Regardless of the Zr/ButNC stoichiometric ratio, the reaction
proceeds Via pathway b at low temperature for 1 and 3, while
for 2, which is unreactive at low temperature, the only
compound formed at room temperature, Via pathway a, is 13.
For the p-tolyl substituent, pathway b is preferred, since we
observe only very limited formation of the η2-imine even under
forced conditions. Gentle heating (60 °C) induces the intramo-
lecular coupling17 of the two η2-iminoacyl derivatives 15 and
16 to the corresponding enediamido complexes 17 and 18. The
rate, the mechanism, and the theoretical treatment of such an
intramolecular coupling of the η2-iminoacyls 15 and 16 will be
discussed in the next section.
(i) The Calix[4]arenezirconium Moiety. In all of the
complexes 5, 13-15, and 17 (Figures 1-5 and Tables 3 and
4), the hexacoordination of the metal and the differences in the
Zr-O bond distances and angles greatly distort the O4 core from
the planarity, typical of the cone conformation of the calix[4]-
arene. In particular, the rather short Zr-O1 and Zr-O3 bond
distances along with the wide bond angles that are close to
linear, Zr-O1-C1 and Zr-O3-C10 reveal a Zr-O multiple
bond,15,17,18 while zirconium interacts with O2 and O4 via a
normal single bond. The calixarene ligand assumes a slightly
flattened cone conformation with the opposite A and C rings
pushed outward and the opposite B and D rings toward the
cavity of the macrocycle, as indicated by the dihedral angles
they form with the “reference plane” through the methylene
Structural Studies. Details of the synthesis and character-
ization of all of the compounds listed in Schemes 1 and 2 are
(16) (a) Durfee, L. D.; Hill, J. E.; Kerschner, J. L.; Fanwick, P. E.;
Rothwell, I. P. Inorg. Chem. 1989, 28, 3095-6. (b) Durfee, L. D.; Fanwick,
P. E.; Rothwell, I. P. Folting, K.; Huffmann, J. C. J. Am. Chem. Soc. 1987,
109, 4720-2.
(17) Chamberlain, L. R.; Durfee, L. D.; Fanwick, P. E.; Kobriger, L.
M.; Latesky, S. L.; McMullen, A. K.; Steffey, B. D.; Rothwell, I. P.; Folting,
K.; Huffman, J. C. J. Am. Chem. Soc. 1987, 109, 6068-6076.
(18) Latesky, S. L.; McMullen, A. K.; Niccolai, G. P.; Rothwell, I. P.
Organometallics 1985, 4, 902.

ReactiVity on a Calix[4]arene Oxo Surface
J. Am. Chem. Soc., Vol. 119, No. 41, 1997 9715
Table 2. Chemical Shifts of 1-18 in C6D6 at 298 K
cmpd
But
MeO
exo-CH2
endo-CH2
∆CH2
J (Hz)
sym
1
2
3
4
5
6
7
8
9
11
12
13
14
15
16
17
18
1.46, 0.77
1.45, 0.70
1.44, 0.84
1.45, 0.74, 0.77
1.44, 0.72, 0.74
1.39, 0.75, 0.68
1.44, 0.69, 1.46, 0.73
1.10, 1.27
1.45, 0.80
1.45, 0.72
1.45, 0.86
1.45, 0.86
1.48, 0.78, 1.43
1.48, 0.93
1.49, 0.89
1.48, 0.81
1.48, 0.94
3.69
3.37
3.35
4.43, 4.20
4.45, 4.03
4.47, 4.32
4.11, 4.0
3.90
3.95
