Amphiphilic dendrimers - Pure · Amphiphilic Dendrimers PROEFSCHRlFT ter verkrijging van de graad...

Amphiphilic dendrimers

Citation for published version (APA):Roman Vas, C. (1999). Amphiphilic dendrimers. Technische Universiteit Eindhoven.https://doi.org/10.6100/IR523682

DOI:10.6100/IR523682

Document status and date:Published: 01/01/1999

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:[email protected] details and we will investigate your claim.

Download date: 07. Nov. 2020

https://doi.org/10.6100/IR523682

https://doi.org/10.6100/IR523682

https://research.tue.nl/en/publications/amphiphilic-dendrimers(a9be18bf-536c-4ed5-a8b9-54f59a14a727).html

Amphiphilic Dendrimers


PROEFSCHRlFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de Rector Magnificus, prof.dr. M. Rem, voor een commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op woensdag 23 juni 1999 om 16.00 uur

door

Cristina Román V as

geboren te Tarragona, Spanje

Dit proefschrift is goedgekeurd door de promotoren:

prof.dr. E.W. Meijer

en

prof.dr. G.R. Newkome

Copromotor:

dr.ir. M.H.P. van Genderen

Omslag: Cristina Román V as, Ben Mobach, TUE

Druk: Universiteitsdrukkerij, TUE

CIP-DATA LIBRARY TECHNISCHE UNIVERSITEIT EINDHOVEN

Román V as, Cristina

Amphiphilic Dendrimers I Cristin a Román V as.

Eindhoven: Technische Universiteit Eindhoven, 1999.

Proefschrift.

ISBN 90-386-2571-5

NUGI 813

Trefwoorden: Amfifiele dendrimeren I supramoleculaire architecturen I polymeer aggregaten

Subject headings: Amphiphilic dendrimers I supramolecular assemblies I polymerie aggregates

rtorecen las primaveras

de risas y de aguaceros

como jlorecen los campos

de la tierra que más quiero.

':S'obre el os euro abiS!ltO en que te mece.r"

Manoio Garcfa Garcfa-Pérez

A mis padres, a J avi y Carlos A Paul

por vuestro apf!JoJ por vuestro carino.

Chapter 1


1.1. Introduetion

Contents

1.2. Dendritic-linear amphiphilic polymers

1.3. Unimolecular dendritic micelles

1.4. Aim and scope of the thesis

l.S. References

Chapter 2

Experimental Techniques

2.1. Introduetion

2.2 Monolayers

2.2.1. Langmuir films

2.2.2. Langmuir-Blodgett films

2.2.3. Self-assembled monolayers

2.3. Aggregation in salution

2.3.1. Critica] aggregation concentration

Pyrene fluorescence technique

2.4. Vesicles

2.4.1. Phase transitions

Fluorescence depolarization

Pyrene excimer formation

2.4.2. Turbidity measurements

2.5. References

Chapter3

3

7

10

14

17

18

18

19

21

22

22

23

24

25

27

32

36

39

Aggregation Behaviour of Polystyrene-Poly(propylene imine) Dendrimers

3.1. Introduetion 43

3.2. Monolayers 45

3.2.1. Langmuir films 45

3.2.2. Self-assembled monolayers 48

3.3. Aggregation in solution

3.3.1. Transmission electron microscopy

3.3.2. X-ray diffraction

3.3.3. Critica) aggregation concentration

3.4. Yesiele properties

3.4.1. Bilayerfluidity

Microcalorimetry



3.4.2. Bilayer structure

Dis tribution of porphyrins in the bilayer

Grientalion of porphyrins in the bilayer

3.5. Metal complexes

3.6. Conclusions

3.7. Experimental

3.8. References

Chapter4

Solid-State Mieropbase Separation of Polystyrene-Poly(propylene imine)

Dendrimers

4.1. Introduetion

4.2. Microphase separation

4.2.1. Materials

4.2.2. Mieropbase separation in PS-dendr-(COOH)n

Smal! angle X-ray scattering measurements

Transmission electron microscopy

Spin-coated thin films

4.3. Conclusions

4.4. Experimental

4.5. References

49

49

52

53

57

57

58

58

60

60

61

63

65

68

69

74

77

79

82

85

85

90

91

92

94

95

Chapter 5

Self-Assernbly of Alkyl-Modified Dendrirners

5.1. Introduetion 99

5.2. Mieropbase separation of palmitoyl functionalized dendrirners 104

5.3. Aggregation behaviour of amphiphilic dendrirners at the interface 106

5.3.1 Monolayers at the air/water interface 106

5.3.2 Langmuir-Blodgett films 109

5.4. Aggregation behaviour of arnphiphilic dendrirners in solution 110

5.4.1. Yesiele properties 111

Electron micrascopy 111

Dynamic light scattering 113

X-ray dif.fraction 114

pH dependenee of aggregation 115

Osmotic behaviour 116

5.4.2. Critica) aggregation concentration 117

5.4.3. Phase transition temperature 118

5.4.4. Microviscosity 124

Discussion on phase transition and microviscosity 126

5.5~ Conclusions; Jlexible dendrimers 128

5.6. Experimental 130

5.7. References 133

Chapter6

Suprarnolecular Assemblies of Palrnitoyl Poly(propylene irnine) Dendrirners

and Surfactants

6.1. Introduetion

6.2. Supramolecular assemblies

6.2.1. CTAB/DAB-dendr-(NHCO-(CH2)14-CH3)64 aggregates

6.2.2. Formation of aggregates

6.2.3. Microviscosity in the aggregates

6.2.4. Non-ionic and anionic surfactants

6.2.5. Azobenzene-containing dendrimers

137

138

138

140

141

144

147

6.2.6. Conclusions

6.3. Host-guest system.s. Encapsulation of dyes

6.3.1 Encapsulation of carboxyfluorescein

6.3.1 Encapsulation of ANTS

6.4. pH controlled release of encapsulated dyes

6.5. Conclusions

6.6. Experimental

6.7. References

Epilogue

Summary

Resumen

Curriculum vitae

List of publications

Dankwoord

Agradecimiento

149

149

150

153

154

154

157

160

163

165

167

169

170

171

174

Abstract

1


"Science proceeds ftrst by open exploration, the time of initia/ discoveries. Th en follows

reconnai.rsance, gatbering etlidence on a broad front, pursuing leads to the jàr corners of

the prob/ent, and generating preliminary hypotheses. Finally, !f warranted by the

resttlts of the exploration and reconnairsance, there comes the time for detailed studies

and the testing of more informed I?; pothes es. "

J.F. Wilford

Dendrimers are versafile molecular structures due to their multifunctionality and specific

shape. These unique propenies make them attractive molecules for their use as building

blocks in larger, organized structures of higher complexity. The introduetion of amphiphilic

character in dendritic molecules brings the necessary driving forces (based in e.g.

electrostalie forces, hydrogen bonding and van der Waals interactions) to obtain

spontaneous supramolecular assemblies with unusual and unprecedented properties. In this

first chapter a shon introduetion is given to the area of dendritic molecules with amphiphilic

propenies.

1.1. Introduetion

Nanostructures with dimensions in the 1-100 nm range1 are playing a dominant role in

bringing together the disciplines of biology, chemistry and physics. In biology, nanostructures

are the smallest expression of life. Proteins, viruses, and bacteria are nanosized structures

which have been self assembied from smaller units. Although individual atoms in the

subunits are covalently bonded, assembly of these subunits is maintained by non-covalent

interactions, such as van der Waals, hydrogen-bonding, electrostatic and hydrophobic

interactions. From the chemist' spoint of view, nanosized structures are giant arrangements of

macromolecules with molar mass in the order of 108, composed of millions of atoms. To

ChapterJ

obtain such structures via covalent bonding only is very difficult to achieve for a synthetic

chemist. However, chemists have made important progress in creating self-organized and

supramolecular matcrials in the size domain of nanostructures by the non-covalent assembly

of macromolecules. This progress may yield advanced materials. In the biologica] world there

are numerous examples of self-organizing supramolecular assemblies that may function as

molecular receptors, catalysts and carriers. Being the main components of membranes,

amphiphilic molecules and the assemblies thereof are a special class of systems with

profound interest. In the last decades, many efforts have been made to imitate biologica!

functions? Successful pioneering studies by Ringsdorf,3 Fendler,4 Kunitake5 and others6 have

led to a better understanding of how small molecules assembie to form higher aggregates such

as membranes and how diffusion of small molecules into these membranes occurs.

Macro molecular architectures may provide further guidance for other efforts to develop novel

self-assembling synthetic organizations not found in nature. The development of this new

class of functional matcrials demands innovative chemica] tailoring. In the Jast decades,

advances in synthetic chemistry and characterization techniques permitted a rapid

development of a new kind of well-defined, high molecular weight molecules denominated as

dendrimers.7•

8•

9•

10 Dendrimers are attractive macromolecules whose novel architecture and

properties make them very appropriate for their use as building blocks in supramolecular

assemblies.ll As early as 1986, an exciting example of such assemblies was reported by

Newkome et al.12 Arborol dendrimers (Figure 1.1) form linear networks and exhibit unusual

micellar and gelation-type properties.

HO~) t(OH/OH

HO HO~NCO corl::'-E HO~NO~CH~ntONH H

fK) >K)r-2_,d , ~H H

~~ 6H

Figure 1.1. Example of a Newkome 's arborol didendron. 12

2

Amphiphi/ic Dendrimers

Small changes in the molecular architecture of these molecules can dramatically alter their

macroscopie properties. When the dendrimers were added in low concentrations to water,

gelation occurs due to hydrophobic interactions, packing effects and maximization of H

bonding. The ability of these compounds to induce gel formation was shown to be affected by

the length and rigidity of the spacer. The linear hyperbranched networks formed were clearly

visible by transmission electron microscopy (TEM) analysis of the gel. Rod-shaped

aggregates with a monodisperse diameter of ca. 4 nm were observed. The alkyl ebains of the

molecule associate due to hydrophobic interactions in the interlor of the aggregates, while the

dumbbell-shaped arborols are probably stacked in an orthogonal way to form long thin tubes.

The hydroxy endgroups are directed towards the aqueous phase. H-bonding between the

amide groups may also play a role in the packing and contribute to the stability of the

aggregates. Insertion of an alkyne unit in the core introduces a helical and scissors-Iike

morphology in the stacked array.13 The rigidity of the hydrophobic tail may play an important

role in this behaviour; the molecules may in this case prefer a non-orthogonal stacking.

Benzene-based arborols also yield H-bonded aggregates in the form of spherical micelles. The

critica! micelle concentration was found at 2 mM with dynamic light scattering (DLS)

techniques. 14

Tomaha et al. introduced concepts for the use of dendrimers as hosts for small

molecules and for their use as templates to mimic biologica! systems.7'

15 Varlation of the

dendrimer surface allows dendrimers as building blocks for supramolecular assemblîes. This

might include templating groups for organizing lipid layers or bilayers; cationic surfactants

have been successfully organized on the surface of dendrimers. A cooperative effect was

noted for higher generation P AMAM dendrimers according to photoluminescence

techniques, 16 as well as the introduetion of pores for controlling the en try and exit of guest

molecules, which gave rise to the design of the dendritic box by Johan Jansen in our

laboratories in 1995P Construction of networks of dendritic building blocks has potential

applications in such different areas of matcrials science as molecular electronics, information

processing and biomolecular and liquid crystal engineering.8•

18

1.2. Dendritic-linear amphiphilic polymers

Hybrids of linear polymers and dendrimers are unusual bleek copolymers because they

combine in one molecule the random coil conformation of a long flexible chain with the

3

Chapterl

restricted conformation and flexibility of dense globular dendrimers. Fréchet et al. prepared

the first polymer-dendrimer block copolymer by rea.cting monofunctional narrow molecular

weight distribution PEO with poly(benzyl ether) dendritic wedges (Figure 1.2).19 ABA block

copolymers were also prepared with varying PEO block lengtbs and different dendrimer

generation numbers. The solubility and aggregation of these copolymers was strongly affected

by the molecular weight ratio of the Iinear block to the dendritic block, as well as the size of

the dendrimer. In genera!, copolymers containing low generation dendrons tended to form

unimolecular micelles, while higher generations formed micelles from the aggregation of

various molecules.

Figure 1.2. Fréchet' s AB and ABA PEO-<iendrimer block copofymers.

An illustration of the various phases that can he observed is represented in Figure 1.3 for the

case of an ABA type block copolymer, where A is the hydrophobic dendrimer and B the

hydrophilic PEO. The exact location of the phase boundaries is dependent on the polymer

concentration. The boundary between the monomolecular and micellar phases depends on

whether the experimental concentration is above or below the critica! micelle concentration.

The boundary between micellar salution and gel is defined by the equilibrium between loops

and bridges formed by the soluble PEO blocks. A similar but less complete phase diagram

was also obtained for AB type polymer-dendrimer block copolymers. However, in that case

no gel phase is produced for lack of bridging.

4

PEO monomolecular solutlon

10000 0

5000 0

1000

micellar solutlon

0

• •

[G-1 I [G-2) [G-3) [G-4}


Figure 1.3. Phase diagram of ABA poly(benzyl ether) dendrimer PEO in CH30H:H20 ( 1:1 ). 19b

The single focal point of the dendrimer can be used to initiate the polymerization of

monomer, e.g. of E-caprolactone. The experimental and calculated molar mass agrees with the

ratio of monomer feed and initiator and the molar mass distribution was narrow ?0 They also

used the focal functional group of the dendritic wedge to initiate living free radical

polymerization of styrene. For the TEMPO mediated free radical polymerization of styrene,

the focal point is converted to a TEMPO derivative. 21 The same wedges were used to

polymerize 4-acetoxystyrene. The resulting copolymer can be hydrolyzed to an amphiphilic

block copolymer consisting of a hydrophilic poly(vinylphenol) and hydrophobic poly(benzyl

ether) dendrimer.

Dendrimers prepared by the divergent metbod are grown from a preformed polymer

having a functional endgroup. The first synthesis by this metbod used glycine modified PEO

and poly(a,E-lysine) to grow a poly(a,E-L-lysine) dendrimer.22 Up to four generations of

lysine were synthesized. At this generation the molar mass of the poly(a,E-L-lysine)

dendrimer and the PEO block are comparable. These amphiphilic block copolymers were

found to decrease the surface tension of water and form micelles in solution. Hammond and

coworkers reported the synthesis of a new series of hybrid linear-dendritic diblock

copolymers with PEO as the linear block and PAMAM as the dendritic block?3 Intrinsic

viscosity and GPC techniques were used to study the behaviour of these copolymers in

5

Chapterl

aqueous solution, which was found to be dependent on the length of the PEO linear block.

The results obtained for diblocks with Jonger PEO chain Jength suggest the formation of a

unimicellar like structure formed by wrapping of the PEO chain around the dendrimer block.

Differential scanning calorimetry (DSC) analysis of these diblock copolymers indicated that

mieropbase separation took place in the bulk, although no bulk morphology was described.

Recently, poly(propylene imine) dendrimers have been synthesized on an amino-terminated

poly(2-methyl-2-oxazoline) by Okada and coworkers. These structures present the possibility

to vary hydrophilicity and hydrophobicity by changing the substituents of the polyoxazoline

tail and the endgroups of the dendrimer. 24

Poly(propylene imine) dendrimers have been constructed stepwise onto an amine

functionalized polystyrene for the first time by Jan van Hestin our laboratorles (Figure 1.4 )?5

NH, NH,

Figure 1.4. Fourth generation PS-poly(propylene imine) dendrimer block copolymer.

The resulting block copolymers present a high versatility, since the dendritic headgroup can

be varied in size and chemica) functionality. The aggregation behaviour of amphiphilic

molecules depends on the nature and the concentration of the amphiphile, the nature of the

solvent, and the method of preparation. The thermodynamically preferred structure formed for

a certain surfactant has been rationalized by considering its headgroup and tail volume ratio?6

The aggregation behaviour of PS-dendrimer diblock copolymers was found to be

qualitatively in agreement with these theoretica] calculations. Van Hest analyzed the

structures formed by PS-poly(propylene i mine) dendrimers by means of DLS and TEM?5•

27

PS-dendr-(NH2)4 is soluble in organic solvents, where it forms spherical inverted micelles.

6

Amphiph1ïic Dendrimers

Block copolymers with 8, 16, and 32 amine endgroups are soluble in water and form vesicles,

cylinders and spherical micelles, respectively. These shapes are the classica! cylindrical,

lamellar and spherical phases of block copolymers in the solid state. However, the boundary

between the phases occurs at very different volume fractions, due to the different packing

requirements of the linear polymer and the spherical dendrimer at the interface, as it will be

described in Chapter 4.

Due to the commercial interest in amphiphilic block copolymers and the potentially

interesting properties of these novel macromolecules, the characteristics of amphiphilic linear

polymer-dendrimer systems has been extensively studied in our group and will be described

in Chapters 3 & 4 of this thesis.

1.3. Unimolecular dendritic micelles

Modification of the dendrimer endgroups with aliphatic tails gave rise to a new kind of

amphiphilic molecules. Alkyl modified dendrimers were compared to micellar systems for the

first time by Newkome in 1985, who observed the solubilization of different probes in the

hydrophobic interior of these water-soluble Micellanols.Z8

af....,, ~,....,._,.,.....afqft

af....,ft

Figure 1.5. Example of one ofNewkome's unimolecular micelles.28

A micelle has two concenttic spherical regions: an inner core consisting of closed-packed

solvent-incompatible components and an outer shell of solvent-compatible (swollen)

7

Chapterl

moieties. Dendritic unimolecular micelles can be described as covalently fixed

microdomains, which mimic either regular or inverse micelles, depending on the

compatibility of the dendrimer surface with water. Appropriate dendrimer surface and interior

design offers many possibilities for unimolecular mimicry of micelles. Wîth this approach,

the headgroup multiplicity is determined by synthesis, in contrast to self-assembling micellar

systems, where the aggregation numbers are determined by free-energy minimization and/or

head-to-tail packing parameters. Micellanoli9 are water soluble structures that possess a

dendritic aliphatic interior and a hydrophilic exterior, consisting of carboxylic acid

ammonium salts, Pi gure l.S. UV -vis and steady-state as well as time-resolved fluorescence

measurements of various probes in aqueous solutions in the presence of unimolecular

micelles confirmed the presence of hydrophobic microdomains in water, similar to the ones

found in micellar solutions of small surfactants, even at extremely low concentrations of

dendrimcr (4·10-7 M). The dimensions of these unimolecular micelles could be estimated by

means of electron microscopy techniques. EM also showed the absence of higher aggregates

or clustering of micelles in basic aqueous solution.30 These unimolecular micelles were used

in a variety of applications, e.g. electrokinetic capillary chromatography.31

The group of Fréchet also prepared amphiphilic molecules containing multiple

dendritic blocks attached to the endsof a long PEG star polymer (Figure 1.6).32

Figure 1.6. Fréchet's star block copolymer.

Depending on the solvent polarity, these systems can adopt many different conformations. In

THF they were reported to form unimolecu lar micelles that have a hydrophilic core consisting

8

Amphiphilic Denddmers

of compactly packed PEG arms, surrounded by a loose hydrophobic shell of dendritic

wedges. This is similar to the behaviour observed for dendrimer-PEG-dendrimer triblock

copolymers. In polar media, the molecules adopt a conformation with the PEG arms forming

loops around the dendrimers to envelop the dendritic wedges. The dendritic blocks collapse to

form a new core. These molecules were able to respond to changes in the surrounding

medium forming unimolecular micelles with different core-shell structures (Figure 1.7). A

potential application of these stimuli-responsive molecules involves their use as solvent

specific encapsulation agents.

THF

Figure 1.7. Stimuli-responsive star copolymers with dendritic groups at the periphery.

Fréchet et al. also prepared hydrophobic dendrimers based on polyether dendritic wedges

containing a molecular probe as core and carboxylic acid units as endgroups (Figure 1.8).33

Although the molecules were water soluble, the À.max of the absorption of the probe indicated

that the core is protected from being in contact with water. These molecules were also able to

host small aromatic molecules, such as pyrene, in their hydrophobic interiors, even at

concentrations of dendrimer as low as 5·1 o-7 M. The guest molecules could be released by

solubilization of the system in organic solvents.

9

Chapterl

Figure 1.8. Micellar carboxylic acid-temzinated Fréchet's dendrimer.

Recently, the group of Fréchet synthesized a new kind of unimolecular micelle, based on

poly(benzyl ether) dendrimers, containing alkyl chain as endgroups and hydroxyl

functionalities in their interior that are available for hydrogen bonding interactions with other

species and that provide the dendritic interior with a polar microenvironment (see Figure

1.9).34

Figure 1.9. Fourth generation Fréchet' s unimolecu/ar micelle.

10

Amphiphi/ic Dendrimers

The catalytic activity of these molecules bas been tested for bimolecular nucleophilic

substitutions and elimination reactions of alkyl halides. Significantly, the elimination reaction

was catalyzed by the dendrimers to an even greater extent than the SN2 reactions, with

conversions of lOO% within 2 days, while the control reactions all give 0% conversion after

that time. An alternative unimolecular inverted micelle has been also prepared by this group,

in which the apolar outer shell surrounds polar tetraethyleneglycol spacers in the interior. The

second, third and fourth generation activated dendrons (Figure 1.1 0) were used in convergent

synthesis to obtain the respective dendritic molecules. The increased flexibility and

hydrophilicity of these dendrimers introduce new possibilities for encapsulation of larger

polar molecules, metal ions and other catalytic moieties.34

Figure 1.10. Fourth generation tetraethyleneglycol dendron.

In our laboratories, unimolecular inverted micelles based on alkyl-modified poly(propylene

imine) dendrimers were first prepared by Sandra Stevelmans and Jan van Hest.35 In this new

type of structures the core is hydrophilic, whereas the shell has a hydrophobic nature (see

Figure 1.11 ).

l1

ChapterJ

Figure 1.11. Schematic representation of an alkyl modified dendrimer (left) and an inverted micelle (right).

These molecules are well soluble in organic solvents such as THF and CHCh. Dynamic light

scattering (DLS) experiments confirmed the absence of cluster formation in diluted CH2Cb

solutions. Single partiele behaviour was found with a hydrodynamic diameter of ca. 2-3 nm.

Water-soluble dyes such as bengal rose can be easily used as guest for the hydrophilic

dendritic interior of the molecules. Liquid-liquid extraction experiments performed by

Maurice Baars in our group confirmed an acid-base interaction between the tertiary amines of

the dendritic interior and acid-functionalized water-soluble dyes.36 Extraction of small

molecules into supercritical carbon dioxide has also been reported using dendritic

unimolecular micelles with a fluorinated shell.37

Surprisingly, alkyl modified dendrimers can be also solubilized in aqueous solutions, where

they form well-defined supramolecular aggregates. The properties of these interesting

structures will bedescribed in Chapter 5. Alkyl-modified dendrimers are able to interact with

low molecular weight surfactants to form supramolecular complexes in aqueous solutions.

The combination of these systems with the ability of dendrimers to hold guest molecules

opens new routes towards the achievement of advanced materials. The possibilities and limits

of this approach will be discussed in Chapter 6.

12

Amphiphtïic Dendrimers

1.4. Aim and scope of the thesis

The rational design of supramolecular systems using amphiphilic dendrimers is a field of

chemistry with infinite possibilities for fundamental new discoverles and practical

applications. The use of dendrimers as building blocks for the construction of networks has

potential applications in such different areas of matcrials science as molecular electronics,

biomolecular and liquid crystal engineering and information processing.8•

38 The aim of this

thesis is to obtain detailed information about the behaviour and particular characteristics of

amphiphilic dendrimers by applying well-known techniques generally used in the study of

classica! surfactants; once more evidence is collected and the insight into the special

characteristics of these macromolecules becomes better, we can proceed to explore potenrial

applications for the most unusual and interesting properties of this new class of amphiphilic

molecules based on dendrimers. Furthermore, pioneering experiments have been performed

on dendritic unimolecular micelles, such as monolayer studies, vesicle formation, interactions

with low molecular weight surfactants, encapsulation and release of probes, etc. introducing

new unprecedented possibilities for the application of dendrimers as building blocks in the

field of supramolecular chemistry and ultimately, advanced materials.

Chemica! modifications of amphiphilic molecules permit us to change the

functionality and size of the molecules and, therefore, to control the physical and chemica!

properties of the aggregates formcd by the amphiphile. Many different techniques wiJl be

applied to study the behaviour of amphiphilic dendrimers in solution. The theoretica) basis of

some of these techniques in the study of amphiphilic dendritic block copolymers will be

described in Chapter 2.

In Chapters 3 and 4, the amphiphilic properties of polystyrene-poly(propylene imine)

dendrimer block copolymers in solution, at the air/water interface and in the solid state will

be reported. In solution, the aggregation of the amphiphiles is highly dependent on the

generation of the dendrimer block.25 The physical characteristics of the aggregates wiJl be

studied using the techniques described in Chapter 2. Small angle X-ray scattering (SAXS)

measurements and transmission electron microscopy (TEM) show evidence of mieropbase

separation of the block copolymers in the solid state. The micro-lattice morphology was

found to be highly dependent on the dendrimer generation as well?9

The second kind of dendritic amphiphiles (based on poly(propylene imine) dendrimers

modified with palmitoyl ebains) will be treatcd in Chapters 5 and 6. The formation of vesicles

13

Chopterf

in acidic water by palmitoyl modified dendrimers40 and the characteristics of this new type of

polymerie vesicles will be the topic of Chapter 5. The behaviour of palmitoyl modified

dendrimers at the air/water interface will be described as well. The arrangement of small

molecular weight surfactants around palmitoyl-modified dendrimers to form stabie

aggregates41 will be reported in Chapter 6. The description of encapsulation of small dyes in

these surfactant/dendrimer aggregates is included in Chapter 6 as wel!.

l.S. References

Lindsey, J.S. New 1. Chem. 1991, 15, 153.

2 Lehn, J.M. Angew. Chem. Int. Ed. Engl. 1988, 27, 89. (b) Lehn, J.M. Science 1985, 227, 849. (c)

Vögtle, F.; Weber, E. Angew. Chem. Int. Ed. Engl. 1979, 18, 753. (d) Vögtle, F.; Weber, E. Host-guest

complex chemistry 1-lll. Top. Curr. Chem. 1981, 98; 1982, JOl; 1984, 121. (e) Vögtle, F.; Weber, E.

Riomimetic and bioorganic chemistry 1-/l/. Top. Curr. Chem. 1985, 128; 1986, 1 32; 1986, 136.

3 Ringsdorf, H.; Schlerb, B; Venzmer, J. Angew. Chem. Int. Ed. Engl. 1988, 27, 113.

4 Fendler, J.H. Membrane mimetic chemistry, Wiley, New York, 1982.

5 Kunitake, T.; Okashata, Y. 1. Am. Chem. Soc. 1977, 99, 3860.

6 (a) Fuhrhop, J.H.; Mathiu, J. Angew. Chem. Int. Ed. Eng/. 1984, 96, 124. (b) Fuhrhop, J.H.; David,

H.H.; Mathiu, J.; Liman, U.; Winter, H.J.; Boekema, E. 1. Am. Chem. Soc. 1986, 108, 1785. (c)

Takigawa, D.Y.; Tirrell, D.A. Macromolecules 1985, 18, 338.

7 Tomalia, D.A.; Naylor, A.M.; Goddard lil, W.A. Angew. Chem. Int. Ed. Eng/. 1990,29, 138.

8 Newkome, G.R.; Moorefield, C.N.; Vögtle, F. Dendritic molecules: concepts, synthesis, perspectives.

VCH, Weinheim, 1996.

9 Fréchet, J.M.J.; Hawker, C.J. in Comprehensive Polymer Science, 2"d suplement, G.Ailen, Ed.

Pergamon, Elsevier Science, Oxford, 1996.

10 De Brabander-van den Berg, E.M.M.; Meijer, E.W. Angew. Chem., Int. Ed. Eng/. 1993, 32, 1308.

11 Tsukruk, V. T. Adv. Mater. 1998, JO, 253. (b) Zhao, M.; Tokuhisa, H.; Crooks, R. M. Angew. Chem.,

Int. Ed. Eng/. 1997, 36, 2596. (c) Tokuhisa. H.; Zhao, M.; Baker, L. A.; Phan, V. T.; Dermody, D. L.;

Garcia, M. E.; Peez, R. F.; Crooks, R. M.; Mayer, T. M. 1. Am. Chem. Soc. 1998, 120,4492.

12 (a) Newkome, G.R.; Baker, G.R.; Saunders, M.J.; Russo, P.S.; Gupta, V.K.; Yao, Z.Q.; Miller, J.E.;

Bouillion, K. 1. Chem. Soc., Chem. Commu1z. 1986, 753. (b) Newkome, G.R.; Baker, G.R.; Arai, S.;

Saunders, M.J.; Russo, P.S.; Theriot, K.J.; Moorefield, C.N.; Rogers, E.; Miller, J.E.; Lieux, T.R.;

Murray, M.E.; Phillips, B.; Pascal, L. J. Am. Chem. Soc. 1990, I 12, 8458.

13 Newkome, G.R.; Moorefield, C.N.; Baker, G.R.; Behera, R.K.; Escamilla, G.H.; Saunders, M.J. Angew.

Chem., Int. Ed. Eng/. 1992,31,917.

14


14 Newkome, G.R.; Yao, Z.Q.; Baker, G.R.; Gupta, V.K.; Russo, P.S.; Saunders, M.J. J. Am. Chem. Soc.

1986, 108, 849.

15 Tomalia, O.A.; Hall, B.M.; Hedstrand, O.M. Macromolecules 1987,20, 1164.

16 (a) Kumar, C.V.; Barton, J.K.; Turro, N.J. J. Am. Chem. Soc. 1985, 107, 5518. (b) Caminati, G.; Turro,

N.J.; Toma1ia, O.A. J. Am. Chem. Soc. 1990, ll2, 8515. (c) Ottaviani, M.F.; Turro, NJ.; Jockusch, S.;

Tomalia, O.A. J. Phys. Chem. 19%, 100, 13675. (d) Ottaviani, M.F.; Turro, N.J.; Jockusch, S.;

Tomalia, O.A. Colloid Sulfaces A: Physicochem. Eng. Aspects 1996, 115, 9. (e) Ottaviani, M.F.;

Andechaga, P.; Turro, N.J.; Tomalia, O.A. J. Phys. Chem. 1991, 101, 6057. (f) Caminati, G.; Turro,

N.J.; Tomalia, O.A. J. Am. Chem. Soc. 1990, 112, 8515. (g) Watkins, O.M.; Sayed-Sweet, Y.;

Klimash, J.W.; Turro, N.l; Tomalia, O.A. Langmuir 1997, 13, 3136. (h) Ottaviani, M.F.; Oaddi, R.;

Brustolon, M.; Turro, N.J.; Tomalia, O.A, Appl. Mag. Res. 1997,13, 347.

17 (a) Jansen, J.F.G.A.; de Brabander-van den Berg, E.M.M.; Meijer, E.W. Science 1994,226, 1226. (b)

Jansen, J.F.G.A.; de Brabander-van den Berg, E.M.M.; Meijer, E.W. Macromol. Symp. 1996, 102, 27.

18 (a) Crooks, R.M., Ricco, A.J. Acc. Chem. Res. 1998,31,219. (b) Percec, V.; Chu, P.; Ungar, G.; Zhou,

J. J. Am. Chem. Soc. 1995, ll7, 11441. (c) Lorenz, K.; Hölter, D.; Stühn, B.; Mül1haupt, R.; Frey, H.

Adv. Mater. 19%, 8, 414. (d) Cameron, J. H.; Facher, A.; Lattermann, G.; Oiele, S. Adv. Mater. 1997,

9, 398. (e) Baars, M. W.; Sön~jens, S. H. M.; Fischer, H.M.; Peerlings, H. W. 1.; Meijer, E. W. Chem.

Eu~ J. 1998,4,2456.

19 Gitsov, 1.; Wooley, K.L.; Fréchet, J.MJ. Angew. Chem. Int. Ed. Engl. 1992, 31, 1200. (b) Gitsov, 1.;

Fréchet, J.MJ. MacrorruJlecules 1993, 26, 6536. (c) Gitsov, 1.; Wooley, K.L.; Hawker, C.J.; Ivanova,

P.T.; Fréchet, J.M.J. Macromolecules 1993, 26, 5621. (d) Fréchet, J.MJ.; Gitsov, I. Macromol. Symp.

1995, 98, 441.

20 Gitsov, 1.; Jvanova, P.T.; Fréchet, J.M.J. Macromol. Rapid Commun. 1994, 15, 387.

21 Leduc, M.R.; Hawker, C.J.; Dao, J.; Fréchet, J.MJ. J. Am. Chem. Soc. 1996, I 18, 11111.

22 Chapman, T.M.; Hillyer, G.L.; Mahan. EJ.; Shaffer, K.A. J. Am. Chem. Soc. 1994, 116, 11195.

23 Iyer, J; Fleming, K.; Hammond, P. Macromolecules 1998,31, 8757.

24 Aoi, K.; Motoda, A.; Okada, M. MacromoL Rapid Commun. 1997, 18, 945.

25 van Hest, J.C.M.; Delnoye, O.A.P.; Baars, M.W.P.L.; van Genderen, M.H.P.; Meijer, E.W. Science

1995, 268, 1592. (b) van Hest, J.C.M.; Oe1noye, D.A.P.; Baars, M.W.P.L.; Elissen-Román, C.; van

Genderen, M.H.P.; Meijer, E.W. Chem. Eur. J. 1996, 2, 1616. (c) van Hest, J.C.M.; Baars, M.W.P.L.;

Elissen-Román, C.; van Genderen, M.H.P.; Meijer, E.W. Macromolecules 1995,28, 6689.

26 Israelachvili, J .N. intermolecu/ar & surface force !I Academie, New Y ork 1992.

27 van Hest, J .C.M. New molecular architectures basedon dendrimers. PhO Thesis, Eindhoven University

ofTechnology, 1996.

28 Newkome, G.R.; Yao, Z.Q.; Baker, G.R.; Gupta, V.K. J. Org. Chem. 1985,50, 2003.

15

Chapter 1

29 (a) Newkome, G.R.; Baker, G.R.; Moorefield, C.N.; Saunders, M.J. Polym. Preprints 1991, 32, 625. (b)

Newkome, G.R.; Moorefield,.C.N.; Baker, G.R.; Johnson, A.L.; Behera, R.K. Angew. Chem. Int. Ed.

EngL 1991, 30, 1176.

30 Newkome, G.R.; Mooretield, C.N.; Baker, G.R.; Saunders, M.J.; Grossman, S.H. Angew. Chem. int.

Ed. Engl. 1991. 30, 1178.

31 Kuzadal, S.A.; Monning, C.A.; Newkome, G.R.; Moorefield, C.N. J. Chem. Soc., Chem. Commun.

1994, 2139.

32 Gitsov, I.; Fréchet, J.M.J. J. Am. Chem. Soc. 1996, I /8, 3785.

33 Hawker, C.J.; Wooley, K.L.; Fréchet, J.M.J. J. Chem. Soc., Perkin Trans. i1993, 1287.

34 Piotti, M.E.; Hawker, C.; Fréchet, J.M.J.; Rivera, F.; Dao, J.; Bond, R. Polymer Preprinrs 1999, 40,

410.

35 Stevelmans, S.; van Hest, J.C.M.; Jansen, J.F.G.A.; Meijer, E.W.; van Boxtel, D.; de Brabander-van

den Berg J. Am. Chem. Soc. 1996, ii8, 7398.

36 Baars, M.W.P.L.; Froehling, P.E.; Meijer, E.W. Chem. Commun. 1997, 1959.

37 Cooper, A.I.; Londono, J.D.; Wignall, G.; McLain, J.B.; Samulski, E.T.; Lin, J.S.; Dobrinyn, A.;

Rubinstein, M.; Burke, A.L.C.; Fréchet, J.M.J.; DeSimone, J.M. Nature 1997, 389, 368.

38 (a) Bhyrappa, P.; Young, J.K.; Moore, J.S.; Suslick, K.S. J. Am. Chem. Soc. 1996, ii8, 5708. (b)

Knapen, J.W.J.; van der Made, A.W.; de Wilde, J.C.: van Leeuwen, P.W.N.M.; Wijkens, P.; Grove,

D.M.; van Koten, G. Nature 1994, 327, 659. (c) Reetz, M.T.; Lohmer, G.; Schwickardi, R. Angew.

Chem., int. Ed. Eng/. 1997, 36, 1526. (d) Bolm, C.; Dertien, N.; Seger, A. Synlett. 1996, 387. (e)

Mekelburger, H.-B.; Rissanen, K.; Vogtle, F. Chem. Ber. 1993, i26, 1161. (f) Sadamoto, R.; Tomoka,

N.; Aida, T. J. Am. Chem. Soc. 1996, i 18, 3978.

39 Román, C.; Fischer, H.R.; Meijer, E. W. Macromolecules, submitted.

40 Schenning, A.P.H.J.; Elissen-Román, C.; Weener, J.W.; Baars, M.W.P.L.; van der Gaast, S.J.; Meijer,

E.W. J. Am. Chem. Soc. 1998, 120, 8199.

41 Donners, J.J.J.M.; Román, C.; Heywood, B.R.; Meijer, E.W.; Nolle, R.N.; Schenning, A.P.H.J.;

Sommerdijk, N.A.J.M. Nature, submitted.

16

2 Experimental Techniques

'We are all slaves of our time, scientiftc trainin& and experience."

J.H. Fendier

Abstract

A wide variety of supramolecu/ar systems, including monolayers, Langmuir-Blodgett and

self-assembled films, micelles, polymerie vesicles, cast bilayers and other systems can be

obtained by the self-assembly of amphiphilic molecules. Nanosized particles and organized

films can be generaled in situ using well-developed methodologies. The application of

sophisticated techniques, such as the Langmuir-Blodgett method, atomie force microscopy,

electron microscopy, light scattering and the use of fluorescence probes, has facilitated the

better understanding of the nanoscopic materials generated. The development of advanced

techniques for the characterization of aggregates in salution and organized systems on

sulfaces allow us to achieve a higher knowledge and control of the size and morphology of

the aggregates, which is ultimately necessary to obtain advanced materials. In this Chapter,

a short description is given of the experimental techniques used in the investigations

peiformed in this thesis.

2.1. Introduetion

Colloidal chemistry is panicularly well suited for the development of nanostructured

matcrials science since colloidal aggregates (i.e., monolayers, multilayers, vesieles and

micelles) have often served as containers andlor templates for the construction of advanced

materials.1 Nanostructured matcrials (like biominerals) are constructed in nature in a stepwise

manner, going from the atomie to the macroscopie scale? At each step, the components are

held together by specific intcractions and are organized for optima! overall performance.

Many new supramolecular systems have been obtained from a large variety of synthetic

amphiphilic and polymerie molecules. Several innovative techniques like atomie force

microscopy,3 cryo-electron microscopy and X-ray diffraction, along with established

methodologies, like Langmuir techniques, luminescence of molecular probes, light scattering,

etc. have permitted the characterization of these self-assembled systems and their constituents

Chapterê

on an atomie and molecular scale. Investigation of synthetic systems has led to a better

understanding of many properties of the aggregates, including phase transitîons, phase

separatîon, osmotic actîvity, fluidity, permeability, fusion and substrate mobilities into and

out of the hydrophobic domains. Using a subdivision of the supramolecular systems into

monolayers, aggregates in solution in genera!, and vesicles in particular, we wiJl discuss in

the following section the backgrounds of the most relevant techniques as used throughout this

thesis.

2.2. Monolayers


Monolayers can be easily prepared by spreading a chloroform salution of an amphiphile on

water. After evaporation of the solvent, a film of known composition and weight is left on a

surface of known area. The hydrophilic headgroups of the molecules are immersed in the

water phase while the hydrophobic parts remain on the surface. The equipment developed

almost a century ago by Langmuir4 is basically the same as the technique used today to order

the monolayer into a well-defined 2D solid crystalline lattice. A rectangular trough is filled

with water and the monolayer is prepared by spreading the amphiphile solution on one side of

a movable harrier. The surface area can be controlled by rnaving the harrier. The surface

pressure is the reduction of the surface tension on the pure subphase at the air interface (y0) by

the monolayer film: n = Yo y.

4 3 2

I t::

A (Á2/molecule)

Figure 2.1. Suiface pressure-area isotherm for an ideal single-component monolayer. Region 1: gaseaus state; 2: transition region; 3: condensed state; and 4: supercompressed state.

18


The state of the monolayer at various n (surface pressure) and A (area) values can be

described by an isotherm. In the most ideal case, the isotherm will present 4 different regions

(see Figure 2.I). In region I, a single, homogeneous surface phase is present. The surfactant

molecules are spread on the water surface, relatively far from each other. In this region of the

isotherm, the film is comparable to a gas. Region 2 is a transition region in which the gaseous

phase coexists with a liquid phase; n is constant and independent of the film area A. In

region 3, the surface pressure is a function of the film area and the film is compressed to a

single condensed phase. The surfactant molecules in this phase have properties strikingly

similar to those of bulk Iiquids. In region 4, further compression results in the transition of

the surfactants into a two-dimensional solid state; individual molecules are densely packed

and direct their hydrophobic chains to the air phase. Further increase in the compression

introduces mechanica) instability and results in a monolayer collapse, accompanied by a

deercase in n. The pressure at which a monolayer collapses is characteristic for every

surfactant. Langmuir isotherms are generally given as surface pressure (in mN/m) vs

molecular area A (in Á 2/molecule) plots. ll-A isotherms provide helpful insight into the

molecular packing of surfactants in monolayers. A steep slope in the ll-A curve, for instance,

is indicative of strong chain-chain interactions and tight packing.

In order to access the molecular parameters of monolayers in a Langmuir experiment, it is

necessary to work in an environment with scrupulous cleanliness. Por the elimination of

surface-active contaminants, continua! sweeping of the surface of the water in the trough

must be done until no change in film pressure can be detected when the surface area is

reduced to I% of the full area of the trough. Th is process may take several hours. The

equipment and glassware has to be cleaned meticulously and solvents and solutions have to

be very pure and dust-free.


Monolayers and multilayers on solid substrates have been extensively studied since their

initia! preparation by Katherine Blodgett in the early 30s. Her experiments, together with

those of Langmuir, resulted into a major discipline in surface science, of which actvances are

reported in numerous review articles and books.5

Moving a clean substrate through a Langmuir monolayer (on a air/water interface)

results in the transfer of the monolayer onto a solid support. Hydrophobic substrates

19

Chapter2

preferentially attraet the tails of surfaetants and the monolayer is transferred during

immersion. On the contrary, po lar substrates have better interactions with the surfactant

headgroups and monolayers are transferred during raising of the plate (see Figure 2.2).

Figure 2.2. Schematic representation ofZ-type LB-jilmformation on a hydrophilic substrate.

The effectiveness of monolayer deposition onto a substrate is given by the transfer ratio TR = AT/As, where AT is the deercase in surface area oecupied by the monolayer at the aqueous

solution-air interface resulting from transfer to the substrate at a constant surface pressure,

and As is the area of the substrate that becomes coated by the monolayer. Deposition is

considered to be ideal for TR 1 ± 0.05. X-ray diffraction measurements, ellipsometry and

AFM teehniques have been successfully used for the charaeterization of LB films. AFM is a

recently developed technique that offers morphological information of the sample surface on

the nanometer scale. Tapping atomie force microscopy has proved to be a very useful tooi to

image soft samples.6 The cantilever oscîllates, in a tapping mode, vertically near its resonance

frequency so that the tip makes contact with the sample surface only briefly in each cycle of

oscillation. This method reduces lateral forces during scanning, preventing sample damage

and making it possible to image the structural details of weakly bound surface layers.

