Self-assembly and anion encapsulation properties of cavitand-based coordination cages

Article abstract of DOI:10.1021/ja0103492

Two novel classes of cavitand-based coordination cages 7a-j and 8a-d have been synthesized via self-assembly procedures. The main factors controlling cage self-assembly (CSA) have been identified in (i) a P-M-P angle close to 90° between the chelating ligand and the metal precursor, (ii) Pd and Pt as metal centers, (iii) a weakly coordinated counterion, and (iv) preorganization of the tetradentate cavitand ligand. Calorimetric measurements and dynamic ¹H and ¹⁹F NMR experiments indicated that CSA is entropy driven. The temperature range of the equilibrium cage-oligomers is determined by the level of preorganization of the cavitand component. The crystal structure of cage 7d revealed the presence of a single triflate anion encapsulated. Guest competition experiments revealed that the encapsulation preference of cages 7b,d follows the order BF₄^- > CF₃SO₃^- ? PF₆^- at 300 K. ES-MS experiments coupled to molecular modeling provided a rationale for the observed encapsulation selectivities. The basic selectivity pattern, which follows the solvation enthalpy of the guests, is altered by size and shape of the cavity, allowing the entrance of an ancillary solvent molecule only in the case of BF₄^-.

Full text of DOI:10.1021/ja0103492

J. Am. Chem. Soc. 2001, 123, 7539-7552
7539
Self-Assembly and Anion Encapsulation Properties of Cavitand-Based
Coordination Cages
Federica Fochi,† Paola Jacopozzi,† Elina Wegelius,‡ Kari Rissanen,‡ Pietro Cozzini,§
Elena Marastoni,† Emilia Fisicaro, Paola Manini,| Roel Fokkens,& and Enrico Dalcanale*,†
Contribution from the Dipartimento di Chimica Organica e Industriale, UniVersita` di Parma, Parco Area
delle Scienze 17/A, I-43100 Parma, Italy, Laboratory of Organic Chemistry, UniVersity of JyVa¨skyla¨,
P.O. Box 35, FIN-40351 JyVa¨skyla¨, Finland, Dipartimento di Chimica Generale ed Inorganica, Chimica
Analitica, Chimica Fisica, UniVersita` di Parma, Parco Area delle Scienze 17/A, I-43100 Parma, Italy,
Dipartimento Farmaceutico, UniVersita` di Parma, Parco Area delle Scienze 27/A, I-43100 Parma, Italy,
Dipartimento di Clinica Medica, Nefrologia e Scienze della PreVenzione, UniVersita` di Parma,
Via Gramsci 14, I-43100 Parma, Italy, and Laboratory of Supramolecular Chemistry and Technology,
MESA+ Research Institute, UniVersity of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands
ReceiVed February 7, 2001
Abstract: Two novel classes of cavitand-based coordination cages 7a-j and 8a-d have been synthesized via
self-assembly procedures. The main factors controlling cage self-assembly (CSA) have been identified in (i)
a P-M-P angle close to 90° between the chelating ligand and the metal precursor, (ii) Pd and Pt as metal
centers, (iii) a weakly coordinated counterion, and (iv) preorganization of the tetradentate cavitand ligand.
Calorimetric measurements and dynamic 1H and 19F NMR experiments indicated that CSA is entropy driven.
The temperature range of the equilibrium cage-oligomers is determined by the level of preorganization of the
cavitand component. The crystal structure of cage 7d revealed the presence of a single triflate anion encapsulated.
Guest competition experiments revealed that the encapsulation preference of cages 7b,d follows the order
BF4- > CF3SO3- . PF6- at 300 K. ES-MS experiments coupled to molecular modeling provided a rationale
for the observed encapsulation selectivities. The basic selectivity pattern, which follows the solvation enthalpy
of the guests, is altered by size and shape of the cavity, allowing the entrance of an ancillary solvent molecule
-
only in the case of BF4
.
Introduction
In particular, metal-directed self-assembly is emerging as one
of the most promising approaches to the generation of complex
supramolecular architectures, due to the large number of
coordination motifs and multidentate ligands available. Mol-
ecules with a wide variety of topologies and shapes have been
constructed in this way, including boxes, cages, catenanes,
dendrimers, grids, helicates, rotaxanes, and others.5 Multidentate
pyridine ligands6 are the most popular building blocks for metal-
assembled molecular containers, followed by cathecolates7 and
others.8 Cavitand-based coordination cages are receiving in-
Container molecules are unique among synthetic molecular
receptors because of their peculiar encapsulation properties.1
A wide range of guests of different size, shape, and charge have
been trapped within their interior in a time scale which can span
from microseconds to forever. Desired features of container
molecules comprise selectivity in guest encapsulation, control
of guest orientation and dynamics within the cage, and revers-
ibility, which allows guest uptake and release under controlled
conditions. Equipped with such features, many potential ap-
plications can be envisioned for molecular containers, ranging
from specific drug release to catalysis2 and memory storage
devices.3
(2) Kang, J.; Hilmersson, G.; Santamar´ıa, J.; Rebek, J., Jr. J. Am. Chem.
Soc. 1998, 120, 3650-3656.
(3) van Wageningen, A. M. A.; Timmerman, P.; van Duynhoven, J. P.
M.; Verboom, W.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Chem. Eur.
J. 1997, 3, 639-654.
The central issue of their preparation has been tackled both
via covalent synthesis and, more recently, via self-assembly.4
(4) (a) Conn, M. M.; Rebek, J., Jr. Chem. ReV. 1997, 97, 1647-1668.
(b) MacGillivray, L. R.; Atwood, J. L. Angew. Chem., Int. Ed. 1999, 38,
1018-1033. (c) Jasat, A.; Sherman, J. C. Chem. ReV. 1999, 99, 931-967.
(5) For recent reviews on this topic, see: (a) Lehn, J.-M. Supramolecular
Chemistry: Concepts and PerspectiVes; VCH: Weinheim, Germany, 1995;
pp 144-160. (b) Baxter, P. N. W. In ComprehensiVe Supramolecular
Chemistry; Lehn, J.-M., Chair Ed.; Atwood, J. L., Davis, J. E. D., MacNicol,
D. D., Vogtle, F., Executive Eds.; Pergamon Press: Oxford, U.K., 1996;
Vol. 9, Chapter 5, pp 165-211. (c) Fujita, M. In ComprehensiVe
Supramolecular Chemistry; Lehn, J.-M., Chair Ed.; Atwood, J. L., Davis,
J. E. D., MacNicol, D. D., Vogtle, F., Executive Eds.; Pergamon Press:
Oxford, U.K., 1996; Vol. 9, Chapter 7, pp 253-282. (d) Fujita, M. Chem.
Soc. ReV. 1999, 27, 417-425. (e) Caulder, D. L.; Raymond, K. N. Acc.
Chem. Res. 1999, 32, 975-982. (f) Saalfrank, R. W.; Demleitner, B. In
Transition Metals in Supramolecular Chemistry; PerspectiVes in Supramo-
lecular Chemistry; Sauvage, J. P., Ed.; Wiley-VCH: Weinheim, 1999, Vol.
5, pp 1-51. (g) Leininger, S.; Olenyuk, B.; Stang, P. J. Chem. ReV. 2000,
100, 853-908.
* Corresponding
enrico.dalcanale@unipr.it.
author.
Fax
+39-0521-905472.
E-mail:
† Dipartimento di Chimica Organica e Industriale, Universita` di Parma.
‡ University of Jyva¨skyla¨.
§ Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica,
Chimica Fisica, Universita` di Parma.
Dipartimento Farmaceutico, Universita` di Parma.
| Dipartimento di Clinica Medica, Nefrologia e Scienze della Prevenzione,
Universita` di Parma.
& University of Twente.
(1) (a) Cram, D. J.; Cram, J. M. Container Molecules and their Guests,
Monographs in Supramolecular Chemistry; Stoddart, J. F., Ed.; Royal
Society of Chemistry: Cambridge, 1994; Vol. 4. (b) Sherman, J. C.
Tetrahedron 1995, 51, 3395-3422. (c) Rebek, J., Jr. Chem. Soc. ReV. 1996,
255-264. (d) Warmuth, R. J. Inclusion Phenom. Mol. Recognit. Chem.
2000, 37, 1-38.
10.1021/ja0103492 CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/13/2001

7540 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Fochi et al.
Chart 1
creasing attention9 due to the versatility of cavitand platforms
in terms of synthetic modularity and molecular recognition
phenomena. In all cases they have been employed as tetradentate
ligands by introducing four ligand moieties at the upper rim.
In a previous communication we reported the metal-directed
self-assembly of new cavitand-based coordination cages in
which two tetracyano cavitands are connected through four PdII
or PtII square-planar complexes.9a Here we present (i) prepara-
tion and characterization of a whole family of cages, including
the X-ray crystal structure of one of them, (ii) the factors
controlling the self-assembly process in terms of metal, chelating
ligand, counterion, geometry, and preorganization of the mo-
lecular components, (iii) evidences that the self-assembly process
is entropy driven, (iv) self-selection of the cavitand components,
and (v) the anion encapsulation properties of the cages.
racyano cavitands 1a-c and 2a,b were prepared from the
corresponding tetrabromo cavitands using the Rosenmund-von
Brown reaction (see Experimental Section). Cavitands 1a-c are
conformationally rigid (C4V-cone symmetry) with the four nitriles
preorganized in a fixed, diverging spatial orientation. The
presence of four ethylene bridges in cavitands 2a,b imparts a
limited degree of conformational mobility which, in solution,
allows 2a,b to rapidly interconvert between two equivalent
flattened cone conformations.10 Upon cooling to 223 K in
CD2Cl2, no splitting of the proton signals was observed for 2a,b,
indicating that the interconversion between the two flattened
cone C2V conformers through the C4V symmetry cone conformer,
in which all four nitriles have the same angle with respect to
the C4 axis, is fast in the temperature range of cage formation.
Square-planar cis-metal bis(triflate) complexes 3a-g and 4a
were prepared via chelation of the corresponding MCl2 salt,
followed by exchange of the anionic ligands with AgOTf (OTf
) triflate) in CH2Cl2. In the case of Pd(dppb)(OTf)2 3h, ligand
substitution of Pd(C6H5CN)2Cl2 with dppb11 was required to
afford the complex Pd(dppb)Cl2, followed by reaction with
AgOTf. M(dppp)X2 complexes 5a-d and 6a-c were obtained
via ligand exchange of M(dppp)Cl2 with the corresponding AgX
salts, while Pd(dppp)(CH3COO)2 6d was formed by reaction
of palladium acetate with dppp. Platinum complexes 5b and
5d are stable only in THF solution and were used immediately
after their preparation.
Results and Discussion
Synthesis of the Molecular Components (Chart 1). Tet-
(6) For recent examples of molecular containers assembled using
multidentate pyridine ligands, see: (a) Takeda, N.; Umemoto, K.; Yamagu-
chi, K.; Fujita, M. Nature 1999, 398, 794-796. (b) Olenyuk, B.; Whiteford,
J. A.; Fechtenko¨tter, A.; Stang, P. J. Nature 1999, 398, 796-799. (c)
Kusukawa, T.; Fujita, M. J. Am. Chem. Soc. 1999, 121, 1397-1398. (d)
Ikeda, A.; Yoshima, M.; Udzu, H.; Fukuhara, C.; Shinkai, S. J. Am. Chem.
Soc. 1999, 121, 4296-4297. (e) Hiraoka, S.; Fujita, M. J. Am. Chem. Soc.
1999, 121, 10239-10240. (f) Olenyuk, B.; Levin, M. D.; Whiteford, J. A.;
Shield, J. E.; Stang, P. J. J. Am. Chem. Soc. 1999, 121, 10434-10435.
(7) For molecular containers assembled using multidentate cathecolate
ligands, see ref 5e.
(8) For miscellaneous ligands, see: (a) Saalfrank, R. W.; Burak, R.; Breit,
A.; Stalke, D.; Herbst-Irmer, R.; Daub, J.; Porsch, M.; Bill, E.; Mu¨ther,
M.; Trautwein, A. X. Angew. Chem., Int. Ed. Engl. 1994, 33, 1621-1623.
(b) Mann, S.; Huttner, G.; Zsolnai, L.; Heinze, K. Angew. Chem., Int. Ed.
Engl. 1996, 35, 2808-2809. (c) Saalfrank, R. W.; Bernt, I.; Uller, E.;
Hampel, F. Angew. Chem., Int. Ed. Engl. 1997, 36, 2482-2485. (d)
Hartshorn, C. M.; Steel, P. J. Chem. Commun. 1997, 541-542. (e) Lee, S.
B.; Hong, J.-I. Tetrahedron Lett. 1998, 39, 4317-4320. (f) Abrahams, B.
F.; Egan, S. J.; Robson, R. J. Am. Chem. Soc. 1999, 121, 3535-3536.
(9) (a) Jacopozzi, P.; Dalcanale, E. Angew. Chem., Int. Ed. Engl. 1997,
36, 613-615. (b) Fox, D. O.; Dalley, N. K.; Harrison, R. G. J. Am. Chem.
Soc. 1998, 120, 7111-7112. (c) Fox, O. D.; Drew, M. G. B.; Beer, P. D.
Angew. Chem., Int. Ed. 2000, 39, 136-140. (d) Cuminetti, N.; Ebbing, M.
H. K.; Prados, P.; de Mendoza, J.; Dalcanale, E. Tetrahedron Lett. 2001,
42, 527-530.
Self-Assembly and Characterization of the Coordination
Cages. The typical procedure for cage formation is shown in
the case of cavitand 1a-c and metal precursors 3f-g (Scheme
1): by mixing the two components in a 1:2 molar ratio at room
temperature in solvents such as CH2Cl2, CHCl3, or acetone cages
7a-e were obtained in quantitative yield. Cages 7a-d were
recovered in pure form by simple evaporation of the solvent,
(10) Cram, D. J.; Karbach, S.; Kim, H.-E.; Knobler, C. B.; Maverick, E.
F.; Ericson, J. L.; Helgeson, R. C. J. Am. Chem. Soc. 1988, 110, 2229-
2237.
(11) Ligand abbreviations: dppm ) bis(diphenylphosphino)methane;
dppe ) 1,2-bis(diphenylphosphino)ethane; dppp ) 1,3-bis(diphenylphos-
phino)propane; dppb ) 1,4-bis(diphenylphosphino)butane.

