15-Membered triolefinic macrocycles as stabilizers of palladium(0) nanoparticles

Article abstract of DOI:10.1039/b608673j

15-Membered triolefinic macrocycles form complexes with palladium(0). However, metal nanoparticles are formed instead of discrete complexes if substituents provided with the ability to stabilize nanoparticles are incorporated into the structure of the mac

Full text of DOI:10.1039/b608673j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
15-Membered triolefinic macrocycles as stabilizers of palladium(0)
nanoparticlesw
Anna Serra-Muns, Roger Soler, Elena Badetti, Paula de Mendoza, Marcial
´
˜
Moreno-Manas,z Roser Pleixats,* Rosa M. Sebastian* and Adelina Vallribera*
Received (in Montpellier, France) 19th June 2006, Accepted 24th July 2006
First published as an Advance Article on the web 31st August 2006
DOI: 10.1039/b608673j
15-Membered triolefinic macrocycles form complexes with palladium(0). However, metal
nanoparticles are formed instead of discrete complexes if substituents provided with the ability to
stabilize nanoparticles are incorporated into the structure of the macrocycle. Ancillary
substituents include fluorous and polyoxyethylenated chains. The role of initial organometallic
coordination in the early steps of nanoparticle formation is underlined.
(iii) entrapment: in polymers (e.g. polyvinylpyrrolidone), cy-
Introduction
clodextrins, or dendrimers, although electrostatic stabilization
also operates with polymers.11 In all cases the protecting
agents ought to interact in an attractive manner with the
surface of the metal.
The 15-membered triolefinic macrocycles 1 (Fig. 1) form stable
palladium(0) and platinum(0) complexes 21 and are reminis-
cent of the cyclic butadiene trimers, the 12-membered
cyclododeca-1,5,9-trienes 3, which coordinate nickel(0).2
Coordination of one metal atom in macrocycles 1 takes place
by the three olefinic double bonds, the nitrogen atoms of
sulfonamides being devoid of coordinating ability. Macro-
cycles 1 can be prepared from trans-1,4-dibromobutene and
aromatic sulfonamides. Since many sulfonamides or their
related sulfonyl chlorides are commercially available, the
properties of compounds 1—solubility, crystallinity, electro-
chemical properties—can be tailored varying the aromatic
ring.1 Moreover, macrocycles 1 or their palladium complexes
have found use in heterogeneous catalysis when anchored to a
mesostructured silica,3 and as liquid crystals.4 The 15-mem-
bered cyclic triacetylenes related to 1 also coordinate palla-
dium(0) and cycloisomerize to hexasubstituted benzenes under
metal catalysis.5 In summary, a vast array of properties can be
conferred to macrocycles 1 depending on the nature of the
substituents in the aromatic ring, while maintaining the co-
ordinating ability.
Ketone 4 is a fluorous analogue of dibenzylideneacetone,
dba, (H instead of C8F17), the ligand of palladium in the family
of stable complexes Pd2(dba)3 ꢀ L (L = solvent or dibenzylide-
neacetone). When we attempted to reproduce the preparation
of Pd2(dba)3 ꢀ L with ketone 4, we obtained palladium nano-
particles instead of the expected discrete complex.6a,b
We thought that the different outcome of the reac-
tions—discrete complex from dba and nanoparticles from
4—was due to the fluorinated chains. This poses the intellec-
tual problem of understanding the reason for the difference.
Since then, we felt that initial coordination of an initial metal
atom was helpful for the formation of nanoparticles albeit not
necessarily, since also other non-coordinating fluorous com-
pounds do stabilize nanoparticles.6,7 Although kinetic work
on the growing of nanoparticles has been published,11,12 little
is known on the process of binding together the first few metal
atoms, e.g. less than ten.13
Coordinating macrocycles 1 fulfil the requirements to test
our idea that initial coordination of the metal with the double
bonds promoted the formation of nanoparticles if appropriate
stabilizing groups were bound to the aromatic rings. We have
chosen macrocycles bearing long fluorinated chains, which
could provide more examples of the curious stabilizing effect
of these chains and macrocycles having polyoxyethylenated
chains because stabilization of palladium nanoparticles by
Some time ago we found in a truly serendipitous manner
that heavily fluorinated ketone 4 (Fig. 1) stabilized palladium
nanoparticles.6 Other fluorous compounds, even fluorinated
hydrocarbons, stabilize nanoparticles of palladium,6b,c gold6d,7
or metal oxides.8 Stabilization of nanoparticles is customarily
divided into three categories:9,10 (i) electrostatic stabilization
by cationic and anionic surfactants; (ii) steric stabilization by
compounds possessing a functional group endowed with high
affinity for metals: thiols, sulfides, amines and phosphines; and
ligands
containing
polyoxyethylenated
chains
has
´
Department of Chemistry, Universitat Autonoma de Barcelona,
Cerdanyola, 08193 Barcelona, Spain. E-mail: roser.pleixats@uab.es;
Fax: +34 935811265; Tel: +34 935812067
w Electronic supplementary information (ESI) available: Full experi-
mental details on the preparation of macrocycles 5, 11 and 12. 1H and
13C NMR of 11 and 17, ESI-MS of 11, MALDI-TOF MS of 17. High
Resolution Transmission Electron Microscopy (HRTEM) and Elec-
tron Diffraction (ED) of nanoparticles of Entries 2, 4–13. p-XRD of 7
and nanoparticles of Entry 9. See DOI: 10.1039/b608673j
z Deceased on 20th February 2006.
Fig. 1 Structures of macrocycle 1 and other pertinent compounds.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1584 | New J. Chem., 2006, 30, 1584–1594

View Article Online
precedents.14 Moreover, hydrogenation of the three double
bonds of products of type 1 give another series of macrocycles
which obviously cannot form metal complexes of type 2.
Consequently, hydrogenated macrocycles should be less effec-
tive towards formation of nanoparticles.
The most used method for the formation of palladium(0)
nanoparticles involves reduction of a palladium(II) salt by an
appropiate reducing agent in the presence of a stabilizer.9
However, in a preliminary work aimed at the preparation of
macrocyclic palladium complexes 2 using our standard meth-
odology,1 which involved the use of Pd(dba)2 as a palla-
dium(0) source, we had observed the formation of black
solids containing metal nanoparticles when fluorous or poly-
oxyethylenated chains were present in the macrocycle.
Therefore, we decided to explore further this new method
for the preparation of palladium(0) nanoparticles and to
determine the role of initial organometallic coordination in
the formation of palladium nanoparticles using a new type of
stabilizing macrocyclic compound possessing the above men-
tioned polyfluorinated and polyoxyethylenated chains.
Scheme 2 Preparation of palladium(0) nanoparticles. See Table 1 and
the Experimental section.
nanoparticles. For the sake of clarity in the schematic pre-
sentation of results, we summarize the synthesis of the differ-
ent macrocycles in Scheme 1, the experiments leading to
palladium nanoparticles in Scheme 2 and the obtaining of
palladium(0) macrocyclic triolefinic complexes in Scheme 3.
Pivotal compound 515 and its conversion into fluorous macro-
cycle 616 have been reported elsewhere. We have now intro-
duced a significant improvement in the preparation of
macrocycles of type 1 in general and 5 in particular: trans-
1,4-dichloro-2-butene is used instead of the corresponding
dibromo compound. In one step of the synthesis a large excess
of dihalogenobutene is required and elimination of the low-
boiling dichlorobutene is much easier than elimination of
Results and discussion
Thus, we prepared 6, 7, 11 and 12 (Scheme 1) as candidates for
nanoparticle stabilization. We synthesized also hydrogenated
macrocycles 9 and 10 (Scheme 1), which should not form
Scheme 1 Preparation of macrocycles; reagents and conditions. (i)
To obtain 6 and 9: HSCH2CH2C8F17, Cs2CO3, n-BuNCl, THF, room
temperature. (ii) To obtain 7 and 10: HO(CH2CH2O)45CH3, NaH,
THF, room temperature. (iii) Pd–C 10%, PtO2 ꢀ H2O, THF, room
temperature.