LB films provide a route to precise two-dimensional molecular architecture, and

therefore to advanced nanostructured materials. However, the mechanism of monolayer

transfer to substrates is not yet completely understood. The experimental difficulties

associated with the creation of defect-free, stable, and long-lasting structures in the

dimensions required for device construction are an important drawback. Selective

polymerization of monolayers provides a potentially viabie approach to fulfil these

requirements. Early experiments with surfactants containing polymerizable functionalities

showed the formation of very stabie LB films, though they contained more defects than non-

20

Experimenta/ Techniques

polymerized surfactants.7 The use of pre-polymerized, well defined surfactants based on

dendrimers is lîkely to improve the results in this field8 (see Chapter 5).


Self-assembled monolayers (SAMs) are spontaneously formed upon the immersion of asolid

substrate into an organic salution of surfactants. The simplicity of this method and its

potentiality for scale-up have attracted the attention of many research groups in the last

decade.9 Wettability is an extremely sensitive and simple surface characterization technique

often used to characterize SAMs. A clean hydrophilic surface (i.e. silicon dioxide) is wettable

by apolar solvent such as water and therefore has a low solid-liquid contact angle. Coating

by a surfactant yields a more hydrophobic surface, and hence deercases the wettability.

Amphiphilic molecules containing for instanee -SH or -SiCl3 headgroups form very stabie

SAMson gold and Si02 substrates, respectively.10 Qualitatively, this can be observed by the

incomplete spreading of a droplet of the polar solvent on the substrate and an increase in the

solid-liquid contact angle. Quantitatively, the contact angle, e, is related to the solid-vapour

(Ysv), the liquid-vapour ("(Lv) and the solid-liquid (ysd interfacial tensions by Young's

equation11 (see Figure 2.3) :

COS e = (J',v I',Jf ~v Equation 2.1

Figure 2.3. Schematic representation of a liquid droplet spreading in contact with a solid surface. e is the contact ang/e; 'Y.s"v. the so/id/vapour interfaciaf free energy; rLv, the fiquid/vapour Înterfaciaf free energy; and 'Y.s"L the solid/liquid free energy. The relationship between these parameters for a stationary droplet at thermadynamie equilibrium is described by the Young 's equation (Equation 2.1).

Contact-angle measurements only provide information on the differences of surface tensions

and not on their absolute values. Van der Waals, electrostatic, and other weak interactions are

21

Chapter2

the driving forces for the organization of surfactant molecules in well-packed, ordered

structures on SAMs.


Amphiphiles are molecules that contain a hydrophobic part as well as a hydrophilic part. In

water, the pol ar groups of the molecules are hydrated and in contact with the aqueous phase.

The hydrophobic ebains tend to minimize the contact with water by sticking together and

forming small microdomains in solution. The structure of the aggregates can include spherical

and cylindrical micelles, single or multilamellar vesicles and inverted structures depending on

parameters such as the molecular structure of the amphiphiles, the hydratîon of the

headgroup, interrnolecular interactîons, concentration and the temperature12 (see Figure 2.4).

Figure 2.4. Schematic representation ofvarious jonns of surfactant aggregation.

2.3.1. Critica! aggregation concentration

The driving force for aggregation of amphiphiles in water is often referred to as the

hydrophobic effect, 13•

14 in which a subtie balance between entropie and enthalpie effects is

operative. Despite its importance, but due to this subtlety, a full understanding of the

hydrophobic effect is stilllacldng. Aggregation is accompanied by a decrease in entropy as a

result of smaller translational freedom of the amphiphile in the aggregate. However, the

release of water molecules from the highly ordered hydration mantles around the hydrophobic

part of the molecules compensates for this negative entropy effect. As aresult the net change

in entropy of the system is positive. The aggregation process is somelimes also driven by the

22


favourable change in enthalpy, i.e. the enthalpy of transferring a hydrophobic molecule from

the polar water phase to the hydrophobic environment of the aggregate. But this solvophobic

effect can be partially compensated by the unfavourable enthalpy effect due to loss of

hydrogen honds between water molecules in the hydration sphere of the amphiphile as wel!,

so the net enthalpy contribution is very dependent on the particular system and other system

variables such as temperature. When low molecular mass surfactants are studied as a tunetion

of the concentration, it is observed that micelles or other aggregates exist only above a certain

minimum concentration, the critica! aggregation concentration (cac). The cac can be defined

as the concentration below which only monomeric molecules are in solution and above which

both aggregates and monomeric species can be found in solution. In genera!, arnphiphilic

polymers exhibit the same type of behaviour in solution. However, critica! aggregation

phenomena occur at much lower mol ar concentrations than in low molecular mass surfactants

due to cooperativity and the molar mass of the polymers. Many techniques are available for

the determination of critica! aggregation concentrations. In principle, any physical property

that depends on the partiele size or the number of particles in solution can be used for the

deterrnination of the cac. In genera], breaks or discontinuities in plots of properties such as

surface tension, conductivity, osmotic pressure, interfacial lension or light scattering as a

tunetion of concentration have been used for this purpose. Critica! aggregation concentrations

can also be determined very effectively from the change in speetral characteristics of dye

probes, such as pyrene, added to the surfactant solution.15

Pyrenejluorescencetechnique

In the presence of hydrophobic microdomains in solution (such as micelles or other

aggregates), pyrene is preferentially solubilized within the interior of the aggregates, while

below the cac, pyrene is molecularly dissolved in water. With increasing amphiphile

concentration in the presence of pyrene, there is an increase in the quanturn yield of the

t1uorescence, as well as changes in the vibrational fine structure of the emission spectrum as

the intensities suffer significant perturbations on going from polar tonon-polar solvents.16 At

the very low concentrations of pyrene used (ca. 5·10-7 M), the effect of pyrene on the

aggregation of arnphiphiles is negligible. 17 Furtherrnore, in the excitation spectrum the (0,0)

transition band shifts from 335 to 340 nm as the pyrene goes from water to the hydrophobic

domains. If we represent the intensity of theemission spectra as a tunetion of the arnphiphile

23

Chapter2

concentration, we can directly obtain cac 1•17 From the excitation spectra we can obtain cac2

by representing the ratio l34oll335 vs Jog C. The intensity ratio between the first (Ih with A. =

373 nm) and the third (IJ, with A. = 393 nm) emission peaks is known to correlate well with

solvent polarity.16 The I1 peak, which arises from the (0,0) transition from the lowest excited

electronic state, is a ''symmetry-forbidden" transition that can be enhanced by the distortion

of the 1t-electron cloud. The I3 peak is not forbidden and therefore relatively independent of

the solvent. From the ratio 11/I3 information can be obtained about the polarity of the

microenvironment where the pyrene molecules are located. This value ranges from 1.6 for

water to 0.9 fora polystyrene film and about 0.6 for apolar solvents such as cyclohexane.

2.4. V esicles

Closed bilayer aggregates formed by surfactants represent sophisticated models for biologica!

membranes and have been therefore extensively investigated.18 Swelling of thin surfactant

films in water produces large multilamellar vesicles (MLVs) with diameters between 1 and 8

!J-m. Sonication of MLVs above their phase transition temperature results in the formation of

rather uniform, small unilamellar vesicles (SUVs) with diameters varying from 300 to 600 Á.

Typically, each vesicle contains 8·104-105 surfactant molecules. SUVs can also be prepared

by injecting an organic solution (THF/ethanol) of the surfactant through a small syringe into

water. SUVs can be prepared from pre-polymerized surfactants as well (see Chapters 3 and 5

for vesicles based on dendeitic surfactants).

Vesicles can contain many guest species in their compartments. Hydrophobic

molecules can be distributed among the hydrocarbon bilayer and polar molecules may move

relatively freely in vesicle-entrapped water pools, particularly if they are electrostatically

repelled from the inner surface of the vesicle. Small ions can be easily bound to the

oppositely charged vesicle surfaces, and species with the same charge as the vesicle can be

anchored onto the surface by long hydrocarbon tails. Below their phase transition

temperatures, surfactants are forming highly ordered solid states. Above the phase transition

temperature, the surfactants are separated from each other and change their configurations;

they are in the liquid state. Temperature-induced phase transition is an important property of

vesicles, offering possibilities for permeability control and molecular recognition.

24

Experimental Technigues

2.4.1. Phase transitions

Phospholipid membranes are characterized by well-defined phase transitions.19 These

transitions can be caused by changes in the solvent content of the membrane (lyotropic phase

transitions) or in temperature (thermotropie phase transitions). The possibility of isotherrnal

regulation of these transitions by the use of ionic or other solute interactions shows very

interesting applications in the fields of biologica! systems. Starting at low temperatures, the

lipid phase presents a lamellar crystalline state (lc). In this state, the lipid chains are in an all

trans conforrnation. The increase in temperature causes therrnal and rotational chain

excitations, resulting in an increase of the volume occupied by the ebains and hence, the

packing density decreases. When the so-called subtransition temperature (Ts) is reached, the

bilayer undergoes a transition from a crystalline phase to a lamellar gel or ordered bilayer,

where molecules are less tightly packed. An area increase occurs in the subtransition, more

significantly in the interfaciallipid area than in the hydrocarbon chains.

Lc ~~ltlflf ~ Ts Subtransition Temp.

LP WMM~M •1 ~ T1 Chaîn Melting Temp.

La. \lllllllf Figure 2.5. Schematic representation of the various lamellar bilayer states in phospholipid membranes. Lc is a crystalline phase, L13 is a gel phase and La is the jluid phase. The chains represented on the right are tilted with respect to the bilayer normal.

Increasing the temperature above T., the hydrocarbon ebains start to oscillate faster,

descrihing a cylindrical rotation along their long axes. The lipid ebains are packed on a

hexagonal lattice. As a consequence of this packing, the chain area and the molecular cross

25

Chapter2

section normal to the ebains are not much bigger than that in the crystalline state (about 0.2

nm2 in the crystalline state and 0.4 nm2 in the gel state), compared to the interfacial area per

molecule which increases significantly (0.65 nm2 in phosphatidylcholine). A longitudinal

displacement of each molecule related to its neighbour usually takes places in order to

maintain the close packing when the interfacial area is increased so much. In addition, some

long-range interactions may take place. As a result of all these effects, the membrane adopts a

wavy appearance: an undulating surface with a fairly long periooicity (see Figure 2.5).

A B

;: ;: J/ '*

//

* :

__..

Figure 2.6. Change from an all-trans conformation (A) toa gauche conformation (B) in the alkyl chain of a phospholipid molecule.

Below T., the hydrocarbon ebains are ordered and approximately keep their maximum

extension. Above T., increasing temperatures cause this order to disappear as a consequence

of the chain melting or gel-to-liquid crystalline phase transition, which happens at the chain

melting temperature, T1• There is evidence of the bilayer chain melting transition happening in

a step-wise way.19 First, the hydrocarbon ebains experience trans-gauche isomerizations

(Figure 2.6). The distortions in the chain orientation are transferred from chain to chain and

provoke density fluctuations, which cause increased lipid mobility through lateral expansion

of the alkyl chains. Thus, the whole molecule (hydrocarbon interior and polar-apolar

interface) moves even faster, showing orientational changes and higher-speed rotational

motions. As a consequence, the polar part of the molecule gets in touch with water, whose

molecules are introduced in between the polar heads, thereby increasing the interlamellar

separation. This last step seems to start at the polar heads and is transmitted all over the

whole chain, resulting in a fluid bilayer (Lu).

26


From a thermodynamic point of view, the chain-melting transition occurs when the decrease

in the bilayer cohesive energy, due to lateral expansion and to the energy cost of creating

chain rotational isomers, is compensated by the entropie reduction in free energy due to chain

isomerism.

At the liquid crystalline state, amphiphilic molecules show, on the one hand,

orientational ordering and, on the other hand, rotational motions. The viscosity and

permeability of the bilayer are determined by the orientations and conformations of the

amphiphilic molecules. Fluidity studies of liposomes as membrane models have established

that the most important determinants for the phase transition temperature in lipids are chain

length, degree of unsaturation of the phospholipid alkyl chains, nature of the polar headgroup

and degree of hydration.Z0 The type of bond by which the alkyl ebains are coupled to the

glycerol backbone in phospholipids influences the transition temperature, di-ether derivatives

having higher transition temperatures than the corresponding di-esters. It is well known that

intermolecular hydrogen honds in the membrane can have small effects in the shift of

transition temperature (1-3 °C). In the case of strong interlipid ionic bridges, e.g. by Ca2+ or

some other divalent ions, larger effects are found, with Tt shifts as great as 50 °C.18 a

There are many techniques that have been used to study the membrane fluidity in

different systems, like NMR,21 electron spin resonance (ESR),22 differential scanning

microcalorimetry,23 use of radioactive tracers24 and fluorescence spectroscopy_27•25

•26 The

fluorescence teehniques most used to study the phase transition temperature are fluorescence

depolarization29 and pyrene excirner formation.27 These two techniques have been applied to

several systems, such as phospholipid26·30 membranes, micelles,28 polymerized and non

polymerized vesicles of several amphiphiles28 by means of the introduetion of fluorescent

probes within the membranes. The probes chosen for this purpose are small, often organic

molecules, with very illustrative fluorescence emission properties that strongly depend on the

characteristics of the immediate environment around the fluorophore. This property makes

the use of fluorescence a very useful technique as structural changes in the membranes are

expected to be quite localized.


The orientations and conformations of the amphiphilic molecules determine the structure of

the bilayer, which affects the viscosity and permeability of the membrane in the aggregates.

27

Chapter2

Fluorescence is known to be sensitive to factors such as polarity and viscosity of the

immediate environment of the probe. Hydrophobic fluorescence probes are located in the

bilayer and provide information about its physical properties. Changes in the microviscosity

of the bilayer can be investigated by the fluorescence depolarization technique?9

In the common fluorescence of aromatic compounds, the processes of absorption and

emission of electromagnetic radiation are associated with transition dipoles of a well-defined

orientation on the molecular frame. 29 When polarized exciting light is directed towards a

chromophore whose excitation dipole moment is aligned with the electric dipole moment of

the exciting light, the chromophore will preferentially absorb this light Excitation will take

place in the direction of polarization; absorption oscillators perpendicular to the direction of

incident light will be not excited. Since the absorption process is much faster than molecular

rotation, the use of oriented exciting light creates a population of preferentially oriented

excited fluorophores. Since the emission of a pboton by the excited fluorophore often requires

a much longer time (the lifetime 't, of the excited state) than does absorption, the fluorophore

can reorient before emission occurs. In this case, the emitted pboton will no longer be

polarized parallel to the exciting photon. The resulting polarization of fluorescence is often

defined in terms of the steady-state fluorescence anisotropy.

Light ~ Souree

Excitation Monochromator

High Viscosity

11111

Polarizer 1

LowViscosity

1////

Polarizer 2

Cell

PhotoMultiplier

Emission Monochromator

Figure 2.7. Principle of the depolarization fluorescence technique. Experimental set-up: 1-vertical polarizer for the excitation beam; 2- vertical polarizer for theemission beam. ln low viscosity media, the emission oscillator can change direction during the fluorescence lifetime.

28


The polarization of fluorescence is usually expressed as the anisotropy value, r

Equation 2.2

Where Ivv and Ivh are vertical and horizontal polarizations resulting from excitation with

vertically polarized light, respectively. Ivv + 2Ivh is the total fluorescence emission. The

polarization ratio Iv = Ivv I Ivh• has to be divided by ~ (Ih= Ihv I ~h), which is a correction factor

for polarization artifacts of the photomultiplier and has no physical meaning (Ihv and Ihh are

vertical and horizontal polarization resulting from excitation with horizontally polarized light,

respectively), resulting r = (Iv -Ih)I(Iv + 2Ih).

From equation 2.2, it is clear that the greater the extent of rcorientation of the fluorophore

during the lifetime of its excited state, the smaller will be the observed fluorescence

anisotropy, to the extent that r = 0 for complete fluorophore reorientation (Ivv = Ivh).

Therefore, the polarization of fluorescence can be regarded as a measure of the viscosity of

the medium in which the fluorescent probe is present. If the viscosity of the medium is high,

the orientation of excited fluorescent molecules wiJl hardly change within their lifetime. As a

result the orientation of the absorption and emission oscillators of the probe wiJl not change

very much and the fluorescent light will be polarized. In a less viscous environment, the

fluorescence molecule will rotate within the fluorescence lifetime and the fluorescent light

will be (partly) depolarized (Figure 2.7).

In the case of a fixed chromophore in a frozen system,

r = r0 = (3 coi a-1)1 5 Equation 2.3

where the intrinsic anisotropy (r0) is determined only by the electronic distributions of the

excited and ground states, since these determine the angle a. between the excitation and

emission dipoles of the molecule. From Equation 2.3, we obtain ro= 0.4 as the maximum

anisotropy value for the fluorescence anisotropy (a.= 0°).

29

Chapter2

Figure 2.8. All-trans-1,6-diphenyl-1,3,5-hexatriene (DPH) molecule.

All-trans-I ,6-diphenyl-1 ,3,5-hexatriene (DPH) is the most frequently employed probe for

fluorescence polarization. The molecule is a polyene hydrocarhon with a stabie all-trans

configuration and a rod-like shape. The cis-trans isomerizations of DPH, which are

accompanied by a loss in fluorescence intensity, strongly depend on the viscosity of the

medium. Therefore, in a viscous environment, where cis-trans isomerizations are greatly

inhibited, the fluorescence quanturn yield of DPH approaches 1 and for all practical purposes

the all-trans DPH can he considered as the exclusive configuration that contributes to the

fluorescence.30 The intrinsic characteristics of DPH make this molecule a very suitable dye

for fluorescence depolarization studiesY The transition dipole of the fluorescence and the last

absorption hand lay close to parallel (a 14°) to the long axis of the molecule and the

fluorescence depolarization therefore reflects almost exclusively the angular displacement of

this axis. The intrinsic anisotropy value of DPH in a frozen system is therefore very large (r0

= 0.362) and the molecule has a very long fluorescence decay time ("Co= 1 1.4 ns).

The relation hetween fluorescence depolarization and the hydrodynamic parameters of the

fluorophore is described by the Perrin equation:32

IQ Tr -=1+C(r)-r 11

Equation 2. 4

T is the absolute temperature (in K). C(r) is a parameter which relates to the molecular shape

and the location of the transition dipoles of the rotating fluorophore. This term expresses the

contribution of each mode of rotation of the probe to the recorded fluorescence

depolarization, and it includes only structural parameters of the probe molecule. C(r) can he

experimentally evaluated by measuring the dependenee of rofr on T·'t!TJ in a reference

30


solution.* Lifetime t' and C(r) are essenrial for the determination of the microviscosity Tl in

macroscopie units. The lifetime t' can be determined indirectly from the temperature profile of

the fluorescence intensity I (I = Ivv + 2 Ivh). 31 At low temperatures, dynamic quenching

processes are suppressed and the fluorescence quanturn yield approaches its maximum value

(which equals 1 in the absence of static quenching processes) and the corresponding I reaches

its maximum value at Io. Since I is directly proportional to t', Io should correspond to t'o,

which for DPH is equal to 11.4 ns. A plot of I vs T yields a value for Io from the upper plateau

of the curve, which can be used for derivation of 'C at any given temperature.

Equation 2.5

In aliphatic synthetic22•33 and biologicae4 membranes, the values obtained by this indirect

determination and by direct measurements have been proven to be in very good agreement.

The membrane microviscosity Tl can bc regarded as a mechanical barricr imposed by the alkyl

chains that can control transport processes. It is wc11 described by assuming complete analogy

with a hydrocarbon fluid, which cnables thc application of wcll-known exprcssions for

macroscopie viscosity. Tbc viscosity of lincar hydrocarbon liquids deercases with temperature

in an exponential manner, which agrees well with the empirica] relation

Tl= A. e/J.EfRT Equation 2.6

in which R is the gas constant (8.3145 J·mor1-K1), A is a constant charactcristic of the

* The calibration curve of DPH obtained in American White Oil USP35 is characterized by C(r) valnes between

10-3 and 7.5·10-4 (P ·K-1·ns-1) obtained at high and low r values. This calibration curve is described by the

following empirical equations:

:!...+ ~- 827 = !..!_ for r > 0.16 r' r 1j

~- 1680 = '!:..:!.. for r < 0.16 r 11

31

Chapter2

system related to the entropy of flow activation and àE is the flow activation energy (J·mol

1). In an aliphatic membrane with an invariant phase a plot of In 11 vs lff should yield a

straight line with a slope from which the flow activation energy àE (J·mor1) can be obtained.

A phase transition is indicated by a change in àE which is displayed by a break in the plot of

In 11 vs lff.


A second method to investigate the transition temperature, as well as to estimate the

microviscosity in the hydracarbon bilayer, is the use of pyrene as a probe. Pyrene has been

frequently used to determine the lateral diffusion within biologica! membranes35 and also in

synthetic ones.27 Pyrene has many advantageous properties: it is the most comprehensively

studied of the excimer formation dyes, it has a high quanturn yield, its solubility in water is

extremely low27c and it has a long fluorescence lifetime. In dilute solution, pyrene exhibits a

structured violet fluorescence emission band, with a characteristic transition at 395 nm. As

the molecular concentration of the salution is increased, the molecular fluorescence quanturn

yield decreases and a broad structureless blue fluorescence appears at 480 nm.36 This

emission band is due to the fluorescence of excited dimers, produced by the collisional

interaction of an excited molecule with another pyrene molecule in the ground state:

It has been reported that the pyrene excimer may adopt a sandwich-like structure, with an

interplan ar di stance between the conjugated ringsof approximately 3-4 Á.27

10.7 A

Figure 2.9. Schematic representation of a pyrene excimer and calculated dimensions.

32

Expertmental Techniques

A general kinetic scheme for excimer formation has been proposed:37

Scheme 2.1. Mechanism of pyrene excimer formation via a bimolecular reaction between a pyrene excited molecule and another one in the ground state.

The wavy arrows indicate the processes of fluorescence emission, and the straight arrows

indicate radiationless processes. For the excimer, the rate constants are indicated in italic (kj;

k1), and for monomerk pyrene in regular type (kf, kt). In which k, is the second order rate

constant for the excimer formation; kf and kJ> the transition probabilities for the radiative

decay of the excited monoroer and the excimer, respectively; kt and k~> the probabilities for

the radiationless transitions; and kd is the rate constant which corresponds to the dissociation

of the excimer into the excited monoroerand the ground state monomer.

Several considerations have been taken into account to apply this model for the excimer

forrnation?7a

• The excimer formation in fluid media is a diffusion controlled process.

• At temperatures up to ca. 60 oe the dissociation rate constant, kd, of the pyrene excimer is

small compared to the transition rate litO of the complex into two unexcited molecules

(klto << 1) at low temperatures.

• Every collision between an excited and a ground state molecule is effective and leads to

excimer formation.38

• The diffusion of a molecule can be considered as a random process and the diffusion

coefficient is related to the total number of jumps per second in the diffusing system.

• The second order excimer formation rate constant is proportional to the number of

collisions per second between the diffusing particles, which is related to the diffusion

coefficient of the molecules.

• The ratio of the fluorescence quanturn yields of the excimer and the monoroer are directly

related to thesecondorder rate constant k, by:

33

Chapter2

------ka kr (1 + kd fo )

Equation 2. 7

By combining the previous considerations, the following cxpression has been obtained by

Galla and Sackmann:27"

Equation 2.8

in which, Ddiff is the diffusion coefficient of molecules performing a lateral diffusion parallel

to the membrane surface; 39 À is the length of a diffusional jump; de, the critica] van der W aais

diameter of the diffusing molecule (de ""8 A for pyrene); K = 0.8 = let/Jml lmt/Je, represents the

proportionality coefficient between fluorescence intensity and quanturn yield and it only

depends on the speetral distribution of the monomer and the excimer emission; kf/k1 is

approximately 0.1 and it is an intrinsic property of pyrene, measured in nonane;40 'ZQ

corresponds to the lifetime of the excited dimer; and c to the concentration of the diffusing

particles (molecules/A2), calculated as the number of pyrene moles per alkyl chain divided by

the area per alkyl chain.

The dimensions of a system based upon palmitoyl chain-derivatives are the following:

The length of a diffusional jump (À) is considered to be approximately equal to the distanee

between two palmitoyl chains, assuming maximal bilayer density and cylindrical symmetry

for the palmitoyl chains: À= 8 A. The lengthof a palmitoyl chain has been determined by X

ray measurements as 22 Á and the area is 25 A2•41

22Á

Figure 2.10. Schematic representation of a pyrene molecule in between two palmitoyl chains.

34


After all these considerations, the initia! expression can be reduced to the following ones:

D diff = 0.25 k0

Equation 2.9 Equation 2.10

Ddiff and ka depend on temperature following the Arrhenius law:

Equation 2.11

therefore, the activation energy for the diffusion process can also be estimated.

The Einstein-Smoluchowski theory42 for diffusion-controlled processes relates the second

order rate constant of excimer formation to the viscosity of the medium where the reaction

takes place through the following expression:

s_!!!_(pa) 30001} b

Equation 2.12

pa/b = 1 for pyrene, where p is the probability that each collision between two pyrene

molecules is effective and leads to excimer formation, (p = 1 for pyrene), a is the interaction

radius between two pyrene molecules and b is the Stokes radius for the pyrene molecule; both

radii are considered to be the same in pyrene molecules. 11 is the viscosity of the medium

surrounding the pyrene molecule (microviscosity of the hydracarbon bilayer).

To conclude this introduction, it is important to remark that the calculated values of

microviscosity, diffusion coefficient and activation energy are only estimations, and not

absolute values. The theories of diffusion-controlled processes can predict qualitatively the

rate of such processes as a function of temperature, viscosity, lifetime, etc. However,

quantitative comparison between theory and experiment is rather difficult to obtain because

of the uncertainties in evaluating essential parameters such as the interaction radii, diffusion

coefficients, and the probability per collision that a reaction will take place. It has to be

pointed out as well that for pyrene excimer experiments the assumption has been made that

pyrene excimer formation is a diffusion controlled process. Nevertheless, the collision theory

supposes that the number of molecules inside the hydracarbon bilayer is so high that, most of

the pyrene collisions take place among pyrene molecules within the hydracarbon bilayer.

35

Chapter2

However, at low concentrations of pyrene, the number of molecules colliding with the alkyl

ebains may be significant43 and hence, the formation of the excimer can be hindered, leading

to errors in the determination of thesecondorder rate constant of dimer formation. We must

not forget that the basic features of alkyl chain ordering within bilayer membranes is not an

easy issue to understand. Even when many theoretica! and experimental studies have been

publisbed in the last 20 years, still the high complexity of the system to solve has made

approximate treatment of the problem inevitable.

2.4.2. Turbidity Measurements

There are many techniques that can be used to determine the sizes and geometry of uni- and

multilamellar vesicles and Jiposomes, basedon Rayleigh's scattering Iaw.44'45 Light scattering

methods have been used to determine sizes and shapes of cells, subcellular particles and

phospholipids dispersed in water, and turbidity measurements have been used to study several

biologica! membranes as wel! as lipid vesicles and the gel to liquid crystalline phase

transition in Iipidic liposomes. Turbidity measurements are based on the assumption that

when a sample is irradiated, the transmitted light is decreased if scattering of this radiation by

the particles of the sample occurs.

The total intensity of scattered light may be related to the transmission of a radlation beam

through a dispersion of scattering particles. In a similar way to the Beer-Lamhert's law for

absorption, the transmitted light intensity I can be written as:

Equation 2.13

where the attenuation constant 't is called the turbidity, d is the path-length and A is the

absorbance of the sample, measured with a spectrophotometer. Applying the Rayleigh-Gans

Debye theory of light scattering, 46 for hollow spherical particles47'

48 with radius R * smaller

than ')J20, being À the wavelength in vacuum of the incident beam, the turbidity is expressed

as:

* Sonicated vesicles are on average smal] unilamellar vesicles. A Jinear relationship was found when the

turbidity 1: was pl otted vs I f').. 4, indicating that the vesicle solutions contained mainly unilamellar structures.

36


m = n/n is the relative refractive index, where nv is the refractive index of the partiele and n

that of the medium. N is the number of isotropie particles per unit volume, V the volume of

material in a particle, and Q is the dissipation factor, which a tunetion of the partiele radius

and shell thickness and of the scattering vector, q : (41tll!À) sin( ()12), where e is the angle of

observation measured from the incident beam.

Based on this expression, calculations on the turbidity of a model vesicle can be done. These

calculations were performed for model vesicles with a bilayer thickness of 5 nm and

refractive index49 n0 = 1.42 (n = 1.333 for water, m = 1.065) for incident light with À = 450

nm. The calculations reveal that the turbidity increases with increasing partiele radius (see

Figure 2.11). The turbidity is proportional to the square of the bilayer volume (V2) and

directly proportional to the dissipation factor Q. With increasing size, the bilayer volume

increases but the dissipation factor Q decreases with increasing vesicle radius (see Figure

2.11).

10-"T------------------,

. I • I •

I .

.. "•' ,. ..• -·-....

•······•·· ... .. .. .

10""'+-~..-~r-..---,~--r~-.-~-r---r-....--l 0 W ~ U W 100 lW I~ lU

Yesiele diameter (nm)

Q

0.1

•••• • • . ._ •.. , . '···-.

' ..... ... •• ••• ••••••

0 20 40 60 80 100 120 140 !60

V esicle diameter (nm)

Figure 2.11. Left: logarithmic plot of single partiele turbidity as function of the vesicle radius for il = 450 nm. Right: logarithmic plot of dissipation factor as function of the vesicle radius for il = 450 nm.

37

Chapterl!

From this calculation it becomes evident that the scattering ability of a single scatterer

(looking at the dissipation factor only) decreases as the size increases. Nonetheless, the

overall turbidity increases due to the increase of scattering volume. This results in the Tyndall

effect. For a given volume of the scattering material, NV, dispersed as small particles, the

scattered intensity is directly proportional to the volume of the individual particles; the

coarser they are, the more intense is the scattering.50 This is valid for smal! particles. For

particles that are large compared with the wavelength, the turbidity decreases with the partiele

radius.

Swelling of a vesicle is restricted to a eertaio radius.51 Some calculations were performed on

the model vesicles with restricted swelling. The upper limit of swelling was chosen to a

situation corresponding to a volume increase of 30%. For different values of the initia] radius

of the vesicles, the increase of turbidity was calculated over this range of swelling.

~ -e B ll.)

ü ·.: til

""

7.0xl0·1•

6.0xJ0' 18

5.0xl0.18

4.0xl0. 18

3.0x10.18

2.0x 1 0' 18'+-~-.---.---.~~..--~-,-~--.-~-....--~-.--~-1 0.95 1.00 1.05 1.10 1.15 1.20 1.25 1.30 1.35

Relative Volume

Figure 2.12. lnfluence of swelling on the turbidity of a single vesicle. 0 Initia[ radius = 50 nm; 0 initia! radius = 40 nm.

From this plot it becomes clear that the turbidity is linearly dependent on the volume of the

vesicle when undergoing restricted swelling. Turbidity measurements allow monitoring of

volume changes and can therefore be used to estimate the osmotic behaviour of vesicles.

38


2.5. References

(a) Kallury. K.M.R.; Thompson, M.; Tripp, C.P.; Hair, M.L. Langmuir 1992, 8. 947. (b) Rubînstein, I.;

Steinberg. S.; Tor, Y.; Shanzer, A.; Sagiv, J. Nature 1988, 332, 426. (c) Whitesides, G.M.; Laibinis,

P.E. Langmuir 1990, 6, 87. (d) Eychmüller, A.; Katsikas, L; Weller, H. Langmuir 1990, 6, 1605. (e)

Ishikawa, Y.; Kunitake, T. J. Am. Chem. Soc. 1991, I I3, 1749.

2 (a) Osada, Y.; Ross-Murphy, S.B. Scientific American 1993, 82. (b) Ugelstad, J.; Berge, A.; Ellingsen,

T.; Schrnid, R; Nilsen, T.N.; Mork, p.C.; Stenstad, P.; Homes, E.; Olsvik, 0. Prog. Polym. Sci. 1992,

17, 87. (c) Bedard, RL; Bowes, C.L.; Coomgs, N.; Holmes, A.J.; Jiang, T.; Kirkby, S.J.; Macdonald,

P.M.; Malek, A.M.; Ozin, G.A.; Petrov, S.; Plavac, N.; Rarnik, RA.; Steele, M.R.; Young, D. J. Am

Chem. Soc. 1993,I/5,2300.

3 Manne, S.; Hansma, P.K.; Massie, J.; Elings, V.B.; Gewirth, A.A. Science 1991, I52, 183.

4 Langmuir, I. J. Am. Chem. Soc. 1917, 39, 1848.

5 (a) Blodgett, KB. J. Am. Chem. Soc. 1935, 57, 1007. (b) Ulman, A.; An introduetion to ultrathin

organic films jrom Langmuir-Blodgett to self-assembly. Academie Press, Boston, 1991. (c) Gaines,

G.L. lnsoluble monolayers at liquid-gas inteifaces. Interscience, New York, 1966. (d) Roberts, G.G.

Langmuit-Blodgettfilms. Plenum Press, New York, 1990.

6 (a) Rugar, D.; Hansma, P. Phys. Today 1990, 43, 23. (b) Tamayo, J.; Garcia, R Langmuir 1996, 12,

4430.

7 Rolandi, R.; Paradiso, R.; X u, S.Q.; Palmer, C.; Fendler, J.H. J. Am. Chem. Soc. 1989, 1 I I, 5233.

8 Schenning, A.P.H.J.; Elissen-Román, C.; Weener, J.W.; Baars, M.W.P.L.; van der Gaast, S.J.; Meijer,

E.W. J. Am. Chem. Soc. 1998, 120, 8199.

9 (a) Sagiv, J. J. Am. Chem. Soc. 1980, 102, 92. (b) Bain. C.D.; Whitesides, G.M. Langmuir,1989, 5,

1370.(c) Kim, Y.-T.; McCarley, R.L.; Bard, A.J. J. Phys. Chem. 1992,96, 7416.

10 (a) Sagiv, J. 1. Am. Chem. Soc. 1980, 102, 92. (b) Netzer, L; Sagiv, J. J. Am. Chem. Soc.1983, 105

674. (c) Bain, C.D.; Troughton, E.B.; Tao, Y.T.; Evall, J.; Whitesides, G.M.; Nuzzo, R.G. J. Am.

Chem. Soc. 1989, 1 I I, 321. (d) Bain, C.D.; Whitesides, G.M.; Science 1988, 240, 62. (e) Rubistein, 1.;

Steinberg, S.; Tor, Y.; Shanzer, A.; Sagiv, J. Nature 1988, 332, 426.

11 Whitesides, G.M.; Laibinis, P.E. Langmuir 1990, 6, 87.

12 lsraelachvilli, J.N.; Mitchell, DJ.; Ninham, B.W. J. Chem. Soc. Far. Trans. 2 1976, 72, 1525.

13 (a) Tanford, C. The hydrophobic effect, Wiley lnterscience, New York, 1973. (b) Evans, D.P.; Niham,

B.W. ], Phys. Chem. 1986, 90,226.

14 Blokzijl, W.; Engberts, J.B.F.N. Angew. Chem. Int. Ed. Eng/. 1993,32, 1545.

15 Wilhelm, M.; Zhao, C.L.; Wang, Y.; Xu, R; Winnik, M.A.; Mura, J.L.; Riess, G.; Croucher, M.D.

Macromolecules 1991, 24, 1033.

16 Kalyanasundaram, K.; Thomas, J.K. J. Am. Chem. Soc.l977, 99, 2039.

39

ChapterZ

17 Astafieva, I.; Xong, X.F., Eisenberg, A. Macromolecules 1993, 26, 7339.

18 (a) Fendler, J.H. Membrane mimetic chemistry. Wiley, New York 1982. (b) Fendler, J.H. Science 1984,

223, 888. (c) Kunitake, T. Angew. Chem. Int. Ed. Eng/. 1992, 3I, 709. (d) Ringsdorf, H.; Schlarb, B.;

Venzmer, J. Angew. Chem. Int. Ed. Engl. 1988,27, 113. (e)

19 (a) Larsson, K. Chem. Phys. Lipids 1977, 20, 225. (b) Falkovitz, M.S.; Seu1, M.; Frisch H.L.;

M<.:Connell, H.M. Proc. Natl. Acad. Sci. USA 1982, 79,3918.

20 Cevc, G.; Marsh, D. Phospholipid Bilayers: Physical Principles and Models. Wiley, New York 1987.

21 Ebelhäuser, R Makromol. Chem., Rapid Commun. 1984,5,403.

22 (a) Sackmann, E.; Träuble, H. J. Am. Chem. Soc. 1972, 94, 4483. (b) Devaux, P.; McConnell, H.M. J.

Am. Chem. Soc. 1972, 94,4475. (c) Stier, A.; Sackmann, E. Biochim. Biophys. Acta 1973,311,400.

23 Blandarner, M.J.; Briggs, B.; Cullis, P.M.; Green, J.A.; Waters, M.; Soldi, G.; Engbens, J.B.F.N.;

Hoekstra, D. J. Chem. Soc. Faraday Trans. 1992, 88, 3431.

24 (a) Rigaud, L.; Gary-Bobo, C.M.; Lange, Y. Biochim. Biophys. Acta 1972, 266, 72. (b) Radda, G.K.;

Vanderkooi, J. Biochim. Biophys. Acta 1972,265, 509.

25 Pownall, H.J.; Smith, L.C. J. Am. Chem. Soc. 1973, 95,3136.

26 Shinitzky, M.; Barenholz, Y. Biochim. Biophys. Acta 1978,515, 367.

27 (a) Galla, H.J.; Sackmann, E. Biochim. Biophys. Acta 1974, 339, 103. (b) Soutar, A.K.; Pownall, H.J.;

Hu, A.S.; Smith, L.C. Biochemistry 1974, 13, 2828. (c) Vanderkooi, J.M.; Callis, J.B. Biochemistry

1974, 13, 4000.

28 Roks, M.F.M. Applications of Poly(iminomethylenes) in Bilayer Membranes. Ph.D. Thesis, University

of Utrecht 1987.

29 (a) Feofilov, P.P. The Physical Basis of Polarized Emission. Consultants Bureau, New York 1961. (b)

Steinberg, I.Z. in Biochemica/ Fluoresanee Concepts, Eds. Chen, R.F.; Edelhoch, H. Marcel Dekker,

New York 1975.

30 (a) Cehelnik, E.D.; Cundall, R.B.; Lockwood, J.R.; Palmer, T.F. Chem. Phys. Lett. 1974, 27. 586. (b)

Hudson, B.; Kohier, B. Annu. Rev. Phys. Chem. 1974,25,437.

31 Shinitzky, M.; Barenholz, Y. J. Biol. Chem. 1974, 249, 2652.

32 (a) Perrin, F. J. Phys. Radium 1926. 7, 390. (b) Spencer, R.; Weber, G. J. Chem. Phys. 1970, 52, 1654.

33 Lenzt, R.B.; Barenho1z, Y; Thompson, T.E. Biochemistry 1976, 15,4521.

34 Shattil, S.J.; Cooper, R.A. Biochemist1y 1976, 15, 4832.

35 Galla, H.J.; Luisetti, J. Biochim. Biophys. Acta 1980,596, 108.

36 (a) Birks, J.B. Photophysics of Aromatic Molecules. Wiley-Interscience, London 1970. (b) Birks, J.B.

Organic Molecular Photophysics. Vol2. Wîley-Interscience, London 1975.

37 (a) Förster, T. Angew. Chem. 1969, 81,364. (b) Förster, T.; Kasper, K. Z Phys. Chem. 1954, 1, 275.

38 Birks, J.B; Dyson, D.J.; Munro, J.H. Proc. R. Soc. 1963, A275, 575.

39 In Equation 2.8, k, bas the dimension cm2·Ç

1• Usually second order rate constants are given in units of

L·mo1-1·s-1• A value of 1.0 cm2·s-1 corresponds toa value of 1.34·1014 L·mor1·s-1

.

40


40 Förster, T.; Seidel, P. Z. Phys. Chem. 1965,48, 58.

41 Small, D.M. The physical chemistry of lipids, Plenum Press. New York 1986.

42 (a) Debye, P. Trans. Electrochem. Soc. 1942, 82, 205. (b) Umberger, J.Q.; La Mer, V.K. J. Am Chem.

Soc. 1945,69, 1089. (c) Noyes, R.M. Prog. Reaction Kinetics 1961, 1, 131.

43 Chong, C.S.; Colbow, K. Biochim. Biophys. Acta 1976. 436, 260.

44 (a) Rayleigh, J.W.S. Phil. Mag. 1881, 12, 81. (b) Rayleigh, J.W.S. Proc. R .. Soc. Lond. Ser. A 1914,

90, 219. (c) Kano, K.; Romero, A.; Djermouni. B.; Ache, H.J.; Fend1er, J.H. J. Am. Chem. Soc. 1979,

101,4030.

45 (a) Van de Hulst, H.C. Light Scattering by Small Particles. John Wiley & Sons. London 1957. (b)

Kerker, M. The Scattering of Light. Academie Press, New York 1969.

46 Gans, R. Ann. Physik 1925, 76, 29.'

47 (a) Wang, J.; Hallett, F.R. Applied Opties 1995, 34, 5010. (b) Tinker, 0. Chem. Phys. Lipids 1972, 8,

230. (c) Hallett, F.R.; Watton, J.; Krygsman, P. Biophys. J. 1991,59, 357.

48 (a) Oster, G.; Riley, D.P. Acta Cryst. 1952, 5, 1. (b) Pecora, R.; Aragon, S.R. Chem. Phys. Lipids 1974,

13, I. (b) Wyatt, P.J. in Methods in Microbiology; Eds. Norris, J.R.; Ribbons, D.W. Academie Press,

New York, 1973.

49 Bangham, A.D.; Hili, M.W.; Miller, N.G.A. in Methods in membrane biology vol i. Ed. Kom, E.D.,

Plenum Press, New York, 1974.

50 Kerker, M.lnd. Eng. Chem. 1968, 60, 31.

51 Dominique Hubert, perstmal communications.

41

3 Aggregation Behaviour of Polystyrene-Poly(propylene

imine) Dendrimers

Abstract

'To leam about or;ganized assemblies, and ultimatefy about nature, 011e must for.rake

single-molecule or;ganic chemistry and confront the assemblies themselves despite the

difftculties, complications, and uncertainties that wi/1 undoubtedly enst~e."

F.M.Menger

The amphiphilic behaviour of polystyrene-poly(propylene imine) dendrimer diblock

copolymers, a new class of amphiphilic macromolecules obtained by combining well-defined,

nearly monodisperse polystyrene with poly(propylene imine) dendrimers, has been

investigated. Monolayer measurements showed a generation-dependent amphiphilic

behaviour, whereas transmission electron microscopy showed a distinct effect of amphiphile

molecular structure on aggregation behaviour. Critica[ association concentrations were in

the range of pM as determined with luminescence techniques. The optica{ properties of small

probes located in the aggregates of polystyrene-poly(propylene imine) dendrimer diblock

copolymers provided ample information about the organization of the amphiphiles in the

supramolecular assemblies; the characteristics of these polymerie bilayers resembie a

polystyrene glassy matrix. Polystyrene-poly(propylene imine) dendrimers have much in

common with traditional block copolymers regarding their size and stability, but besides that,

they offer the possibility to tune the aggregate shape and its chemica/ functionality on the

surface as welt.

