Metal-Containing Dendritic Polymers Fiona J. Stoddart and Thomas Welton* Department of Chemistry, Imperial College of Science, Technology and Medicine, South Kensington, London SW7 2AY, UK. Abstract Metal-containing dendrimers (metallodendrimers) have attracted a great deal of attention recently and their study is becoming a growing field. Many workers have entered the field and it is rapidly developing. In this review, the preparation, characterisation and applications of metal-containing denrimers are discussed. The principal methodologies for the preparation of dendrimers are first demonstrated and then the derivatisation of organic dendrimers to form suitable potential ligands is presented. Finally the formation of transition-metal complexes of the dendrimers is discussed. The manuscript is organised such that the metallodendrimers are discussed by donor element in the dendrimer. As one might expect, phoshine and nitrogen-donor complexes have dominated this initial phase of synthesis. However, there are reports of metallodendrimers with a wide variety of donor atoms. In 1
59
Embed
Start Here - Imperial College London · Web viewHydrolysis of ester-terminated PAMAM dendrimers with Group 1 metal (Na+, K+, Cs+ and Rb+) hydroxides resulted in the formation of salts
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Metal-Containing Dendritic Polymers
Fiona J. Stoddart and Thomas Welton*
Department of Chemistry, Imperial College of Science, Technology and Medicine, South
Kensington, London SW7 2AY, UK.
Abstract
Metal-containing dendrimers (metallodendrimers) have attracted a great deal of attention
recently and their study is becoming a growing field. Many workers have entered the field
and it is rapidly developing. In this review, the preparation, characterisation and applications
of metal-containing denrimers are discussed. The principal methodologies for the
preparation of dendrimers are first demonstrated and then the derivatisation of organic
dendrimers to form suitable potential ligands is presented. Finally the formation of
transition-metal complexes of the dendrimers is discussed. The manuscript is organised such
that the metallodendrimers are discussed by donor element in the dendrimer. As one might
expect, phoshine and nitrogen-donor complexes have dominated this initial phase of
synthesis. However, there are reports of metallodendrimers with a wide variety of donor
atoms. In the few years since the first metallodendrimers were prepared the field has moved
rapidly towards potential applications, and this has been noted.
1
Introduction
Over the last twenty years, a new class of polymers known as dendrimers has fascinated
many chemists. This review concentrates on those dendrimers that contain metals.
However, a brief introduction to dendrimers in general and the major approaches to their
syntheses is given. More detailed reviews on this subject have been given elsewhere.1
The term dendrimer is derived from the Greek word dendra meaning tree. These
highly branched macromolecules have compelling molecular structures that are reminiscent
of patterns often observed in nature and particularly those found in trees and in coral.
Dendrimers – also called arborols2 or cascade3 molecules – exhibit controlled patterns of
branching and ideally are monodisperse, i.e.; all the molecules should have exactly the same
molecular masses, constitutions and average dimensions. The larger dendrimers, which have
globular structures, carry many close-packed surface end groups and contain internal cavities.
The interest in dendritic polymers stems from the possibility that their architectures, which
differ from those of traditional linear step-growth polymers, offer exciting prospects of new
applications.4
Before 1940, branched molecular structures had been considered to be responsible for
the insoluble and intractable materials formed during polymerisations.Error: Reference
source not foundb These materials were largely ignored since it was invariably impossible to
isolate discrete molecular compounds and assign them definite structures.
In 1978 Vögtle and co-workers published a synthetic strategy which involved the
“cascade-like” synthesis of acyclic, branched polyamines.5 The synthesis, which is illustrated
in Scheme 1, began with an exhaustive Michael addition of the monoamine 1 to acrylonitrile,
leading to the annexation of two branches per amino group, thus affording the bisnitrile 2.
The nitrile groups were then reduced to amine functions, using cobalt(II)/sodium
borohydride to give the bisamine 3. Repetition of these two steps afforded the hexa-2
branched tetraamine 5, via the tetranitrile 4. Although this synthesis was not continued
beyond this point because of problems encountered in the reduction step, the principle that
repeated cycles of reactions could lead to controlled polymer growth had been demonstrated.