Scheme 3 Preparation of macrocyclic palladium(0) complexes. Re-
agents and conditions: (i) PEG-2000 monomethyl ether, NaH, THF,
room temperature; (ii) Pd(dba)2 or Pd(PPh3)4, refluxing THF.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1584–1594 | 1585

View Article Online
dibromobutene. The new preparation of 5 is fully described in
the Supplementary Information.w Compound 6 was prepared
by nucleophilic aromatic substitution of fluoride with
1H,1H,2H,2H-perfluorodecanethiol. The same type of reac-
tion performed with polyethylene glycol 2000 (PEG 2000)
monomethyl ether afforded 7. Compounds 9 and 10, devoid
of olefinic coordinating ability, were prepared in two steps
consisting of catalytic hydrogenation of 5 into 8 followed by
SNAr-type reaction16 with 1H,1H,2H,2H-perfluorodecane-
thiol or polyethylene glycol 2000 (PEG 2000) monomethyl
ether to afford 9 and 10, respectively. Macrocycles 11 and 12
(Scheme 1) were prepared by standard methods.w
As we have mentioned before, macrocycles having many
different substituents in the arenesulfonyl moiety form discrete
stable palladium(0) complexes when treated with sources of
palladium(0) such as Pd2(dba)3 ꢀ L or Pd(PPh3)4.1,17 In con-
trast, macrocycles 6, 7 and 12 —but not 11—afforded directly
metal nanoparticles by the same treatment (Scheme 2, Table
1). The nanoparticles prepared in our work are frequently
stabilized by macrocycles possessing already a coordinated
central palladium atom, in other words, the stabilizing shield is
made of complexes 13 or 14; these materials are denoted as
Pdn/(13) or Pdn/(14). But other metal nanoparticles were
surrounded by free macrocycles and denoted as Pdn/(6),
Pdn/(7) and Pdn/(12). In the first situation two types of
palladium atoms are present in the material, some atoms form
the nanoparticles, whereas some others are coordinated to the
macrocycles in the protecting shield. We have also found that
under some conditions palladium(0) nanoparticles were stabi-
lized by both free macrocycles and their palladium(0) com-
plexes, such as in the case of Pdn/(6 + 13) and Pdn/(12 + 15)
(see Scheme 2). NMR spectra distinguish very well free
macrocycles from their palladium(0) complexes since free
macrocycles present signals for the olefinic protons and car-
bons at d ca. 5.50–5.80 and ca. 124 ppm, whereas in palladium
complexes strong upfield shifts are observed, up to
d
2.5–4.40—depending on the proton considered—and 78–83
ppm, respectively (see the Experimental section and Supple-
mentary Informationw). The complexity of the NMR spectra
of palladium(0) complexes of macrocycles 1 stems from the
generation of chirality on the olefinic carbon atoms upon
complexation. This is better understood if the palladacyclo-
propane formulation is adopted. Depending on which face of
the olefin coordinates, the chirality will be R,S or S,R in the
former olefinic carbon atoms. Since the three olefins do not
give rise to identical chirality on coordination, this produces
non-equilibrating isomers when two aryl groups are identical
and the third is different. This problem has been thoroughly
discussed.17b
When fluorous macrocycle 6 was treated with Pd(dba)2 in
refluxing THF nanoparticles containing complex 13 in the
stabilizing shield were formed (Entries 1–3, Table 1). This
behaviour is in sharp contrast to the general behaviour of
macrocycles 1 that afford stable palladium(0) complexes of
type 2 without problems.1,17 Nanoparticles of Entries 1 and 2
presented signals in 1H-NMR for the olefinic protons only at d
ca. 2.84, 3.76, 4.02, indicating clearly the presence of a
palladium atom coordinated to the three olefins of the macro-
cycle. Free macrocycle 6 is present up to 30% in the stabilizing
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1586 | New J. Chem., 2006, 30, 1584–1594

View Article Online
shield of nanoparticles of Entry 3 as evidenced by a signal at
d = 5.62. When more severe conditions were adopted (heating
at 140 1C in a closed reactor for about one day, Entry 4), the
nanoparticulated material showed only the presence of macro-
cycle 6 as a stabilizing agent.
A sample of complex 13 (see the structure in Scheme 2)
could be obtained when the black solution from the reaction of
macrocycle 6 with Pd(dba)2 in refluxing THF was directly
chromatographed through silica gel. The chromatography
decomposes the nanoparticulated material, the non-coordi-
nated palladium being retained on top of the column.
Nanoparticles of Entries 1 (Fig. 2) and 3 (Fig. 3) are of good
quality, small and disperse. Once more fluorous compounds
stabilize small nanoparticles. HRTEM permits the observa-
tion of details of the particles. Nanoparticles of Entry 2
(Supplementary Informationw) are more agglomerated
although in general are well defined.
Compound 11 (Scheme 1) did not form nanoparticles,
probably the fluorine loading was insufficient. Instead, com-
plex 17 (see Scheme 3 for the preparation of macrocyclic
palladium complexes) was isolated without problems when it
was treated with Pd(dba)2 or Pd(PPh3)4 in refluxing THF.
In other words, macrocycle 11 exhibits the standard coor-
dination ability characteristic of this family of 15-membered
macrocycles.17
Fig. 3 Nanoparticles Pdn/(6 + 13) of Entry 3. (a) TEM image (scale
bar = 20 nm); (b) size distribution (average 2.94 ꢃ 0.39 nm); (c)
HRTEM (scale bar = 5 nm); (d) electron diffraction (0.23 nm between
Pd layers).
nanoparticles presented a certain tendency to agglomerate.
Similar results were obtained when fluorous macrocycle 12
was treated with Pd(PPh3)4 at room temperature for several
days (Entry 6). In contrast, when 12 was reacted with Pd(dba)2
at higher temperature (140 1C in a closed reactor) only the free
macrocycle 12 was observed by NMR in the stabilizing shield
(Entry 7).
In contrast, macrocycle 12 (Scheme 1) has enough fluorous
character to stabilize nanoparticles and direct formation of
nanoparticulated material was found upon treatment with a
palladium(0) source, whereas isolation of complex 15 was not
possible (see Pdn/(12 + 15) and Pdn/(12) in Scheme 2 and
Table 1). Thus, in the experiment performed with 12 and
Pd(dba)2 in refluxing THF, the stabilizing shield was made of a
93 : 7 mixture of complex 15 (olefinic signals at d = ca. 2.81,
3.80, 4.06) and free ligand 12 (d = 5.64) (Entry 5). The
Fluorous compound 9 (Scheme 1), devoid of coordinating
ability, does not form nanoparticles under the same experi-
mental conditions used for the triolefinic fluorous macrocycle
6. This reinforces our hypothesis: the initial coordination of a
first palladium atom is a favourable factor for nanoparticle
formation.
Polyoxyethylenated triolefinic macrocycle 7 (Scheme 1) is
also a good stabilizer of palladium(0) nanoparticles. Thus,
when 7 was treated with Pd(dba)2 in refluxing THF, nano-
particles Pdn/(14) were formed featuring complex 14 in the
stabilizing shield (Entry 8). Similar results were found by
treatment of 7 with Pd(PPh3)4 or Pd(dba)2 in THF at room
temperature (Entries 11 and 12 of Table 1). But if the
temperature is raised to 140 1C in a closed reactor or if the
solvent is permitted to dry (ca. 110 1C of bath temperature),
the palladium coordinated with the olefins is lost and Pdn/(7)
material is formed with macrocycle 7 as stabilizer (Entries 9
and 10 of Table 1).
Thus, for both fluorous and polyoxyethylenated macro-
cycles the effect of the temperature on the outcome of the
process appears to be the same: nanoparticles are formed in
refluxing THF and even at room temperature, the correspond-
ing macrocyclic palladium(0) complex being the only or the
major stabilizer present in the protecting shield in these cases.
At higher temperatures (140 1C) the palladium coordinated to
the olefins is lost and the stabilizing shield of the nanoparticles
is formed by the free triolefinic macrocycle. Thus, the initially
Fig. 2 Nanoparticles Pdn/(13) of Entry 1 (Table 1). (a) TEM image
(scale bar = 20 nm); (b) size distribution (average 3.93 ꢃ 0.50 nm); (c)
HRTEM (scale bar = 5 nm); (d) electron diffraction (0.23 nm between
Pd layers).
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1584–1594 | 1587

View Article Online
Fig. 4 Sequence of events in the formation of nanoparticles Pdn/(6) and Pdn/(13).
coordinated palladium atoms form metal nanoparticles upon
heating, although partial decomposition to black palladium
cannot be ruled out, which should be eliminated by filtration
through Celite or PTFE filter (see Experimental). Fig. 4
describes a sequence of steps compatible with these observa-
tions.
