Surface-Confined Single Molecules: Assembly and Disassembly of Nanosize Coordination Cages on Gold (111)

Article abstract of DOI:10.1002/chem.200305570

A cavitand functionalized with four alkylthioether groups at the lower rim, and four tolylpyridine groups on the upper rim is able to bind to a gold surface by its thioether groups, and forms a coordination cage with [Pd(dppp)(CF₃SO₃)₂] by its pyridine groups. The cavitand or the cage complex can be inserted from solution into a self-assembled monolayer (SAM) of 11-mercaptoundecanol on gold. The inserted molecules can be individually detected as they protrude from the SAM by atomic force microscopy (AFM). The cages can be reversibly assembled and disassembled on the gold surface. AFM can distinguish between single cavitand and cage molecules of 2.5 nm and 5.8 nm height, respectively.

Full text of DOI:10.1002/chem.200305570

FULL PAPER
Surface-Confined Single Molecules: Assembly and Disassembly of Nanosize
Coordination Cages on Gold (111)
Edoardo Menozzi,[a] Roberta Pinalli,[a] Emiel A. Speets,[b] Bart Jan Ravoo,[b]
Enrico Dalcanale,*[a] and David N. Reinhoudt*[b]
Abstract:
A
cavitand functionalized
into
a
self-assembled monolayer
dividually detected as they protrude
from the SAM by atomic force micros-
copy (AFM). The cages can be rever-
sibly assembled and disassembled on
the gold surface. AFM can distinguish
between single cavitand and cage mole-
cules of 2.5 nm and 5.8 nm height, re-
spectively.
with four alkylthioether groups at the
lower rim, and four tolylpyridine
groups on the upper rim is able to bind
(SAM) of 11-mercaptoundecanol on
gold. The inserted molecules can be in-
to
a gold surface by its thioether
Keywords:
cavitands
microscopy
cage compounds
scanning probe
self-assembly
¥
groups, and forms a coordination cage
with [Pd(dppp)(CF3SO3)2] by its pyri-
dine groups. The cavitand or the cage
complex can be inserted from solution
¥
¥
¥
single-molecule studies
Introduction
tionalized coordination cages to gold substrates in microme-
ter-size patterned self-assembled monolayers (SAMs),[4]
which were prepared by soft lithographic techniques.[5] The
attachment of single molecular containers able to encapsu-
late ions and neutral molecules on solid surfaces is attractive
for single-molecule addressing.[6]
Self-assembly, in its various forms, is emerging as a key tech-
nology for the formation of two- and three-dimensional
structures.[1] The appeal of self-assembly resides in the ther-
modynamic control of the process, which leads to the exclu-
sive formation of the desired molecular architecture under a
given set of conditions. The reversibility of self-assembly
conveys interesting properties, such as self-repairing ability
and responsiveness to external stimuli. Of the various self-
assembly protocols, metal-directed self-assembly is particu-
larly appealing, because of the large number of different
structural motifs and bond energies that are available
through coordination chemistry.[2] There are relatively few
publications that describe the transfer of self-assembly pro-
tocols from the solution phase to surfaces.[3] Our groups
have recently reported on the attachment of thioether-func-
Here we report on the synthesis of a cavitand, functional-
ized with four alkylthioether groups at the lower rim, and
four tolylpyridine groups at the upper rim. We describe the
self-assembly of coordination cages from the cavitands with
[Pd(dppp)(CF3SO3)2] both in solution[7] and immobilized on
gold surfaces. For the formation of coordination cages on
gold, two different strategies leading to homocages and het-
erocages have been explored. Immobilized homocages
result from the direct insertion of cages formed in solution
into an SAM of 11-mercaptoundecanol (MU) on gold. Im-
mobilized heterocages result from the insertion of the thio-
ether-footed cavitand into an MU SAM, followed by assem-
bly of cages by complexation of a different cavitand from
solution. The insertion and assembly processes were moni-
tored by atomic force microscopy (AFM) by measuring the
height of individual cavitands and cages.
[a] Dr. E. Menozzi, Dr. R. Pinalli, Prof. E. Dalcanale
Dipartimento di Chimica Organica ed Industriale
and Unit‡ INSTM, Universit‡ degli
Studi di Parma, Parco Area delle Scienze 17/A
43100 Parma (Italy)
Fax : (+39)0521-905-472
E-mail: enrico.dalcanale@unipr.it
Results and Discussion
[b] E. A. Speets, Dr. B. J. Ravoo, Prof. D. N. Reinhoudt
Laboratory of Supramolecular Chemistry and Technology
MESA+ Institute for Nanotechnology and Faculty of Science
and Technology, University of Twente, POBox 217
7500 AE Enschede (The Netherlands)
Synthesis: At the upper rim of cavitand 6 (Scheme 2) there
are four phenylpyridine groups as bridging units between
the phenolic OHs, and four alkylthioether chains at the
lower rim. The presence of four outward oriented phenyl-
pyridine groups is pivotal for cage self-assembly (CSA) by
Fax : (+31)53-489-4645
E-mail: smct@utwente.nl
2199
Chem. Eur. J. 2004, 10, 2199 2206
DOI: 10.1002/chem.200305570
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPER
Scheme 1. Protection/deprotection scheme for the synthesis of resorcinarene 4.