3.1. Introduetion

The physical behaviour of amphiphilic block copolymers is, despite extensive investigations, 1

far from well understood. This is partly due to the large number of parameters that have a

critica! influence on aggregation. Additionally, polymer systems are not as well-defined with

respect to molecular weight and shape as the traditional surfactants,2 making research in this

Chapter3

field even more difficult. Hence, to be able to investigate the aggregation behaviour of

polymerie systems, it is necessary to generate amphiphilic structures that are well-defined in

molecular weight and in architecture. Dendrimers have gained a growing scientific interest as

building blocks in new molecular architectures.3 The combination of hydrophilicity, highly

branched structure and chemica! functionality makes poly(propylene imine) dendrimers very

interesting building blocks to obtain well-defined amphiphilic superstructures. The properties

of dendrimers offer us not only the possibility to change the size and molecular weight of the

molecule by choosing different generations, but they also allow us to access new molecular

structures and self-assembling systems that can be designed to perform one specific function.4

PS-dendr-(CN)16

PS-dendr-(COOH) 16

Figure 3.1. Schematic representation of Jour different fonctionalities in the headgroup of polystyrene-poly(propylene imine) dendrimer ofthe fourth generation.

44

Aggregation Behaviour of PS-Dendrimers

Amphiphilic properties of dendrimers are among the most explored within the area of

well-defined dendritic structures. Important contributions to this field have been the synthesis

of structures containing dendrimers and linear macromolecules.5 Fréchet's dendritic-linear

block copolymers,6 Chapman's hydraarnphiphiles7 and the amphiphilic polymers as described

by Zhong and Eisenberg8 -the latter can be regarcled as the first approach towards

polystyrene-dendrimer structures with variabie polar head group size--already show the

versatility of the introduetion of dendrimers into amphiphilic molecules. A new type of

amphiphilic block copolymers from the combination of a polystyrene chain and a

poly(propylene imine) dendritic headgroup has been introduced by our laboratories.9 Jan van

Hest prepared in a stepwise synthesis five different generations of amine terminated

dendrimers onto an amine functionalized polystyrene: from PS-dendr-(NHz)z up to PS-dendr

(NH2)32. Their aggregation behaviour was found to be generation dependent, as the theory of

Israelachvili predicts. 10 Two modification reactions were performed on polystyrene

poly(propylene imine) dendrimers (Figure 3.1). Acid hydralysis of the nitrile functionalized

intermediates PS-dendr-(CN)n with n = 2-32 resulted in carboxylic acid functionalized

dendrimers. By methylation of PS-dendr-(NHûn with CH3I, polycationic headgroups were

obtained in which all the primary and tertiary arnines were quatemized. These hybrid

polystyrene-dendrimer block copolymers fill the gap between low molecular weight

surfactants and amphiphilic polymers. Following the initia! studies by van Hest, 9 we describe

in this chapter the arnphiphilic behaviour of these molecules in great detail at the air/water

interface, the generation dependency of their aggregation behaviour and the characteristics of

the aggregates formed in aqueous solutions.

3.2. Monolayers

Surface monolayer studies can provide arnple information about the packing and orientation

of amphiphiles that cannot be obtained otherwise. The defined conditions of the air/water

interface allow the surface activity of the superamphiphiles and the dimensions of the

dendritic headgroups to be explored.


Monolayer experiments have been performed on films of PS-dendr-(+N(CH3)3)n ,with n = l,

45

Chapter3

2, 4, 8, and 16. The behaviour of these surfactants was very similar to the one showed in

previous studies by the PS-dendr-(NH2)n series.11 The monolayers were prepared by

spreading a chloroform solution of amphiphile on a water surface. After evaporation of the

organic solvent, a film of known weight and composition is left on a surface of known area.

The hydrophilic headgroups of the molecules are immersed in the aqueous subphase whereas

the hydrophobic PS part points away from the surface. The isotherms recorded for PS-dendr

(+N(CH3)3)n, with n = I, 2 and 4, displayed a sharp increase of surface pressure upon

compression, indicating the formation of a condensed film before collapse took place (Figure

3.2). The headgroup of these molecules is smal! in comparison with the polystyrene chain and

the isotherm is probably dominated by steric interactions between polystyrene chains on the

surface. In early experiments on PS-dendr-(NH2)n, with n 1, 2 and 4, similar behaviour has

been observed with a Brewster Angle Microscope.11 Isotherms of PS-dendr-(NH2)n were not

reversible since layers of polystyrene chains slide over each other after collapse took place,

forming a very thick film that broke up when decompression was applied.

60

~ 50 s -z s 40 '-"

~ ::I 30 "' "' ~ ~ 20

10

0

0 100

'\

200

\ \

300 400

Area (Á2/molecule)

500 600 700

Figure 3.2. Gompression isotherms at 20 oe of PS-dendr-(+ N(CH3)J)", ( - - ) n = 1; (--) n = 2: (- - -) n = 4; ( -------) n:::: 8; ( ··· --···-) n = 16.

46


The compression isotherms of compounds PS-dendr-(+N(CH3h)s and PS-dendr

[N(CH3h)16 showed the formation of a very stabie film before collapse took place at 210 and

290 Á 2 per molecule, respectively. At the third and fourth generation, the headgroups are

large enough to interact with the water subphase and there is an increase in the area per

molecule on the corresponding isotherms. In this case, the molecules are probably oriented

perpendicularly to the air/water interface. The form of the isotherms revealed the formation of

a stable, compressible monolayer. It has been shown before that dendrimers are flexible

structures, which can alter shape in response to increased surface pressure. 12

Extrapolation of the steepest part of the curve to zero pressure yields the area per molecule A0

at vanishing surface pressure. The measured values amounted to 308 Á2 for PS-dendr

(+N(CH3h)s and 458 N for PS-dendr-(+N(CH3)3)16, which correspond to a headgroup

diameter of 20 and 24 Á, respectively (see Figure 3.3). These diameter values are of the same

order of magnitude as the hydrodynamic radius of DAB-dendr-(NHûn dendrimers, with n = 16 and 32, respectively, obtained from SANS measurements as well as from molecular

simulations in the gas phase.13 This agreement supports the assumption that the dendritic part

of the amphiphiles at Ao is in a fully extended conformation due to a strong interaction with

the water subphase and to electrostatic repulsions due to the polycationic nature of the

headgroup and the consequent high local concentration of positive charges.

Dendrimer diameter

DAB·dendr·(NH2) 16 r= 9.5 A

DAB-dendr·(NH 2)a2 r= 12.9 A

PS·dendr·(+N(CH s}3)8 r=9.9A

PS·dendr-(+N(CHs}(!)16 r= 12.1 A

Figure 3.3. Schematic representation of the molecular dimensions of DAB-dendr-(NH2)" and PSdendr-(N(CH3h)" molecules. IJ we assume that the cross-section of the dendrimer part of the amphiphiles is a circle in a Langmuir monolayer, diameters of 20 A and 24 A are found for PSdendr-(N(CH3h)11, with n = 8 and 16, respectively.

47

Chapter3


Due to their high degree of branching, dendrimers represent a partienlar class of polymers

with an overall globular molecular structure. The size and well-defined surface of the

dendrimers provide them with properties of colloidal particles, like a high wettability.14 The

interaction of the dendrimer block with a specific surface depends on the flexibility of the

branches and on the chemica) nature of the substrate used.15 Due to this particular cohesive

property of dendrimers, self-assembled monolayers (SAMs) could be obtained by dipping of

hydrophilic substrates, such as silicon and glass wafers, in chloroform solutions of PS-dendr

(COQH)0, with n = 4, 8 and 32. The wetting behaviour of polystyrene-poly(propylene imine)

dendrimers and the film properties have been characterized by atomie force microscopy

(AFM) and ellipsometry. Smooth films were obtained with a low surface roughness and

heterogeneity (Figure 3.4). The film thickness was approximately 2 nm as calculated by

ellipsometry for PS-dendr-(COOH)0 , with n = 4, 8 and 32; these results indicate that the

polymerie films are monolayers. The dipping technique allows the polymers to interact with

the hydrophilic substrate and to reach an equilibrium on the surface. Attractive forces

between the polar dendrimer block and the hydrophilic silicon oxide surface of the substrate

yield an optimum wetting of the surface. The dendritic part of the polymer interacts with the

substrate and the PS ebains are exposed to the air, making the surface highly hydrophobic.

350nrn

0

0 350nrn

Figure 3.4. AFM picture of a self-assembled monolayer of PS-dendr-(COOHh prepared by dipping ofthe siliconwaferin a chloroform salution ofpolymer. The surface was completely covered in this sample. Phase contrast is shown in this picture in the non-contact tapping mode, which gives a AFM image of the surface topology.

48

Aggregation Behaviour of PS~Dendrimers

When a water droplet was deposited onto a coated glass substrate, it was clearly observed that

the droplet has a higher surface tension than in the case of a non-coated glass and the surface

contact angle changed from approximately 30" to ca. 50° (Figure 3.5). These results indicate

that amphiphiles based on dendrimers are very suitable matcrials for the formation of well

defined smooth layers and coating of hydrophilic surfaces.

Figure 3.5. Pictures of a water droplet on a glass substrate. a) and c) Non-coated glass before dipping in polymer solution. b) and d) Coated glass after dipping 15 min in a chloroform salution of PS-dendr-( COOHh


The amphiphilic and aggregation behaviour of polystyrene-poly(propylene imine) dendrimers

was studied with different techniques: transmission electron microscopy (TEM) was used to

examine the aggregates formed by the different generations in aqueous solutions, while X-ray

diffraction was performed on cast films of vesicle solutions to determine the bilayer

thickness; the critica! association concentrations were detected with the pyrene-probe

luminescence technique.

3.3.1. Transmission electron microscopy

Aggregation of PS-dendr-(NH2)0 , PS-dendr-(COOH)n and PS-dendr-(+N(CH3)3)n. with n = 8,

49

Chapter3

16, and 32, in water was studied with transmission electron microscopy by using freeze

fracture and Pt shadowing techniques. In case of the third and fourth generations, with n = 8

and 16, vesicle formation was found; the vesicle diameters varied between 15 and 125 nm

with a mean vesicle diameter between 20-30 nm in all cases (Figures 3.6, 3.7 and 3.8).

Figure 3.6. TEMpicturesof an aqueous solution of PS-dendr-(COOH)a: a) Freeze Fracture; b) Pt shadowing.

u w ~ ~ w ~ m Yesiele Diameter {nm)

Figure 3.7. Vesicle size distribution diagram of aqueous solutions of PS-dendr-(COOH)a. Modal value: 25 nm, mean vesicle diameter: 24 nm, polydispersity: 1.2 (population: 200)

0 20 40 60 80 lOO lW I~

Yesiele Diameter (nm)

Figure 3.8. Pt shadowing TEM picture of an aqueous solution of PS-dendr-tN(CH3) 3)J6 and size distributton diagram. Modal value: 25 nm, mean vesicle diameter: 27 nm, polydispersity: 1.2 (population: 200).

50


The observed diameters indicate that the vesicles formed are small and probably of

unilamellar constitution principally. It is weiJ known that the physical characteristics of the

vesicles produced depend strongly on preparation procedure.16 During the study of vesicles of

PS-dendr-(NH2)8 it was observed that sonication of the solutions produced a higher

population of small vesicles (sec Figure 3.9).

W ~ W W ~ ~ l~ ~ lW D m Yesiele Diameter (nm) Yesiele Diameter (nm)

Figure 3.9. Size distribution diagram of an aqueous solution of vesicles of PS-dendr-(NH2)8: a) befare sonication. Modal value: 70 nm, mean vesicle diameter: 58 nm, polydispersity: 1.2 (population: 200); b) after 1 hour sonication. Modal value: 30 nm, mean vesicle diameter: 48 nm, polydispersity: 1.2 (population: 200).

However, the possibility that the very small spherical aggregates observed (diameter ca. 15

nm) consist of micellar structures instead of vesicles can not be disregarded. The presence of

micellar aggregates in coexistence with vesicular structures has been observed before for

systems with mixed surfactants. 17 In genera!, the shape of the aggregates depends on the

intermolecular forces and on the geometrie packing properties (steric interactions) of the

molecules. The packing properties of simple molecules, such as small molecular weight

surfactants, are reasonably easy to model. But in the case of large amphiphilic molecules,

such as block copolymers, other important parameters Iike hydrophilic head-group volume

and compressibility, hydrapbobic tail length and volume and the interactions that affect these

values (e.g. degree of protonation of the head-group) can not be disregarded, increasing the

complexity of the analysis of the packing of block copolymers in solution.

PS-dendr-(COOH)32 showed spherical micelles with diameters between I 0 and 20 nm

(Figure 3.10), which are in agreement with the dimensions expected for two molecules of

polystyrene-dendrimer. It is noteworthy that the micelles did not disintegrate when the

51

Chapter3

solvent was evaporated, indicating a high stability of the aggregates, primarily due to their

polymerie nature.

Figure 3.10. TEM picture of spherical micelles of PS-dendr-(COOH)32 with Pt shadowing.

The aggregates formed by PS-dendr-(+N(CH3)3)n and PS-dendr-(COOH)n remained stabie in

solution for at least one month when the concentration of the solution was not higher than ca.

w-5 M; for higher concentrations the formation of clusters did finally lead to partial

precipitation. The stability of the aggregates is remarkable: vesicle and micellar solutions

remained stabie and they did not collapse by sample treatment, which includes evaporation of

the solvent. Even spherical micelles could be made visible with the TEM-techniques used. 18

3.3.2. X-Ray diffraction

X-Ray diffraction has been carried out on cast films of vesicle solutions of PS-dendr-(NH2)n,

PS-dendr-(COOH)n and PS-dendr-(+N(CH3)3)n, with n = 8 and 16, on a silicon plate. By slow

evaporation of the solvent, the vesicles dried on the solid substrate forming very thin lamellae

that presumably have the same thickness as the vesicle bilayer. These measurements give us

valuable information about the organization of the molecules in the absence of solvent and a

good approximation of the bilayer dimensions.

In Figure 3.1 1, the diffraction curve obtained fora cast film of PS-dendr-(N(CH3)3)16

is depicted. A first order peak at 29 = 0.75°, which corresponds to a bilayer thickness of

ca. 12 nm was found. No clear higher order peaks could be obtained due to the very low

amount of material on the substrate and probably also because of the low long-range ordering

in the sample.

52

.. .


0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

29

Figure 3.11. X- ray diffraction peak of cast films of PS -dendr-t N( CH3)3) 16; relative humidity = 20%, bilayer thickness ca. 12 nm.

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

29

Figure 3.12. X-ray diffraction peak of cast films of PS-dendr-(COOH)8• -0- relative humidity =

90%, bilayer thickness ca. 14 nm; -•- relative humidity = 20%, bilayer thickness ca. 12 nm.

The diffraction pattems of the polystyrene-acid functionalized dendrimer diblocks display a

behaviour similar to the previous case. A bilayer thickness of ca. 12 nm at a relative humidity

of the air of 20% was found as well (see Figure 3.12). The relative humidity of the air was

53

Chapter3

varied between 90% and 0%, to sec whether the hydration layer of the dendrimer headgroup

could introducesome changes in the layer thickness, but since the signals were very broad, no

accurate quantitative analysis of the results was possible. However, a clear shift of the first

order peak to smalter e values seems to indicate that swelling of the layers took place at high

humidity percentages.

Interestingly, when the cast films of PS-dendr-(NHûn. with n = 8 and 16, were

examined, no signals of microscopie ordering were found at all in a wide range of diffraction

angles (see Figure 3.13). By changing the relative humidity in the air between 0% and 90%,

no changes were observed either, indicating that afterwards absorbed water is not able to

introduce ordering in the samples, even after 24 h under an atmosphere at RH = 90%. These

unexpeeted results suggest that when the solvation layers are removed from the dendrimer

part, this block becomes miscible with the polystyrene part and the layered microstructure

disappears. This explanation was confirmed later when the behaviour in solid state of

polystyrene-poly(propylene imine) dendrimers was investigated (see Chapter 4).

1 '

,--._

::i cd '-' l.l s::: :::1 • 0 u ..

• •

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

29

Figure 3.13. X-ray diffraction curve of cast films of PS-dendr-(NH2)16. Relative humidity 20%.

3.3.3. Critica! aggregation concentration

Many techniques can be used for the determination of critica! association concentrations

(cac's), however notall of them are sensitive enough to detect the on set of aggregation if this

occurs at very low concentrations. Since the cac's of block copolymers are usually much

54


lower than those of low molecular mass surfactants,19 we used pyrene as a fluorescent probe

and calculated the effective cac's from the changes in the speetral characteristics of pyrene20

as function of surfactant concentration (cf. Chapter 2).

300 320 340 360 360 380 400 420 440 460

À(nm) À(nm)

Figure 3.14. Excitation ( left) and emission (right) fluorescence spectra of pyrene in the presence of PS-dendr-(COOH)8• The polymerconcentration rangesfram 2·la-" M to 6·10-5 M.

If we represent the intensity of the emission spectra as a function of the block

copolymer concentration, we can directly obtain cac1?1 From the excitation spectra we can

obtain cac2 by repcesenting the ratio l34ollm vs log C. The intensity ratio between the first (l1

at /... = 373 nm) and the third (13 at /... = 393 nm) emission peaks is known to correlate wel!

with solvent polarity?0 This value ranges from 1.6 for water to 0.9 fora polystyrene film and

about 0.6 for nonpolar solvents such as cyclohexane.

The experimental results for polystyrene-poly(propylene imine) dendrimers, with n

8 and 16, obtained from excitation and emission spectra, are shown in Figure 3.15. The cac

values are represented in Table 3.1. No cac's could be measured for the lower generations

because these products are insoluble in water. Only the higher generations are soluble in

water, since small dendritic headgroups are not able to compensate for the low solubility of

the polystyrene chain in water.

55

ChapterJ

1.6 1.6

a 2.0 b 1.5

1.4 1.4 1.5

1/13 J.o'"'Jlm 1.2 1.0 1.2

1.0 0.5 1.0 .5

0.0 -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4

log C (M) JogC(M)

1.5 1.6 IA

I /1 1.2 1.0 1.4 lyJI'm 1 3

1.0 1.2 0.5

5

0.8

-8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4 JogC (M) logC(M)

1.4 f 2.0

e 2.0 • • • 0

1.4 • 1.5 1.2 1.5

11!1, 1.2 JyJI3JS

1.0 1.0

1.0 0.5 1.0 .5

0

0.0 -9 -8 -7 -6 -5 -4 -9 -8 -7 -6 -5 -4

log C (M) log C (M)

Figure 3.15. Plots offluorescence intensity ratio (o) /34r/1335 (right axis) and (•) 1/13 (left axis) from excitation and emission spectra of pyrene, respectively in aqueous solutions of a) PS-dendr-(NH2)B; b) PS-dendr-(NH2) 16; c) PS-dendr-(COOH)8; d) PS-dendr-(COOH)16; e) PS-dendr-(N(CH3h)8; f) PS-dendr-(N(CH3)3)16·

56


Table 3.1. Cac values of polystyrene-poly(propylene imine) dendrimers expressed in J.Imol/l.

from emission spectra (cac1) from excitation spectra (cac2)

PS-dendr-(NH2)8 1.10 0.40

PS-dendr-(NH2)16 0.48 0.72

PS-dendr-(COOH)8 3.10 1.80

PS-dendr-(COOH)16 0.48 0.44

PS-dendr-(+N(CH3)3)gl1s 2.10 0.36

PS-dendr-c+N(CH3h)16I31 1.50 0.60

No significant differences were found between the cac's for different generations and

functionalities, as the length of the hydrophobic chain remains the same for all the block

copolymers. It has been pointed out before21'

22 that the onset of micellation in amphiphilic

block copolymers is mainly determined by the nature and the length of the hydrophobic

block. Also for most low molecular mass surfactants in aqueous solutions, the free energy of

micellation is proportional to the lengthof the alkyl chain.23

3.4. V eside properties

Vesicles are closed bilayers that contain a small amount of water in their interior. Vesicles

formed by polystyrene-poly(propylene imine) dendrimers are expected to have different

properties than vesicles formed by classica! surfactants, such as DODAB (didodecyl dimethyl

ammonium bromide). To study the characteristics of these "polymerie" vesicles, several

techniques have been used. Micro-DSC, excimer formation, fluorescence depolarization and

ESR were performed to investigate the properties of the bilayer of the vesicles formed by PS

dendr-(NH2h·

3.4.1. Bilayer fluidity

The phase transition temperature T1 of vesicles is a very important parameter, because

properties of the bilayer like stability and membrane microviscosity undergo drastic changes

upon going from the solid, gel-like state to the fluid, liquid-crystalline-like state. The most

57

Chapter3

direct way to determine a phase transition is by using differential scanning calorimetry (DSC).

Indirect methods involve the utilization of fluorescent probes localized within the apolar

chain of the bilayer that can provide important information about their surrounding

environment.

Microcalorimetry

Differential scanning calorimetry is the most effective method to find transitions in solid

samples that involve heat exchange; in the case of solutes at low concentrations, a more

sensitive technique is required to detect phase transitions. Microcalorimetry was developed

for enhanced sensitivity. This technique makes use of a calorimeter fitted with a

nanovoltmeter that permits us to visualize very smal! transitions and to operate with smal!

amounts of sample. 24

The Tg of solid polystyrene-poly(propylene imine) dendrimers in the bulk was found

at ca. 77 oe by DSC measurements.9 However, the Tg of polystyrene in such a different

environment and packing as in the bilayer of the vesicle is not known. In an attempt to

determine the Tg or to detect any phase transition in the bilayers of the aggregates, DSC was

measured of an aqueous solutions of PS-dendr-(NH2) 8 and PS-dendr-(COOH)8• However, no

transition was detected between 5 and 95 °C. To makesure that this absence of signa! was not

due to a Jack of sensitivity of the DSC device, microcalorimetrie experiments were performed

on the same samples. Again no signa! was obtained in a temperature range 5-95 oe. To

confirrn the absence of transitions in the bilayers forrned by polystyrene, other methods were

applied. These techniques involve the use of molecular probes, such as diphenylhexatriene

(DPH) and pyrene.


As it has been indicated befare (see Chapter 2), the formation of pyrene excimer within the

bilayer is a technique frequently used to obtain inforrnation about the microviscosity of apolar

environments. For PS-dendr-(NH2) 8 vesicles no pyrene excimer formation was observed even

using pyrene concentrations up to 10-3 M (!),even at temperatures up to 80 oe (Figure 3.16).

These results indicate that the viscosity in the bilayer is very high and pyrene molecules are

not able to diffuse to give excimer formation. The rigidity of the bilayer did not change with

58


temperature, since the polystyrene ebains maintained the same glassy state between 20 and

80°C.

140

120

100

~ 80 :;i o:i

60 ~

E - 40

20

0 400 500 600

A (nm)

Figure 3.16. Fluorescence emission spectra of pyrene in PS-dendr-(NH2) 8 vesicles. [Pyrene] = f(T3

M. Pyrene I PS-dendr-(NH2) 8 molar ratio = 20; T"' 80 "C.

The addition of a good solvent for the polystyrene part, such as toluene, did show a

remarkable effect in the formation of pyrene excimer (sec Figure 3.17). The organic solvent

molecules penetrate in the bilayer and act as a "local solvent" diluting the polystyrene chains

and allowing the fluorescent probe to move more freely. At tol u ene concentrations higher

than 4.7·10-5 M ([PS-dendr-(NH2)s]/[toluene] = 1), phase separation was observed and

toluene was probably partially dispersed in the water phase.

c

360 400 440 480 520 560 600

A (nm)

Figure 3.17. Normalized fluorescence emission spectra of pyrene in PS-dendr-(NH2JB vesicles. [Pyrene} = 2·10-5 M. Pyrene I PS-dendr-(NH2)s molar ratio= 0.4; T"' 25 "C. a) No tofuene added; b) PS-dendr-(NH2)s I tofuene molar ratio= 5; c) PS-dendr-(NH2)sl toluene molar ratio 1.

59

Chapter3


Fluorescence depolarization is a very useful technique to measure the rotatien of a molecule

located within the bilayer of a vesicle (see ehapter 2). When diphenylhexatriene (DPH) was

solubilized in an aqueous solution of PS-dendr-(NH2)8 , a large anisotropy value (r = 0.25)

was found, reflecting a high microviscosity in the bilayer. The results found by changing the

temperature were consistent with the previously described experiments. No change in the

anisotropy value was found in a temperature range from 5 to 90 oe, indicatîng again that the

probe was located in a highly rigid, glass-like environment that did not undergo any changes

with temperature.

For known anisotropy and lifetime values, the microviscosity 11 can be calculated

using empirica) equations obtained by calibration with a solvent of known viscosity in a wide

range of temperatures (see ehapter 2). An average microviscosity value of 6.4 poise was

found for the bilayer of PS-dendr-(NH2)8• However, it is important to remark that the results

were difficult to reproduce and different samples gave different r values, although the rvalues

did not change in any case with temperatures between 5 and 90 oe.

The effect of addition of an organic solvent into the bilayer has been described before.

This phenomenon could be observed with fluorescence anisotropy as well by the addition of

small volume fractions of toluene. The anisotropy value for DPH in a PS-dendr-(NH2) 8

solution at 20 oe was r = 0.24, but when 0.1% toluene in volume was added ([PS-dendr

(NH2)n]/[toluene] ""5) the r value deercases to 0.

3.4.2. Bilayer structure

The orientation and conformation of the amphiphilic molecules determine the structure of the

bilayer, which affects the viscosity and permeability of the vesicular membrane. It was

concluded from the experiments described before (vide supra, paragraph 3 .4.1) that the

bilayer possesses a very high rigidity and microviscosity and that the chains do not show any

phase transition in the temperature range between 5 and 95 oe. However, these experiments

do not provide information about the conformation of the polystyrene chain within the

bilayer. To obtain more information about the organization of the surfactants in the

membrane, ESR anisotropy studies were performed in cast films of polystyrene

poly(propylene imine) dendrimer vesicles, using porphyrin molecules as probe.25

60


CHs CH,

CHs

1

Figure 3.18. 1: 5,10,15,20-Tetrakis(4-(methyl)phenyl)porphyrin. 2: 5,10,15,20-Tetrakis(4-( polystyrene )phenyl)porphy rin.

Two different molecules were used: an apolar porphyrin ring (1), which is expected to

be located into the apolar bilayer, and a porphyrin ring substituted with polystyrene chains of

the same composition and length as the polystyrene chains of the diblock copolymers in the

bilayer (2). The latter one is expected to be anchored to the bîlayer and to be located in the

centre of the rnernbrane. The ethanolffHF injection rnethod was used toprepare the samples

and they were all sonicated. This method has proved to be very effective for the incorporation

of porphyrins into vesicles.25•26 Electron rnicrographs of the resulting solutions showed that

the incorporation of porphyrins did not affect the forrn of the vesicles. A combination of UV

vis, fluorescence and ESR rneasurements has been used to study the spatial distribution of

these two molecules into the vesicle bilayers.

Distribution of porphyrins in the bilayer

The UV -vis absorption spectra of 1 and 2 both showed a maximurn at 424 nrn. Upon

changing the porphyrin concentratien in the bilayer from R = 7500 to R = 100 (R =

molecular ratio polystyrene-poly(propylene irnine) dendrimer/porphyrin), the Àmax of the UV

vis spectra did not change, indicating that formation of aggregates in the bilayer did not take

place even at high concentrations. Also, no shift of the Àmax was observed in the fluorescence

ernission. The fluorescence intensity is expected to deercase drastically when the porphyrins

are forming aggregates, as a consequence of self-quenching?7 In DODAB vesicles the

fluorescence ernission was totally quenched at R = 250 (Iflo = 0.15). In this case, no

considerable self-quenching was detected in a range R = 4000-100. These results indicate

61

Chapter 3

that the porphyrins are isolated in the bilayers of PS-dendr-(NH2)8 vesicles and possesss a

limited mobility.

The use of external quenchers can give us information about the location of

porphyrins within the bilayers. Quenching of the fluorescence of porphyrins in the PS-dendr

(NH2)s vesicles was studied with iodide ions. The magnitude of the interaction quencher

porphyrin is given by the Stem-Volmer quenching constant. Linear Stern-Volmerplots were

obtained up to a quencher concentration of 10-4 M (see Figure 3.19).

14

1.3

1.2

10 11 1.1

1.0

0.9 0 5xl0 -5 lxiO 4 lxiO ·4 2xl0 4

u-J (M)

Figure 3.19. Stern-Volmerplot of the fluorescence intensity ratio It/1 vs quencher concentration for porphyrins 1 (•) and 2 (0).

The fluorescence in organic solution of the compounds 1 and 2 is strongly quenched by 1

ions. The Stem-Volmer quenching constant in organic solution CHCh/CH30H (2:1) is

higher for 1 (K.v = 2370 M-1) than for the more sterically hindered compound 2 (Ksv = 1760

M-1). Surprisingly, in vesicle solutions at R = 500 both porphyrins have quenching constants

in the same order of magnitude, indicating that both porphyrins can be easily quenched by

iodide in the bilayers. Probably, the polycationic nature of the surfactafit headgroup due to

protonation of the primary amine endgroups in water induces the iodide anions to adsorb to

the surface of the vesicle in a much higher concentration than in the case of traditional

surfactants. In that case, and since the hydrophobic chains of the copolymer are not very

densely packed, it is feasible to think that the iodide ions are able to penetrate the rnembrane

and interact with the porphyrins.

62


Orientation ofporphyrins in the bilayer

The orientation of the porphyrins in the bilayer was investigated with ESR spectroscopy.

Therefore, the porphyrins were complexed with capper (II) ions.

Ct;,

3

Figure 3.20: 3: Copper 5,10,15,20-tetrakis(4-(methyl)phenyl)porphyrin. 4: Copper 5,10,15,20-tetrakis(4-(polystyrene)phenyl)porphyrin.

Following the sameprocedure as for the preparation of films for XRD measurements (see this

chapter, 3.2), solid films of PS-dendr-(COOH)n and PS-dendr-(~(CH3)3)n vesicles

containing porphyrins 3 or 4 were obtained on a synthetic film (Mylar film)?8 From XRD

experiments, it was known that these cast bilayers form lamellar structures with their planes

oriented parallel to the substrate surface. Based on the coordinate system described by van

Esch et al.,29 the orientation of the moleculescan bedescribed by studying the contributions

to the measured spectra of the gn and g.l tensors with the help of computer simulations. •

ESR spectra of 3 and 4 in PS-dendr-(COOH)s and PS-dendr-~(CH3)3)s cast films

were measured at angles a = 0" and a = 90" between the bilayer normal and the magnetic

field (see Figure 3.21). In all cases, contributions of both tensors gil and g.l are present in

equal quantities, similar to spectra obtained from powder samples, where no predominant

orientation of the porphyrins in a particular direction is present. From these results we can

* The square-planar synunetry of the porphyrin ligand system around the copper (11) ion gives rise to a

strong anisotropy of both the g tensor and the copper hyperfine splitting tensor A. The tensors in the

directions x and y are degenerate (g, = gy; A,= Ay) and they are labelled g~ and A1., respectively. The

g, and A, tensors are written as gn and A1. The g11

tensor is split into four lines with A11

"' 200 G. The

tensor g1. is visible as a single line, since A1. 20 Gis generally not resolved.

63

Chapter3

conclude that the porphyrins do not have a preferentlal orientation on the bilayers, resulting in

an isotropie system.

a

b

2500 3000 3500 4000

(G)

Figure 3.21. ESR ~pectra of 3 in a cast film of PS-dendr-(COOH)8 (R = 40) with (a) the magnetic field perpendicular to the bilayer normal, a= 90" and (b) the magnetic field parallel to the bilayer normal, a = 0".

Interestingly, the hyperfine splitting due to coupling of the copper nucleus with the nitrogen

ligands was clearly visible in all measurements. This is only possible if the porphyrins are

magnetically diluted (distance between copper nuclei> 20 Á).3° Kunitake et al. showed that

in the case of anionic porphyrins dispersed in bilayers of cationic amphiphiles, it was possible

to obtain non-aggregated porphyrins.29b They were able to observe the monomeric copper

ESR spectrum with resolved nitrogen hyperfine splitting. The tendency of porphyrins to form

aggregates in bilayers is a well known phenomenon. Only in the case of sterically hindered

porphyrins or porphyrins that have strong electrostatic interactions with the amphiphiles

aggregation can be prevented. The presence of monomeric species is only possible for very

low concentmtions of porphyrin in the bilayer (R > 2000), while in our case the concentration

of porphyrin in the bilayer was considerably higher (R = 40). Since no interactions between

the apolar porphyrins 3 and 4 and the polar dendritic headgroups are expected to take place,

presumably steric interactions account for the absence of aggregate formation. The bilayer of

the vesicles possesses a very high rigidity and microviscosity in a very broad range of

64


temperatures, which hinders the diffusion of small hydrophobic molecules in the membrane.

However, 1t-1t interactions between the aromatic units of the polystyrene chain and the probes

can play a role as well.

Taking into account that no phase transition was found for polystyrene

poly(propylene imine) dendrimers, that no evidence has been found of orientation of the

porphyrins in the bilayer, and that no aggregation of porphyrins takes place even at very high

concentrations in the membrane, we can conclude that the porphyrin molecules are probably

entrapped in a bilayer that very much resembles a glassy polystyrene matrix. The atactic

polystyrene chains are probably not able to form a densely packed, crystalline structure and,

therefore, the characteristics of these polymerie bilayers resembie much more a polystyrene

glassy matrix than a traditional lipidic membrane.

3.5. Metal cornplexes

Polymers containing metal-coordinating groups are interesting because they can be used to

obtain metal-containing particles of nanoscopic dimensions.31 Recently, Tonny Bosman in

our group has reported on multiple complexation of Cu(ll), Zn(ll), and Ni(ll) with

poly(propylene imine) dendrimers as multi(tridentate) ligands.32 Following an analogous

procedure, Cu(ll) roetal ions could be complexated using the primary amines of PS-dendr

(NH2)g as ligands. The titration of PS-dendr-(NH2)g in CH30H/CH2Cb (2: I v/v) with CuC]z

was foliowed by UV-vis and ESR spectroscopy.

0.14

0.12 • 0.10

0.08

"' .r;, <o.oe

0.04

0.02

• 0.00

0

equivalents Cu(II)

Figure 3.22. Tirration curvefor PS-dendr-(NH2)8 with CuCl2 in CH30HICHCl3 (2:1 vlv).

65

Chapter3

UV -vis spectroscopy showed a broad band at 700 nm, indicative of the presence of the

complex, and typical for a five-coordinated Cu(IT) species. The titration curve is represented

in Figure 3.22 and reveals the formation of 1 :2 complexes, resulting in molecules with

3.5 ± 0.5 copper i ons per dendrimer with 8 amine endgroups.

The titration of PS-dendr-(NHz)8 with CuCb was followed with ESR spectroscopy as

wel!. Characterization of the copper complex es with ESR spectroscopy showed a single broad

peak at 3200 G, indicative of the presence of more than one copper ion per molecule.

Increasing the copper concentration resulted in the appearance of the ESR signa] of free

copper.32 These results are in full agreement with the UV-vis measurements, indicating the

formation of a 1:2 complex.

Aqueous dispersions of [Cu4 PS-dendr-(NH2)8]Cls were slightly opalescent and the

aggregates remained stabie in solution for very long periods of time (months). The

aggregation of copper complexated amphiphiles in water was investigated with TEM. Small

globular structures were observed with an oblate shape (Figure 3.23a). At high magnifications

a highly ordered structure could be observed in the surface of the aggregates (Figure 3.23b).

A larnellar structure was present with periodicities of 8 nm. It is not clear if this pattem in the

nanostructures is caused by a change in the type of aggregation that produces mieropbase

separation of the copper containing en ti ties.

b

-50nm

Figure 3.23. TEM picture of a UT5 M aqueous salution of [CU-i PS-dendr-(NH2)8]Cl8• The sample was not stained.

66

Aggregation Behaviour of PS-De.ndrimers

Stabie sölutions in toluene could be obtained as well. High resolution TEM showed the

presence of large (ca. 550 nm) rather monodisperse aggregates of [Cu16 PS-dendr

(NHz)32]Cb2 (see Figure 3.24).

Figure 3.24. HREM picturesof [Cu16 PS-dendr-(NH2h2 ]Cl32 in toluene solution, the sample was not stained. a) Group of aggregates; b) intact sphere; c) one of the aggregates that has burst open; d) high magnification of the opening of one sphere. The high contrast of the dark particles around the opening may be due to copper in the particles.

The sample consisted of large spheres, presumably water-filled inverted micelles, the majority

of which have burst open, probably due to the drying process. Figure 3.24a shows a low

magnification image with several single spheres. Part of the content of the spheres has been

projected out of the aggregates and is also visible near the openings where they broke. In

Figure 3.24b, an image is shown of one of the few intact spheres in the sample. In Figures

3.24c and 3.24d, two images are shown at higher magnification of spheres that have

exploded. Part of the content of the sphere that has been expelled seems to contain small dark

dots that may possible be copper containing particles?3

67

Chapter3

3.6. Conclusions

The results presented in this chapter, together with the initial reports of Jan van Hest, 11 allow

us to obtain a general view on the amphiphilic behaviour of a series of different block

copolymers of polystyrene-poly(propylene imine) dendrimers. The combination of

hydrophilicity, highly branched structure and chemica! functionality makes poly(propylene

imine) dendrimers very interesting building blocks to be used as polar part in amphiphilic

block copolymers. The properties of dendrimers offer us not only the possibility to change the

size and molecular weight of the molecule by changing the generation, but also the ability to

synthesize a wide range of compounds with new properties and applications by chemically

modifying the dendritic part of the block copolymer. Polystyrene-poly(propylene imine)

dendrimer diblock copolymers form stabie Langmuir films. The isotherms recorded for low

generations are dominated by steric interactions between polystyrene chains on the surface. In

the case of the third and fourth generation, the headgroups are larger and they interact with

the water subphase causing an increase in the area per molecule on the corresponding

isotherms. Extrapolation of the curves to zero pressure yields a hypothetical cross-section per

molecule at vanishing surface pressure. Polystyrene-poly(propylene imine) dendrimers are

also able to form self-assembled monolayers on a solid substrate from an organic solution.

The interaction of the hydrophilic groups of the dendrimer with the hydrophilic substrate was

strong enough to afford a completely covered surface. From monolayer experiments valuable

information was obtained about the structure and behaviour of the molecules on the interface.

Previous studies of the amphiphilic behaviour of PS-dendr-(NHûn clearly showed a change

in aggregation type from inverted micellar structures for the first and second generation (n = 2

and 4), through vesicles 3rd and 41h generation (n = 8 and 16), to spherical micelles for the 51

h

generation dendrimer (n = 32).9 XRD measurements give us valuable information about the

association of the molecules in the absence of solvent and a good approximation of the

bilayer thickness. The critica! aggregation concentrations are in the same order of magnitude

for all polystyrene-poly(propylene imine) dendrimer diblock copolymers, being dependent on

the chain length of the apolar part, rather than on the size or functionality of dendrimer

headgroup. The rigidity in the interior of the aggregates was very high and there was no

evidence found of a melting transition in the PS phase in the temperature range measured (5-

90 "C). The addition of a plasticiser for the polystyrene part, such as toluene, showed a

remarkable drop in the viscosity of the hydrophobic domains. Changes in the chemica!

68


functionality of the dendrimer block did not introduce appreciable changes in the aggregation

form of the block copolymers in water, since the shape of the aggregates is mainly dependent

on the headgroup-tail size-balance. However, the introduetion of new functionalities on the

dendrimer block is important to achieve cationic or anionic groups, ligands for metal

complexation, etc. on the aggregate surface, which wiJl int1uence the kind of interactions of

the aggregates in aqueous solutions with hydrophilic charged surfaces or other molecules in

solution. The complexation of transition metal ions on the headgroups of polystyrene

poly(propylene imine) dendrimers usîng primary amine endgroups as ligands is possible.

These metallic amphiphiles formed stabie aggregates in aqueous, as well as in organic

solutions. TEM pictures of these amphiphiles, showed small dots or highly ordered structures

in the surface of the aggregates, confirming the formation of nanoscopic metallic clusters in

solution. Furthermore, the aggregates formed by polystyrene-poly(propylene imine)

dendrimer diblock copolymers have proved to possess a remarkable stability, much higher

than in the case of lipasomes or low molecular weight surfactants, due to their polymerie

nature. These features introduce many possibilities for the use of polystyrene-poly(propylene

imine) dendrimer aggregates in catalysis and biologica) sciences.

3.7. Experimental

Materials

PS-dendr-(NH2)n, PS-dendr-(COOH)n. and PS-dendr-CN(CH3) 3) 5 with n = 4, 8, 16 and 32, and amine

functionalized polystyrene (PS-0-(CH2) 3-NH2) have been synthesized in our laboratory, analogous to

a Iiterature procedure that has been publisbed in more detail elsewhere.JJ Commercial p.a. grade

solvents were used, unless otherwise indicated. DPH, pyrene and 5,10,15,20-tetrakis(4-

carboxyphenyl)porphyrin (5) were purchased from Aldrich. The porphyrins were kindly provided by

Albert Schenning. The synthesis of 5,10,15,20-tetrakis(4-methylphenyl)porphyrin (1) was carried out

in the laboratorles of the University of Nijmegen and has been described before in literature.34

5,10,15,20-Tetrakis( 4-(polystyrene)phenyl)porphyrin 35

This porphyrin was synthesized starting from 5,10,15,20-tetrakis(4-carboxyphenyl)porphyrin. 1.429 g

(0.12 mol) thionyl chloride was added to 86,6 mg (0.109 mmol) of porphyrin. The reaction mixture

was refluxed for 2 h. The solution was evaporated at low pressure. The resulting acid chloride

functionalized porphyrin and 1.78 g (5.4 mmo I) PS-O-(CH2)J-NH2 were dissolved in 50 ml of dry

CHzClz with 1 ml N(CH2CH3)3. The reaction mixture was stirred at room temperature for 2 hours.

69

Chapter3

The solution was then filtered and evaporated under reduced pressure. Further purification was

achieved by column chromatography on silica with 1% EtOH/CH2Ch as eluent. Yield: 1.22 g (82%)

of2 as a purple powder. GPC Mw: 12285 g/mol (D 1.1). 1H NMR 15: 0.6-1.2 (9H, Bu-(CHrCHPh)n);

1.2-1.7 (Cfu-CHPh); 1.7-2.3 (CHz-CHPh); 3.5 (8H, 0-(CHz)z-CJ::h-NHCO) (2H, 0-Cfu-(CHz)z

NHCO); 6.3-7.3 (CHz-CHPh); 7.9 (8H, 3,5-phenyl); 8.2 (8H, 2,6-phenyl); 8.8 (8H, ~-pyrrole) ppm.

UV-vis (CHCb) À: 421,516,551,590,645 nm.

Copper 5,10,15,20-tetrakis(4·(methyl)phenyl)porphyrin (3) and Copper 5,10,15,20-tetrakis(4·

(polystyrene)phenyl)porphyrin (4)

5,10,15,20-Tetrakis(4-(methyl)phenyl)porphyrin (127 mg, 0.19 mmo!) and 150 mg (0.75 mmo!)

Cu(CH3C02) 2 were dissolved in 5 mi DMF and the mixture was refluxed ovemight until all free-base

porphyrin was consumed ( foliowed by UV -vis). After cooling to room temperature the reaction

mixture was poured into water and extracted with 30 mi CHCJ,. The organic layer was washed 2x

with 50 ml water and 3x with 50 mi of 5% aqueous sodium carbonale solution. The organic layer was

dried with Na2S04, filtered and evaporated at low pressure. Yield: 124 mg (90%) as a dark red

powder. UV-vis (CHCh) À: 419, 540. Copper 5,10,15,20-tetrakis(4-(polystyrene)phenyl) porphyrin

was prepared in the same way from 44 mg (3.2 Jlmol) of 2 and 2.64 mg (13.2 Jlmo1) Cu(CH3C02)2.

Yield: 4.03 mg (91%) as a dark red powder. UV-vis (CHCh) À: 414,541 nm.