N
H2N NH2
N
N
NH2H2N
R
N
NC NC
N
N
CNCN
R
H2N
N
NH2
R
NC
N
CN
R
NH2
1
R
2
Co(III)/NaBH4
Co(III)/NaBH4
3
4 5
AcOH MeOH
AcOH MeOH
CN
CN
Scheme 1. “Cascade-like” synthesis of acyclic, branched polyamines
In 1981, Denkewalter et al.6 patented the synthesis of highly branched polylysine derivatives.
Each member of this series of compounds was monodisperse, consisting of branching units
of differing lengths. From 1985 onwards, two research groups, one headed by Tomalia7 and
the other by Newkome,Error: Reference source not found,8 simultaneously developed
families of dendrimers synthesised using this divergent method (see below). In 1990,
Fréchet and Hawker9 employed a different method, the convergent approach (see below), to
prepare poly(aryl ether) dendrimers.
3
Dendritic Structure
Figure 1 depicts the structure of a typical dendrimer. The following points must be
considered and, where appropriate, adapted when describing the structures of dendrimers:-
(i) There is a central point known as the initiator core: in the dendrimer shown in Figure
1, four branches emanate from a core and so the core multiplicity (Nc) is four.
(ii) Each branch contains further branching sites: in the example illustrated in Figure 1,
the degree of branching (Nb) is two.
(iii) Each new layer of branches that are constructed upon old branch points is called a
generation (G): generations are numbered at 0, 1, 2, 3 ... and so on.
(iv) The branch cell unit lengths (l) are determined by the choice of branched monomers.
l
G = 2
G = 1
Figure 1. Schematic representation of a dendrimer
4
The number of monomer units in a dendrimer increases exponentially as a function of the
generation. As the dendrimer grows in size, the end groups reside closer and closer to one
another. Eventually, this branch-growing process results in surface congestion, a feature that
prevents further growth from all branch points with the consequence that the dendrimer can
no longer be monodisperse. The highest generation at which the dendrimer is still potentially
monodisperse is described as its “starburst limit”.
5
Dendrimer Synthesis
Dendrimers are constructed in stages using repetitive synthetic strategies. Both the divergent
and convergent approaches to dendrimer synthesis have advantages and disadvantages.
The Divergent Approach
The synthetic approach to dendrimer formation, which has become known as the divergent
method, emerged during the period 1978-1987 with many of the seminal contributions
coming from Newkome Error: Reference source not found,Error: Reference source not found and Tomalia.Error:
Reference source not found The basic concept, which is that of starting at the core and
working outwards in a divergent fashion to create a highly branched structure, has
subsequently been developed and exploited by many research groups world-wide.10 An
illustration of the divergent approach to the synthesis of a dendrimer is shown in Scheme 2.
8 x
Coremolecule
First-generationdendrimer
Second-generationmolecule
4 x
Scheme 2. Schematic representation of the divergent synthesis of dendrimers
6
A multifunctional core molecule – in this case, one with four functional groups – is reacted
with four monomer molecules to give the first generation dendrimer. Repetitive addition of
similar building blocks – usually achieved by a protection-deprotection procedure – affords
successive generations. It is important to ensure that each set of reactions leading to these
new generations has been completed before the next cycle of reactions is commenced, if
defects in the dendritic structure are to be avoided.
Using the divergent approach, it is possible to prepare up to tenth generation
dendrimers with molecular weights of the order of 700,000 and with more than 3,000 end
groups per molecule.11 The advantage of the divergent method is that the production of
several grams of dendrimer is easily attainable since, with each subsequent generation, the
molar mass of the dendrimer is greatly increased.