Scheme 2. Synthesis of the cavitand ligand 6, and self-assembly of nanosize coordination cages 7a and 7b.
coordination to Pd or Pt metal precursors,[8] while the four
thioether chains at the lower rim are necessary to attach the
compound to a gold surface.[9] In order to obtain cavitand 6,
the thioether-footed resorcinarene 4 was synthesized and
then bridged with 4,4’-(a,a’-dibromotolyl)pyridine (5).
The bridging group 5 was obtained by radical bromination
of the methyl group of tolylpyridine,[10] by using N-bromo-
succinimide (NBS) in the presence of benzoyl peroxide as
an initiator. By mixing tolylpyridine, NBS, and benzoyl per-
oxide in a suitable ratio, the reaction gave 4-(a,a’-dibromo-
tolyl)pyridine as a major product with traces of tribrominat-
ed tolylpyridine as a byproduct. The reaction was carried
out in the presence of a large amount of radical initiator to
prevent attack at the aromatic rings, and to favor the bromi-
nation of the methyl group.
lane as a reagent and triethylamine as a base at room tem-
perature in THF. Resorcinarene 2 was obtained in a 49%
yield. The second step was the anti-Markovnikov addition
of 1-decanethiol to the double bonds of 2 in presence of a
stoichiometric quantity of 9-BBN in THF as a solvent at
room temperature. Resorcinarene 3 with four sulfide chains
at the lower rim, was obtained in 71% yield. Finally, the
silane groups in 3 were removed with HCl (36% in metha-
nol), and the desired resorcinarene 4 was obtained in 32%
overall yield starting from 1. This synthetic procedure repre-
sents a general route to thioether-footed cavitands in all
cases in which the bridging units at the upper rim are BBN-
sensitive.
This molecule was used for the preparation of the phenyl-
pyridine functionalized thioether-footed cavitand (6)
(oooo isomer). The synthesis of this deep-cavity cavitand[12]
was performed by bridging the resorcinarene 4 with 4,4’-
(a,a’-dibromotolyl)pyridine (5) in the presence of K2CO3 as
a base. The reaction gave the desired oooo isomer (41%
yield) as the only product with four tolylpyridyl groups
The thioether footed resorcinarene 4 was synthesized fol-
lowing a protection/deprotection scheme (Scheme 1).[11] The
first step was the protection of the hydroxyl moieties of re-
sorcinarene 1 by trimethylsilane groups to avoid reaction
with 9-BBN; this was achieved by using chlorotrimethylsi-
2200
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2199 2206

Nanosize Coordination Cages on Gold
2199 2206
pointing outwards from the cavity (Scheme 2).[13] The stereo-
chemical control of the reaction is assured by the bulkiness
of the bridging units, which can be accommodated only in
the outward orientation.
the self-assembly process were not detected. Addition of
one equivalent of [Pt(dppp)(CF3SO3)2] to a solution of cavi-
tand 6 (spectrum a) in [D6]acetone (1:1 ratio) showed that
the only species present in solution are the free cavitand
and the cage (labeled by colored dots); there was no evi-
dence of oligomers or partial complexation products (spec-
trum b). When a second equivalent of metal complex was
added to the same solution (final ratio 1:2 ligand/metal com-
plex), the cage was the only product present in solution
(spectrum c). This is consistent with the rapid formation of
the thermodynamically favored cage with respect to other
species. 31P NMR spectra exhibited a sharp singlet, with ap-
propriate Pt coupling satellites
General procedure for the self-assembly of the thioether-
functionalized coordination cages 7a and 7b: Before engag-
ing in surface-confined self-assembly, cage self-assembly
(CSA) of cavitand
6
was tested in solution with
[M(dppp)(OTf)2] (M=Pt, Pd) as metal precursors to verify
that the thioether chains do not interfere with the cage for-
mation by complexing the metal centres (Figure 1). The self-
for cage 7b, which indicates
that all the DPPP phosphorus
atoms are equivalent; this
therefore confirms the symme-
try of the cage. Further evi-
dence of cage formation was
given by MALDI analysis,
which contained a prominent
À
peak at m/z=[MÀCF3SO3 ]+
for both cages.
Surface assembly: Having es-
tablished CSA of 7a and 7b in
solution, we extended this pro-
tocol to surface-directed CSA.
Two different approaches have
been followed. First, the direct
insertion of the preformed cage
7a into a monolayer of thiols
(homocages insertion), and
second the insertion of cavitand
6 into a monolayer of thiols fol-
lowed by cage self-assembly on
the preformed SAM (surface
self-assembly of heterocages).
Insertion or assembly of immo-
bilized individual cages was
monitored by tapping-mode
atomic force microscopy (TM-
1
Figure 1. Self-assembly of cage 7b in solution monitored by H NMR spectroscopy.
AFM). Self-assembled mono-
layers of MU were prepared on
1
assembly of cages 7a and 7b was monitored by H NMR
flame-annealed gold surfaces. These hydroxy-terminated
SAMs are favored over methyl-terminated alkylthiolate
SAMs, because they cause less physisorption of cavitand
molecules.[4] The monolayers were analyzed by TM-AFM
before being used for insertion experiments to ensure that
no undefined features were present on the surface. Insertion
experiments were performed according to procedures previ-
ously used in our groups,[14] although solvents and insertion
times had to be adjusted for the adsorbate molecules. The
insertion of molecules occurs at random, and this method is
an excellent and easy strategy to perform scanning probe
microscopy on immobilized single molecules and assem-
blies.[15]
spectroscopy (Figure 1). Mixing the two components in a 1:2
molar ratio at room temperature in acetone gave cages 7a
and 7b, which were obtained in quantitative yields. There-
fore, the thioether chains at the lower rim do not interfere
1
with CSA. In all cases H NMR spectra clearly showed the
formation of a new set of signals indicative of the presence
of only one highly symmetric compound (D2h symmetry).