Copper complexes

PS-dendr-(NH2)s (9.75 mg, 2.4 Jlmol) was dissolved in 9 mi of 2:1 CHCl:JCH30H. After addition of

1 mi of 9.6 mM CuC]z solution in CH30H (1.3 mg, 9.6 JlffiOI), the solution immediately tumed to a

blue colour bel on ging to an absorption maximum at 680 nm indicative of the complexation of copper

with the primary amines. After evaporation of the solvent, the product was washed with water. Yield:

9.6 mg (87%) as a blue-green powder. UV-vis (2:1 CHCI:JCH30H) À: 680 nm. The aqueous solutions

were prepared by the injection method. 0.46 mg [Cu4 PS-dendr-(NH2) 8)Cl8 was dissolved in 100 Jll

THF/EtOH 2: I and injected into I 0 mi warm water (55 "C), to give blue-green opalescent solutions.

Monolayers

Monolayer experiments were performed at 20 oe on a home-built Langmuir trough at the Laboratory

of Physical Chemistry and Colloidal Science, Wageningen Agricultural University. The surface

pressure was measured using the Wilhelmy plate method. On the subphase (Milli-Q water), 50 mi of

a stock solution of the amphiphile (0.15 mM) in CHCI3 was spread and allowed to evaporate. The

rate of compression was 50 mm2/s. In order to establish the stability of the monolayer, the surface

70


pressure was maintained at a constant value of I 0 mNim for 8 hours. The monolayer area was found

to remain constant, indicating thatthe monolayers are stable.

AFM Experiments

Atomie force microscopy was done at the lnstituto de Microelectrónica de Madrid (eentro Superior

de Investigaciones eientfficas) on a sample prepared by dipping a Si single crystal wafer on a eHel3

solution of amphiphile during I5 min. The Si wafer was previously treated at 1000 oe for 12 hours to

obtain a silicon oxide layer on the surface of ca. 90 nm The micrographs were measured by means of

a Multimode Nanoscope lil (Digital lnstruments) operaled in the tapping mode at a resonance

frecuency of 300-350 kHz. The measurements were performed under ambient conditions using

Nanosensors Si cantilevers with a spring constant of 35-45 Nlm. The tip radius was 20-50 nm.

The contact angle experiments were done by depositing a single droplet of 50 IJ.l water onto a clean

glass wafer. After dipping of the glass into a chloroform solution of PS-dendr-(eOOH)8 during I5

min, the wafer was rinsed with chloroform and dried with compressed air. Another water droplet of

50 IJ.] was then deposited on the coated glass for comparison with the first droplet The contact angles

were measured from the pictures and they are only approximated values.

Preparation of Samples

Aqueous solutions were prepared by the following procedure: the amphiphiles were dissolved in 2 mi

of THF. After actdition of water (25 mi), the organic solvent was evaporated and stabie aggregates

were formed in water. The concentrations ranged from approximately 10-8 to 10-4 M.

Electron Microscopy

A droplet of aqueous solution of polystyrene-poly(propylene imine)dendrimer was placed on a eu

grid, covered with formvar or mica, and allowed to dry for I minute, after which the droplet was

removed. Pt-shaded samples were prepared by covering the dried sample with Pt using a Balzers

Sputter unit Freeze fractured samples were prepared by actdition of a droplet of the amphiphile

dispersion onto a golden microscope grid (I 50 mesh), placed between 2 copper plates and fixated in

supercooled liquid pentane. Sample holders were placed in a Balzers freeze etching system BAF

4000 at 10-7 Torrand heated to -105 oe. After fracturing, the samples were etched for 1 min. (.6.T

20 oe), shaded with Pt and covered with carbon. Replicas were allowed to heat up to room

temperature and Ieft on 20% sulphuric acid for 16 h. After rinsing with water they were allowed to

dry. All samples were studied using a Philips TEM 201 (60 kV). The samples for HREM were

prepared by allowing a droplet of toluene solution of [eul(,PS-dendr-(NH2h2]el32 to dry at ambient

conditions on a copper grid. These samples were nol further stained. HREM was performed using a

71

Chapter3

Philips CM 30 T electron microscope operated at 300 kV at the National Centre for High Resolution

Electron Microscopy, Delft University ofTechnology.

XRD Measurements

Cast bilayers for X-ray diffraction measurements were prepared from aqueous vesicle solutions of

polystyrene-poly(propylene imine) dendrimer diblock copolymers. Aliquots of the colloidal

dispersions (2 ml) were left to dry on Si wafers in a desiccator over sodium hydroxide. These Si

single crystal wafers, cut along the (501) plane, were used for low angle X-Ray measurements. The

specimen chamber was rnaintained at 20 oe and flusbed with air of variabie relative humidity, RH=

0-90%, during the XRD measurements. The patterns were digitally recorded. Peak positions were

obtained using peak fitting JANDEL software. Smal] angle X-ray diffraction measurements were

performed using a Huber D8211 goniometer. These high accuracy diffractometer is equipped with a

Cu LFF XRD Philips tube, variabie divergence and antiscatter slits, and an energy dispersive Si/Li

Kevex detector, which enable a high peak to background ratio.

Critical association concentrations

Steady-state fluorescence spectra were run in a Perkin-Elmer Luminescence spectrometer LS 508 in

the right-angle geometry (90° collecting opties) using slit openings of 5 nm for emission and 2.5 nm

for excitation. 1 cm square quartz cells were filled with ca. 3 ml solution, with [pyrene] = 4.8 x 10-7

M. For fluorescence emission spectra Àex was 339 nm, for excitation spectra À.cm was 390 nm. The

block copolymer was dissolved in an emulsion of tetrahydrofuran and water with agitation. After

removal of the organic solvent in the rotatory evaporator at 35 oe, a stock solution of the polymer in

water was obtained. All samples were prepared by adding a known amount of pyrene in acetone to a

series of empty 10 ml volumetrie flasks; after evaporation of the acetone, known amounts of the stock

solution of amphiphile were added and diluted with distilled water in order to obtain final polymer

concentrations between 10-4 and 10-9 M. The flasks were sealed and stirred ca. 20 h at room

temperature to allow the pyrene and the aggregates to equilibrate.

Microcalorimetry

Differential scanning microcalorimetry measurements were carried out on a Microcal MC-2

microcalorimeter (Microcal, Amherst MA), fitted for enhanced sensitivity with a nanovoltmeter. The

heat capacities of pure water and vesicle solutions were compared, the sample and reference volumes

being 1.2 mi.

72

Aggregation Behawour of PS-Dendrimers

Fluorescence probes

For the pyrene excimer formation experiments, the samples were purged with argon and stored in the

dark to rninhnize pyrene degradation. Stock solutions of pyrene in acetone (10 mi 6 mM) and

polystyrene-poly(propylene imine) dendrimer diblock copolymers in water (25 mi 0.5 r.tM) were

prepared. ealculated amounts of pyrene solution were allowed to dry into a flask under vacuum, then

2.5 mi of vesicle solution was added. After stirring overnight, the fluorescence excitatîon and

ernission spectra were recorded. Final amphiphile and pyrene concentrations amounted to 5·1 o-s and

10~7-10-5 M, respectively. The sample was thermostated by using a thermocouple in the sample

holder conneeled toa RTE 110 Neslab thermostat, wîth an ethylene glycol/water (lil, v/v) bath. The

temperature was controlled by a thermometer in the sample holder connected to the computer. The

measurements were made in a range of temperatures of 5 to 80 oe.

For the fluorescence anisotropy measurements, DPH was dissolved together with the polystyrene

poly(propylene imine) dendrimers by the ethanol injection method (the desired amount of dendrimer

is dissolved in 100 1-11 ethanol/THF (1/2 v/v); the resulted solution is warmed up to ca. 55 oe and

injected into 10 mi of water, preheated to 60 oe, while stirring). The vesicle solutions were placed in

a bath-type sonicator and sonicated for 30 rninutes at room temperature. The ratio amphiphile to

probe was 250:1. For fluorescence ernission spectra Àex = 382 nm, for the excitatîon spectra Àem = 430

nm. In all cases the anisotropy was measured in a range of temperature between 5 and 90 oe by

taking steps of 5 oe and allowing the solutions to equilibrate during 15 min.

At high temperatures, dynamic quenching processes are enhanced and the fluorescence lifetime is

decreased. lf the time scale at which we are "measuring" the molecular motion becomes shorter, the

anisotropy value becomes higher. A correction for this phenomenon has to be made. As only 90"

steady-state fluorescence measurements have been performed, the lifetîme ( T) has to be determined

indirectly, by making the approximation that the higher fluorescence intensity (/0) corresponds with

the maximal fluorescence decay time, which for DPH is equal to 11.4 ns. The lifetime for each

temperature was calculated according to the empirica! equation T = To I I 10.

Fluorescence quenching

Samples for self-quenching experiments were prepared by the ethanol injection method and sonicated

for 15 min. The porphyrin concentration was maintained constant at 0.1 1-lM for 1 and 2, while the

amphiphile to porphyrin ratio R varled between 3000 and 100. The Àmax of absorption of the

prophyrins was chosen as excitation wavelength (424 nm). Quenching of the fluorescence emission

with iodide ions was studied by titration of 2.5 mi of a PS-dendr-(NH2). vesicle solution (R = 500)

with 50 r.tl aliquots of a 0.9 rnM KI solution. The fluorescence intensity was measured 5 min after

each ad dition of quencher and the data were corrected for the volume increments.

73

Chapter3

ESR measurements

Yesiele solutions containing copper porphyrins were prepared by the ethanol injection method, with

amphiphile concentration 0.8 mM and the con centration of the porphyrins 2·1 o-5 M. These

dispersîons were driedon a Mylar film in a dessicator over NaOH. The resulting films were carefully

cut into 3 x 20 mm strips. Approximately 20 strips were stacked and fixed with Cellotape on a quartz

tube. This tube was placed in a Bruker ESP 300 ESR spectrometer, equipped with an Oxford flow

cryostat. Spectrometer conditions: modulation amplitude 4 G, modulation frequency 100 kHz,

microwave power 30 dB, microwave frequcncy 9.174 GHz, receiver gain 8·105, temperature 40 K.

3.8. References

(a) Tuzar, Z.; K.ratochvil, P. Micelles of Block and Graft Copolymers in Solutions, Surface and Colloid

Science, Vol. 15, Ed. E. Matijevic, Plenum Press, New York 1993. (b) Riess, G.; Nervo, J.; Rogez, D.

Polym. Eng. Sci. 1977, 8, 634. (c) Gallot, Y.; Selb, J., Marie, P.; Rameau, A. Polym. Prep. Am. Chem.

Soc. Div. Polym. Chem. 1982, 23, 16. (d) Desjardins, A.; Eisenberg, A. Macromolecules 1991, 24,

5779. (e) Xu, R.; Winnik, M.A.; Riess, G.; Chu, B.; Croucher, M.D. Macromolecules 1992,25, 644. (f)

Denton, J.M.; Duecker, D.C.; Sprague, E.D. J. Phys. Chem. 1993, 97, 756. (g) Toncheva, V.;

Velichkova, R.; Trossaert, G.; Goethals, E.J. Polym. Int. 1993, 31, 335. (h) Moffin, M.; Khougaz, K.;

Eisenberg, A. Acc. Chem. Res. 1996,29, 95. (i) Zhang, L.F.; Eisenberg, A. Science 1995,268, 1728. (j)

Spatz, J.P.; Möbmer, S.; Möller, M. Angew. Chem. 1996, 108, 1673.

2 (a) Lucassen-Reynders, E.H. Anionic suifactants, physical chemistry of swfactant action. Surfactani

series vol. ll, Marcel Dekker, New York 1981. (b) Fendler, J.H. Membrane mimetic chemistry:

characterizations and applications of micel/es, microemulsions, monolayers, bilayers, vesicles, host

guest systems and polyions, Wiley Interscience, Chichester 1982. (c) Buckingham, S.A.; Garvey, C.l.;

Warr, G.G. J. Phys. Chem. 1993, 97, 10236.

3 (a) Tomalia, D.A.; Naylor, A.M.; Goddard lil, W.A Angew. Chem. 1990, 102, 119. (b) Fréchet, J.M.J.

Science 1994, 263, 1710. (c} Tomalia, D.A Adv. Mater. 1994, 6, 529. (d) Newkome, G.R.; Nayak, A.;

Behera, R.K.; Moorefield, C.N.; Baker, G.R. 1. Org. Chem. 1992, 57, 358. (e) Hawker, C.l.; Wooley,

K.L.; Fréchet, J.M.J. 1. Chem. Soc .. Perkin. Trans /1993, 1287. (f) Percec, V.; Chu, P.; Kawasumi, M.

Macromolecules, 1994,27, 4441. (g) Watanabe S.; Regen, S.L. J. Am. Chem. Soc. 1994, I 16, 8855. (h)

Zimmennan, S.C.; Zeng, F.; Reichert, D.E.C.; Kolotuchin, S.V. Science, 1996, 271, I 095.

4 (a} Zimmerman, S.C.; Zeng, F.; Reichert, D.E.C.; Kolotuchin, S.V. Science, 1996, 271, 1095. (b) Huck,

W.T.S.; van Veggel, F.CJ.M.; Reinhoudt, D.N. Angew. Chem. Int. Ed. Engl., 1996,35, 1213.

5 (a) Newkome, G.R.; Yao, Z.-q.; Baker, G.R; Gupta, V.K. J. Org. Chem. 1985, 50, 2003. (b) Tomalia,

D.A.; Berry, V.; Hall, M.; Hedstrand, D. Macromolecules 1987, 20, 1164. (c) Newkome, G.R.;

Moorefield, C.N.; Baker. G.R.; Johnson, A.L.; Behera, R.K. Angew. Chem. 1991, 103, 1205. (d)

74


Tanaka, N.; Tanigawa, T.; Hosoya, K; Kimata, K.; Araki, T.; Teraba, S. Chem. Lett. 1992, 959. (e)

Newkome, O.R.; Young, J.K; Baker, O.R.; Potter, R.L.; Audoly, L.; Cooper, D.; Weiss, C.D.

Macromolecules 1993,26, 2394. (f) Wooley, KL.; Hawker, O.J.; Fréchet, J.M.J. J. Am. Chem. Soc.

1993, 115, 11496. (g) Newkome, O.R.; Moorefield, C.N.; Keith, J.; Baker, O.R.; Escamillo, 0. Angew.

Chem. 1994, 106, 701. (h) Kuzdzal, S.A.; Monning, C.A.; Newkome, O.R.; Moorefield, C.N. J. Chem.

Soc. Chem. Commun. 1994, 2139. (i) Muijselaar, P.O.H.M.; Claessens, H.A.; Cramers, C.A.;. Jansen,

J.F.O.A; Meijer, E.W.; de Brabander-van den Berg, E.M.M.; van der WaJ, S. J. High Resol. Chrom.

1995, 18, 121. (j) Oitsov, 1.; Wooley, KL.; Fréchet, J.M.J. Angew. Chem. 1992, 104, 1282. (k) Oitsov,

1.; Wooley, K.L.; Hawker, CJ.; lvanova, P.; Fréchet, J.MJ. Macromolecules 1993, 26, 5621. (I)

Oitsov, 1.; Fréchet, J.M.J. Macromolecules 1994, 27, 7309.

6 (a) Oitsov, 1.; Wooley, KL.; Fréchet, J.M.J. Angew. Chem. Int. Ed. Eng/. 1992, 31, 1200. (b) Oitsov, L;

Fréchet, J.MJ. Macromolecules 1993, 26, 6536.

7 Chapman, T.M.; Hillycr, O.L.; Mahan, EJ.; Shaffer, K.A. J. Am. Chem. Soc. 1994, 116, I 1195.

8 Zhong, X.F.; Eisenberg, A. Macromolecules 1994, 27, 1751. (b) Zhong, X.F.; Eisenberg, A.

Macromolecules 1994, 27, 4914.

9 (a) van Hest, J.C.M.; Delnoye, D.A.P.; Baars, M.W.P.L.; Elissen-Román, C.; van Oenderen, M.H.P.;

Meijer, E.W Chem. Eur. J. 1996, 2, 1616. (b) van Hest. J.C.M.; Delnoye, D.A.P.; Baars, M.W.P.L.;

van Oenderen, M.H.P.; Meijer, E.W. Science 1995,268, 1592. (c) van Hest, J.C.M., Baars, M.W.P.L.;

Elissen-Román, C.; van Oenderen, M.H.P.; Meijer, E.W. Macromolecules, 1995,28,6689.

JO (a) lsraelachvili, J.N.; Mitchell, D.; Ninham, B. Biochim. Biophys. Acta 1977, 470, 185. (b)

Israelachvili, J.N.; Marcelja, S.; Hom, R. Rev. Bîophys. 1980, 13, 121. (c) lsraelachvili, J.N.; Mitchell,

DJ.; Ninham, B.W. J. Chem. Soc., Faraday Trans.l11976, 72, 1525.

11 van Hest, J.C.M. New Molecular Architectures based on Dendrimers. PhD Thesis, University of

Eindhoven 1996.

12 (a) Saville, P. M.; Reyno1ds, P. A.; White, J. W.; Hawker, C. J.; Fréchet, J. M. J.; Wooley, K. L.;

Penfo1d, J.; Webster, J. R.P. J. Phys. Chem. 1995, 99, 8283. (b) Kampf, J.P.; Frank; C.W.; Malmström,

E.E.; Hewker, C.J. Langmuir 1999, 15, 227. (c) Sheiko, S.S.; Buzin, A.I.; Muzafarov, A.M.; Rebrov,

E.A.; Oetmanova, E.V. Langmuir 1999, 14,7468.

13 (a) de Brabander, E.M.M.; Brackman, J.; Muré-Mak, M.; de Man, H.; Hogeweg, M.; Keulen, J.;

Scherrenberg, R.; Coussens, B.; Mengerink, Y.; van der Wal, Sj. Macromol. Symp. 1996, 102, 9. (b)

Scherrenberg, R.; Coussens, B.; van Vliet, P.; Edouard, 0.; Brackman, 1.; de Brabander, E.M.M.;

Mortensen, K Macrouw/ecules 1998, 31,456.

14 Israelachvili, J. lmemwlecular mui Swjace Forces, Academie Press; London, 1992.

15 (a) Tsukruk, V. T. Adv. Mater. 1998, JO, 253. (b) Bliznyuk, V. N.; Rinderspacher, F.; Tsukruk, V. V.

Polymer 1998, 39, 5249. (c) Tsukruk, V. V.; Rinderspacher, F.; Bliznyuk, V. N. Langmuir 1997, 13,

2171. (d) Esumi, K.; Ooino, M. Langmuir 1998, 14, 4466.

75

Chapter3

16 Lasic, D.D. Biochem. J. 1988, 256, l.

17 Spink, C.H.; Lieto, V.; Mereand, E.; Pruden, C. Biochemistry 1991, 30, 5104.

18 Regev, 0.; Kang, C.; Khan, A. J. Phys. Chem. 1994,98,6619.

19 Wilhelm, M.; Zhao, C.L.; Wang, Y.; Xu, R.; Winnik, M.A.; Mura, J.L.; Riess, G.; Croucher, M.D.

Macromolecule.~ 1991,24, 1033.

20 Kalyanasundaram, K.; Thomas, J.K. J. Am. Chem. Soc.1977, 99, 2039.

21 Astafieva, I.; Xong, X.F.; Eisenberg, A. Macromolecules 1993,26, 7339.

22 Marko, J.F.; Rabin, Y. Macromolecules 1992, 25, 1503.

23 (a) Rosen, M.J. Swfactants and lnteifacial Phenomena, John Wiley & Sons; New York, 1989. (b)

Mayers, D. Suifactant Science and Technology, VCH Publishers Inc.; New York, 1988. (c) Lindman,

B.; Wennerstrom, H. Top. Curr. Chem. 1980, 87, l.

24 (a) Smits, E.; Blandamer, M.J.; Briggs, B.; Cullis, P.M.; Engberts, J.B.F.N. Rev. Trav. Chim. Pays-Bas

1996, 115, 37. (b) Hoehne, G.W.H.; Hemminger, W.; Flammerschein, H.J. Differentlal Scanning

Calorimetry: an Introduetion for Practitioners. Springer; Berlin, 1996.

25 van Esch, J.H.; Feiters, M.C.; Peters, A.M.; Nolte, R.J.M. J. Phys. Chem. 1994, 98,5541.

26 (a) Kremer, J.M.H.; van de Esker, M.W.J.; Pathmamanoharan, C.; Wiersema, P.H. Biochemistry 1977,

16, 3932. (b) Rupert, L.A.M.; Hoekstra, D.; Engbens, J.B.F.N. J. Am. Chem. Soc. 1985, 107, 2628.

27 (a) Kasba, R., Rawls, H.R.; Asraf EI-Bayoumi, M. Pure Appl. Chem. 1965, JJ, 371. (b) Hunter, C.A.;

Sanders, J.M.K.; Stone, A.J. Chem. Phys. 1989, 133, 395.

28 Nakashima, N.; Ando, R.; Kunitake, T. Chem. Lett. 1983, 1577.

29 (a) Van Esch, J.H. Studies on Synthetic Bilayer Aggregates. In Search of Supramolecu/ar Catalysts.

PhD Thesis, University of Nijmegen 1993. (b) Ishikawa, Y.; Kunitake, T. J. Am. Chem. Soc. 1991, 113,

612. (b) Michaeli, S.; Hugerat, M.; Levanon, H.; Bemitz, M.; Natt, A.; Neumann, R. J. Am. Chem. Soc.

1992, 114, 3612.

30 (a) Smith, T.D.; Pilbrow, J.R. Coord. Chem. Rev. 1974, 13, 173. (b) Palmer, G. Biochem. Soc. Trans.

1985,548.

31 (a) Andrews, M.P.; Ozin, G.A. Chem. Mater. 1989, 174, I. (b) Kaneko, M.; Tsuchida, E. J. Polym. Sci.,

Macromol. Rev. 1981, 16,397. (c) Biswas, M.; Mukherjee, A. Adv. Polym. Sci. 1994, IJ5, 91.

32 (a) Bosman, A.W.; Schenning, A.P.H.J.; Janssen, R.A.J.; Meijer, E.W. Chem. Ber./Receui/1997, 130,

725. (b) Bosman, A.W. Dendrimers in Action. PhD Thesis, Eindhoven University ofTechnology 1999.

33 Dr. P.J. Kooyman, National Centre for High Resolution Electron Microscopy, Delft University of

Technology, personal communications.

34 (a) van der Made, A.W. On the Epoxidation of Alkenes with Hypochlorite Catalysed by Manganese

(Ill) Tetraarylporphyrins. PhD Thesis, University of Utrecht 1988. (b) Little, R.G.; Anton, J.A.; Loach,

P.A.; Ibers, J.A. J. Heterocycl. Chem. 1975, 12, 343.

35 Schenning, A.P.H.J. personal communications.

76

4 Solid-State Mieropbase Separation of Polystyrene

Poly(propylene imine) Dendrimers

Abstract

"Higher levels of complexity are distinguished by some genuine!J new ftatures and

activities, and these require distinctive theories, language, and concepts to describe

them"

A.R. Peacocke

Diblock copolymers consisring of polystyrene and acid1unctionalized poly(propylene imine)

dendrimers have been found to self-assemble spontaneously into regu/ar microdomains.

Hybrid dendrimer-linear chain block copolymers yield highly asymmetrie molecules which

display a behaviour at the strong segregation limit that dif.fers from the well-known all-linear

block copolymers. The morphological features are used to detennine the relationship

between the architecture of the blocks and the arrangement of the molecules in the bulk.

Micro-lattice formation of these materials was examined by using small-angle X-ray

scattering and transmission electron microscopy techniques. By increasing the dendrimer

generation, the structures changed from hexagonally packed cylinders with polystyrene as

matrix to lamellar phases of different packing density. Good agreement between TEM and

SAXS results was obtained and the micro-lattice morphology was found to be highly

dependent on the dendrimer generation.

4.1. Introduetion

The rapid development of synthetic chemistry allows us to access new molecular structures

that can be designed to explore the role of polymer topology on the physical and chemica!

properties of macromolecules in genera11 and block copolymers in particular? Inthelast few

years an increasing interest emerged in complex copolymer architectures, like triblocks,3

stars,4 dendrimers,5 rings,6 hydra-7 and superamphiphiles8 of weB-controlled structures and

dimensions; thereby it is possible to acquire a better understanding about the relation between

Chapter4

the chemica] structure and the morphology of block copolymers. Bates et al. pointed to the

complexity of the phase behaviour of block copolymers.9 They found new microstructures

that were not described before and explained this behaviour in terros of compositional and

conformational symmetry. The composition of the polymer and differences in the

conformational and volume-filling characteristics of each block are important parameters.

Non-linear miktoarm star polymers have been synthesized to investigate the influence of

chain architecture on polymer properties.10 Comparison with linear block copolymers showed

that the macromolecular architecture not only strongly affects the morphology of the domain

borders, but can introduce new morpbologies as well. Ring polymers have been used to study

relaxation and diffusion mechanisms.6" Star polymers with multiple arms have been applied

also as model for polymerie micelles. Their viscoelastic behaviour was compared with

analogous linear polymers.4" Series of macrocycles with different Mw, like macrocyclic

poly(2-vinylpyridine), have been synthesized and their hydrodynamic size and stiffness have

been compared with their linear precursors in the melt. 11 Gnanou et al. studied the

possibilities offered by living ring-opening metathesis polymerization of macromonoroers to

engineer novel macromolecular architectures.12 By this metbod amphiphilic branched

structures with novel topologies have been prepared. The synthesis and morphological studies

of well-defined block ionomers, carried out by different researchers, 13 have provided much

important information about the aggregation mechanisms of ionic groups to form domains.

The first example of carboxylate- and dicarboxyethyl-terminated polystyrene

synthesized by Eisenberg et al. can be regarded as the zeroth and first generation dendeitic

superamphiphile, respectively.14 The aggregation behaviour of these ionic copolymers in

organic solvents was extensively investigated. Diblock copolymers with one dendritic block

and one Iinear block have been prepared by seveml groups. Newkome et al. studied the

aggregation of hydrophilic arborols bonded at both ends to a long hydrocarbon chain. 15 The

preparation of polymerie surfactants with hydrophobic terminal residues on a lysine

dendrimer anached to a PEO tail (hydra-amphiphiles) and their surfactant properties have

been described as well.7 Fréchet et al. have reported dendritic-linear diblocks and triblocks

with hydrophobic and hydrophilic termini. 16 The therrnal behaviour of these copolymers was

investigated with DSC and it was found that when the dendrimer block was the majority

block, phase mixing took place, while in the case of the linear PEO block ha ving the highest

mass, mieropbase separation was found. More recenûy, the group of Hammond reported the

78

Soliel State Microphase Separation of PS-Dendrimers

synthesis of an amphiphilic dendritic block copolymer with PEO as the linear block and

PAMAM as the dendritic block. 17 Thermal characterization indicated that these diblock

copolymers exhibit some degree of mieropbase separation as well, though no bulk

morphology was described. The authors suggested that the differences found in the melting

temperature of the PEO block were due to changes of crystallite sizes on account of variations

on the block copolymer sizes or geometries, or to an increased extent of mixing for larger

dendrimer blocks.

In Chapter 3 the aggregation behaviour in aqueous solution of amphiphilic block

copolymers from the combination of a hydrophobic linear polymer (polystyrene) and

hydrophilic poly(propylene imine) dendrimers has been described.8·18 A remarkable

generation dependency was observed for these amphiphilic polymers that qualitatively

foliowed lsraelachvili's theory on the relationship between molecular shape and aggregation

form. 19 In an analogous way, the contiguration of this new type of polymers is expeeted to

intlucnee significantly the micro-lattice formation in the solid phase as well. In this chapter,

the mieropbase separation of polystyrene-dendrimer bloek eopolymers in the solid state is

described. Morphological studies in the solid phase have been performed for the first time in

this type of dendritic-linear block copolymers. These results yielded a better understanding of

the intlucnee of macromolecular architecture on the structure of micro-lattices of block

copolymers.

4.2. Microphase separation

Block copolymers provide the possibility to incorporate different physical properties into a

single molecule, and thereby to avoid the macroscopie phase separation that commonly

accompanies any attempt to blend chemically different homopolymers. However, the most

interesting properties of block copolymers are manifested when incompatibility effects are

present between the two blocks. The characteristic feature of diblock copolymers (A-B) is the

presence of strong unfavourable interactions between unlike sequences even when the

unfavourable interactions between unlike monomers are relatively weak. As a result,

sequences tend to segregate, but as they are chemically bound even a complete segregation

can not lead to a macroscopie phase separation as in mixtures of two homopolymers. In the

case of a sufficiently strong incompatibility, microphase separation occurs and microdomains

rich in monomer A and in monomer B are formed. When mieropbase separation occurs, the

79

Chapter4

microdomains are not located randomly but form a rather regular arrangement giving rise to a

periodic structure, a micro-lattice. The geometry of the microstructure is largely controlled by

the volume fraction of the A block, $A, versus that of the B block, <!>a· Conformational

asymmetry between A and B blocks plays also a significant role in determining the geometry

of the lattice.

A schematic representation of mieropbase separation in an AB-type block copolymer

in which both blocks are non-crystalline is given in Figure 4.1. In this case, the domains are

composed of A blocks and the matrix of B blocks. The space in the domains and the matrix is

packed with A and B blocks, respectively, to a density corresponding to the bulk state. The

intermediate regions represented in the picture, correspond to an interfacial region in which

mixing of A and B blocks can occur.

Figure 4.1. Schematic representation of microphase separation in an AB block copolymer in which both components are non-crystalline.

Figure 4.2. Effect of composition on block copolymer morphology: (a) spheres of A in matrix of B; (b) cylindersof A in matrix of B; (c) alternaring A and B lamellae; (d) cylindersof B in matrix of A; (e) spheres of Binmatrix of A.

For block copolymers A-B in which both blocks are non-crystalline, there are three

basic types of domain possible: spheres, cylinders and lamellae.20 Based on theoretica]

80

Solid State Microphase Separation of PS-Dendrimers

backgrounds and on empirica! findings, in general a gradual change in structure as described

in Figure 4.2, can be expected when the composition of the copolymer varles from

predominantly B to predominantly A.

60r-T-r--.-----,.--..,-,......,

Ordered 50

Figure 4.3. Schematic phase diagram for block copolymers; (bcc: body-centered cubic array of spheres; hex: hexagonally packed cylinders; ODT: order-disorder transition).

For a given block copolymer, the type of morphology, or phase diagram, depends on N, the

total degree of polymerisation; fA NA I (NA + NB). the fractional composition; and X· the

Flory-Huggins A-B interaction parameter.21 * In the most simple case a schematic phase

diagram can be based on the mean-field model of Leibier (see Figure 4.3).22 To a first

approximation, the ODT (order-disorder transition) is dctermined by thc dimensionless

product xN, as a function of fA· Within the ordered state, the dashed lines indicate

approximate boundaries between different morphologies, or order-{)rder transitions (OOTs).

Recently, it bas become clear that other variables are also important, like the statistkal

* The Flory-Huggins segment-segment parameter X is defined as X= Z8W ArJkT, where Z is the number

of contacts between segrnents of A and of B, 8 W AB is the Gibbs energy balance associated to the

energetic process of break:ing the contacts between segments A-A and B-B to generale A-B contacts; k

is the Boltzman constant and T the temperature. X is related to the change in enthalpy in the process of

mixture of blocks A and B as !J,.HAB = nkT<PA<!>sx. where <1> is the volume fraction of one block and n is

the total number of segments.

81

Chapter4

segment Jength of the blocks and differences in the conformational and volume filling

characteristics of each block.9

Several techniques like transmission electron microscopy, X-ray scattering and neutron

scattering can provide information about symmetries, periodicities and size of the

microdomains. 23

The dependencies of the equilibrium morphology on molecular parameters, such as molar

mass, fraction of monomers A in the polymer and chemica! nature of copolymer monomers,

have been studied in detail.24 However, not much is known about the influence of geometry

and conformational factors of the polymer in the symmetry of micro-lattices. By using

dendritic structures in one of the blocks, we can study this phenomenon in some detail.

4.2.1. Matenals

The mieropbase separation of polystyrene-poly(propylene imine) dendrimers of generations

2-5, with 4 and 32 primary amine endgroups respectively (Figure 4.4), were investigated with

SAXS and XRD.

Figure 4.4. Schematic representation of PS-dendr-(NH2Jn dendrimers. The tail of the superamphiphile consistsof an atactic polystyrene chain of M,. 3.2 kg/mol.

82


Surprisingly, no evidence was found of microphase separation in any of the samples.

Theoretica! calculations on the x parameter for a copolymer consisting of styrene and

propylene imine blocks suggest that the incompatibility effects between the two block are not

strong enough to induce segregation in the solîd phase?5 In order to introduce higher polarity

into the dendrimer block, methylated compounds (Figure 4.5) were also investigated in the

solid phase.

Figure 4.5. Schematic representation of the fourth generation methylated dendrimer [PS-dendr(N(CH_ihhól h1 .

The introduetion of po lar groups into the dendritic branching points, in the form of quatemary

ammonium salts, induces stronger incompatibility effects between the polystyrene block and

the polar dendritic block. In this case microphase segregation does occur, giving rise to the

formation of microlattices in the bulk. Lamellar structures with a layer thickness of ca. 10 nm

were visualized for PS-dendr-tN(CH3)3)4, PS-dendr-tN(CH3)3)8 and PS-dendr

tN(CH3)3)J6, using transmission electron microscopy teehniques (see Figure 4.6). An

alternative method to introduce polarity in the dendrimer block was the use of metal

complexation. It has been described in Chapter 3 that the primary amines of the dendrimer

block can act as effeetive ligands for transition metals, such as Cu2+. Complexation of the

endgroups of PS-dendr-(NH2)n. with n = 4, 8 and 16, was carried out with CuCh. As in the

case of the salts, introduetion of metal ions in the dendrimer block produces segregation of

the blocks in the bulk and microphase separation occurs. In the ease of [Cu2PS-dendr

(NH2)4] Cl4, the excess of CuCh that did not react with the primary amines formed rather

83

Chapter4

monodisperse salt clusters homogeneously distributed in the bulk. In the matrix, the

formation of an organized arrangement seems to appear, however, from the electron

micrographs this structure is difficult to define. [Cu4PS-dendr-(NH2) 8] Cl8 forms lamellar

structures in the solid phase. The excess of salt that was not complexated to the polymer was

expelled from the bulk and it was found at the surface of the solid phase in the form of

globular clusters.

Figure 4.6. TEM picture of a solid film of PS-dendr-(N(CH3h)s'showing lamellar phases. The sample was nat stained.

Changes in the chemica! functionality of the dendrimer block cause incompatibility effects

between the two blocks as well and produce mieropbase separation. To study the effect of

polar groups in the segregation of these block copolymers, acid functionalized polystyrene

poly(propylene imine) dendrimers (see Figure 4.7) were studied in detail.

The synthesis of PS-dendr-(COOH)n block copolymers is based on the use of a well-defined

monodisperse atactic polystyrene chain, obtained by anionic polymerization, as core moleeule

for building the poly(propylene imine) dendrimer using the standard sequence.26 This has

been reported elsewhere.18 Characterization was carried out with IR, 1H-NMR, 13C-NMR and

MS techniques. Based upon analytica! characterization in non-aqueous solutions and in the

solid state, it is assumed that the block copolymers have the amine and carboxylic acid form

and not the zwitterionic form, as they may have in water, however, some acid-base reaelions

84

So/id State Microphase Separation of PS-Dendrimers

can not be ruled out. The morphology of these new non-linear block copolymers has been

examined by using SAXS, TEM and XRD techniques.

PS-dendr-(COOH)4 0\Sl'loH COOH l.r ?_OCH \1 NJ'"'COOH

)~ f -'COOH

~~N ...,...J-.,..."'--cooH

. @ ~'·ç-~·l~::: ~ t ............ cooH r \ '""COOH

L a..,"-.cooH ) t;ooH H"cooH

PS-dendr-(COOH),6 PS-dendr-(C()OH)32

Figure 4.7. Schematic representation of PS-dendr-(COOH),. dendrimers.

4.2.2. Mieropbase separation in PS-dendr-(COOH)n

Small angle X-ray scattering measurements

The type of lattice formed on a crystalline sample can be determined by comparing the

observed sequence of d spacings with characteristic sequences of model lattices (Table 4.1).

The d spacings are determined by the Bragg equation, 27•

28

nÀ = 2d sin 8 Equation 4.1

where the 0 angle is half the scattering angle at which the diffraction peak is observed and À

is the wavelength of the radiation.

85

Chapter4

Table 4.1. Ratiosof consecutive Bragg spacings for different model morphologies.

Morphology

Bcc packed spheres'

Fee packed spheresb

Hexagonally packed cylinders

Lamellae

Ratios

1 : 0.706: 0.577: 0.500

1 : 0.866: 0.612: 0.521

1 : 0.577 : 0.500 : 0.378 : 0.333

1 : 0.500 : 0.333 : 0.250 : 0.200

a) Body-centered cubic array of spheres. b) Face-centered cubic array of spheres.

The SAXS profiles of the acid functionalized polystyrene-dendrirner diblock copolyrners for

4 different generations of dendrirner PS-dendr-(COOH)n are shown in Figure 4.8.

b

Int

Int

!.OxJO"' 2.0x!o·' 3.0x!O"' 4.0xl0° 5.0x!O' 6.0xJO"' 7.0x!O~

q (k')

Figure 4.8. Smal/ angle X-ray scattering intensity profiles for solid films of a) PS-dendr-(COOH)4,· b) PS-dendr-(COOH)R; c) PS-dendr-(COOH)J6 and d) PS-dendr-(COOH)J2.The scattering vector, q is equal to (2n:l:t)-sin8, where 8 is the scattering angle. *) The positions marked with an asterisk correspond to positions at which higher order peaks are expected for the geometry of the corresponding lattice in each case.

86

So/ld State Mlcrophase Separatlon of PS-Dendrlmers

The scattering of PS-dendr-(COOH)4 shows two clear diffraction peaks. The higher order

reflections are found at multiples of J3 and J4 of the first maximurn (ratios d2/d1 = 0.577,

d3/d1 = 0.500), although the peak corresponding to the 200 position is hardly visible. This

sequence of spacings corresponds to the geornetry of hexagonally packed cylinders (hpc). In

the case of PS-dendr-(COOH)4, the volurne of the dendrirner block is srnall cornpared with

the volurne of the polystyrene block and the dendritic part segregates, forrning cylinders

which are surrounded by a PS matrix. The distance between the eentres of these cylinders is

86 A, as deterrnined frorn the ex perimental data. Therefore the radius of the cylinders is 46 A and consequently the conformation of the PS blockis between all-trans and random coil. A

cylindrical structure is norrnally observed for linear block copolyrners that possess a volurne

fraction of the apolar blockof ca. 0.7-0.8.29 However, in the case of PS-dendr-(COOH)4, the

rnalar fraction of the PS block (/ps) is nearly 0.90 (see Table 4.2). At this high value, an

ordinary block copolyrner with a low degree of polyrnerization is near the order-disorder

transition, and a spherical rnorphology is expected to appear. Hence in this case, the x value

must be quite high, resulting in a large xN parameter. Additionally, due to the branched forrn

of the dendritic block, the interface is forced to curve towards the PS side supporting the

formation of well-ordered hexagonally packed cylinders (hpc), instead of a cubic array of

spheres.

Table 4.2. Characterization of the morphology of the different microlattices by analysis of the SAXS * signals. fis the molar fraction of one of the blocks in the polymer.

*

Block Copolyrner fps dz/ d1 d3 I d1 Mw (g/rnol)

PS-dendr-(COOH)4 0.87 0.57 0.50 3676

PS-dendr-(COQH)8 0.75 0.50 0.25 4288

PS-dendr-(COQH)16 0.60 0.50 0.25 5320

PS-dendr-(COOH)32 0.42 0.55 7608

Probably, the molecular density of the polystyrene block in the Jattice will not be constant for the

different generation copolyrners and therefore we will use in this Chapter /Ps ( molar ft action of the PS

block in the polyrner), rather than the volume fractîon, 1/Jps.

87

Chapter4

B

A

Figure 4.9. For the same copolymer composition, there is a greater crowding on the A side of the interface for the dendritic structure that leadstoa change of curvature away from the A side.

In the SAXS profile of PS-dendr-(COOH)8, the peak sequence confirms that the microphase

separation takes place in a lamellar structure. X-ray diffraction pattems show two reflections

corresponding to the 100 and 200 signals, as expected for a one-dimensional lattice. The

molar fraction of the PS block is 0.75, which in the case of a linear block copolymer would

correspond with an hpc morphology. For PS-dendr-(COOH)8, as in the case of PS-dendr

(COOH)4, the interface PS-dendrimer probably becomes flatter than a corresponding linear

block copolymer of the same composition due to the more compact shape of the dendritic

block (see Figure 4.9). The distance found between two layers is determined at 114 Á. In this

lamellar lattice the molecules occupy a cylindrical volume19 (see Figure 4.1 0).

1 ~57

Figure 4.10. Scheml.ltic representation of PS-dendr-(COOH)x and PS-dendr-(COOH)J6 molecules on a layer.

88

Sol!d State Microphase Separation of PS-Dendrimers

The PS chains are probably in a more stretched conformation than in PS-dendr-(COOH)4 ,

adapting their form to the available space between the molecules for an optima! packing. If

we consider the PS chains in the all-trans conformation, the measured layer thickness

corresponds to chains tilted (calculated angle ca. 45°) or interdigitated within the bilayer. A

more feasible explanation is that, since the polystyrene chains are atactic, they are not able to

pack in a well-defined pattem and they are present in a glassy phase with a conformation in

between all-trans and random coil.

PS-dendr-(COOH)16 is also found to be ordered in a lamellar phase. The distance between

two layers is in this case determined at 105 A. As the volume of the dendritic part increases,

the cross sec ti on of the cylindrical space where the polymers are located will increase too, due

to the increase in diameter of the volume occupied by the dendrimer block (see Figure 4.10).

Moreover, the PS ebains will have more space to adopt a more random conformation than in

the case of PS-dendr-(COOH)s; this effect leads to a shorter distance between the planes in

the lamellar phase. Probably because of the limited conformational freedom of the dendrimer

block, the PS block is able to adopt its conformation in order to produce an optima! packing

in the lattice.

lnterestingly and contrary to that expected, the long-range spatial order is less

pronounced for the higher generations, i.e. larger molar weights of the polymer. The signals

obtained for PS-dendr-(COOH)32 are less defined than for the smaller generation dendrimers.

No clear second order peaks are visible. Since the degree of macroscopie orientation is low in

this sample, it is difficult to determine whether the morphology of the lattice is of a lamellar

or cylindrical nature. At q = 0.036 A -I a very weak signa! seems to appear. In view of the

ratio of the first and se~ond peaks being did1 "'0.55 (this signa! could correspond toa 110

reflection), the structure is proposed to be of the hpc type. For PS-dendr-(C00Hh2, the

volume of the dendrimer part is large compared with the polystyrene part and the mieropbase

separation seems to be energetically favourable, but the interface is not as well-defined as in

the case of smaller generations, and other structures, like lamellae, can not be discarded

completely.

89

Chapter4

Transmission electron microscopy

Transmission electron microscopy (TEM) bas been extensively used for characterization of

microdomain structures. To enhance contrast between mieropbases the samples were stained

with ruthenium tetraoxide vapour. Figure 4.11 shows typical TEM pictures obtained for

ultrathin sections of the films. The dark parts in the photographs correspond to dendrimer

domains, selectively stained. Both highly ordered as well as somewhat disordered regions

were observed in all samples. This could be due to sample preparation; however, the

disordered regions were clearly larger as the dendrimer generation of the sample increased.