CaVitand-Based Coordination Cages
J. Am. Chem. Soc., Vol. 123, No. 31, 2001 7541
Scheme 1
while 7e precipitated out of the reaction medium. All cages are
stable both in the solid state and in solution and soluble in
chlorinated and aromatic solvents. A tetrachloroethane solution
of 7b was kept overnight at 100 °C and then the cage was
recovered unchanged.
individuation of competitive oligomeric species, and (iii) the
presence of solvent molecules trapped within the cages. For
- 2+
7a-e ES-MS showed prominent [M - 2CF3SO3
- 3CF3SO3
]
and [M
]
- 3+ peaks; the molecular ions could not be detected
since their molecular weight exceeded the limit of the instru-
ment.
Their 1H NMR spectra showed the complete absence of
absorptions belonging to both precursors and the formation of
a new set of signals, indicative of the presence of a single highly
symmetric compound (D4h symmetry).9a In all cases, upon
formation of the cage, upfield shifts were observed for the inner
methylene bridge and the methyne protons of the cavitand
moiety, while aromatic protons were shifted downfield. The
upfield shift of the methyne protons is particularly indicative
since it can be attributed to shielding caused by the close
proximity of dppp aromatic rings in the formed cage (see crystal
structure). The downfield shift of the aromatic protons instead
is diagnostic of nitrile coordination to the metal. Likewise the
methylene signals of dppp were shifted downfield upon com-
plexation. Nitrile-metal coordination was also demonstrated
by FT-IR spectroscopy: the nitrile stretching band of the cages
is shifted to higher wavenumbers than that of the free cavitands.
The high symmetry of the cages was confirmed by their 31P
NMR spectra, which exhibited sharp singlets, with appropriate
Pt satellites for Pt complexes 7c-e, indicating the equivalence
of the eight dppp phosphorus atoms. The triflate counterions
of all cages instead experience two different environments: their
19F NMR spectra exhibited two singlets in a 7:1 integral ratio,
the bigger one at about -78 ppm, typical of noncoordinated
ionic triflates,12 and a smaller one shifted upfield at about -82
ppm, indicative of the encapsulation of one triflate anion inside
the cage.
MALDI-TOF spectrometry was carried out on cage 7d to
exclude the presence of species having a different stoichiometry
of the components consistent with D4h symmetry, such as wider
cages formed by four cavitands and eight organometallic
precursors. In agreement with the proposed structure, the [M
- +
- CF3SO3
] signal at m/z 5981 is the dominant peak of the
spectrum (Supporting Information) and the only one present
above 4000 Da. Signals corresponding to different aggregates
were not observed in the spectrum.
Suitable crystals of 7d for X-ray analysis were obtained from
a dichloromethane-benzene solvent mixture. Two coordination
cages A and B (Figure 1) having slightly different conformations
were found in the unit cell. Both cages are filled with a single
triflate anion, without any ancillary solvent molecule present.
The distances and angles relevant to the shape and coordination
geometry are defined in drawing I. The internal cavity resembles
an oblate sphere: the long axis a, defined as the distance be-
tween opposite Pt atoms, is 13.5(1) Å for cage A and 13.7(1)
Å for cage B. The short axis b is 12.1(1) Å for cage A and
11.9(1) Å for cage B and is defined as the distance between
the centroids of the four resorcinarene methine carbons. These
distances are in agreement with the dimensions of the cavity
calculated using the GRASP program. The included triflate
cannot escape from the lateral portals without breaking a
coordinative bond.
Additional evidences for the self-assembly of the cages were
obtained by using electrospray mass spectrometry (ES-MS).13
ES-MS proved to be a powerful technique for the investigation
of coordination cages in terms of (i) characterization, (ii)
Platinum metals have slightly distorted square-planar coor-
dination spheres: the angles around Pt range from 87(1) to
92(1)° and from 175(1) to 179(1)°. Moreover, the nitriles are
not perfectly aligned in the direction of the Pt metals, the â
angles CtNsPt being 168(1), 172(1), 168(1), and 166(1)° for
cage A and 170(1), 165(1), 169(1), and 170(1)° for cage B. In
addition, the nitriles are slightly bent from the aromatic rings
toward the Pt metals with C(Ar)sCtN angles from 174(1) to
179(1)°. The nonideal value of the angle R [30.1(3)° < R <
(12) Lawrence, G. A. Chem. ReV. 1986, 86, 17-33.
(13) For studies of noncovalent aggregation by ES-MS, see: (a) Marquis-
Rigault, A.; Dupont-Gervais, A.; Van Dorsselaer, A.; Lehn, J.-M. Chem.
Eur. J. 1996, 2, 1395-1398. (b) Hirsch, K. A.; Wilson, S. R.; Moore, J. S.
J. Am. Chem. Soc. 1997, 119, 10401-10412. (c) Schalley, C. A.; Mart´ın,
T.; Obst, U.; Rebek, J., Jr. J. Am. Chem. Soc. 1999, 121, 2133-2138.

7542 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Fochi et al.
Table 1. Cage Self-Assembly (CSA) Using Cavitand 1b as
Tetradentate Ligand and Organometallic Precursors 3a-h and 4a
entry
organometallic precursor
outcome
1
2
3
4
Pd(dppm)(OTf)2 (3a) (72.7°)a
Ni(dppe)(OTf)2 (3b) (86.9°)a
Pd(dppe)(OTf)2 (3c) (85.8°)a
Pt(dppe)(OTf)2 (3d) (85.1°)a
no CSA
no CSA
no CSA
oligomers formation at
300 K; partial CSA
at 373 K
5
6
7
8
Ni(dppp)(OTf)2 (3e)
no CSA
CSA (7b)
CSA (7d)
partial CSA at 300 K;
CSA at 373 K (7f)
partial CSA at 300 K;
CSA at 328 K (7g)
Pd(dppp)(OTf)2 (3f) (90.6°)a
Pt(dppp)(OTf)2 (3g) (89.3°)a
Pd(dppb)(OTf)2 (3h) (97.6°)a
9
Cis-Pt(PEt3)2(OTf)2 (4a)
a P-M-P angle of M(dppx)Cl2 complexes taken from the crystal
structures.15
A. In the case of cage B one triflate fluorine atom interacts
strongly with some resorcinarene carbon atoms [F488...C204
) 2.24(3) and F488...C205 ) 2.61(3) Å]. The displaced
positions of the included anion with respect to the cavity center
are due to interactions of the CF3 moiety of the anion with the
cavitand concave surface.
In addition to two cages and the included triflate anions, 14
triflate anions and 25 solvent benzene molecules together with
folded long alkyl chains of the resorcinarenes fill the interstices
of the unit cell.
Control of the Self-Assembly Process. The self-assembly
process considered in this work leads to molecular cages via
formation of four square-planar transition metal complexes. In
every complex two adjacent coordination sites are occupied by
nitriles, each belonging to a separate cavitand. The cis coordina-
tion geometry, a necessary condition for cage self-assembly
(CSA), is imposed using dppp as chelating ligand. Therefore,
the following structural parameters have been examined as the
key factors controlling the self-assembly process: (i) choice of
chelating ligand, transition metal, and counterions of the metal
precursor, (ii) preorganization and stiffness of the tetradentate
cavitand ligand.
31.4(3)°], imposed by the rigid structure of the cavitand, does
not allow the nitriles to assume the most favorable alignment
with respect to the Pt atoms. The distances of Pt(II) ions from
the best planes through the nitrogen and phosphorus atoms
bonded to Pt are 0.03(1), 0.05(1), 0.03(1), and 0.03(1) Å for
Pt1, Pt2, Pt3, and Pt4, respectively, indicating nearly no strain
in the coordination. The average bond distance of Pt...N is
2.05(1) Å and Pt...P 2.25(1) Å.
Triflate anions included into the cavities are highly disordered
between two positions and badly resolved from the X-ray data.
In both positions the CF3 of the triflate points toward the
cavitand scaffold. The shortest triflate-resorcinarene distances
[F478...C34 ) 3.15(4) and F478...C30 ) 3.16(3) Å] reveal weak
interactions between triflate and the resorcinarene part of cage
The influences of the chelating ligand and of the type of metal
have been analyzed by mixing cavitand 1d with organometallic
triflate precursors 3a-h and 4a, under standard self-assembly
conditions. The results are reported in Table 1. All the chosen
complexes assume square-planar coordination geometry in
solution, except for Ni(dppp)(OTf)2 3e, which in solution is in
equilibrium between the tetrahedral and square-planar forms.14
The effect of varying the chelate ring size has been examined
in the Pd series (entries 1, 3, 6, and 8). The stepwise increase
in the number of carbon atoms of the chelating ligand leads to
a progressive widening of the P-Pd-P angle and to a
consequent reduction of the CN-Pd-NC angle in the complex
(see for comparison the P-M-P angle of the M(dppx)Cl2
complexes in Table 1 taken from their X-ray crystal structures).15
The results clearly indicate that the CSA is highly sensitive to
distortion from the optimal square-planar arrangement of the
four ligands provided by dppp, particularly when the P-M-P
angle is smaller than 90°. In fact, in the case of dppb (3h, entry
8) partial cage formation was observed at 300 K, which became
complete at 373 K (the reasons of this temperature dependence
(14) Van Hecke, G. R.; DeW. Horrocks, W., Jr. Inorg. Chem. 1966, 5,
1968-1974.
(15) The complete set of X-ray crystal structures is available only for
the M(dppx)Cl2 series (see Steffen, W. L.; Palenik, G. J. Inorg. Chem. 1976,
15, 2432-2438 and reference cited therein). The reported P-M-P angles
have been obtained through the Cambridge Data Bank.
Figure 1. Side and top views of the molecular structure of 7d showing
the two orientations A and B assumed by the included triflate anion.
Solvent benzene molecules, alkyl chains of the cavitands, external
triflate anions, and hydrogen atoms are omitted for clarity.