This method is not without its drawbacks. As the dendrimer grows in size, the
number of end groups involved in the reaction increases and the likelihood of incomplete
growth steps leading to defects in the structure becomes greater. It is often difficult to detect
the precise extent of conversion from one generation to the next. As a consequence,
imperfect samples of dendrimers, which are virtually impossible to purify and characterise,
since they may differ only slightly from the desired monodisperse samples, are obtained.
Therefore, if the divergent method is to be employed successfully, extremely efficient and
high-yielding reactions are required in order to ensure the production of dendrimers with low
polydispersities. This often poses a great synthetic challenge.
The Convergent Approach
Fréchet and Hawker first proposed an alternative approach to dendrimer syntheses, known as
the convergent method.Error: Reference source not found,12 Here, the reverse of the
divergent method is applied; the synthesis starts at what will eventually become the periphery 7
of the dendrimer and progresses inwards. Surface units are linked together increasingly with
more monomers until a wedge-shaped molecule is generated, carrying a reactive group at its
apex. The final step of the synthesis involves attaching the desired number of wedges to a
multifunctional core. This approach is illustrated in Scheme 3.
The attraction of the convergent method lies in the fact that only a small number of
molecules are involved in the reaction steps that form each successive generation. In
contrast, increasing numbers of molecules are involved in the reactions in the later stages of a
synthesis using the divergent approach. Large excesses of reagents and slight impurities can
also be avoided, without sacrificing high yields and, because of easier purification, reactions
no longer need to be as efficient, meaning that a much larger choice of reaction types are
available.
+
+
Two of I
II
Two of II Repeat
n times
Three wedges
+
WedgeDendrimer
I
Scheme 3. Schematic representation of the convergent synthesis of dendrimers
8
The main disadvantage of the convergent approach is that it is not accompanied, after
each reaction cycle, by the marked increase in molar mass, which is observed in the case of
the divergent method. The total number of steps involved in the construction of the
dendrimer using the convergent method is not actually reduced compared with that needed in
the divergent approach, yet significantly more starting material is required. Also the higher
generation dendritic wedges can experience severe steric problems when reactions to attach
their reactive apex groups to core molecules are attempted. Thus, the convergent approach
has been found to be less useful than the divergent one for the synthesis of dendrimers
approaching their starburst limit.
Metallodendrimers
During the last decade, those working with dendrimers have switched their focus from the
initial synthetic directions explored mainly by organic chemists to a more applied emphasis.
Thus, metallodendrimers are becoming of interest from a materials science perspective
because of their unique physical properties, leading to potential photophysical and catalytic
applications. Metallodendrimers show substantial structural diversity and their properties and
applications are wide-ranging. Metallodendrimers may be classified by where the metal
appears in the dendrimer, at the centre, as connectors, as branching units, or as peripheral
units of the dendrimer.13 However, here the metallodendrimers are classified by ligand type,
i.e., the particular ligand which complexes the metal centre to/within the dendrimer – thus
viewing them from the perspective of the inorganic chemist. In many of the following
illustrations, only one section of the dendrimer has been portrayed and a “W” within a
wedge-shaped motif represents other dendritic arms identical to the one which has been
drawn out in full.
9
Dendrimers as Counter ions
Perhaps the simplest way in which to include metals in a dendritic structure is to use the
dendrimer as a counterion for a well-defined metal or metal complex. The metal may bind to
a surface site on the dendrimer (exo-receptor) or to a site within the internal cavaties of the
dendrimer (endo-receptor).
