Syntheses and phase-transfer properties of dendritic nanocarriers that contain perfluorinated shell structures

Article abstract of DOI:10.1002/chem.200305305

Perfect dendrimers that contain perfluorinated shells have recently attracted attention because they have been shown to encapsulate polar molecules in supercritical CO₂ and catalytically active metal nanoparticles in perfluorinated solvents. Moreover, they can then be easily separated after reaction from the biphasic organic/fluorous system. In this paper several dendritic architectures that contain perfluorinated shells were derived by covalent modification of glycerol dendrimers ([G0.5]-[G3.5]), hyperbranched polyglycerol, and polyethyleneimine. These core-shell architectures show interesting physicochemical properties. For example, they are soluble in fluorinated solvents, they are able to transport different guest molecules, and they display thermomorphic behavior. The transport capacity of these molecular nanocarriers increases significantly when amino groups are present in the core. Certain functionalized polyethyleneimines that contain perfluorinated shells show high transport capacities (up to 3 dye molecules per nanocarrier) in perfluorinated solvents. Moreover, these perfluoro-functionalized dendritic polyethyleneimines can act as templates that stabilize nanoparticles; for example, encapsulation and subsequent chemical reduction of Ag^I ions. Silver nanoparticles with a narrow size distribution (3.9 ± 1 nm) have been prepared and characterized by transmission electron microscopy. Furthermore, it has been demonstrated that the encapsulated guest molecules remain accessible to small molecules after transport into the fluorous phase. Therefore, dendritic nanocarriers that contain perfluorinated shells are currently being investigated as polar environments in nonpolar reaction media such as fluorous phases and supercritical CO₂, in particular, for application in homogenous catalysis.

Full text of DOI:10.1002/chem.200305305

FULL PAPER
Syntheses and Phase-Transfer Properties of Dendritic Nanocarriers That
Contain Perfluorinated Shell Structures
Abel Garcia-Bernabÿ,[a] Michael Kr‰mer,[b] Bõla Ol‡h,[b] and Rainer Haag*[a]
Abstract: Perfect dendrimers that con-
tain perfluorinated shells have recently
attracted attention because they have
been shown to encapsulate polar mole-
cules in supercritical CO2 and catalyti-
cally active metal nanoparticles in per-
fluorinated solvents. Moreover, they
can then be easily separated after reac-
tion from the biphasic organic/fluorous
system. In this paper several dendritic
architectures that contain perfluorinat-
ed shells were derived by covalent
modification of glycerol dendrimers
([G0.5] [G3.5]), hyperbranched poly-
glycerol, and polyethyleneimine. These
core-shell architectures show interest-
ing physicochemical properties. For ex-
ample, they are soluble in fluorinated
solvents, they are able to transport dif-
ferent guest molecules, and they dis-
play thermomorphic behavior. The
transport capacity of these molecular
nanocarriers increases significantly
when amino groups are present in the
core. Certain functionalized polyethyl-
eneimines that contain perfluorinated
shells show high transport capacities
(up to 3 dye molecules per nanocarrier)
in perfluorinated solvents. Moreover,
these perfluoro-functionalized dendritic
polyethyleneimines can act as tem-
plates that stabilize nanoparticles; for
example, encapsulation and subsequent
chemical reduction of AgI ions. Silver
nanoparticles with a narrow size distri-
bution (3.9ꢀ1 nm) have been prepared
and characterized by transmission elec-
tron microscopy. Furthermore, it has
been demonstrated that the encapsulat-
ed guest molecules remain accessible
to small molecules after transport into
the fluorous phase. Therefore, dendritic
nanocarriers that contain perfluorinat-
ed shells are currently being investigat-
ed as polar environments in nonpolar
reaction media such as fluorous phases
and supercritical CO2, in particular, for
application in homogenous catalysis.
Keywords: dendrimers
¥ fluorous
biphasic catalysis ¥ host guest sys-
tems ¥ hyperbranched polymers ¥
supramolecular chemistry
Introduction
media such as fluorous phases and supercritical carbon diox-
ide (scCO2). The advantage of these alternative reaction
media is the simple workup, which allows products to be
easily recovered and catalysts to be recycled.[3] As a result
of their chemical inertness, perfluorinated compounds are
also considered to be environmentally friendly.
In this paper we describe simple and efficient approaches to
dendritic core-shell architectures that contain covalently
linked perfluorinated shells. These molecular nanocarriers
provide polar environments in nonpolar reaction media, and
some candidates show interesting phase-transfer behavior
with respect to polar molecules in the fluorous phase.
Dendritic core-shell architectures have attracted increased
interest because of their ability to act as unimolecular nano-
carriers for catalysts, drugs, and other guest molecules.[1,2]
More recently, such nanostructures have also been used to
generate polar environments within alternative reaction
Frÿchet and DeSimone reported the first example of a
molecular nanocarrier based on a perfect polyamine den-
drimer with a covalently bonded perfluorinated shell.[4]
More recently, Crooks et al. prepared dendrimer-encapsulat-
ed metal nanoparticles in the fluorous phase by electrostatic
attachment of perfluorocarboxylic groups to a poly(amido-
amine) (PAMAM) dendrimer.[5] In both cases, the defined
core-shell architecture and a pronounced polarity gradient
of the dendrimer seem to play an important role in their sol-
ubility and transport behavior. Many examples of dendrim-
er-encapsulated guest molecules have been reported,[2] some
of which have been metal nanoparticles (e.g. Cu, Pt, Ag,
Au, and Pd), and in some cases, their applications towards
homogenous catalysis has been demonstrated.[6] These nano-
particles can be prepared by chemical reduction of the cor-
responding encapsulated metallic cations. For example, a
poly(amidoamine) (PAMAM) dendrimer with an ionically
[a]Dr. A. Garcia-Bernabÿ, Prof. Dr. R. Haag
Organische Polymerchemie, Universit‰t Dortmund
Otto-Hahn-Strasse 6, 44227 Dortmund (Germany)
Fax : (+49)0231-755-6148
E-mail: rainer.haag@uni-dortmund.de
[b]Dipl.-Chem. M. Kr‰mer, Dr. B. Ol‡h
Freiburger Materialforschungszentrum (FMF)
und Institut f¸r Makromolekulare Chemie
Hermann-Staudinger-Haus, Stefan-Meier-Strasse 31
79104 Freiburg (Germany)
2822
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200305305
Chem. Eur. J. 2004, 10, 2822 2830

2822 2830
bonded fluorous shell was used to stabilize Pd nanoparticles
Results and Discussion
ꢁ
in the fluorous phase, and this was then applied in C C
bond forming reactions.[5] However, in some cases (i.e. in
polar solvents) these electrostatic complexes can be unsta-
ble. Therefore, our approach was based on preparing cova-
lently linked perfluorinated shells that give rise to an easily
accessible and highly branched polymer scaffold.