3.96
4.08
3.73
4.37, 4.01
4.12
3.98
3.52
3.64
3.18
3.24
3.19
3.15, 3.27
3.19, 3.18
3.15
4.46
4.46
4.60
4.42, 4.45
4.22
4.43
4.4-4.6
4.29
4.50
4.40
4.43
4.26
1.28
1.22
1.41
(1.22)
1.03
1.32
12.5
12.5
12.5
12.5
12.4
12.4
C2V
C2V
C2V
Cs
Cs
Cs
∼3.4-3.2
∼1.2
∼12.5
C1
3.25
1.04
12.6
C2V
C2V
C2V
C2V
C2V
C1
C2
C2
C2V
C2V
3.22
3.14
3.15
3.08
3.46-3.26
3.35, 3.42
3.42
3.36
3.37
1.28
1.26
1.28
1.18
2-1.3
(1.27)
1.43
1.22
1.47
12.4
12.5
12.4
12.4
12.0, 12.4
11.8
11.9
12.2
12.4
4.64, 4.93, 4.74, 5.3
4.40, 4.72
4.72, 4.89
4.58
4.84
Table 3. Selected Bond Distances (Å) and Angles (deg) for
Complexes 5 and 13
Complex 5
Zr1-O1
Zr1-O2
Zr1-O3
Zr1-O4
1.966(6)
2.323(8)
1.966(6)
2.314(7)
Zr1-O5A
Zr1-C47
O5A-C47
2.084(11)
2.210(12)
1.32(2)
O5A-Zr1-C47
O2-Zr1-O4
O1-Zr1-O3
Zr1-O1-C1
Zr1-O2-C13
Zr1-O3-C20
35.5(4)
153.1(3)
120.2(3)
171.6(6)
119.3(6)
159.0(6)
Zr1-O4-C27
Zr1-O5A-C47
Zr1-O5B-C47
Zr1-C47-O5B
Zr1-C47-O5A
118.2(5)
77.5(8)
77.4(9)
69.4(9)
67.0(7)
Complex 13
molecule A
molecule B
Zr1-O1
Zr1-O2
Zr1-O3
Zr1-O4
Zr1-N1
Zr1-C51
N1-C47
N1-C51
1.967(6)
2.355(5)
2.020(5)
2.466(6)
2.048(7)
2.261(8)
1.482(12)
1.434(11)
1.960(6)
2.339(5)
2.026(5)
2.527(6)
2.043(7)
2.281(8)
1.467(11)
1.425(9)
Figure 1. A SCHAKAL view of complex 5. Disorder has been omitted
for clarity.
N1-Zr1-C51
O2-Zr1-O4
O1-Zr1-O3
Zr1-O1-C1
Zr1-O2-C13
Zr1-O3-C20
Zr1-O4-C27
Zr1-N1-C51
C47-N1-C51
Zr1-C51-N1
38.5(3)
150.3(2)
120.4(2)
174.6(5)
115.0(4)
162.0(5)
110.2(4)
78.8(4)
38.0(3)
150.1(2)
120.2(2)
173.9(5)
114.6(4)
163.0(5)
109.5(4)
80.1(4)
127.5(7)
62.7(4)
127.5(7)
61.9(4)
interaction. The guest molecule is oriented nearly parallel to
the B and D rings (dihedral angles 19.9(6) and 18.6(6)°,
respectively) and perpendicular to the A and C rings (dihedral
angles 87.0(5) and 87.6(6)°, respectively). The host-guest
interactions could be interpreted in terms of CH3/π interac-
tions19 occurring between the protons of the methyl groups of
the host and the guest aromatic ring, as is inferred from the
following distances: C34‚‚‚C75, 3.82(3) Å; H342‚‚‚C75, 2.70
Å; C35‚‚‚C73, 3.92(3) Å; H352‚‚‚C73, 2.69 Å; C42‚‚‚C75,
3.73(3) Å; H422‚‚‚C75, 2.58 Å; C43‚‚‚C73, 3.80(4) Å;
H432‚‚‚C73, 2.67 Å.
Figure 2. A SCHAKAL view of molecule A of complex 13.
carbon atoms C7, C14, C21, and C28 (Scheme 3, Table 5a).
The elliptical section of the cavity is indicated by the distances
between opposite p-carbon atoms (Table 5b). The most
pronounced ellipticity of the cavity is observed for complex
13, as evidenced by the short C10‚‚‚C24 distance [5.740(10)
Å] and the high B∧D angle (Table 5). As already observed in
similar cases,9 solvent inclusion is impossible in such an
elliptical cavity, and in fact, complex 13 is the only one to
crystallize without a guest molecule.
