Chemistry and physics of supramolecular systems
Organic Chemistry
ORC-20806
Ton Marcelis Organic Chemistry Ernst Sudhölter Organic Chemistry CONTENTS
CHAPTER 1: INTRODUCTION...... 3 1.1 ORDERED SYSTEMS...... 3 1.2 MOLECULES AND MATERIALS ...... 4 1.3 SUPRAMOLECULAR SYSTEMS ...... 6 CHAPTER 2: SURFACTANTS...... 7 2.1 INTRODUCTION ...... 7 2.2 SURFACTANT PARAMETER ...... 8 2.3 MICELLES ...... 10 2.3.1 Micelle models...... 11 2.3.2 Dynamics of micelles ...... 11 2.3.3 Counter ion binding...... 12 2.4 FACTORS THAT INFLUENCE THE CMC...... 13 2.5 METHODS FOR DETERMINING THE CMC ...... 14 2.5.1 Surface tension...... 14 2.5.2 Conductivity...... 15 2.5.3 Light scattering...... 15 2.5.4 Solubilisation ...... 15 2.6 APPLICATIONS OF MICELLES ...... 16 2.6.1 Cleaning agents ...... 16 2.6.2 Micellar catalysis...... 17 2.6.3 Application of micelles in chiral separations...... 19 2.7 GENERAL PHASE BEHAVIOUR OF SURFACTANTS...... 22 2.8 LYOTROPIC LIQUID CRYSTALS ...... 24 2.9 SPECIAL AGGREGATES OF SINGLE-TAILED SURFACTANTS...... 26 CHAPTER 3: VESICLES...... 28 3.1 INTRODUCTION ...... 28 3.2 PREPARATION OF VESICLES...... 30 3.2.1 Preparation of vesicles by sonication ...... 30 3.2.2 Preparation of vesicles by the alcohol injection method...... 31 3.2.3 The preparation of giant vesicles...... 31 3.3 VISUALISATION OF VESICLES BY ELECTRON MICROSCOPY ...... 32 3.4 PHASE TRANSITIONS IN AMPHIPHILIC BILAYERS...... 33 3.5 TRAPPING OF GUEST MOLECULES IN A VESICLE COMPARTMENT...... 34 3.6 FUSION OF VESICLES ...... 35 3.7 SOLUBILISATION ...... 36 3.8 APPLICATIONS OF VESICLES...... 38 3.8.1 Photodynamic therapy ...... 38 3.8.2 DNA-transfection...... 39 3.8.3 Controlled drug release...... 39 3.9 LITERATURE ...... 40 CHAPTER 4: MONOLAYERS ...... 41 4.1 INTRODUCTION ...... 41 4.2 LANGMUIR TROUGH...... 42 4.3 PHASE BEHAVIOUR...... 42 4.4 RELATION BETWEEN STRUCTURE AND MONOLAYER BEHAVIOUR...... 44 4.5 VISUALISATION OF MONOLAYERS...... 45 4.6 MOLECULAR RECOGNITION AT MONOLAYERS...... 47 4.6.1 Chiral recognition of taste compounds...... 47 4.6.2 Interactions with chiral amino acids ...... 47 4.6.3 Biotine - streptavidine interactions...... 48 4.7 APPLICATIONS OF MONOLAYERS...... 49 4.8 LITERATURE ...... 49 4.9 INTERMEZZO: AMPHOTROPES ...... 50
- 1 - CHAPTER 5: POLYMERS...... 57 5.1 INTRODUCTION ...... 57 5.2 POLYMERISATION REACTIONS...... 57 5.3 MECHANISMS FOR THE ADDITION POLYMERISATION...... 58 5.3.1 Radical polymerisation...... 59 5.3.2 Ionic polymerisation ...... 60 5.4 NON-LINEAR POLYMERS ...... 61 5.5 LINEAR POLYMERS FROM DIFFERENT MONOMERS...... 64 5.6 GENERAL PROPERTIES OF LINEAR POLYMERS...... 65 5.6.1 Molecular weight...... 65 5.6.2 Phase behaviour ...... 66 5.7 LITERATURE ...... 68 CHAPTER 6: LIQUID CRYSTALS...... 69 6.1 PHASES AND ORDERING ...... 69 6.2 TYPES OF LIQUID CRYSTALS...... 71 6.3 GENERAL DESCRIPTION OF LIQUID CRYSTALLINE PHASES ...... 72 6.4 METHODS FOR INVESTIGATING LIQUID CRYSTALS...... 73 6.4.1 Polarisation microscopy ...... 73 6.4.2 Differential Scanning Calorimetry (DSC) ...... 74 6.4.3 Röntgen reflection analysis...... 77 6.5 DESCRIPTION OF SOME COMMON LIQUID CRYSTALLINE PHASES...... 78 6.5.1 Nematic phase...... 78 6.5.2 Chiral nematic phase...... 79 6.5.4 Smectic phases...... 80 6.5.5 Chiral smectic phases ...... 83 6.5.6 Discotic liquid crystalline phases ...... 84 6.6 STRUCTURE - PROPERTY RELATIONS ...... 85 6.7 APPLICATIONS OF LIQUID CRYSTALS...... 87 6.8 LIQUID CRYSTALLINE POLYMERS...... 89 6.8.1 Main chain liquid crystalline polymers...... 89 6.8.2 Side chain liquid crystalline polymers...... 90 6.9 AN EXAMPLE OF RESEARCH ON LIQUID CRYSTALLINE POLYMERS ...... 92 6.9.1 Liquid crystalline polymers with a smectic-B ordering...... 93 6.9.2 A multilayer on a surface...... 94 6.10 AMPHOTROPICS (SURFACTANTS WITH THERMOTROPIC LIQUID CRYSTALLINE PROPERTIES)...... 94 6.11 LITERATURE ...... 98 6.12 INTERMEZZO: SPECIAL PLASTICS...... 99
- 2 - Chapter 1: Introduction
1.1 Ordered systems
There is a need for miniaturisation in diverse technology areas, for example in microelectronics, but also for mechanical and medical applications. The reasons are often energy reduction and the wish to make the devices portable. In electronics shorter connections also leads to increased speeds of information transfer. A lot of effort and technological innovation is required to make smaller structures from larger ones and also to make small structures in materials. Structures smaller than 1 µm (10-6 m) are very difficult to achieve with classical methods, but also techniques like etching and photolithography are often not useful in this domain. Another approach could be to build these structures from molecules. The typical size of molecules is around 1 nm (10-9 m). In order to obtain technologically interesting structures, a large number of molecules would have to aggregate, preferably spontaneous, to structures that have the desired properties, for example, conducting wires. This building up of materials from molecules is often called supramolecular chemistry or nanotechnology. In both areas, decrease of the size of larger structures and the building up of structures from molecules, considerable progress is made in the last decade. However, it will probably take several years before developments in both directions start to overlap and one can truly speak of nanotechnology. Research that is conducted at this moment already leads to new materials with special properties and applications or for example very small devices like pumps or for example a complete gas chromatograph on a chip.