The signals corresponding to the CH2 a of sulfur in the thio-
ether chains, were not shifted by the addition of the metal
complex to a solution of cavitand 6. The only shifts recorded
are those of the hydrogens, which correspond to the phenyl-
pyridine groups, which are shifted downfield as a conse-
quence of the coordination to the metal. The coordination
cage was the only product observed and intermediates in
In order to verify the correctness of the experimental
data, X-ray data from alkyl-footed cavitands were used to
2201
Chem. Eur. J. 2004, 10, 2199 2206
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

E. Dalcanale, D. N. Reinhoudt et al.
FULL PAPER
estimate the height of cavitand 6 (2.5 nm) and cage 7a
(5.8 nm) considering the contribution of the thioether chains
in a backfolded conformation to the total length (see
Table 1).[8] The MU molecules in the SAM are tilted to opti-
mize the van der Waals interactions and the packing stabili-
ty. The thickness of the MU monolayer is about 0.8 nm, due
to this tilted orientation.[16] On the basis of these considera-
tions, 7a is expected to protrude beyond the OH-terminated
surface by at least 5.0 nm, and about 1.7 nm for 6.
image shows the presence of objects protruding from the
MU SAM. The section analysis shows some variation in the
height of these objects (Figure 2b). We conclude that these
objects are individual cages inserted into, and protruding
from the SAM. The average height of the cage is 5.27Æ
0.59 nm (average of 22 measurements), which is consistent
with the dimensions obtained from the X-ray structure of an
identical cage with C6H13 alkyl feet.[8] The variation in
height can be attributed to the flexible conformation of the
sulfide chains at the lower rim of the cages, which can lead
to variation in protrusion length of the cages from the SAM,
and also to the measurement technique (note that the AFM
tip is very large compared to the molecule).
The surface self-assembly of heterocages was achieved by
inserting cavitand 6 into a MU SAM, and performing CSA
on the immobilized cavitand (schematically shown in Fig-
ure 3a and 4). The insertion into a MU SAM was achieved
by soaking the SAMs for one hour into a solution (0.25mm)
of cavitand molecules 6 in dichloromethane. Extensive
washing (dichloromethane, ethanol, and water) ensured the
removal of all physisorbed material. TM-AFM experiments
and section analysis showed the presence of objects protrud-
ing from the monolayer (Figure 3b). Section analysis re-
vealed that these entities had a height of 1.87Æ0.18 nm
(average of 25 measurements). From X-ray and CPK
models, the total height of cavitand 6 is about 2.5 nm
(Table 1). Since the molecules are inserted into a SAM of
Table 1. Theoretical and experimental heights of cavitand and cages.[a]
Cavitand and
cages
Height from X-ray Protrusion
Measured height
(AFM)
data[8]
2.5
height[b]
cavitand 6
1.7
5.0
4.0
1.87Æ0.18
5.27Æ0.59
3.96Æ0.32
homocage 7a 5.8[c]
heterocage 8
4.8[d]
[a] All heights in nm. [b] Protrusion heights are calculated by subtracting
the thickness of the MU SAM (0.8 nm) from the X-ray data. [c] Value
calculated for the upper thioether chains in the backfolded conformation
(1.8 nm). [d] Value calculated for the hexyl chains in the all-trans confor-
mation (0.8 nm).
The homocages insertion process of thioether-footed cage
7a is schematically shown in Figure 2a. A clean MU SAM is
soaked into a solution of cage 7a (0.25mm) in dichlorome-
thane for one hour at room temperature. After extensive
rinsing with dichloromethane, ethanol, and water, the
sample surface was analysed by AFM (Figure 2b). The 3D
Figure 3. a) Schematic representation of cavitand 6 insertion and AFM
analysis on MU SAM; b) TM-AFM image and section analysis of a MU
SAM with inserted phenylpyridine functionalized thioether-footed cavi-
tand 6.
Figure 2. a) Schematic representation of the insertion of homocage 7a in
a MU SAM; b) TM-AFM height image and section analysis of a MU
SAM with inserted homocages 7a.
2202
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2199 2206

Nanosize Coordination Cages on Gold
2199 2206
Figure 4. a) Schematic representation of the surface assembly of heterocages 8 from surface bound cavitands with cages in solution; b) TM-AFM image
and section analysis of a MU SAM with heterocages 8 (green and black) and cavitands (red).
0.8 nm height, they are expected to protrude from the SAM
by approximately 1.7 nm. Hence, we contend that we can
observe individual cavitand molecules protruding from the
SAM. The height differences which are measured for the
cavitands on the surface can be caused by the factors dis-
cussed above.
al success of the capping reaction is not surprising consider-
ing that very dilute conditions do not favor a complete
ligand exchange.