Synthesis of glycerol dendrimers: To generate the molecular
nanocarriers 3a d, three generations of the four-armed glyc-
erol dendrimers 2a c, which give rise to a more dense cen-
tral core architecture, were prepared. For this purpose, a re-
action sequence similar to the one we recently developed
for three-armed glycerol dendrimers and pseudo-dendrimers
was utilized (Scheme 2).[9] The synthesis is based on a
simple iterative two-step protocol that involves allylation of
an alcohol and subsequent catalytic dihydroxylation of the
allylic double bond. Under phase-transfer conditions, the al-
lylation reproducibly affords high yields of the product even
when polyols are employed.[15] Catalytic dihydroxylation of
the double bond with osmium tetroxide and N-methyl mor-
pholine-N-oxide (NMO) as co-oxidant[16] completes the se-
quence and leads to the formation of new glycerol units on
every available alcohol functionality (Scheme 2). As previ-
ously reported for a three-armed trimethylolpropane core,
both transformations can be
The dendrimer topology allows a clear differentiation be-
tween the interior branching scaffold (the core) and the end
groups in the periphery (the shell). For many applications
the difficult accessibility of dendrimers through costly multi-
step syntheses represents
a
major problem.[7] Hyper-
branched polymers that can be obtained in one reaction
step from the polymerization of ABm or latent ABm mono-
mers are currently being considered as possible alterna-
tives.[8] In contrast to the perfectly branched glycerol den-
drimers (degree of branching (DB)=100%),[9] hyper-
branched polyglycerol (PG)[10] and hyperbranched polyeth-
yleneimine (PEI)[8a,11] (Scheme 1) are randomly branched,
carried out in aqueous phase
and are suitable for a wide
range of core molecules.[9] The
use of a four-armed core mole-
cule gave rise to high conver-
sions (>90%) but lower isolat-
ed yields (60 80% per se-
quence) of the glycerol den-
drimers 2a c. This might be the
result of steric crowding and
cross-linking of the allylated in-
termediates 1a d.
Dendritic polyglycerols that
contain perfluorinated shells:
Four glycerol dendrimers with
perfluorinated shells 3a d were
obtained by free-radical addi-
tion of tridecafluorooctane-1-
thiol to the perallylated inter-
mediates 1a d; this afforded
Scheme 1. Hyperbranched polyglycerol (PG) and polyethyleneimine (PEI). The depicted structures contain
25 40% linear functional groups and 30 40% terminal groups, and represent only small idealized fragments
of the large polymer cores.
but still have well-defined dendritic structures with a DB of
60 to 75%. They are prepared in a one-step process, are
readily available on a large scale, and have relatively low
polydispersities (ꢂ2).[8a,11,12] Upon selective functionaliza-
tion of these hyperbranched polymers special environments
are created in the core of these dendritic architectures.[13]
These can then act as molecular nanocarriers[14] to encapsu-
late and transport polar guests (e.g. drugs, dyes, metal salts).
Here we describe an efficient approach to access novel
dendritic architectures that contain perfluorinated shells
through covalent modification of perfect glycerol dendrim-
ers, hyperbranched polyglycerol (PG), and hyperbranched
polyethyleneimine (PEI). These core-shell architectures that
contain perfluorinated shells are soluble in several fluorous
solvents and can transport ions, polar guest molecules, and
metal nanoparticles. In addition, some of these molecular
nanocarriers have been applied in phase-transfer reactions,
for example, between aqueous and fluorous phases.
the thioether linkages with high conversions (>95%,
Scheme 3). Also, various other hyperbranched polyglycerols
(DB ꢃ60%)[10] of different molecular weights were func-
tionalized with a perfluorinated shell (5a d) using the same
synthetic strategy, but the sequence began with the perally-
lated hyperbranched polyglycerols 4a d.[9] The degree of
functionalization in these cases was 83 94%.[17] All the den-
dritic polyglycerols[18] with perfluorinated shells (3b d, 5a
d), apart from [G0.5]( 3a), are soluble in perfluorinated sol-
vents such as perfluorobenzene, perfluorohexane (FC-72),
and perfluoro-2-butyltetrahydrofuran (FC-75), as well as in
partially fluorinated solvents such as trifluorotoluene
(Table 1). However, good solubility at room temperature
was only observed in perfluorobenzene, trifluorotoluene,
and FC-72.
Unexpectedly, the transport capacity of the dendritic poly-
glycerols 3 and 5 was very poor, and seemed to be inde-
pendent of the core size (see below, Table 2). Only dendritic
Chem. Eur. J. 2004, 10, 2822 2830
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2823

R. Haag et al.
FULL PAPER
75%.[11] To generate a dendritic
polyamine with perfluorinated
shells, PEI was modified by al-
kylation (6) and amidation (7),
as well as by amidation and
subsequent thiol addition ap-
proach (9). The reaction condi-
tions for these transformations
were problematic because of
the extreme polarity difference
between the polar PEI core and
the perfluorinated reactant. As
a result, the solvent had to be
carefully chosen or omitted. Al-
kylation of PEI with heptadeca-
fluoro-10-iododecane to give
the amphiphilic polyamine
6
was performed in acetonitrile
under reflux for 15 h using po-
tassium carbonate as base
(Scheme 4).
Unfortunately,
even though an excess of the al-
kylating agent was used, con-
version of the available amino
groups was only 29% (accord-
ing to elemental analysis). Ami-
dation of PEI to give the poly-
amide 7 was performed in bulk
at 1408C using heptadeca-
fluoroundecanoic acid and a ni-
trogen stream to remove the
water formed during the reac-
tion (Scheme 3). In this case,
according to elemental analysis,
conversion was about 56%. To
generate the amphiphilic mo-
lecular nanocarriers 9, PEI first
underwent amidation with 10-
undecenoic acid to give the
polyalkenyl derivative 8 with a
conversion of 67% (according
Scheme 2. Synthesis of perfect glycerol dendrimers up to generation 3.5 using a four-arm core. The first step
involves allylation of the OH groups with allyl chloride (1a d: G0.5, G1.5, G2.5, G3.5) under phase-transfer
conditions followed by dihydroxylation of the double bonds with catalytic amounts of OsO4 and NMO as co-
oxidant (2a c: G1, G2, G3): i) Bu4NBr, NaOH, allyl chloride, toluene/water, 24 h, 508C; ii) OsO4 (cat.), NMO,
acetone/water/tert-butanol, 20 h, 258C.
1
to H NMR spectroscopy). Sub-
sequently, to elongate the alkyl
chains and obtain a less pro-
nounced polarity gradient, the
terminal double bonds of the
polyglycerol 5d, which contains an amine group in the
core,[10b] was able to transport polar guest molecules, but its
transport capacity was still low in comparison to alkyl-modi-
fied hyperbranched polyglycerols.[13a,g]
intermediate 8 were reacted with a tridecafluorooctane-1-
thiol reagent using the same protocol as described for 3 and
5 (Scheme 4). In this case, as observed by 1H NMR spec-
troscopy, conversion of the double bonds into thioethers was
quantitative.
Dendritic polyethyleneimines that contain perfluorinated
shells: As it appeared that amine groups in the dendritic
structure increased the transport capacity of the resulting
dendrimer, we synthesized three dendritic architectures
based on hyperbranched polyamine core structures that con-
tain perfluorinated shells (6, 7, and 9, Scheme 4). Hyper-
branched polyethyleneimines (PEI, Scheme 1) are commer-
cially available polymers with a DB in the range of 65 to
All the dendritic architectures that contained perfluorinat-
ed shells were subsequently evaluated for their solubility in
perfluorinated solvents, phase-transfer behavior, and ther-
mal properties.
Solubility and thermomorphic behavior: The solubility of
dendritic architectures is highly dependent on the structural
features of the shell. For many applications, partial (40
2824
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2822 2830

Phase-Transfer Properties of Dendritic Nanocarriers that Contain Perfluorinated Shell Structures
2822 2830
Table 1. Degree of functionalization and solubility of dendritic architectures that contain perfluorinated shells.[a]
Sample
Gener-
ation [Mn]functional-
ization [%]
Degree of
Shell structure
Trifluoro-
toluene
Perfluoro-
benzene
FC-72
FC-75
CHCl3
3a
3b
3c
3d
5a
5b
5c
5d
6
G0.5
G1.5
G2.5
G3.5
2000
5000
15000
4500[c]
4500
4500
4500
>98
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)2(CF2)7CF3
408C
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
n.s.