Size of structures (nm)
1000 materials
100 Nano structures 10
molecules 1 year 2000 2050
Figure 1: Hypothetical time path in which nanostructures could be reached from materials or from molecules.
- 3 - A living organism is a very beautiful example of an ordered supramolecular system. Besides the build-up of organs from cells, the cell itself is build-up from constituent parts that can be considered as supramolecular systems.
Figure 2 Schematic representation of a cell membrane
Cell membranes are bilayer systems that are build up from amphiphilic phospholipids. DNA contains stacked bases held together by hydrogen bonds. Proteins and enzymes have an ordered structure, by which it can perform its structuring and catalytic properties. Other parts like the cytoskeleton also consist of ordered molecular structures. Furthermore, various receptors are present in the cell or on its outer surface (enzymes, sensors, ion channels, and glycoproteins) whose function strongly depends on their mutual ordering or the ordering (interaction) with other molecules. In this respect, much can be learned from nature. Nanostructures with the same complexity as living cells are much too complicated to mimic or reproduce from simple molecules, at least for the moment. However, it is possible to make and study ordered molecular systems that are spontaneously from “simple” molecules that have interesting and useful properties. In this course, we will look at how ordered molecular structures are “spontaneously” formed from certain molecules. We will look at what kind of material properties these structures have, how the molecular properties influence the material properties and we will treat some applications of these materials.
1.2 Molecules and materials
The materials that we know consist of atoms (or molecules). Therefore, it is logical to suppose that the properties of the materials are determined by the properties of the atoms (or molecules). We can feel and see materials but not molecules and therefore it is difficult to get a feeling for
- 4 - them. For a long time atoms and molecules were more philosophical ideas than concrete physical entities whose existence was proven without doubt. The thought experiment of the Greek philosopher Democritos is well known. He stated that if you divide a material like gold often enough in two parts, you end up with particles that can not be divided further without losing its properties. He called these smallest particles, which still retain the properties of the bulk material atoms (which means something like “can not be parted”). The English word particle that chemists and physicists often use for the smallest structures still reminds us of this. According to present day definitions and insights atoms (or molecules: electrically neutral groups of atoms that are connected with each other in a defined way) ultimately have influence on the properties of a material. Many of the material properties are however, more determined by the mutual interactions between the molecules of the material, that is they are determined by aggregates of molecules. It has been calculated that a materials property like the melting point of for example water starts to have some meaning if a group of at least 6 H2O molecules is involved.
Many of the observable and measurable properties of materials are therefore determined by the mutual interactions between the molecules of the material. The most important interactions are: a) Vanderwaals interactions b) Dipole interactions c) Hydrogen bonds d) Electrostatic interactions (interactions between ions in salts, etc) The observable macroscopic properties (materials properties) are strongly determined by the molecular properties and particularly by their structure. By structure of an (organic) molecule, we mean the way it is build-up from atoms, the configuration around the atoms and their conformation. Furthermore, the materials properties are strongly influenced by the temperature and/or the concentration in a second material or a mixture of materials, for example a solvent. The mutual interactions between molecules are responsible for the phase in which the material is present. Phases are states of matter in which the materials may be present. The best known phases are the solid state, liquid state and gas state. For a certain compound this state depends on temperature and pressure. For the matter that consists of H2O molecules, we know three well- known phases, i.e. ice, water and water vapour. At normal pressure there is a phase transition from ice to water (and back) at 0°C and from water to water vapour at 100°C.
- 5 - 1.3 Supramolecular systems
In the last paragraph, we saw that the mutual interactions between molecules determine the material properties. In several cases, these mutual interactions lead to special structures with special (functional) properties. These special aggregates are often present as a function of temperature (for example liquid crystals). This is because temperature induces motion of whole molecules (translation and rotation) or molecular fragments (vibration, etc) that is often opposed to an optimal favourable interaction. Special aggregates can also be formed in the presence of a solvent. In particular water induces the formation of special aggregates, especially by molecules that have an apolar part that does not want to be surrounded by water molecules and a polar part that particularly likes water molecules. Examples are surfactants that spontaneously form structures like micelles, bilayer vesicles, etc in the presence of water. For (bio)polymers like proteins, polysaccharides and nucleic acids the balance between mutual interactions of the monomeric units (intramolecular) and interactions between the surrounding water (intermolecular) is responsible for the special structure (folding) these molecules often have. This special structure is also important for their function. For biopolymers the structure and therefore the properties have to be in an exact balance that is influenced by temperature, pH, the presence of salts, etc. In addition, the presence of surfaces plays an important role in the aggregate structure and functionality of biopolymers. Furthermore, special aggregation structures like monolayers are formed at surfaces. That is why in this course also adhesion, adsorption and self-assembly of in particular biopolymers at surfaces (interfaces) is treated. The interfaces can be solid-liquid, but also air-liquid and liquid-liquid. Of special importance are surfaces consisting of bilayer surfactants as found in cell membranes. This whole area, which deals with the mutual interactions between molecules that lead to functional aggregates, is often called supramolecular chemistry or nanotechnology. A good understanding of these interactions is of great importance for applications of these molecules in liquid crystals technology, detergents, nanotechnology, etc. In combination with adhesion and adsorption behaviour, supramolecular chemistry is important in application areas like food technology, biotechnology, environmental technology, etc.
- 6 - Chapter 2: Surfactants
2.1 Introduction
Molecules in which the polarity in different parts of the molecule is very different are often called amphiphiles. This word is derived from Greek: amphi (both sides) and philein (love). One part of the molecule likes to be in a polar environment like water, whereas the other part prefers an apolar environment. In an aqueous medium one part is therefore hydrophilic, the other part hydrophobic. Consequently, these molecules like to be at an interface (solid – liquid, liquid – liquid or liquid – gas) especially when the polarities at the two sides of the interface are very different. Amphiphiles are therefore surface-active agents, hence their name. In the following table some groups of surfactants are given.