(ii) The Guest Solvent. Complex 5. The C71‚‚‚C76 guest
benzene molecule enters the cavity of the complex molecule
with its C71‚‚‚C74 axis along the molecular axis, as indi-
cated by the value of the Zr‚‚‚C71‚‚‚C74 angle (173.3(8)°). The
Zr‚‚‚C71 distance of 5.03(2) Å rules out any possible Zr‚‚‚H
Complex 14. One CH2Cl2 solvent molecule is hosted in the
calixarene cavity. It is oriented nearly parallel to the B and D
rings (dihedral angles 22.4(7) and 25.2(5)°, respectively) and
perpendicular to the A and C rings (dihedral angles 88.1(7) and
(19) Andreetti, G. D.; Ugozzoli, F. In ref 1b, pp 88-122.

9716 J. Am. Chem. Soc., Vol. 119, No. 41, 1997
Giannini et al.
Table 4. Selected Bond Distances (Å) and Angles (deg) for
Complexes 14, 15, and 17
Complex 14
2.014(7) N1-C51
Zr1-O1
Zr1-O2
Zr1-O3
Zr1-O4
Zr1-N1
Zr1-N2
1.488(13)
1.443(11)
1.227(14)
1.526(16)
1.324(15)
2.373(6)
2.025(8)
2.364(5)
2.154(9)
2.159(9)
N2-C52
N3-C53
C51-C52
C52-C53
N1-Zr1-N2
O2-Zr1-O4
O1-Zr1-O3
Zr1-O1-C1
Zr1-O2-C13
Zr1-O4-C27
Zr1-N1-C51
82.1(3)
146.2(3)
95.3(3)
178.3(6)
121.6(5)
122.1(5)
107.1(6)
Zr1-N2-C52
C53-N3-C64
N3-C53-C52
Zr1-O3-C20
N1-C51-C52
N2-C52-C51
C51-C52-C53
N2-C52-C53
83.3(5)
130.0(10)
175.5(11)
171.0(6)
100.0(8)
117.5(8)
120.4(9)
121.6(10)
Complex 15
Figure 3. A SCHAKAL view of complex 14. Disorder has been
omitted for clarity.
Zr1-O1
Zr1-O2
Zr1-N1
2.054(4)
2.450(5)
2.249(4)
Zr1-C52
N1-C52
2.259(6)
1.271(8)
N1-Zr1-C52
O1-Zr1-O1′ a
O2-Zr1-O2′
Zr1-O1-C1
32.7(2)
91.9(1)
143.8(1)
169.4(4)
Zr1-O2-C13
Zr1-N1-C52
Zr1-C52-N1
120.0(3)
74.0(3)
73.2(3)
Complex 17
Zr1-O1
Zr1-O2
Zr1-O3
Zr1-N1
2.031(5)
2.401(3)
2.037(4)
2.104(7)
Zr1-N2
N1-C55
N2-C56
C55-C56
2.097(7)
1.507(9)
1.450(15)
1.385(13)
N1-Zr1-N2
O2-Zr1-O2′′ b
O1-Zr1-O3
Zr1-O1-C1
Zr1-O2-C13
80.4(3)
146.2(1)
94.5(2)
174.4(4)
120.7(2)
Zr1-O3-C20
Zr1-N1-C55
Zr1-N2-C56
N2-C56-C55
173.1(4)
90.3(4)
89.5(6)
118.6(9)
a Prime denotes a transformation of -x, y, 0.5 - z. b Prime denotes
a transformation of x, 0.5 - y, z.
Scheme 3
Figure 4. A SCHAKAL view of complex 15. Prime denotes a
transformation of -x, y, 0.5 - z. Disorder has been omitted for clarity.
around the 2-fold axis. The disorder prevents any sound
discussion on the nature of interactions with calixarene. The
mean plane through the disordered positions forms dihedral
angles of 18.7(1)° with the symmetry-related B and D rings.
Complex 17. The C71‚‚‚C76 guest benzene molecule lying
on the mirror plane enters the cavity aligning the C71‚‚‚C74
axis along the molecular axis of the complex, as indicated by
the value of the Zr‚‚‚C71‚‚‚C74 angle (178.8(7)°). The
molecule is perpendicular from symmetry requirements to the
A and C rings and nearly parallel to the B and D rings (dihedral
angle 19.9(5)°). The host-guest interactions are of the same
kind as those observed in complex 5, the shortest contacts
involving the C34 methyl carbon of the host and the C74,
C75 carbon atoms of the guest are C34‚‚‚C74, 3.989(9) Å;
Figure 5. A SCHAKAL view of complex 17. Prime denotes a
transformation of x, 0.5 - y, z. Disorder has been omitted for clarity.