X
n.s.
rflx
rflx
X
X
X
n.s.
X
X
>98
>98
>98
rflx
rflx
rflx
n.s.
rflx
rflx
rflx
n.s.
n.s.
X
rflx
rflx
rflx
rflx
rflx
rflx
n.s.
92[b]
85[b]
83[b]
94[b]
29[d]
56[d]
67[b]
X
n.s.
n.s.
rflx
n.s.
7
9
CO(CH2)2(CF2)7CF3
CO(CH2)10S(CH2)2(CF2)7CF3
X
X
438C
658C
X
[a]X =soluble at 208C, rflx=soluble at reflux, n.s.=not soluble. [b]From 1H NMR measurements. [c]PG-core contains a stearylamine starter. [d]By el-
emental analysis.
Even though both partially fluorinated amides 7 and 9 are
insoluble in the perfluorinated solvent FC-75 at room tem-
perature, they become soluble in FC-75 at higher tempera-
tures (43 and 658C, respectively). In contrast, polymer 6 is
only slightly soluble even at the boiling point of FC-75 (99
1078C) (see Table 1); this can be attributed to compound 6
only having a relatively low fluorine content (29%).
Nanocarriers 7 and 9 have been observed to encapsulate
polar bromophenol blue dye in the pure perfluorinated sol-
vent FC-75 at elevated temperatures (Figure 1), even
Figure 1. Thermomorphic behavior of dendritic nanocarrier 9 in FC-75 in
the presence of encapsulated bromphenol blue: left, 708C (clear solu-
tion); and right, 208C (suspension).
though bromophenol blue is not itself soluble in FC-75 at
these temperatures. This reversible thermomorphic[19] behav-
ior at about 608C has possible application in biphasic cataly-
sis, whereby the guest could be separated from its molecular
transporter by simple precipitation (Figure 1).[3]
Scheme 3. Synthesis of glycerol dendrimers with perfluorinated shells 3a
d. The double bonds of the allyl ether dendrimers 1a d were functional-
Phase-transfer behavior of dendritic polyamines that contain
ized with perfluorinated alkenethiols (e.g. [G2.5]dendrimer 3c).
perfluorinated shells: The phase-transfer behavior of the
dendritic nanocarriers that contain perfluorinated shells was
60%) shell modification is sufficient to adjust the polarity
and solubility with respect to specific solvents. Although the
dendritic polyethers 3 and 5 show high solubilities in most
perfluorinated solvents (Table 1), the dendritic polyamines 7
and 9, which contain a degree of functionalization higher
than 50%, are only soluble in a,a,a-trifluorotoluene and
perfluorobenzene. To dissolve the least functionalized poly-
amine 6 in trifluorotoluene, the mixture has to be heated up
to 808C (accelerated by ultrasound), whereas compound 9
dissolves readily in hexafluorobenzene or trifluorotoluene.
studied for different guest molecules. The respective amphi-
philic dendritic polymer was dissolved in trifluorotoluene,
perfluorobenzene, or FC-75, and the solution was then treat-
ed with a polar guest dissolved in aqueous solution. We ob-
served significantly higher transport capacities (see below)
for these amphiphilic PEIs in comparison to PGs that con-
tain perfluorinated shells (cf. Table 2). For example, the
polar dye bromophenol blue can be encapsulated and trans-
ported within the polyamine based architectures in hexa-
fluorobenzene or a 1:2 mixture of trifluorotoluene/FC-75
Chem. Eur. J. 2004, 10, 2822 2830
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2825

R. Haag et al.
FULL PAPER
Transport capacity of dendritic
structures that contain per-
fluorinated shells: To quantify
the transport capacities of the
perfluorinated nanocarriers (3,
5, 6, 7, and 9) an aliquot of
each dendrimer was dissolved
in trifluorotoluene, hexafluoro-
benzene, or chloroform, and
this was then saturated with an
excess of congo red. The UV/
Vis absorption spectrum of the
encapsulated congo red dye in
solution was then measured as
previously described.[13a,g] Satu-
ration of the fluorinated phase
was achieved by adding solid
congo red dye to the solution.
On the basis that the concentra-
tion of the respective dendritic
polymer in the perfluorous so-
lution is known, and that the
extinction coefficient for both
the encapsulated dye and the
dye dissolved in water can be
assumed to be similar, one can
Scheme 4. Synthesis of dendritic architectures 5, 6, 7, and 9 starting from hyperbranched PG and PEI (cf.
Scheme 1). In compounds 4d and 5d the core contains a stearylamine starter: i) Bu4NBr, NaOH, allyl chloride, calculate the number of encap-
toluene/water, 24 h, 508C; ii) HSCH2CH2(CF2)5CF3, AIBN, 22 h, 708C; iii) ICH2CH2(CF2)7CF3, K2CO3,
CH3CN, 15 h, reflux; iv) HOOCCH2CH2(CF2)7CF3, 4 h, 1408C; v) CH2CH(CH2)8COOH, 20 h, 1208C;
vi) HSCH2CH2(CF2)5CF3, AIBN, 16 h, 1008C.
sulated guest molecules for
each dendritic nanocarrier by
using
a
calibration curve
(Table 2).[15] It is noteworthy
that the transport capacities of
these perfluorinated core-shell
architectures were found to be
lower than the maximum trans-
port capacity we recently ob-
served for selectively modified
PGs[13a] and PEIs,[13b] which con-
tained longer alkyl chains. This
Table 2. Transport capacities of congo red dye at 208C in dendritic nanocarriers that contain perfluorinated
shells.
Polymer
Degree of
functional-
ization [%]benzene
Shell structure
toluene
Number of encapsulated dye molecules[a]
hexafluoro-
trifluoro-
CHCl3
3a
3b
3c
3d
5a
5b
5c
5d
6
100
100
100
100
92
85
83
94
29
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)3S(CH2)2(CF2)5CF3
(CH2)2(CF2)7CF3
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
again demonstrates that
a
<0.01
strong polarity gradient and a
relatively thick shell (1 2 nm)
are necessary for high transport
capacity. In addition, a degree
<0.01
0.01[b]
n.s.
0.05ꢀ0.01
1.0ꢀ0.3
0.13ꢀ0.03
7
9
56
67
CO(CH2)2(CF2)7CF3
CO(CH2)10S(CH2)2(CF2)7CF3
0.5ꢀ0.1
n.s.
0.25ꢀ0.06
3.0ꢀ0.8
[a]n.d. =not determined, n.s.=not soluble. [b]PG-core contains a stearylamine starter.
(Figure 2). The polar amine groups situated within the core
of nanocarrier 9 provide a polar microenvironment that
allows efficient complexation and transport (see below).
Furthermore, the color of the encapsulated pH-indicator
dye changed from blue to yellow after addition of acid to
the aqueous phase (pH 1), and returned to blue after addi-
tion of base. This clearly demonstrates that the encapsulated
guest molecule is still accessible to small molecules (i.e.
acid/base) upon encapsulation, and that complete phase
transfer occurs from the aqueous to the fluorous phase. Sim-
ilar results have been observed for the dendritic nanocarrier
7.