Table I: Types of single-tail surfactants with examples cationic N-alkylpyridinium halides N CnH2n+1 X
Alkylammonium halides R3N CnH2n+1 X
e.g. CTAB Me3N C16H33 Br (cetyltrimethylammonium bromide)
Alkylphosphonium R3P CnH2n+1 X halides
anionic metal alkylsulfates ROSO3 M
e.g. SDS C12H25OSO3 Na (sodium dodecylsulfate) O metal alkanoaten R M O 2 metal alkylphosphates R–OPO3 2M
zwitterionic alkylbetaines R N O O
alkylsulfobetaines R N OSO3
alkylammonium-N-oxides R N O
non-ionic polyoxyethylene alcohols R(OC2H4)n-OH
e.g. Brij 35 C12H25–O–(CH2CH2)23–OH
alkyl ethers of carbohydrates HO O HO O e.g β. -D-octylglucoside HO HO
Often, but not always, amphiphiles consist of a hydrophilic head group and one or more hydrophobic tails. The head group can be a cationic, anionic, zwitterionic or non-ionic group. The - 7 - hydrophobic tails often consist of a hydrocarbon chain, straight or branched and with or without double bonds, but they can also be completely or partially fluorinated or consist of a siloxane chain. Due to the hydrophobic groups amphiphiles tend to shield their hydrophobic parts from water. This can be accomplished in several ways, depending on the structural properties but also on other factors like temperature, pH and the presence of dissolved species like salts. Common types of aggregates are micelles and bilayer structures. Micelles are spherical aggregates in which usually about 50 to 100 molecules are present in such a way that the hydrophobic tails are in the middle surrounded by the hydrophilic head groups that are in contact with water. In bilayers, the amphiphiles are ordered in such a way that the molecules form two layers in which the hydrophobic parts are directed to each other and the hydrophilic head groups are on the outside in contact with water. When the bilayer is curved in such a way that a spherical shell is formed, we call these aggregates vesicles. A certain volume of water is separated from the rest of the aqueous solution by the bilayer membrane. The major constituents of cell membranes are bilayer forming amphiphiles, the phospholipids. It is not exaggerated to state that life as we know it would not be possible without amphiphiles. Due to their special properties, natural and synthetic amphiphiles are of utmost importance in many areas.
Some examples: Detergents (soaps) form micelles in which fatty substances and other dirt particles are solubilized in water. Natural detergents like bile acids help with the digestion of food by dispersing the fat components in nutrients. Its surface-active properties are used to stabilise emulsions, for example pigments in paint, fats in mayonnaise, halvarine, cremes, cosmetics, etc. Vesicle forming compounds are used to encapsulate compounds and release them only when this is desired. In pharmaceutics, this is often called “controlled release”; vesicles are often called liposomes in this field. Because many of these useful applications are determined by the special aggregation behaviour of amphiphiles, fundamental knowledge of the physico-chemical behaviour, their unusual properties and their phase behaviour are is of great importance for students in chemistry and molecular sciences.
2.2 Surfactant parameter
If we want to know which type of aggregate is formed by a certain surfactant, we have to take a closer look at the structure of the surfactant. The apolar tails tend to stick together and fill the space in the most efficient way, although this is counteracted somewhat by thermal movements. On the other hand, the head groups are hydrated and in case of ionic surfactants also carry the same charges. Therefore, they tend to repulse each other. For a good contact between the tails, it is favourable when the head group is not too bulky in relation to the cross-section of the tail and that the tail is long. Based on these considerations a packing parameter was defined by Israelachvili that
- 8 - relates the cross sectional area of the head group (ao) with the average cross-sectional area of the tail expressed as volume divided by length (vo/l). The packing parameter is now vo/aol.
Figure 1 Relation between packing parameter and type of aggregate.
When the packing parameter is about 1 the best packing is in the form of sheet-like bilayers or vesicles. With normal alkyl tails and normal head groups this is usually obtained when the polar head group contains two unbranched alkyl tails. When there is one alkyl tail present, the area of the head group is much larger than the cross section of the alkyl tail. Therefore, the molecule has in fact a cone shape and the packing parameter is <1/3. This shape is favourable for micelle formation.
- 9 - When the cross section of the tails is larger than of the head group the packing parameter is >1 and the form is favourable for the formation of reversed micelles. Amphiphiles that form reversed micelles are for example compounds that have three alkyl tails or two branched alkyl tails. They have their polar head groups in the middle of the spherical aggregate and are thus able to stabilise (droplets of) water in an apolar solvent. Although the packing parameter is very useful to estimate what type of aggregate can be formed, the determination of the actual packing parameter is not easy. Especially, the determination of ao is difficult because it is influenced by the extent of the hydration shell and this is again dependent on the presence and concentration of other ions in the solution and often also on the pH and temperature. Furthermore, it is an oversimplification to picture a surfactant as a symmetrical cone or cylinder. From the figure, it can be seen that single-tailed surfactants generally form micelles in water. In this chapter, we will only discuss single-tailed micelle-forming surfactants, whereas in the next chapter double-tailed surfactants that form vesicles are treated. First, we will treat the micellization behaviour. Then the behaviour at high concentrations will be treated, where so-called lyotropic liquid crystalline phases are formed. Finally, so-called amphotropic molecules will be treated that combine liquid crystalline behaviour and lyotropic properties.
2.3 Micelles
When a simple compound is dissolved in water, the molecules are completely solvated by water and move around independently from one another. When a salt is dissolved the electrical conductivity of the solution increases almost linearly with concentration. For micelle-forming compounds like sodium dodecylsulphate (SDS) it is found that the conductivity increases linearly until a certain concentration (~8 mM). Above this concentration, the conductivity also increases linearly, but with a much smaller slope. For many other physical properties, a change in solution properties is also found at this concentration. This concentration is related to the formation of micelles and is called the critical micelle concentration (CMC). y t i v i t c u d n o c CMC
concentration Figure 2 Change in physical properties of solution when reaching the CMC.