5.5 ppm had been observed in the Pdn/(14) nanoparticles. In
contrast, this signal is absent in the complex 14 formed from
complex 16 (Scheme 3). We presume that the complex stabiliz-
ing the palladium nanoparticles in Entries 8, 11 and 12 of
Table 1 is not 14, but a related complex in which the olefins are
partially decomplexed and probably the polyoxyethylenated
chain acts as a coordinating agent. This would explain the
unexpected thermal stability of 14 and that formation of
nanoparticles is not observed by thermal treatment of this
complex.
When a THF solution of fluorous macrocyclic complex 13
was heated overnight at 140 1C in a sealed tube, a grey solid
containing nanoparticles stabilized by the corresponding free
ligand, Pdn/(6), was obtained (Scheme 2). This fact seems to
confirm in this case the sequence of events described in Fig. 4,
the formation of nanoparticles from the palladium(0) complex
and then the loss of the coordinated palladium due to the high
temperature.
Moreover, we tested the behaviour of the polyoxyethy-
lenated compound 10 (Scheme 1), devoid of a triolefinic
coordinating core. We found that it was able to form palla-
dium(0) nanoparticles, Pdn/(10), by reaction with Pd(dba)2 in
refluxing THF, although the content of metal is low (Entry 13
of Table 1). This is in sharp contrast with the behaviour of
fluorous compound 9. The coordinating ability of the oxygen
atoms present in the polyoxyethylenated chains should be the
reason for such a difference. In the polyoxyethylenated macro-
cycles the triolefinic core is thus not essential for nanoparticle
formation.
Pure macrocyclic polyoxyethylenated palladium complex 14
is not available under the usual conditions, because nanopar-
ticles Pdn/(14) are formed as we have mentioned before.
Alternatively, compound 14 has been obtained by an SNAr-
type reaction of complex 1618 with PEG-2000 monomethyl
ether in THF at room temperature in the presence of sodium
hydride (Scheme 3).
In contrast with the behaviour of fluorous macrocyclic
complex 13, polyoxyethylenated macrocyclic complex 14 re-
sulted to be more stable and no decomplexation and no
nanoparticle formation was observed after the heating for
one day of a THF solution of 14 at 140 1C. After addition
of Pd(dba)2 to this solution and further heating at 140 1C for
about 30 h, partial decomplexation was observed by 1H NMR
(70% of 14, 30% of 7), but no nanoparticles were detected by
HRTEM. Therefore, for triolefinic macrocycles bearing poly-
oxyethylenated chains the sequence of events seems to be
different from that in the fluorous macrocycles and probably
the initial coordination to the three carbon–carbon double
bonds of the triolefinic core is not the first step in the process
of nanoparticle formation. This thermal stability was intri-
guing us after the observation of nanoparticles in Entries 8–10
In all cases where nanoparticles were formed, electron
diffraction of the samples showed the characteristic pattern
of face-centred cubic palladium(0) with the average d-spacing
values given in the electronic Supplementary Information.w
In our previous studies of formation of metal nanoparticles
(Pd, Au)6c,d with fluorinated stabilizers, powder X-ray diffrac-
tion provided us with significant information. Solid fluori-
nated stabilizers showed well-defined diffractograms in the low
angle range 2–20 of the 2y scale, which indicated a degree of
order that was basically preserved in the nanoparticles, which
presented diffractograms showing similar ordered structures.
This observation made us to postulate a mechanism of stabi-
lization based on the entrapment of the palladium and gold
nanoparticles within the fluorinated compound. In this work,
we have also performed the powder X-ray diffraction of
polyoxyethylenated triolefinic macrocycle 7 and the nanopar-
ticles Pdn/(7) of Entry 9 of Table 1 (See Supplementary
Informationw). Compound 7 exhibits diffraction peaks indi-
cating a long-range ordered structure and the same crystalline
phases of this polyoxyethylenated compound exist in the
related nanoparticle sample Pdn/(7). Therefore, we assume a
similar mechanism of stabilization for the fluorous and the
polyoxyethylenated macrocycles described in this work, which
should be an entrapment of the palladium(0) nanoparticles
within the solid framework of the mentioned compounds.
Long-range ordered structures for compounds bearing
1
of Table 1. For this reason, we monitored by H NMR the
reaction of 14 and Pd(dba)2 (molar ratio 1 : 2.2) in THF at
140 1C in a sealed tube. After 20 min signals corresponding to
coordinated olefins at 4.60 (dd) were present, together with a
signal at 5.55 (apparent d) corresponding to free olefins.
Similar results were obtained after 3 h and after 5 h. After
20 h of reaction the signal at 4.6 ppm corresponding to the
coordinated olefins was still present, together with a broad
singlet at 5.59 (corresponding to free olefins of macrocycle 7),
which was of higher intensity than another broad singlet at
5.52 ppm. The presence of a residual signal of free olefins at ca.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1588 | New J. Chem., 2006, 30, 1584–1594

View Article Online
polyoxyethylenated19 and fluorinated chains6c,d,20 have been
previously reported. These ordered structures provided by
these type of chains must be in the origin of the competitive
formation of nanoparticles when palladium(0) complexation
with the macrocycles 6, 7 and 12 is tested. The 1H NMR
spectra allows us to know the nature of the stabilizing species
(free macrocycle or macrocyclic complex) present in the solid
nanomaterial and in solution. While an entrapment stabiliza-
tion mechanism is postulated for the solid material, the kind of
interaction acting in solution remains unclear.
in 79% yield in the first run. However, reactions with 4-
bromoacetophenone failed. Recovery of the catalytic material
was achieved by addition of diethyl ether followed by centrifu-
gation and separation of the organic layer. The insoluble
nanoparticles were introduced in the following reaction.
Conclusions
15-Membered triolefinic macrocycles 1 are organometallic
coordinating agents through the olefinic bonds.17 This stan-
dard behavior is modified if substituents that stabilize nano-
particles (fluorous chains and polyoxyethylenated chains) are
introduced in the aromatic sulfonamide moieties of the macro-
cycles. In such cases nanoparticles between 2.9 and 5.6 nm of
diameter are formed. Therefore, we formulate this hypothesis:
nanoparticle formation is favoured if the stabilizing molecule
has the ability to coordinate an initial metal atom. This ability
is secured in the fluorous macrocycles by the triolefinic core. In
the polyoxyethylenated macrocycles these chains contain co-
ordinating oxygen atoms, the triolefinic core being non-essen-
tial for nanoparticle formation. However higher content of
palladium(0) in the nanoparticulated material have been
found when the triolefinic macrocycle is present.
Metal nanoparticles in general and palladium nanoparticles
in particular, have found application in catalysis due to their
high specific surface.9p We have described fluorous biphasic
systems that permitted recovery and reutilization of palladium
nanoparticles as catalysts in Suzuki cross-coupling and in
Mizoroki–Heck reactions.6a Now, we have tested two batches
of nanoparticles stabilized by macrocycles featuring different
stabilizing chains. Our experiments are not biphasic; however,
the catalytic material was recovered in a different manner and
reutilized. Thus, nanoparticles Pdn/(13) of Entry 2 catalyzed
the Mizoroki–Heck reaction of iodobenzene with n-butyl
acrylate (Scheme 4). Excellent conversions of iodobenzene
were secured in six consecutive runs using the recovered
sample of nanoparticulated material from the previous run.
Recovery was achieved by evaporation of the mixture and
extraction of all reaction products with cold acetonitrile. The
insoluble nanoparticles were used in the following reaction.
Butyl cinnamate was isolated in 94% yield in the third run.
The loss of activity of the catalytic material was evident in the
sixth run. The recovered material from the fifth run was
examined by HRTEM showing particles of about 100 nm.
Whereas this size is considered in the limit of the nanometric
scale, agglomeration of the original catalyst was evident,
which can explain the loss of activity. On the other hand the
recovered catalytic material was Pdn/(6) (1H-NMR monitor-
ing) or in other words, free macrocycle 6 was the only
stabilizing compound around the metal nanoparticles.
In general, smaller size nanoparticles are obtained when
fluorous macrocycles are used as stabilizing agents. The way of
addition of the palladium(0) source or the amount of metal
added have small effect on the size of the nanoparticles.