CaVitand-Based Coordination Cages
J. Am. Chem. Soc., Vol. 123, No. 31, 2001 7543
Table 2. Cage Self-Assembly (CSA) Using Cavitand 1b as
Tetradentate Ligand and M(dppp)X2 5a-d and 6a-d as
Organometallic Precursors
cavitand nitriles giving rise to open oligomers. The behavior
of Ni(dppp)(OTf)2 3e is more difficult to rationalize: its
reluctance to enter the self-assembly game can be due either to
the weakness of the CN-Ni bond or to the presence in solution
of the tetrahedral form, which is not compatible with cage
formation.
anion
Pd
Pt
CF3SO3- (3f-g)
BF4- (5a-b)
PF6- (5c-d)
CSA (7b)
CSA (7h)
CSA (7k)
no CSA
no CSA
no CSA
CSA (7d)
CSA (7i)
CSA (7j)
no CSA
Another basic factor to be considered for CSA is the selection
of the counterions, which must be weakly coordinated to the
metal precursor in order to allow the exchange with the nitriles
of the cavitand ligands. For this reason triflate complexes were
selected first. The results of a series of experiments performed
on cavitand 1b with metal precursors M(dppp)X2 5a-d and
6a-d are summarized in Table 2. CSA proceeded nicely with
noncoordinating counterions such as BF4- and PF6- but failed
with the acetate, trifluoroacetate, and tosylate anions, which are
too strongly coordinated to the metal to undergo ligand exchange
with the nitriles. To check further this hypothesis, the following
experiment was devised (Scheme 2): Tetrabutylammonium
trifluoroacetate was added to cages 7b and 7d respectively in
the correct 8:1 stoichiometric ratio required to replace all nitriles,
recording 1H NMR and 19F NMR spectra of the resulting
mixtures immediately after the addition of the salt. The 1H NMR
spectrum revealed complete cage disassembly with release of
the free cavitand 1b, while the 19F NMR showed the formation
of the M(dppp)(CF3COO)2 complexes 6a,b and of tetrabutyl-
ammonium triflate. This experiment clearly indicates that
CF3COO- anions readily replace the nitriles in the cage metal
complexes, therefore inhibiting CSA.
CF3COO- (6a-b)
TsO- (6c)
CH3COO- (6d)
will be analyzed in the following paragraph). In the other two
cases (entries 1 and 3) the CSA was completely suppressed at
any temperature. The X-ray crystal structure of 7d supports these
results, showing almost no strain in the coordination sphere of
the four Pt complexes. Likewise, exchange of dppp with dppe
in the Pt complexes (compare entries 4 and 7) destabilizes the
cage with respect to open oligomers. By comparison, the use
of ethylenediamine as chelating ligand in the form of Pt(en)-
(OTf)2 totally inhibited CSA, giving no reaction with 1c. The
partial CSA observed in the case of 4a to give cage 7g (entry
9) at room temperature demonstrates that the absence of a
chelating ligand destabilizes the cage with respect to alternative
open species.
The influence of the transition metal has been highlighted
in the M(dppe)(OTf)2 and M(dppp)(OTf)2 series (entries 2-7).
The observed trend in cavitand nitriles coordination (Pt > Pd
> Ni) correlates with the relative strength of the CN-M dative
bonds. In the favorable case of dppp as chelating ligand, both
Pd and Pt complexes led to the formation of 7b,d, while for
dppe only the Pt complex 3d was able to coordinate to the
The preorganization of the cavitand ligand is another essential
ingredient of the self-assembly recipe. The influence of the
Scheme 2
Scheme 3

7544 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Fochi et al.
Table 3. Influence of Cavitand Preorganization on Cage
Self-Assembly (CSA)
Figure 2. Equilibrium cage-oligomers for 8b in C2D2Cl4 monitored
1
by H NMR as a function of temperature (1:2 mixture of 2b and 3g):
(9) diagnostic signals of the cage; (4) solvent and water resonances.
cyano substituents at the upper rim, in the presence of 3 equiv
of Pt(dppp)(OTf)2, led to the quantitative formation of capsule
10 (Scheme 4). Interestingly in this case no inclusion of a triflate
counterion was detected by 19F NMR. This experimental
evidence can be interpreted in two ways: (a) the cavity no longer
has a thermodynamic affinity for triflate; (b) the guest exchange
in and out the cavity is fast on the NMR time scale, due to the
presence of a wide portal in the equatorial region of the capsule,
large enough to allow the escape of the triflate ion from the
interior of the cage. In the second case a change in the chemical
shift (∆δ) of all triflates should be observed, as the average of
two situations (bound and free triflate). The δ value of the bound
triflate must be measured in order to evaluate the ∆δ. Since we
have been unable to determine the chemical shift of the bound
triflate in 10 from low-temperature spectra, we cannot distin-
guish between the two possibilities.
a The CSA is partial at 300 K and complete at 373 K. b The CSA is
partial at 300 K and complete at 353 K.
rigidity of the tetradentate cavitand ligand and the relative
orientation of the four cyano substituents on the cage self-
assembly have been evaluated, introducing conformationally
mobile ethylene-bridged cavitands 2a,b (Scheme 3). In the C4V-
cone conformer the four nitriles of 2a,b assume an even more
diverging spatial orientation with respect to their methylene-
bridged analogues, i.e., they form a wider R angle relative to
the C4 symmetry axis (see drawing I),16 closer to the 45° angle
ideal for the formation of strainless square-planar complexes.
Computer modeling (Spartan 5.1, Wavefunction Inc.) of cages
7 and 8 suggested the formation of less strained square-planar
complexes in the case of 8. Table 3 reports the real outcome of
the self-assembly process between cavitand 2b and several
organometallic precursors compared to the case of 1b. In the
case of 2b, CSA is possible only with Pt complexes, which
provide sufficient coordinative strength toward nitriles to freeze
the cavitand in the required cone conformation. Their Pd
analogues failed to give CSA under the same conditions. Also
in this case inclusion of a single anion has been detected by
19F NMR. A second interesting feature connected to the use of
ethylene-bridged cavitands 2a,b is the temperature effect on
CSA: in all cases the process is partial at 300 K and complete
only above 350 K (Figure 2). This unusual temperature effect
led us to investigate the thermodynamics and kinetics of CSA.
The different behavior of 2b vs 1b in the self-assembly process
points out that, in terms of preorganization of the cavitand
ligands, conformational rigidity is more important than optimal
coordination geometry.
Thermodynamics and Kinetics of the CSA. The self-
assembly process of methylene-bridged cages 7b and 7d has
1
been monitored by H NMR and FT-IR. In particular in the
1
case of H NMR experiments, the stepwise addition of metal
precursors 3f-g to cavitand 1b led to the appearance of the
diagnostic peaks of the cages and to the disappearance of those
of the limiting reactant.9a Likewise for FT-IR the changes in
nitrile absorption in the triple bond stretching region of the
spectrum were monitored (see Supporting Information). In both
cases the self-assembly process turned out to be highly
cooperative: the coordination cage was the only product
observed even in the presence of an excess of either one of the
two reactants. No intermediates in the self-assembly process
were observed. This is consistent with the rapid formation of
the thermodynamically favored cage with respect to other
species. Variable temperature 1H NMR was performed on cage
7d between 213 and 353 K. Only below 253 K did oligomeric
Finally, CSA does not require the formation of all four square-
planar complexes to be effective. Cavitand 9, having only three
Scheme 4

CaVitand-Based Coordination Cages
J. Am. Chem. Soc., Vol. 123, No. 31, 2001 7545
Figure 3. Equilibrium cage-oligomers for 8b in C2D2Cl4 monitored
1
by H NMR as a function of temperature using a 1:1 stoichiometric
ratio of 2b and 3g: (9) diagnostic signals of the cage; (0) signals of
uncomplexed cavitand 2b; (4) solvent and water resonances.
species started to compete with 7d, as indicated by broadening
of the spectrum and the appearance of new peaks.
This picture changed dramatically when cavitand 2b was used
in the self-assembly process: the 1H NMR spectra recorded by
mixing 2b and Pt(dppp)(OTf)2 3g in a 1:1 stoichiometric ratio
at 300 K exhibited broad signals, as well as signals belonging
to cage 8b (Figure 3). Upon heating to 353 K the spectrum
became sharp, showing the presence of 8b plus unreacted
cavitand 2b. Higher temperatures did not lead to further changes.
When the temperature was lowered again to 300 K, the spectrum
returned to the original pattern; this conveys a reversible process.
In the case of a 1:2 stoichiometric ratio, the formation of
oligomers is still favored at 300 K, in accordance with ES-MS
data, while at 353 K the equilibrium is shifted toward the
formation of 8b (Figure 2). The same behavior was observed
using chelating ligands forming P-M-P angles which deviates
from the optimal 90° value (see entries 4 and 9 of Table 1).
Calorimetric experiments were conducted in tetrachloroethane
as solvent at different temperatures to quantify the enthalpic
contribution to the self-assembly. Both processes are strongly
endothermic: ∆HL ) + 57.7 ( 0.7 kJ mol-1 for cage 7d
formation at 298 K and ∆HL ) + 60 ( 2 kJ mol-1 for cage 8b
formation at 343 K,17 the temperature at which 8b is almost
completely assembled. The formation of both cages is therefore
entropy driVen, with the temperature range of the equilibrium
cage-oligomers governed by the level of preorganization of the
cavitand components.18
Figure 4. ES-MS spectra at 298 K of acetone solutions of (a) cage 7d
(100 µM) and (b) cage 8b (100 µM) recorded under the same operating
conditions.
as [cage 8b]2+, and m/z 2519.5, corresponding to [2b2‚(Pt-
complex)3]2+ complex (Figure 4b).
The unusual temperature dependence of the CSA was further
substantiated by dynamic 19F NMR studies performed on cages
7d and 8b. Since the encapsulation of a triflate anion is
diagnostic of cage formation, monitoring the resonance of the
included triflate anion as a function of temperature is a simple
and clean way to trace CSA. The overall size and the dimensions
of the equatorial portals of the cavity are similar in methylene-
and ethylene-bridged cages; therefore their encapsulation prop-
erties are comparable. Anion encapsulation follows the same
1
temperature dependence observed by H NMR. For cage 7d
triflate encapsulation is temperature independent from 253 K
upward (Figure 5a), while for cage 8b it increases with
temperature from 253 K, where it is negligible, to 353 K, where
it is complete (Figure 5b). The two cages manifest the same
type of temperature dependence, only 7d requires lower
temperature to disassemble. This implies that CSA of 7d is more
strongly entropically driven than that for cage 8b.
This unusual temperature dependence of self-assembly pro-
cesses has been observed in hydrogen-bonded capsules hosting
neutral guests19 and in supramolecular clusters encapsulating
cations.20 In both examples, the origin of this unusual behavior
has been attributed to the large entropic gain associated with
the release of solvent upon encapsulation. Also in our case the
endothermic nature of CSA can be rationalized in terms of
desolvation of the cage components and particularly of the
included anion21 in the self-assembly process, which leads to
the entropic overriding of the unfavorable positive ∆HL of
formation. Less evident is the reason for the enthalpic opposition
when the encapsulation leads to a partial, electrostatic favorable,
charge neutralization (the host charge is reduced from 8+ to
7+). A convincing explanation has been given by Raymond and
ES-MS experiments were performed in acetone at room
temperature to confirm the existence in solution of other adducts
of 2b plus Pt(dppp)(OTf)2 besides cage 8b. In the case of 7d
- 2+
the dominant species in acetone detected as [M - 2CF3SO3
]
was m/z 2916.0, identified as [cage 7d]2+ (Figure 4a). By
contrast, in the case of 8b, the primary species was observed at
m/z 2066.3 and corresponds to [2b2‚(Pt-complex)2]2+ either in
the form of partially formed cage or open dimer. The other two
major species detected were respectively m/z 2971.8, identified
(16) In ref 10 several crystal structures of cavitands are reported, from
which is possible to estimate the average C4-aromatic ring R angle of 29°
for methylene-bridged cavitands and 41° for ethylene-bridged cavitands.
A close average C4-aromatic ring R angle of 31° [range 30.1(3)° < R <
31.4(3)°] has been derived for methylene-bridged cavitand 1b from the
crystal structure of 7d.
(17) Cage 8b at 298 K is in equilibrium with many oligomeric species;
therefore it is not possible to separate the various contributions to the overall
heat of reaction.
(18) For the thermodynamics of self-assembly, see: (a) Chi, X.; Guerin,
A. J.; Haycock, R. A.; Hunter, C. A.; Sarson, L. D. J. Chem. Soc., Chem.
Commun. 1995, 2563-2565. (b) Ercolani, G. J. Phys. Chem. B 1998, 102,
5699-5703.
(19) (a) Kang, J.; Rebek, J., Jr. Nature 1996, 382, 239-241. (b) Meissner,
R.; Garcias, X.; Mecozzi, S.; Rebek, J., Jr. J. Am. Chem. Soc. 1997, 119,
77-85.
(20) (a) Parac, T. N.; Caulder, D. L.; Raymond, K. N. J. Am. Chem.
Soc. 1998, 120, 8003-8004. (b) Parac, T. N.; Scherer, M.; Raymond, K.
N. Angew. Chem., Int. Ed. 2000, 39, 1239-1242.
(21) For an example of entropy-driven anion complexation, see: Berger,
M.; Schmidtchen, F. P. Angew. Chem., Int. Ed. 1998, 37, 2694-2696.