Figure 4.11. TEM picture of stained solid films of a) PS-dendr-(COOH)4 section across the cylinders present in the lattice; b) PS-dendr-(COOH)4 section along the cylinders present in the lattice; c) PSdendr-(COOH)R. lamellar lattice and d) PS-dendr-(COOHh& lamellar lattice.

90


Fîgures 4.1Ia and 4.1 lb show two different cross sections of the same sample, PS-dendr

(COOH)4; in Figure 4.1la the plane of the section is perpendicular to the cylinders and the

hexagonal packing of the cylinders is clearly observed. From thîs picture the distance between

two eentres is ca. 8 nm, being in agreement with the spacing obtained from the position of the

I 10 peak in SAXS measurements (d2 = 43 À). In Figure 4.1 I b the cross sectional plane is

parallel to the cylinders, and the same di stance can be measured.

Figures 4.11 c and 4.11 d show the lamellar microdomains of PS-dendr-(CQOH)8 and

PS-dendr-(CQOH)16, respectively. Both samples exhibit amorphous regions between the

well-defined morphologies. The distance between the 1amellae is in both cases ca. 10 nm.

However, these values are estimated and do not account for the change in the apparent

spacing with possible orientation of the domains with viewing angle in the TEM.

Unfortunately, no TEMpicturesof PS-dendr-(COOH)32 could be made. The ultrathin

section of the film could not be obtained because of the high hydrophilicity of the polymer, so

that instead of floating in water after the process of cutting it interacted with it, sinking in the

solvent.

Spin-coated thin films

The process of self-assembly of diblock copolymers in the bulk is driven by the total free

energy minimization, with enthalpie and entropie contributions. In thin films, there are

additional contributions to the overall free energy from surface and interfacial energies of the

two blocks and the substrate. The self-assembly process in thin films can be described by

three different phenomena: morphology, surface wetting and surface topology.30

Tapping-mode AFM has been widely used for high-resolution imaging of polymer surfaces.

We examined the nanostructures and morphologies of several PS-dendr-(COOH)n spin

coated films on silicon substrates. Polymerie films were made on silicon wafers by spin

coating from chloroform solutions PS-dendr-(CQOH)4 and PS-dendr-(COOH)32 produced

non-uniform films, sec Figure 4.12. The second generation dendrimer film presenled holes on

the surface with a depth of approximately I 0 nm. The fifth generation dendrimer sample

presented a morphology of islands with rather large size polydispersity. The height of these

islands was ca. 20 nm.

91

Chapter4

~!Onm

I Onm

0.00 J.OO 2.00 ~-tm

0.00 2.50

2.50

0.00

5.00 J.liD

i30nm

Onm

Figure 4.12. Height AFM images of spin-coated films on silicon wafers. a) PS-dendr-(COOH)4,

holes are formed on the polymerie surface. The contrast covers height variations in the 0-10 nm range. b) PS-dendr-(COOH).12, islands are formed by rearrangement of the polymer on the surface. The contrast covers height variations in the 0-30 nm range.

Polymer properties such as viscosity, Tg. molecular motions, etc. are different in thin films

than in the bulk. In thin films of block copolymers, the polymer experiences additional

contributions to the free overall energy from interfacial energies of the two blocks and the

two different surfaces, substrate and air. It is well-known that in the case of thin films of

block copolymers (see Figure 4.13), islands or holes are formed on the surface when the total

thickness is different from nD, where D is the equilibrium domain thickness in the film and n

is an integer number. The height (or depth) of the islands (or holes) is equal to D. Such a

rearrangement of the polymer-air interface is a direct consequenee of the incompressibility of

the polymer which dictales that the total volume remains constant. In thin films, where the air

interface is easily deformable, dislortion of the morphology of the polymer is less

energetically favourable than rearrangement of the surface. This could be an explanation for

the heterogeneaus surface topology found in the case of spin-coated block copolymers.

Average film thickness

Figure 4.13. Schematic representation of island formation in a diblock copolymer film at equilibrium. The arrows indicate material movement that should have occurred at the surface for the is land to form. The height of the is land is equal to the domain thickness D.

92

Solld State Microphase Separation of PS-Dendrimers

4.3. Conclusions

In this chapter, we have studied the solid state properties of a novel type of hybrid dendritic

Iinear block copolymers using SAXS, TEM and XRD. PS-dendr-(NH2)n, did not show any

evidence of microphase separation for any generation, analogously to the XRD measurements

performed on dried vesicle solutions (cast films) described in Chapter 3. These results suggest

that no segregation takes place between styrene and propylene imine monomers in the bulk,

which is in agreement with theoretica! calculations of the X parameter for this system.25

Modification of the amines, by methylation or complexation with metal ions, to introduce

ionic groups in the dendrimer, presumably increases the x parameter of the system and

induces segregation of the blocks. By comparing the experimental SAXS results of the

different generations of PS-dendr-(COOH)n. it is striking that the second order peaks are

weaker for the higher generations, which is assumed to be due to the poor regularity within

the films. This assumption is corroborated by transmission electron microscopy techniques.

This effect is probably caused by interpenetration of the PS in the poly(propylene imine)

dendrimer phases, confirming the non-ionic character of the acid-functionalized

poly(propylene imine) dendrimer block. SAXS and XRD experiments performed with

primary amine functionalized dendrimers demonstrated that only upon introduetion of

charges in the dendrimer, the amine-functionalized dendrimer is incompatible with

polystyrene. Consequently, the microphase separation observed for acid functionalized

dendrimers is more likely to be due to the segregation of carboxylic endgroups from the PS

phase than to incompatibility effects between the PS block and the tertiary amines of the

dendrimer block. The degree of polymerization NA increases with the different generations

dendrimer, but the larger the dendrimer block becomes, the Iarger also becomes the distance

between the interface and the pol ar carboxylic end groups. The interface between the different

domains is therefore not well defined, yielding a more mixed material and the material can be

regarded as a "triblock" copolymer.

By now comparing the observed morpbologies for the dendritic-Iinear polymers with

the ones expected for linear diblock copolymers, it is clear that architecture significantly

influences the microdomain structure at a given volume fraction. It is characteristic that in the

case of fps = 0.87, Iinear diblocks form spherical morphologies, while copolymer PS-dendr

(COOH)4 has a cylindrical structure of dendrimers in a PS matrix. In the case of !Ps "" 0.75, a

93

Chapter4

cylindrical structure will be expected for a linear diblock copolymer, while PS-dendr

(COOH)8 forms a lamellar structure. It has been observed before that the phase diagram in the

strong segregation limit for block copolymers of type AnB is shifted toward higher volume

fractions ofthe linear B block.10 This aggregation behaviour can be explained in terrus of the

local preferred curvature of the A-B interface as shown in Figure 4.9. The higher the

dendrimer generation, the higher the curvature towards the PS phase will be. Dendrimers are

highly branched polymers and the high degree of branching of these molecules signitïcantly

limits their conformational freedom.31 Recently, comparison experimentsof dendrimers with

their analogous linear molecules32 did show that dendrimers possess much lower

hydrodynamic volume values than linear molecules with the same molar weight, indicating

that dendrimers behave as flexible, globular molecules.

In summary, we have shown that for asymmetrie diblock copolymers based on a

dendritic block, microdomains can be formed in which the dendritic block segregation is

probably due to incompatibility effects between the carboxylic groups and the polystyrene

block. By increasing the dendrimer generation, the structures changed from hexagonal to

lamellar and the !ow-range spatial order decreased in the lattices. The phase diagram in the

strong segregation limit for dendritic-linear diblock copolymers is shifted towards higher

volume fractions of the linear block with respect to the theoretica! phase diagram, calculated

for linear block copolymers. These new macromolecules provide the necessary materials for

morphological studies that can reveal the influence of molecular architecture and chemica)

functionality on the structure of microphase-separated block copolymers.

4.4. Experimental

Materials

PS-dendr-(NHz)., PS-dendr-(N(CH3)3)n and PS-dendr-(COOH)., with n 4, 8, 16 and 32 have been

synthesized in our laboratory analogously to a literature procedure that has been publisbed in more

detail elsewhere.8·

18 Organic solvents (Biosolve, p.a.) were used without further purification. CuC}z

was purchased from Aldrich.

Preparation of samples

Solid films of polystyrene-poly(propylene imine)dendrimers with 4, 8, 16 and 32 endgroups, were

prepared by dissolving ca. 25 mg of polymer in 1 mi toluene. The solutions were allowed to dry at

94

Solid State Mk:rophase Separation of PS-Dendrimers

atmospheric pressure and later in vacuum to remove all the solvent. The resulting solid films were

stored under vacuum to prevent absorption of water. The samples were not further pressed or heated.

SAXS measurements

Small angle X-ray scattering measurements were performed using Cu Ka radiation (À = 1.54 Á), a

Riguka Denki small angle goniometer and a small-angle flat-focus camera using a point focus

colli meter. The accessible range of scattering veetors q (21t!À) sin 8 is 0.005 < qlk1 < 0.1, where 8

is the scattering angle.

Electron microscopy

Transmission electron microscopy was carried out using a JEOL 2000-FX operated at 80 kV. Thin

samples for TEM with an approximate thickness of approximately I 00 nm were cut at -122 oe. Staining was perforrned by exposure of the films to Ru04 vapour.

AFM measurements

Polymerie films were made on silicon wafers [100] by spin coating from 1.3 mM chloroform

solutions of polystyrene-dendrimer copolymers of the second and fifth generation, (0.32% and 0.62%

polymer mass fraction for PS-dendr-(COOH)n, with n 4 and 32, respectively). The images were

taken at the Instituto de Microelectrónica de Madrid (Centro Superior de Investigaciones Cientfficas)

with a Multimode Nanoscope lil (Digital Instruments) operated in the tapping mode at a resonance

frequency of 300-350 kHz. The measurements were performed under ambient conditions using

Nanosensors Si cantilevers with a spring constant of 35-45 Nlm. The tip radius was 20-50 nm.

4.5. References

Schlüter, A.-D. Synthesis of Polymers. Vol. 20 of Materials Scie1u:e and Technology: A comprehensive

treatment; Wiley-VCH: Weinheim; New York; Chichester; Brisbane; Singapore; Toronto, 1999.

2 (a) Pitsikalis, M.; Pispas, S.; Mays, J.W.; Hadjichristidis, N. Adv. Polym. Sci. 1998, 135, I. (b)

Hadjichristidis, N.; Poulos, Y.; Avgeropoulos, A. MaCI·om. Symp. 1998, 132, 207. (c) Muthukumar, M.;

Ober, C.K.; Thomas, E.L. Science 1997,277, 1225.

3 Stupp, SJ. ; LeBonheur, V.; Walker, K.; Li, L.S.; Huggins, K.E.; Keser, M.; Amstutz, A. Science 1997,

276,384.

4 (a) Roovers, J.; Zhou, L.L.; Toporowski, P.M.; Van der Zwan, M.; latrou, H.; Hadjichristidis, N.

Macromolecules 1993, 26, 4324. (b) Fukui, H.; Degucji, T.; Sawamoto, M., Higashimura, T.

95

Chapter4

Macromolecules 1996, 29, 1131. (c} Floudas, G.; Pispas, S.; Hadjichristidis, N.; Pakula, T.;

Erukhimovich, I. Macromolecules, 1996,29.4142.

5 (a) Tomalia, O.A.; Dupont Durst, H. Top. Curr. Chem. 1993, 165, 193. (b) Newkome, G.R. Advances

in Dendritic Macromolecules, Vol. 1; JAI Press lnc.: London, 1994.

6 (a) Roovers, J. Macromolecules 1988,21, 1517. (b) Ma, J. Macromol. Symp. 1995, 91, 41. (c) Beinat,

S.; Schappacher, M.; Deffieux, A. Macromolecules, 1996,29, 6737.

7 Chapman, T.M.; Hillyer, G.H.; Mahan, E.J.; Shaffer, K.A. J. Am. Chem. Soc. 1994, 116, 11195.

8 (a) van Hest. J.C.M.; Delnoye, D.A.P.; Baars, M.W.P.L.; van Genderen, M.H.P.; Meijer, E.W. Science

1995, 268, 1592. (b) Van Hest, J.C.M.; Delnoye, D.A.P.; Baars, M.W.P.L.; Elissen-Román, C.; van

Genderen, M.H.P; Meijer, E.W. Chem. Eur. J. 1996,2, 1616.

9 Bates, F.S.; Schulz, M.F.; Khandpur, A.K.; Föster, S.; Roseda1e, J.H.; Almdal, K.; Mortensen, K.

Faraday Discuss. 1994. 98, 7.

10 (a) Hadjichristidis, N.; Tselikas, Y.; Iatrou, H.; Efstratiadis, V.; Avgeropoy1os, A. J. Macrom. Sci.

Pure Appl. Chem. 1996, A33, 1447. (b) Y. Tselikas; H. Iatrou; N. Hadjichristidis; K.S. Liang; K.

Mohanty; DJ. Lohse J. Chem. Phys. 1996, 105, 2456.

11 Hogen-Esch, T.E.; Sundararajan, J; Toreki, W. Makromol. Chem., Macromol. Symp. 1991, 47, 23.

12 (a) Cloutet, E.; Fillaut, J.L.; Gnanou, Y.; Astruc, D. Chem. Comm. 1996, 2047. (b) Gnanou, Y. J.

Macrom. Sci.-Reviews Macrom. Chem. Phys. 1996, C36, 77. (c) Heroguez, V.; Gnanou, Y.; Fontanille,

M. Macromolecules 1997,30,4791.

13 (a) Venkateshwaran, L.N.; York, G.A.; De Porter, C.D.; MeGrath, J.E.; Wilkes, G.L. Polymer 1992, 33,

2277. (b) Pineri, M.; Eisenberg, A. Structure and Properties of lonomers; NATO Advanced Study

Institute Series 198; D. Reidel Publishing Co.: Dordrecht, Holland, 1987. (c) Gauthier, S.; Eisenberg,

A. Macromolecules 1987, 20, 760.

14 Zhong, X.F.; Eisenberg, A. Macromolecules 1994, 27, 1751. (b) Zhong, X.F.; Eisenberg, A.

Macromolecules 1994,27,4914.

15 (a) Newkome, G.R.; Lin, X.; Yaxiong, C; Escamilla, G H. J. Org. Chem. 1993, 58, 3123. (b)

Newkome, G.R.; Baker, G.R.; Saunders, M.J.; Russo, P.S.; Gupta, V.K.; Yao, Z.Q.; Miller, J.E.;

Bouillon, KJ. Chem. Soc., Chem. Commun. 1986, 752. (e) Newkome, G.R.; Yao, Z.Q.; Baker, G.R.;

Gupta, V.K.; Russo, P.S.; Saunders, M.J. J. Am. Chem. Soc. 1986, 108, 849.

16 (a) Hawker, C.J.; Wooley, K.L.; Fréchet, J.M.J. J. Chem. Soc .. Perkin Trans. 1 1993, 1287. (b) Gitsov,

I.; Wooley, K.L.; Hawker, CJ.; Ivanova, P.T.; Fréchet, J.M.J. Macromolecules 1993, 26, 5621. (c)

Gilsov, I.; Fréchet, J.M.J. Macromolecules 1994, 27, 7309. (d) Leduc, M.R.; Hawker, C.J.; Dao, J.;

Fréchet, J.M.J. J. Am. Chem Soc. 1996, 118, 11111.

17 Iyer, J; Fleming, K.; Hammond, P. Macromolecules 1998,31, 8757.

18 (a) van Hest, J.C.M.; Baars, M.W.PL; Elissen-Román, C.; van Genderen, M.H.P.; Meijer, E.W.

Macromolecules, 1995, 28, 6689. (b) van Hes!, J.C.M. New Molecular Architectures based on

Dendrimers. PhD Thesis, Eindhoven University of Technology 1996.

96


19 lsraelachvelli, J.N.; Mitchell, D.J.; Niharn, B.W. J. Chem. Soc. Faraday Trans. 1976, 72, 1525.

20 Brown, R.A.; Masters, A.J.; Price, C.; Yuan, X.F. Compr. Polym. Sci. 1989, 2, 155.

21 (a) Flory, P.J. J. Chem. Phys. 1942, 10, 51. (b) Huggins, M. J. Phys. Chem. 1942,46, 151. (c) Huggins,

M. J. Am. Chem. Soc. 1942,64, 1721.

22 Leibler, L. Macromolecules 1980, 13, 1602.

23 (a) Lodge, T. Mikrochim. Acta 1994, 116, L (b) Brown, RA; Masters, A.J.; Price, C.; Yuan, X.F.

Compren. (Adv.) Polym. Sd. 1989, 2, 155.

24 Meier, DJ. Block Copulymers. Science and Technology., Harwood Academie Publishers: London, New

York, 1983.

25 Dr. H.R Fischer, personal communications.

26 de Brabander-van den Berg, E.M.M.; Meijer, E.W. Angew. Chem., lm. Ed. Engl. 1993,32, 1308.

27 Oster, G.; Riley, D.P. Acta Crystallogr. 1973, 6, 81.

28 Keiler, A.; Olugosz, J.; Folkes, M.l; Pedemonte, E.; Scalisa, F.P.; Willrnounth, F.M. J. Phys. (Orsay,

Fr.) 1971, 32, 295.

29 (a) Bates, F.S. Science, 1991, 251, 898. (b) Leibler, L. Macromolecules, 1980, 13, 1602. (c)

Fredrickson, G.H.; Helfand, E. J. Chem. Phys. 1987,87,697.

30 Koneripalli, N.; Singh, N.; Levicky, R.; Bates, F.S., Gallagher, P.D.; Satija, SK Macromolecules

1995, 28, 2897.

31 (a) de Brabander, E.M.M.; Brackrnan, J.; Muré-Mak, M.; de Man, H.; Hogeweg, M.; Keulen, J.;

Scherrenberg, R.; Coussens, B.; Mengerink, Y.; van der Wal, Sj. Macromol. Symp. 19%, 102, 9. (b)

Scherrenberg, R.; Coussens, B.; van Vliet, P.; Edouard, G.; Brackrnan, J.; de Brabander, E.M.M.;

Mortensen, K. Macromolecules 1998,31.456.

32 Hawker, CJ.; Malrnstrom, E.E.; Frank, C.W.; Kampf, J.P. J. Am. Chem. Soc. 1997, 119, 9903.

97

5 Self-Assembly of Alkyl-Modified Dendrimers

Abstract

':At each level o/ organization, types o/ behatiour open up which are entire!J new and

basical!J unpredictable from a detailed ana!Jsis o/ the entities which make up the

higher level .rystems. "

P. W. Anderson

The self-assembly of amphiphilic dendrimers, basedon poly(propylene imine) dendrimers of

jive different generations modified with long aliphatic chains, is described in this chapter.

Stabie monolayers we re formed at the air/water intelface in which the dendritic suifactants

presumably adopt a cylindrical shape; all the chains are aligned perpendicularly to the water

suiface and the dendritic poly(propylene imine) care faces the aqueous phase. Electron

microscopy and dynamic light scattering measurements peiformed on acidic aqueous

solutions of the amphiphiles, showed the formation of vesicles with diameters varying

between 20 and 800 nm. X-ray dif.fraction of cast films of these aggregates revealed bilayer

thicknesses of 48-54 A, while a phase transition was detected by using 2 different

luminescence techniques. The theoretica[ volumes of the dendritic amphiphi/es we re in good

agreement with those calculated from the monolayer experiments and X-ray dif.fraction.

Hence, the amphiphilic dendrimers within the aggregates in salution have the same highly

asymmetrie conformation as proposed for the air/water inteiface. Calculations showed that

the dendritic poly(propylene imine) care in the aggregates is distorted and that the axial ratio

(rh : ra) ranges from 1 : 2.5 for the first generation to approximately 1 : 8 for the three

highest generations dendrimer. This behaviour introduces a new concept conceming the

flexibility of the poly(propylene imine) dendrimers and their ability to farm self-assembled

structures.

5.1. Introduetion

The construction of supramolecular architectures having well-defined shapes and dimensions

by the self-assembly of molecules is a topic of great current interest.1•

2'

3•

4'

5 Dendrimers are

attractive building blocks to forrn such rnatenals because they are well-defined both in

molecular weight and in architecture.6•

7 Furtherrnore, due to their catalytic,8 binding,9 and

Chapter5

optica! properties, 10 supramolecular assemblies of this type of molecules have potential

applications in the fieldsof drug delivery, chemica! sensors and photosensitive materials.

The assembly of dendrimers in monolayers or multilayers on solid surfaces has been

discussed in several studies. These studies were performed simultaneously with our studies

discussed in this chapter and all the results were disclosed independently in the last two to

three years. A survey of the different structures proposed for dendrimers at mono- and

multilayers is given in Figure 5.1. Sheiko et al. used SFM to visualize monolayers of

carbosilane dendrimers modified with hydroxyl end groups. A complete wetting of the

substrate was observed due to the adsorption of the hydroxyl groups to the mica surface. 11

The thicknesses found for these monolayers were approximately half the expected

(theoretical) values. These results point to strong deformation of the surface-bound

dendrimers (Figure 5.1 a).

a b

c d

e f

Figure 5.1. Schematic representation of different modes of ".flattened" dendrimers; a) adsorbed dendrimers with suiface-interacting end groups; b) interacting multilayer dendrimer films; c) mixed monolayer; d) proposed modelfor the organization ofmesogenic poly(propylene imine) dendrimers into smectic layered phases; e) compressed dendrimer Langmuir bilayer; f) Langmuir monolayer of an alkyl modified dendrimer as discussed in this chapter.

100

Self-Assembly of Alkyi-Modif;ed Dendrimers

Tsukruk et al. have observed the deformation of PAMAMs in monolayers on silicon

surfaces. 12•

13 The PAMAMs are collapsed and highly compressed, resulting in flattened, disc

like structures. The observed deformation can be explained by electrastatic interactions

between the terminal cationic functional groups (-NH3+) and the activaled (negatively

charged) substrate. In the same way, monolayers of carboxylated PAMAMs are flattened on

positively charged surfaces. 14'

15 In the case of multilayer films of oppositely charged

PAMAMs (-NH3+ and -C02 endgroups) compression of the dendrimers due to electrastatic

interactions between the layers is observed as wel! (Figure 5.1 b ). 13

Crooks et a/. 16'

17 performed interesting experiments using mixed monolayers of

dendrimers and alkanethiols on gold substrates. PAMAM dendrimers adsorbed on gold were

flattened due to multiple Au-amine interactions. However, when alkanethiols were added to

the dendrimer monolayer, a mixed monolayer is formed on the gold surface in which the

PAMAMs adopt a prolate contiguration due the stronger thiol-Au interaction when compared

to the amine-Au interaction18 (Figure 5.lc).

The deformation of dendrimers has been predieled by Mansfield19 in a Monte Carlo

study, considering the adsorption of dendrimers on a surface at different interaction strengths.

The calculations show a flattening of the dendrimer shape with increasing adsorption

strengths. As depicted in the 'phase diagram' in Figure 5.2, the mode of adsorption of the

dendrimers is dependent on adsorption strength and on the generation number (higher

generation dendrimers have more interaction sites per molecule and, therefore, a better chance

to be adsorbed).

strong adsorption

desorption

Generation

Figure 5.2. 'Phase diagram' of dendrimer behaviour in adsorbed monolayers and the corresponding dendrimer shapes, according to Mansfield. 19

101

Chapter5

Several groups have reported that dendrimers can be deformed to "pancake-shaped"

structures when LC-behaviour is induced by modifying the outer functionalities with

mesogenic groups.20 A study by Baars et al. on poly(propylene imine) showed that

dendrimers modified with alkoxy-cyanobiphenyl moieties form smectic phases?1 Since the

calculated SA-Iayer spacings are independent of dendrimer generation, a completely distorted

dendrimer conformation is proposed. X-ray analysis confirmed the structure given in Figure

5.ld.

The assembly of dendritic molecules on the air/water interface has been investigated

by several authors. Saville et al. have investigated polyether wedges with a benzylic alcohol

function at the core.22 The isotherms of larger generations indicate the formation of stabie

monolayers. Campression of the fourth generation polyether dendrimer results in the

formation of a stabie bilayer. In this bilayer, the dendrimers are compressed laterally,

presenting an ellipsoid shape which is twice as high as broad (Figure 5.le). Neutron

reflectivity studies on analogues with perdeuterated end groups indicate that the terminal

benzyl groups are located at the top of the lower layer. These studies confirm that

experimental conditions can induce asymmetry in the shape of dendrimers.

Amphiphilic P AMAM dendrimers have been studied on the air/water surface by

Tomalia et a/.23 The PAMAMs with aliphatic end groups of varying lengths (6, 8, 10, and 12

carbon atoms) also display the linear behaviour between the molecular area at the compressed

state and the number of end groups per molecule. Tomalia et al. explain their findings in a

model in which the lower generations are asymmetrie, while the higher generations act as

hydrophobic spheroids floating on the air/water interface. Since no indication for the latter

behaviour is found, and based on our results as presenled in this Chapter, it is proposed here

that the amphiphilic PAMAM dendrimers of high generation are highly distorted with all

aliphatic end groups pointing upwards.

The preparation of dendritic inverted unimolecular micelles as a new class of

macromolecular structures has been previously reported.24 These macromolecules are based

on hydrophilic poly(propylene imine) dendrimers (DAB-dendr-(NH2)0 , with n = 4-64) in

which all the primary amines of generation 1-5, respectively, are terminated with

hydrophobic alkyl ebains (Figure 5.3). These and related compounds were able to encapsulate

guest molecules and could be used as very effective extractants in liquid-liquid

extractions.Z5• 26 Due to their amphiphilic nature, i.e. a hydrophilic poly(propylene imine)

102

Self-Assembly of Alkyi-Modified Dendrimers

core and a hydrophobic hydrocarbon shell, this type of dendrimers is a suitable building block

for the construction of supramolecular architectures in solution and at the air/water interface.

In this chapter, the self-assembly of these amphiphilic dendrimers will be reported. It became

clear during this study that these macromolecules were able to alter their conformation

completely.

Figure 5.3. DAB-dendr-(NHCO-(CH2)u-CH.i)64·

The structure changes from a globular inverted micellar arrangement to a cylindrical

amphoteric shape, in which the dendeitic poly(propylene imine) part acts as apolar headgroup

and the alkyl chains are packed together in a parallel fashion, forming a hydrophobic unit.

This observation represents an unconventional view concerning the conformational freedom

of dendrimers. The high ratios found (1 : 7) for the minor and major radii of the dendritic part

of the molecule support the findings by others of dendrimers at interfaces, however in our

case the highly distorted conformation is not the result of external stimuli, but of self

assembly ?7

103

Chapter5

5.2. Mieropbase separation of palmitoyl functionalized dendrimers

The palmitoyl functionalized dendrimers of all generations, except the first, are

semicrystalline solids. The phase transition as determined with DSC is independent of

generation, being ca. 70 °C.24b The microscopie structures of palmitoyl functionalized

poly(propylene imine) dendrimers were examined by transmission electron microscopy.

Figure 5.4 shows typical TEM pictures obtained for the ultrathin sections of the films of

DAB-dendr-(NHCO-(CH2)14-CH3) 0 , with n = 16 and 64, stained with ruthenium tetraoxide.

The dark parts in the photographs probably correspond to the dendrimer region. Large

domains of highly ordered regions could be observed in all samples, indicating a long-range

microphase separation in the films.

Figure 5.4. TEMpicturesof the solid films of a) DAB-dendr-(NHCO-(CH2hrCH1h6 and b) DABdendr-(NHCO-(CHz)14-CH3)M

Figures 5.4a and 5.4b show the lamellar micro-dornains of DAB-dendr-(NHCO-(CH2) 14-

CH3)16 and DAB-dendr-(NHCO-(CH2)14-CH3)64 , respectively. The interlamellar distancc is

in both cases ca. 5 nm. This value was corroborated by XRD measurements on cast films

prepared from aqueous solutions (vide infra). Morphological studies of DAB-dendr-(NHCO

(CH2)14-CH3)n in thc solid phasc have been performed using SANS techniques.28 Neutron

scattering also confirmed the formation of lamellar Iattices of palmitoyl functionalizcd

dendrimers in thc bulk.

104


Mieropbase separation of carbosilane dendrimers modified with perfluorinated ebains

has been recently reported by Stühn and coworkers?9 The molecules synthesized by these

authors consisted of a highly flexible carbosilane core, which is unable to crystallize due to

the branched structure of the dendrimer, and perfluorohexyl endgroups. The endgroups are not

miscible with the core and may be considered as rigid rods. The strong tendency of the

perfluorinated endgroups to form layers caused deformation of the dendritic core, leading to

mieropbase separation in the bulk phase. In the case of palmitoyl modified poly(propylene

imine) dendrimers, the endgroups consist of flexible aliphatic ebains instead of stiff

perfluorinated groups. The mieropbase separation is not caused by strong segregation farces.

Presumably, this conformation is driven by van der Waals interactions between hydrocarbon

chains, which are able to produce lamellar structures as the ones proposed in Figurc 5.5. This

proposal for the structure of the morphology in the solid state also explains the DSC data of

the thermal behaviour of alkyl modified dendrimers reported by Jan van Hest. The phase

transition temperature of DAB-dendr-(NHCO-(CHz)m-CH3) 0 with m = 8-14 was investigated

and it was found to continuously increase with increasing chain length, independently of the

number of endgroups.24b The melting of these semi-crystalline matcrials is not generation

dependent, but alkyl chain length dependent

Figure 5.5. Schematic representation of two possible organizations of DAB-dendr-(NHCO-(CH2hr CH3)" molecules in the solid phase. On the left: half of the palmitoyl chains of one molecule are directed up and the other half down. On the right: all the chains of one molecule are directed towards the same side (up or down).

105

Chapter5

5.3. Aggregation behaviour of amphiphilic dendrimers at the interface

5.3.1. Monolayers at the air/water interface

The self-assembly of the different generation dendrimers at the air/water interface was

investigated by the Langmuir technique. The results are shown in Figure 5.6. The isotherms

of palmitoyl modified dendrimers display a sharp increase of the surface pressure upon

compression, which is indicative for the formation of stabie monolayers. This stability was

further supported by performing experiments at a constant pressure (20 mN/m), in which no

decreasein area per molecule was observed after 12 hours. Decompression isotherms of the

monolayers showed irreversibility due to the formation of aggregates on the surface of the

platinum plate when the barrier was moved backwards.

70

60

50

40

Ê z 30 E '-'

100 150 200 250 300 350 t::

20 Area(cm)

JO

0

0 500 1000 1500 2000 2500 3000

Area/Molecule (Á2)

Figure 5.6. Campression isotherms of different generation palmitoyl functionalized poly(propylene imine) dendrimers on a water surface at 25 °C. Campression and decompression isotherms of the fourth generation dendrimer are given in the inset. Barrier Speed== 50 mm2/s.

Brewster Angle Microscopy (BAM) revealed the formation of pre-aggregates on the water

surface before compression. The molecules are not uniformly distributed when they are

106

Self-Assembly of A/kyi-Modified Dendrimers

spread out on the water smface, probably due to van der Waals interactions between the

aliphatic chains.

Tabl.e S.J. Measured and theoretica! molecular area as function of the dendrimer generation.

Dendrimer Generation 2 3 4 5

Experimental area a (Á 2) 107 214 442 814 1600

Theoretica! areab (Á2) 100 200 400 800 1600

a)Determined by extrapolation of the steep rise in surface pressure to zero pressure. b) The area of a palmitoyl chain, in an all-trans confonnation, is 25 Á 2•

2000

Dendrimer End Groups

Figure 5.7. Experimental molecular area of palmitoyl modified dendrimers as a function of dendrimer generation on the water sutface. The straight line represems theoretica[ values, calculated by taking n times (n = number of dendrimer end groups) the cross-section area of the end groups (25 tF for palmitoyl chains).

The experimental area per molecule for the various dendrimers was estimated by

extrapolation of the steep rise in surface pressure to zero pressure?2b The molecular area for

every generation palmitoyl dendrimer shows a linear increase with the number of endgroups

ofthe dendrimer (Figure 5.7, Table 5.1). Assuming a molecular area of 25 Á2 (a value often

found for alkylcarboxylates) for one palmitoyl chain,30 the molecular area obtained for the

different generation dendrimers corresponds precisely with n x 25 Á2 (n = 4, 8, 16, 32, and

64).

107

Chapter5

The results obtained for our dendritic amphiphiles at the air/water interface can be

explained by assuming one single model. All the generations have presumably an alignment

in which the hydrophilic dendritic poly(propylene imine) core is pointing to the aqueous

phase and in which all the hydrophobic tails attached to the dendritic core are oriented in a

parallel fashion perpendicular to the water surface. The poly(propylene imine) core

presumably has a nearly flat conformation because in that case all the attached ebains count

for the observed molecular area (Figure 5.8) and as a consequence, a linear increase with the

number of alkyl ebains attached to the different generations dendrimer is observed.

Air

Water

Air

Water

Figure 5.8. Schematic representation of the organization of amphiphilic dendrimers in a monolayer on the water surface.

In their independent study, Tomalia et al. used hydrophobically modified

poly(amidoamine) (PAMAM) dendrimers at the air/water interface. Their behaviour was

descrîbed by two models assuming that higher generation dendrîmers are spherical.23 For the

lower generation dendrimers, it was proposed that the hydrophilic dendrîmer interior interacts

with the aqueous subphase while the hydrophobically modified terminal end groups

reorganize to extend outwards away from the water surface. The higher generation

dendrimers were proposed to act like hydrophobic spheroids floating at the air/water

interface. However, a critica! evaluation of their results shows that the PAMAMs can be

described using one asymmetrie model, the one described for our dendrimers as wel!.

108

Self-Assembly of Alky/-Modified Dendrimers


Amphiphilic dendrimers offer great potenrial as interfacial modification agents because of

their polymerie nature and high content of functional end-groups. The formation of layers

based on dendritic molecules is of particular interest because of the possibility to introduce

large changes in the surface properties of different substrates (wettability, adhesion, optica)

properties, etc.) with a small amount of materiaL It was possible to transfer monolayers of the

amphiphilic dendrimers to hydrophilic glass substrates. Z-type deposition (deposition takes

place during upstroke dipping only) was observed with a transfer ratio between 0.8-1.0 (5

layers were deposited).

25

Figure 5.9. Transfer ofthefifth generation palmitoyl dendrimer onto a glass plate.

Monolayers of alkyl-functionalized dendrimers can be also transferred to a silicon

oxide substrate with high transfer efficiency. An LB film was prepared out of a monolayer of

DAB-dendr-(NHCO-(CH2) 14-CH3) 64 on water. The thickness of the LB film was measured

by ellipsometry and found to be ca. 30 A, which is a value in very good agreement with the

calculated height for monolayers ha ving the alkyl ebains in an all-trans conformation directed

towards the air (26 À) and with the experimental values of 27 A found by X-ray diffraction

on cast films of DAB-dendr-(NHCO-(CH2) 14-CH3) 64 (vide infra).Z7 This value is close to, but

109

Chapter5

somewhat larger, than the 25 A found for the single layer of the solid phase (see 5.2).

Deposition experiments were performed at a constant surface pressure of 25 mN/m (collapse

takes place on the water surface upon compression at 50 mN/m). The silicon wafer (9 x 5

mm) was dipped 5 mm in the subphase at a constant rate of 2 mm2/s. The films obtained by

this procedure were smooth and homogeneous at llll scale. AFM images of the monolayer

show the presence of small holes on the surface (at 25 mN/m, 15% of the surface is still free).

b)

~~-----------------------------,

~~·-/'\. § o~----~------f1--------~----,H

~-r-J ~+-----~-----.-----T----~----~ . 0 100 200 300 400

nm

0

Figure 5.10. a) Height AFM image of an LB monolayer of DAB-dendr-(NHC0-{CH2) 14-CH3)M in tapping mode. The formation of preaggregates is clearly shown in this picture, probably due to van der Waals interactions between the aliphatic chains. This effect has been previously observed by Brewster Angle Microscopy techniques on monolayers on water. b) Transversal section of the 3-D AFM image. The monolayer height is 20 A, probably due to tilring of the palmitoyl chains.

These results indicate that at the beginning of the rise in the isotherm (surface pressure vs

surface per molecule), the molecules are probably not uniformly distributed on the water

surface, but form clusters. The alkyl ebains are probably not yet completely stretched

perpendicularly to the water substrate, but tilted to some extent.

5.4. Aggregation behaviour of amphiphilic dendrimers in solution

When the amphiphilic dendrimers were dispersed in water (pH < 9), opalescent solutions

were obtained. Dispersions of the first and second generation dendritic amphiphiles

precipitated within one day, whereas the higher generation aggregates were stabie for several

weeks as was judged by measuring the time dependent change in turbidity at A= 450 nm.31

IlO

Se!f-Assembly of Alkyi-Modified Dendrimers

Several techniques were used in order to obtain information about the molecular organization

and the dynamics of palmitoyl dendrimers in the aggregates and these wil! be discussed

below.

5.4.1. Yesiele properties

Electron microscopy

Four different visualization techniques have been used to investigate the aggregation

behaviour of palmitoyl dendrimers in acidic aqueous solutions. TEM pictures of DAB-dendr

(NHCO-(CH2)14-CH3)n obtained with the Pt-shadowing technique displayed the presence of

spherical aggregates with diameters of 20-140 nm (Table 5.2, Figure 5.11). The formation of

clusters was commonly observed, probably due to the preparation method, which involved

draining and evaporation of solvent. Using uranyl acetate for negative staining in the

preparation of the samples of the 51b generation dendrimers, the micrographs revealed the

presence of spherical particles that resembied vesicles. Cluster formation was also observed

by using this second sample preparation technique. It has been argued before that these

techniques can not be used to identify aggregates in aqueous solution because they are

accompanied by dehydration of the solutes, so the native specimen can be modified before it

is subjected to imaging with the electron microscope. Therefore, we decided to perform cryo

TEM measurements. It is well known that cryo-electron microscopy is the most reliable

technique to observe colloids in their natura! hydrated state, because it provides an

unimpaired vision of the individual aggregates in solution.32 Cryo-TEM confirmed the

presence of spherical aggregates in solution (Figure 5.11). The contrast was high, probably

due to their polymerie nature which yields a dense materiaL The material was very sensitive

to the electron beam, making it difficult to make the micrographs without destruction of the

sample. No clear indications were found that permit us to distinguish between vesicles, disks

or other types of aggregates with spherical shape using cryo-TEM.

lll

Chapter5

-IOOnm

Figure 5.11. Electron microscopy pictures of 10-5 M disperslons of the fifth generation palmitoyl modified dendrimer in 0.1 N HCl aqueous solutions: (A) TEM picture; sample shaded with Pt. (B) TEM picture; sample stained with uranyl acetate.(C) and (D) Cryo-electron photographs. (E) and (F) Scanning electron microscopy (SEM) pictures.

ll2

Self-Assembly of A/ky/-Modified Dendrimers

The partiele size observed was about 60-130 nm, which is in the same order of magnitude as

found with the previous two techniques. Striations could be observed with a thickness of ca. 5

nm. This thickness is comparable with the distances found by X-ray diffraction experiments

carried out on cast films of acidic aqueous dispersions of palmitoyl dendrimers and the layer

thickness of the dendrimers in the bulk as measured by TEM, and presumably corresponds to

bilayer aggregates (vide infra). The presence of larger aggregates could also be possible.

Particles with diameters exceeding 200 nm are not visible with cryo-TEM as they are

excluded from the very thin film in which the sample is observed (not thicker than 200 nm).

When the samples were not sonicated, Jarger spherical aggregates could be observed using

scanning electron microscopy techniques (SEM). The maximum vesicle diameter detected

was in the range of 800 nm. The possibility to observe these spherical aggregates by SEM

gives us the advantage to obtain 3 dimensional pictures, and therefore to distinguish between

spheres and disks, excluding the Jatter.

Dynamic light scattering

The partiele size distributions of aggregate solutions of DAB-dendr-(NHCO-(CH2) 14-CH3)n

were also determined with dynamic light scattering (DLS). The average partiele diameter

value for palmitoyl modified dendrimers of generations 3 and 4 was 340 nm with a relatively

broad distribution (average width 40 nm). These results indicate the presence of clusters in

solution at rather high concentrations (> 10-5 M) which could be slightly minimized upon

dilution.

Table 5.2. Experimental datafor the vesicle size ofpalmitoyl modified dendrimers after sonication.

Dendrimer Generation 3 4 5

(TEM) Size Vesicles (nml _c 35-200 35-130 20-140

(DLS) Size Vesicles (nm)d 330 340 350 330 160

a) Measured immediately after preparation due to theîr instabîlity. b) Partiele size obtained by TEM. c) The size of aggregates of the first generation dendrirner could not be estimated due to the high degree of cluster formation. d) Partiele size rneasured by dynamic light scattering. The rnean size indicates a high degree of cluster formation in solution. DLS samples were filtraled with I IJ.m polysulphonate filters before the measurement in order to eliminale dust particles in solution.

113

Chapter5

X-ray diffraction

X-ray diffraction was carried out on cast films of acidic aqueous dispersions of palmitoyl

modified dendrimers dried on a silicon plate in vacuum (Figure 5.12). The diffraction patterns

displayed clear periodicities between 4.8 and 5.4 nm for the different generations, with in

some cases additional periodicities (Table 5.3). These distances correspond very well with the

striations of 5 nm observed by cryo-TEM (vide supra).

Table 5.3. Experimental X-Ray datafor the different generations palmitoyl modified dendrimers.

Dendrimer Generation 2 3 4 5

Bilayer Thickness (nm)a 5.1 5.1 (4.1) 4.8 (4.2) 5.7 (5.0) 5.4

a) The values between parentheses are additional periodicities corresponding to lower intensity signals in the X-Ray diffraction pattems.

Since the extended molecular length of a palmitoyl chain as estimated from CPK molecular

models is approximately 2.2 nm, the observed thickness corresponds to that of a bilayer in

which the dendrimers possess a non-spherkal shape. The additional periodicities in the

diffraction patterns can be due to small domains in which the bilayers are tilted or

interdigitated.

c

c

0

El fJD ca

Cl Cl c a c c

a c D D

a

2 4

29

Figure 5.12. X-ray dijfraction curve of castfilms of DAB-dendr-(NHCO-{CH2)u-CH3)64.

ll4

6


pH dependenee of aggregation

There are two principal driving forces for the formation of the dendrimer aggregates in

aqueous solution. First, the strong tendency of the alkyl ebains to avoid contact with water

molecules; this is the hydrophobic effect. Second, the protonation of the tertiary dendritic

amines that makes the dendritic part more polar. It is not known which one of these two

effects dominates the formation of vesicles.

The protonation of the first three generation DAB-dendr-(NH2)n dendrimers bas been

stuclied by 15N-NMR.33 We can compare the dendrimer with a sphere containing different

concentric shells, where every shell represents one different generation (every shell

corresponds to the tertiary amines of the same generation). The process of protonation of a

dendrimer can he described in a stepwise mode; first the outer shell is protonated, foliowed

by the core shell; the rest of the shells are protonated sequentially in such a way that the free

energy of the system is minimized by avoiding repulsive coulombic interactions between

neighbouring positive charges. At pH values Iower than pH = 4, all the amines were

protonated. Although protonation studies have been performed on the first three primary

amine dendrimer generations only, a similar behaviour is expected for the fifth generation

dendrimer. However, there are no quantitative values for the degree of protonation of the

tertiary amines in the interior of alkyl substituted dendrimers as a function of pH. In order to

ensure the full protonation of the dendritic molecules, experiments have always been carried

out at pH = I, at which the formation of vesicles was observed in previous studies. The full

protonation of the amine interior of the dendrimer at pH values less extreme than pH = 1

should also lead to the formation of vesicles.

To obtain more information about the influence of protonation in the process of

vesicle formation, small aliquots of a stock solution of the fifth generation dendrimer in

THF/EtOH (2/1 v/v) were injected in a series of aqueous solutions with different pH values,

in a range between pH = 1 and pH = 13. The resulting aggregates were investigated using

TEM. The results are shown in Table 5.4. The fifth generation dendrimer was chosen for this

purpose because of the higher stability of the resulting vesicles, compared to other

generations.