Our method of forming nanoparticles is different from other
methods currently used that are based on: (i) the reduction of a
high valence metal salt in the presence of the stabilizer; (ii) the
hydrogenation of an unsaturated ligand in a metal(0) complex; or
(iii) vapour metal deposition. Our method, here described, is
based in the interchange of ligand between a discrete complex
(Pd(dba)2 or Pd(PPh3)4) and macrocycles of type 1. Indeed
macrocycles 1 are better ligands than dba and therefore, transfer
of the initial palladium atom to the macrocycle is favoured.1 In
the case of Pd(PPh3)4, triphenylphosphine is a better ligand than
macrocycles 1.1 However, transfer of metal to the macrocycle is
favoured by oxidation of the phosphine to its oxide, a process
difficult to avoid even by working under an inert atmosphere.
Two of these new nanoparticuled materials have been tested
as catalyst or precatalyst of a Mizoroki–Heck reaction in
homogeneous media, being easily recovered and reused. Sev-
eral cycles have been performed and small variations in their
catalytic activity have been observed. These phosphine-free
palladium catalysts offer the advantage of superior stability in
the face of oxidation. Phosphines are readily oxidized to the
corresponding phosphine oxides, which can prevent the easy
recovery and recycling of the catalyst. Further experiments
will be done in the future in order to study more deeply the
catalytic activity of these new materials.
Nanoparticles Pdn/(7) of Entry 10 containing 3.25% of metal
were tested also in the Mizoroki–Heck reaction of iodobenzene
with n-butyl acrylate (Scheme 4). Excellent conversions of
iodobenzene were secured in five consecutive runs using the
same sample of nanoparticulated material, the reaction time
being the same in all runs (24 h). Butyl cinnamate was isolated
Experimental
General methods and materials
Elemental analyses for C, H, N, S and Pd are the average of
two determinations. IR spectra were determined by the Atte-
nuated Total Reflectance (ATR) mode.
Scheme 4 Heck reaction catalyzed by Pdn/(13) (Entry 2) and Pdn/(7)
nanoparticles (Entry 10).
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1584–1594 | 1589

View Article Online
HRTEM analyses were performed in the Servei de Micro-
scopia of the Universitat Autonoma de Barcelona, in a Hitachi
1,6,11-Tris[4-(1H,1H,2H,2H-perfluorodecylthio)phenylsulfo-
nyl]-1,6,11-triazacyclopentadecane, 9. Anhydrous cesium car-
bonate (1.5 g, 4.6 mmol), was warmed up at 50 1C in vacuum
(ca. 1 mmHg) for 30 min. After cooling under an argon
atmosphere THF (5 mL) and 1H,1H,2H,2H-perfluorodeca-
nethiol (0.75 mL, 2.5 mmol) were added. Then, a solution of 8
(360 mg, 0.054 mmol) and tetrabutylammonium chloride (25
mg, 0.086 mmol) in THF (6 mL) was added via cannula with
stirring under an inert atmosphere. The dark-orange mixture
was stirred overnight at room temperature. The solvent was
eliminated by means of a nitrogen stream. The residue was
washed by a solvent system made of perfluorooctanes (FC-77)
(20 mL), ethyl acetate (20 mL) and water (15 mL) with
vigorous stirring for 1 h. The washed solid was filtered and
recrystallized from THF and finally washed again with more
FC-77 to afford 808 mg (74%) of 9 as a white powder, mp
´
´
H-7000 model at 100 kV and in a JEOL JEM-2010 model at
200 kV.
The HRTEM measurements were made by sonication of the
nanoparticulate material in hot THF for about ten minutes;
then one drop of the finely divided suspension was placed on a
specially produced structureless carbon support film having a
thickness of 4–6 nm and dried before observation.
p-XRD measurements were carried out on a Siemens D5000
diffractometer with an area detector operating under Cu Ka
(1.5406 A) radiation, at the Laboratori de Cristal.lografia i
Raigs X at the Institut de Ciencia de Materials de Barcelona.
Compounds 515, 616 and 1718 have been previously reported.
The preparation of macrocycles 11 and 12 is described in the
Supplementary Information.w HRTEM and electron diffrac-
tion images for materials of Entries 2 and 4–13 of Table 1 and
p-XRD of 7 and nanoparticles of Entry 9 of Table 1 are
available as Supplementary Information.w
164–186 1C. IR (ATR): 1333, 1201, 1149 cmꢄ1 1H-NMR
;
(CDCl3, 250 MHz): d = 1.71 (br s, 12 H), 2.34–2.60 (m, 6H),
3.08 (br s, 12H), 3.18–3.27 (m, 6H), 7.38 (d, J = 8.6 Hz, 6H),
7.70 (d, J = 8.8 Hz, 6H). Anal. calcd for C60H48F51N3O6S6: C,
34.84; H, 2.34; N, 2.03, S, 9.30; found: C, 34.73; H, 2.08; N,
1.93; S, 9.49.
Syntheses of macrocycles
(E,E,E)-1,6,11-Tris[4-(methoxy(polyoxyethylene)phenylsulfonyl
]-1,6,11-triazacyclopentadeca-3,8,13-triene, 7. A solution of PEG-
2000 monomethyl ether (1.820 g, ca. 0.91 mmol) in THF (5 mL)
was added under inert atmosphere over a stirred suspension of
sodium hydride (59 mg of 60% suspension in mineral oil, 1.47
mmol) in THF (1 mL). The mixture was stirred for 10 min and
then a solution of macrocycle 5 (200 mg, 0.29 mmol) in
anhydrous THF (10 mL) was added dropwise. The stirring was
continued for one day. The solvent was evaporated and chloro-
form was added. The suspension was filtered through a wool pad
and the chloroform was evaporated to afford a brown solid that
was washed with hexane to afford 1.6 g (ca. 82%) of macrocycle
7 contaminated with PEG-2000 monomethyl ether; IR: 2885,
1,6,11-Tris[4-methoxy(polyoxyethylene)phenylsulfonyl]1,6-11-
triazacyclopentadecane, 10. A solution of PEG-2000 mono-
methyl ether (3.5 g, ca. 1.75 mmol) in anhydrous THF (10
mL) was added under an inert atmosphere over a stirred
suspension of sodium hydride (113 mg of 60% suspension in
mineral oil, 2.82 mmol) in THF (2 mL). The mixture was stirred
for 10 min and then a solution of macrocycle 8 (388 mg, 0.56
mmol) in THF (20 mL) was added dropwise. The mixture was
stirred overnight. An excess of NaH was added in two portions
(2 ꢂ 56 mg, 1.41 mmol) in order to complete the reaction during
the next two days. After filtration, the solvent was evaporated
to afford a brown solid that was washed with hexane to afford a
pink solid of 10 (3.80 g, 98%). IR: 2879, 1593, 1466, 1341, 1101
cmꢄ1; 1H-NMR (CDCl3, 250 MHz): d = 1.69 (s, 12H), 3.02 (s,
12H), 3.38 (s, 12H), 3.56–3.64 (m, 578H), 3.87–3.90 (m, 12H),
4.18 (m, 6H), 6.99 (d, J = 8.5 Hz, 6H), 7.7 (d, J = 9.0 Hz, 6H);
MALDI-TOF MS: m/z 5900 (39 ꢂ 3 CH2CH2O units) to 7440
(50, 51, 51 CH2CH2O units) separated by 44 Da [M + Na+],
the more intense peak occurs at 6561 (44 ꢂ 3 CH2CH2O units).
1
2859, 1591, 1337, 1149, 1106, 1056, 955, 840 cmꢄ1; H-NMR
(CDCl3, 250 MHz): d = 3.34 (s, 9H), 3.50–3.69 (broad absorp-
tion, ca. 522 H), 3.85–3.91 (m, 12H), 4.16 (t, J = ca. 4.4 Hz, 6H),
5.54 (br s, 6H), 6.97 (d, J = 8.8 Hz, 6H), 7.66 (d, J = 8.7 Hz,
6H); MALDI-TOF MS: m/z 5982 (40 ꢂ 3 CH2CH2O units) to
7348 (50 ꢂ 3 CH2CH2O units) separated by 44 Da [M + Na+],
the more intense peak occurs at 6728 (45 ꢂ 3 CH2CH2O units).