86.0(5)°, respectively), so as to orientate the hydrogen atoms
toward the center of the B and D rings (CT1‚‚‚C68, 3.397(12)
Å; CT2‚‚‚C68, 3.520(16) Å; H682‚‚‚CT1, 2.45 Å; CT2‚‚‚H681,
2.57 Å; CT1 and CT2 refer to the centroids of the B and D
rings, respectively).
Complex 15. The macrocycle shows an elliptical cone
conformation more symmetric than those observed for com-
plexes 5 and 13 (Table 5 and above). The toluene molecule
enters the cavity not aligning its molecular axis with that of the
complex, resulting in a statistical distribution of the molecule

ReactiVity on a Calix[4]arene Oxo Surface
J. Am. Chem. Soc., Vol. 119, No. 41, 1997 9717
Table 5. Comparison of Selected Dihedral Angles for Complexes 5, 13-15, and 17
(a) Dihedral Angles (deg) between Planar Moietiesa
13b
5
14
15
17
E∧A
E∧B
E∧C
E∧D
A∧C
B∧D
140.7(2)
110.2(2)
130.5(2)
108.2(2)
91.2(2)
139.4(2)
100.1(2)
136.3(2)
94.1(3)
95.8(3)
165.8(2)
[138.8(2)]
[97.5(2)]
[138.1(2)]
[95.4(2)]
[96.9(3)]
[167.1(2)]
138.5(2)
110.6(2)
142.1(2)
108.7(2)
100.6(3)
140.7(3)
145.4(2)
108.6(2)
145.4(2)
108.6(2)
110.8(2)
142.7(2)
139.3(1)
109.9(1)
141.6(1)
109.9(1)
100.9(1)
140.2(2)
141.6(3)
(b) Contact Distances (Å) between p-Carbon Atoms of Opposite Aromatic Ringsc
5
13
14
15
17
C4‚‚‚C17
C10‚‚‚C24
9.294(12)
7.096(12)
9.652(11)
5.740(10)
9.472(15)
7.154(12)
9.835(10)
7.044(10)
9.508(8)
7.171(7)
a The E reference plane refers to the least-squares mean plane defined by the C7, C14, C21, and C28 bridging methylenic carbons. b Values in
square brackets refer to molecule B. c In complex 15, C17 and C24 should be read C4′ and C10′, respectively (the prime ) -x, y, 0.5 - z). In
complex 17, C24 should be read C10′ (the prime ) x, 0.5 - y, z).
H341‚‚‚C74, 3.05 Å; C34‚‚‚C75, 3.753(9) Å; and H341‚‚‚C75,
2.81 Å. The Zr‚‚‚C71 distance is 4.947(8) Å (Zr‚‚‚H71,
4.02 Å).
The discussion will essentially focus on the kinetic study and
on the mechanistic and theoretical interpretation of some aspects
of the migratory insertion of carbon monoxide (see the formation
of η2-metal-bonded ketones in Scheme 1) and isocyanides (see
the formation of η2-metal bonded imines, η2-iminoacyls, and
their coupling in Scheme 2) into the Zr-C bonds of complexes
1-3.
(iii) The Organic Fragment Bonded to Zirconium. Com-
plex 5. In spite of the disorder involving the O5 oxygen atom,
the values of the Zr-O5 and Zr-C47 bond distances are
consistent with an asymmetric η2-C,O interaction of the dibenzyl
ketone to zirconium [Zr-O5A, 2.084(11) Å; Zr-O5B, 2.119-
(17) Å; Zr-C47, 2.210(12) Å]. The O5-C47 bond distance
(O5A-C47, 1.32(2) Å; O5B-C47, 1.24(2) Å) due to the
molecular disorder does not have a correct estimation.