- 10 - 2.3.1 Micelle models At this concentration, several molecules join to form an aggregate, in which the molecules in the aggregate are in equilibrium with the non-aggregated molecules in solution. When more surfactant molecules are dissolved the concentration of micelles increases and the concentration of free surfactant molecules stays practically constant. This equilibrium can be described as follows:
Kn nS Mn
In order to describe a micelle in this way, the so-called “mass action model” the association constant of each step from monomer to micelle should be known. Experimentally this is not feasible. According to another description, the “phase separation model”, the micellar and the aqueous phase are treated as separate phases, between which exchange of molecules takes place. Both models can be used for a thermodynamic treatment of micelles, however they both do not describe all aspects of micelles accurately. The “mass action model” is better for small micelles, whereas for large micelles the “phase separation model” is better. Many experiments have indicated that micelles consist of about 20-100 molecules that form a spherical aggregate with a cross section of 30-50Å (3-5 nm). In the course of time different models have been made for the structure of the micelle. The interior of a micelle is made up of the apolar tails that behave as a liquid. Apolar compounds can be dissolved in this apolar environment. Measurements have indicated that many gauche conformations are present in the alkyl tails, even more than when the molecule is present as a free monomer. Furthermore, it has been shown that even CH2-groups close to the ends of the tails are on average often in contact with water.
Figure 3 Models for micelles. The models on the left side are more realistic than those on the right side.
2.3.2 Dynamics of micelles Micelles are very dynamical particles, that are build-up and broken down very fast. Within microseconds, individual surfactant molecules go from micelle to solution and back. Within milliseconds micelles can completely disintegrate or be formed. It is difficult to visualise micelles with techniques like electron microscopy because they are so dynamical but also because they are - 11 - so small. With a recent development in electron microscopy, cryo-electron microscopy, it is possible to detect and make images of micelles. With this technique, droplets of sample are cooled very fast to -100°C and the water vitrifies without crystallising (glass-formation). This ensures that the structures are maintained in a frozen in state and thus can be observed. As expected micelles are spherical particles with diameters of about 5 to 10 nm.
– Figure 4 Cryo-electron microscopic image of C21H43–OSO3 micelles.
2.3.3 Counter ion binding Although many ionic surfactants are completely split in ions as monomers in water, micelles have a charge that is much less than the number of surfactant molecules. Due to the high charge density many counter ions stay near the head groups and as such are more or less part of the micelle. The degree of counter ion binding is variable and is for micelles often between 50 and 80%. Furthermore, the experimentally determined value is to a certain degree dependent on the technique used to obtain this value. It is not always clear if a counter ion at a certain distance from the micelle is really free or still part of the micelle.
Question: A certain physical property of a micellar solution can be described with the following formula: φ = α [S] + β [M],
in which φ is the property to be measured, [S] is the concentration of free surfactant, [M] is the concentration of micelles and α and β are the relative contributions of S and M to the property. With conductivity measurements of hexadecyltrimethylammonium bromide (CTAB), the contributions to the conductivity come almost only from the bromide ions. a) Draw the graph that can be expected when the conductivity is plotted against surfactant concentration. b) b) Calculate the degree of counter ion binding when the slope before the CMC equals 80 µS/Mol and after the CMC 15 µS/Mol. Contributions from other ions than bromide can be neglected.
The nature of the counter ion and its concentration has an influence on the counter ion binding. This influences the repulsion between the head groups and therefore the effective cross section of the
- 12 - head group ao. The fact that counter ion binding to micelles occurs can be used to concentrate other ions in the shell around the micelle. This shell of counter ions and head groups is also called the Stern layer.
2.4 Factors that influence the CMC
- A micelle exists as a result of two counteracting effects. The apolar chains like to aggregate under the influence of water and the polar head groups like to be as far apart as possible. Therefore, it is logical to assume that the CMC increases with more polar head groups and decreases with longer alkyl chains. For ionic surfactants, the nature of the head group has little influence on the CMC, when the counter ion is univalent. Most important is the presence of a charge. Multivalent counter ions keep the head groups closer together, which results in lower CMC values for these surfactants.
Table II CMC values at 25°C for single-tailed amphiphiles with a dodecyl tail. Surfactant CMC / mM
(C12H25SO4)2Cu 1.2
(C12H25SO4)2Mg 1.8
C12H25SO4Na 8.1
C12H25SO4Li 8.9
C12H25SO3Na 10
C12H25CO2K 12.5
C12H25NH3Cl 14.7
C12H25N(CH3)3Br 16
C12H25N(CH3)3Cl 17
C12H25(OC2H4)4OH 0.046
C12H25(OC2H4)6OH 0.087
C12H25(OC2H4)8OH 0.109
+ - C12H25(CH3)2N O 2.1 + - C12H25N(CH3)3N -(CH2)2SO3 3.6 + - C12H25N(CH3)3N -(CH2)2CO2 5.3
For non-ionic surfactants, a much lower CMC is found and the size of the polar group is here much more important. Larger (more polar) groups increase the CMC. For surfactants with zwitterionic head groups, values for the CMC are found that are intermediate between those of ionic and non- ionic surfactants.
- 13 - - When the length of the hydrophobic tail is changed this has a pronounced influence on the CMC. A plot of log(CMC) of a certain type of micelle-forming surfactant versus the length of the tail gives a straight line with a slope of ~-0.5 for non-ionic and about -0.3 for ionic surfactants.
Figure 5 Change of log CMC as a function of the tail length (nc) for some types of surfactants.
Question: Estimate the CMC of hexadecyltrimethylammonium bromide using the relation given above when it is given that the CMC of dodecyltrimethylammonium bromide is 16 mmol/l.
Other structural variations like branching in the alkyl tail or the presence of double bonds have little influence on the CMC. In general the most important is the number of carbon atoms in the apolar chain.
- When a surfactant contains groups like –NH2, –NR2, –COOH that can be protonated or deprotonated, the CMC is strongly dependent on the degree of ionisation, and therefore of the pH. The CMC strongly increases with ionisation of the head group. - Addition of salts to ionic surfactants decreases the repulsion between the head groups. This makes micelle formation easier, so the CMC decreases. The influence of salts on non-ionic surfactants is more complex. Some salts like KCl, NaCl, etc decrease the CMC by “salting out” the surfactants, which makes them less soluble in water. Other salts like Mg- and Al-salts complex to a certain degree with the polyoxyethylene groups. Therefore, the hydratation of the head groups increases and the head groups also become slightly charged, thus the CMC increases.