Preparation of palladium(0) nanoparticles
1,6,11-Tris[4-fluorophenylsulfonyl]-1,6,11-triazacyclopentadec-
ane, 8. A stirred mixture of macrocycle 5 (519 mg, 0.76 mmol),
10% Pd–C (24 mg) and platinum(IV) oxide monohydrate (43
mg) in THF (15 mL) was hydrogenated at atmospheric pressure
for 2 h (1H NMR monitoring). The mixture was filtered through
Celite that was washed with more THF (3 ꢂ 15 mL). The
solvent was evaporated and the white solid residue was washed
with pentane (30 mL) and dried to afford 486 mg (93%) of 8,
mp 191–193 1C; IR (ATR) 2954, 2867, 1590, 1492, 1334, 1156,
1092, 841, 692 cmꢄ1; 1H-NMR (CDCl3, 250 MHz): d = 1.72 (br
s, 12H), 3.07 (br s, 12H), 7.20 (apparent t, J = 8.6 Hz, 6H), 7.79
(dd, J = 9 and 5 Hz, 6H); 13C-NMR (CDCl3, 62.5 Mz): d =
27.2, 50.5, 116.7, (d, J = 22.9 Hz), 130.2 (d, J = 9.5 Hz), 134.8
(d, J = 3.8 Hz), 165. 4 (d, J = 253 Hz); Anal. calcd for
C30H36F3N3O6S3: C, 52.39; H, 5.28; F, 8.29; N, 6.11; S, 13.99;
found: C, 52.17; H, 5.08; N, 6.01; S, 13.89.
Nanoparticles Pdn/(13) (Entry 1). A solution of Pd(dba)2 (89
mg, 0.154 mmol) in THF (10 mL) was added via syringe with
stirring under an inert atmosphere into a boiling solution
of (E,E,E)-1,6,11-tris[4-(1H,1H,2H,2H-perfluorodecylthio)-
phenylsulfonyl]-1,6,11-triazacyclopentadeca-3,8,13-triene,
6
(205 mg, 0.097 mmol) in THF (70 mL). The mixture was
refluxed with stirring. After three days more Pd(dba)2 (47 mg,
0.082 mmol) in THF (4 mL) was added. The mixture was
heated to reflux for two more days. The black solution was
then filtered first through Celite and then through a pad of
PTFE of 0.2 mm of filter pore. The filtrate was completely
evaporated with a stream of nitrogen to afford 330 mg of a
black solid containing dba and traces of 6 (1H-NMR monitor-
ing) and palladium nanoparticles (HRTEM monitoring). The
solid (200 mg) was digested for two days with chloroform
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1590 | New J. Chem., 2006, 30, 1584–1594

View Article Online
(14 mL). The suspension was filtered through PTFE of 0.2 mm
of filter pore and the solid was washed with cold chloroform (3
ꢂ 2 mL) and dried to afford 112 mg (70% yield) of black Pdn/
(13). The IR and 1H-NMR spectra were identical with those of
complex 13 (vide infra). Anal. calcd for C60H42F51N3O6S6Pd2
(Pd1131): C, 31.67; H, 1.86; N, 1.85; S, 8.46; F, 42,59; Pd, 9.36;
found: C, 31.12; H, 1.63; N, 1.81; S; 8.74; F, 42.81; Pd 10.2 ꢃ
0.2 (ICP analysis).
pentane and dried to afford 50 mg of brown nanoparticles of
composition Pdn/(12 (8%) + 15 (92%)).
Nanoparticles Pdn/(12) (Entry 7). A mixture of macrocycle
12 (50 mg, 0.027 mmol), Pd(dba)2 (36 mg, 0.064 mmol) and
THF (1 mL) was heated overnight at 140 1C in a closed
reactor. Half of the solvent was evaporated and 5 mL of
pentane were added. The mixture was stirred for 10 min and
the solid was filtered and washed twice more with a mixture of
THF : pentane (0.5 : 5 mL) in order to remove the last traces of
dba. The black solid was dried and 42 mg of nanoparticles of
composition Pdn/(12) were obtained. IR and 1H-NMR spectra
were identical with that of 12.
Nanoparticles Pdn/(13) (Entry 2). The same procedure as for
Entry 1 but different working up were followed. After the five
reaction days the solution was filtered through the PTFE filter.
The filtrate was concentrated to 20 mL and left two days at
ꢄ20 1C. The resulting black precipitate was filtered and
washed with cold THF (2 ꢂ 5 mL) to afford 239 mg of
nanoparticles (58%). Anal. found: C, 33.45; H, 1.81; N,
1.96; S, 9.37; Pd, 6.8.
Nanoparticles Pdn/(14) (Entry 8). A mixture of macrocycle 7
containing PEG-OMe (300 mg,o0.045 mmol), Pd(dba)2 (57
mg, 0.104 mmol) and THF (2.3 mL) was refluxed for one day.
Then the mixture was filtered through Celite using THF as
eluent. The solvent was evaporated to afford a black solid that
was washed with hexane and dissolved in water. The aqueous
solution was partitioned again with hexane to remove the last
traces of dba. The water was finally evaporated at reduced
pressure and the resulting solid was washed once again with
hexane to afford 283 mg of nanoparticles of composition
Pdn/(14) (46% yield with respect to Pd); Anal. found: C,
52.12; H, 8.53; N, 0.54; S, 1.11; Pd 1.78 ꢃ 0.04.
Nanoparticles Pdn/(6 + 13) (Entry 3). The same procedure
as for Entry 2, but Pd(dba)2 was added in one portion and the
reaction time was three days. The yield was 32%. Anal. found:
Pd, 3.47.
Nanoparticles Pdn/(6) (Entry 4). A mixture of macrocycle 6
(103 mg, 0.049 mmol), Pd(dba)2 (65 mg, 0.12 mmol) and THF
(30 mL) was heated 22 h at 140 1C in a closed reactor. The
solid was filtrated through Celite and the filtrate was evapo-
rated. The solid was washed several times with cold THF in
order to eliminate dba. Grey nanoparticles of composition
Pdn/ (6) were obtained (49 mg). IR and 1H-NMR spectra were
identical with that of 6.
Nanoparticles Pdn/(7) (Entry 9). A mixture of macrocycle 7
containing PEG-OMe (300 mg, o0.045 mmol), Pd(dba)2 (57
mg, 0.104 mmol) and THF (1.0 mL) was heated at 140 1C in a
closed reactor for one day. Then the mixture was filtered
through Celite using THF as eluent. The solvent was evapo-
rated to afford a black oil that was washed with hexane and
dissolved in water. The aqueous solution was partitioned
again with hexane to remove the last traces of dba. The water
was finally evaporated at reduced pressure and the resulting
black oil was washed with hexane to afford 253 mg of solid
nanoparticles of composition Pd/7 (11% yield with respect to
Nanoparticles Pdn/(12 + 15) (Entry 5). A mixture of 12
(0.11 g, 0.056 mmol) and Pd(dba)2 (0.051 g, 0.089 mmol) in
THF (6 mL) was refluxed for two days. NMR monitoring
indicated that some free 12 remained. Then more Pd(dba)2
(0.11 g, 0.19 mmol) was added and the mixture was refluxed
three more days. The formed black solid was filtered and
washed with THF and chloroform till total elimination of dba
to afford 0.099 g of brown nanoparticles of composition Pdn/
(12 (7%) + 15 (93%)) (76%); decomp. at 179–185 1C;
1
Pd). IR and H-NMR spectra were identical with that of 7;
Anal. found: C, 52.80; H, 8.71; N, 0.68; S, 1.23; Pd 0.48 ꢃ 0.05.
Nanoparticles Pdn/(7) (Entry 10). A mixture of macrocycle 7
containing PEG-OMe (1.3 g,o0.20 mmol), Pd(dba)2 (147 mg,
0.25 mmol) and THF (1.0 mL) was refluxed for one day. Then
more Pd(dba)2 (57 mg, 0.10 mmol) was added and refluxed for
6 days. More Pd(dba)2 (57 mg, 0.10 mmol) was added and the
mixture was left to reflux for a long weekend of 4 days. In the
meantime, accidentally, the solvent was evaporated. The crude
mixture was filtered through Celite using THF as eluent. The
solvent was evaporated to afford a black oil, which was
washed with hexane to eliminate dba. A black solid was
obtained of composition Pdn/(7) (906 mg, 61% yield with
IR(ATR): 1201, 1152 cmꢄ1 1H-NMR (CDCl3, 250 MHz):
;
d = 1.6–1.9 (m, 4H), 2.81 (t, J = 11.5 Hz, 2H), 3.10 (d, J =
8.0 Hz, 1H), 3.11 (d, J = 6.2 Hz, 1H), 3.70–3.86 (m, 3H), 4.06
(t, J = 10.6 Hz, 2H), 4.69 (d, J = 14.1 Hz, 3H), 4.86 (d,
J = 8.4 Hz, 2H), 7.70 (t, J = 8.0 Hz, 3H), 7.81 (d, J = 8.0 Hz,
3H), 8.00–8.10 (m, 6H), traces of free 12 (ca. 7%) were
detected by a broad singlet at d = 5.64. Anal. found: C,
28.60; H, 6.14; N, 1.78; S, 1.19; Pd 23.0.