At pH values between I and 9, TEM experiments using the Pt-shadow technique

showed the presence of particles with diameters of the same order of magnitude as the ones

observed before at pH= I (30-150 nm). At pH= 9, only partial protonation of the tertiary

115

Chapter5

amines of the dendrimer is expected, but this low degree of protonation seems to be enough

to maintain the aggregates in solution for months. At pH ;?: 10, the presence of macroscopie

irregular particles was observed only. In addition, solutions at pH value 10 and higher were

very unstable and precipitation of the polymer took place almost immediately. At pH ;?: 10,

the protonation degree is very low or there is no protonation of the tertiary amines at all, and

the polymer is too apolar to remain stabie in aqueous solution, leading to the formation of a

precipitate.

Table 5.4. Vesicle diameters of DAB-dendr-(NHCO-{CH2) 14-CH3) 64 in aqueous solutions at different pH values, as measured by TEM.

pHa 2 3 4 5 6 7 8 9

diam.b (nm) 40-170 30-170 30-140 30-120 40-190 40-120 30-120 40-250 40-170

a) At pH> 9, precipitation of DAB-dendr-(NHCO-{CH2) 14-CH3) 64 occurs. b) Minimum and maximum vesicle diameters found withEM are indicated.

Osmotic behaviour

Osmotic experiments were performed to see whether an inner aqueous compartment is

present in the vesicle. A concentratien difference between the inner cernpartment and the

surrounding fluid medium should lead to the flow of solvent from the less concentrated

region to the region with a higher concentration until the equilibrium is reached, i.e. the same

concentrations are present in both regions. If the alkyl bilayer in the dendritic vesicles

behaves as an osmotically active membrane, the vesicle volume is expected to change due to

swelling in hypotonic media. 34 Turbidity measurements were carried out in order to study the

change in size of the DAB-dendr-(NHCO-(CH2) 14-CH3) 64 vesicles when different

concentrations of sucrose are present inside and outside the vesicle. 35 Sucrose was chosen for

this experiment due to its high solubility in water. The fifth generation dendrimer was

injected in a sucrose aqueous solution and the resulting vesicles were dissolved in aqueous

solutions with lower concentrations of sucrose.

116

Self-Assembly of A/ky/-Modified Dendámers

6.0

,•'

.... · 5.5 .... /./

'CIC 5.0

4.5

4.o+-~.,--~.,.-~,.-~,.--,---.----,.--,

0 5 10 15 20 25 30 35

1/[S]

Figure 5.13. Unear representation of the specific turbidity as a function of the sucrose concentration outside the DAB-dendr-(NHCO-(CH2) 14-CH3) 64 vesicles.

The plot of the absorbance vs the inverse of the concentration of the solute outside the

vesicles is depicted in Figure 5.13. Results show that the sucrose concentration outside the

vesicles and the absorbance are inversely proportional, repcesenting a qualitative change in

the size of the vesicles. This is probably due to an increase in the partiele size, caused by the

permcation of solvent through the membranes. Therefore, an osmotic process presumably

took place. The experiment was repeated several times with different concentrations of

sucrose surrounding the vesicles. However, the osmotic behaviour was very difficult to

reproduce, probably due the very small extent of swelling of the vesicles or to some

experimental sourees of error, like multilamellar vesicles, dust particles or clustering of the

aggregates.

5.4.2. Critical aggregation concentranon

The critica! aggregation concentration (cac) of palmitoyl modified dendrimers was

determined with pyrene as probe molecule.36 In the presence of hydrophobic microdomains in

solution (such as micelles or vesicles), pyrene is preferentially solubilized into the interior of

the aggregates. With increasing amphiphile concentration in presence of pyrene there is an

increase in the quanturn yield of the fluorescence as well as changes in tbc vibrational fine

structure of the emission spectrum as the intensities suffer significant perturbations on going

from polar to nonpolar environments.37 Furthermore, in the excitation spectrum the (0,0)

117

Chapter5

transition band shifts from 334 to 340 nm as the pyrene goes from water to the aliphatic

cores. From these speetral changes in the fluorescence of pyrene we can directly obtain the

cac.

Tabk 5.5. Critical aggregation concentrations of DAB-dendr-(C0-(CH2) 14-CH3)", with n 4-64, expressed in moVl and in gil.

Dendrimer Generation

cac (M) · 107

cac (g/1) · 103

0.1

1.3

2

6.2

1.7

3

6.3

3.4

4

6.3

7.0

5

2.2

4.9

The cac values for DAB-dendr-(NHCO-(CH2) 14-CH3)0 ), with n 4-64 obtained from pyrene

fluorescence spectra are shown in Table 5.5. They were all very low (a factorlO--'~ in M and

10-3 in g/1, when compared with low molecular weight surfactants as CTAB), as expected for

amphiphilic polymers. The differences between the various generations do not follow an

obvious trend. It has to be pointed out that the aggregates of the first and second generation

dendrimers were much less stabie and they formed precipitates after a few minutes following

injection in water.

5.4.3. Phase transition temperature

Fluorescence depolarization38 and pyrene excimer formation39 were used to study the phase

transition temperature of palmitoyl functionalized dendrimers. For the fluorescence

depolarization diphenylhexatriene (DPH) was dissolved in an aqueous vesicle solution of

DAB-dendr-(NHCO-(CHzh4CH3)n, with n = 16, 32, and 64, at pH = 1 by the ethanol

injection method. Very high anisotropy values (r = 0.25) were found at room temperature for

palmitoyl modified dendrimer solutions of the 3rd, 41h and 5th generation, reflecting a high

microviscosity in the bilayer at room temperature. At that temperature the alkyl ebains are

below their phase transition temperature, which corresponds to a state of stiff, tightly packed,

fully-extended chains. Below the phase transition temperature, DPH shows a high degree of

orientation and restricted motion. Upon increasing the temperature the anisotropy decreases

from ca. 0.25 to 0.10 in a stepwise manner (Figure 5.14). This transition is assigned to the gel

118

Se/f-Assembly of Alky/-Modified Dendrimers

to liquid-crystalline phase transition of the alkyl chains in the bilayer. The presence of a

transition temperature (Tt) indicates that the amphiphilic molecules are arranged in well

organized vesicles, where an ordered bilayer is present. The amphiphilic dendrimer can not

show a phase transition if the aggregate is in a non-organized state. The phase transitions

found were at 34, 35, and 39 oe for DAB-dendr-(NHCO-(CH2)14-CH3) 0 , with n 16, 32, and

64, respectively. The phase transition temperature increases with dendrimer generation,

indicating a lower mobility in the case of palmitoyl chains bonded to a highly distorted

dendritic headgroup. The anisotropy values for higher generations are somewhat higher

indicating a higher rigidity in the bilayer with increasing generation.

0.25

0.10

0.25

;;.-, 0. 0 ..... -0

"' ·a < 0.10

0.25

0.15

n= 16

• • •

•

n=32

n=64

0 10 20 30 40 50 60 70 80

Teq

Figure 5.14. Fluorescence anisotropy plotsjor DPH in acidic aqueous solutions. (pH= 1) of DABdendr-(NH-CO-(CH2)u-CH3),., with n = 16, 32, and 64.

ll9

Chapter5

The pyrene excimer metbod was also used to investigate the transition temperature, and to

estimate the microviscosity in the hydrocarbon bilayer. Pyrene bas been frequently used to

determine the lateral diffusion within biologica! membranes40 and also in synthetic ones.'9•

The values obtained for DAB-dendr-(NHCO-(CHz)14-CH3)n with n = 16, 32, and 64 will

allow us to compare the propenies of dendritic bilayers of each generation with each other

and with lipid membranes measured using the same technique and approximations. In all

generations, pyrene showed two emission maxima, one very sharp and clear at 394 nm,

corresponding to the monomer, and a broad peak at 468 nm, corresponding to the excimer

emission (Figure 5.15). From graphs of the excimer intensity to the monoroer intensity ratio

(L/Im) vs temperature the second order rate constant of excimer formation (ka) cao be

determined in the interval where diffusion controls the process for each temperature. From

these values, the transition temperature (T1) can be estimated (Chapter 2).

100

80

Int 60

40

20

360 400 440 480 520 550

nm

Figure 5.15. Pyrene fluorescence spectra for [pyrene] = 8.64 JO~ M, T = 25-65 oe in the hydrocarbon bilayer of DAB-dendr-(NHCO-{CH2)u-CH3)64 vesicles.

The fifth generation palmitoyl modified dendrimer was injected in water with different

contents of pyrene within the bilayers. The L/Im ratio clearly increased with temperature (see

Figure 5.16), so the processis diffusion controlled between 25 and 60 oe. At approximately

60 °C, the intensity ratio reaches a plateau at each pyrene concentration due to a deercase in

the excimer formation rate, probably as a consequence of thermal dissociation at high

temperatures (Chapter 2). Neither time nor temperature history affected the measurements,

indicating that the system was thermodynamically stable.

120

-- r= 1.4 -o- r= 1.2

--.r=LO

-'<I- r = 0. 8


, T=25cc

• T=30°C x T= 35°C • T=40"C

-+-- r=0.6 --1- r = 0.4 1/lm

• T 45°C ,. T=50"C " T= SS"C

--r=0.2 • T 60°C • T = 65° C

0.2

30 40 50 60 70

T ("C) [pyrene] x 10·12 (molecules/cm2)

Figure 5.16. Left plot: Temperafure dependenee of the ratio 1/1., for different concentrations of pyrene in the bilayer of the fifth generation palmitoyl modified dendrimer ( r = molar ratio of pyrene per dendrimer molecule). Right plot: Concent ration dependenee of the ratio 1/lm in the bilayer of the fifth generation.

The values of the second order rate constant of excimer formation were obtained from the

slopes of the lines in the plot of IJim vs concentration of pyrene in the bilayer (see Chapter 2).

By plotting ka vs temperature, a phase transition temperature of 48 oe can be obtained (see

Figure 5.17).

Teq 't x 107 (s) ka x 10~ (cm2/ s)

15.0 25 7.00 0.6 ~

'('

30 3.80 1.2 11)

"3 u 11)

35 2.20 2.2 ö 10.0 E

40 1.40 4.0 ;::;-E

45 1.00 6.3 ~ 0

50 0.80 8.9 >< 5.0 ..:.rt."

55 0.70 11.3

60 0.65 13.7 0.0

65 0.62 14.8 10 20 30 40 50 60 0 Teq

Figure 5.17.Plot of the second order reaction rate constant, kro of pyrene excimer formation in a bilayer of the fifth generation palmitoyl modified dendrimer vs T. The excimer lifetimes represented in the Tablewere obtainedfrom lirerature (ref 39).

121

Chapter5

Pyrene excimer formation within the bilayers of the fourth generation alkyl modified

dendrimer was foliowed using the sarne method. The pyrene to dendrimer ratio varied from

0.7 to 0.1, while the pyrene to palmitoyl chain ratio was the sameasin the previous case. The

IJim ratio clearly increased with temperature (see Figure 5.18), so the process is ditfusion

controlled between 25 and 55 °C. Thermal dissociation of the excimer begins to take place at

approximately 55 oe.

0.5 -11- r= 0.7

~~-•·'' 0.4 -4-- r=0.5

~ -"l-r=0.4 0.3 ~::::::::~ Jjlm

....- r 0.1 0.2

~ 0.1

20 30 40 50 60 70

T ("C)

0.5

0.4

0.3 ljlrn

0.2

0.1

0.0 2 4 6 8

(pyrene) x 10·12 (molecules/cm')

I T=25'C 11 T=30"C x T=35"C + T=40°C • T=45"C v T=50°C "' T=55'C • T=60°C • T=65'C

10

Figure 5.18. Left plot: Temperature dependenee of the ratio 1/lm for different concentrations of pyrene in the bilayer of DAB-dendr-(NHCO-(CH2)14-CHJ).12. (r = ratio pyrene per dendrimer molecule). Right plot: Pyrene concentration dependenee of the ratio!/[., in the bilayer of the fourth generation dendrimer.

10.0

~ <U 7.0 ~ <U 0 E

;;-5.0 e

!:!..-"" 0

:< .lè"' 2.0

0.0 20 30 40 50 60 70

T ("C)

Figure 5.19. Second order reaction rate constant, k"' of pyrene excimer formation vs T. The transition temperafure in the bilayer of the fourth generation palmitoyl modified dendrimer is 43 oe.

122

Self-Assembly of A/ky/-Modified Dendritners

To obtain the values of the second order excimer formation rate constant, the plots of the

excimer to monomer intensities ratio vs pyrene concentration are represented in Figure 5.18.

In this case, the linear regression fits were less accurate than in the case of the fifth generation

dendrimer (with R == 0.998), due to thermal dissociation of excimer at temperatures higher

than 55 oe. By plotting ka VS temperature (Figure 5.19), a phase transition temperature of 43

OC can be obtained for DAB-dendr-(NHeO-(eH2)14-eH3h2.

In the case of the third generation alkyl modified dendrimer, the pyrene to dendrimer ratio

varied from 0.35 to 0.05. The L/I.n ratio increased with temperature until 40 oe (see Figure

5.20). At high concentrations of pyrene (r > 0.2), there is a strong reduction in the rate of

excimer formation above the

0.4

OJ

0.2 lflm

0.1

10 20 30 40 50 60 T("C)

0.5

-•- r=0.35 -o- r=0.30

-- r=0.25 0.4

--';1- r = 0. 20

-- r=0.15 03 -+- r=O.IO -- r = 0.05 lil",

0.2

0.1

2345678 [pyrene] x J0·12 (moleeules/crn2)

• T= I5°C I T=20°C

111 T= 25° C x T=30°C + T=35°C • T=40°C v T=45°C .A T=50° C o T= 55° C • T=60"C

Figure 5.20. Temperafure dependenee of the ratio 1/lm for different concentrations of pyrene in the bilayer of DAB-dendr-(NHCO-{CH2) 14-CH3) 16. (r = ratio pyrene per dendrimer molecule). Right plot: Pyrene concentration dependenee af the ratio 1/lm in the bilayer of the third generation.

In the case of the third generation dendrimer, pyrene farms clusters at low

concentrations (r,., 0.20). Therefore DAB-dendr-(NHeO-(eH2)w-CH3)16 seems to be able to

form a rather regular crystal structure below Tt. containing only a small number of free

pockets.41'

42 For the higher generations, the solubility of pyrene in the bilayer was higher.

Ibis anornalous decrease of leflm at T1 is probably due to the low solubility of pyrene in the bilayer.

Pyrene is able to forrn two-plane crystals that can also lead to excirner formation and hence to

fluorescence ernission with the sarne speetral characteristics as the excirner forrned by birnolecular

collision. The microscopie crystallites of pyrene dissolve upon going frorn the crystalline state to the

liquid crystalline state, leading to a decrease in the excirner formation rate, and thus the intensity ratio

decreases.

123

Chapter5

Since the diameter of the vesicles has been found to be independent of dendrimer generation,

the strain present in small vesicles will be larger for the higher generations and probably

prevents the formation of a very regular crystal structure. The plot of the intensity ratio vs

pyrene concentration yields the second order excimer formation rate constant for each

temperature (see Figure 5.20). From the plot of these values versus temperature a transition

temperature of 40 oe was obtained (Figure 5.21)

K

..... 2.0

I 0 20 30 40 50 60 70 T(OC)

Figure 5.21. Secorul order reaction rate constant, k., of pyrene excimer formation vs T. The transition temperafure in the bilayer ofthe third generation palmitoyl modified dendrimer is 40 ÓC.

5.4.4. Microviscosity

Applying the diffusion model described before (see ehapter 2), the diffusion coefficients can

be calculated for the corresponding values of ka obtained by the pyrene method. By platting

the Dctîff values on a logarithmic scale as a tunetion of the temperature, the curves obtained

indicate that there is a phase transition (see Figure 5.22). The activation energies for the

diffusion of pyrene within each bilayer in the crystalline state were obtained by fitting the data

to the Arrhenius law at temperatures below the T1• There were no large differences between

dendrimer generations: Ea = 100.8 kJ/mol for the fifth, 102.5 kJ/mol for the fourth and 102.1

kJ/mol for the third generation dendrimer. From the values of ka, the values of the

microviscosities of the bilayers of the three dendrimer generations were estimated at

temperatures between 25 OC and 65 oe. The microviscosity in the bilayers decreases with

temperature until the transition temperature, after which a constant value is reached (Figure

5.22).

124

I ·e·'Ó

~ ~ 1-e-w

a I

c

c:."

2.9 3.0 3.1 3.2 3.3 3.4 3.5 3.6

T-1 • 103 (K·1)


10 0

8 a

& 6

-~ ~ 4 ii ·;;: e 4 •

.::: :2: 4

2 • t A • t t t 0

0 • 300 310 320 330 340

T (K)

Figure 5.22. Left plot: Arrhenius plot for the dijJusion coefficients in the three last dendrimer generations. Right plot: Microviscosity temperature dependenee for the bilayers of the three highest dendrimer generations . .A. DAB-dendr-(NHCO-{CH2) 14-CH3)J6 ; 0 DAB-dendr-(NHCO-{CH2) 14-

CHJ)32; e DAB-dendr-(NHCO-{CH2)u-CH3)64.

The viscosity of linear hydrocarbon liquids as well as most simpte solvents decreases

with the temperature in an exponential way, which agrees well with the empirical relation:

1J A. ellE/RT Equation 5.1

By fitting these curves, flow activation energies ~ = 75 kJ/mol, 96 kJ/mol and 100 kJ/mol

were obtained for DAB-dendr-(NHC0-(CH2) 14-CH3)0 , with n = 16, 32, and 64, respectively.

Below the phase transition (25 °C), the values found for 11 were 8, 9, and 8 P for n = 16, 32,

and 64, respectively. In the fluid state (65 oq the values found for 11 were 0.4, 0.6, and 0.8 P

for n = 16, 32, and 64, respectively. The values of microviscosity of the bilayer are in the

same order of magnitude as the viscos i ties found in fluid phospholipid biomembranes. 38• 39

•

From the tluorescence depolarization experiments, the microviscosity 11 in the bilayers can be

calculated for every temperature fora known anisotropy value (see Chapter 2). In the bilayers

formed by DAB-dendr-(NHCO-(CHût4-CH3)n, the values found for 11 (25 "C) were 4, 5, and

6 P for n 16, 32, and 64, respectively. Inthefluid state the values found for 11 (65 oq were

0.3, 0.4, and 0.6 P for n = 16, 32, and 64, respectively (Figure 5.23).

125

Chapter5

2.8 3.0 3.2 3.4 3.6

1/T x 10' (K1)

Figure 5.23. Tempera/ure dependenee of microviscosity plotted as ln 11 vs. 1/T of DAB-dendr(NHC0-{CH2)14-CH3)"' with •, n=64; 0, n=32; .Á., n=l6.

Discussion on phase transition and microviscosity

The transition temperatures obtained by means of the DPH tluorescence depolarization

technique are slightly lower than the values obtained by the pyrene excimer formation

technique (Figure 5.24). This difference can be caused by a different location of the probes

into the bilayers. Even when both probes are very hydrophobic molecules and wil! be located

in the most apolar part of the bilayers, DPH is a smaller molecule than pyrene and could be

preferentially located inside free pockets present in the bilayer.

50

45 :==; ~ 40 u 0

"-" 35 E-<

30

3 4 5

Dendrimer Generation

Figure 5.24. Biklyer phase transition temperature as a Junction of dendrimer generation, experimentally found using: --e-- pyrene excimer jonnation techniques and- 0- fluorescence depolarization techniques.

126


Nevertheless, the values obtained by each kind of experiment can be compared for the

different generations because they have been measured using the same technique and

approximations. We observe in both experiments that the transition temperature found for the

bilayer formed by palmitoyl modified dendrimers increases with the dendrimer generation.

These results suggest that the T1 of the bilayer is dependent on the conformational

freedom of the alkyl chains, which is different for every dendrimer generation probably due to

deformation of the dendritic headgroup. In the case of the higher generation, the

conformational freedom of the molecule is expected to be more restricted than in the case of

lower generation dendrimers, due to the high deformation of the dendritic part of the

molecule in the vesicles. In the gel-like state, the headgroups are probably able to form inter

and intramolecular hydrogen bonds in the amide linkage, condensing the hydrocarbonlwater

interface of the vesicle and introducing rigidity on the hydracarbon region as weiL29 The

hampered dynamics in the dendritic part of the molecule induce rigidity in the palmitoyl

chains, and presumably increase the harrier for chain melting. Additionally, vesicles

composed of dendritic amphiphiles could be compared with polymerized vesicles. Formation

of vesicles with polymerizable surfactauts has been extensively investigated due to the high

stability that polymerization introduces in the aggregates.43 Likewise, stiffness is also

introduced in the vesicular structures due to the large number of covalently bonded chains.

An increase in the dendrimer generation can beregardedas an increase in the polymerization

degree of the vesicle and therefore, stability and stiffness of the aggregates most likely

increases for high dendrimer generations.

The observed decrease in the formation of pyrene crystals at room temperature for

high generations is related to the packing of the ebains in the bilayer. The distribution of

pyrene molecules may not be random within the bilayer, since new pyrene molecules could be

easily introduced close to a previously introduced pyrene molecule, because of the access it

bas "opened" to the bilayer interior. Therefore, the conglomeration of pyrene molecules in the

same site within the bilayer may probably result in the formation of crystals because of the

proximity of one to the other. A higher probability that pyrene molecules coincide in the same

site within the bilayer apparently exists in low generations, whereas a less packed bilayer for

the higher generations may allow pyrene molecules to enter the bilayer more easîly.

Considering the lower solubility of pyrene in the bîlayer of the third generation, it can be

127

Chapter5

proposed that at room temperature, the packing density within the bilayer of the third

generation dendrimer is larger than in high generations.

5.5. Conclusions; flexible dendrimers

In this chapter it has been described how poly(propylene imine) dendrimers

functionalized with hydrophobic palmitoyl chains assembie in stabie monolayers at the

air/water interface.27 In the assemblies, the dendrimers adopt a cylindrical, amphoteric shape,

in which the ellipsoid dendritic moiety acts as a polar headgroup and the alkyl chains arrange

in a parallel fashion to form an apolar tail.

\H20 pH<9\

Figure 5.25. Schematic representation of monolayer and bilayer formation of amphiphilic dendrimers in aqueous solutions.

Upon dispersion of the amphiphilic dendrimers in aqueous solutions, the poly(propylene

imine) unit becomes protonated; this leads to a more extended conformation of the dendritic

core as a result of Coulombic repuls ion. In the aggregates, the dendrimers are thought to have

similar conformations as those observed at the air/water surface. The hydrophilic protonated

dendritic part faces the aqueous phase, while the aliphatic chains are packed in a parallel

fashion to form an apolar bilayer (Figure 5.25). Within this assumption, the degree of

128


flattening of the dendritic poly(propylene imine) part in the supramolecular assemblies can be

calculated from the experimental data by assuming that it bas the shape of an oblate spheroid,

where r. is the major radius and rb the minor radius (Figure 5.26).

54Á

Figure 5.26. Degree of flatlening ofthe poly(propylene imine) part of DAB-dendr-(NHCO-(CH2 ) 14 -

CH3)M The major radius (Table 5. 7), r"' was calculated from the molecular area (Table 5.1) found by monolayer experiments. (in the case of DAB-dendr-(NHCO-(CH2 ) 14- CH3) 64: 1600 A2 = n: · r/; ra = 22.6 A). The length of palmitoyl chains basedon molecular modeling is 22 A. The thickness of the bilayer of DAB-dendr-(NHC0-(CH1 ) 14- CH3 }64 (Table 5.3), measured by XRD, is 54 A. This value corresponds to twice the length ofthe palmitoyl chain and 3 times rb (2 x 22 + 3 rb 54 A, assuming that the two layers of amphiphilic dendrimers are positioned as illustrated and not interpenetrated) yielding a minor radius ofrb = 3.3 Afor DAB-dendr-(NHCO-(CH2 ) 14 -CH3h4·

Table 5.7. Molecular dimensions ofthe different generations dendrimer.

calc. hydrod. calc.molecular exp.molecular major radius minor radius axial ratio Dendr.

radius vol. vol. fa rb r.: rb generatio11

(Á)a (Á 3)b (Á 3)c (Á)d (Á)e

1 6.3 3000 2730 5.8 2.3 2.5: 1

2 7.7 5820 5460 8.3 2.3 3.6: 1

3 9.5 11400 10600 11.9 1.3 9: 1 i

4 I2.9 24500 20350 16.1 2.0 8 : I

5 I 16.1 48580 43200 22.6 3.3 7: I a) Dendnmer hydrodynam1c rad1us obtamed WJth molecular sJmulations m the gas phase and SANS

measurements (ref. 44). b) Theoretica! volume of dendritic ampbipbiles was calculated by ta.ldng the sum of tbe

volume of palmitoyl ebains (one cbain basedon CPK models = 484 Á3) and tbe volume of tbe dendrimer part

(based on dendrimer hydrodynamic radius). c) Obtained by multiplying half tbe bilayer thickness (X-ray

measurements, Table 5.3) witb the molecular area (monolayer experiments, Table 5.1 ). d) r, was calculated from

the molecular area by considering tbe sectional area of the dendrimer part to be a circle. e) rb was calculated by

maldng tbe approximation that tbe a.lkyl ebains are in an all-trans conformation (Figure 5.26).

129

Chapter5

The calculated values for ra and rb are given in Table 5.7. The calculations showed that the

shape of the dendrimer is distorted and that the axial ratio (rb : r.) ranges from 1 : 2.5 for the

first generation to approximately 1 : 8 for the three highest generations dendrimer. Thus, the

dendritic headgroup has a flattened, far from globular, ellipsoid shape.

Fora long time, dendrimers have been presented as spherical objects with a symmetrie

architecture. As a matter of fact, dendrimers do not always behave as might be expected from

this primary idea of globularity. Most dendrimers possess flexible branches that can adopt

different conformations, implying that the end groups can fold back into the interior of the

molecule. More surprisingly, the flexibility in dendritic molecules--even in bulky higher

generation dendrimers-allows these molecules to adopt shapes that are far from spheres.

Such shapes are only observed when dendrimers are exposed to 'extemal stimuli' or by self

assembly, i.e. secondary interactions that force dendrimers into specific supramolecular

arrays. Flattened dendritic structures can therefore be found in monolayers and in amphiphilic

molecules.

5.6. Experimental

Materials

DAB-dendr-(NHCO-(CH2)t4-CH3)n, with n = 4, 8, 16, 32 and 64, have been synthesized in our

laboratory, analogous to a literature procedure24 that has been publisbed in more detail elsewhere.

THF (Biosolve, p.a.) was distilled over Na!K/benzophenone under an argon atmosphere. Ethanol

(Biosolve, p.a.) and chloroform (Biosolve, p.a.) were used without further purification. Buffer

solutions (Merck), D(+)-sucrose (Janssen 99+%), pyrene (Aldrich, 98%) DPH (Aidrich, 98%) and

uranyl acetate (Merck) were used as received.

General Methods

UV spectra were recorded on a Perkin Elmer UV NIS/NlR Lambda 900 Spectrometer. A Branson

2210 sonication bath was used for the preparation of disperslons in acidic water. Steady-state

fluorescence spectra were run in a Perkin Elmer Luminescence spectrometer LS 50B in the right

angle geometry (90° collecting opties) using slit openings of 5 nm for emission and 2.5 nm for

excitation. Dynamic light scattering experiments were performed in a Malvern Autosizer Instrument,

5 mW laser at 633 nm and the samples were filtrated with I 11m polysulphonate filters before the

measurement in order to eliminate dust particles in solution. Low angle X-ray powder diffraction

measurements were performed with a self-made instrument (Netherlands lnstitute for Sea Research,

130

Self-Assembly of Alky/-Modified Dendrimers

NIOZ, Texel). This high accuracy e-e diffractometer is equipped with a Huber D8211 goniometer, a

Cu LFF XRD Philips tube, variabie divergence and antiscatter slits, and an energy dispersive Si/Li

Kevex detector, which enables a high peak to background ratio.

Preparation of solid films

Solid films of DAB-dendr-(NHCO-(CH2) 14-CH3) 0 , with n = 4, 16 and 32, were prepared by melting

ca. 15 mg of polymer at 130, 80 and 90 °C, respectively. The solutions were allowed to cool toroom

temperature. The films were slored under vacuum to prevent absorption of water. Transmission

electron microscopy was carried out using a JEOL 2000-FX operated at 80 kV. Thin samples for

TEM with an approximate thickness of approximately 100 nm were cut at -122 °C. Staining was

performed by exposure of the films onto Ru04 vapour.

Langmuir-Blodgett Experiments

Monolayer experiments were performed on a thermostated, home-built trough at 20 oe. The surface

pressure was measured using the Wilhelmy plate method. Plates made of platinum or filter paper

gave identical results. On the water subphase (pH = 7), 50 mi of a solution of the amphiphiles in

CHCh was spread and allowed to evaporate. In order to establish the stability of the monolayer, the

surface pressure was maintained at a constant value of 10 mN/m. The monolayer area was found to

remain constant for more than 8 hours, indicating that the monolayers are stable. The rate of

compression was 50 mm2/s. Deposition experiments were performed at a constant surface pressure of

25 mN. A glass slide of 7.4 x 1.5 cm was dipped 20 mm in the subphase at a constant speed of 4

mm2/s. Brewster Angle Microscope experiments were carried out with a NFT BAM1 instrument,

manufactured by Nano Film Technology, Goettingen. The instrument was equipped with a lOmV

He-Ne laser with a beam diameter of 0.68 mm operating at 632.8 nm. Reflections were detected

using a CCD camera.

AFM Experiments

Atomie force microscopy was done in Langmuir-Blodgett films that were prepared as has been

described before. The micrographs were measured by means of a Multimode Nanoscope III (Digital

Instruments) operated in the tapping mode at a resonance frequency of 300-350 kHz. The

measurements were performed under ambient conditions using Nanosensors Si cantilevers with a

spring constant of 35-45 N/m. The tip radius was 20-50 nm.

131

Chapter5

Preparation of V es iele Solutions

The desired amount of dendrimer was dissolved in 100 111 ethanolffHF (1/2 v/v). The resulting

solution was warmed up to ca. 55 oe and injected into 10 m1 of a buffer solution of pH = 1 (preheated

at 60 oq while stirring. The vesicle solution was placed in a bath-type sonicator and sonicated for 10

minutes at room temperature.

Electron Microscopy

Samples for transmission electron microscopy were prepared by the negative staining and Pt-shading

methods. A droplet of acidic aqueous solution (pH 1) was placed on a Cu-grid, covered with

formvar, and allowed to dry for 1 minute, after which the droplet was removed with filter paper.

Negative staining was performed by actdition of a droplet of a 1 w% uranyl acelate solution during 1

minute. Pt-shaded samples were prepared by covering the dried sample with Pt using a Balzers

Sputter unit (Pt layer thickness 2 nm). These samples were studiedusinga PhilipsTEM 201 (60 kV)

(University of Nijmegen). For cryo-electron microscopy, thin films were prepared by dipping a 700-

mesh eopper grid in the suspension and blotting with filter paper to remove the excess of liquid. The

thin films that formed between the bars of the grid were vitrifled by plunging into melting ethane.

The vitrified films were slored under liquid nitrogen and observed at -170 oe with a Philips CM 12

microscope at low dose conditions, I 20 kV (University of Maastricht). In case of palmitoyl modified

dendrimers, the material was very sensitive to the electron beam. The samples for SEM were

prepared by allowing a droplet of vesicle solution to evaporate in a carbon substrate and covering the

dried sample with a thin layer of Au. Scanning electron microscopy measurements were performed

with a JEOL JSM 840A Microscope with an acceleration voltage of 20 kV.

Preparation of Cast Bilayersfor XRD

Acidic aqueous dispersions of amphiphilic dendrimers were prepared by the ethanol injection metbod

described before. Special Si single crystal wafers, cut along the (501) plane, were used for low angle

measurements. Aliquots of the colloidal dispersions (2 ml) were left to dry on these Si wafers in a

desiccator over sodium hydroxide. The specimen ehamber was maintained at 20 oe and flusbed with

air of 20% relative humidity during the XRD measurements. The patterns were digitally recorded.

Peak positions were obtained using peak fitting JANDEL software.

Fluorescence Measurements -1

Square 1 cm quartz cells were filled with ca. 3 m1 solution, with [Py] = 4.8 x 10 M. In the case of

fluorescence emission spectra Àex = 339 nm, for excitation spectra Àem = 390 nm. All samples were

prepared by ad ding a known amount of pyrene in acetone to a series of empty 10 m1 volumetrie

132

Self-Assemb/y of Alkyi-Modified Dendrimers

flasks; after evaporation of the acetone, known amounts of a stock solution of polymer were added

and diluted with a buffer of pH = 1 in order to obtain final dendrimer concentrations between 10-4

and 1 o-9 M. The flasks were sealed, proteeled from light and stirred for ca. 20 h at room temperature

to allow the pyrene and the aggregates to equilibrate.

For the pyrene excimer formation experiments, the fluorescence spectra of pyrene were recorded for

each pyrene concentration for three different dendrimer generations. Pyrene and dendrimer were

dissolved logether in THF/ethanol and injected in warm water at pH = 1. The pyrene-to-dendrimer

ratio was kept between 2.2·1 o-2 and 3·1 o-3 pyrene molecules per palmitoyl chain, which means molar

ratios of pyrene to dendrimer of 1.40-0.19 for the fifth generation, 0.70-0.09 for the fourth and 0.35-

0.05 for the third generation. The sample was thermostated by using a thermoeouple in the sample

holder connected to an RTE 110 Neslab thermostal with an ethyleneglycol/water (l/1, v/v) bath. The

temperature was controlled by a thermometer in the sample holder connected to the computer. The

measurements were made at increasing temperature in 5 oe intervals and kept constant to an accuracy

of ±0.1 oe, between 25 and 65 "e at pH = I.

For the fluorescence anisotropy measurements, DPH was dissolved together with the polymer in an

aqueous buffer pH = I, by the ethanol injection method. The ratio of alkyl ebains attached to the

poly(propylene i mine) dendrimer toprobe was 25:1. For fluorescence emission spectra Àcx = 382 nm,

for the excitation spectra Àcm = 430 nm. In all cases the anisotropy was measured in a range of

temperature between 10 oe and 75 oe by taking steps of 5 oe and allowing the solutions to

equilibrate during 15 min.

Osmotic Experiments

A stock solution of DAB-dendr-(NH-eO-(eH2) 14eH3) 64 in water was prepared by injecting 8.202 mg

of DAB-dendr-(NH-eO-(eH2)I4eH3) 64 dissolved in 50 !ll methanoi/THF (112 v/v) into 5 ml of a 0.50

M sucrose solution (preheated at 60 oq while stirring. Small aliquots (100 lll) of this stock solution

were added to 3.25 ml of solutions containing 0.03-0.25 M sucrose. JO minutes after mixing, the

absorption was measured at 450 nm.

5.7. References

(a) Whitesides, G.M.; Mathias. J.P.; Seto, C.T. Science 1991, 254, 1312. (b) Whitesides, G.M.;

Simanek, E.E.; Mathias, J.P.; Seto, C.T.; Chin, D.N.; Mammen, M.; Gordon, D.M. Acc. Chem. Res.

1995, 28, 37.

2 Lehn, J.-M. Angew. Chem .. lnt. Ed. Engl. 1990,29, 1304.

133

Chapter5

3 Stupp, S.l., LeBonheur, V.; Walker, K.; Li. L.S.; Huggins, K.E.; Keser, M.; Amstutz, A. Science 1997,

276, 384.

4 (a) Zimmerman, S.C.; Zeng, F.; Reichert, D.E.C.; Kolotuchin, S.V. Science 1996, 27i, 1095. (b) Huck,

W.T.S.; van Veggel, F.C.J.M.; Reinhoudt, D.N. AnRew. Chem., int. Ed. EnRl. 1996, 35, 1213. (c)

Huck, W.T.S.; Hulst, R.; Timmerman, P.; van Veggel, F.C.J.M.; Reinhoudt, D.N. AnReW. Chem .. int.

Ed. Engl. 1997, 36, 1006. (d) van Hest, J.C.M.; Delnoye, D.A.P.; Baars, M.W.P.L.; van Genderen,

M.H.P.; Meijer, E.W. Science 1995. 268, 1592. (e) van Hest, J.C.M.; Delnoye, D.AP.; Baars, M.W.P.

L.; Elissen-Román, C.; van Genderen, M.H.P.; Meijer, E.W. Chem. Eur. 1. 1996,2, 1616.

5 Zeng, F.; Zimmerman, S.C. Chem. Rev. 1997, 97, 1681.

6 (a) Tomalia, D.A Adv. Mater. 1994, 6, 529. (b) Tomalia, D.A.; Naylor, A.M.; Goddard lil, W.A

Angew. Chem., Int. Ed. Engl. 1990,29, 138. (c) Hawker, C.J.; Fréchet J.M.J. 1. Am. Chem. Soc. 1990,

112, 7638. (d) Newkome, G.R; Lin, X. Macromolecules 1991, 24, 1443. (e) lssbemer, J.; Moors, R.;

Vogtle, F. Angew. Chem., int. Ed. Engl. 1994, 33, 138.

7 (a) Newkome, G.R; Yao, Z.Q.; Baker, G.R.; Gupta, V.K. J. Org. Chem. 1985, 50, 2003. (b) Tomalia,

D.A.; Berry, V.; Hall, M.; Hestrand, D. Macromolecules 1987, 20, 1164. (c) Newkome, G.R.;

Moorefield, C.N.; Baker, G.R; Saunders, M.J.; Grossman, S.H. Angew. Chem., Int. Ed. Engl. 1991,30,

1176. (d) Wooley, K.L.; Hawker, G.J.; Fréchet, J.M.J. 1. Am. Chem. Soc. 1993, JJ5, 11496. (e)

Newkome, G.R; Young, J.K.; Baker, G.R.; Potter, R.L.; Audoly, L.; Cooper, D.; Weiss, C.D.

Macromolecules 1993,26, 2394. (I) Hawker, C.J.; Wooley, K.L.; Fréchet, J.M.J. 1. Chem Soc., Perkin

Trans 1 1993, 1287.

8 (a) Bhyrappa, P.; Young, J.K.; Moore, J.S.; Suslick., K.S. 1. Am. Chem. Soc. 1996, 118, 5708. (b)

Knapen, J.WJ.; van der Made, A.W.; de Wilde, J.C.; van Leeuwen, P.W.N.M.; Wijkens, P.; Grove,

D.M.; van Koten, G. Nature 1994, 327, 659. (c) Reetz, M.T.; Lohmer, G.; Schwickardi, R Angew.

Chem., Int. Ed. Eng!. 1997,36, 1526. (d) Bolm, C.; Derrien, N.; Seger, A. Synlett. 1996, 387.

9 Jansen, J.F.G.A.; de Brabander-van den Berg, E.M.M.; Meijer, E.W. Science 1994, 226, 1226.

10 (a) Mekelburger, H.-B.; Rissanen, K.; Vogtle, F. Chem. Ber. 1993, 126, I 161. (b) Sadamoto, R.;

Tomoka, N.; Aida, T. 1. Am. Chem. Soc. 1996, JJ8, 3978

IJ Sheiko, S. S.; Muzafarov, A.M.; Winkler, R. G.; Getmanova, E.V.; Eckert, G.; Reineker, P. Langmuir

1997,13,4172.

12 Tsukruk, V. V.; Rinderspacher, F.; Bliznyuk, V. N. Langmuir 1997, 13, 2171.

13 Tsukruk, V. T. Adv. Mater. 1998, JO, 253.

14 Bliznyuk., V. N.; Rinderspacher, F.; Tsukruk, V. V. Polymer 1998,39, 5249.

15 Esumi, K.; Goino, M. Lagmuir 1998, 14,4466.

16 Zhao, M.; Tokuhisa, H.; Crooks, R. M. Angew. Chem., Int. Ed. Eng/. 1997,36,2596.

17 Tokuhisa, H.; Zhao, M.; Baker, L. A.; Phan, V. T.; Dermody, D. L.; Garcia, M. E.; Peez, R. F.; Crooks,

R. M.; Mayer, T. M. J. Am. Chem. Soc. 1998, 120, 4492.

134

Self-Assembly of Alky/-Moclified Dendrimers

18 Hierlemann, A.; Campbell, J. K.; Baker, L. A.; Crooks, R. M.; Ricco, A. J. J. Am. Chem Soc. 1998,

120,5323.

19 Mansfield, M. L. Polymer 1996, 37, 3835.

20 (a) Percec, V.; Chu, P.; Ungar, G.; Zhou, J. J. Am. Chem. Soc. 1995, 117, 11441. (b) Lorenz, K.;

Hölter, D.; Stühn, B.; Müllhaupt, R.; Frey, H. Adv. Mater. 1996,8, 414. (c) Cameron, J. H.; Facher, A.;

Lattennann, G.; Diele, S. Adv. Mater. 1997, 9, 398.

21 Baars, M. W.; Söntjens, S. H. M.; Fischer, H. M.; Peerlings, H. W. I.; Meijer, E. W. Chem. Eur. J.

1998, 4, 2456.

22 (a) Saville, P.M.; White, J. W.; Hawker, C. J.; Wooley, K. L.; Fréchet, J. M. J. J. Phys. Chem. 1993,

97, 293. (b) Saville, P.A.; Reynolds, P.A.; White, J.W.; Hawker, CJ.; Fréchet, J.MJ.; Wooley, K.L.;

Penfold, J.; Webster, J.R.P. J. Phvs. Chem. 1995, 99, 8283.

23 Sayed-Sweet, Y.; Hedstrand, D.M.; Spinder, R.; Tomalia, D.A. J. Mater. Chem. 1997, 7, 199.

24 (a) Stevelmans, S.; van Hest, J.C.M.; Jansen, J.F.G.A; van Boxtel, D.AFJ.; de Brabander-van den

Berg, E.M.M.; Meijer, E.W. J. Am. Chem. Soc. 1996, 118, 7398. (b) van Hest, J.C.M. New Molecular

Architectures basedon Dendrimers. PhD Thesis, Eindhoven University ofTechnology 1996.

25 Baars, M.W.P.L.; Froehling, P.E.; Meijer, E.W. Chem. Comm. 1997, 1959.

26 Cooper, Al.; Londono, J.D.; Wignall, G.; MeClain, J.B.; Samulski, E.T.; Lin, J.S.; Dobrinyn, A.;

Rubinstein, M.; Burke, A.L.C.; Fréchet, J.M.J.; DeSimone, J.M. Nature 1997,389, 368.

27 Schenning, A.P.HJ.; Elissen-Román, C.; Weener, J.W.; Baars, M.W.P.L.; van der Gaast, SJ., Meijer,

E.W. J. Am. Chem. Soc. 1998, 120, 8199.

28 Ramzi, A; Bauer, B.J.; Scherrenberg, R.; Froehling, P.; Joosten, J.; Amis, EJ. submitted to

Macromolecules.

29 (a) Lorenz, K.; Frey, H.; Stühn, B.; Mülhaupt, R. Macromolecules 1997, 30, 6860. (b) Stark, B.; Stühn,

B.; Frey, H.; Lach, C.; Lorenz, K.; Frick, B. Macromolecules 1998,31,5415.

30 (a) Smal!, D.M. The Physical Chemistry ~f lipids, Plenum Press., New York 1986. (b) Fendler, J.H.

Memhrane Mimetic Chemistry, Wiley, New York 1982.