Control experiments: Since we started from clean MU
SAMs and used extensive rinsing to remove physisorbed
material, we contend that the features observed in AFM are
really inserted cavitands or cages. To confirm these results
and to exclude artefacts, two control experiments have been
carried out: a) insertion of molecules without interaction
sites, and b) disassembly of the cages on gold.
The first control experiment was performed by using cavi-
tand 10 and cage 9 with inert hexyl tails instead of thioeth-
ers at the lower rim . A full MU SAM was soaked into a sol-
ution (0.25mm) of cavitand 10 or cage 9 for one hour. After
successive rinsing with dichloromethane, ethanol, and water,
the substrates were analyzed by TM-AFM (Figure 5). The
analysis clearly shows that no detectable features are pres-
ent on the gold surface. Hence, no insertion or absorption of
molecules occurs in the absence of a functional group with
affinity for gold, and the thioether functions in cavitand 6
and cage 7a are required for binding to the gold substrate.
The second control experiment consisted of the disassem-
bly of cage 7a immobilized on gold. A gold substrate with
inserted homocages (route 1) was exposed to a solution of
triethylamine in ethanol (5mm solution) for one hour (Fig-
The next step was the assembly of the heterocage 8 (Fig-
ure 4a). By using the cavitands already attached to the sur-
face, nanosize cages could be assembled on the gold surface.
Exposing an SAM of 6 to a solution (0.25mm) of cage 9,[8]
with eight hexyl chains at the lower rim of the resorcinarene
skeleton, the assembly of heterocage 8 on gold was obtained
(Figure 4a). The formation of cages 8 on gold required
ligand exchange with cages 9 present in solution. Ligand ex-
change between cages has already been proven in solu-
tion.[7g] These experiments were performed by using the
cage of the cavitands without thioether-chains to make sure
that the cages seen on the surface were assembled in two
steps on the surface, and not directly inserted from solution
(see control experiments). Rinsing with dichloromethane,
ethanol, and water ensured the removal of all physisorbed
material. AFM experiments and section analysis showed the
presence of two different features: i) objects with a height of
about 4 5 nm, and ii) objects with a height of about 2 nm
(Figure 4b). The different heights of the objects suggest that
both uncapped cavitands and heterocages are present. Parti-
2203
Chem. Eur. J. 2004, 10, 2199 2206
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

E. Dalcanale, D. N. Reinhoudt et al.
FULL PAPER
Figure 5. a) Cavitand 10 and cage 9 used for the control experiments;
b) AFM image of 11-mercaptoundecanol monolayer after 1 h in a solu-
tion (0.25mm) of cavitand 10; c) AFM image of 11-mercaptoundecanol
monolayer after 1 h in a solution (0.25mm) of cage 9.
Figure 6. a) Schematic representation of the disassembly of the surface
bound cages 7a; b) TM-AFM height image of partly disassembled cages
7a in a MU SAM.
ure 6a). Triethylamine was able to shift the equilibrium to-
wards the formation of [Pd(dppp)(NEt3)2(OTf)2] plus free
cavitand by competing with Pd coordination centres.[7g]
AFM indicated that more than half of the cages were disas-
sembled, leaving the free cavitand molecules as shown in
Figure 6b. This experiment confirms that metal-directed
self-assembly of cages is a reversible process, also on sur-
faces.
Experimental Section
General methods: All commercial reagents were ACS reagent grade and
used as received. All solvents were dried over 3 and 4 ä molecular
sieves. 1H NMR spectra were recorded on Bruker AC300 (300 MHz)
spectrometer at 300 K, and all chemical shifts (d) were reported in parts
per million (ppm) relative to the proton resonances; this resulted from
incomplete deuteration of the NMR solvents. 31P NMR spectra were re-
corded on a Bruker AMX400 (162 MHz), and all chemical shifts were re-
ported in ppm relative to external 85% H3PO3 at 0.00 ppm. Melting
points were obtained with an electrothermal capillary apparatus and
were uncorrected. Mass spectra of the organic compounds were meas-
ured with a Finningan-MAT SSQ 710 spectrometer by using CI (chemical
ionization) technique. Matrix-assisted laser desorption ionization time-of-
flight (MALDI-TOF) mass spectra were obtained on a PerSeptive Bio-
systems Voyager DE-RP spectrometer equipped with delayed extraction.
Column chromatography was performed by using silica gel 60 (Merck
70 230 mesh). 4-(4’-Tolyl)pyridine and resorcinarene 2 were prepared ac-
cording to literature procedures.[10,17] Metal precursors were prepared
from the corresponding dichlorobis derivatives following established pro-
cedures.[18]
Conclusion
Molecular containers of nanosize dimensions have been as-
sembled in a controlled fashion on gold. The assembly in
two steps, and the disassembly of the Pd-coordinated cage
compound were monitored by TM-AFM. We envisage that
the reversibility of the system can be exploited to reversibly
encapsulate guests in the cage on the surface. Also, the for-
mation of heterocages with different substituents in the
upper and lower part opens new possibilities. For example,
the thioether chains on the top half of homocages might be
used to attach gold colloids from solution and prepare−sand-
wich× structures that may find application in nanoscale elec-
tronic device preparation. Alternatively, the top half of het-
erocages might be equipped with further functionalities for
specific interaction with molecules or nanoparticles in solu-
tion. Both approaches are currently being pursued in our
laboratories.