7546 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Fochi et al.
fast on the human time scale at ambient temperature only for
7b, a further evidence of the greater strength of the Pt-NC
dative bond compared to the Pd-NC one.
Self-Selection of the Cavitand Subunits. The relative
stability of cages 7d and 8b has been assessed and exploited in
competition experiments. The selective formation of cage 7d
was proven by adding 2 equiv of Pt(dppp)(OTf)2 3g to a solution
of a 1:1 molar ratio of cavitands 1b and 2b in CDCl3. The self-
assembly process, monitored by 1H NMR, gave cage 7d as the
only product, while the flexible ethylene-bridged cavitand 2b
remained unchanged in solution. The same reaction performed
in the presence of an excess of Pt(dppp)(OTf)2 at 330 K led to
the exclusive formation of both homocages 7d and 8b, with no
trace of the heterocage having both cavitands as components
(Scheme 6). Selective disassembly of Pt cages was also
realized: reaction of equimolar amounts of cages 7d and 8b in
CDCl3 with the competitive ligand NEt3, in sufficient quantity
to disassemble only one cage, left 7d unchanged, while 8b was
completely disassembled. This was clearly visible from the 1H
NMR spectrum which showed the characteristic signal patterns
of 2b and 7d. The small mismatch between the biting angles
of the nitriles of the two cavitand ligands is sufficient to suppress
heterocage formation. This is a further evidence of how sensitive
CSA is to the preorganization of the molecular components,
both in terms of process activation and product selectivity.
Selectivity in Anion Encapsulation. The Born equation23
allows for estimation of the free energy of solvation of an ion
of radius r (Å)24 and charge z in a given solvent. For monovalent
anions such as BF4-, PF6-, and CF3SO3 in chloroform, we
-
have ∆G°m,solv ) -550r-1 kJ mol-1. Since encapsulation
requires desolvation of the anion, our expectation was that
inclusion selectivity in a given cage would have been dictated
by the relative entalphy of solvation of the anions, thus favoring
the larger, less solvated ones. Accordingly, in our case the
Figure 5. Variable temperature 188 MHz 19F NMR spectra of (a) cage
7d and (b) cage 8b in CDCl3.
co-workers20a by taking into account the large enthalpic costs
associated with ion desolvation. According to this interpretation,
anions with small solvation enthalpies should be preferentially
encapsulated, provided that they fit into the cavity. To verify
this hypothesis, we undertook a series of anion exchange
experiments to test the selectivity in anion encapsulation. The
results are reported in a following paragraph.
-
-
predicted selectivity should have been CF3SO3 > PF6
>
BF4-. Table 4 reports the observed selectivity in anion
encapsulation by methylene-bridged cages for competition
experiments such as the one depicted in Scheme 7, monitored
at two different temperatures by 1H and 19F NMR. Two trends
are observed: (i) at 300 K in the presence of stoichiometric
amounts of the competing ion, the encapsulation preference is
in the order BF4- > CF3SO3- . PF6-, which becomes complete
for BF4- in the presence of an excess of (NBu4+)(BF4-); (ii) at
333 K under stoichiometric conditions the slight prevalence of
Since the anion exchange experiments require dissociation/
recombination of the coordination cages, the kinetic stability
of Pd and Pt cages deserves special attention. A rough estimate
of the dissociation/recombination time scale can be obtained
1
by monitoring via H NMR the ligand exchange between a
preformed cage and a competitive ligand. This estimate, albeit
not accurate, is sufficient for the purpose of choosing the optimal
anion exchange conditions.22 Two set of experiments were
performed directly in the NMR tube. Addition of cavitand 1c
to a CDCl3 solution of cage 7b at 300 K led to the immediate
formation of three new species in solution, identified respec-
tively as cavitand 1b, Pd heterocage 11, and Pd homocage 12,
and to the decrease of 7b (Scheme 5). The five species were
identified through the diagnostic signals of the methylene bridge
CHin and resorcinarene CH and ArH protons, and their presence
in solution was confirmed by ES-MS. Further addition of 1c
led to an increase in the concentration of 1b, 11, and 12 with
concomitant decrease of 7b. In the same experiment performed
on Pt cage 7d, the ligand exchange is negligible immediately
after the addition, with an equilibration time of over 4 h at 300
K. The same experiment, repeated at 330 K, led to immediate
ligand exchange. Therefore, the dynamics of dissociation/
recombination, slow on the NMR time scale in both cages, is
-
-
BF4 vs CF3SO3 is reversed in favor of the latter.
The discrepancy between observed and expected selectivity
in anion encapsulation led us to further investigate the matter
both experimentally and via computer modeling. ES-MS turned
out to be, once again, a very valuable investigation tool, enabling
us to detect the presence of solvent molecules trapped within
the cages.25 For all cages having CF3SO3 and PF6 as
counterions, ES-MS gave [M - 2X]2+ and [M - 3X]3+ as
prominent ions, both in chloroform and acetone as solvents,
while in the case of BF4- the base peaks correspond to the m/z
values expected for [CHCl3@M - 2BF4]2+, [CHCl3@M -
3BF4]3+, [acetone@M - 2BF4]2+, and [acetone@M - 3BF4]3+
respectively (see Figure 6 in the case of chloroform). These
results imply that only in the case of encapsulated BF4- is there
-
-
(23) (a) Moore, W. J. In Physical Chemistry, 5th ed.; Longman Group
Limited: London, 1976; p 430. (b) Marcus, Y. In Ion SolVation; Wiley &
Sons: Chichester, 1985.
(24) Even if the Born equation is strictly valid only for spherical ions,
(22) For a detailed NMR study of the exchange process in hydrogen-
bonded cali[4]arenes capsules, see: Mogck, O.; Pons, M.; Bo¨hmer, V.; Vogt,
W. J. Am. Chem. Soc. 1997, 119, 5706-5712.
it gives a good approximation of ∆GL
also for nonspherical ones.
(25) Schalley, C. A.; Castellano, R. K.; Brody, M. S.; Rudkevich, D.
m, solv
M.; Siuzdak, G.; Rebek, J., Jr. J. Am. Chem. Soc. 1999, 121, 4568-4579.

CaVitand-Based Coordination Cages
J. Am. Chem. Soc., Vol. 123, No. 31, 2001 7547
Scheme 5
Scheme 6
-
room enough in the cavity for one molecule of solvent, while
in the other two cases the anion is included alone, as proven
for the case of CF3SO3 by the crystal structure.
Molecular modeling was carried out to confirm the experi-
mental findings using the GRASP program.26 From the crystal
structure of 7d the volume of the internal cavity has been
estimated in 248 Å3 with a spherical probe of diameter 1.7 Å.27
The van der Waals molecular volumes of BF4-, PF6-, CF3SO3
,
acetone, and chloroform were also calculated with GRASP using
the same probe. If a single guest is encapsulated, the resulting
occupancy factor28 is small, well below the reported optimal
value of 0.55 (Table 5).26b When two species (anion + solvent)29
(26) (a) Nicholls, A.; Sharp, K.; Honig, B. Proteins 1991, 11, 281-
296. (b) Mecozzi, S.; Rebek, J., Jr. Chem. Eur. J. 1998, 6, 1016-1022.
-

7548 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Fochi et al.
1
Table 4. Guest Inclusion Ratio Based on H NMR and 19F NMR
-
-
-
-
(NBu4+)(BF4
)
(NBu4+)(PF6
)
(NBu4+)(BF4
excess
(40 equiv)
)
(NBu4+)(PF6
excess
(40 equiv)
)
stoichiometric
(8 equiv)
stoichiometric
(8 equiv)
cage
T (K)
-
-
-
-
-
Pt cage 7d
300
CDCl3‚BF4- > CF3SO3
CF3SO3- . PF6
only CDCl3‚BF4
CF3SO3- > PF6
BF4-/CF3SO3- ) 1.3/1.0
-
-
-
-
Pt cage 7d
Pd cage 7b
Pd cage 7b
333
300
333
CF3SO3- > CDCl3‚BF4
CF3SO3- . PF6
CDCl3‚BF4- . CF3SO3
CF3SO3- > PF6
CF3SO3-/BF4- ) 1.2/1.0
CF3SO3-/PF6- ) 1.3/1.0
-
-
-
-
CDCl3‚BF4- > CF3SO3
only CF3SO3
only CDCl3‚BF4
only CF3SO3
BF4-/CF3SO3- ) 2.7/1.0
-
-
-
CF3SO3- > CDCl3‚BF4
only CF3SO3
CDCl3‚BF4- . CF3SO3
only CF3SO3
CF3SO3-/BF4- ) 1.2/1.0
Scheme 7
are encapsulated together, the geometrical constraints imposed
by the shapes of the guests must also be considered to evaluate
the possible occupancy of the cage. The problem has been
approached using the docking functions of Spartan (version 5.1,
Wavefuction Inc.) and Sybyl (version 6.5, Tripos Inc.) softwares,
implemented with the GRASP van der Waals molecular volumes
of the guests. Both programs led to the same results for the
-
energy-minimized structures of BF4-‚CHCl3@cage, PF6
‚
CHCl3@cage, and CF3SO3-‚CHCl3@cage (Figure 7 in which
part of the cavity walls have been cut away for easy of
visualization). The pictures clearly show that for the larger
triflate and hexafluorophosphate anions there is overlap among
the van der Waals spheres of the two guests, absent in the case
of tetrafluoborate anion. The resulting steric hindrance prevents
-
-
the inclusion of the solvent molecule for CF3SO3 and PF6
.
ES-MS and molecular modeling together provide a consistent
molecular description of the observed encapsulation properties
-
of cavitand-based coordination cages. Only for BF4 is the
cavity sufficiently wide to accommodate an ancillary solvent
molecule, thus reducing the enthalpic cost required for the
-
desolvation of BF4-. This renders BF4 the preferred encap-
sulated anion at 300 K. The change in selectivity observed upon
increasing the temperature to 333 K is the result of the higher
-
cage occupancy factor of BF4-‚CHCl3 with respect to CF3SO3
.
At higher temperatures the entropic term favors CF3SO3-, since
its movements are less restricted in the confined space of the
cavity with respect to the couple BF4-‚CHCl3. This additional
entropic cost for BF4-‚CHCl3 is sufficient to reverse to
selectivity pattern observed at 300 K, as theoretically predicted
by Mecozzi and Rebek for neutral guest encapsulation.26b To
the best of our knowledge, this is the first time that such an
effect has been experimentally observed.
Conclusions
Figure 6. ES-MS spectra of CHCl3 solutions of (a) cage 7i (50 µM)
and (b) cage 7j (50 µM).
In this paper we have presented a complete characterization
of cavitand-based coordination cages and a comprehensive study
of the factors affecting the self-assembly process. The following
Table 5. Molecular Volumes and Occupancy Factors for Anionic
and Neutral Guests in Cage 7d
(27) The 1.7 Å probe is the smallest one which does not fall out of the
lateral portals of the cage.
(28) The occupancy factor of a guest molecule in the cavity of a host is
defined as the ratio of the van der Waals molecular volume of the guest to
the volume of the cavity calculated on the van der Waals interior surface
of the host (ref 19b).
guest
vol (Å3)
occupancy factor
-
BF4
PF6
37
66
86
61
68
0.15
0.27
0.35
0.25
0.27
-
-
CF3SO3
acetone
(29) The possibility of encapsulating two anions in one cage has been
excluded due to electrostatic repulsion.
chloroform