2.5 Methods for determining the CMC
2.5.1 Surface tension As already indicated by their name, surfactants have a strong affinity for the water-air interface. This lowers the surface tension of water and this surface tension decrease becomes more with increasing concentration of free surfactant. When micelles are formed, the free surfactant - 14 - concentration does not change much anymore and so the surface tension becomes constant. The break in the curve is taken as the CMC.
2.5.2 Conductivity This method is very useful for ionic surfactants with no or only a small amount of added electrolyte. The conductivity increases linearly with the concentration of dissolved ions. Above the CMC most of the ions are present in the aggregates (both surfactant ions and counter ions that are present in the Stern-layer), therefore the conductivity increase after the CMC is much smaller. Conductivity also gives information about counter ion binding.
2.5.3 Light scattering Small dissolved molecules hardly scatter light. When larger particles are present, the scattering strongly increases. Although micelles do not scatter much due to their small size, their presence can be detected with this method and thus this can be used for determining the CMC. Additionally information can be obtained with this technique about the size and shape of the aggregates. Light scattering is also a very useful technique for investigating the size and shape of vesicles and macromolecules like proteins and synthetic polymers.
2.5.4 Solubilisation Apolar compounds that do not dissolve in water sometimes dissolve in micelles. A red compound like Orange-OT dissolves in the interior of micelles, but not in an aqueous solution with only free surfactant molecules. By stirring solutions of different concentrations of a surfactant with some solid Orange-OT and then measuring the absorption of the filtered solutions one can easily determine the CMC.
CH3 Abs N CMC N OH Orange-OT conc surfactant Change in absorption of dissolved OOT as function of concentration
Figure 6
Compounds like alkanes also dissolve well in the interior of micelles. However, the dissolving capacity is very limited. Although long-tailed alcohols do not act as micelle-formers, they can act as so-called cosurfactants. In the presence of micelles, they are taken up in the micelles with their
- 15 - apolar tails in the interior and the polar -OH group in the Stern-layer. As a result the effective packing parameter is adjusted and the curvature of the micelle decreases. This increases the size of the micelle and so more of an apolar compound can be dissolved in a micelle than without cosurfactant. Such a solution is sometimes called a micro-emulsion.
Besides the methods mentioned above, there are several other techniques that can be used to determine the CMC, among which are NMR- and fluorescence-techniques
2.6 Applications of micelles
2.6.1 Cleaning agents The cleaning action of micellar solutions (soaps) is based on a number of properties. a) Micellar solutions can solubilise apolar compounds in their interior; the capacity is however limited. Dirt b) More important is the fact that surfactants adsorb at a water/fat interface and because of that stabilise an Fibre surface emulsion of fat in water. This process is called Dissolved surfactants dispersion. It is helped by agitation and higher temperatures. This lowers the viscosity of fat and breaks it more easily in smaller fragments. Solid
particles like soil and dust are dispersed in water by Enclosure the adsorption of surfactant molecules at their surface. Especially ionic surfactants are suitable, because the particles that are adsorbed by ionic surfactants
become charged and repel each other. This avoids Solubilisation aggregation of these particles.
dirt Solubilized dirt
Fibre surface Adsorption layer Detergent action in time
Figure 7 Change of contact angle by surfactants Figure 8 The washing process
Surfactants lower the surface tension of both the substrate-water and oil-water interface. This changes the contact angle between oil and the substrate, which makes it easier for the oil to be released from the substrate. Substrates can be all kinds of surfaces like skin, clothing fibres, etc. The surface that becomes exposed due to the changed contact angle quickly becomes covered by surfactant molecules from the solution. This process is sometimes called wetting.
- 16 - The solubilizing and dispersing effects of surfactants are applied in many ways. Some applications are: a) Cleaning of polluted soil. Rinsing soil with non-ionic surfactants results in a removal of hydrophobic pollutants like oil and phenols of more than 90%, whereas rinsing with water alone has no effect. b) Partly exhausted oil fields can be exploited much longer by injecting surfactant-containing solutions near the oil wells. The surfactants disperse the oil droplets to form an emulsion, which can be pumped up. For this it is important that the surfactant molecules do not adsorb too much to the soil particles like clays or to metal ions, because this would reduce the flow too much and too much surfactant would get lost. Often anionic surfactants are used that have poorly co- – + ordinating (= metal-binding) head groups. Examples are R–(OCH2–CH2)n–SO3 Na c) In the exploration of certain ores, use is made of surfactants. The ore is dispersed in water with surfactants. By blowing air through the slurry, foam containing the desired ore particles is transported to the surface and collected. This technique is sometimes called “froth flotation” It is important to choose surfactants that disperse the desired minerals and not the other particles. For the isolation of sulphides, xanthates are used for example. These adhere specifically to sulphides under the influence of oxygen between pH 1 and 4 due to the following reaction:
S S RO M + H O RO + H–S–M + 1/2 O 2 2 S H S S xanthate metal sulfide complex
Due to this reaction the metal sulphide containing ore particles are specifically surrounded by surfactant molecules and can thus be separated. Worldwide, more than 2 billion tons of ore are obtained by flotation techniques.
2.6.2 Micellar catalysis. Apolar compounds tend to gather in the apolar (interior) parts of the micelle. This leads to a local increase in concentration. If such a compound can undergo a reaction with an ionic reagent and the head groups have a charge opposite to that of the reagent, a high concentration of reagent will be present in or near the micelle through counter ion binding. This leads of course to an increase of reaction rate, that is catalysis by the presence of micelles. It was found for example that the rate of hydrolysis of p-nitrophenyl heptanoate at pH 8.5 in a micellar solution of octadecyldimethylhydroxyethylammonium bromide could be increased by a factor of 105. When the reaction constant k is plotted versus the surfactant concentration, there is a strong increase in rate at the CMC. Above the CMC, the rate gradually decreases again, because with an increasing concentration of micelles the local concentrations of the reagents decrease again.
- 17 - O
C6H13 O2N O k
CH 2 Br C18H37 N CMC OH CH3 [surfactant]
Figure 9 Catalysis of an ester hydrolysis under the influence of micelles
Question: Write down the reaction equation of the hydrolysis reaction described above. What do – + you expect that occurs, when instead of this surfactant C18H35OSO3 Na is used as a micelle-former.