For monolayer preparation and rinsing p.a. grade solvents were used as
received. Water was purified by a Millipore (MilliQ, Q2) system. MU
(97%, Aldrich) was used as received.
Gold substrates were obtained from Metallhandel Schroer GmbH,
Lienen, Germany. Samples consisted of an 11î11 mm glass substrate
with 5 nm chromium adhesion layer, and on top of that 200 nm gold.
Prior to use, substrates were rinsed with dichloromethane, dried in N2
flow, and annealed in a hydrogen (purity 6) flame. After annealing, the
slightly cooled substrates were first immersed in ethanol. After five mi-
nutes, the substrates were rinsed with ethanol and immersed into the MU
solution.
AFM-analyses were performed on a Nanoscope III multimode AFM
(Digital Instruments, Santa Barbara, CA) in the tapping mode by using
the E-scanner (~12 micron). Silicon cantilevers (nanosensors) with a stiff-
2204
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2199 2206

Nanosize Coordination Cages on Gold
2199 2206
ness between 37 and 56 NmÀ1 were used. Measurements were performed
at frequencies slightly lower than the natural resonance frequency of the
cantilever in air (~300 kHz). All AFM analyses were performed at room
temperature and ambient pressure.
0.046 mmol, 41%). Rf =0.7; m.p. 2438C; 1H NMR (CDCl3, 300 MHz):
À
d=0.87 (t, 12H; R CH3), 1.26 (m, 104H; (CH2)13), 1.48 (brm, 8H;
À
À
À
CH3 CH2 R), 1.57 (m, 16H; R CH2CH2S), 2.36 (brm, 8H;
À
À
À
ArCH CH2), 2.50 (m, 16H; R CH2 S), 4.99 (t, 4H; ArCH, J=8.0 Hz),
5.56 (s, 4H; ArCH), 6.74 (s, 4H; ArH), 7.31 (s, 4H; ArH), 7.51 [bd (AA’
part of a AA’XX’ system), 8H; PyHm), 7.72 (d (AA’ part of a AA’BB’
system), 8H; PhH, J=8.3 Hz), 7.83 (d (BB’ part of a AA’BB’ system),
Trimethylsylane protected resorcinarene (2): NEt3 (5.6 mL, 40.3 mmol)
and ClSi(CH3)3 (3.2 g, 25.3 mmol) were added to a solution of resorcinar-
ene 1 (3.0 g, 2.88 mmol) in dry THF (20 mL) cooled to À108C. The solu-
tion was stirred at 08C for 2 h and then at room temperature for 16 h.
After evaporation of the solvent under vacuum, the brown crude product
obtained was purified by flash column chromatography on silica by using
CH2Cl2/hexane (8:2 v/v) as an eluent to give compound 2 as a white solid
(2.3 g, 1.4 mmol, 49%). Rf =0.2; m.p. 80 838C; 1H NMR (CDCl3,
300 MHz): d=0.39 (s, 72H; Si(CH3)3), 1.27 (m, 52H; R-(CH2)6), 1.72 and
8H; PhH, J=8.3 Hz), 8.68 ppm (brd (XX’ part of a AA’XX’ system),
1
À
8H; PyHo); H NMR ([D6]acetone, 300 MHz): d=0.86 (t, 12H; R CH3),
À
À
1.28 (m, 104H; (CH2)13), 1.51 (brm, 8H; CH3 CH2 R), 1.57 (m, 16H;
À
À
À
R CH2CH2S), 2.50 (m, 24H; ArCH CH2 +R CH2S), 5.03 (t, 4H;
ArCH, J=8.1 Hz), 5.65 (s, 4H; ArO-CH), 7.02 (s, 4H; ArH), 7.65 (d
(AA’ part of a AA’XX’ system), 8H; PyHm, J=4.6 Hz), 7.82 (m (AA’
part of a AA’BB’ system), 12H; PhH+ArH), 7.92 (d (BB’ part of a
AA’BB’ system), 8H; PhH, J=8.2 Hz), 8.63 ppm (brd (XX’ part of a
AA’XX’ system), 8H; PyHo); MS (CI): m/z (%): 2401 [MH+, (100)].
À
1.83 (brm, 8H; R CH2CH=CH2, diastereotopic protons), 2.04 (q, 8H;
À
À
ArCH CH2), 4.41 (t, 4H; ArCH, J=7.9 Hz), 4.97 (m, 8H; R CH=CH2),
À
5.83 (m, 4H; CH=CH2 R), 6.02 (brs, 2H; ArH), 6.20 (brs, 2H; ArH),
6.30 (brs, 2H; ArH), 7.18 ppm (brs, 2H; ArH); MS (CI): m/z (%): 1620
General procedure for cage formation (7a and 7b): Cages 7a and 7b
were assembled by mixing cavitand 6 with different metal precursors
[M(dppp)(OTf)2] (M=Pt, Pd) in a 1:2 molar ratio at room temperature
in acetone. In all cases, removal of the solvent in vacuum gave the de-
sired cage in quantitative yields. Cage 7a (M=Pd): 1H NMR ([D6]ace-
[MH+, (100)].