Figure 2. Encapsulation of bromophenol blue by compound 9 in aqueous
and trifluorotoluene/FC-75 (1:2) solutions. The first picture depicts that
the dye is soluble in water (pH 5). In the second photo, the bromphenol
blue is encapsulated within the dendritic nanocarrier 9 at pH 5. Upon ad-
dition of acid (pH 1), a color change is observed as a result of the encap-
sulated dye undergoing an acid base reaction. The reaction is reversible
upon addition of base.
2826
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2822 2830

Phase-Transfer Properties of Dendritic Nanocarriers that Contain Perfluorinated Shell Structures
2822 2830
of functionalization of about 50% seems to be optimum for
efficient transport. We have also observed a significant sol-
vent dependence in these encapsulation experiments (see
Table 2). A significant increase in transport capacity was ob-
served for the dendritic polyamide 9 when the solvent was
changed from trifluorotoluene to hexafluorobenzene. This
might be the result of the alkyl perfluoroalkyl shell under-
taking a different orientation in each respective solvent.
Two factors that seem to play an important role in the
mechanism of encapsulation are: 1) the flexibility and orien-
tation of both the shell and core at the interface of the two
phases (aqueous and perfluorous phase); and 2) the strong
interaction of the polar guest (e.g. anionic dyes) with the
polar groups in the core of the dendritic architecture. The
flexibility and orientation of the perfluorinated shells are
determined by the degree of functionalization, the degree of
branching, and the polarity gradient between the core and
the shell. The low transport capacities observed for PGs 3
and 5 can be attributed to the weak interactions that arise
between the polar (anionic) guests and the polyether polyol
core, whereas the higher transport capacities of PEIs 7 and
9 arise because of the strong interactions between the anion-
ic species and the amino groups in the core of the dendritic
architecture.
Thermal behavior of dendritic architectures that contain
perfluorinated shells: The interesting properties observed in
solution for dendritic architectures that contain fluorinated
shells raised the question whether as bulk materials they
possess ordered phases. The thermal behavior of the den-
dritic architectures was studied by differential scanning calo-
rimetry (DSC) to obtain insight into their bulk structures.
The results are summarized in Table 3, and a typical DSC
diagram of polyamide 7 is shown in Figure 4. While the PG-
Table 3. Glass transition temperatures Tg and melting temperatures for
perfluorinated PEI. Tg values were obtained by extrapolation to a heat-
ing rate of 08Cminꢁ1
.
Polymer
Tg [8C]
Tm [8C]
3a
3b
3c
3d
5a
5b
5c
5d
6
68
47
71
52
41
23[a]
21[a]
26[a]
48
50
ꢁ14[b]
150
75
ꢁ10
14
7
9
ꢁ3[a]
80
[a]Weak Tg at a heating rate of 368Cminꢁ1. [b]PG-core contains a
stearylamine starter.
Encapsulation of metal salts and nanoparticle formation:
Silver cations have been encapsulated in PEIs that contain
perfluorinated shells because dendritic polyamines have
good ligand properties. The encapsulated silver cations were
then reduced with an excess of sodium triacetoxyborohy-
dride according to a previously described procedure.[13b]
Upon reduction of the encapsulated silver salt an orange-
brown color was observed (455 nm); this indicated the for-
mation of silver nanoparticles. The transmission-electron mi-
croscopy (TEM) image (Figure 3) of encapsulated silver
Figure 4. DSC diagram of PEI perfluoroamide 7. Heating rate was
368Cminꢁ1
.
based systems displayed none, or in the case of 5a c, only
very small glass transitions, the PEI cores that contained
perfluorinated shells showed glass transitions in the range of
ꢁ10 to 148C. In contrast to the results obtained for per-
fluorinated carbosilane dendrimers,[20] liquid crystalline be-
havior could not be detected for either core polymer type
(PG or PEI).
Figure 3. TEM image and particle size distribution of silver nanoparticles
encapsulated in nanocarrier 9. The bar in the TEM image is equivalent
to 100 nm.
nanoparticles reveals that the nanoparticles formed are
spherical and have a narrow distribution. For dendrimer 9,
the silver particles obtained had diameters of 3.9ꢀ1 nm
(Figure 3). This demonstrates that dendritic polyamines that
contain perfluorinated shells can act as templates and stabi-
lizers[13b,d] for silver nanoparticles even in the fluorous phase.
Moreover, this effect can then be used either in the prepara-
tion of catalytically active nanoparticles[6,13f] or in bacterio-
static applications.[13d]
Conclusion
To study the physicochemical properties of molecular nano-
carriers that have extremly nonpolar peripheries, we synthe-
sized a number of novel dendritic architectures that contain
covalently linked perfluorinated shells from glycerol den-
Chem. Eur. J. 2004, 10, 2822 2830
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2827

R. Haag et al.
FULL PAPER
(CH2=C), 135.3 ppm (CH2=C); MS (70 eV, EI): m/z (%): 81 (40), 421.1
(39), 639.5 (40), 753.5 (100) [M+H]+.
drimers [G0-G3], commercially available hyperbranched
polyglycerols (PG), and polyethyleneimines (PEI). While
perfluorinated PGs are soluble in many perfluorinated sol-
vents, perfluorinated PEIs are soluble only in some per-
fluorinated solvents such as hexafluorobenzene or hot FC-
75. These amphiphilic core-shell architectures are able to
provide polar environments in very nonpolar reaction media
such as fluorous phases. Phase-transfer behavior between an
aqueous/fluorous phase interface was demonstrated by the
encapsulation and transport of polar molecules (e.g. dyes
and metal salts) into the fluorous phase. It is noteworthy
that high transport capacities were only observed for the
dendritic PEIs that contained perfluorinated shells, while
poor transport was detected for the corresponding PG-
based systems independent of their core size. The different
transport behavior observed clearly depends on the guest
molecule and the core of the dendritic architecture undergo-
ing different interactions. We also demonstrated that the en-
capsulated guest molecules remain accessible to small mole-
cules after transport into the fluorous phase. As a result,
this behavior can be used for the generation of polar nano-
reactors in very nonpolar reaction media such as fluorous
phases and supercritical CO2. In some cases (e.g. 9) a ther-
momorphic behavior was also observed. This was important
because it would allow the host guest system to be separat-
ed from the perfluorinated solvent by precipitation at room
temperature. Evaluation of these molecular nanocarriers as
homogeneous supports for catalysts is currently in progress.
[G2.5] Allyl ether dendrimer 1c: (M(C89H148O28)=1664 gmolꢁ1): 4.7 g
(63%). 1H NMR (300 MHz, CDCl3): d=0.82 (TBAB traces), 1.22
(TBAB traces), 3.30 3.80 (m, OCH2), 3.96 (d, OCH2CH=C), 4.10 (d,
OCH2CH=C), 5.11 (m, CH2=C), 5.21 (s, CH2=C), 5.26 (s, CH2=C),
5.85 ppm (m, CH2=CHCH2); 13C NMR (300 MHz, CDCl3): d=29.6
(TBAB traces), 45.4 (C(CH2)4), 70.3 (OCH2), 71.3 (OCH2), 71.9 (OCH2),
72.3 (OCH2), 78.6 (OCH2), 116.7 (CH2=C), 134.7 (CH2=C), 135.2 ppm
(CH2 =C); MS (70 eV, EI): m/z: 1688.1, 1574, 1459.9, 855.5.