In some ways micellar catalysis strongly resembles the mode of action of enzymes. An important reason why enzymatic rates are so large is that substrate and reagent are brought together in a high local concentration. Additionally enzymes bring the reactants together in a good orientation for reaction, which is something these simple surfactants can not do.
Example: In the post cold-war era large quantities of nerve gasses (combat gasses) have become obsolete in various arsenals in East and West. It is a huge problem to safely and cheaply dispose of these compounds. Several nerve gasses contain a phosphorous-fluorine bond of which GD (or Soman) is just one example. Its mode of action is related to a disturbance of the transfer of nerve signals, which causes an inhibition of the relaxation of muscles. These compounds can be made harmless by hydrolysis under the influence of base. The problem is that these compounds do not dissolve in water, which makes contact with water ineffective. It is also known that the hydrolysis of phosphates is catalysed by metal ions. Based on this information Fred Menger developed a system with which these phosphate compounds can be hydrolysed efficiently. He developed a compound that complexes with copper ions and thus forms a surfactant (Atlanta-2 or A-2). This surfactant forms micelles.
- 18 - H C CH3 CH3 3 N N CH3 H3C 2+ P F H C Cu CH H C "Atlanta-2" 3 3 3 H O GD
F F 2+ 2+ 2+ 2+ – 2+ Cu O P OH Cu Cu O P HO Cu2+ Cu O P OH Cu R R' R R' R R'
kaCtaaltyatliysctihc es tsetapp
Figure 10 Micellar catalysis of the hydrolysis of GD by Atlanta-2 surfactants
The nerve gas GD dissolves well in the apolar interior of the micelles. The hydrolysis proceeds as follows: even in a weakly alkaline or neutral environment Cu2+ binds hydroxide ions from the aqueous bulk in the Stern-layer because counter ions are required for the cationic micelles. The OH– - group is close to the polar phosphate group of GD and is able to attack the phosphorous atom. The negative charge that develops on the oxygen of the phosphate group can be compensated for by another Cu2+-ion. Reformation of the P=O double bond leads to a release of fluoride. At pH 7 and 25ºC the halftime in this micellar system was about 50 seconds, whereas without surfactant even at pH 10 the halftime of the reaction is about 60 hours, despite the much higher average concentration of hydroxide ions. It is clear that this surfactant system offers many advantages for a cheap and safe destruction of these dangerous nerve gasses.
2.6.3 Application of micelles in chiral separations Enantiomers have the same physical properties, like melting point, boiling point and solubility. This makes an easy separation of these compounds difficult. For a successful separation it is always necessary to choose a route via a diasteriomeric compound or a diastereomeric complex, because diastereomers have different physical properties, which makes separation possible. After separation, the diastereomers are converted back into the original compounds. A well-known example is the formation of diastereomeric salts in the separation of R- and S-α-phenylethylamine with optically pure (R,R)-tartaric acid. In first instance the diastereomeric salts (R,R,R) and (S,R,R) are formed. These can be separated by selective crystallisation. After the separation, the phenyethylamine is liberated by reaction of a base. The tartaric acid also becomes available again and can be used for a subsequent separation cycle. Presently there is a lot of interest in the preparation of enantiomerically pure products for applications in medicine and in agrochemicals. This is because many governments enforce rigorous rules for the use of these chemicals. Because usually only one of the enantiomers is the active ingredient, the other enantiomer is waste. In favourable cases, it is only physiologically inactive, but in unfavourable cases it can also be toxic or possess undesirable physiological activities. - 19 -
COOH + + H OH HO H H H H C NH2 H N 3 2 CH3 COOH
S R R,R
COO- COO- H OH H OH HO H HO H H + NH + H COOH H3C 3 H N COOH 3 CH3 S R,R R R,R Figure 11 Racemate separation of R- and S-α-phenylethylamine with (R, R)-tartaric acid
In the following Micellar Enhanced UltraFiltration (MEUF) system, a clever use is made of diastereomeric complex formation for the separation of (R) and (S) amino acids.
Figure 12 Separation of D-phenylalanine (•) and L-phenylalanine () with the use of cholesteryl-L- glutamate as chiral selector that is anchored in micelles. The chiral selectors have a different affinity for the different enantiomers. The unbound enantiomers can pass the membrane and the micelles can not.
As diastereoselective complexes there are now not salts, but complex formation occurs in a micelle. The amino acid enantiomer that is not bound to the micelle is present in the water phase. With the use of a membrane ultrafiltration system, micelle and not bound amino acids can be separated, because water and free amino acids can pass the membrane, while micelles can not. A system that has been investigated in detail at the Laboratories of Organic Chemistry and Process Engineering is based on the chemical recognition of the receptor (S)-5-cholesteryl glutamate. - 20 - O O
HO O H NH2 chiral ligand ester group hydrofobic anchor
Figure 13 Structure of the receptor (S)-5-cholesteryl glutamate.
In the presence of copper (II) ions, this receptor forms diastereomeric complexes with amino acids.
H H H O * N O O Chol Cu O O N R O H H selectand in selector in S-configuration (L) R-configuration (D) H H H O * N O O Chol Cu O O N O H H R selectand in selector in S-configuration (L) S-configuration (L)
Figure 14 Both diastereomeric complexes of the receptor with (R)- en (S)- phenylalanine.
In these diastereomeric complexes the amino and carboxylate groups act as a ligand around copper(II). There is a preference for a flat square co-ordination geometry around copper in which the amino groups (and also the carboxylate groups) are in trans positions. The difference in the stability of the complexes seems to be strongly determined by the geometrical positioning of the rest group from the bound amino acid group with respect to the cholesteryl group of the receptor. The receptor itself is dissolved in a non-ionic surfactant NNP-10. This is 4-n- nonylphenyl(poly)ethyleneoxide. Thus, the diastereomeric complex formation takes place in the micellar medium. OH
H19C9 O O 9
Figure 15 The structure of NNP-10
In the presence of an ultrafiltration membrane, the racemic mixture of phenylalanine can be separated. A very high selectivity is found where the R-phenylalanine is bound to the micelle 5 times better than the S-enantiomer. This selectivity can not only be explained based on a difference
- 21 - in diastereomeric complex formation constant. The special micellar medium significantly enhances the selectivity.