Thioether-footed trimethylsylane protected resorcinarene (3): 9-BBN
(11.36 mL, 5.68 mmol; commercial solution 0.5m in THF), and 1-decane-
thiol (11.8 mL, 56.8 mmol) were added to
a solution of 2 (2.3 g,
1.42 mmol) in dry THF (15 mL) cooled to À108C. The solution was stir-
red at room temperature for 16 h. After evaporation of the solvent under
vacuum, the crude product was dissolved in CH2Cl2 (20 mL), and washed
with water (3î20 mL). The organic layer was dried on MgSO4, filtered,
and the solvent evaporated under vacuum. After recrystallization from
ethanol, compound 3 was obtained as a white solid (2.3 g, 0.99 mmol,
70%). M.p. 428C; 1H NMR (CDCl3, 300 MHz): d=0.35 (s, 72H;
Si(CH3)3), 0.88 (t, 12H; CH3), 1.26 (m, 112H; (CH2)14), 1.55 (m, 16H;
À
tone, 300 MHz): d=0.85 (t, 24H; R CH3), 1.26 (m, 224H; (CH2)14), 1.46
À
À
(m, 32H; R CH2CH2S), 1.55 (m, 16H; ArCH CH2), 2.50 (m, 40H;
À
R CH2S+PCH2CH2), 3.41 (brm, 16H; PCH2CH2), 4.93 (t, 8H; ArCH,
J=8.2 Hz), 5.53 (s, 8H; ArO CH), 7.00 (s, 8H; ArH), 7.38 7.43 (m,
48H; ArHmdppp+ArHpdppp), 7.47 (d (AA’ part of a AA’XX’ system),
16H; PyHm, J=5.8 Hz), 7.69 (d (AA’ part of a AA’BB’ system), 16H;
PhHPy, J=8.5 Hz), 7.76 7.85 (m, 56H; ArHodppp+ArH+(BB’ part of a
À
AA’BB’ system), PhHPy), 9.06 ppm (brd (XX’ part of
a AA’XX’
À
À
À
system), 16H; PyHo); 31P NMR ([D6]acetone, 162 MHz): d=12.6 ppm;
MALDI-TOF MS: m/z (%): found 7918 [MÀCF3SO3, (100)]+ ; calculated
R CH2CH2S), 1.78 (m, 8H; ArCH CH2), 2.50 (t, 16H; R CH2-S), 4.36
(t, 4H; ArCH, J=7.7 Hz), 5.97 (brs, 2H; ArH), 6.15 (brs, 2H; ArH),
6.25 (brs, 2H; ArH), 7.12 ppm (brs, 2H; ArH); MS (CI): m/z (%): 2318
[MH+, (100)], 2143 [MH+-1-decanethiol, (30)].
7918.26, in which M=C428H528O40N8Pd4P8S16F24 (8067.32 amu). Cage 7b
1
À
(M=Pt): H NMR ([D6]acetone, 300 MHz): d=0.85 (t, 24H; R CH3),
1.27 (m, 224H; (CH2)14), 1.46 (m, 32H; R CH2CH2S), 1.55 (m, 16H;
ArCH CH2), 2.48 (m, 40H; R CH2S+PCH2CH2), 3.51 (brm, 16H;
PCH2CH2), 4.93 (t, 8H; ArCH, J=8.0 Hz), 5.53 (s, 8H; ArO CH), 7.00
Thioether-footed resorcinarene (4): HCl 36% (2.0 mL) was added to a
solution of 3 (2.3 g, 0.99 mmol) in methanol (40 mL). The pink solution
was stirred at room temperature for 16 h. The reaction was quenched
with water (15 mL), and the white solid formed was washed with water
to neutrality. After drying at the vacuum pump, compound 4 was ob-
À
À
À
À
(s, 8H; ArH), 7.38 7.45 (m, 48H; ArHmdppp+ArHpdppp), 7.49 (d (AA’
part of a AA’XX’ system), 16H; PyHm, J=6.6 Hz), 7.71 (d (AA’ part of
a AA’BB’ system), 16H; PhHPy, J=8.6 Hz), 7.76 7.85 (m, 56H; Ar-
Hodppp+ArH+(BB’ part of a AA’BB’ system), PhHPy), 9.09 ppm (brd
(XX’ part of a AA’XX’ system), 16H; PyHo); 31P NMR ([D6]acetone,
162 MHz): d=À8.1 ppm (JPÀPt =3050 Hz); MALDI-TOF MS: m/z (%);
tained in pure form as a white solid (1.6 g, 0.92 mmol, 93%). M.p. 222
1
À
2258C; H NMR (CDCl3, 300 MHz): d=0.88 (t, 12H; R CH3), 1.28 (m,
À
112H; (CH2)14), 1.57 (m, 16H; R CH2CH2S), 2.21 (brm, 8H;
À
À
ArCH CH2), 2.49 (t, 16H; R CH2S), 4.3 (brt, 4H; ArCH), 6.09 (brs,
4H; ArH), 7.21 (brs, 4H; ArH), 9.28 (brs, 4H; OH), 9.58 ppm (brs, 4H;
OH); MS (CI): m/z (%): 1737 [MH+, (30)].
found 8273 [MÀCF3SO3 (100)]+
; calculated 8272.9, in which M=
428H528O40N8Pt4P8S16F24 (8421.96 amu).