Complex 13. The η2-bonded imine shows an asymmetric
bonding mode to the metal, the Zr-N1 and Zr-C51 bond
distances being 2.048(7) [2.043(7)] and 2.261(8) [2.281(8)] Å,
respectively. The N1-C51 bond distance [1.434(11) Å] is, as
expected, consistent with a single bond.16b
Complex 14. The parameters listed in Table 4 support the
structure drawn for 14, and rule out any interaction of the metal
with the cumulene C52-C53-N3 moiety. The organic frag-
ment acts as an enediamido ligand. The Zr, N1, C51, C52,
and N2 five-membered chelate ring assumes an envelope
conformation with the N2 atom protruding by 1.049(8) Å from
the plane through Zr, N1, C51, and C52.
The Kinetic Study of the η2-Iminoacyl Coupling. As far
as the mechanistic interpretation of the η2-iminoacyl coupling
reaction is concerned, we refer, as a guideline, to the funda-
mental work by Rothwell.20
The rates of conversion of 15 and 16 into 17 and 18,
respectively, were obtained by 1H NMR measurements, as
reported in the Supporting Information. The intramolecular
coupling followed first-order kinetics in both cases. The
activation parameters [∆Hq (kJ mol-1), Ea (kJ mol-1)] obtained
from the kinetic measurements are, respectively, 106 ( 2, 109
( 2 for complex 15 and 113 ( 8, 116 ( 8 for complex 16.
Although in our compounds we are dealing with a pseudo-
octahedral coordination of the metal, this does not seem to have
a significant impact on the energy pathway of the coupling.
Our parameters are, in fact, very close to those of Rothwell, as
is our mechanistic interpretation (Vide infra).20
Complex 15. The two η2-iminoacyl groups are η2-bonded
to zirconium, in a trans arrangement, as is found in the few
cases of bis-η2-iminoacyls reported thus far, with very close
Zr-C and Zr-N bond distances [Zr1-N1, 2.249(4) Å; Zr1-
C52, 2.259(6) Å]. The N1-C52 bond distance [1.271(8) Å] is
in agreement with a double bond.15
Complex 17. The enediamido17 ligand forms a metallacycle
which is nearly planar and forms a dihedral angle of 31.8(3)°
with the reference plane. The complex possesses a crystallo-
graphically imposed m symmetry with the mirror plane running
through Zr, O1, O3, N1, and C51. This requires a statistical
distribution of the N2 and C54‚‚‚C61 atoms over two positions
about the mirror plane.
Theoretical Interpretation of the Migratory Insertion
Reaction of CO and RNC. Extended Hu¨ckel calculations21,22
have been performed with the aim of understanding the major
differences which exist between the [Cp2Zr]2+ and the [calix-
[4]-(OMe)2(O)2Zr]2+ fragments in driving the migratory inser-
tion of carbon monoxide and isocyanides into the metal-carbon
bonds of the [ZrR2] functionality. To this purpose, we examine
in detail, emphasizing the major differences between CO and
RNC, the following items: (i) the formation of the η2-metal-
bonded acyl and its conversion to η2-ketone; (ii) the formation
of the η2-bonded iminoacyl and either its transformation in η2-
imine or its reductive coupling to the enediamido.
The [calix[4]-(OMe)2(O)2Zr]2+ fragment has already been
considered from a theoretical point of view,9 and its frontier
orbitals are depicted in Figure 6, where we report, for
comparison, the five metal-based empty orbitals of [Cp2Zr].2+
The two lowest frontier orbitals of [Cp2Zr]2+ (1a1 and 1b1)
have the same symmetry properties of those of the [calix[4]-
(OMe)2(O)2Zr]2+. However, while the former fragment has only
another low-lying empty orbital (2a1) available for bonding with
additional ligands, the latter fragment has two (1b2, 1a2), and
this has important consequences on its chemical behavior.
Discussion
All reactions displayed in Schemes 1 and 2 belong to four
general classes, namely: (i) the migratory insertion of carbon
monoxide or isocyanides into a metal-alkyl bond with the
consequent formation of η2-acyl or η2-iminoacyl functionalities;
(ii) the alkyl-/aryl- migration to an η2-acyl or η2-iminoacyl
metal-bonded moiety, leading to the corresponding η2-ketone
or η2-imine; (iii) the intramolecular coupling of two η2-
iminoacyls to an enediamido ligand; (iv) the migratory insertion
of ketones and isocyanides into the M-C bond of η2-metal-
bonded ketones and imines.