2.7 General phase behaviour of surfactants
In the previous paragraph, we discussed the properties of micellar solutions of surfactants. These diluted solutions act as isotropic liquids, that means that they for example do not show birefringence. However, at higher concentrations and at different temperatures other types of ordering can be present. A useful way of visualising the solution properties of a surfactant is by making a phase diagram, in which on the x-axis the concentration of surfactant is plotted and on the y-axis the temperature. For many compounds, the solubility increases gradually upon increasing the temperature. For solid surfactants, the same is true until a certain temperature is reached. Above that temperature, the solubility increases much more rapidly. This temperature is called the Krafft-temperature or Krafft point.
Figure 16: Change of solubility and CMC with temperature. TKr = Krafft point
At this point, the solubility becomes equal to the critical micelle concentration; now the concentration of free surfactants in water does not increase anymore, only the concentration of micelles. For many applications of micelles, it is necessary that the Krafft temperature is below the temperature of the application, because below this temperature no micelles are present. The Krafft temperature of ionic micelles is strongly dependent on the nature of the counter ion. The Krafft temperatures of some dodecyl sulphates with different counter ions are given in the following Table. Table III Krafft temperatures (°C) of some dodecyl sulphates Na+ Mg2+ Ca2+ Ba2+ – C12SO4 9 25 50 105 - 22 -
In hard water, i.e. with Ca2+-ions the consequence is that those anionic surfactants (sulphates and also long chain carboxylates that are present in many soaps) do not dissolve anymore. This is for example the cause of a rim of soap rests in a bath.
In order to solve this “problem” there are several possibilities: i) Addition of extra surfactant: the precipitate is dissolved in the additional micelles; this may cost a substantial amount of additional detergent. ii) Addition of more univalent metal salts, for example NaCl (bath salt of washing machine salt), the counter ion binding is for the greatest part taken care of by the univalent metal ions which cause the calcium ions to stay dissolved. iii) Addition of non-ionic surfactants or the use of only non-ionic surfactants; These are less sensitive to metal ions. iv) Use of complexing agents that bind polyvalent metal ions and keep them dissolved. In the past phosphates were often used for this. Presently organic complexing agents are used or zeolites. The Krafft point is closely related to the structure of the surfactant. Compounds that fit easily in a crystal lattice will have a high Krafft point. This are for example surfactants with an ionic or a compact polar head group and with an unbranched alkyl tail. A disturbance of the packing in the crystal lattice gives a less high lattice energy and thus a lower melting point. Therefore, the Krafft temperature is lowered by branching or double bonds in the alkyl tails and also by large head groups that are present in non-ionic head groups.
For non-ionic surfactants, a so-called cloud point is often found in the low-concentration area and at increased temperatures. A micellar solution of non- ionic surfactants becomes cloudy above a certain temperature. Phase separation occurs and a water- rich phase and a surfactant-rich phase are formed.
Upon further increase of the temperature, a clear phase Figure 17 Phase diagram of C10E5. In the is formed again, often far above 100 C at high shaded area phase separation exists. pressure. The reason for this is that the ethyleneoxide chains are hydrated by water and this water is better structured than bulk water. Upon increasing the temperature, the hydrogen bonds in this hydration mantle are broken down and the entropy of the solution rapidly increases. The consequence is that phase separation occurs. At much higher temperatures, the general solubility has increased so much that a clear solution forms again.
- 23 - 2.8 Lyotropic liquid crystals
At high concentrations of surfactant in water, other aggregates than micelles are formed. A micellar phase is also called the L1-phase. When the volume fraction of water decreases the micelles will get in contact with each other and form a more or less regular stack. This type of ordering is called cubic (I1). With less water, the free energy can be minimised further by a transformation to a hexagonally packed cylindrical phase (H1), followed by a lamellar phase. (Lα). At even lower water contents reversed phases can be formed, in which the water is for example in the middle of the cylinders or micelles.
Figure 18 Changes in aggregate form and structure by changing (increasing) the surfactant concentration
It can be calculated when one phase should theoretically be transformed into another phase. For example, in a closed packed arrangement of spheres 74% of the space is taken up by the spheres and with cylinders this is 91%). In practice, this depends on the nature of the surfactant and also on for example the temperature. For a certain surfactant, some phases are not found and sometimes other (more exotic) phases are found. The most often found phases are given in Table IV, in order of decreasing water content. In Figure 19 the phase behaviour of a non-ionic surfactant as a function of water content and temperature is given. Due to the ordering of the molecules in the different phases, they exhibit properties that have a strong resemblance to those of liquid crystals. A number of phases are anisotropic, which means that the properties depend on the direction through the sample in which the properties are determined. The phases behave like a (viscous) liquid, because the molecules are able to move with respect to each other, just like in liquid crystals.
- 24 - Table IV: The most important lyotropic phases in water/surfactant systems arranged according to decreasing water content and the symbol that is often used for it.
Symbol Geometry
I1 Cubic
H1 Hexagonal
V1 Bicontinuous Cubic
Lα Lamellar
V2 Bicontinuous Cubic (reversed)
H2 Reversed Lamellar
I2 Cubic
Figure 19 Phase diagram of a
non-ionic surfactant C16E8
Due to this resemblance, these phases are sometimes also called lyotropic liquid crystalline phases.
We shall now discuss some properties of the most common phases. The micellar phase (L1) is isotropic, as are the cubic phases; i.e. the properties are independent of the direction. Hexagonal phases are anisotropic, which means that the properties of domains with the same ordering are dependent of direction. With polarisation microscopy, typical fan-shaped textures are visible. In such a domain the cylinder-like structures are parallel (Figure 20). Lamellar phases are also anisotropic and often show a thread-like texture by polarisation microscopy. In thin films of these materials, the sheets (or lamellae) are often parallel to the glass slides. This so-called homeotropic ordering has as a consequence that the refractive indices in the directions parallel to the glass slide are equal, so no rotation of the polarisation plane occurs. This means that the observation field remains dark. At fault lines and domain boundaries the polarisation plane is rotated, this leads to the thread-like texture.
Different cubic phases can occur. They are characterised by their isotropic behaviour. The I1 and I2 phases resemble stacked arrays of micelles and reversed micelles. The bicontinuous cubic phases
(V1 and V2) can be considered as micelles (spheres) that are open at the positions where they are in contact with each other. For the V1-phase, this means for example that the apolar interior of one micelle is in contact with the interior of neighbouring micelles. The water-phases are also in contact with each other throughout the sample; hence, the name bicontinuous cubic phase.