31 Chong, C.S.; Colbow, K. Biochim. Biophys. Acta 1976, 436, 260.

32 (a) Frederik, P.M.; Stuart, M.C.A.; Bomans, P.H.H.; Lasic, D.D. in Handhook of Nonmedical

Applications of liposomes, Lasic, D.D; Barenholz, Y. Ed. CRC Press, Boca Ratón, New York

1996. (b) Kilpatrick, P.K.; Miller, W.G.; Talmon, Y. in Swfactants in Solution. Vo/4, Mittal,

K.L; Bothorel, P. Plenum Publishing Corp 1986.

33 Koper, G.J.M.; van Genderen, M.H.P.; Elissen-Román, C.; Baars, M.W.P.L.; Meijer, E.W.; Borkovec,

M. J. Am. Chem. Soc. 1997, ll9, 6512.

34 (a) Bangham, AD.; De Gier, J.; Greville, G.D. Chem. Phy. lipids 1967, 1, 225. (b) Blok, M.C.; Van

Deenen, L.L.M.; De Gier, J. Biochim. Biophys. Acta 1967,433, L

35 Kom, E.D. Methods in Membrane Biology. Plenum Press, New York 1974.

135

Chapter5

36 (a) Wang, L.Y.; Winnik, M.A. Langmuir 1990, 6, 1437. (b) Wilhem, M.; Zhao, C.; Wang, L.Y.; Xu,

R.; Winnik, M.A. Macromolecules 1991, 24, 1033. (c) Yekta, A., Duhamel, J.; Brochard, P.;

Adiwidjaja, H.; Winnik, M.A. Macromolecules 1993, 26, 1829. (d) Astafieva, I.; Zhong, X.F.;

Eisenberg, A. Macromolecules 1993,26, 7339.

37 Kalyasundaram, K.; Thomas, J.K. J. Am. Chem. Soc. 1977, 99, 2039.

38 Shinitzky, M.; Barenholz, Y. Biochim. Biophys. Acta 1978,515, 367.

39 (a) Galla, HJ.; Sackmann, E. Biochim Biophys. Acta 1974, 339, 103. (b) Soutar, A.K.; Pownall, H.L

Hu, A.S.; Smith, L.C. Biochemist1y 1974, 13, 2828. (c) Vanderkooi, J.M.; Callis, J.B. Biochemist1y

1974, 13, 4000.

40 Galla, H.J.; Luisetti, J. Biochim. Biophys. Acta 1980,596, 108.

41 Zana, R. Surfactant Solutions: New Methods of Investigation. Marcel Dekker, New York 1986.

42 Lakowicz, J.R. Prindples of Fluorescence Spectroscopy. Plenum Press, New York 1983.

43 (a) Finkelmann, M.; Schafheutle, A. Colloid Polym. Sci. 1986, 264, 786. (b) Hamid, S.M.; Sherrintong,

O.C. Polymer 1987, 28, 332. (c) Ringsdorf, H.; Schlarb, B.; Venzmer, J. Angew. Chem. Int. Ed. Eng/.

1988, 27, 113. (d) Fendler, J.H. Membrane-mimetic approach to advanced materials Springer, Berlin

1994. (e) Fendler, J.H. Interactions and reaelions in polymerized vesicles Research Triangle Park,

Arrny Research Office 1988. (f) Fendler, J.H., Membrane mimetic chemist1}' : characterizations and

applications of micelles, microemulsions, morwlayers, bilayers, vesicles, host-guest systems, and

polyions Wiley-Interscîence, Chichester 1982.

44 (a) de Brabander, E.M.M.; Brackman, J.; Muré-Mak, M.; de Man, H.; Hogeweg, M.; Keulen, J.;

Scherrenberg, R.; Coussens, B.; Mengerink, Y.; van der Wal, Sj. Macromol. Symp. 1996, 102, 9. (b)

Scherrenberg, R.; Coussens, B.; van Vliet, P.; Edouard, G.; Brackman, J.; de Brabander, E.M.M.;

Morrensen, K. Macromolecules 1998,31,456.

136

6 Supramolecular Assemblies of Palmitoyl Poly(propylene

imine) Dendrimers and Surfactants

"Row a boatinto that tall tJD!Sferious grass jus/ because_you have heard a strange bird cal! there."

RHoffmantl

Abstract

When a mixture of DAB-dendr-(NHCO-{CH2) 14-CH3Jri4 and low molecular weight surfactants

is dispersed in water, stabie self-reinforced aggregates are formed. Pyrene fluorescence,

TEM, and DLS measurements show the formation of nanosized assemblies. A layer of

surfactant is constructed around the palmitoyl modified dendrimer, probably due to

interdigitation of the alkyl chains of the surfactant and the alkyl chains in the shell of the

dendritic core. The cac of the surfactants can be lowered by a relatively smal! amount of

palmitoyl modified dendrimer acting as a template and the surfactantldendrimer aggregates

obtained are thermodynamically stable. Fluorescence depolarization studies point to the

formation of semicrystalline domains in the interior of the aggregates. The chemica/

functionality of the nanoparticle's surface can be controlled by the proper choice of the

surfactant used. Smal! molecules could be enclosed as guest in the dendritic interior by

extraction in a liquid-liquid system, foliowed by the formation of supramolecu/ar assemblies

with surfactants in aqueous solutions. By using this new concept, very effective encapsulntion

of acidic probes was achieved, while attempts on pH-dependent release are discussed.

6.1. Introduetion

Molecular recognition and self-assembly are two important areas in supramolecular

chemistry. 1 Supramolecular assemblies allow us to obtain chemica) structures with nanometer

dimensions in a controlled manner. The construction of particles in this range is currently

receiving considerable attention because of applications in fields of inforrnation storage,

catalysis and sensors? One of the strategies toprepare nanostructures is the self-assembly of

amphiphilic molecules in water. With this concept an enorrnous variety of architectures can

be built.3 Dendrimers are attractive building blocks because they are well-defined,

Chapter6

multifunctional macromolecules with a specific size.4 In this Chapter, the use of amphiphilic

dendrimers in the construction of nanosized architectures and their use in host-guest

chemistry is described.

Earlier Tomalia and Turro et al.5 have shown that ionic surfactants can interact with

PAMAM dendrimers in aqueous solutions. PAMAM dendrimers with carboxylic acid

endgroups were found to interact with cationic surfactants to form different suprarnolecular

structures in water. The nature of these structures varied as a function of the dendrimcr sizc

and the concentrations of both the dendrimer and the surfactant.6 Interactions of primary

amine terrninated P AMAMs with anionic surfactants generate supramolecular assemblies that

may hold small hydrophobic molecules. Larger molecules with hydrophobic dodecyl chain as

dendrimcr core were able to accommodate anionic surfactants in the dendritic interior. The

interaction of the surfactants with the dendrimers was pH depcndent, as lower pH served to

create more positively charged sites for interaction with the negatively charged surfactants.7

In this Chapter, the complexation behaviour of low molecular weight surfactants with

inverted unimolecular micelles based on poly(propylene imine) dendrimers (DAB-dendr

(NHCO-{CHz)I4-CH3)n) will be reported. Investigations on the solubility of palmitoyl

modified dendrimers in water in the presence of low molecular wcight surfactants suggest a

co-operative organization of small amphiphiles and the alkyl chains of the dendritic micelle.

The dendritic macromolecule acts as a matrix for surfactant aggregation due to hydrophobic

intcractions. This is the first exarnple of the formation of supramolecular assemblies of

surfactants with the endgroups of a dendrimer, based on hydrophobic effects instead of

electrostatic forces. Using this concept, stabie nanosized assemblies with a functional surface

can be constructcd based on dendritic building blocks.

6.2. Supramolecular assemblies

6.2.1. CT A BID AB-dendr-(NHCO-( CH2)14-CH3)64 aggregates

When a mixture of DAB-dendr-(NHCO-{CHz)t4-CH3)64 and CTAB (hexadecyltrimethyl

ammonium bromide) in organic solution (THF!EtOH 2/1 v/v) is dispersed in water at 55 oe by injection with a thin-necdle-syringe, stabie aggregates are formed. The solubility of

palmitoyl modified dendrimers with low molecular weight surfactants in water suggest a co

operative organization of the small amphiphiles and the alkyl chains of the dendritic micelle.

138

Supramolecu/ar Assemblies Surfactant/Dendrimer

The dendritic macromolecule acts as a matrix in which the surfactant complexes by

hydrophobic interactions (see Figure 6.1).

+

Figure 6.1. Schematic representation of the formation of a layer of surfactani around the dendrimers due to interdigitation of the alkyl chains of the low molecular weight surfactant and the alkyl chains in the shell ofthe dendritic core. Approximately 85 dendrimers are self-assembled in one aggregate.

Electron microscopy and AFM experiments showed globular aggregates with an average

diameter of 20 nm (Figure 6.2) for aqueous solutions with a ratio of concentrations of CTAB

versus DAB-dendr-(NHCO-(CH2)14-CH3)64 of 70. The samediameter was found in solution

for this complex with dynamic light scattering measurements. The diameter indicated that

these assemblies between surfactant and dendrimer are formed by clusters of dendrimers.'

0.00 Jlffi

2.00 Jlm

0.00 Jlm

2.00!-lm

40.0nm

O.Onm

Figure 6.2. Height AFM image ofCTABI DAB-dendr-(NHCO-(CH2h,,-CH3)64 aggregates in aqueous solution. The sample was prepared by putting a droplet of salution on a silicon wafer and the measurement was done inside the droplet. The contrast covers height variations in the 0-30 nm range. The stripes are artefacts caused during the measurement.

' With molecular modeling, a radius of 1.6 nm for the fifth generation dendrimer is calculated while the lengthof

a palmitoyl chain as estimated from CPK molecular roodels is 2.2 nm.

139

Chapter6

Knowing that the calculated volume of one DAB-dendr-(NHCO-(CH2)14-CH3)64 molecule is

48580 N (see Chapter 5, Table 5.7), a rough estimation can be made for the number of

dendrimers in the clusters; assuming that the volume occupied by the surfactant layer is

negligible compared with the volume of the dendrimers, it was calculated that one aggregate

contains approximately 85 dendrimers on the average.

6.2.2. Formation of aggregates

The formation of supramolecular aggregates of DAB-dendr-(NHCO-(CH2)14-CH3)n in water

with low molecular weight surfactants was further investigated using pyrene as a fluorescent

probe. DAB-dendr-(NHCO-(CHz)14-CHûn, with n = 4, 8, 16, 32, and 64, and the cationic

surfactant CT AB were added to an aqueous solution of pyrene, keeping the concentration of

CTAB constant and under the cmc of CTAB (8.1·10-3 M). The ratio of vibration-band

intensities 1dh of pyrene fluorescence, which is an indicator of the polarity of the pyrene

environment,8 was foliowed as a function of dendrimer concentration (Figure 6.3). The curves

obtained show the shape of a conventional cmc plot. However, at very low concentrations of

dendrimer, supramolecular assemblies of dendrimer and surfactants are already formed,

possessing a 11113 ratio of 1.3, being lower than the one corresponding to free pyrene in water

(11/13 = 1.5) (Figure 6.3). When the concentration of dendrimer increases an onset appears in

the curves, indicating that pyrene is located in a much more apolar environment. Finally, the

ratio 11/13 of pyrene becomes constant (I 1/13 = 0.8).

1.8 2.0

1.6 1.8

1.4 1.6

1.4 1.2

1338/ 1m 1.0 1

1 I 1

3

1.2

1.0

0.8 0.8

0.6 0.6

0.4 0.4 -9 -8 -7 -6 -5 -4 -3 -9 -8 -7 -6 -5 -4 -3

logC(M) logC (M)

Figure 6.3. Values of lm/lmfrom excitation spectra (right) and 1/hfrom emission spectra (left), as a function of dendrimer concentration for a constant concentration of CTAB (JO~ M). Symbols correspond to DAB-dendr-(NHCO-(CH2) 1.CH3)" with: à, n = 4; 0, n = 8; V, n = 16; •. n = 32 and•. n = 64.

140


From the excitation spectra, the same changes in the speetral characteristics of pyrene are

observed, due to changes in the polarity of its microenvironment (Figure 6.3). These results

suggest that at least two different kinds of aggregates are formed, depending on the

CTAB/dendrimer ratio.

We propose a model that involves two different kind of aggregates. In the first region,

when the CTAB/dendrimer ratio is large, free surfactants, as wel! as "primary micelles"

containing dendrimers are present. In this situation the aliphatic shell of the dendrimer is

presumably completely saturated with surfactants. This implies that the cmc of CTAB can be

lowered by a relatively small amount of dendrimer present in solution. In the second region,

when the CT AB/dendrimer ratio is lower, clusters of dendrimers are probably formed in the

interior of these micelles, in which pyrene can be located in very apolar environments, wetl

protected from contact with water molecules (Figure 6.4). These CTAB/dendrimer aggregates

can be diluted by a factor of one hundred and still the pyrene is located in the hydrophobic

environment (I1/I3 of pyrene = 0.8). The supramolecular aggregates remain stabie in solution

for weeks

Figure 6.4. Schematic representation of the formation of a cluster of dendrimers when the molecular ratio surfactantldendrimer decreases.

6.2.3. Microviscosity in the aggregates

To obtain more information about the intemal organization of the hydrophobic domains in the

aggregates, diphenylhexatriene (DPH) is used as a hydrophobic molecular probe for

fluorescence depolarization studies (see Chapter 2).

141

Chapter6

0.35

• • 0.30 • • • • • • 0.25 0 0 0

0 :>. 0 §"0.20 0 0 • '-

0

0 • "' ·= 0.15 0 0 • <( • 0

0.10

0.05

o.oo+--.--.--.--.--.--.--.--.---.--.--r----.----,--, 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85

T("C)

Figure 6.5. DPH anisotropy values measured by fluorescence depolarization as a function of the temperature. e, OA/DAB-dendr-(NHC0-{CH2)u-CH3) 64 system; 0, CTABIDAB-dendr-(NHCO(CH2)u-CH3)M system.

DPH was solubilized in aqueous solutions of DAB-dendr-(NHCO-(CH2)w-CH3)64 with

octadecyl amine (OA) and CTAB as surfactants, respectively. A very high anisotropy value (r

= 0.32) was found for the OA/DAB-dendr-(NHCO-(CH2)14-CH3)64 system, reflecting a high

rigidity in the hydrophobic domains of the aggregates. In the case of CTAB/DAB-dendr

(NHCO-(CHûJ4-CH3)64 systems, the anisotropy value was lower (r = 0.24), indicating a

Jower microviscosity in the hydrophobic domain of the aggregates than in the previous case,

or the presence of more free pockets in between the aliphatic chains due to a lower chain

packing density that could permit the dye to move more freely.

In both cases, a smal! phase transition seems to appear at ca. 38 °C, foliowed by another

phase transition at 60 °C. It has to be pointed out that, in the case of CT AB, the curve is much

smoother and the transitions are more difficult to detect. The transition at 38 oe is very close

to that of the bilayers of vesicles of DAB-dendr-(NHCO-(CH2) 14-CH3)n as measured with

DPH (Chapter 5), while the transition at 60 oe is within the range of the melting point of

DAB-dendr-(NHCO-(CH2)14-CH3)n in the solid phase (Chapter 5). These results indicate

that the probe is probably located in at least two different regions in the hydrophobic part of

the aggregates.

142


70 x 85 -=85 ...

=20nm

Figure 6.6. Schematic representation of the formation of supramolecular aggregates surfactants/DAB-dendr-(NHCO-(CH2hrCH3)w The resulting aggregate can be compared with a "filled" micelle subsequently, yielding a very stable "self-reinforced" collotdal particle.

The cores of these aggregates consist presumably of clusters of DAB-dendr-(NHCO

(CHû14-CH3)64 in which amorphous material coexists with small domains where the

dendrimers are forming semi-crystalline lamellae.9 In Chapter 5, the spontaneous mieropbase

separation of DAB-dendr-(NHCO-(CHû14-CH3)n compounds to form lamellar structures has

been described. Based in these findings, a model is proposed to explain the results obtained

by fluorescence depolarization (see Figure 6.6).

Structural heterogeneities in the hydrophobic regions result in a different packing density for

probe molecules. Near the surface, where the surfactants are located, the packing density is

presumably low and the rotational motions of the probe are less hindered, and a lower degree

of order involves a phase transition at lower temperature. Probe molecules located within the

hydrophobic domains, where the packing density is higher, will most likely have a strongly

restricted rotation. It is important to point out that the hydrocarbon region near to the hydrated

headgroups in the aggregates is presumably more polar than the hydrocarbon interior in the

cores due to water penetration and the limited thickness of this region (the length of a

palmitoyl chain in an all-trans conformation is 22 A and the length of an octadecyl chain is

no Jonger than 25 Á). Consequently, the intemal part most likely contains a higher population

of the very apolar DPH probe than the exterior of the aggregate. Different populations should

also give different r values. Since only steady-state measurements were performed, no

information about the reorientational fluorophore motions could be obtain and the

populations of DPH in these two positions are unknown. Nevertheless, this phenomenon does

143

Chapter6

not influence the results obtained, from which we can conclude that the presence of two phase

transitions is an evidence of the formation of microdomains in the aggregates that display

distinct chain packing density.

6.2.4. Non-ionic and anionic surfactants

Non-ionic and anionic surfactants can also be used for the formation of supramolecular

aggregates with dendrimers. Two different kind of non-ionic surfactants were used to

investigate their interactions with DAB-dendr-(NHCO-(CH2) 14-CH3) 64 (see Figure 6.7).

1~ OH

2

Figure 6.7. Non-ionic surfactants used to fonn aggregates with DAB-dendr-(NHCO-(CH2)14-CH3)64

1: Dodecylthioglucose (DTG); 2:a-methoxy, ~tetradecyloxyl hexaethyleneglycol (E06C14).

Pyrene fluorescence measurements showed that non-ionic surfactants are also able to form

aggregates with DAB-dendr-(NHCO-(CH1)w-CH3)64 in an analogous way as CTAB. When

the samples containing 70 surfactant molecules per molecule DAB-dendr-(NHCO-(CHz)14-

CH3)64 were diluted, the surfactants were still able to form stabie aggregates, even at

concentrations of I o-7 M, far below their corresponding cmc' s.

After a mixture of anionic surfactant SDS (sodium dodecylsulphate) (concentration

under cmc of 9.2·10-4 M) and DAB-dendr-(NHCO-(CH2) 14-CH3)64 was injected in an

aqueous salution a prccipitate was immediately formed. This behaviour was additionally

foliowed with fluorescence measurements of pyrene, of which the emission intensity

decreased when the concentration of dendrimer was increased. Presumably, the hydrapbobic

pyrene molecules precipitated tagether with the surfactant/dendrimer assembly. The

precipitation might indicate that the orientation of the negatively charged SDS molecules is

opposite to that of the amphiphiles in the previous cases, as illustrated in Figure 6.8. Due to

electrastatic interactions the negatively charged sulphate headgroups are pointed to the

positively charged amino groups with the hydrapbobic aliphatic ebains facing the water

phase, creating an architecture with an apolar periphery.

144


+

Figure 6.8. Schematic representation of the formation of water insoluble SDS/DAB-dendr-(NHCO(CH2hrCHJ)64 aggregates.

On the other hand, when the palmitoyl modified dendrimer was injected into an

aqueous salution of SDS (concentration under cmc), stabie aggregates were formed, pyrene

was located into hydrapbobic microdomains and did not disappear from solution since its

fluorescence intensity in the salution did not decrease after 24 hours. Presumably, only

hydrophobic interactions take pi ace in aqueous solution, instead of the electrastatic forces, or

acid-base interactions, that take place in organic media and attract the surfactant headgroups

towards the polar core of the dendritic molecule. 11 In tlris case the same kind of aggregates

are formed as in the case of cationic and non-ionic surfactants, where the dendrimers are

covered by a surfactant layer that prevents the palmitoyl modified dendrimer from

precipitating in water (see Figure 6.9).

+

Figure 6.9. Schematic representation of the formation of water soluble SDS/DAB-dendr-(NHCO(CH2)u-CHJ)64 aggregates.

In order to obtain more information about the interaction of the surfactants with the

dendrimers, 10-(1-pyrenyl)decanoic acid (PDA), an anionic surfactant with a fluorescent

probe covalently bound to the tail was used (Figure 6.10).

145

Chapter6

HO, c 11 0

Figure 6.10. 10-( 1-Pyrenyl)decanoic acid (PDA).

When DAB-dendr-(NHCO-(CH2)14-CH3) 64 was added to an aqueous solution of PDA

micelles, the speetral characteristics of the pyrene chromophore changed immediately and a

broad peak corresponding to excimer formation appeared in the emission spectra (see Figure

6.11 a). The formation of excimer is presumably due to the formation of pyrene dimers in the

surface of the palmitoyl modified dendrimer. These results indicate that the surfactants are

able to interact with the alkyl ebains of the dendrimer, and that they are probably not

uniformly distributed around the hydrophobic shell of the dendrimer molecule. Sonication

experiments produced a deercase in the intensity of the excimer peak, indicating that the

distribution of surfactants around the dendrimer could be influenced to some extent. After one

hour of sonication, the excimer intensity to monomer intensity ratio did not change anymore

and the system reached equilibrium. After that, the plot of the intensity ratio of the excimer

peak to the monomer peak as a function of the temperature was reversible, indicating that the

aggregates are thermodynamically stabic (see Figure 6.1 1 b ).

40

35 b • e •

30 0 0 a

0

IJim 25 • Int 0

0

• 0

................ _ 360 400 440 480 520 550 20 30 40 50 60 70

À. (nm) T("C)

Figure 6.11. a) Nonnalized spectra of PDA at afixed concentration of 7-Jo-5 M. ·········· PDA in water; -- PDA in the presence of DAB-dendr-(NHCO-(CH1)u-CH3)64. b) Plot of the 1/lm ratio as ajunetion of the temperature, for the PDA/DAB-dendr-(NHCO-{CH2) 14-CH3). system; •. heating curve; 0, cooling curve.

146


6.2.5. Azobenzene-containing dendrimers

In a second type of experiment, the chromophores were located in the alkyl chains of

the dendrimer. Therefore, an azo-benzene modified dendrimer of the fifth generation was

used as designed, synthesized and characterized by Jan Willem Weener in our group12 (see

Figure 6.12). In this class of amphiphilic dendrimers, the alkyl chains contain an azo-benzene

chromophore so that information about the relative orientation of the alkyl chains with respect

to each other can be monitored by UV -vis spectroscopy. 13

• = -

Figure 6.12. Schematic representation of 11-[4-(4-hexyloxypherrylazo)phenyloxy]undecanoyl functionalized dendrimer of the fifth generation (AD64).

Aqueous solutions of AD64 show an absorption maximum at 312 nm, indicative of H-type

aggregates of the azo-benzene units14 and one shoulder at 350 nm, corresponding to

monomeric azo-benzene. When CTAB/AD64 mixtures with different surfactant to dendrimer

ratios were dispersed in water, interaction of the cationic surfactants with the dendrimer was

clearly supported by UV-vis measurements (see Figure 6.13). The UV-vis spectra of the azo

benzene units show a decrease in the n:-1t interactions between chromophores, presumably

due to intercalation of the aliphatic chains of the surfactant in between the alkyl chains of the

dendrimer, and dissalution of dendrimer clusters into smaller aggregates. UV-vis

measurements of the aggregates by varlation of temperature showed reversible curves for

heating and cooling. These results indicate that, as in the case of PDA/DAB-dendr-(NHCO

(CHz)t4-CH3)64, these systems are thermodynamically stabie as well.

147

Chapter6

2.30

a

2.20

0

Abs :::; 2.10 --" "

2.00

• 1.90

300 340 380 420 460 500 0 2 4 6

A(nm) SurfactantI dendrimer ratio (x 103)

Figure 6.13. a) Normalized UV-vis spectra of AD64 for different CTABIAD64 ratios (60-1300). The relatively intensity of the monomeric chromophores related to the 11:-stacked chromophores, increases with increasing suifactant concentration. b) Plot of the 11:-stacked to monomeric intensity ratio vs suifactant to dendrimer ratio.

Trans-cis photoisomerization of the azo-benzene units by irradiation with light at 365 nm

was possible for samples with high surfactant to dendrimer ratio (Figure 6.14). Stacked azo

benzene chromophores are not able to undergo trans-cis isomerizations. Interpenetration of

the alkyl ebains of the surfactants between the alkyl ebains of the dendrimers "dissolves" the

aggregates of azo-benzene units and allow the monomeric chromophores to undergo

photoisomerization due toa higher mobility of the chains.

a

Abs lRANS Abs

~

" ~ CIS

+ --275 320 360 400 440 480 500 275 320 360 400 440 480 500

À(nm) À(nm)

Figure 6.14. Nonnalized UV-vis spectra of AD64 -- before and ········ after irradiation with light at 365 nm. a) CTABIAD64 = 70; b) CTABIAD64 = 1300.

148


6.2.6. Conclusions

These results suggest that the surfactant/DAB-dendr-(NHCO-(CH2)14-CH3)64 aggregates

possess a rigid structure and could be regarded as fitled micelles. The polymerie nature of

palmitoyl dendrimers confers rigidity and stability to the aggregates. Furthermore, the

possibility to cover the outer layer of the aggregates with different functionalities, just by

carefully choosing the surfactant, opens multiple possibilities for the construction of

nanoscopic materials. Surfactants with polymerizable headgroups can be used for the

construction of a well defined polymerie shell around the dendrimer matrix. Positively

charged surfactants form polycationic structures that could for instanee be used in the

formation of molecular sieves15 or in the development of gene transfer devices. 16 Recently,

octadecylamine/DAB-dendr-(NHCO-(CH2)14-CH3)64 aggregates have been used by Jack

Donners as self-reinforced template for the oriented nucleation of inorganic materials,9

similar to the way biomaterials such as shells and nacre are produced in biologica! systems. 17

Formation of spheroids of amorphous calcium carbonate were observed. These are

remarkable results since the coexistence of amorphous and crystalline calcium carbonale is

only found in nature.18

6.3 Host-guest systems. Encapsulation of dyes

Poly(propylene imine) dendrimers have been shown to be efficient hosts for anionic

molecules. The dendritic box is a well-known example of the ability of dendrimers to

accommodate small molecules intheir interior. 19 DAB-dendr-(NH-CO-(CH2)14-CH3)n have

been used as extractants in liquid-liquid systems, due to the strong interaction of the dendritic

interior with smal! molecules containing acid functionality. It has been reported that dendritic

unimolecular micelles with a fluorinated shell can transport small molecules into supercritical

carbon dioxide within their cores.20 The amphiphilic behaviour of alkylated dendrimers

makes it possible to extract water-soluble anionic dyes into an organic phase as wel!. Maurice

Baars in our laboratories used alkyl modified poly(propylene imine) dendrimers as very

effective extractants of anionic dyes like Fluorescein, 4,5,6,7-Tetrachlorofluorescein,

Erythrosin B and Bengal Rose from an aqueous phase into organic phases as dichloromethane

and toluene. Since it was not possible to extract cationic dyes, like Ethidium Bromide, it was

149

Chapter6

concluded that the interaction between solute and dendritic structure was mainly determined

by their acid-base properties and therefore it is related to their tertiary amine interior. 11

6.3.1. Encapsulation of carboxyfluorescein

The high extraction efficiency of anionic molecules from aqueous to organic phases by

palmitoyl modified dendrimers and their capability of being stabilized in water with low

molecular weight surfactants should lead to the effective encapsulation of anionic molecules

within dendritic micelles in aqueous solution. The proposed mechanism involves first an

extraction of the anionic molecules from aqueous solution into an organic phase using

dendrimer micelles. Secondly, the dendritic micelles, presumably containing the entrapped

anionic guest molecules, are stabilized in water by actdition of low molecular weight

surfactants. As a result, unimolecular micelles containing anionic guest molecules surrounded

by a surfactant shell are present in aqueous solution (Figure 6.15). DAB-dendr-(NH-CO

(CH2)14-CH3)64 was used to extract carboxyfluorescein from the aqueous phase into

dichloromethane. Carboxyfluorescein (CF) is a dianionic xanthene dye with high absorption,

quanturn yield and solubility in water. It is moreover very much less hydrophobic than its

parent compound fluorescein and is therefore expected to show a lower tendency to be

located within the dendrimer alkyl chains. Fluorescence studies can provide valuable

information about Jocal pH and dye concentration?1

Figure 6.15. Proposed mechanism for the extraction of anionic molecules into organic media and encapsulation of the probes in the interior of the dendrimer by stabilization with low molecular weight surfactanis in water.

150


The extraction efficiency has been proved to bedependenton pH (Figure 6.16).The pKa and

hydrophobicity of the dye are likely to account for this pH dependency in extraction curves. 11

80

60 "d 'Qj ·;;..

40 = 0 ·.:::: u <:<:: .!:::: 20 ~ Q)

~

0

2 4 6 8

pH

10 12

HO

COOH

Carboxyfluorescein 14

Figure 6.16. Left: Extraction yield plot of the extraction of carboxyfluorescein molecules with DABdendr-(NH-CO-(CH2)14-CH3)64, as ajunetion of pH.

The dry residue, obtained by evaporation of the organic solvent after the extraction,

was mixed with CTAB in a surfactant/dendrimer ratio of 70 and injected into water. In this

aqueous solution carboxyfluorescein presented a different fluorescent behaviour than in an

aqueous salution of the same concentration of free CF with CT AB/DAB-dendr-(NHCO

(CHz)J4-CH3)64 aggregates, used as reference (see Figure 6.17). The Àmax of the fluorescent

spectra of encapsulated CF were shifted 8 nm to higher wavelengths, due to changes in the

local environment of the probe in the interior of the dendrimer. A decrease in the quanturn

yield of the fluorescence spectra was also observed, due to quenching by the tertiary amines

in the interior of the dendrimer. *22

* Probably, an excited-state electron-transfer reaction takes place between the tertiary amines in the interlor of

the dendrimer and tbe hosted dye. Photosensitized oxidation of tertiary amines by xanthene dyes has been

observed before (ref 22).

151

Chapter6

•••• 0 0

Int

400 440 480 520 560 600

'A. (nm)

Figure 6.17. Fluorescence excitation and emtsswn spectra of carboxyfluorescein in aqueous solutions. ············· In the presence of CTABIDAB-dendr-(NH-CO-(CH2)u-CH3)64 aggregates; ---Encapsulated by extraction into CTABIDAB-dendr-(NH-CO-(CH2)u-CH3) 64 aggregates.

Several experiments were performed with different ratios of dye to dendrimer

molecule from 0.1 to 10, in order to study the effect of local concentration of dye into the

dendritic interior. lt was observed that the dendritic interior produces a red shift of 8 nm in

the case of on the average one dye molecule per dendrimer, due to the local environment

changes. An additional red shift of 18 nm and se1f-quenching effects were found in the case

of approximately 7 CF molecules per dendrimer, due to the high local concentratien of dye?3

In this last case, the dye molecules are probably not all located in the dendeitic interior, but

with their aromatic rings located between the alkyl chains, near the dendrîmer core?4 The

fluorescence intensity of two different aqueous solutions of encapsulated carboxyfluorescein

was foliowed with time in order to see whether encapsulated carboxyfluorescein leaked from

the dendritic core. The leakage percentage will be defined as the percentage of increase in

intensity taking the intensity of a reference solution of free CF in water as 100%, and the

initia! intensity of the encapsulated CF at time zero as 0%.

In the case of low loading in the dendrimers (CF/DAB-dendr-(NHCO-(CH2)14-

CH3)64 = 0.07), there is practically no leakage of dye after four days. In the case of highly

loaded dendrimers, there is probably a slow ditfusion out of the aggregates of some CF

molecules that are not strongly bound to the tertiary amines of the dendrimer. After the first

two days, the system becomes stabie and further leakage is negligible.

152


~ 41llday

~ " Cl) " "' bO ... "' "' -0% ~ ~ leakage ~

time (min) time (min)

Figure 6.18. Changes in the intensity of encapsulated CF at room temperature, expressed in % leakage, at ratios CF/DAB-dendr-(NHC0-(CH2) 14-CH3)r>4 of0.07 (left plot) and 7 (right plot).

6.3.2. Encapsulation of ANTS

The anionic dye 8-aminonaphthalene-1 ,3,6-trisulphonic acid (ANTS) fulfils convenient

characteristics for the extraction with dendritic micelles; it possesses acid functionalities and

ît is highly soluble in water. After extraction of ANTS in two-phase systems and

complexation of the host-guest system with cationic surfactants in aqueous solutions, the

fluorescence intensity of ANTS was measured and compared with a reference salution with

the same concentration of ANTS, in which the system CTAB/ DAB-dendr-(NHCO-{CH2)14-

CH3)64 was present. The results are plotted in Figure 6.19.

The fluorescence emission of ANTS was strongly quenched by the tertiary amines of

the dendrimer interior. So we can conclude that, similar to the case of CF, ANTS could be

effectively encapsulated into the dendritic molecules by the extraction method.

Int

I

'' ' ' I \ I \

I I I

I I I

I I I .

I I

I I

I I

I

' ' ' ' ' I I . . .

' I I . .

. I

I I I

I I

I \ I

I . ' ' ' ' ' ' ' I \ 0

I

' ' '

0

0

0

\ ...... ~ ~~====~~-~--~~==~====:==---~ ' I

I I

300 400 500 600 À(nm)

Figure 6.19. Fluorescence excitation and emission spectra of ANTS in aqueous solution, ............. in the presence of CTAB/DAB-dendr-(NH-CO-(CH2) 14--CH3)64 aggregates; --- encapsulated by extraction into CTAB/DAB-dendr-(NH-C0-(CH2) 14-CH3) 64 aggregates.

153

Chapter6

In order to investigate the accessibility of encapsulated molecules by molecules free in

solution, quenching experiments were performed. Quenching assays are often used for

membrane fusion and permeability studies.25•26 The fluorescent emission of the probe 8-

aminonaphthalene-1,3,6-trisulphonic acid (ANTS) can he quenched by complexation with the

cationic molecule p-xylene-bis-pyridinium bromide25 (DPX) schematically depicted in Figure

6.20.

2.0

1.5

IJl

0.0 LOxw--• 2.0xlo-' 3.0xlo-'

[DPX] (M)

ANTS

DPX

S03H

Figure 6.20. Left: Stern-Volmer plot of the fluorescence intensity ratio lr/I vs quencher concentrationfor aqueous solutions offree ANTS (•) and encapsulated ANTS (0). Right: schematic representation of the quenching assay, with ANTS as probe and DPX as quencher.

The Stem-Volmer quenching constant of free ANTS is 2500 M-1, while for the encapsulated

ANTS the Stem-Volmer constant is 70 M-1• Quenching experiments demonstrate that

encapsulated ANTS molecules can not he dynamically quenched by DPX, presumably due to

the steric and electrostatic harrier composed by the surfactant/DAB-dendr-(NHCO-(CH2) 14-

CH3)64 aggregates.

6.4. pH controlled release of encapsulated dyes

The dendritic micelles can he dissolved in water by using low molecular weight surfactants,

and there is evidence that smal! molecules can be encapsulated in the interior of these

aggregates. In Chapter 5, it was described how alkyl modified dendrimers are able to self

assemble in acidic aqueous solutions to form vesicles. By combining these features, it could

he possible that a decrease of the pH in the solution originates a rearrangement of the

154

Supramolecu/ar Assemblies Surfoctant /Dendrimer

aggregates in solution, due to protonation of the dendritic interior of the molecules, to form

vesicle structures. If that could happen, it could be possible to enclose dyes in the dendritic

interior by extraction and release them later by a pH controlled method; vesicles would

expose the guest from the dendritic interior to the aqueous solution. This process is

schematically represented in Figure 6.21.

Dye Extraction

Dendritic Micelles in CH2Cl2

Encapsulated Dye in Water

Stabilized Dendritic Micelles

in Water

FreeDye in Acid Solution

Vesicles in Water pH= 1

Figure 6.21. Schematic representation of the encapsulation of dyes by the extraction method and pilcontrolled release.

Such a system could have potential applications in the fields of medicinal, cosroetics and

environmental chemistry. In order to investigate the applicability of this nanoscopic delivery

system, fluorescent studies on systems containing encapsulated CF were made by changing

the pH of the solution.

In solutions with one CF molecule per dendrimer, a change in pH from 7 to 1 by

acidification with HCl provokes a decrease in the tluorescence intensity of the probes due to

protonation of the molecule (see Figure 6.22a).Z1 Remarkably, when highly loaded

CTAB/DAB-dendr-(NHCO-(CH2) 14-CH3) 64 aggregate solutions (with at least 7 CF

molecules per dendrimer core) were acidified, the tluorescence emission intensity of CF did

not decrease but increased with a factor of 5 with respect to the intensity maximum inside the

dendritic core (Figure 6.22b). This increase in theemission quanturn yield could be due to the

partlal release of CF to the bulk solution. If the probe leaves the host, the high local

concentration effect disappears and the quenching of tluorescence only accounts for

155

Chapter6

protonation and not for self-quenching effects or electron transfer of the dendritic tertiary

amines.

Int

12

a 10

8

6

4

2

0~-r~~~,-~~~,-~~~ 400 420 440 460 480 500 520 540 560 580 600

À. (nm)

Int

50 b

40 I' ,,

I \ I ' I

\ I ' \ I ' I

\ I

30 I \ f ... \ I ' I ...

I ' I

\ I

\ I

20 I

'~ \

' \

\ ' \ ' \ '

JO

0~~~~~~--~~rT~~~~ 400 420 440 460 480 500 520 540 560 580 600

À. (nm)

Figure 6.22. Fluorescence spectra of encapsulated CF molecules-- at pH = 7 and·· ....... aft er acidification at pH = 1. a) CFIDAB-dendr-(NHCO-{CH2)u-CH3)64 molecular ratio = 0.7; b) CFIDAB-dendr-(NHCO-(CH2)14-CH3)64 molecular ratio= 7.

Dialysis experiments were performed to wash out the free dyes, but the presence of low

molecular weight surfactants makes the use of dialysis membranes impossible, as was

confirmed by reference experiments using CF in CT AB solutions, where the probe was not

able to diffuse trough the pores of the membrane.

The complexity of the system, where many different surface active molecules are

present and numerous factors can influence the observed results, makes the interpretation of

the information obtained highly speculative. Moreover, extreme conditions as pH close to 1

arenotabundant in biologica) systems; the pH conditions in the blood stream and cells varies

in between 7 and 4. Furthermore, the use of multiple systems with many different molecules

presents many drawbacks for applications in medicinal chemistry, where pure

macromolecular compounds are preferred.

6.5. Condusions

Investigations on the solubility of palmitoyl modified dendrimers in water in the presence of

low molecular weight surfactants suggest a co-operative organization of smal! amphiphiles

and the alkyl ebains of the dendritic micelle. The dendritic macromolecule acts as a matrix for

156


the formation of surfactant aggregates due to hydrophobic interactions. Depending on the

molecular ratio of surfactant to dendrimer, monomeric aggregates or clusters of dendrimers

can be obtain in solution. Preliminary results indicate that, in the case of clusters, the radius

of the aggregates depends on the type of surfactant used, due to the different interactions

between the surfactant's headgroups.* Different kind of surfactants can be used to form stabie

surfactant!DAB-dendr-(NHCO-(CH2) 14-CH3)64 aggregates. Consequently, it is possible to

obtain very stabie aggregates of nanoscopic dimensions with a well defined chemica!

functionality on the surface. The polymerie nature of these aggregates yields rigid,

semicrystalline particles with a narrow size distribution. The construction of stabie

nanoscopic building blocks of known size and chemica! properties, makes it possible to

obtain functional matcrials in a controlled way.

The ability of dendrimers to act as a host for small molecules has been studied in

detailY The higher generations of dendrimers can adopt a globular structure with cavities

inside, that makes it possible toentrap small molecules in their interior. The driving forces for

the small molecules to interact with the dendritic interlor can have different character, such as

hydrophobic, hydrogen-bonding, metal coordination or electrostatic interactions. The

extraction of molecules from the aqueous phase into an organic solution is based in acid-base

interactions. Using this method, small molecules with an acid functionality can be

encapsulated into the alkyl modified dendrimers and stabilized in water by forming

aggregates with low molecular weight surfactants. The exact location of the probes is not

known, but there is evidence of effective physical encapsulation. Further reseach is needed to

study the scope and limitations of this novel encapsulation technique.

6.6. Experimental

Materials

The synthesis of DAB-dendr-(NHCO-(CH2) 14-CH3)n has been reported before.24"28 11-[4-(4-

Hexyloxyphenylazo)phenyloxy]undecanoyl functionalized dendrimer of the fifth generation (AD64),

E06C14 and DGT were kindly provided by Jan-Willem Weener, Henk Janssen and Rob Peerlings,

* When the surfactani used is cationic (CT AB), the modal size of the aggregates, studied by TEM and DLS, is

ca. 20 nm. In the case of a non-ionic surfactant (OA), the size of the aggregates is a factor of 10 larger: ca. 200

nm.9

157

Chapter6

respectively and they were synthesized in our laboratorles followîng the procedures reported

elsewhere. 12'29 The surfactanis SDS and CT AB were purchased from SIGMA. OA and Pyrene were

obtain from Aldrich. CF was commercially available from Fluka and PDA, ANTS and DPX from

Molecular Probes.

AFM measurements

AFM measurements were taken by Dr. W.P. Veilinga at the Department of Computational and

Experimental Mechanics, Polymer Technology in the Eindhoven University of Technology. The

images were taken with a Multimode Nanoscope lil (Digital Instruments) operaled in the tapping

mode. The measurements were performed in a droplet of CT AB/DAB-dendr-(NHCO-(CH2) 14-CH3) 64

aqueous solution deposited onto a silicon wafer at room temperature.

Pyrene Experiments

DAB-dendr-(NHCO-(CH2) 14-CH3)., with n = 4, 8, 16, 32, and 64, and cationic surfactant CTAB

(hexadecyltrimethylammonium bromide) were added to an aqueous solution of pyrene, with [pyrene]

= 4.8 x I o-7 M, maintaining the concentration of CT AB constant at 1.1·1 0-4 M and under the cmc

(the cmc of CTAB is 8.1·10-3 M). The concentration of DAB-dendr-(NHCO-(CH2) 14-CH3)n varied

from 3.1·10-9 M to 6.7·10-5 M. Steady-state fluorescence spectra were run in a Perkin-Elroer

Luminescence spectrometer LS 50B in the right-angle geomctry (90° collecting opties). For

fluorescence emission spectra Àcx was 339 nm, for excitation spectra Àem was 390 nm. The

surfactant/DAB-dendr-(NH-CO-(CH2)w-CH3)n mixture was dissolved in 100 IJ.] of THF/EtOH (2/1

v/v) and injected in I 0 mi aqueous solution of pyrene at 55 °C. The samples were stirred ca. 30 min at

room temperature to allow the pyrene and the aggregates to equilibrate.

In the case of non ion ie surfactants, the experiments were carried out in an analogous way, with the

coneentrations of surfactants: [DTG] = 1.0·10-4 M (cmc of DTG = 2.0·10-4 M); [E06C14] = 1.3·10·5

M (cmc of E06C14 = 2.5·10-5 M). Anionic surfactani SDS was first dissolved in water; the

concentration was kept constant at 5.0·10-4 M (cmc of SDS = 9.2·10-4 M). DAB-dendr-(NHCO

(CH2)I4-CH3)64 was added to the surfactani solution using the injection method.

PDA

To 5 mi of a PDA aqueous solution with concentration 7 ·10-5 M (cmc of PDA = 2.0·10-7 M), 0.112

mg DAB-dendr-(NHCO-(CH2) 14-CH3) 64 was added using the injection method. The fluorescence

emission was foliowed using Àex = 339 nm. The samples were heated using a thermocouple in the

sample holder conneeled to an RTE 110 Neslab thermostat, with an ethylene glycol/water (1/1 v/v)

bath. The temperature was measured by a thermometer in the sample holder connected to the

158


computer. The measurements were made in a range of temperatures of 20 to 70 "C. The samples were

sonicated for 1 h in a Branson 2210 sonication bath at room temperature.