C
4,4’-(a,a’-Dibromotolyl)pyridine
(5):
4-(4’-Tolyl)-pyridine
(7.4 g,
SAM preparation: All glassware used was cleaned in piranha solution
(3 parts concentrated sulphuric acid, 1 part 30% hydrogen peroxide) for
at least 15 minutes. Warning: piranha solution should be handled with
caution. It has been reported to detonate unexpectedly. Contact with or-
ganic solvents must be avoided. After removing from piranha the glass
was rinsed extensively with MilliQ water to remove all acid. Gold sub-
strates for AFM measurements were flame-annealed with a hydrogen
flame (grade 6), and after cooling the subtrates were kept in ethanol sol-
ution for five minutes. Substrates were immersed with minimal delay into
a solution (1.0 mm) of MU in ethanol, and kept overnight at room tem-
perature. The substrates were removed from the solution and extensively
rinsed with pure water, ethanol, and dichloromethane. Before insertion
experiments, the MU SAMs were checked for the presence of nanome-
ter-sized features by TM-AFM. To ensure that pollution particles on the
SAM were not mistaken for inserted cavitands, only SAMs with less
than ~3 features per 1mm2 were used.
44.0 mmol) and benzoyl peroxide (0.53 g, 2.0 mmol) were added under
nitrogen to a suspension of N-bromosuccinimide (15.6 g, 87.0 mmol) in
degassed CCl4 (300 mL). The mixture was stirred under reflux in a nitro-
gen atmosphere for 16 h. The suspension was filtered, the solid was
rinsed with CH2Cl2, and the resulting liquid was washed with saturated
Na2CO3 solution. The liquid was then dried on K2CO3 and evaporated.
The crude product was purified by column chromatography on silica by
using ethyl acetate/hexane (8:2 v/v) as an eluant to give compound 5 as a
1
yellow solid (5.9 g, 18.0 mmol, 41%). Rf =0.35; m.p. 100 101 C; H NMR
(CDCl3, 300 MHz): d=6.70 (s, 1H; CHBr2), 7.74 (d (AA’ part of a
AA’XX×system), 2H; PhH, Jo =6.5 Hz, Jm =2.0 Hz), 7.80 (d (AA’ part of
a AA’XX’ system), 2H; PyHm, Jo =6.6 Hz, Jm =1.7 Hz), 8.00 (d (XX’ part
of a AA’XX’ system), 2H; PhH, Jo =6.5 Hz, Jm =2.0 Hz), 8.82 (d (XX’
part of a AA’XX’ system), 2H; PyHo, Jo =6.6 Hz, Jm =1.7 Hz); MS (CI):
m/z (%): 328 [M+, (30)]; 248 [(MÀBr)+, (100)].
Phenylpyridine bridged thioether-footed cavitand (6) (oooo isomer):
Compound 5 (0.23 g, 0.69 mmol) and K2CO3 (0.19 g, 1.38 mmol) were
added under nitrogen to a solution of resorcinarene 4 (0.2 g, 0.11 mmol)
dissolved in dry DMA (15 mL). The mixture was stirred in a sealed tube
at 808C for 16 h. The reaction was quenched by the addition of water
(10 mL), and the mixture was extracted with CH2Cl2 (15 mL). The organ-
ic layer was washed with water (3î15 mL), dried on K2CO3, and then
the solvent evaporated. The black crude product obtained was purified
by column chromatography on silica by using CH2Cl2/ethanol (9:1 v/v) as
an eluent to give compound 6 (oooo isomer) as a yellow solid (0.11 g,
Insertion of cages and cavitands was obtained by soaking the 11-mercap-
toundeacol monolayer into a solution (0.25mm) of cage 7a or cavitand 6
in dichloromethane for one hour at room temperature. Self-assembly of
heterocage 8 was obtained by exposing a SAM of cavitand 6 to a solution
(0.25mm) of cage 9 in dichloromethane for one hour. Homocage disas-
sembly was obtained by soaking the MU SAMs with inserted cage 7a
into a solution(~5mm) of Et3N in CH2Cl2 for one hour.
All monolayers were extensively rinsed with pure ethanol and dichloro-
methane, and dried in a nitrogen flow before the AFM analyses.
2205
Chem. Eur. J. 2004, 10, 2199 2206
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

E. Dalcanale, D. N. Reinhoudt et al.
FULL PAPER
Fokkens, E. Dalcanale, J. Am. Chem. Soc. 2001, 123, 7539 7552;
f) N. Cuminetti, M. H. K. Ebbing, P. Prados, J. de Mendoza, E. Dal-
canale, Tetrahedron Lett. 2001, 42, 527 530; g) L. Pirondini, F. Ber-
tolini, B. Cantadori, F. Ugozzoli, C. Massera, E. Dalcanale, Proc.
Natl. Acad. Sci. USA 2002, 99, 4911 4915; h) R. G. Harrison, O. D.
Fox, M.-O. Meng, N. K. Dalley, L. J. Barbour, Inorg. Chem. 2002, 41,
838 843.
Acknowledgement
This work was supported by CNR (Nanotechnology Program), MURST
(FIRB-Nanoorganization of Materials with Magnetic and Optical Proper-
ties), and the Nanolink program of MESA+. The instrumental facilities
at the Centro Interfacolt‡ di Misure G. Casnati of the University of
Parma were used.
[8] R. Pinalli, E. Dalcanale et al., unpublished results.
[9] E. U. Thoden van Velzen, J. F. J. Engbersen, P. J. de Lange, J. W. G.