Nanoparticles Pdn/(12 + 15) (Entry 6). A mixture of 12 (50
mg, 0.027 mmol) and Pd(PPh3)4 (34 mg, 0.029 mmol) in THF
(5 mL) was stirred at room temperature overnight. Monitoring
by 1H-NMR revealed that not all the macrocycle was com-
plexed. Then, more Pd(PPh3)4 (20 mg, 0.017 mmol) was added
and the new mixture was stirred for 24 h more. Then, more
Pd(PPh3)4 (30 mg, 0.026 mmol) was added and the mixture
was stirred for 3 more days. The black mixture was filtered
through a filter of polytetrafluoroethylene (PTFE) of 0.2 mm of
pore. The filtrate was evaporated and the solid washed with
1
respect to Pd). IR and H-NMR spectra were identical with
that of 7; Anal. found: C, 52.84; H, 8.58; N, 0.57; S, 1.15; Pd
3.25 ꢃ 0.06.
Nanoparticles Pdn/(14) (Entry 11). A mixture of 7 (200 mg,
o0.030 mmol) and Pd(PPh3)4 (35 mg, 0.030 mmol) in THF (5
mL) was stirred at room temperature for 24 h. Monitoring by
1H-NMR revealed that not all the macrocycle was complexed.
Then, more Pd(PPh3)4 (35 mg, 0.030 mmol) was added and the
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1584–1594 | 1591

View Article Online
new mixture was stirred for 24 h more. The solvent was
evaporated, the residual solid was dissolved in THF (1.5
mL) and precipitated with pentane (10 mL). The supernatant
liquid was eliminated via cannula and the process was per-
formed up to five times in order to eliminate all the triphenyl-
phosphine oxide. The solution was filtrated through Celite
using THF as eluent and the solvent was evaporated to afford
brown nanoparticles of composition Pdn/(14) (157 mg, 36%
yield with respecto to Pd). MALDI-TOF: m/z = 1332–2699
[PEG-OMe + Na]+, 3919–4974 [disubstituted macrocycle-Pd
+ Na]+ and 5891–7384 [14 + Na]+. Anal. found: C, 54.21;
H, 8.56; N, 0.45; S, 0.83; Pd 1.48 ꢃ 0.06.
mp 187–196 1C; IR (ATR): 1337, 1242, 1203, 1166, 1142 cmꢄ1
;
1H-NMR (CDCl3, 250 MHz): d = 1.65–1.85 (m, 4H), 2.35–2.6
(m, 6H), 2.85 (br t, J = 15.2 Hz, 2H), 3.05–3.11 (m, 2H), 3.15–3.25
(m, 6H), 3.76–3.80 (m, 2H), 4.02 (broad t, J = 14 Hz, 4H), 4.65
(br d, J = 6.8 Hz, 4H), 4.79 (d, J = 13.3, Hz 2H), 7.34 (m, 6H),
7.69 (d, J = 8.4 Hz, 4H), 7.76 d, J = 8.4 Hz, 2H). MALDI-TOF:
m/z 2168.184 (M + 1)+, 2190.163 (M + Na)+, 2206.123 (M +
K)+. Anal. calcd for C60H42F51N3O6PdS6: C, 33.23; H, 1.95; N,
1.94; S, 8.87; found: C, 33.69; H, 1.66; N, 1.92; S, 9.24.
E,E,E-1,6,11-Tris[4-(methoxy(polyoxyethylene)phenylsulfo-
nyl]-1,6,11-triazacyclopentadeca-3,8,13-triene palladium(0), 14.
A solution of PEG-2000 monomethyl ether (393 mg, 0.197
mmol) in 4 mL of THF was added under inert atmosphere
over a stirred suspension of sodium hydride (13 mg of 60%
suspension in mineral oil, 0.32 mmol) in THF (0.5 mL). The
mixture was stirred for 10 min and then a solution of palla-
dium(0) complex 16 (50 mg, 0.063 mmol) in THF (4 mL) was
added dropwise. The mixture was stirred overnight. An excess
of NaH (13 mg, 0.317 mmol) was added in order to complete
the reaction during the next day. After filtration, the solvent
was evaporated to afford a white solid that was washed with
hexane to afford 390 mg (89%) of 14. IR: 2880, 1593, 1465,
Nanoparticles Pdn/(14) (Entry 12). A mixture of 7 (300 mg,
o0.045 mmol), Pd(dba)2 (56 mg, 0.095 mmol) in THF (2.3
mL) was stirred overnight at room temperature. Then 5 mL of
THF were added and the mixture was filtered through Celite
using THF as eluent. The solvent was evaporated to afford a
black solid, which was dissolved in water (ca. 30 mL) and
filtered to eliminate Pd(dba)2 and dba. The filtrate was eva-
porated, the residue was dissolved in CH2Cl2 and the resulting
solution dried with anhydrous Na2SO4. Then the mixture was
filtered and the solvent was evaporated to afford brown
nanoparticles of composition Pdn/(14) (200 mg, 23% yield
with respect to Pd). Anal. found: C, 52.62; H, 8.98; N, 0.48; S,
0.83; Pd 1.16 ꢃ 0.03.
1
1341, 1102 cmꢄ1. H-NMR (CDCl3, 250 MHz): d = 3.38 (s,
17H), 3.64 (m, 670H), 4.14 (m, 6H), 4.62 (d, J = 13 Hz, 3H),
4.76 (d, J = 12.5 Hz, 1H), 6.96 (d, J = 8.2 Hz, 6H), 7.67 (d,
J = 8.5 Hz, 4H), 7.74 (d, J = 7.2 Hz, 2H). MALDI-TOF MS:
m/z 6175 (40, 40, 41 CH2CH2O units) to 7454 (50 ꢂ 3
CH2CH2O units) separated by 44 Da [M + Na+], the more
intense peak occurs at 6706 (44, 44, 45 CH2CH2O units).
Nanoparticles Pdn/(10) (Entry 13). A mixture of 10 (300 mg,
o0.045 mmol) and Pd(dba)2 (57 mg, 0.104 mmol) in THF (2.3
mL) was refluxed overnight. Then 5 mL of THF were added
and the mixture was filtered through Celite using THF as
eluent (ca. 80 mL). The solvent was evaporated to afford a
brown solid that was dissolved in water and the solution
filtered. The water was evaporated and the solid obtained
was dissolved in CH2Cl2. The new solution was dried with
anhydrous Na2SO4, filtered and the solvent was evaporated to
afford clear brown solid nanoparticles of composition Pdn/(10)
(281 mg, 8% yield with respect to Pd). IR and 1H-NMR
spectra were identical with that of 10. Anal. found: Pd
0.33 ꢃ 0.007.
1,6-Bis-[2,4,6-(triisopropyl)phenylsulfonyl]-11-[4-(perfluorooctyl)
phenylsulfonyl]-1,6,11-triazacyclopentadeca-3,8,13-triene palladium
(0), 17 (mixture of two isomers). A mixture of macrocycle 11 (0.11
g, 0.084 mmol) and freshly prepared tetrakis(triphenylphosphi-
ne)palladium(0) (0.116 g, 0.100 mmol) in THF (30 mL) in a
Schlenk tube was refluxed for 20 h. The crude was filtered
through Celite and the solvent was evaporated to afford an oil
that upon digestion with hexane gave 0.052 g (40%) of complexes
17, mp 145–148 1C; IR(ATR) 1246, 1216, 1151 cmꢄ1; 1H-NMR
(CDCl3, 250 MHz, see Supplementary Informationw): d = 1.24
(br s, 36H), multiplets for 24H at 1.6–2.2, 2.80–2.96, 2.88, 3.22,
3.73–3.94, 4.0–4.25, 4.45–4.85, 7.17 (s, 4H), 7.74 (d, J = 8.5 Hz,
2H), 7.85–8.05 (two doublets of relative intensity ca. 1 : 3, J = 8.3
Hz, 2H); 13C-NMR (CDCl3, 62.5 MHz): d = 23.5, 24.8, 29.2,
29.6, 29.7, 34.2, nine peaks assigned to the methylenic ring carbon
atoms at 43.8, 45.4, 46.4, 47.7, 47.9, 48.4, 49.1, 49.8, 51.0, nine
peaks assigned to the olefinic carbons of both isomers at 78.0,
78.3, 79.2, 79.4, 79.9, 80.0, 82.3, 83.9, 84.5, aromatic carbons at
123.7, 126.9, 127.1, 127.7–127.8 (br signal), 129.8 (t, J = 22 Hz),
130.5, 130.8, 130.9, 130.9, 132.8 (t, J = 24 Hz), 142.2,
142.8, 151.0, 151.2, 152.9, 153.0, 153.0; MALDI-TOF MS:
m/z = 1403.8 [M]+, 1425.8 [M + Na]+, 1441.8 [M + K]+.