(20) Durfee, L. D.; McMullen, A. K.; Rothwell, I. P. J. Am. Chem. Soc.
1988, 110, 1463-1467 and references therein.
(21) Hoffmann, R.; Lipscomb, W. N. J. Chem. Phys. 1962, 36, 2179.
(22) Hoffmann, R. J. Chem. Phys. 1963, 39, 1397.

9718 J. Am. Chem. Soc., Vol. 119, No. 41, 1997
Giannini et al.
Figure 6. Frontier orbitals of the [calix[4](OMe)2(O)2Zr]2+ fragment
Figure 7. Total energy profiles for the CO approach toward [calix-
[4](OMe)2(O)2ZrMe2] and [Cp2ZrMe2], and the CNMe approach to
[calix[4](OMe)2(O)2ZrMe2].
(left), compared with the [Cp2Zr]2+ fragment (right).
Scheme 4
Scheme 5
First, these four low-lying d-orbitals can accommodate up to
eight electrons, and this is fully exploited in the bis-η2-iminoacyl
compounds 15 and 16. This behavior is at variance with that
of [Cp2Zr]2+ which can accommodate a maximum of six
electrons without violating the effective atomic number (EAN)
rule. Moreover, while in [Cp2Zr]2+ all of the three low-lying
frontier orbitals lie in the symmetry plane bisecting the Cp-
Zr-Cp angle (xz); in [calix[4]-(OMe)2(O)2Zr]2+ only two out
of the four lowest-lying orbitals are coplanar, thus favoring a
facial over a meridional structure when such a fragment is bound
to three additional ligands.
In the hexacoordinate [calix[4]-(OMe)2(O)2ZrR2] compounds,
1-3, the two alkyl groups lie in the yz plane. Only two of the
low-lying d-orbitals are used, notably 1a1 and 1b2 (dyz) which
interact, respectively, with the in-phase and out-of-phase
combinations of the two σ orbitals of the alkyl groups, leaving
as LUMO the 1b1(dxz) orbital. Its spatial extension suggests
that the most favorable approach of the CO or ButNC nucleo-
philic species occurs along a line in the xz plane forming an
angle of approximately 45° with the z-axis. We first simulated
the initial stages of the CO attack by performing extended
Hu¨ckel calculations on the [calix[4]-(OMe)2(O)2ZrMe2]‚‚‚CO
system with different CO to metal distances, from 5.0 to 2.5 Å,
along three possible directions, i.e., a line in the xz plane forming
an angle of approximately 45° with the z-axis, a line in the yz
plane bisecting the Me-Zr-Me angle, and a line in the yz plane
perpendicular to the z-axis, respectively, A, B, and C in Scheme
4. For the first two approaches, we relaxed the Me-Zr-Me
angle at each point along the attack pathway; for the second
approach, we also relaxed the dihedral angle between Me-Zr-
Me and O-Zr-O planes and the angular position of CO, while
in the third approach we independently relaxed the angles of
the two Me groups and of CO. It was discovered that pathway
A is by far energetically more favorable than B and C (with a
barrier of only ca. 0.2 eV Vs 2.2 and 2.4 eV, respectively). The
total energy profile for approach A is compared in Figure 7
with that for the approach of CO toward [Cp2ZrR2] (for which
a C-like approach is known to be the most favorable).12c We
see that the energy barrier for [calix[4]-(OMe)2(O)2ZrMe2] is
slightly lower than for [Cp2ZrR2], with the corresponding
precoordinated CO complex always lower in energy. This
suggests a more electrophilic character of the [calix[4]-(OMe)2-
(O)2Zr]2+ Vs [Cp2Zr]2+, in agreement with the Mulliken charges
calculated on the zirconium atom (+2.1 and +1.5, respectively).
We then considered the hypothetical η2-acyl intermediate
[calix[4]-(OMe)2(O)2Zr(η2-MeCO)(Me)] in the three possible
idealized structures (D, E, and F), and in a plausible η1-acyl
geometry, G, as shown in Scheme 5.