CaVitand-Based Coordination Cages
J. Am. Chem. Soc., Vol. 123, No. 31, 2001 7549
-
Figure 7. CPK representation of the vertical cross-section of cage 7d with encapsulated (A) CHCl3‚BF4-; (B) CHCl3‚PF6-; (C) CHCl3‚CF3SO3
.
in ppm relative to the carbon resonance of the deuterated NMR
solvent: CDCl3 (77.0 ppm). 31P NMR spectra were recorded at 81 and
161 MHz, and all chemical shifts were reported in ppm relative to
external 85% H3PO4 at 0.00 ppm. 19F NMR spectra were recorded at
188 MHz, and all chemical shifts were reported relative to external
CFCl3 at 0.00 ppm. Melting points were obtained with a Electrothermal
capillary melting points apparatus and were not corrected. Mass spectra
of the organic compounds were measured with a Finningan-MAT SSQ
710 spectrometer using the CI (chemical ionization) technique. Mi-
croanalyses were performed by the service of Parma University. Thin-
layer chromatography (TLC) was performed on commercially prepared
silica gel 60F254 plates purchased by Merck, and column chromatog-
raphy was performed using Merck 230-400 mesh silica gel 60. Flash
column chromatography was performed with silica gel 60 (Merck 230-
400 mesh).
structural parameters have been found to be the key factors
controlling CSA: (i) a P-M-P angle close to 90° between the
chelating ligand and the metal precursor; (ii) Pd and Pt as metal
centers; (iii) a weakly coordinated counterion; (iv) preorgani-
zation of the tetradentate cavitand ligand. CSA is extremely
sensitive to structural variations in the metal precursor: small
deviations from the optimal parameters completely switch off
the process. On the other hand, the level of preorganization of
the cavitand ligand influences the equilibrium cage-oligomers,
favoring cage formation. The ethylene-bridged cages 8 are
thermodynamically less stable than the methylene-bridged ones
7, due to the residual conformational mobility of 2, which
requires an entropic price to be paid for freezing 2 in the cone
conformation necessary for CSA. In addition, a mismatch
between the biting angles of the nitriles of the two cavitand
ligands is not tolerated by CSA, suppressing heterocage forma-
tion.
Materials. All commercial reagents were ACS reagent grade and
used as received. Diethyl ether and THF were dried and distilled over
Na/benzophenone. All the other solvents were dried over 3 and 4 Å
molecular sieves. All solvents were freeze-thaw-pump degassed three
times before use. All octols,31 tetrabromo octols, and tetrabromo
methylene/ethylene-bridged cavitands10,32 were prepared following
modified literature procedures (Supporting Information). Complexes
3a-h and 4a were prepared from the corresponding dichlorobis
derivatives following a established procedures.33
1
Dynamic H and 19F NMR experiments and calorimetric
measurements clearly indicate that CSA is entropy driven, with
the enthalpic cost of desolvating the molecular components, and
particularly the encapsulated anion, compensated by the large
entropic gain associated with the release of the solvent molecules
into the bulk. Desolvation is pivotal also in determining
selectivity in anion encapsulation, together with cavity size.
Dimensions and shape of the cavity perturb the basic selectivity
pattern, dictated by the relative solvation enthalpy of the guests,
by allowing the entrance of a solvent molecule only in the case
of BF4-, thus favoring it with respect to the other ions.
The potential of this self-assembly process in terms of
quantitative, selective, and reversible formation of molecular
cages can be fully exploited when all the critical factors
determining the outcome of the process are understood and
rationalized. In this respect, the self-assembly protocol emerging
from this study has been employed in the surface-controlled
generation of coordination cages, using self-assembled mono-
layers (SAMs) as molecular components.30
General Procedure for the Synthesis of Tetracyanocavitands. To
a stirred solution of tetrabromo cavitand (2 mmol) in 1-methyl-2-
pyrrolidinone was added CuCN (1.39 g, 15.5 mmol). The reaction
mixture was heated at reflux (210 °C) for 12 h and then cooled to 80
°C. A mixture composed of water (85 mL), hydrochloric acid (29 mL),
and FeCl3 (5.11 g, 31.5 mmol) was poured in it, and the solution was
stirred for 1 h. After cooling to room temperature, the resulting brown
precipitate was filtered and washed with water.
7,11,15,28-Tetracyano-1,21,23,25-tetrahexyl-2,20:3,19-dimetheno-
1H,21H,23H,25H-bis[1,3]dioxocino[5,4-i:5′,4′-i′]benzo[1,2-d:5,4-d′]-
bis[1,3]benzodioxocino (1a). Purification of the crude by chromatog-
raphy on silica gel (eluant CH2Cl2/hexane/acetone 95:2:3, Rf ) 0.6)
gave cavitand 1a as a white solid (70% yield): mp >300 °C; 1H NMR
(CDCl3, 300 MHz) δ 0.89 (t, 12H, CH3), 1.25-1.43 [m, 32H, CH2-
(CH2)4 CH3], 2.19 [m, 8H, CHCH2(CH2)4CH3], 4.57 (d, 4H, OCHinO,
J ) 6.7 Hz), 4.80 [t, 4H, CHCH2(CH2)4CH3], 6.08 (d, 4H, OCHoutO,
J ) 6.7 Hz), 7.26 (s, 4H, ArH); 13C NMR (CDCl3, 75 MHz) δ 13.9
(CH3), 22.5, 27.5, 29.2, 31.7 (CH2), 36.2 (CH), 98.8 (OCH2O), 104.4
(CN), 111.9 (Cipso), 124.6 (Cp), 139.0 (Cm), 156.2 (Co); IR (KBr) 2233
Experimental Section
General Methods. All reactions were conducted under an argon
atmosphere. FT-IR spectra were recorded on a Nicolet 5PC spectrom-
eter. NMR spectra were recorded on Bruker XP200, AC300, and
AMX400 spectrometers. 1H NMR were recorded at 300 and 400 MHz,
and all chemical shifts (δ) were reported in ppm relative to the proton
resonance resulting from incomplete deuteration of the NMR solvent:
CDCl3 (7.24 ppm), CD2Cl2 (5.32 ppm), C2D2Cl4 (6.0 ppm). 13C NMR
spectra were recorded at 75 MHz, and all chemical shifts were reported
(31) Tunstad, L. M.; Tucker, J. A.; Dalcanale, E.; Weiser, J.; Bryant, J.
A.; Sherman, J. C.; Helgeson, R. C.; Knobler, C. B.; Cram, D. J. J. Org.
Chem. 1989, 54, 1305-1312.
(32) Roma´n, E.; Peinador, C.; Mendoza, S.; Kaifer, A. E. J. Org. Chem.
1999, 64, 2577-2578.
(33) (a) Fallis, S.; Anderson, G. K.; Rath, N. P. Organometallics 1991,
10, 3180-3184. (b) Stang, P. J.; Cao, D. H.; Saito, S.; Arif, A. M. J. Am.
Chem. Soc. 1995, 117, 6273-6283.
(30) Levi, S.; Guatteri, P.; van Veggel, F. C. J. M.; Vancso, J.; Dalcanale,
E.; Reinhoudt, D. N. Angew. Chem., Int. Ed. 2001, 40, 1892-1896.