UV-vis measurements in the azobenzene modified dendrimer

11-[ 4-( 4-Hexyloxyphenylazo )phenyloxy ]undecanoyl functionalized dendrimer of the fifth generation

( AD64) was mixed with CT AB surfactants and injected in water at 55 "C to obtain samples of 2.5 ml.

The concentration ofthe azobenzene dendrimer in the solutions was 2.7·10-6 M, the concentration of

surfactani varled from 0 to 1.2· 10-2 M. The photoisomerizatlon from trans to cis was performed by

irradiation of the samples at 360 nm. UV -vis measurements were performed on a Perkin-Eimer UV

vis Lambda 3B spectrometer. The temperature variabie experiments were made using the same

instruments as in the case of fluorescence measurements.


For the t1uorescence anisotropy measurements, DPH was dissolved tagether with the

surfactantJDAB-dendr-(NHCO-(CH2)14-CH3) 64 aggregates by the injection method. The solutions

were placed in a bath-type sonicator and sonicated for 30 minutes at room temperature. The ratio

surfactani to palmitoyl dendrimer was 70:1, being [DAB-dendr-(NHCO-(CH2)14-CH3) 64] 6.8· 10 -s

M; the concentration of probe was [DPH] = 3.4·1 o-5 M. For fluorescence emission spectra À", 382

nm, for the excitation spectra À.:m = 430 nm. In both cases the anisotropy was measured in a range of

temperature between 20 oe and 75 oe by taking steps of 5 oe and allowing the solutions to

equilibrate during 15 min.

Encapsulation experiments

The extraction of CF from water to CH2Ch solutions was performed with volumes of 4.5 ml in a 1: 1

ratio. The concentratîon of the probewas 10 4 10-<> M, the concentration of DAB-dendr-(NHCO

(CH2)J4-CH3)64 in the organic phase was 10-5 M. The efficiency of the extraction was 70%. The

extraction of ANTS in aqueous salution with concentration I o-4 M was performed with 5 ml of a

dendrimer salution in dichloromethane with a concentration of 10-5 M. The efficiency of the ANTS

extraction was 88%. The organic solvent was evaporated under vacuum at room temperature. The

organic phase was separated after the extraction and the solvent was evaporated in vacuo. The dry

residue was mixed with CTAB to obtain a final CTAB/DAB-dendr-(NHCO-(CH2) 14-CH3) 64 ratio of

70:1. This mixture was dissolved in water by the injectîon method. Fluorescence measurements were

recorded using À.:x = 507 nm for CF and 350 nm for ANTS. The slit openings for emission and

excitation were 5 nm. For the quenching experiments, the emission intensities were measured after 4

j.l] aliquots of a DPX stock salution with [DPX] = 6.6· 10-3 M were added to 2.5 ml of ANTS

159

Chapter6

solutions, giving increasing DPXIANTS ratios from 0 to 3.6. Upon addition of DPX stock solution,

the fluorescence intensity was corrected for the change in volume.

6.7. References

(a) Lehn, J.-M. Angew. Chem., Int. Ed. Engl. 1988, 27, 90. (b) Lehn, J.-M. Pure Appl. Chem. 1994, 66,

1961.

2 (a) Whitesides, G.M.; Mathias, J.P; Seto, C.T Science 1991,254, 1312. (b) Lehn, J.-M. Angew. Chem.,

Int. Ed. Eng/. 1990,29, 1304.

3 Fendler, J.H. Membrane Mimetic Chemistry Wiley, New York 1982.

4 (a) Fréchet, J.M.l Science 1994, 263, 1710. (b) Tomalia, D.A.; Naylor, A.M.; Goddard lil. WA

Angew. Chem., Int. Ed. Eng/. 1990, 29, 138. (c) Newkome. G.R.; Lin, X. Macromolecules 1991, 24,

1443. (d) Issbemer, J.; Moors, R.; Vögtle, F. Angew. Chem., Int. Ed. Eng/. 1994, 33, 138.

5 Caminati, G.; Turro, N.l; Tomalia, D.A. J. Am. Chem. Soc.1990, 112, 8515.

6 (a) Kumar, C.V.; Barton, J.K.; Turro, N.J. J. Am. Chem. Soc. 1985, 107, 5518. (b) Ottaviani, M.F.;

Turro, N.J.; Jockusch. S.; Tomalia, D.A. J. Phys. Chem. 1996, 100, 13675. (c) Ottaviani, M.F.; Turro,

N.J.; Jockusch, S.; Tomalia, D.A. Colloid Suifaces A: Physicochem. Eng. Aspects 1996, 115, 9. (d)

Ottaviani, M.F.; Andechaga, P.; Turro, N.J.: Tomalia, D.A. J. Phys. Chem. 1997, 101, 6057.

7 (a) Watkins, D.M.; Sayed-Sweet, Y.; Klimash, J.W.; Turro, N.J.; Tomalia, D.A wngmuir 1997, 13,

3136. (b) Ottaviani, M.F.; Daddi, R.; Brustolon, M.; Turro, NJ.; Tomalia, D.A, Appl. Mag. Res. 1997,

13,347.

8 Kalyasundaram, K.; Thomas, J.K. J. Am. Chem. Soc. 1977. 99, 2039.

9 Donners, J.J.J.M.; Román, C.; Heywood, B.R.; Meijer, E.W.; Nolte, RJ.M.; Schenning, A.P.HJ.;

Sommerdijk, N.A.J.M. submitted to Nature.

10 (a) van Genderen, M.H.P; Baars, M.W.P.L.; Elissen-Román, C.; de Brabander-van den Berg, E.M.M.;

Meijer, E.W. Abstr. Pap. Am. Chem. Soc. 1995, 20, 179. (b) Koper, G.J.M.; van Genderen, M.H.P.;

Elissen-Román, C.; Baars, M.W.P.L.; Meijer, E.W.; Borkovec, M. J. Am. Chem. Soc. 1997, 1/9,6512.

11 Baars, M.W.P.L.; Froehling, P.E., Meijer, E.W. Chem. Commun. 1997, 1959.

12 Schenning, A.P.HJ.; Elissen-Román, C.; Weener, J.W.; Baars. M.W.P.L.; van der Gaast, SJ.; Meijer,

E.W. J. Am. Chem. Soc. 1998, 120, 8199.

13 (a) Kunitake, T.; Shimomura, M. J. Am. Chem. Soc. 1982, 104, 1757. (b) Song, X.; Perlstein, J.;

Whitten, D.G. J. Am Chem. Soc. 1997, 119, 9144.

14 Shimomura, M.; Kunitake, T. Chem. Lett. 1981, 1001.

15 Tanev, P.T.; Chibwe, M.; Pinnavaia, J.T. Nature 1994,368,321.

16 (a) Kukowska-Latallo, J.F.; Bielinska, A.U.; Johnson, J.; Spindler, R.; Tomalia, D.A.; Baker, J.RJ.

Proc. Nat/. Acad. Sci. USA 1996, 93, 4897. (b) Bielinska, A.U.; Kukowska-Latallo, J.F.; Johnson, J.;

Tomalia, D.A.; Baker, J.R.J. Nucleic Acids Res. 1996, 24, 2176.

160


17 (a) Mann, S. Nature 1993, 365, 499. (b) Addadi, L.; Weiner, S. Angew. Chem. Int. Ed. Eng!. 1992, 31,

153.

18 Aizenberg, J.; Lambert, 0.; Addadi, L. Weiner, S. Adv. Mater. 1996,8, 222.

19 Jansen, J.F.G.A.; de Brabander-van den Berg, E.M.M.; Meijer, E.W. Science 1994, 226, 1226.

20 Cooper, A.I.; Londono, J.D .. Wignall. 0.; McLain, J.B.; Samulski, E.T.; Lin, J.S.; Dobrinyn, A.;

Rubinstein, M.; Burke, A.L.C.; Frechet, J.M.J.; DeSimone, J.M. Nature 1997,389,368.

21 Graber, M.L.; Dilillo, D.C.; Friedman, B.L.; Pastoriza Mufioz, E. Analytica/ Biochemistry 1986, 156,

202.

22 (a) Bartho1omeu, R.F.; Davidson, R.S. J. Chem. Soc. 1970, 2347. (b) Archut, A.; Azzellini, O.C.,

Balzani, V.; De Cola, L.; Vögtle, F. 1. Am. Chem. Soc. 1998, 120, 12187.

23 Weinstein, J.N.; Yoshikami, S.; Henkart, P.; Blumenthal, R.; Hagins, W.A. Science 1977, 195, 489.

24 Stevelmans. S.; van Hest, J.C.M.; Jansen, J.F.G.A.; van Boxtel, D.A.F.J.; de Brabander-van den Berg,

E.M.M.; Meijer, E.W. 1. Am. Chem. Soc. 1996, Jl8, 7398.

25 Ellens, H.; Bentz, J.; Szoka, F.C. Biochemistry 1984,23, 1532.

26 (a) Parente, R.A.; Nir, S.; Szoka, F.C. Biochemistry 1990, 29, 8720. (b) Stamatos, L.; Laventis, R.;

Zuckermann, M.J.; Silvius, J.R. Biochemistry 1988, 27, 3917. (c) Leventis, R.; Diacovo, T.; Silvius,

J.R. Biochemistry 1987, 26, 3267. (d) Manchefio, J.M.; Ofiaderra, M.; Martfnez del Pozo, A.; Dfaz

Achirica, P.; Andreu, D.; Rivas, L.; Gavilanes, J.G. Biochemistry 1996, 35, 9892.

27 (a) Naylor, A.M.; Goddard, W.A.I.; Kiefer, G.E.; Tomalia, D.A. 1. Am. Chem. Soc. 1989, I IJ, 2339.

(b) Newkorne, G.R.; Moorefie!d, C.N.; Baker, G.R.; Saunders, M.J.; Grossman, S.H. Angew. Chem.,

Int. Ed. Engl. 1991, 30, 1178. (c) Hawker, C.J.; Wooley, K.L.; Fréchet, J.M.J. 1. Chem. Soc., Perkin

Trans.] 1993, 2/, 1287. (d) Mattei, S.; Seiler, P.; Diederich, F.; Gramlich, V. Hel. Chim Acta 1995,

78, 1904. (e) Jansen, J.F.G.A.; de Brabander-van den Berg, E.M.M.; Meijer, E.W. Adv. Mater. 1995, 7,

561.

28 van Hest, J.C.M. New Molecular Architectures based on Dendrimers. PhD Thesis, Eindhoven

University of Technology 1996.

29 Shinoda, K.; Yamaguchi, T.; Hori, R. Bull. Chem. Soc. Jap. 1961,34,237.

161

Epilogue

Dendrimers are attractive molecules due to their well-defined architecture and high

functionality. Many efforts have been made inthelast 15 years to improve the synthesis and

characterization of this new kind of macromolecules. In the past, several modifications have

been performed in the chemical structure of dendrimers in order to introduce specific

functions like liquid crystallinity, chirality, etc., with potentlal applications in catalysis,

energy transfer processes, drug delivery, gene transfer, host-guest chemistry, etc.

Since the pioneer work of Newkome in 1985, special attention has been focused on modified

dendrimers that present amphiphilic properties due to their ability to change their

conformation to interact with surfaces or to form suprarnolcular assemblies. This kind of

dendrimers has potentlal applications in numerous fields of science, such as environmental

chemistry, polymer therapeutics, cosmetics, coatings, advanced nanostructured materials, etc.

The aim of this thesis has been to study the amphiphilic properties of hydrophobically

modified poly(propylene imine) dendrimers and their potentlal applications. Poly(propylene

imine) dendrimers present some important advantages for the modification with hydrophobic

units to obtain amphiphilic macromolecules. They are relatively easy to modify due to their

high content of active functional groups, such as primary amines, and they possess a high

number of sites that can be easily protonated in aqueous media, therefore containing a high

polar hydrophilic character. In a previous studyin our group, Jan van Hest used two different

approaches to obtain well-defined amphiphilic molecules based on poly(propylene imine)

dendrimers.

In a first approach, a monofunctional polystyrene chain was used as core for the

synthesis of poly(propylene imine) dendrons of different generations to obtain amphiphilic

diblock copolymers. The behaviour of these block copolymers in water and in the bulk, have

been studled and described in this thesis. PS-Dendrimer diblock copolymers have much in

common with linear block copolymers such as size, stability of the aggregates in water and

stiffness of the hydrophobic cores, but besides that, they present new advantages, like the

possibility to tune the aggregate shape and its chemica! functionality as well in a controlled

manner. Preliminary results, not showed in this thesis, indicated that PS-Dendrimer block

copolymers are able to form microemulsions in organic solvents such as toluene.

Microemulsion polymerization of styrene using PS-Dendrimer block copolymers as

surfactants may be possible without the phase separation problems usually involved in this

163

process, when low molecular weight surfactants are used for that purpose. On the other hand,

the multiple-step synthesis is Jaborious and time consuming and it could represent a harrier

for their u se on a large scale.

Therefore, we may conclude that PS-Poly(propylene imine) dendrimers are suitable

surfactants fortheir use in specific functions where stability, control on the kind of aggregate,

and well-defined chemica] structure with a high content of functional groups is required.

In a second approach, the endgroups of poly(propylene imine) dendrimers were

modified with alkyl chains to obtain dendritic unimolecular micelles. Alkyl modified

dendrimers are very feasible molecules due to their flexibility and, on the eontrary than in the

case of PS-Dendrimer block copolymers, they are easy to obtain. 3 years after the first

synthesis was published, numerous possibilities are found for their application in many

different areas and processes such as Jiquid-liquid extraction of dyes, as additives in

polymerie matrixes to improve the solubility of small molecules, or to LC matcrials to

influence their response to electrical fields, photo-responsive aggregates for drug delivery,

self-reinforced aggregates as templates for biomineralization, gel formation, polymer

electronics, etc.

This type of amphiphilic dendrimers represent a new kind of flexible, well-defined building

blocks. They have attracted many interest due to their versatility and easy production,

showing many possibilities for their use in numerous applications. Therefore we can expect

that, in a near future, they wiJl play an important role in disciplines as varied as polymerie

matcrials and biomedical sciences.

164

Summary

Chemica) modifications of dendrimers have yielded a new kind of amphiphilic molecules

with interesting properties. In this thesis the physical and amphiphilic properties of

hydrophobically modified poly(propylene imine) dendrimers is described. Introduetion of

different functional groups in the dendrimer permits us to change the nature and size of the

molecule and, therefore, to control the physical and chemica) properties of the aggregates

formed by the amphiphile. Many different techniques can be used (e.g. electron microscopy,

Langmuir film isotherm measurements, turbidity measurements, fluorescence probe studies,

etc.) to obtain important parameters, like critica] micelle concentration, dimensions of the

headgroup, diffusion of small molecules into the aggregates, aggregates stability, form, size,

and permeability, etc. The application of these techniques in the study of amphiphilic

dendrimers is discussed in Chapter 2.

In Chapters 3 and 4, the amphiphilic properties polystyrene-poly(propylene imine) dendrimer

block copolymers in solution, at the air/water interface and in the solid state are described.

Endgroup modification of the dendrimer permits us to change the functionality of the

headgroup and, therefore, to control the physical and chemica] properties of the

superamphiphile. Small angle X-ray scattering (SAX.S) measurements and transmission

electron microscopy (TEM) show evidence of mieropbase separation of the block copolymers

in the solid state. The micro-lattice morphology was found to be highly dependent on the

dendrimer generation. In solution, TEM shows that the aggregation of the amphiphiles is

highly dependent on the generation of the dendrimer block as well. At the interface,

165

amphiphilic dendrimers offer great potential as interfacial modification agents because of

their polymerie nature and high content of functional end groups. The formation of layers

based on dendritic molecules is of particular interest because of the possibility to introduce

large changes in the surface properties of different substrates (wettability, adhesion, optica!

properties, etc.) with a small amount of materiaL

The second kind of dendritic amphiphiles (based on poly(propylene imine) dendrimers

modified with palmitoyl chains) is treated in Chapters 5 and 6. Generally, dendrimers are

presented as spherical objects with a well-defined architecture. But in the case of the alkyl

modified dendrimers the shape is only on the average sphericaL In aqueous solutions at pH <

9, the dendrimer adopts a highly distorted conformation, far from spherical; the dendritic core

of the molecule acts as head group and the long aliphatic chains as hydrophobic part of the

molecule, as a result very stabie vesicles are obtained. These findings introduce a new

concept conceming the flexibility of the dendrimers and their ability to form self-assembled

structures.

Investigations on the solubility of palmitoyl modified dendrimers in water in the presence of

low molecular weight surfactants suggest a cooperative organization of small amphiphiles

and the alkyl chains of the dendritie micelle. The dendritic micelle acts as a matrix in which

the surfactant complexes by hydrophobic interactions. These supramolecular aggregates

remain stabie in water for several weeks, the dendrimer can be used as guest molecule for the

encapsulation of small guests, while the peripheral shell can be used as template for

mineralization or polymerization of the small surfactant headgroups. In the Epilogue, the

scope and limitations of amphiphilic dendrimers basedon poly(propylene imine) dendrimers

is discussed.

166

Resumen

Las modificaciones quîmicas en dendrimeros han producido un nuevo tipo de moleculas con

interesames caracteristicas anfifîlicas. En esta tesis se describen las propierlades fisicas y

tensoactivas de poli(propileno imino) dendrîmeros modificados con grupos hidrofóbicos. La

introducción de diferentes grupos tuneionales en el dendrîmero nos permite cambiar la

naturaleza quimica y el tamafio de la molecula y, por ello, control ar las propiedades fîsicas y

quîmicas de los agregados formados por el surfactante. Numerosas técnicas pueden ser usadas

(microscopia electrónica, curvas isotérmicas en monocapas de Langmuir, medidas de

turbidez, fluorescencia, etc.) para obtener importantes parámetros tales como concentración

critica de micela, dimensiones de la cabeza del amfifilo, difusión de pequefias moleculas en

los agregados, estabilidad de los agregados, forma, tamafio, permeabilidad, etc. La aplicación

de estas técnicas en el estudio de dendrîmeros anfifflicos es discutida en el Capftulo 2.

En los Capitulos 3 y 4, se describen las propiedades anfifflicas de los copolîmeros de bloque

poliestireno-poli(propileno imino) dendrimero en disolución, en la interfase aire/agua y en el

estado sólido.

PS-dendr-(COOH),.

Los resultados de medidas de difracción de rayos X (SAXS) y mieroscopla de transmisión de

electrones (TEM) han demostrado que en el estado sólido se produce microseparación de

fases de los copolimeros de bloque. La morfologîa de esta red microscópica depende de la

generación del dendrimero. En disolución, fotograflas de TEM demuestran que la forma de

los agregados depende también de la generación del dendrimero de que se trate. En la

interfase, los dendrîmeros ofrecen un alto potencial en su uso como agentes tensoactivos

167

debido a su naturaleza macromolecular y a su alto contenido en grupos funcionales. La

formación de pelkulas basadas en moleculas dendriméricas tiene un interés particular ya que

ofrece la posibilidad de alterar las propiedades del substrato (impermeabilidad, tensión

Un segundo tipo de dendrimeros anfifflicos (basados en poli(propileno imino) dendrimeros

modificados con cadenas alquflicas) se trata en los Capftulos 5 y 6. Comunmente, los

dendrimeros se representan como moleculas esféricas con una estructura qufmica muy bien

definida. Sin embargo, en el caso de dendrîmeros que contienen cadenas alqm1icas, Ja forma

sólo es esférica bajo determinadas condiciones. En disoluciones acuosas a pH < 9, el

dendrîmero es capaz de adoptar una conformación altamente distorsionada, que dista mucho

de ser esfériea; el nucleo dendrimérico de la molecula ejerce la función de cabeza de un

anfifilo y las largas cadenas alifáticas la de la cola del anfifilo, el resultado es la formación de

un agregado vesicular altamente estable. El descubrimiento de este drástico cambio en

conformación del dendrfmero, introduce un nuevo concepto en Jo que concieme a la

flexibilidad de los dendrfmeros y su capacidad para formar agregados de forma espontanea.

Los resultados de las investigaciones realizadas sobre la solubilidad de dendrîmeros

palmitîlicos en agua cuando se afiaden detergentes, sugieren una organización cooperativa de

los detergentes y las cadenas palmitflicas del dendrîmero. El denddrnero actua como un

soporte en el cual los surfactantes se organizan debido a interacciones hidrofóbicas. Estos

agregados supramoleculares son estables en agua durante semanas. El denddrnero se puede

utilizar para alojar y encapsular pequefias moleculas en su interior, ruientras que la capa

periférica se puede usar como soporte para mineralización o polimerización de los

detergentes. En el Epflogo, se discute sobre las posibilidades y limitaciones de los anfifilos

basados en poli(propileno imino) dendrfmeros.

168

Curriculum vitae

Cristina Román Vas werd geboren op 20 mei 1969 te Tarragona

(Spanje). Na een VWO-opleiding aan de Centro de Ensefîanzas

Integradas in Alcalá de Henares (Madrid) begon zij met de studie

Scheikundige Wetenschappen aan de Universiteit Alcalá de Henares

(Madrid). Na de toekenning in 1992 van een jonge onderzoekers beurs

van het Spaanse Ministerie voor Opleiding en Wetenschap, voor

onderzoek aan de synthese en optische eigenschappen van aromatische polyesters in de

vakgroep Fysiche en Makromoleculaire Chemie in de Universiteit Alcalá de Henares

(Madrid), kreeg zij een Europese Brasmus beurs voor een afstudeerstage binnen de vakgroep

Fysiche en Makromoleculaire Chemie in de Rijks Universiteit Leiden (RUL). De opdracht

betrof onderzoek naar het dielectrische en elektro-optische Kerr effect in tertiair water-in-olie

microemulsies. In 1993 werd deze opleiding afgerond. In 1994 begon zij het

promotieonderzoek in de vakgroep Makromoleculaire en Organische Chemie aan de

Technische Universiteit Eindhoven. De resultaten van het onderzoek staan beschreven in dit

proefschrift.

169

List of publications

Natura[ abundance 15N-NMR spectroscopie investigations of poly(propylene imine) dendrimers. Van Genderen, M.H.P.; Baars, M.W.P.L.; Elissen-Román, C.; de Brabander-van den Berg, E.M.M.; Meijer, E.W. PMSE 1995, 73, 336. Abstr. Pap. Am. Chem. Soc. 1995, 20, 179.

Polystyrene-poly(propylene imine) dendrimer block copolymers: a new class of amphiphiles. Van Hest, J.C.M.; Elissen-Román, C.; Baars, M.W.P.L.; Delnoye, D.A.P.; van Genderen, M.H.P.; Meijer, E.W. PMSE 1995, 73, 281. Abstr. Pap. Am. Chem. Soc. 1995, JO, 149.

Acid-functionalized amphiphiles derived from polystyrene-poly(propylene imine)dendrimers, with a pH-dependent aggregation. Van Hest, J.C.M.; Baars, M.W.P.L.; Elissen-Román, C.; van Genderen, M.H.P.; Meijer, E.W. Macromolecules 1995, 28, 6689.

Polystyrene-poly(propylene imine) dendrimers: synthesis, characterization, and association behavior of a new class of amphiphiles. Van Hest, J.C.M.; Delnoye, D.A.P.; Baars, M.W.P.L.; Elissen-Román, C.; van Genderen, M.H.P.; Meijer, E.W. Chem. Eur. J. 1996, 2, 1616.

Amphiphilic block copolymers based on quaternized poly(propylene imine) dendrimers. Elissen-Román, C.; van Hest, 1 .C.M.; Baars, M.W.P.L.; van Genderen, M.H.P.; Meijer, E.W. PMSE 1997, 77, 145.

Protonafion mechanism of poly(propylene imine) dendrimers and some associated oligo amines. Koper, G.J.M.; van Genderen, M.H.P.; Elissen-Román, C.; Baars, M.W.P.L.; Meijer, E.W.; Borkovec, M. J. Am. Chem. Soc. 1997,119, 6512.

Amphiphilic dendrimers as building blocks in supramolecu/ar assemblies. Schenning, A.P.H.J.; Elissen-Román, C.; Weener, J.W.; Baars, M.W.P.L.; van der Gaast, S.J.; Meijer, E.W. J. Am. Chem. Soc. 1998, 120, 8199.

Microphase Separation of Diblock Copolymers consiStmg of Polystyrene and AcidFunctionalized Poly(Propylene !mine) Dendrimers. Román, C.; Fischer, H.R.; Meijer, E. W. Macromolecules 1999, submitted.

Amorphous calcium carbonare and calcite can coexist in a synthetic system. Donners, J.J.J.M.; Román, C.; Heywood, B.R.; Meijer, E.W.; Nolte, R.N.; Schenning, A.P.H.J.; Sommerdijk, N.A.J.M. Nature 1999, submitted.

170

Dankwoord

Het voltooien van een promotie is natuurlijk niet alleen de verdienste van één persoon. Er zijn

veel mensen die op een directe of indirecte manier een belangrijke bijdrage geleverd hebben

aan het werk beschreven in dit proefschrift en ik wil ze daarvoor van harte bedanken.

Alleeerst dank aan mijn promotor Prof. Bert Meijer. Ik kon altijd bij hem terecht voor allerlei

problemen, ongeacht hoe druk hij het had. Ik kreeg van hem niet alleen een beter inzicht in de

wetenschap, maar hij wist ook moed aan mij te geven, wanneer de resultaten van het

onderzoek tegenvielen, zodat ik met een vernieuwd animo en enthousiasme terug aan het

werk ging. Marcel van Genderen wil ik bedanken voor de kritische en heldere noot die hij aan

de begeleiding van mijn werk heeft toegevoegd, alsmede voor de effectieve en snelle

correctie van mijn hoofstukken. Prof. Newkome, I want to thank you for coming all the way

to The Netherlands, to take part in my examination committee and for the corrections to my

manuscript. I appreciated the interest you showed in my work. Prof. Nolte wil ik graag

bedanken voor zijn deelname aan de kerncommissie en het enthousiasme dat hij voor mijn

onderzoek getoond heeft Ook alle andere leden van de commissie wil ik in dit dankwoord

bedanken.

Afstudeerders en stagiaires, bedankt voor het waardevolle werk dat jullie hebben verricht.

Linda Anssems, wiens onderzoek aan PS-dendr-(NHz)8 vesicles een belangrijke bijdrage

heeft geleverd aan Hoofstuk 3. Alicia Marcos Ramos en Beatriz Ruiz Fernández, de Brasmus

studenten, voor hun enthousiasme en doorzettings vermogen in de moeilijke wereld van host

guest chemistry. Ook (samen met Elena) bedankt voor de spaansesfeer in het lab (El ultimo

de la fila, Seguridad Social, Jarabe de palo en andere muzikale talenten) die mij zo leuke

herinneringen bracht en mij ook een beetje op de hoogte gehouden heeft van de laatste trends

in mi tierra. Ik wens jullie allemaal het beste in de toekomst.

Hartruut Fischer heeft een belangrijke bijdrage gehad aan Hoofstuk 4. Bedankt voor de SAXS

en TEM metingen in de vaste fase. Albert Schenning, "Appie", voor zijn nuttige advies. Een

groot deel van het werk beschreven in Hoofstukken 5 en 6 is te danken aan zijn innovatieve

wetenschappelijke inzicht.

171

Talloze geweldige mensen in andere groepen en universiteiten hebben mij geholpen met hun

ervaring in verschillende gebieden van de chemie. In Nijmegen, Huub Geurts en Peter

Buynsters, die mij altijd openhartig en behulpzaam ontvangen hebben toen ik daar TEM

metingen ging doen. Martin Feiters en Paul van Kan, bedankt voor de ESR anisotropie

metingen. Nico Sommerdijk en Jack Donners, de eerste vanaf het begin, de tweede in een

later stadium van mijn promotie, bedankt voor de hele nuttige discussies en samenwerking

waaruit zoveelleuke resultaten zijn gekomen. In Wageningen, Edwin Currie voor jouw hulp

met monolagen en de gezellige gesprekken over, onder andere, de "oude tijden in Leiden".

Joost Maas, zonder jou was het voorbereiden van de AFM samples niet gelukt. Sjerry van der

Gaast, voor de XRD metingen en voor jouw warme ontvangst op het (toen) koude en

winderige Texel. In Maastricht, Prof. Peter Frederick en Paul Bomans voor de cryo-TEM

metingen en de nuttige discussies over de resultaten. In Madrid, Ricardo Garcia en Javier

Tamayo voor de AFM metingen en vooral Álvaro San Paulo die ondanks de "meet-marathon"

zijn humor nooit verloor (dat geldt ook voor Brigitte), het was een plezier met jullie te

werken. In Leiden, Wiebke Sager voor de experimenten met microemulsies. Ger Koper voor

de theoretische achtergrond in de protonatie van polyamines, die tot een mooie publikatie

geleid heeft; Ger, ook bedankt voor de vriendelijke ontvangst en steun in Leiden in mijn

vroegere tijden als onderzoekster in Nederland, (toen mijn Nederlands nog on verdraagbaarder

was dan nu). Patricia Kooyman, in Delft, voor de HREM metingen. Wim Meijberg, in

Groningen, voor de MicroDSC.

In de TUE, Aissa Ramzi for useful discussions about mieropbase separation in palmitoyl

dendrimers; Pascal Oberndorff voor de SEM metingen; Willem Veilinga voor de AFM

metingen in oplossing en Dominique Hubert voor de DLS metingen en de verhelderende

uitleg over de turbiditeits metingen.

Het dendrimeren cluster wil ik bedanken voor een heel plezierige samenwerking, niet alleen

op het wetenschappelijke vlak, maar ook op het persoonlijke. Jan van Hest heeft mij met

enthousiasme en geduld geholpen en begeleid in een vroeg stadium van dit werk, toen ik nog

maar pas in de diepe wateren van de amphiphiel dendrimeren dook. Rob, bedankt voor de

"zoete" surfactanten. Maurice, je bent altijd behulpzaam en open voor discussie geweest. Jan

Willem, jouw steun, meestal met een scherpe geest gebracht, is belangrijk geweest voor mij.

Onze reis naar London blijkt achteraf niet alleen heel nuttig te zijn geweest, maar het was ook

ontzettend leuk. Tonny Bosman, "el Rico", bedankt voor jouw hulp met de metaal-

172

complexatie experimenten en voor de humor die je overal verspreidt. Met jou was er altijd

iets te beleven in het Lab. Succes met jouw promotie, compafiero de fatigas.

Mijn kamergenoten, Henk (bedankt voor het maken van de ethyleen glycol surfactanten, jouw

visie van de wetenschap en het leven waardeer ik zeer), Anja (in onze reis naar Grenoble heb

ik echt plezier gehad, niet alleen van het lekkere eten, maar ook van jouw gezelschap, je bent

een fantastische reisgenote),en last but not least Ron, bedankt voor jullie aangename

kameraadschap.

Jef Vekemans, Rene lanssen en Rint Sijbesma, wil ik bedanken voor hun wijze advies,

wanneer het over synthetische, physisch organische of supramoleculaire chemie ging. Bas

voor de synthese van verbindingen. Marwijn voor de hulp met de polarizers in de

fluorescentie anisotropie metingen. Koen voor de hulp in het eeuwige gevecht tussen mens en

computer. Joost van Dongen voor de MS-metingen. Hanneke, Ingrid, Hans en Henk voor het

zorgen dat alles altijd op rolletjes liep.

Ky (het was een plezier om het kerstdiner met jou te organiseren. Wanneer spreken we weer

af om te gaan dansen?), Emiel, Toine, Elena, Francesca, Maurice, Henk, Monique en

Marwijn voor het lekkere koken en de gezellige avonden.

De bovengenoemde en alle andere collega's van TOC/SMO die zo'n prettige werksfeer

gecreeerd hebben, wil ik speciaal bedanken. Ik zal jullie missen.

Ten slotte, ik dank Paul (mi chico), voor al de liefde, het geduld, de steun en het begrip die jij

getoond hebt in deze drukke jaren. Mijn ouders (Lauren en Maria), mijn broers (Javier en

Carlos) en andere familieleden voor de belangstelling in mijn werk, de genegenheid en de

ondersteuning waarmee ze me altijd omsingeld hebben, zelfs op 2000 km afstand ...

Zonder jullie was het nooit gelukt.

173

Agradecimiento

En la realización de un doctorado no Ie corresponde todo el mérito a una sóla persona. Hay

muchas personas que de una forma directa o indirecta han contribuide de una manera

importante en el trabajo descrito en esta tesis; por ello quisiera corresponderles con mi más

sincero agradecimiento.

En primer Jugar mi directer de tesis Prof. Bert Meijer. Siempre pude acudir a él cuando tuve

un problema, independientemente de Jo ocupado que estuviese. No sólo me ayudó a

desarrollarme como cientffica, sino que además supo inspirarme valor y confianza cuando los

resultados de la investigación no eran los deseados, de forma que podia volver al trabajo con

un renovado ánimo. A Marcel van Genderen Ie agradezco el tinte crftico y conciso que supo

afiadir a la supervisión de mi trabajo, asf como su rapidez y efectividad en la corrección de la

tesis. Prof. Newkome, quisiera agradecerle su interés en mi trabajo, asf como el largo viaje

que va a realizar a los Paises Bajos para tornar parte en el comité examinador de mi tesis. Al

Prof. Nolte quisiera agradecerle el entusiasmo con que ha seguido mi trabajo y por participar

en el comité examinador. De la misma forma, quiero dar las gracias a todos los demás

mierobros del comité.

Tesinandos y estudiantes, gracias por el valioso trabajo que habeis realizado. Linda Anssems,

cuya investigación sobre las vesfculas de PS-dendr-(NH2)8 ha significado una importante

contribución al Capitulo 3. Alicia Marcos Ramos y Beatriz Ruiz Femández, las becarias

Erasmus, por su entusiasmo y tenacidad en el dificil mundo de la qufmica supramolecular.

También (junto con Elena) gracias por el ambiente espafiol que reinó en el laboratorio (El

ultimo de lafila, Seguridad Social, Jarabe depalo y otros talentos) que me trajo tan buenos

recuerdos y me puso un poco al dfa de las ûltimas tendendas en mi lejana tierra. Os deseo

todo Jo mejor en el futuro.

Hartrnut Fischer ha aportado una importante contribución al trabajo descrito en el Capftulo 4.

Gracias por las medidas de SAXS y TEM en la fase sólida. Albert Schenning, "Appie", por

sus ûtiles consejos. Gran parte del trabajo descrito en los Capftulos 5 y 6 se debe a su

innovativo talento cientffico.

174

Son muchas las fantásticas personas de otros grupos o uni versidades que me han ayudado con

su experiencia en distintas areas de la qufmica. En Nijmegen, Huub Geurts y Peter Buynsters,

que siempre me recibieron con los brazos abiertos cuando iba a utilizar el microscopio

electrónico. Martin Feiters y Paul van Kan, gracias por las medidas de anisotropfa en ESR.

Nico Sommerdijk y Jack Donners, el primero desdeel principio, el segundo al final de mi

doctorado, gracias por las ótiles discusiones y la colaboración de la que han surgido tan

buenos resultados. En Wageningen. Edwin Currie por tu ayuda con las monocapas y las

agradables charlas sobre, entre otras cosas, nuestros "viejos tiempos en Leiden". Joost Maas,

sin ti no habrfa sido posible la preparación de las rouestras para AFM. Sjerry van der Gaast,

por las medidas de XRD y por tu cálida acogida en la (por entonces) tormentosa y frfa isla de

Texel. En Maastricht, Prof. Peter Frederick y Paul Bomans por las medidas en cryo-TEM y

las esclarecedoras discusiones sobre los resultados. En Madrid, Ricardo Garcîa y Javier

Tamayo por las medidas de AFM y sobre todo Álvaro San Paulo que ha pesar del AFM

maratón nunca perdió su sentido del humor (lo mismo que Brigitte Folmer), fue un placer

trabajar con vosotros. En Leiden, Wiebke Sager por los experimentos con microemulsiones.

Ger Koper, por los cálculos teóricos en la protonación de poliaminas, que resultaron en un

buen articulo; Ger, también gracias por la amistosa acogida y apoyo en Leiden en mis

primeros moroentos como investigadora en Holanda, (cuando mi holandés era todavia más

insoportable que ahora). Patricia Kooyman, en Delft, por las tomas en el microscopio

electrónico de aha resolución. Wim Meijberg, en Groningen, por las de Microcalorimetrîa.

En la Universidad de Eindhoven, Aissa Ramzi por los consejos sobre la microseparación de

fase en los dendrfmeros palmfticos. Pascal Obemdorff por las tomas de SEM; Willem

Veilinga por las medidas de AFM en disolución y Dominique Hubert por las de DLS y las

aclaraciones sobre las medidas de turbidez.

Al grupode dendrimeros quiero agradecerles la amistosa colaboración, no sólo en el terreno

profesional, sino también en el personal. Jan van Hest me ayudó con entusiasmo y paciencia

al principio de la tesis, cuando acababa de sumergirme en las profundas aguas de los

dendrimeros tensoactivos. Rob, gracias por los "dulces" surfactantes. Maurice, siempre

dispuesto a ayudar y a dialogar. Jan Willem, tu apoyo, casi siempre acompaiiado de una

mente despierta, ha sido importante para mi. Nuestro viaje a Londres no sólo resulto

extremadamente ótil al final, sino que además fue muy divertido. Tonny Bosman, "el Rico",

gracias por tu ayuda con los experimentos en complejos metálicos y por ese buen humor que

175

derrocbas. Contigo siempre babfa fiesta en el laboratoria. Suerte con tu tesis, campaftera de

fatigas.

Mis compafieros de oficina, Henk (gracias por los surfactantes, valoro mucbo tu visión de la

ciencia y de la vida), Anja (me lo pase muy bien en nuestro viaje a Grenoble, no solo por la

deliciosa comida, sino también por tu compafiîa. Eres una compafiera de viaje estupenda) y

por ûltimo, pero no por eso menos, Ron, gracias a todos por vuestro compafierismo.

lef Vekemans, Rene lanssen y Rint Sijbesma, quiero agradeceros vuestros sabios consejos

cuando sobre sfntesis, ffsica orgánica o supramolecular se trataba. Bas por la sintesis de

compuestos. Marwijn por la ayuda con los polarizadores en las medidas de anisotropfa de

fluorescencia. Koen por tu ayuda en la etema lucba entre el hombre y su ordenador. Joost van

Dongen por las medidas de masa. Hanneke, lngrid, Hans y Henk por cuidar de que todo vaya

siempre sobre ruedas.

Ky (fue un placer organizar la cena de navidad contigo. ;,Cuándo quedamos de nuevo parair a

moverel esqueleto?), Emiel, Toine, Elena, Francesca, Maurice, Henk, Monique en Marwijn

por las ricas cenas y agradables sobremesas.

A todos los nambrados y otros colegas delgrupode orgánica que ban contribuido a crear este

buen ambiente quisiera daros las gracias en especial. Os voy a ecbar de menos.

Por ûltimo, gracias a Paul (mi chica), portodoel amor, paciencia, apoyo y comprensión que

me bas dado en estos afios tan atareados. A mis padres (Lauren y Marîa), hermanos (Javier y

Carlos), y demás mierobros de la familia por el interés que habéis demostrado en mi trabajo y

el carifio y apoyo con el que siempre me babéis rodeado, incluso a 2000 km de distancia ....

Sin vosotros nunca habrfa sido posible.

176

STELLINGEN

behorende bij het proefschrift


door

Cristina Román V as

1. Een dendrimeer doosje dat slechts gedurende een fractie van een seconde "dicht gaat",

bezit niet de benodigde eigenschappen voor toepassing in drug delivery systemen.

A. Archut, G.C. Azzellini, V. Balzani, L. De Cola, F. Vögtle J. Am. Chem. Soc. 1998, 120, 12187. A.

Archut, F. Vögtle, L. De Cola, G.C. Azzellini, V. Balzani, P.S. Ramanujam, R.H. Berg Chem. Eur. J.

1998, 4, 669.

2. Polyethyleen oxides zijn betere aniongeleiders dan kationgeleiders en zijn dus niet de

polyelektrolieten par excellence voor lithiumbatten jen.

G. S. Mac Glashan, Y.G. Andreev, P.G. Bruce Nature 1999, 398, 792. P.P. Soo, B. Huang, Y. Jang,

Y.M. Chiang, D.R. Sadoway, A.M. Mayes J. Electrochem. Soc. 1999, 146, 32. R. Frech, S. Chintapalli,

P.G. Bruce, C.A. Vincent Macronu;lecules 1999, 32, 808.

3. De lage oplosbaarheid van pyreen in polypropyleenimine dendrimeren beperkt in hoge

mate het gebruik van deze gastheer-gast systemen voor gecontroleerde afgifte.

G. Pistolis, A. Malliaris, D. Tsiourvas, C.M. Paleos Chem. Eur. J. 1999,5, 1440.

4. Een rijke fantasie te samen met literair talent mag niet de wetenschappelijke inhoud

overschaduwen.

D.A. Tomalia Adv. Mater. 1994,6,529. Macromol. Symp. 1996, JOl, 243.

5. Een smerige, onoplosbare brij op de bodem van een kolf verdient soms meer aandacht

dan het oorspronkelijke gewenste produkt.

L. H. Backeland J. lnd. Eng. Chem. 1909, /, 149, 204, 545; Chem. Ztg. 1909,33, 857.

6. Of je nu het DNA molecuul bestudeert of de manier waarop het onthuld is, het geheim

van het leven is complementariteit.

J. D. Watson. W. Gratzer Nature 1997, 386, 344. J. D. Watson, The double helix 1968. F. Crick What mad pursuit : a personal view of scientific discovery. New Y ork : Basic Books, 1988. F. Crick Time 1993, 141,58.

7. La cocina espaiiola no sólo es rica, sino que además podrfa ser la clave para una larga

vida.

M. Jang, L. Cai, G.O. Udeani, K.V. Slowing, C.F. Thomas, C.W.W. Beecher, H.H.S. Fong, N.R. Farnsworth, A.D. Kinghom, R.G. Metha, R.C. Moon, J.M. Pezzuto Science 1997, 275, 218. A.I. Romero Pérez, R.M. Lamuela Raventós, A.L. Waterhouse, M.C. de la Torre Boronat J. Agric. Food Chem. 1996, 44, 2124. A.L. Klatsky, M.A. Armstromg, G.D. Friedman Ann. Intern. Med. 1992, 117, 646. A.L. Klatsky Clin. Exp. Res. 1994, 18, 88. L. Shapiro Newsweek 1994, 124, 56. J. Cox Men's health 1990, 5, 43.

8. Er is geen grotere mislukking dan niets te leren van een gemaakte fout.

9. Het "millennium probleem" toont aan dat ondanks het feit dat onze beschaving reeds

lang bestaat, we blijkbaar nog niet geleerd hebben om vooruit te denken.

10. Het is triest dat in een verenigd Europa separatistische gevoelens vaak leiden tot

lokale conflicten gepaard met geweld.

11. Domme mensen ervaren het heden vanuit het verleden.

Verstandige mensen speculeren op de toekomst.

Wijze mensen leven in het nu.

12. Sobre el dinero no se habla, el dinero se utiliza.

13. In tegenstelling tot de uitdrukking, is er meer geluk in de tien vliegende vogels dan in

die ene in onze hand.

14. Daar waar weinig rechtvaardigheid is, is het gevaarlijk om gelijk te hebben.

Amphiphilic dendrimers - Pure · Amphiphilic Dendrimers PROEFSCHRlFT ter verkrijging van de graad...

Documents

Transcript of Amphiphilic dendrimers - Pure · Amphiphilic Dendrimers PROEFSCHRlFT ter verkrijging van de graad...