Previous calculations12c on the analogous [Cp2Zr(η2-MeCO)-
(Me)] species have taken into account only structures D and E
with the η2 planar moiety lying in the equatorial plane, the only
experimentally observed for these bis-cyclopentadienyl systems,
showing that the O-inside structure D is more stable by ca. 0.2

ReactiVity on a Calix[4]arene Oxo Surface
J. Am. Chem. Soc., Vol. 119, No. 41, 1997 9719
eV. In those calculations, the η1-acyl geometry G was also
proposed as an intermediate in the isomerization pathway
between D and E and was calculated to be less stable than D
by ca. 0.4 eV. In our calculations on [calix[4]-(OMe)2(O)2Zr-
(η2-MeCO)(Me)], we fixed the Zr-C(methyl), Zr-C(acyl), and
C-O distances (2.30, 2.21, and 1.29 Å, respectively) and the
C(acyl)-O-C angle (120°) and optimized the Zr-C-O and
the C(methyl)-Zr-C(acyl) angles, R and â in Scheme 5. The
results have shown that the most stable structure is F, with the
η2 planar moiety lying perpendicular to the yz plane. Moreover,
when we constrained the acyl moiety to lie in the equatorial
plane, we found the η1 geometry G more stable than the η2
geometries D or E (ca. 1 eV higher than C). Note that the
most stable structure F found in these calculations has the same
geometry experimentally observed in the X-ray analysis for the
η2-iminoacyl moieties in the bis-η2-iminoacyl complex 15. The
different geometries calculated for [calix[4]-(OMe)2(O)2Zr(η2-
MeCO)(Me)] and [Cp2Zr(η2-MeCO)(Me)] can be rationalized
in terms of plausible interactions between the frontier orbitals
of each metal fragment and the acetyl moiety. For the [Cp2-
Zr]2+ fragment, only orbitals in the equatorial plane are
available, thus ruling out nonplanar structures such as F, while
for [calix[4]-(OMe)2(O)2Zr]2+ there are vacant low-energy
orbitals lying on both the perpendicular symmetry planes,
suggesting that species with three additional ligands are not
stable in a meridional geometry (such as D or E).
A comparison of the orbital energies of [calix[4]-(OMe)2-
(O)2Zr(η2-MeCO)(Me)] and [Cp2Zr(η2-MeCO)(Me)] aids in the
explanation of the fast migration, in the former hypothetical
species, of the second alkyl leading to the η2-ketone complex.
Hoffmann and co-workers have shown that the reactivity of such
an η2-coordinated acyl can be described in terms of a ‘‘carbe-
nium-type’’ character and is determined by the presence of a
low-lying LUMO, essentially consisting of the π*CO perpen-
dicular to the acyl plane.12c,23a Such an empty low-lying π*CO
makes the acyl carbon strongly electrophilic, thus favoring a
second migration which leads to the η2-ketone. We found that
the π*CO orbital is lower in energy for the bis-methoxycalixarene
complex than for the bis-cyclopentadienyl one (-10.2 Vs -9.3
eV) so that it should undergo more easily the migration of the
second alkyl group having carbanion character. This could
therefore explain why, unlike for [Cp2ZrR2], it was impossible
to isolate the [calix[4]-(OMe)2(O)2Zr(η2-MeCO)(Me)] interme-
diate. Moreover, in the calculated most stable structure F, π*CO
points toward the Me substituent, while for the [Cp2Zr] system,
π*CO is perpendicular to the plane containing the Me group.
The investigation on the migratory insertion of isocyanide
into the same Zr-C bonds led us to stress the major differences
both with CO and with the [Cp2Zr]2+ fragment. The initial
stages of the isocyanide attack have been simulated as be-
fore, by performing calculations on the [calix[4]-(OMe)2-
(O)2ZrMe2]‚‚‚CNMe system varying the carbon(CNMe)-to-
metal distance along a line in the xz plane forming an angle of
approximately 45° with the z-axis (an A-like approach), relaxing
the R-Zr-R angle, the dihedral angle between the MeZrMe
and OZrO planes and the CNMe angular position. The total
energy profiles are reported in Figure 7 and show an energy
barrier higher than that for the CO attack (0.4 eV instead of 0.2
eV), probably due to the higher steric hindrance of the
isocyanide species.