7550 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Fochi et al.
cm-1 (CtN); MS (CI) m/z 973 (M+, 100). Anal. Calcd for
C60H68N4O8: C, 74.05; H, 7.04; N 5.76. Found: C, 73.92; H, 7.23; N,
5.68.
was added the complex ML2(OTf)2 3a-h/4a (0.048 mmol), and the
resulting solution was stirred at room temperature for a few minutes.
The solvent was then removed under vacuum, affording the cages 7a-g
in quantitative yield.
7,11,15,28-Tetracyano-1,21,23,25-tetraundecyl-2,20:3,19-dimetheno-
1H,21H,23H,25H-bis[1,3]dioxocino[5,4-i:5′,4′-i′]benzo[1,2-d:5,4-d′]-
bis[1,3]benzodioxocin (1b). Silica gel column chromatography of the
crude (eluant 90:9:1 methylene chloride:hexane:acetone) gave com-
pound 1b as the major product (TLC Rf ) 0.6, 70% yield) and the
tricyano derivative 9 as a byproduct (TLC Rf ) 0.4, 14% yield).
7a. Pale yellow solid: mp 180 °C (dec); 1H NMR (CDCl3, 300 MHz)
δ 0.84 (t, 24H, CH3, J ) 6.8 Hz), 1.29-1.25 [m, 80H, CH3(CH2)5],
1.50 [bs, 16H, CHCH2CH2(CH2)3CH3], 2.42 (bs, 8H, Ph2PCH2CH2-
CH2PPh2), partially superimposed to 2.39 [m, 16H, CHCH2(CH2)4CH3],
2.99 (bs, 16H, Ph2PCH2CH2CH2PPh2), 4.04 (d, 8H, OCHinO, J ) 7.3
Hz), 4.39 (t, 8H, CH, J ) 8.2 Hz), 6.07 (d, 8H, OCHoutO, J ) 7.3 Hz),
7.31-7.52 (m, 80H, C6H5), 7.94 (s, 8H, ArH); 13C NMR (CDCl3, 300
MHz) δ 14.2 (CH3), 18.7, 22.9 [P(CH2)3P], 28.4, 28.6, 28.8, 30.1, 32.5
(CH2), 36.9 (CH), 99.5 (OCH2O), 124.5 (ArCN), 125.4, 129.9, 130.0,
131.4, 132.8, 133.1, 133.4, 133.6, 133.7, 139.3, 156.9 (Ar); 31P NMR
(CDCl3, 81 MHz) δ 16.7 (s); 19F NMR (CD2Cl2, 188.3 MHz) δ -76.8
1
1b: mp 216-218 °C; H NMR (300 MHz, CDCl3) δ 0.88 (t, 12H,
CH3), 1.26-1.40 [m, 72H, CH2(CH2)9CH3], 2.19 [m, 8H, CHCH2(CH2)9-
CH3], 4.59 (d, 4H, OCHinO, J ) 6.7 Hz), 4.79 [t, 4H, CHCH2(CH2)9-
CH3], 6.08 (d, 4H, OCHoutO, J ) 6.7 Hz), 7.25 (s, 4H, ArH); 13C NMR
(75 MHz, CDCl3) δ 14.1 (CH3), 22.7, 27.6, 29.2, 29.4, 29.5, 29.6, 31.9
(CH2)10, 36.2 (CH), 98.8 (OCH2O), 104.4 (CN), 111.9 (Cipso), 124.7
(Cp), 139.1 (Cm), 156.2 (Co); IR (KBr) 2238 cm-1 (CtN); MS (CI)
m/z 1254 (M+, 100). Anal. Calcd for C80H108N4O8: C, 76.64; H, 8.68;
N 4.47. Found: C, 76.49; H, 8.81; N, 4.38.
-
(s, 21F, CF3SO3 out), -79.9 (s, 3F, CF3SO3-- in); IR (KBr) 2306
cm-1 (CtN); ES-MS (acetone) m/z 2458 [M - 2CF3SO3
[M - 3CF3SO3
]
2+, 1589
-
-
]
3+, [M ) C236H240F24N8O40P8Pd4S8].
1
9: mp 140-143 °C; H NMR (300 MHz, CDCl3) δ 0.87 (t, 12H,
7b. Pale yellow solid: mp 187 °C (dec); 1H NMR (CDCl3, 300 MHz)
δ 0.81 (t, 24H, CH3, J ) 6.8 Hz), 1.20-1.34 [m, 128H, CH3(CH2)8],
1.48 [bs, 16H, CHCH2CH2(CH2)8CH3], 2.43 [m, 16H, CHCH2(CH2)9-
CH3] partially superimposed to 2.55 (bs, 8H, Ph2PCH2CH2CH2PPh2),
2.95 (bs, 16H, Ph2PCH2CH2CH2PPh2), 4.03 (d, 8H, OCHinO, J ) 7.3
Hz), 4.38 (t, 8H, CH, J ) 8.2 Hz), 6.07 (d, 8H, OCHoutO, J ) 7.3 Hz),
7.21-7.60 (m, 80H, C6H5), 7.94 (s, 8H, ArH); 13C NMR (CDCl3, 75
MHz) δ 14.0 (CH3), 18.6, 22.6 [P(CH2)3P], 28.1, 28.4, 28.7, 29.3, 29.5,
29.6, 29.7, 30.2, 31.8 (CH2), 36.6 (CH), 99.0 (OCH2O), 116.9 (Ar),
121.2 (CF3, J ) 319 Hz), 124.0 (ArCN), 124.9, 129.8, 131.2, 132.3,
132.7, 133.3, 138.9, 156.4 (Ar); 31P NMR (CDCl3, 81 MHz) δ 10.1
(s); 19F NMR (CDCl3, 188.3 MHz) δ -78.2 (21F; CF3SO3- out), -81.5
(3F; CF3SO3- in), IR (KBr) 2288 cm-1 (CtN); ES-MS (acetone) m/z
CH3), 1.34 [m, 72H, CHCH2(CH2)9CH3], 2.19 [m, 8H, CHCH2(CH2)9-
CH3], 4.47 (d, 2H, OCHinO, J ) 7.3 Hz), 4.60 (d, 2H, OCHinO, J )
7.4 Hz), 4.76 (m, 4H, CHC11H23), 5.91 (d, 2H, OCHoutO, J ) 7.3 Hz),
6.06 (d, 2H, OCHoutO, J ) 7.4 Hz), 6.59 (s, 1H, ArH), 7.05 (s, 1H,
ArH), 7.29 (s, 3H, ArH); 13C NMR (75 MHz, CDCl3) δ 14.0 (CH3),
22.6-31.9 (CH2)10, 36.2 (CH), 98.8 (OCH2O), 99.3 (OCH2O), 104.0,
104.1 (CN), 112.4 (Cipso), 117.0 (CHb), 120.3, 124.9 (Cp), 138.1, 138.3,
139.1, 140.0 (Cm), 155.0, 155.7, 156.2, 156.7 (Co); IR (KBr) 2233 cm-1
(CtN); MS (CI) m/z 1229 (M+, 100).
7,11,15,28-Tetracyano-1,21,23,25-tetrakis(2-phenethyl)-2,20:3,19-
dimetheno-1H,21H,23H,25H-bis[1,3]dioxocino[5,4-i:5′,4′-i′]benzo-
[1,2-d:5,4-d′]bis[1,3]benzodioxocin (1c). The collected solid was
purified by silica gel flash chromatography, eluting with a mixture of
methylene chloride:hexane:acetone 90:9:1, TLC Rf ) 0.3. Compound
1c was obtained as a white solid (61% yield): mp >300 °C; 1H NMR
(300 MHz, CDCl3) δ 2.52 (m, 8H, CHCH2CH2C6H5), 2.66 (m, 8H,
CHCH2CH2C6H5), 4.61 (d, 4H, OCHinO, J ) 7.5 Hz), 4.91 (t, 4H,
CHCH2CH2), 6.11 (d, 4H, OCHoutO, J ) 7.5 Hz), 7.09-7.38 (m, 24H,
C6H5 + ArH); IR (KBr) 2235 cm-1 (CtN); MS (CI) m/z 1053 (MH+,
100). Anal. Calcd for C68H52N4O8: C, 77.55; H, 4.98; N 5.32. Found:
C, 77.50; H, 4.88; N, 5.48.
8,13,18,32-Tetracyano-5,6,10,11,15,16,20,21-octahydro-1,25,27,-
29-tetrahexyl-2,24:3,23-dimetheno-1H,25H,27H,29H-bis[1,4]dioxonino-
[6,5-j:6′,5′-j′]benzo[1,2-e:5,4-e′]bis[1,4]benzodioxonin (2a). Silica gel
column chromatography of the crude (eluant 97.5:2.5 methylene
chloride:acetone) gave compound 2a as a white solid (67% yield): mp
>300 °C;1H NMR (CDCl3, 300 MHz) δ 0.87 [t, 12H, CHCH2-
(CH2)4CH3], 1.19-1.35 [m, 32H, CHCH2(CH2)4CH3], 2.07 [m, 8H,
CHCH2(CH2)4CH3], 3.80 (m, 8H, OCHinCHinO), 4.58 (m, 8H, OCHoutCH-
outO), 5.22 (t, 4H, CHC6H13), 7.41 (s, 4H, ArH); 1H NMR (acetone-d6,
300 MHz) δ 0.86 [t, 12H, CHCH2(CH2)4CH3], 1.22-1.35 [m, 32H,
CHCH2(CH2)4CH3], 2.27 [m, 8H, CHCH2(CH2)4CH3], 3.89 (m, 8H,
OCHinCHinO), 4.56 (m, 8H, OCHoutCHoutO), 5.30 (t, 4H, CHC6H13),
8.12 (s, 4H, ArH); 13C NMR (CDCl3, 75 MHz) δ 13.9 (CH3), 22.5,
27.5, 29.1, 31.6 (CH2), 33.1(CH), 71.9, 72.0 (OCH2CH2O), 103.9 (CN),
113.5, 128.5, 136.3, 156.5 (Ar); IR (KBr) 2231 cm-1 (CtN); MS (CI,
m/z) 1029 (M+,100). Anal. Calcd for C64H76N4O8: C, 74.68; H, 7.44;
N 5.44. Found: C, 74.50; H, 7.88; N, 5.48.
8,13,18,32-Tetracyano-5,6,10,11,15,16,20,21-octahydro-
1,25,27,29-tetraundecyl-2,24:3,23-dimetheno-1H,25H,27H,29H-
bis[1,4]dioxonino[6,5-j:6′,5′-j′]benzo[1,2-e:5,4-e′]bis[1,4]benzo-
dioxonin (2b). The collected solid was loaded onto the top of a silica
gel flash column; elution with a mixture of methylene chloride:acetone
98:2 afforded the compound 2b as a white solid (81% yield): mp 161-
164 °C; 1H NMR (300 MHz, CDCl3) δ 0.87 [t, 12H, CHCH2-
(CH2)9CH3],1.26[m,72H,CHCH2(CH2)9CH3],2.05[m,8H,CHCH2(CH2)9-
CH3], 3.81 (m, 8H, OCHinCHinO), 4.59 (m, 8H, OCHoutCHoutO), 5.23
(t, 4H, CHC11H23), 7.41 (s, 4H, ArH); IR (KBr) 2231 cm-1 (CtN);
MS (CI) m/z 1311 (MH+, 100). Anal. Calcd for C84H116N4O8: C, 77.02;
H, 8.92; N, 4.28. Found: C, 76.95; H, 9.03; N, 4.48.
2738 [M - 2CF3SO3]2+, 1775 [M - 3CF3SO3]3+, [M ) C276H320
F24N8O40P8Pd4S8].
-
1
7c. Off-white solid: mp 245 °C dec; H NMR (CDCl3, 300 MHz)
δ 0.85 (t, 24H, CH3, J ) 6.8 Hz), 1.25-1.29 [m, 80H, CH3(CH2)5],
1.56 [bs, 16H, CHCH2CH2(CH2)3CH3], 2.44 [m, 16H, CHCH2(CH2)4-
CH3], 2.80 (bs, 8H, Ph2PCH2CH2CH2PPh2), 3.32 (bs, 16H, Ph2PCH2-
CH2CH2PPh2), 4.00 (d, 8H, OCHinO, J ) 7.3 Hz), 4.40 (d, 8H, CH, J
) 8.2 Hz), 6.15 (d, 8H, OCHoutO, J ) 7.3 Hz), 7.22-7.52 (m, 80H,
C6H5), 7.98 (s, 8H, ArH);31P NMR (CD2Cl2, 121 MHz) δ -13.4 (J(Pt-P)
) 3350 Hz); 19F NMR (CD2Cl2, 188.3 MHz) δ -77.0 (21F; CF3SO3
-
out), -80.0 (3F; CF3SO3 in); IR (KBr) 2306 cm-1 (CtN); ES-MS
-
(acetone) m/z 2634 [M - 2CF3SO3]2+, 1706 [M - 3CF3SO3]3+, [M )
C236H240F24N8O40P8Pt4S8].
1
7d. Off-white solid: mp 250 °C dec; H NMR (CDCl3, 300 MHz)
δ 0.82 (t, 24H, CH3, J ) 6.8 Hz), 1.20-1.26 [m, 128H, CH3(CH2)8],
1.49 [bs, 16H, CHCH2CH2(CH2)8CH3], 2.44 [m, 16H, CHCH2(CH2)9-
CH3] partially superimposed to 2.60 (bs, 8H, Ph2PCH2CH2CH2PPh2),
3.11 (bs, 16H, Ph2PCH2CH2CH2PPh2), 3.99 (d, 8H, OCHinO, J ) 7.3
Hz), 4.39 (t, 8H, CH, J ) 8.2 Hz), 6.15 (d, 8H, OCHoutO, J ) 7.3 Hz),
7.18-7.55 (m, 80H, C6H5), 7.97 (s, 8H, ArH); 13C NMR (CDCl3, 75
MHz) δ 13.9 (CH3), 21.0, 21.4 [P(CH2)3P], 28.0, 28.3, 28.6, 29.1, 29.4,
29.5, 29.6, 30.0, 31.7 (CH2), 36.5 (CH), 98.4 (OCH2O), 121.2 (CF3, J
) 319 Hz), 123.8 (ArCN), 129.6, 129.8, 131.8, 132.2, 132.5, 133.3,
138.8, 156.6 (Ar); 31P NMR (CDCl3, 81 MHz) δ -15.7 (J(Pt-P) ) 3317
Hz); 19F NMR (CDCl3, 188.3 MHz) δ -78.1 (21F; CF3SO3 out),
-
-81.6 (3F; CF3SO3 in); IR (KBr) 2290 cm-1 (CtN); MALDI-MS
-
m/z 5981 [M - CF3SO3-], [M ) C276H320F24N8O40P8Pt4S8]; ES-MS
(acetone) m/z 2916 [M - 2CF3SO3]2+, 1894 [M - 3CF3SO3]3+
.
7e. This cage can also be collected by precipitation from chloroform
and filtration (off-white powder): mp 240 °C dec; 1H NMR (CD2Cl2,
400 MHz) δ 2.58-2.65 (m, 24H, CHCH2CH2Ph and Ph2PCH2CH2-
CH2PPh2), 2.80 (m, 16H, CHCH2CH2Ph), 3.04 (bs, 16H, Ph2PCH2-
CH2CH2PPh2), 3.98 (d, 8H, OCHinO, J ) 7.4 Hz), 4.44 (t, 8H,
CHCH2CH2Ph), 6.18 (d, 8H, OCHoutO, J ) 7.4 Hz), 7.20-7.56 [m,
120H, CH2C6H5 + (C6H5)2P(CH2)3P(C6H5)2], 8.09 (s, 8H, ArH); 31P
NMR (CD2Cl2, 121 MHz) δ -13.9 (J(Pt-P) ) 3400 Hz); 19F NMR
-
(CD2Cl2, 188.3 MHz) δ -77.8 (21F, CF3SO3 out), -80.5 (3F,
Procedure for the Self-Assembly of Coordination Cages 7a-g.
To a solution of cavitand 1a-c (0.024 mmol) in 10 mL of CH2Cl2
CF3SO3- in); IR (KBr) 2296 cm-1 (CtN); ES-MS (CHCl3) m/z 2715
[M - 2CF3SO3- 2+, 1761 [M - 3CF3SO3- 3+
] ] , [M ) C252H208F24N8O40P8-