[G3.5] Allyl ether dendrimer 1d: (M(C185H308O60)=3488 gmolꢁ1): 0.67 g
(61%). 1H NMR (300 MHz, CDCl3): d=0.82 (TBAB traces), 1.22
(TBAB traces), 3.20 3.70 (m, OCH2), 3.95 (s, OCH2CH=C), 4.10 (s,
OCH2CH=C), 5.11 (t, CH2=C), 5.20 (s, CH2=C), 5.26 (s, CH2=C),
5.85 ppm (m, CH2=CHCH2); 13C NMR (300 MHz, CDCl3): d=29.6
(TBAB traces), 45.4 (C(CH2)4), 70.4 (OCH2), 71.3 (OCH2), 71.7 (OCH2),
72.0 (OCH2), 72.2 (OCH2), 72.6 (OCH2), 77.4 (OCH2), 78.7 (OCH2), 78.9
(OCH2), 116.6 (CH2=C), 134.8 (CH=C), 135.3 ppm (CH2=C); MS (70 eV,
EI): m/z: 3493, 3378, 3264, 3149, 3035.
General procedure for the synthesis of dendritic polyglycerols by dihy-
droxylation (2a c): CAUTION: Osmium tetroxide is toxic. A 4 %wt
OsO4 solution in water (2 mL) was added to a solution of the polyallyl
ether (0.1 mol allyl equivalents) and N-methylmorpholine-N-oxide
(14.8 g, 0.11 mol) in acetone (50 mL), distilled water (50 mL), and tert-bu-
tanol (10 mL). Initially, an exothermic reaction was observed and it was
necessary at times to cool the reaction mixture in a water bath. The mix-
ture was then stirred for 20 h at 258C. The volatile compounds were re-
moved under vacuum, and the product was further purified by reverse-
phase column chromatography (methanol) to give, after concentration,
pale-yellow oils. The high molecular weight [G3]dendrimer 2c was puri-
fied by dialysis in methanol using a benzylated cellulose membrane
(MWCO 1000, Sigma).
Characterization data
[G1] Glycerol dendrimer 2a: (M(C17H36O12)=432 gmolꢁ1): 9.1 g (103%,
product contained traces of NMO). 1H NMR (300 MHz, D2O): d=2.79
(s, COH), 3.38 (m, CH2O), 3.53 (m, CH2O), 3.76 ppm (quintuplet,
CH2O); 13C NMR (300 MHz, D2O): d=47.5 (C(CH2)4), 55.5 (CH2O),
65.0 (CH2O), 66.1 (CH2O), 72.3 (CH2O), 72.6 (CH2O), 74.7 ppm
(CH2O); MS (70 eV, EI): m/z: 439.0.
[G2] Glycerol dendrimer 2b: (M(C41H84O28)=1024 gmolꢁ1): 8.7 g
(101%, product contained traces of NMO). 1H NMR (300 MHz, D2O):
d=0.74 (s, COH), 1.14 (s, COH), 3.30 3.90 ppm (m, CH2O); 13C NMR
(300 MHz, D2O): d=12.0, 20.7, 27.9, 43.3 (C(CH2)4), 60.7 (CH2O), 68.1
(CH2O), 68.4 (CH2O), 68.7 (CH2O), 70.2 (CH2O), 76.0 ppm (CH2O); MS
(70 eV, EI): m/z: 593.7, 1030.6.
[G3] Glycerol dendrimer 2c: (M(C89H180O60)=2208 gmolꢁ1): 1.9 g (68%,
after extensive dialysis). 1H NMR (300 MHz, D2O): d=0.70 (t, COH),
1.20 (q, COH), 1.50 (m, COH), 3.10 (s, CH2O), 3.16 (s, CH2O), 3.49 (m,
CH2O), 3.75 ppm (m, CH2O); 13C NMR (300 MHz, D2O): d=14.9, 21.3,
25.2, 47.5 (C(CH2)4), 60.2 (CH2O), 60.8 (CH2O), 63.5 (CH2O), 64.7
(CH2O), 66.9 (CH2O), 71.3 (CH2O), 72.5 (CH2O), 72.7 (CH2O), 72.8
(CH2O), 74.3 (CH2O), 80.3 ppm (CH2O); MS (70 eV, EI): m/z: 2099.4,
2173.5, 2247.5.
Experimental Section
Materials: Hyperbranched polyglycerol was prepared as previously de-
scribed by using bis(2,3-dihydroxypropyl)stearylamine and pentaerythri-
tol as initiators.[10b] Polyethyleneimine was obtained from BASF. Per-
fluorinated reagents and solvents were obtained from 3M. Perfluoroben-
zene, a,a,a-trifluorotoluene, and hexaflourobenzene were purchased
from Aldrich. 2,2’-Azobisisobutyronitrile (AIBN) was purchased from
Fluka. Sodium hydroxide, potassium carbonate, and all other solvents
were purchased from Merck. Allyl chloride was received from Solvay.
All reagents and solvents were used without any further purification.
General procedure for the synthesis of dendritic polyglycerols by allyla-
tion (1a d, 4a d): CAUTION: Allyl chloride is toxic. Allyl chloride
(162.8 mL, 2 mol) was added to a solution of tetrahydroxymethylmethane
(13.6 g, 0.4 mol OH groups), tetrabutylammonium bromide (TBAB) as
phase-transfer catalyst, and NaOH (80 g, 2 mol) in distilled water
(80 mL), and the reaction mixture was then stirred for 24 h at 508C.
After the addition of toluene (400 mL), the organic phase was separated,
dried over MgSO4, filtered,and concentrated under vacuum. The crude
product was further purified by column chromatography (silica gel, pe-
troleum ether/ethyl acetate 10:1 to 1:1) to give a colorless oil. In some
cases the product had a tendency to slowly polymerize (especially at
higher generations). For long-term storage, it was necessary to keep the
allyl ether product under an inert atmosphere at ꢁ208C.
Synthesis of perfect glycerol dendrimers (3a d) and dendritic polyglycer-
ols (5a d) with perfluorinated shells: A solution of the polyallyl ether
1a d or 4a d (0.1 mol allyl ether group equivalents) and
3,3,4,4,5,5,6,6,7,7,8,8,8-tridecafluorooctane-1-thiol (0.5 mol) was stirred
under reduced pressure (3 mbar) to remove oxygen from the solution.
After heating to 708C, AIBN (2 mmol) was added under an atmosphere
of nitrogen, and the reaction mixture was stirred for 2 h. After further
addition of the same amount of AIBN, the mixture was stirred for anoth-
er 20 h at 708C, and the solvent was then evaporated to yield a pale-
yellow product. For higher generations (molecular weights), further puri-
fication was achieved by dialysis in trifluorotoluene.
Characterization data
[G0.5] Tetraallyl ether 1a: 19.5 g (66%). The spectroscopic data were in
good agreement with those reported in the literature.[15]
[G1.5] Allyl ether dendrimer 1b: (M(C41H68O12)=752 gmolꢁ1): 6.3 g
(78%). 1H NMR (300 MHz, CDCl3): d=3.30 3.50 (m, OCH2), 3.61 (t,
OCH2), 3.96 (d, OCH2CH=C), 4.08 (d, OCH2CH=C), 5.10 (t, CH2=C),
5.22 (d, CH2=C), 5.84 ppm (m, CH2=CHCH2); 13C NMR (300 MHz,
CDCl3): d=45.6 (C(CH2)4), 70.2 (OCH2), 70.5 (OCH2), 71.3 (OCH2),
71.7 (OCH2), 72.3 (OCH2), 116.5 (CH2=C), 116.8 (CH2=C), 134.8
Characterization data for perfect four-arm dendrimers with perfluorinat-
ed shells (3a d)
[G0.5] Compound 3a: (M(C49H48O4S4F52)=1816 gmolꢁ1): 1.3 g crude.