An alternative preparation was accomplished with excess
Pd(dba)2 in refluxing THF under analogous conditions of
Entry 1 of Table 1. In contrast with the behaviour of fluorous
macrocycle 6, this compound 11 did not give nanoparticles,
but only the triolefinic palladium complex 17.
Preparation of palladium(0) complexes
(E,E,E)-1,6,11-Tris[4-(1H,1H,2H,2H-perfluorodecylthio)phenyl-
sulfonyl]-1,6,11-triazacyclopentadeca-3,8,13-triene palladium(0), 13.
A solution of Pd(dba)2 (138 mg, 0.234 mmol) in THF (15 mL) was
added via syringe and with stirring into a solution of (E,E,E)-
1,6,11-tris[4-(1H,1H,2H,2H-perfluorodecylthio) phenylsulfonyl]-
1,6,11-triazacyclopentadeca-3,8,13-triene, 6 (318 mg, 0.15 mmol)
in THF (115 mL) heated at 60 1C. The mixture was heated at
60 1C with stirring. After three days more Pd(dba)2 (77 mg,
0.127 mmol) in THF (10 mL) was added. The mixture was heated
at 60 1C for two more days. The black solution was then filtered
while hot through a filter of polytetrafluoroethylene (PTFE) of 0.2
mm pore. The filtrate was concentrated to 20 mL and left for two
days at ꢄ20 1C in the freezer. The black solid was isolated by
filtering, washed with cold THF (2 ꢂ 5 mL) and flash-chromato-
graphed through silica gel with hot THF (drying pistol over the
column) as eluent to afford a white solid that was digested with
pentane (2 ꢂ 25 mL) to afford 146 mg of 13 (48% yield),
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1592 | New J. Chem., 2006, 30, 1584–1594

View Article Online
Thermal treatment of palladium(0) complexes
iodobenzene (0.06 mL, 0.5 mmol), butyl acrylate (0.10 mL, 0.7
mmol), tributylamine (0.14 mL, 0.6 mmol), Pdn/(7) of Entry
10 (15 mg, 0.0046 mmol of Pd, 3.25 molar %) and THF (1 mL)
was heated at 90 1C (oil bath) in a closed reactor for 24 h.
After cooling diethyl ether (1 mL) was added, the mixture was
centrifuged and the supernatant organic solution was sepa-
rated. A second centrifugation was performed after addition
of more diethyl ether (2–3 mL). The combined ethereal solu-
tions were washed with diluted aqueous HCl and with water,
dried (sodium sulfate) and evaporated. The obtained oil was
purified by column chromatography through silica gel using
hexanes–ethyl acetate (95 : 5) as eluent to afford n-butyl
cinnamate (74 mg, 79%) as a colourless oil with the same
spectroscopic characteristics as above.
Formation of nanoparticles Pdn/(6) by thermal treatment of
the macrocyclic triolefinic complex 13. Palladium complex 13
(49 mg, 0.023 mmol) in 17 mL of THF was heated overnight at
140 1C. The hot mixture was filtered through a filter of
polytetrafluoroethylene (PTFE) of 0.2 mm of pore and the
filtrate was evaporated. A grey solid was obtained (45 mg),
corresponding to nanoparticles of composition Pdn/(6). IR
1
and H-NMR spectra were identical with that of 6.
Behaviour of complex 14 when heated in the absence and in
the presence of Pd(dba)2. (a) A solution of 14 (106 mg, o0.016
mmol) in THF (2 mL) was placed in a sealed tube and heated
1
at 100 1C overnight, no change being observed by H NMR.
The solution was heated at 140 1C overnight and then at
175 1C for one more night. The complex 14 was recovered
unchanged according to the H NMR spectrum.
Acknowledgements
1
Financial support from the Ministry of Science and Technol-
ogy (then Ministry of Education and Science) of Spain (Pro-
jects 2002BQU-04002, CTQ-2005-04968/BQU and CTQ-2005-
01254/BQU), Generalitat de Catalunya (Projects 2001SGR
(b) A mixture of 14 (100 mg, o0.015 mmol) and Pd(dba)2
(7 mg, 0.019 mmol) in THF (2 mL) was heated overnight at
140 1C (sealed tube). The crude mixture was diluted in THF
and filtered through Celite. The solvent was evaporated and
the resulting solid was taken in water and filtered. The
evaporation of water gave a yellow solid corresponding to
complex 14 contaminated with the free macrocycle 7 (70 : 30
´
00181 and 2005SGR00305) and Universitat Autonoma de Bar-
celona (Project PNL2005-10) is gratefully acknowledged. One
of us (R.M.S.) was incorporated to the research group through
1
a ‘‘Ramon y Cajal’’ contract (MCyT-FEDER/FSE). The Min-
´
by H NMR). Nanoparticles were not observed by TEM.
istry of Education and Science of Spain is gratefully acknowl-
edged for predoctoral scholarships to A. S., R. S. and E. B.
(c) A mixture of 14 (150 mg, o0.022 mmol) and Pd(dba)2
(28 mg, 0.049 mmol) in THF (0.5 mL) was heated at 140 1C in
a sealed tube and the following observations were made by 1H
NMR: after 20 min of reaction signals corresponding to
coordinated olefins at 4.60 (dd) were present, together with a
signal at 5.55 (apparent d) corresponding to free olefins.
Similar results were obtained after 3 h and after 5 h. After
20 h of reaction the signal at 4.6 ppm corresponding to the
coordinated olefins was present, together with a broad singlet
at 5.59 (corresponding to free olefins of macrocycle 7), which
was of higher intensity than another broad singlet at 5.52 ppm.
References
1 For reviews see: (a) M. Moreno-Manas, R. Pleixats, A. Roglans,
R. M. Sebastia
available from http://www.arkat-usa.org.; (b) M. Moreno-Manas,
R. Pleixats, R. M. Sebastia
´
n and A. Vallribera, Arkivoc, 2004, iv, 109–129,
´
n, A. Vallribera and A. Roglans,
J. Organomet. Chem., 2004, 689, 3669–3684.
2 For a review see: G. Wilke, Angew. Chem., Int. Ed. Engl., 1988, 27,
185–206.
3 (a) B. Blanco, A. Mehdi, M. Moreno-Manas, R. Pleixats and C. Reye
Tetrahedron Lett., 2004, 45, 8789–8791; (b) B. Blanco, M. Brissart, M.
Moreno-Manas, R. Pleixats, A. Mehdi, C. Reye, S. Bouquillon, F.
Henin and J. Muzart, Appl. Catal. A, 2006, 297, 117–124.
4 M. Moreno-Manas, C. Reichardt, R. M. Sebastian, J. Barbera
L. Serrano and T. Sierra, J. Mater. Chem., 2005, 15, 2210–2219.
5 A. Torrent, I. Gonzalez, A. Pla-Quintana, A. Roglans, M. Mor-
´
,
Catalytic activity of palladium(0) nanoparticles
´
´
Preparation of n-butyl cinnamate under catalysis by Pdn/(13)
of Entry 2 and recovery of the catalytic material (Run 3). A
mixture of iodobenzene (0.7 mL, 6.3 mmol), butyl acrylate (1.3
mL, 8.9 mmol), tributylamine (1.8 mL, 7.6 mmol), Pdn/(13)
(Entry 2) (21 mg, 0.013 mmol of Pd, 0.2 molar%) and THF
(10 mL) was heated at 90 1C in a closed reactor for 24 h. The
solvent was removed with a nitrogen stream and the resi-
due was extracted with cold acetonitrile filtering through a
cannula. The residual black solid was used in the fourth run.