CaVitand-Based Coordination Cages
J. Am. Chem. Soc., Vol. 123, No. 31, 2001 7551
Pt4S8]. Anal. Calcd for C252H208F24N8O40P8Pt4S8: C, 52.83; H, 3.66;
N, 1.95. Found: C, 52.59; H, 3.83; N, 1.50.
resulting solution was stirred at room temperature for a few minutes.
The solvent was then removed under vacuum, affording the cages 8a,b.
7f. 1H NMR (C2D2Cl4, 300 MHz, 373 K) δ 0.95 (t, 24H, CH3), 1.32-
1.35 [m, 128H, CHCH2CH2(CH2)8CH3], 1.52 [m, 16H, CHCH2CH2-
(CH2)8CH3], 2.46 [m, 32H, CHCH2(CH2)9CH3 + Ph2PCH2CH2CH2-
CH2PPh2], 2.80 [m, 16H, Ph2PCH2(CH2)2CH2PPh2], 4.12 (d, 8H,
OCHinO, J ) 7.4 Hz), 4.49 (t, 8H, CH), 6.40 (d, 8H, OCHoutO, J )
7.4 Hz), 7.47-7.65 (m, 80H, C6H5), 7.95 (s, 8H, ArH); 31P NMR
(C2D2Cl4, 81 MHz, 373 K) δ 28.3; 19F NMR (C2D2Cl4, 188.3 MHz,
8a. H NMR (C2D2Cl4, 300 MHz, 353 K) δ 0.83 (t, 24H, CH3),
1
1.19 [bs, 64H, CHCH2(CH2)4CH3], 2.16 [m, 16H, CHCH2(CH2)4CH3],
2.48 (bs, 8H, Ph2PCH2CH2CH2PPh2), 3.20 (bs, 16H, Ph2PCH2CH2CH2-
PPh2), 3.28 (m, 16H, OCHinO), 4.45 (m, 16H, OCHoutO), 4.90 (t, 8H,
CH), 7.34 (t, 16H, Hp-dppp), 7.45 (m, 32H, Hm-dppp), 7.59 (m, 32H,
Ho-dppp), 7.94 (s, 8H, ArH); 31P NMR (CDCl3, 81 MHz, 353 K) δ
-15.4 (J(Pt-P) ) 3321 Hz); 19F NMR (188.3 MHz, C2D2Cl4, 353 K) δ
-
-
-
-
373 K) δ -76.8 (21F, CF3SO3 out), -80.5 (3F, CF3SO3 in).
7g. 1H NMR (CDCl3, 300 MHz, 328 K) δ 0.88 (t, 24H, CH3), 1.27
[m, 216H, CHCH2(CH2)9CH3 + PCH2CH3], 2.07 (m, 48H, PCH2CH3),
2.33 [t, 16H, CHCH2(CH2)9CH3], 4.73 (t, 8H, CH), 4.85 (d, 8H,
OCHinO, J ) 7.2 Hz), 5.97 (d, 8H, OCHoutO, J ) 7.2 Hz), 7.64 (s, 8H,
ArH); 31P NMR (CDCl3, 81 MHz, 328 K) δ 7.8 (J(Pt-P) ) 3590 Hz);
-76.8 (21F, CF3SO3 out), -80.2 (3F, CF3SO3 in); IR (KBr) 2254
cm-1 (CtN); ES-MS (CHCl3) m/z 2692 [M - 2CF3SO3]2+, [M )
244H256F24N8O40P8Pt4S8].
8b. H NMR (C2D2Cl4, 400 MHz, 353 K) δ 0.89 (t, 24H, CH3),
C
1
1.26-1.32 [m, 128H, CHCH2CH2(CH2)8CH3], 1.42 [bs, 16H, CHCH2-
CH2(CH2)8CH3], 2.28 [m, 16H, CHCH2(CH2)9CH3], 2.55 (bm, 8H,
Ph2PCH2CH2CH2PPh2), 3.20 (bs, 16H, Ph2PCH2CH2CH2PPh2), 3.34 (m,
16H, OCHinCHinO), 4.52 (m, 16H, OCHoutCHoutO), 4.95 (t, 8H, CH),
7.37 (t, 16H, Hp-dppp), 7.47 (m, 32H, Hm-dppp), 7.66 (m, 32H,
Ho-dppp), 8.07 (s, 8H, ArH);31P NMR (C2D2Cl4, 81 MHz, 353 K) δ
-16.1 (J(Pt-P) ) 3337 Hz);19F NMR (C2D2Cl4, 188.3 MHz, 353 K) δ
-
19F NMR (CDCl3, 188.3 MHz, 328 K) δ -79.5 (21F, CF3SO3 out),
-
2+
-81.5 (3F, CF3SO3- in); ES-MS (CHCl3) m/z 2558 [M - 2CF3SO3
] ,
-
1655 [M - 3CF3SO3
]
3+, [M ) C216H324F24N8O40P8Pt4S8].
Procedure for the Self-Assembly of Coordination Cages 7h-j.
To a THF solution of M(dppp)X2 was added 5a-d (M ) Pd, Pt, X )
BF4, PF6, 0.048 mmol), prepared in situ from M(dppp)Cl2 and AgX
(stoichiometric ratio 1:2) cavitand 1b (30 mg, 0.024 mmol), dissolved
in 20 mL of CH2Cl2, and the resulting solution was stirred at room
temperature for a few minutes. The solvent was then removed under
vacuum, affording cages 7h-j in quantitative yield.
-
-
-76.8 (21F, CF3SO3 out), -80.2 (3F, CF3SO3 in); IR (KBr) 2254
cm-1 (CtN); MALDI-MS m/z 6092 [M - CF3SO3-]+ [M )
C
284H336F24N8O40P8Pt4S8]; ES-MS (CHCl3) m/z 2972 [M - 2CF3SO3]2+
.
Procedure for the Self-Assembly of Coordination Cages 8c,d.
To a THF solution of Pt(dppp)X2 5b/5d (X ) BF4, PF6, 0.055 mmol),
prepared in situ from Pt(dppp)Cl2 and AgX (stoichiometric ratio 1:2),
was added cavitand 2b (35 mg, 0.027 mmol), dissolved in 20 mL of
CH2Cl2, and the resulting solution was stirred at room temperature for
a few minutes. The solvent was then removed under vacuum, affording
cages 8c,d.
7h. 1H NMR (CDCl3, 300 MHz) δ 0.82 (t, 24H, CH3, J ) 6.9 Hz),
1.20-1.30 [m, 128H, CH3(CH2)8], 1.49 [bs, 16H, CHCH2CH2(CH2)8-
CH3], 2.35 [m, 16H, CHCH2(CH2)9CH3] 2.54 (bs, 8H, Ph2PCH2CH2-
CH2PPh2), 2.94 (bs, 16H, Ph2PCH2CH2CH2PPh2), 4.12 (d, 8H, OCHinO,
J ) 7.0 Hz), 4.36 (t, 8H, CH, J ) 7.9 Hz), 6.19 (d, 8H, OCHoutO, J )
7.0 Hz), 7.21-7.63 (m, 80H, C6H5), 7.84 (s, 8H, ArH); 31P NMR
(CDCl3, 81 MHz) δ +11.1; 19F NMR (CDCl3, 188.3 MHz) δ -150.4
1
8c. H NMR (C2D2Cl4, 400 MHz, 353 K) δ 0.82 (t, 24H, CH3),
1.18 [bs, 128H, CHCH2CH2(CH2)8CH3], 1.40 [bm, 16H, CHCH2CH2-
(CH2)8CH3], 2.24 [m, 16H, CHCH2(CH2)9CH3], 2.81 (bs, 8H,
Ph2PCH2CH2CH2PPh2), 3.14 (bs, 16H, Ph2PCH2CH2CH2PPh2), 3.45 (m,
16H, OCHinCHinO) 4.49 (m, 16H, OCHoutCHoutO), 4.83(t, 8H, CHC11H23),
7.32 (m, 16H, Hp-dppp), 7.47 (m, 32H, Hm-dppp), 7.65 (m, 32H, Ho-
dppp), 8.14 (s, 8H, ArH); 31P NMR (C2D2Cl4, 81 MHz, 353 K) δ -13.1
(J(Pt-P) ) 3320 Hz);19F NMR (C2D2Cl4, 188.3 MHz, 353 K) δ -145.0
-
-
(28F, BF4 out), -151.3 (4F; BF4 in); MALDI-MS m/z 5191 [M -
BF4-]+, [M ) C268H320B8F32N8O16P8Pd4].
7i. Yellow solid: mp 195 °C dec; H NMR (CDCl3, 300 MHz) δ
1
0.82 (t, 24H, CH3, J ) 6.9 Hz), 1.20-1.27 [m, 128H, CH3(CH2)8],
1.51 [bs, 16H, CHCH2CH2(CH2)8CH3], 2.37 [m, 16H, CHCH2(CH2)9-
CH3] 2.51 (bs, 8H, Ph2PCH2CH2CH2PPh2), 3.06 (bs, 16H, Ph2PCH2-
CH2CH2PPh2), 4.10 (d, 8H, OCHinO, J ) 7.0 Hz), 4.36 (t, 8H, CH, J
) 7.9 Hz), 6.24 (d, 8H, OCHoutO, J ) 7.0 Hz), 7.16-7.60 (m, 80H,
C6H5), 7.92 (s, 8H, ArH); 31P NMR (CD2Cl2, 81 MHz) δ -14.5 (J(Pt-P)
-
-
(4F, BF4 in), -149.1 (28F; BF4 out); ES-MS (CHCl3) m/z 2845
[M‚CHCl3 - 2BF4 ] ]
2+, 1868 [M‚CHCl3 - 3BF4 3+, [M ) C284H336B8-
F32N8O16P8Pt4].
-
-
) 3335 Hz); 19F NMR (CDCl3, 188.3 MHz) δ -150.0 (28F, BF4
-
1
8d. H NMR (C2D2Cl4, 400 MHz, 353 K) δ 0.83 (t, 24H, CH3),
1.07-1.41 (m, 128H, [CHCH2CH2(CH2)8CH3], 1.73 [bs, 16H,
CHCH2CH2(CH2)8CH3], 2.21 [m, 16H, CHCH2(CH2)9CH3] 2.88 (bs,
8H, Ph2PCH2CH2CH2PPh2), 3.07 (bs, 16H, Ph2PCH2CH2CH2PPh2), 3.20
(m, 16H, OCHinCHinO), 4.42 (m, 16H, OCHoutCHoutO), 4.84 (t, 8H,
CHC11H23), 7.32 (m, 16H, Hp-dppp), 7.47 (m, 32H, Hm-dppp), 7.65
(m, 32H, Ho-dppp), 7.94 (s, 8H, ArH); 31P NMR (C2D2Cl4, 81 MHz,
333 K) δ - 141.0 (m, 8P, PF6-, J(F-P) ) 715 Hz), -9.3 (s + d, 8P,
dppp, J(Pt-P) ) 3350 Hz); 19F NMR (C2D2Cl4, 188.3 MHz, 333 K) δ
-71.7 (d, 42F, PF6- out), -76.5 (d, 6F, PF6- in). ES-MS (CHCl3) m/z
-
out), -151.0 (4F; BF4 in); ES-MS (CHCl3) m/z 2788 [M‚CHCl3 -
-
-
2BF4
]
2+, 1830 [M‚CHCl3 - 3BF4
]
3+, [M ) C268H320B8F32N8O16P8-
-
Pt4]; ES-MS (acetone) m/z 2758 [M‚(CH3)2CO - 2BF4
]
2+, 1810 [M‚
-
3+
(CH3)2CO - 3BF4
] .
7k. This cage is unstable both in solution and in the solid state. It
must be stored in the dark under an Ar atmosphere.1H NMR (CDCl3,
300 MHz) δ 0.84 (t, 24H, CH3, J ) 6.9 Hz), 1.20-1.25 [m, 128H,
CH3(CH2)8], 1.40 [m, 16H, CHCH2CH2(CH2)8CH3], 2.35 [m, 16H,
CHCH2(CH2)9CH3], 2.74 (bs, 8H, Ph2PCH2CH2CH2PPh2), 2.92 (bs,
16H, Ph2PCH2CH2CH2PPh2), 3.89 (d, 8H, OCHinO, J ) 7.0 Hz), 4.39
(t, 8H, CHC11H23, J ) 7.6 Hz), 6.14 (d, 8H, OCHoutO, J ) 7.0 Hz),
2959 [M - 2PF6 ] ]
2+, 1924 [M - 3PF6 3+, [M ) C284H336F48N8O16P16-
-
-
Pt4].
7.18-7.54 [m, 80H, (C6H5)2P(CH2)3P(C6H5)2], 7.74 (s, 8H, ArH); 19
F
Self-Assembly of Coordination Cage 10. To a solution of cavitand
9 (0.02 mmol) in 10 mL of CH2Cl2 was added the complex Pt(dppp)-
(OTf)2 3g (0.06 mmol), and the resulting solution was stirred at room
temperature for a few minutes. The solvent was then removed under
vacuum, affording cage 10 in quantitative yield.
-
NMR (CDCl3, 188.3 MHz) δ -71.4 (d, 42F, J(F-P) ) 712 Hz, PF6
-
out), -78.1 (d, 6F, J(F-P) ) 712 Hz, PF6 in).
7j. This cage must be stored in the dark: mp 182 °C dec; 1H NMR
(CDCl3, 300 MHz) δ 0.82 (t, 24H, CH3, J ) 6.9 Hz), 1.20-1.28 [m,
128H, CH3(CH2)8], 1.51 [bs, 16H, CHCH2CH2(CH2)8CH3], 2.37 [m,
16H, CHCH2(CH2)9CH3], 2.51 (bs, 8H, Ph2PCH2CH2CH2PPh2), 3.02
(bs, 16H, Ph2PCH2CH2CH2PPh2), 3.88 (d, 8H, OCHinO, J ) 7.0 Hz),
4.41 (t, 8H, CHC11H23, J ) 7.9 Hz), 6.21 (d, 8H, OCHoutO, J ) 7.0
Hz), 7.16-7.52 [m, 80H, (C6H5)2P(CH2)3P(C6H5)2], 7.80 (s, 8H, ArH);
31P NMR (CD2Cl2, 81 MHz) δ -141.0 (m, 8P, PF6-, J(F-P) ) 712
Hz), -14.0 (s + d, 8P, dppp, J(Pt-P) ) 3333 Hz); 19F NMR (CDCl3,
10: mp 125 °C; 1H NMR (CDCl3, 300 MHz) δ 0.83 (t, 24H, CH3),
1.22-1.24 [m, 128H, CHCH2CH2(CH2)8CH3], 1.45 [bs, 16H, CHCH2-
CH2(CH2)8CH3], 2.27-2.42 [m, 16H, CHCH2(CH2)9CH3], superim-
posed to 2.64 (m, 6H, Ph2PCH2CH2CH2PPh2), 3.17 (bs, 12H, Ph2PCH2-
CH2CH2PPh2), 3.99 (d, 4H, CHin, J ) 6.7 Hz), 4.27 (d, 4H, OCHinO,
J ) 7.2 Hz), 4.42 (t, 4H, CH), 4.50 (t, 4H, CH), 5.81 (d, 4H, OCHoutO,
J ) 6.7 Hz), 6.12 (d, 4H, OCHoutO, J ) 7.2 Hz), 6.35 (s, 2H, ArHa),
7.05 (s, 2H, ArHb), 7.18-7.58 (m, 60H, C6H5), 7.79 (s, 6H, ArHc);
13C NMR (CDCl3, 75 MHz) δ 14.1 (CH3), 18.5, 22.7 [Ph2P(CH2)3-
PPh2], 28.2, 28.3, 28.8, 29.0, 29.3, 29.7, 29.8, 30.0, 30.2, 31.9 (CH2),
36.4, 36.5 (CH), 98.2, 99.3 (OCH2O), 123.1, 123.3 (ArCN), 129.1,
129.3, 129.4, 129.5, 129.7, 129.9, 131.3, 131.6, 131.9, 132.4, 132.7,
132.8, 133.1, 133.2, 133.3, 137.7, 137.8, 138.5, 140.1, 154.5, 156.6,
-
188.3 MHz) δ -71.8 (d, 42F, J(F-P) ) 715 Hz, PF6 out), -78.7 (d,
-
6F, J(F-P) ) 715 Hz, PF6 in); ES-MS (acetone and CHCl3) m/z 2903
-
-
[M - 2PF6 ] ]
2+, 1887 [M - 3PF6 3+, [M ) C268H320F48N8O16P16Pt4].
Procedure for the Self-Assembly of Coordination Cages 8a,b.
To a solution of cavitand 2a,b (0.027 mmol) in 20 mL of CH2Cl2 was
added the complex Pt(dppp)(OTf)2 3g (50 mg, 0.055 mmol) , and the