3
1H NMR (300 MHz, C6F6/CDCl3): d=1.78 (quintuplet, JH,H =7 and 6 Hz,
8H; OCH2CH2CH2S), 2.28 (tt, 3JH,F =18 and 3JH,H =9 Hz, 8H;
2828
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2822 2830

Phase-Transfer Properties of Dendritic Nanocarriers that Contain Perfluorinated Shell Structures
2822 2830
3
CH2CH2CF2), 2.50 2.70 (m, 16H; CH2S), 3.35 (s, 8H; CCH2O), 3.44 ppm
(t, 3JH,H =6 Hz, 4H; OCH2CH2); 13C NMR (75 MHz, C6F6/CDCl3): d=
22.6 (SCH2), 29.1 (SCH2), 29.8 (CH2CH2CH2), 32.3 (t, 1JF,C =22 Hz;
CH2CF2), 40.3 (C(CH2)4), 45.8 (CH2O), 69.6 (CH2O), 110 120 ppm (m,
CF2); MS (70 eV, EI): m/z: 149, 421, 493, 521, 939, 958, 1140.
CH2CF2), 40.3 (d, JF,C =21 Hz, CH2CF2), 68.7 (CH2O), 69.7 (CH2O), 71.1
(CH2O), 72.1 (CH2O), 78.8 (CH2O), 105 125 ppm (m, CF2).
Synthesis of dendritic polyamines that contain perfluorinated shells
Synthesis of the partially fluorinated dendritic polyamine 6: A suspension
of PEI (Mn(core)=4500 gmolꢁ1
, PD=1.3, 0.5 g, 4 mmol of primary
[G1.5] Compound 3b: (M(C105H108O12S8F104)=3792 gmolꢁ1): 1.4 g crude.
1H NMR (300 MHz, C6F6/CDCl3): d=1.80 (m, 16H; OCH2CH2CH2S),
2.29 (tt, 3JH,F =17 and 3JH,H =9 Hz, 16H; CH2CH2CF2), 2.50 2.70 (m,
32H; CH2S), 3.30 3.60 (m, 40H; CCH2O), 3.65 ppm (tt, 3JH,H =6 and
6 Hz, 4H; OCH); 13C NMR (75 MHz, C6F6/CDCl3): d=22.7
(CH2CH2CH2), 28.7 (SCH2), 29.1 (SCH2), 29.2 (SCH2), 29.8 (SCH2), 30.1
amino groups), 1,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8-heptadecafluoro-10-iodode-
cane (4.69 g, 8 mmol), and potassium carbonate (9 g, 65 mmol) in aceto-
nitrile (200 mL) was heated to reflux for 15 h. The reaction mixture was
then washed with water and methanol, and dried to give a brownish
rubber-like material (0.94 g, 60% based on functionalization). An NMR
spectrum could not be recorded because of the poor solubility of the
product in perdeuterated or perfluorinated solvents. Elemental analysis
found (%): C 35.05, H 3.83, N 8.13. Degree of functionalization from
CHN-analysis: 29%.
1
(SCH2), 32.3 (t, JF,C =22 Hz; CH2CF2), 46.0 (C(CH2)4), 68.9 (CH2O), 69.9
(CH2O), 70.6 (CH2O), 71.8 (CH2O), 72.1 (CH2O), 78.5 (CH2O), 115
125 ppm (m, CF2); MS (70 eV, EI): m/z: 617.9, 1115.9, 1687.8, 2199.9,
2314.1, 2791.5, 3516.2.
Synthesis of the partially fluorinated dendritic polyamide 7: A mixture of
[G2.5] Compound 3c: (M(C217H228O28S16F208)=7744 gmolꢁ1): 2.3 g
(70.5%, after dialysis). 1H NMR (300 MHz, C6F6/CDCl3): d=1.84 (m,
32H; OCH2CH2CH2S,), 2.50 (tt, 3JH,F =18 and 3JH,H =9 Hz, 32H;
CH2CH2CF2), 2.80 2.90 (m, 64H; CH2S), 3.40 3.80 ppm (m, 100H;
CCH2O); 13C NMR (75 MHz, C6F6/CDCl3): d=22.7 (CH2CH2CH2), 28.6
(SCH2), 28.7 (SCH2), 29.2 (SCH2), 30.3 (SCH2), 31.6 (SCH2), 32.2 (t,
1JF,C =17 Hz; CH2CF2,), 110 120 ppm (m, CF2); MS (ESI): m/z: 617.9,
1114.0, 1578.0, 2056.0, 2359.1, 2375.1, 2399.2, 2878.7, 2894.7, 3404.4.
[G3.5] Compound 3d: (M(C441H468O60S32F416)=15648 gmolꢁ1): 0.50 g
(57%, after dialysis). 1H NMR (300 MHz, C6F6/CDCl3): d=1.84 (m,
64H; OCH2CH2CH2S), 2.32 (m, 64H; CH2CH2CF2), 2.60 (m, 128H;
CH2S), 3.30 3.90 ppm (m, 212H; CH2O and CHO); 13C NMR (75 MHz,
C6F6/CDCl3): d=22.7 (CH2CH2CH2), 29.2 (SCH2), 29.8 (SCH2), 30.1
(SCH2), 32.23 (t, 1JF,C =22 Hz; CH2CF2), 68.9 (CH2O), 69 (CH2O), 70.1
(CH2O), 71.9 (CH2O), 72.1 (CH2O), 78.7 (CH2O), 78.9 (CH2O), 110
120 ppm (m, CF2); MS (ESI): no reliable peak detection was possible.
PEI (Mn(core)=4500 gmolꢁ1
, PD=1.3, 0.19 g, 1.5 mmol of primary
amino groups) and 4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-heptadecafluor-
oundecanoic acid (1.5 g, 3 mmol) was heated at 1408C for 4 h under an
atmosphere of argon. The reaction mixture was then dissolved in per-
fluorinated benzene (20 mL), filtered, Ionenaustauscher III (0.75 g,
4 mmol OHꢁ gꢁ1) (Merck) was added to the filtrate, and the resultant
mixture was stirred for 3 h. The Ionenaustauscher III was removed by fil-
tration and the solvent was evaporated under vacuum to give a brownish
1
solid (1.57 g, 95% based on functionalization). H NMR (300 MHz, C6F6/
CDCl3): d=0.6 (s), 1.0 (s), 1.3 4.0 ppm (m); 13C NMR (75 MHz, C6F6/
CDCl3): d=28 60 (m, CH2), 100 125 (m, CF2), 172 ppm (s, C=O); ele-
mental analysis found (%): C 34.83, H 3.12, N 4.57. Degree of functional-
ization from CHN-analysis: 56%.
Synthesis of the partially fluorinated dendritic polyalkylamide 9: A mix-
ture of PEI (Mn(core)=4500 gmolꢁ1
, PD=1.3, 3.57 g, 54.3 mmol of
amino group) and 10-undecenoic acid (10 g, 54.3 mmol) was heated at
1208C for 20 h under an atmosphere of argon (stream). The crude amide
8 (11.4 g) was then used without further purification in the next step.