Hydrolysis of ester-terminated PAMAM dendrimers with Group 1 metal (Na+, K+,
Cs+ and Rb+) hydroxides resulted in the formation of salts as white hygroscopic
powders.Error: Reference source not foundd Direct observation of these single dendrimer
molecules by Channel Tunelling Electron Microscopy has been achieved. Further studies
were conducted using carboxylate-terminated PAMAMs and their complexes with Fe3+, Gd3+,
Mn2+, Pr3+ and Y3+ ions.Error: Reference source not foundg
In another investigation,14 which sought to support the molecular mechanics
simulations with experimental evidence, the properties of the carboxylate salts of the half-
generation PAMAM dendrimers were likened to those of anionic micelles. The ability of
these anionic dendrimers to effect the kinetics of the electron-transfer quenching of
photoexcited Ru(phen)32+ has been examined.15 The emission decay of the metal-to-ligand
charge transfer (MLCT) excited state of the probe – Ru(phen)32+ bound to half-generation
PAMAMs was analysed in the presence and absence of the quencher – Co(phen)33+. The
studies showed that the probe lifetimes were enhanced when the complexes were bound to
dendrimers as compared with unbound complexes. It was concluded that the quenching of
dendrimer-bound Ru(phen)32+ by Co(phen)3
2+ occurs at the surface of the dendrimer. These
results indicate that the cationic Ru(phen)32+ binds strongly to the negative surface of the
dendrimer. This has been confirmed by another study study of Ru(phen)32+ labeled with a
nitroxide radical, via –NHC(O)OCH2- or –O(CH2)8O- units, as an EPR probe.16 More
recently, similar results using protonated amino-terminated PAMAMs and Ru(4,7-10
(SO3C6H5)2-phen)34- as the probe have been reported.17 Hence, these systems provide
examples of a dendrimer acting as an exo-receptor.
Phosphorus-Donor Metallodendrimers
Phosphorus-containing dendrimers, in which the core and subsequent branch points are
pentavalent phosphorus atoms and which possess peripheral aldehyde groups have been
prepared by Majoral et al.Error: Reference source not foundd The dendrimers – up to the
tenth generation – were functionalised with phosphino groups and then reacted with
AuCl(tetrahydrothiophene) to give dendrimers with AuCl moieties as the peripheral units.18
The authors note that the reactivity of all generations towards gold complexation is similar,
and therefore, independent of the size of dendrimer used. Most recently, Majoral et al.19
have reported the incorporation of gold into different generational layers of dendritic
molecules. Complexation occurs both at the sulfur-donor P=N-P=N-P=S fragments and the
terminal CH2PPh2 moieties. The dendritic fragment, shown in Figure 2, has been modified
at the generation 1 level to introduce ligands, which are able to coordinate gold. The
metallodendrimer has eighteen internal AuCl units – six at the P=N-P=N-P=S linkages and
twelve at the phosphino groups. The complexes formed can be characterised unambiguously
by 31P NMR spectroscopy. Studies are currently underway to extend this methodology to
incorporate a variety of different metals within the cascade structure of dendrimers.
11
O CH
N N
Me
CH2
CN
N
MeP
CH N
N
Me
P
S
P N PO
O
CN
NMe
PS
CH
N N
MeP
HC N N
Me
P
S
SO
O
O
O
CN
N
Me
NP
O
HCN
NP
Ph
Ph
O
CHN
NMeP
Ph
Ph
S
OArOAr
OArOAr
OArS
OAr
Me
S
POAr
OAr
NP
NP
N
P
Au-ClCl-Au
Cl-Au
H
H
H
WW
W
W W
1 a) Matthews, O. A.; Shipway, A. N.; Stoddart, J. F. Prog. Polym. Sci., 1998,
9, 54. b) Newkome, G. R.; Moorefield, C. N.; Vögtle, F. Dendritic Molecules; VCH:
Weinheim, Germany, 1996 c) Tomalia, D. A. Sci. Am., 1995, 272(5), 62. d) Dvornic,
P. R.; Tomalia, D. A. Chem. Br., 1994 (Aug), 641. e) Tomalia, D. A. Adv. Mat.,
1994, 6, 529. f) Topics in Current Chemistry, 1998, 197, g) Tomalia, D. A.; Durst, H.
D. in Top. Curr. Chem., Springer: Berlin, Germany, 1993, 165, 193. h) Tomalia, D.
A.; Naylor, A. M.; Goddard III, W. A. Angew. Chem., Int. Ed. Engl., 1990, 29, 138.