7552 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Fochi et al.
157.8 (Ar); 31P NMR (CDCl3, 81 MHz) δ -16.6 (J(Pt-P) ) 3340 Hz);
benzene solvent molecules were constrained to the regular hexagons.
One of the phenyl groups of the Pt-cage is disordered between two
positions with site occupation values of 0.5. Altogether five temperature
factors were equalized. One triflate is disordered (occupancy 0.5) over
two orientations and is included into cavity of the Pt-cage. Their
temperature factors were fixed to 0.25. The final R values were R )
0.1115, wR2 ) 0.2485 [I > 2σ(I)], R ) 0.2684, wR2 ) 0.3164 (all
data) for 2438 parameters, Goodness-of-fit on F2 ) 1.017. A final
19F NMR (CDCl3, 188.3 MHz) δ -78.4 (18F, CF3SO3 out only); IR
-
(KBr) 2283, 2295 cm-1 (CtN).
Ligand Exchange Experiments. 1-Pd Cage: cage 7b (29 mg, 0.005
mmol) was dissolved in 1 mL of CDCl3. To this solution was added
cavitand 1c (11 mg, 0.01 mmol) dissolved in 1 mL of CDCl3 stepwise.
The 1H NMR spectrum clearly showed five different species (see
Supporting Information): cage 7b (R ) RI ) C11H23), heterocage 11
(R ) C11H23, RI ) CH2CH2C6H5), cage 12 (R ) RI ) CH2CH2C6H5),
cavitand 1b (R ) C11H23), cavitand 1c (R ) CH2CH2C6H5).
1H NMR (CDCl3, 400 MHz, diagnostic peaks only) δ 3.92, 4.07 (d,
OCHinO of 11), 3.98 (d, OCHinO of 12), 4.02 (d, OCHinO of 7b), 4.37-
4.40 (t, CH of 7b 11, 12), 4.59 (d, OCHinO of 1b), 4.61 (d, OCHinO of
1c), 4.80 (t, CH of 1b), 4.90 (t, CH of 1c), 7.44 (s, ArH of 1c), 7.94
(s, ArH of 7b), 7.96, 8.06 (s, ArH of 11), 8.08 (s, ArH of 12).
ES-MS (CDCl3) of the mixture: m/z 2738 [7b - 2CF3SO3]2+, 2602
difference map displayed the highest electron density of 2.51 e Å-3
,
which is located near to Pt-atoms and triflate anions. The crystal-
lographic data have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication no. CCDC 128551. The
crystals are stable only in the presence of solvent and are destroyed
upon exposure to air within seconds. The selected crystal was placed
with loop in oil.
ES-MS Experiments: Electrospray mass spectra were obtained using
a triple quadrupole Perkin-Elmer API 365 SCIEX spectrometer with
an ionspray interface (mass range <3000 amu). The sample were
introduced at a flow rate of 0.5 mL/h; the ion spray, orifice and ring
potentials were set to 5500, 30, and 400 V, respectively. The scan range
was set to 1500-3000 uma, accumulating 10 scans for each experiment.
MALDI-TOF Mass Spectra. MALDI-TOF mass spectra were
recorded on a PerSpective Biosystem Voyager-DE-RP spectrometer
equipped with delayed extraction.37 A 337 nm UV nitrogen laser
producing 3 ns pulses was used, and the mass spectra were obtained
in the linear and reflectron mode. The samples were prepared by mixing
10 µL of a chloroform solution of the cage with 30 µL of a solution of
1 mg/L hydroxybenzylidene malononitrile [HBM] in chloroform/poly-
(ethylene glycol). One microliter of the solution was loaded on a gold-
sample plate, the solvent was removed in warm air, and the sample
was transferred to the vacuum of the mass spectrometer for analysis.
Calorimetric Measurements. Reaction heats were measured by
means of a Thermometric Thermo Activity Monitor Microcalorimeter
equipped with a flow-mixing cell (steady-state experiments) by mixing
a freshly prepared 1 mM solution of Pt(dppp)(OTf)2 3g in tetrachlo-
roethane with a 0.5 mM solution of either 1b or 2b in the same solvent
at 298 and 343 K, respectively. The solutions were thermostated at the
experience temperature by means of a Heto cryothermostatic bath and
then injected in the cell of the calorimeter by means of a Gilson
peristaltic pump, Minipuls 2, equipped with Viton tubing (internal
diameter 1.5 mm) at a flow rate of 5 × 10-3 g s-1. Heats of reaction
were experimentally corrected for the heats of dilution of both reactant
solutions. The reported values are the average of at least four different
measurements.
[11 - 2CF3SO3]2+, 2538 [12 - 2CF3SO3]2+
.
Ligand Exchange Experiments. 2-Pt cage: cage 7d (31 mg, 0.005
mmol) was dissolved in 1 mL of CDCl3. To this solution was added
cavitand 1c (11 mg, 0.01 mmol) dissolved in 1 mL of CDCl3 stepwise.
Immediately after the addition at 300 K, no ligand exchange was
observed. Equilibration was reached after 4 h. The same experiment
was repeated at 330 K, giving immediate ligand exchange.
The following compounds were detected via 1H NMR: cage 7d (R
) RI ) C11H23), heterocage 13 (R ) C11H23, RI ) CH2CH2C6H5), cage
7e (R ) RI ) CH2CH2C6H5), cavitand 1b (R ) C11H23), cavitand 1c
(R ) CH2CH2C6H5).
1H NMR (CDCl3, 400 MHz, 330 K) δ 3.94-4.05 (d, OCHinO of
7d, 7e, 13), 4.39-4.45 (t, CH di 7d, 7e, 13), 4.59 (d, OCHinO of 1b),
4.62 (d, OCHinO of 1c), 4.82 (t, CH of 1b), 4.89 (t, CH of 1c), 7.45 (s,
ArH of 1c), 8.02 (s, ArH of 7d), 8.02, 8.14 (s, ArH of 13), 8.16 (s,
ArH of 7e).
After standing a few days, a white solid precipitated from the
solution. It was collected and dissolved in CD2Cl2. Its 1H NMR spectrum
was superimposable on that of pure 7e.
ES-MS (CDCl3) of the mixture: m/z 2916 [7d - 2CF3SO3]2+, 2780
[13 - 2CF3SO3]2+, 2715 [7e - 2CF3SO3]2+
.
X-ray Structural Analysis of Cage 7d. A suitable colorless crystal
(0.4 × 0.3 × 0.2 mm) of 2[C134H160N4O8P4Pt2]4+‚8CF3SO3-‚12.5C6H6
was obtained from dichloromethane-benzene, M ) 7106.50, triclinic
P-1, a ) 21.1806(4), b ) 28.9684(7), and c ) 29.2633(7) Å, R )
81.8122(12), â ) 74.7096(12), and γ ) 85.3796(12)°, V ) 17 125.5-
(7) Å3, µ(Mo KR) ) 1.791 mm-1 (based on found atoms, see below),
Z ) 2, Dc ) 1.308 g cm-3, F(000) ) 6860, T ) 173.0(1) K. Data
were recorded with a Nonius Kappa CCD diffractometer, using 396
frames, each frame covering a 1° oscillation with an exposure time of
2 × 20 s. A total of 112391 collected reflections and 77319 unique
reflections [30704 I > 2σ(I)] were used for refinement. The data were
processed with Denzo.34 The absorption correction was made with
SORTAV (Blessing, R. H. Acta Crystallogr. 1995, A51, 33-37, ), but
not applied. Lp correction was applied. Structure solution was done by
direct methods35 and refinement on F2.36 The hydrogen atoms were
calculated to their idealized positions with isotropic temperature factors
(1.2 or 1.5 times the C temperature factor) and were refined as riding
atoms. A total of 34 hydrogen atoms could not be calculated to the
located alkyl groups. In addition, 17 alkylic carbon atoms and their H
atoms could not be located. A total of 78 geometrical DFIX restraints
were needed to make some of the alkyl chains and three triflate anions
chemically reasonable. The phenyl groups of the Pt-cage and the
Acknowledgment. The paper is dedicated to the memory
of Don Cram. This work was supported by CNR (Nanotech-
nology Program) and MURST (Molecular Nanoelectronics
Project). Use was made of instrumental facilities at the Centro
Interfacolta` di Misure G. Casnati of the University of Parma.
Financial support from INSTM to P.J. and from the Finnish
Ministry of Education to E.W. is gratefully acknowledged.
Finally, we are grateful to Prof. J. C. Sherman (UBC) for helpful
suggestions.
Supporting Information Available: Experimental proce-
dures and characterization of tetrabromo cavitand precursors
of compounds 2a,b, metal complexes 3a-h, 4a, 5a-d, and 6a-
d; MALDI-TOF spectrum of cage 7d; FT-IR monitoring of
CSA; X-ray structural information on 7d. This material is
available free of charge via the Internet at http://pubs.acs.org.
(34) Otwinowski, Z.; Minor, W. Methods in Enzymology; Carter, C. W.,
Jr., Sweet, R. M., Eds.; Academic Press: New York, 1997; Vol. 276,
Macromolecular Crystallography, part A, pp 307-326.
(35) Sheldrick, G. M. SHELXS97 - A Program for Automatic Solution
of Crystal Structures; University of Go¨ttingen: Germany, 1997.
(36) Sheldrick, G. M. SHELXL97 - A Program for Crystal Structure
Refinement; University of Go¨ttingen: Germany, 1997.
JA0103492
(37) (a) Karas, M.; Bachmann, D.; Bahr, U.; Hillenkamp, F. Int. J. Mass
Spectrom. Ion Process. 1987, 78, 53-68. (b) Vestal, M. L.; Juhasz, P.;
Martin, S. A. Rapid Commun. Mass Spectrom. 1995, 9, 1044-1050.