Perallylated hyperbranched polyglycerols (4a d)[9]
Compound 4a: (Mn(core)=2000 gmolꢁ1): 1H NMR (300 MHz, CDCl3):
A mixture of polyethyleneimine alkenylamide 8 (11.4 g, 49 mmol of allyl
ether groups) and 3,3,4,4,5,5,6,6,7,7,8,8,8-tridecafluorooctane-1-thiol
(93.38 g, 246 mmol) was stirred under reduced pressure (3 mbar) to
remove oxygen from the solution. The mixture was then heated to 1008C
and AIBN (4 mmol) was added under an atmosphere of nitrogen. The re-
action mixture was stirred for 8 h. After further addition of AIBN
(8 mmol) the mixture was stirred for another 8 h at 1008C, and then the
excess AIBN was deactivated by addition of butylhydroxytoluene. The
polymer was purified by dialysis in chloroform using a benzoylated cellu-
lose membrane (MWCO 1000, Sigma), and the solvent was evaporated
to yield a yellow polymer (29.0 g, 90% over two steps based on function-
alization). 1H NMR (CDCl3): d=1.51 (m, CH2), 1.82 (m, CH2CH2S and
NCOCH2), 2.44 (m, CH2S), 2.26 (m, NCH2), 2.78 (m, NCH2), 2.97 (m,
NCH2), 3.11 (m, NCH2), 3.73 (m, NCH2), 4.30 ppm (m, NCH2); 13C NMR
(CDCl3): d=15.8 (t, J=5 Hz; CH2), 22.9 (t, J=4 Hz, CH2), 23.7 (s, CH2),
26.4 (s, CH2), 28.8 (s, CH2), 29.2 (s, CH2), 29.7 (m, CH2), 30.0 (m, CH2),
31.9 (s, CH2), 32.1 (s, CH2), 32.4 (s, CH2), 32.6 (s, CH2), 32.7 (s, CH2),
36.0 (s, CH2), 36.3 (s, CH2), 36.6 (s, CH2), 37.0 (s, CH2), 105 125 (m,
CF2), 180.0 ppm (s, CO). Degree of functionalization from 1H NMR:
67%.
3
d=3.1 3.6 (m, CH2O, CHO), 3.78 (d, CH2CH=CH2, JH,H=4 Hz), 3.92 (d,
3JH,H =4 Hz, CH2CH=CH2), 4.93 (m, CH=CH2), 5.02 (d, 3JH,H =2 Hz,
CH=CH2), 5.08 (d, 3JH,H =2 Hz, CH=CH2), 5.70 ppm (m, CH=CH2); 13C
NMR (75 MHz, CDCl3): d=45.4 (CCH2) 70.2 (CH2O), 71.2 (CH2O),
71.6 (CH2O), 72.2 (CH2O), 77.3 (CH2O), 77.6 (CH2O), 78.6 (CH2O), 78.8
(CH2O), 116.5 (CH2=CH), 116.6 (CH2=CH), 116.7 (CH2=CH), 134.7
(CH2=CH), 134.8 (CH2=CH), 135.2 ppm (CH2=CH).
Hyperbranched polyglycerols that contain perfluorinated shells (5a d)
Compound 5a: (Mn(core)=2000 gmolꢁ1): 3.1 g crude. 1H NMR
(300 MHz, C6F6/CDCl3): d=1.9 (m, OCH2CH2CH2S), 2.3 (m,
CH2CH2CF2), 2.7 (m, CH2S), 3.3 3.9 ppm (m, CH2O and CHO);
13C NMR (75 MHz, C6F6/CDCl3): d=22.5 (CH2CH2CH2), 28.4 (SCH2),
29.1 (SCH2), 29.7 (SCH2), 30.1 (SCH2), 31.5 (SCH2), 31.7 (SCH2), 31.9
(SCH2), 32.6 (t, 1JF,C =23 Hz; CH2CF2), 68.8 (CH2O), 69.8 (CH2O), 70.8
(CH2O), 72.0 (CH2O), 78.7 (CH2O), 79.0 (CH2O), 105 125 ppm (m,
CF2).
Compound 5b: (Mn(core)=5000 gmolꢁ1): 9.5 g (73%, after dialysis).
1H NMR (300 MHz, C6F6/CDCl3): d=1.9 (m, OCH2CH2CH2S), 2.3 (m,
CH2CH2CF2), 2.7 (m, CH2S), 3.3 3.9 ppm (m, CH2O and CHO);
13C NMR (75 MHz, C6F6/CDCl3): d=22.5 (CH2CH2CH2), 28.4 (SCH2),
Determination of the number of encapsulated guest molecules: Solid
congo red dye was added to a solution of the respective dendritic nano-
carrier 3, 5, 6, 7, or 9 (50 mg) in 5 mL of solvent (see Table 3) and the re-
sultant mixture was stirred for two days. The suspension was filtered and
the absorbance of the solution at 530 nm was measured. The concentra-
tion of congo red was determined from a calibration curve in water. The
transport capacity n (number of encapsulated congo red molecules) was
calculated by using Equation (1).
1
29.1 (SCH2), 29.8 (SCH2), 30.1 (SCH2), 31.7 (SCH2), 32.3 (t, JF,C =22 Hz;
CH2CF2), 68.7 (CH2O), 69.7 (CH2O), 72.2 (CH2O), 78.7 (CH2O), 79.0
(CH2O), 79.5 (CH2O), 79.6 (CH2O), 105 120 ppm (m, CF2).
Compound 5c: (Mn(core)=15000 gmolꢁ1): 7.1 g (41%, after dialysis).
1H NMR (300 MHz, C6F6/CDCl3): d=1.8 (m, OCH2CH2CH2S), 2.3 (m,
CH2CH2CF2), 2.6 (m, CH2S), 3.3 3.9 ppm (m, CH2O and CHO);
13C NMR (75 MHz, C6F6/CDCl3): d=22.6 (CH2CH2CH2), 28.6 (SCH2),
1
29.1 (SCH2), 29.7 (SCH2), 30.1 (SCH2), 31.8 (SCH2), 32.2 (t, JF,C =23 Hz;
½dyeŠ
CH2CF2), 68.7 (CH2O), 69.8 (CH2O), 72.0 (CH2O), 78.6 (CH2O), 78.8
(CH2O), 105 120ppm (m, CF2).
n ¼
ð1Þ
½polymerŠ
Compound 5d: (Hyperbranched polyglycerol with a stearylamine starter
group, Mn(core)=4500 gmolꢁ1): 8.5 g (50%, after dialysis). 1H NMR
(300 MHz, C6F6/CDCl3): d=1.44 (m, OCH2CH2CH2S), 2.03 (m,
CH2CH2CF2), 2.58 (m, CH2S), 2.88 (CH2O, CHO), 3.81 ppm (CH2O and
CHO); 13C NMR (75 MHz, C6F6/CDCl3): d=22.4 (CH2CH2CH2), 29.0
(SCH2), 29.8 (SCH2), 30.2 (SCH2), 32.3 (SCH2), 32.3 (t, 3JF,C =22 Hz;
Chem. Eur. J. 2004, 10, 2822 2830
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2829

R. Haag et al.
FULL PAPER
2000, 6, 2499 2506; c) Y. H. Kim, O. W. Webster, J. Am. Chem. Soc.
1990, 112, 4592 4593; d) B. Voit, J. Polym. Sci. Part A 2000, 38,
2505 2525.
Acknowledgements
The authors thank Dr. Ralf Thomann for the TEM measurements, Ben
Konfitin and Fabricio Malek for their support with respect to synthesis,
and Professor Rolf M¸lhaupt for his generous support. BASF, Solvay,
and 3M are gratefully acknowledged for their donation of chemicals.
R.H. is indebted to the Deutsche Forschungsgemeinschaft, the Fonds der
Chemischen Industrie, for financial support. A.G.-B. thanks the Spanish
Ministery of Education for a postdoctoral grant.