The acetonitrile solution was evaporated and the residue was
purified by column chromatography on silica gel with hex-
anes–ethyl acetate (95 : 5) as eluent, to afford n-butyl cinna-
mate (1.2 g, 94%) as an oil (99% purity by gas
´
´
, J.
´
eno-Manas, T. Parella and J. Benet-Buchholz, J. Org. Chem., 2005,
70, 2033–2041.
6 (a) M. Moreno-Manas, R. Pleixats and S. Villarroya, Organome-
tallics, 2001, 20, 4524–4528; (b) M. Moreno-Manas, R. Pleixats
and S. Villarroya, Chem. Commun., 2002, 60–61; (c) M. Tristany, J.
Courmarcel, P. Dieudonne, M. Moreno-Manas, R. Pleixats, A.
´
Rimola, M. Sodupe and S. Villarroya, Chem. Mater., 2006, 18,
716–722; (d) M. Moreno-Manas, R. Pleixats and M. Tristany, J.
Fluorine Chem., 2005, 126, 1435–1439.
7 (a) E. B. Zuckerman, K. J. Klabunde, B. J. Olivier and C. M.
Sorensen, Chem. Mater., 1989, 1, 12–14; (b) K. J. Klabunde, G.
Youngers, E. J. Zukerman, B. J. Tan, S. Antrim and P. M.
Sherwood, Eur. J. Solid State Inorg. Chem., 1992, 29, 227–260.
8 For a review see: M. Moreno-Manas and R. Pleixats, Fluorous
Nanoparticles, in Handbook of Fluorous Chemistry, ed. J. A.
chromatography): IR 2966, 2867, 1710, 1637, 1172 cmꢄ1
;
1H-NMR (CDCl3, 250 MHz): d = 0.97 (t, J = 7.4 Hz, 3H),
1.44 (m, 2H), 1.70 (m, 2H), 4.21 (t, J = 6.6 Hz, 2H), 6.44 (d,
J = 16.0 Hz, 1H), 7.38 (m, 3H), 7.53 (m, 2H), 7.69 (d,
J = 16.0 Hz, 1H).
´
Gladysz, D. P. Curran and I. T. Horvath, Wiley-VCH, Weinheim,
2004, ch 12.2, pp. 491–507.
9 For reviews and monographs see: (a) L. N. Lewis, Chem. Rev.,
1993, 93, 2693–2730; (b) J. S. Bradley, The Chemistry of Transition
Metal Colloids, in Cluster and Colloids, From Theory to Applica-
tions, ed. G. Schmid, VCH, Weinheim, 1994, pp. 459–544; (c)
Metal Clusters in Chemistry, ed. P. Braunstein, L. Oro and P. R.
Raithby, Wiley-VCH, Weinheim, 1998; (d) Nanoparticles and
Preparation of n-butyl cinnamate under catalysis by Pdn/(7)
of Entry 10 and recovery of the catalytic material. A mixture of
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1584–1594 | 1593

View Article Online
Nanostructured Films. Preparation, Characterization and Applica-
tions, ed. J. H. Fendler, Wiley-VCH, Weinheim, 1998; (e) K. J.
Klabunde and C. Mohs, Nanoparticles and Nanostructured Ma-
terials, in Chemistry of Advanced Materials. An Overview, ed. L. V.
Interrante and M. J. Hampden-Smith, Wiley-VCH, New York,
1998, ch 7, pp. 271–327; (f) J. D. Aiken III and R. G. Finke, J. Mol.
Catal. A, 1999, 145, 1–44; (g) A. C. Templeton, W. P. Wuelfing and
R. W. Murray, Acc. Chem. Res., 2000, 33, 27–36; (h) C. N. R. Rao,
G. U. Kulkarni, P. J. Thomas and P. P. Edwards, Chem. Soc. Rev.,
2000, 29, 27–35; (i) D. Horn and J. Rieger, Angew. Chem., Int. Ed.,
2001, 40, 4330–4361; (j) M. T. Reetz, M. Winter, R. Breinbauer, T.
Thurn-Albrecht and W. Vogel, Chem. Eur. J., 2001, 7, 1084–1094;
13 M. Tromp, J. R. A. Sietsma, J. A. van Bokhoven, G. P. F. van
Strijdonck, R. J. van Haaren, A. M. J. van der Eerden, P. W. N. M.
van Leeuwen and D. C. Koningsberger, Chem. Commun., 2003,
128–129.
14 For representative examples see: (a) C. Luo, Y. Zhang and Y.
Wang, J. Mol. Catal. A, 2005, 229, 7–12; (b) Z. Hou, N. Theyssen,
A. Brinkmann and W. Leitner, Angew. Chem., Int. Ed., 2005, 44,
1346–1349; (c) S.-W. Kim, S. Kim, J. B. Tracy, A. Jasanoff and M.
G. Bawendi, J. Am. Chem. Soc., 2005, 127, 4556–4557; (d) X.-F.
Qiu and J.-J. Zhu, Huaxue Xuebao, 2003, 19, 766–700, Chem. Abst.
139: 200539; (e) K. Naka, M. Yaguchi and Y. Chujo, Chem.
Mater., 1999, 11, 849–851.
(k) F. Caruso, Adv. Mater., 2001, 13, 11–22; (l) H. Bonnemann and
15 M. Moreno-Manas and J. Spengler, Tetrahedron, 2002, 58,
7769–7774.
¨
R. M. Richards, Eur. J. Inorg. Chem., 2001, 2455–2480; (m) C. N.
R. Rao, G. U. Kulkarni, P. J. Thomas and P. P. Edwards, Chem.
Eur. J., 2002, 8, 28–35; (n) Metal Nanoparticles. Synthesis, Char-
acterization and Applications, ed. D. L. Feldheim, and C. A. Foss,
Jr, Marcel Dekker Inc., New York, 2002; (o) A. Roucoux, J. Schulz
and H. Patin, Chem. Rev., 2002, 102, 3757–3778; (p) M. Moreno-
Manas and R. Pleixats, Acc. Chem. Res., 2003, 36, 638–643; (q) For
a comprehensive review on gold nanoparticles see: M.-C. Daniel
and D. Astruc, Chem. Rev., 2004, 104, 293–346.
16 E. Badetti, M. Moreno-Manas, R. Pleixats, R. M. Sebastia
Serra, R. Soler and A. Vallribera, Synlett, 2005, 449–452.
17 (a) S. Cerezo, J. Cortes, E. Lago, E. Molins, M. Moreno-Manas, T.
Parella, R. Pleixats, J. Torrejon and A. Vallribera, Eur. J. Inorg.
Chem., 2001, 1999–2006; (b) A. Pla-Quintana, A. Roglans, J.
Vicente de Julian-Ortiz, M. Moreno-Manas, T. Parella, J.
´
n, A.
´
´
Benet-Buchholz and X. Solans, Chem. Eur. J., 2005, 11,
2689–2697.
18 M. Moreno-Manas, R. Pleixats, J. Spengler, C. Chevrin, B.
10 For a didactic explanation on the structure and properties of
nanoparticles see: J. Cox, Chem. Br., September 2003, 21–25.
11 R. G. Finke, Transition-Metal Nanoclusters, in Metal Nanoparti-
cles. Synthesis, Characterization and Applications, ed. D. L. Feld-
heim and C. A. Foss, Jr, Marcel Dekker, Inc., 2002, ch 2, pp.
17–54.
Estrine, S. Bouquillon, F. Henin, J. Muzart, A. Pla-Quintana
´
and A. Roglans, Eur. J. Org. Chem., 2003, 274–283.
19 X. Li, X. Jun Loh, K. Wang, Ch. He and J. Li, Biomacromolecules,
2005, 6, 2740–2747.
20 (a) G. Chisholm, B. Hay, K. D. M. Harris, S. J. Kitchin and K. M.
Morgan, Dyes Pigm., 1999, 42, 159–172; (b) F. C. Krebs and H.
Spanggaard, J. Org. Chem., 2002, 67, 7185–7192.
12 C. Besson, E. E. Finney and R. G. Finke, J. Am. Chem. Soc., 2005,
127, 8179–8184, and references cited therein.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1594 | New J. Chem., 2006, 30, 1584–1594