the intermediacy of [calix[4]-(OMe)2(O)2Zr(η2-MeNdCMe)-
(Me)], for which four hypothetical forms corresponding to those
of the η2-acyl (see Scheme 5) have been considered. In our
calculations, we kept the Zr-C(methyl), Zr-C(iminoacyl), and
C-N distances and the C(iminoacyl)-N-C and C-C(iminoa-
cyl)-N angles fixed with the values taken from the X-ray
structure of the bis-η2-iminoacyl complex 15 and optimized
the Zr-C-N and the C(methyl)-Zr-C(iminoacyl) angles,
analogous to R and â in Scheme 5. The results have shown
that the most stable structure is again that with the η2-planar
moiety lying perpendicular to the yz plane (C-like).
A
comparison of the orbital energies and Mulliken charges of
[calix[4]-(OMe)2(O)2Zr(η2-MeCO)(Me)] and [calix[4]-(OMe)2-
(O)2Zr(η2-MeNdCMe)(Me)] aids in the explanation of the
different behavior of the [ZrR2] functionality toward the
migratory insertion of carbon monoxide and isocyanides. We
found the π*CO orbital in the η2-acyl species slightly lower than
the π*CN orbital in the η2-iminoacyl (-10.2 Vs -9.9 eV,
respectively) and a much higher charge on the acyl carbon than
on the iminoacyl one (+0.67 Vs +0.26 eV, respectively). The
higher electrophilicity of the acyl carbon could explain the more
facile migration of the second alkyl group to the η2-acyl, leading
to the η2-ketone as the only observed product. For ButNC, the
lower electrophilicity of the iminoacyl carbon opens the possi-
bility of observing the migratory insertion of ButNC into the
second metal-carbon bond to give the bis-η2-iminoacyl com-
plex, before an intramolecular migration of the alkyl group to
the η2-iminoacyl carbon takes place to give an η2-imine species.
We finally turned our attention to the intramolecular C-C
coupling of the two coordinated η2-iminoacyl ligands in
complexes 15 and 16, leading to the formation of the enediamido
complexes 17 and 18. A complete theoretical analysis of this
reaction has not been performed, due to the high number of
geometrical parameters which should be simultaneously con-
sidered to take into account all of the possible degrees of
freedom of the two η2-iminoacyl groups during their coupling.
However, a few qualitative conclusions can be drawn on the
basis of a previous theoretical study on the intramolecular
coupling of two η2-acyl groups in organoactinide compounds
[Cp2M(η2-RCO)2] (M ) Th, U). Indeed, such a study23b has
shown that the energy of the π*CO orbitals is an important factor
in determining the barrier to C-C bond coupling. A similar
argument can be reasonably applied to the coupling of two η2-
iminoacyl groups. Therefore, the relatively high energy of the
π*CN orbital leads to a significant barrier to the C-C coupling
and makes possible the isolation of the bis-η2-iminoacyls, at
least at low temperature.
Acknowledgment. This paper is dedicated to Prof. Dieter
Seebach on the occasion of his 60th birthday. We thank the
Fonds National Suisse de la Recherche Scientifique (FNS, Bern,
Switzerland, grant no. 20-46’590.96), CIBA-GEIGY SA (Basel,
Switzerland), and Fondation Herbette (University of Lausanne,
N.R.) for financial support.
Supporting Information Available: X-ray crystallography
and kinetic measurements, ORTEP drawings, tables giving
crystal data and details of the structure determination, atomic
coordinates, anisotropic thermal parameters, bond lengths, bond
angles for 5, 13-15, and 17; table of activation parameters for
15 and 16 (44 pages). See any current page for ordering and
Internet access instructions.
Also in the migratory insertion of MeNC, we should invoke
(23) (a) Tatsumi, K.; Nakamura, A.; Hoffmann, R. Organometallics 1985,
4, 404. (b) Tatsumi, K.; Nakamura, A.; Hofmann, P.; Hoffmann, R.; Moloy,
K. G.; Marks, T. J. J. Am. Chem. Soc. 1986, 108, 4467.
JA9708225