[9]R. Haag, A. Sunder, J.-F. Stumbÿ, J. Am. Chem. Soc. 2000, 122,
2954 2955.
[10]a) A. Sunder, R. M¸lhaupt, DE Patent 19947631 A1 (Germany),
1998; b) A. Sunder, R. Hanselmann, H. Frey, R. M¸lhaupt, Macro-
molecules 1999, 32, 4240 4246; b) A. Sunder, R. M¸lhaupt, R.
Haag, H. Frey, Macromolecules 2000, 33, 253 254.
[11]M. Kr‰mer, J.-F. Stumbÿ, G. Grimm, B. Kaufmann, U. Kr¸ger, M.
Weber, R. Haag, ChemBioChem 2004, in press.
[12]a) H. Frey, R. Haag, Rev. Mol. Biotechnol. 2002, 90, 257 267; b) A.
Sunder, R. M¸lhaupt, R. Haag, H. Frey, Adv. Mater. 2000, 12, 235
239.
[13]R. Haag, J.-F. Stumbÿ, A. Sunder, H. Frey, A. Hebel, Macromole-
cules 2000, 33, 8158 8166.
[1]For a recent review on host guest chemistry of dendrimers and den-
dritic nanocarriers for drug delivery see: a) M. W. P. L. Baars, E. W.
Meijer, Top. Curr. Chem. 2000, 210, 131 182; b) S.-E. Stiriba, H.
Frey, R. Haag, Angew. Chem. 2002, 114, 1385 1390; Angew. Chem.
Int. Ed. 2002, 41, 1329 1334; c) R. Haag, Angew. Chem. 2004, 116,
280 284; Angew. Chem. Int. Ed. 2004, 43, 278 282.
[2]For recent reviews on dendrimer encapsulated catalysts see: a) D.
Astruc, F. Chardac, Chem. Rev. 2001, 101, 2991 3024; b) R. Kreiter,
A. W. Kleij, R. J. M. K. Gebbink, G. v. Koten, Top. Curr. Chem.
2001, 217, 163 199; c) G. E. Oosterom, J. N. H. Reek, P. C. J.
Kamer, P. W. N. M. van Leeuwen, Angew. Chem. 2001, 113, 1878
1901.
[3]a) C. Tzschucke, C. Markert, W. Bannwarth, A. Hebel, S. Roller, R.
Haag, Angew. Chem. 2002, 114, 4136 4173; Angew. Chem. Int. Ed.
2002, 41, 3694 4001; b) D. J. Cole-Hamilton, Science 2003, 299,
1702 1706; c) D. P. Curran, Angew. Chem. 1998, 110, 1230 1255;
Angew. Chem. Int. Ed. 1998, 37, 1174 1196; d) I. T. Horvµth, J.
Rµbai, Science 1994, 266, 72 74.
[4]A. I. Cooper, J. D. Londono, G. Wignall, J. B. McClain, E. T. Samul-
ski, J. S. Lin, A. Dobrynin, M. Rubonstein, A. L. C. Burke, J. M. J.
Frÿchet, J. M. DeSimone, Nature 1997, 389, 368 371.
[14]a) M. Kr‰mer, J.-F. Stumbÿ, H. T¸rk, S. Krause, A. Komp, L. Delin-
eau, S. Prokohova, H. Kautz, R. Haag, Angew. Chem. 2002, 114,
4426 4431; Angew. Chem. Int. Ed. 2002, 41, 4252 4256; b) R.
Haag, M. Kr‰mer, J.-F. Stumbÿ, S. Krause, A. Komp, S. Prokhorova,
Polymer Preprints 2002, 43(1), 328; c) S.-E. Stiriba, H. Kautz, H.
Frey, J. Am. Chem. Soc. 2002, 124, 9698 9699; d) C. Aymonier, U.
Schlotterbeck, L. Antonietti, P. Zacharias, R. Thomann, J. C. Tiller,
S. Mecking, Chem. Commun. 2002, 3018 3019; e) M. Q. Slagt, S.-E.
Stiriba, R. J. M. K. Gebbink, H. Kautz, H. Frey, G. van Koten, Mac-
romolecules 2002, 35, 5734 5737; f) S. Mecking, R. Thomann, H.
Frey, A. Sunder, Macromolecules 2000, 33, 3958 3960; g) A.
Sunder, M. Kr‰mer, R. Hanselmann, R. M¸lhaupt, H. Frey, Angew.
Chem. 1999, 111, 3758 3761; Angew. Chem. Int. Ed. 1999, 38,
3552 3555.
[15]R. M. Nouguir, M. Mchich, J. Org. Chem. 1985, 50, 3296 3298.
[16]V. VanRheenen, R. C. Kelly, D. Y. Cha, Tetrahedron Lett. 1976, 17,
1973 1976.
[17]Lower degrees of perfluoro functionalization (ca. 40%) have also
been investigated but did not result in higher transport capacities
(0.014 congo red molecules) for the resultant dendritic polyglycerol
architectures.
[5]a) V. Chechik, R. Crooks, J. Am. Chem. Soc. 2000, 122, 1243 1244;
b) L. K. Yeung, R. M. Crooks, Nano Lett. 2001, 1, 14 17.
[6]R. M Crooks, M. Zhao, L. Sun, V. Chechik, L. K. Yeung, Acc.
Chem. Res. 2001, 34, 181 190.
[7]For books and reviews on dendrimer synthesis see: a) G. R. New-
kome, C. N. Moorefield, F. Vˆgtle, Dendritic Molecules: Concepts,
Syntheses, Perspectives, 2nd ed., Wiley-VCH, Weinheim, 2001;
b) Dendrimers and other Dendritic Polymers, (Eds.: D. Tomalia,
J. M. J. Frÿchet), Wiley, Chichester, 2001; c) M. Fischer, F. Vˆgtle,
Angew. Chem. 1999, 111, 934 955; Angew. Chem. Int. Ed. 1999, 38,
884 905; d) A. W. Bosman, H. M. Janssen, E. W. Meijer, Chem.
Rev. 1999, 99, 1665 1688; e) H. F. Chow, T.-K.-K. Mong, F. M.
Nongrum, C.-W. Wann, Tetrahedron 1998, 54, 8543 8668; f) O. A.
Matthias, A. N. Shipway, J. F. Fraser-Stoddart, Prog. Polym. Sci.
1998, 23, 1 56.
[18]In the following, we use the general term ×dendritic architecture× for
functional core shell architectures that originate from hyper-
branched macromolecules that contain a DB in the range 60 75%,
as well as in regard to perfect dendrimers (DB=100%) because
they both have similar properties.
[19]a) D. E. Bergbreiter, Y-S. Liu, P. L. Osburn, J. Am. Chem. Soc. 1998,
120, 4250 4251; b) D. E. Bergbreiter, P. L. Osburn, A. Wilson, E. M.
Sink, J. Am. Chem. Soc. 2000, 122, 9058 9064.
[20]K. Lorenz, H. Frey, B. St¸hn, R. M¸lhaupt, Macromolecules 1997,
30, 6860 6868.
[8]a) H. Frey, R. Haag, in Encyclopedia of Materials, Science and Tech-
nology (Eds.: K. H. J. Buschow, R. H, Cahn, M. C. Flemings, B.
Ilschner, E. J. Kramer, S. Majahan), Elsevier, Oxford 2001,
pp. 3997 4000; b) A. Sunder, J. Heinemann, H. Frey, Chem. Eur. J.
Received: July 4, 2003
Revised: January 14, 2004
Published online: April 22, 2004
2830
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2822 2830