Mahy, D. N. Reinhoudt, J. Am. Chem. Soc. 1995, 117, 6853 6862.
[10] D. L. Comins, A. H. Abdullah, J. Org. Chem. 1982, 47, 4315 4319.
[11] The direct addition of 1-decanethiol to an w-alkene-footed phenyl-
pyridyl cavitand was not successful.
[12] H. Xi, C. L. D. Gibb, E. D. Stevens, B. C. Gibb, Chem. Commun.
1998, 1743 1744.
[13] For a discussion on in-out isomerism in cavitands see: T. Lippmann,
H. Wilde, E. Dalcanale, L. Mavilla, G. Mann, U. Heyer, S. Spera, J.
Org. Chem. 1995, 60, 235 242.
[1] a) J.-M. Lehn, Science 2002, 295, 2400 2403; b) D. N. Reinhoudt, M.
Crego-Calama, Science 2002, 295, 2403 2407; c) G. M. Whitesides,
B. Grzybowski, Science 2002, 295, 2418 2421.
[2] a) M. Fujita, J. Yazaki, K. Ogura, J. Am. Chem. Soc. 1990, 112,
5645 5647; b) M. Fujita, K. Umemoto, M. Yoshizawa, N. Fujita, T.
Kusukawa, K. Biradha, Chem. Commun. 2001, 509 518; c) S. Leini-
ninger, B. Olenyuk, P. J. Stang, Chem. Rev. 2000, 100, 853 908.
[3] a) I. Weissbuch, P. N. W. Baxter, S. Cohen, H. Cohen, K. Kjaer, P. B.
Howes, J. Als-Nielsen, G. S. Hanan, U. S. Schubert, J.-M. Lehn, L.
Leiserowitz, M. Lahav, J. Am. Chem. Soc. 1998, 120, 4850 4860;
b) A. Hatzor, T. Moav, H. Cohen, S. Matlis, J. Libman, A. Vaske-
vich, A. Shanzer, I. Rubinstein, J. Am. Chem. Soc. 1998, 120,
13469 13477; c) A. Hatzor, T. Moav, S. Yochelis, A. Vaskevich, A.
Shanzer, I. Rubinstein, Langmuir 2000, 16, 4420 4423.
[4] S. A. Levi, P. Guatteri, F. C. J. M. van Veggel, G. J. Vancso, E. Dal-
canale, D. N. Reinhoudt, Angew. Chem. 2001, 113, 1945 1948;
Angew. Chem. Int. Ed. 2001, 40, 1892 1896.
[5] Y. Xia, G. M. Whitesides, Angew. Chem. 1998, 110, 568 594;
Angew. Chem. Int. Ed. 1998, 37, 550 575.
[14] A. Friggeri, H. Schˆnherr, H.-J. van Manen, B.-H. Huisman, G. J.
Vancso, J. Huskens, F. C. J. M. van Veggel, D. N. Reinhoudt, Lang-
muir 2000, 16, 7757 7763.
[15] The preparation of ordered self-assembled monolayers of the cavi-
tands on gold was not possible. The cross-sectional area of the phe-
nylpyridine-bridged upper rim is much larger than the cross-section-
al area of the sulfide chains at the lower rim, which makes a well-or-
dered packing on the surface impossible (see also ref. [9]). Surface
absorption by self-assembly results in disordered monolayers, on
which it was not possible to perform cage assembly. −Molecular
ruler× experiments, as described in ref. [4], indicated no height dif-
ference after the cage assembly step.
[6] A. Friggeri, H.-J. van Manen, T. Auletta, X.-M. Li, S. Zapotoczny,
H. Schˆnherr, G. J. Vancso, J. Huskens, F. C. J. M. van Veggel, D. N.
Reinhoudt, J. Am. Chem. Soc. 2001, 123, 6388 6395.
[16] A. Ulman, Chem. Rev. 1996, 96, 1533 1554.
[17] L. Tunstad, J. A. Tucker, E. Dalcanale, J. Weiser, J. A. Bryant, J. C.
Sherman, R. C. Helgeson, C. B. Knobler, D. J. Cram, J. Org. Chem.
1989, 54, 1305 1312.
[18] a) S. Fallis, G. K. Anderson, N. P. Rath, Organometallics 1991, 10,
3180 3184; b) P. J. Stang, D. H. Cao, S. Saito, A. M. Arif, J. Am.
Chem. Soc. 1995, 117, 6273 6283.
[7] For cavitand-based coordination cages see: a) P. Jacopozzi, E. Dalca-
nale, Angew. Chem. 1997, 109, 665 667; Angew. Chem. Int. Ed.
Engl. 1997, 36, 613 615; b) O. D. Fox, N. K. Dalley, R. G. Harrison,
J. Am. Chem. Soc. 1998, 120, 7111 7112; c) O. D. Fox, N. K. Dalley,
R. G. Harrison, Inorg. Chem. 1999, 38, 5860 5863; d) O. D. Fox,
M. G. M. Drew, P. D. Beer, Angew. Chem. 2000, 112, 139 144;
Angew. Chem. Int. Ed. 2000, 39, 136 140; e) F. Fochi, P. Jacopozzi,
E. Wegelius, K. Rissanen, P. Cozzini, E. Marastoni, E. Fisicaro, R.
Received: September 24, 2003
Revised: December 19, 2003 [F5570]
2206
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2199 2206