Question: Describe and schematically depict the structures of the reversed hexagonal, lamellar and cubic phases.
- 25 -
Figure 20 Schematic representation (left) of the hexagonal (above) and lamellar (below) lyotropic liquid crystalline phases and their textures (right) as observed by polarisation microscopy.
2.9 Special aggregates of single-tailed surfactants
The formation of the micellar phase is possible for surfactants that have the correct shape for micelle-formation; i.e. the packing parameter is less than about 0.3. When the packing parameter is slightly higher worm-like or thread-like micelles can be formed. These are long cylinder-like aggregates. The packing parameter can also be influenced by the nature of the counter ion. A counter ion that binds well to the surfactant decreases the charge surplus in the Stern layer and decreases the water mantle around the polar ionic head group. Consequently, the head groups approach each other better, thus increasing the packing parameter. A well-known example is hexadecyl-ammonium (CTA+) bromide micelles. When salicylate anions are added the spherical micelles are transformed to thread-like micelles. These aggregates grow so long that they become intertwined which results in solutions that show so-called visco-elastic behaviour. After agitation, for example stirring, these (viscous) solutions have a tendency to return to the old situation.
- 26 - O
O CH3 H3C N C16H33 OH CH3 + salicylate CTA Addition of certain other counter ions, for example a high concentration of NaCl can have this effect. In the following Figure, the structure of such a micelle is schematically depicted. Also shown is how the entangled thread-like aggregates can become disentangled again by a rearrangement of the individual surfactant molecules.
Figure 21 Schematic representation of thread- (or worm-like) micelles and how the threads break and form again upon entanglement.
Another case is found when a counter ion is added that itself is able to form micelles. Now mixed aggregates are formed in which the packing parameter approaches 1. The charges in the Stern layer are (almost) completely compensated, the water mantle around the ionic head groups is strongly reduced and an additional apolar tail is present. Consequently bilayer or vesicle-like structures are formed. Recently there is a strong interest in these aggregates, which are sometimes called ion-pair amphiphiles (IPA’s).
+ – + + – – + – + – + + – – +
Cationic micelle Anionic micelle Ion-pair amphiphile (IPA) Figure 22 Schematic representation of the formation of ion-pair amphiphiles from anionic and cationic surfactants.
- 27 - Chapter 3: Vesicles
3.1 Introduction
Vesicles are closed bilayer structures in water of amphiphilic molecules. The name “vesicle” comes from Latin and means something like bubble. The bilayer vesicle surrounds a water-filled compartment, in which all kinds of compounds can be dissolved. The bilayer acts as a membrane, that inhibits transport of compounds dissolved in the interior or exterior or lets them pass selectively. Due to their structural resemblance, vesicles are a good representation of the much more complex biomembranes. The structure of the amphiphilic molecules that make up the vesicle wall is characterised by the presence of usually two alkyl chains per polar head group. Therefore, the surfactant parameter S has a value between 0.5 and 1 (see Chapters 1 and 2). Vesicle forming amphiphiles usually have a poor solubility in water. When crystals of those amphiphiles are brought into contact with water a swelling on the outside of the solid material occurs, giving stacks of hydrated lamellar structures, called myelin structures.
Figure 1: Myelin structures formed by letting solid lecithin get in contact with water, viewed by microscopy between almost completely crossed polarises.
The driving force for this process is the favourable hydration energy. By applying mechanical force on these myelin structures, by shaking or vortexing, these lamellae become separated from the solid surface and can form closed bilayer structures. There is a preference to form closed bilayer structures and not open lamellar sheets, because in a closed structure there are no unfavourable interactions between the hydrophobic interiors (at the edges) and water. Unfavourable is the curvature, so vesicles are usually not very small. When starting from myelin structures in first instance usually multilamellar vesicles are formed (MLV; diameter 1-2 µm). Upon ultrasonic treatment much smaller unilamellar vesicles are formed (SUV = small unilamellar vesicle; diameter
- 28 - 300 - 500 nm). Compare these sizes with those of a virus (130 nm) and of a bacillus (700 nm). Micelles are an order of magnitude smaller (only 5-10 nm). Usually these vesicles are not thermodynamically stable, due to the curvature energy. They can merge to larger vesicles and sometimes to hydrated bilayers. Thermodynamically more stable vesicles can be obtained by a process called freeze thawing. Upon freezing the growing ice crystals disrupt and fragment the bilayer walls and upon thawing, these fragments combine again to more stable, i.e. in most cases larger vesicles. This process is usually repeated for a number of times. SUV’s can be regarded as high energy intermediates between the crystalline phase and more stable larger vesicles or hydrated bilayers.
Figure 2: Schematic representation of the formation of vesicles.
Because the preparation of SUV’s by sonication leads to an inhomogeneous distribution of vesicle sizes, these are less suitable for further studies. This is because a small variation in size can lead to different bilayer properties due to the differences in curvature. Research on very large vesicles (so- called giant vesicles with diameters of 5-200 µm, eliminates this problem, because with these large sizes the bilayer properties are not influenced anymore by fluctuations in diameter. An additional advantage is that these vesicles can be directly observed by light microscopy. The size of such giant vesicles is comparable to those of red blood cells (~7 µm) The formation of vesicles was first reported by Bangham in 1964, who investigated phospholipids. Due to their structure, phospholipids are amphiphiles. The vesicles from natural phospholipids are mostly called liposomes in literature. In this course, we will use the more general name vesicles, because this also includes closed bilayer structures made from synthetic amphiphiles. In 1977, the first publication appeared in which the preparation of vesicles was described from synthetic amphiphiles. These were dialkyldimethylammonium bromides. Later is was found that much more types of synthetic amphiphiles were able to form closed bilayer structures when dispersed in water in a correct way. A number of vesicle-forming amphiphiles are given in the Table. All amphiphiles mentioned have two apolar tails. It is however also possible to prepare bilayer vesicles from other
- 29 - molecules. For example, an almost equimolar mixture of anionic and cationic single chained amphiphiles is also able to form vesicles (see Chapter 2).
Table I: Molecular structure of vesicle forming amphiphiles.
Phospholipids: O
R C O CH2 R C O CH O
O H2C O P O X O + X = CH2CH2N(CH3)3 Phosphatidylcholine = lecithine
+ CH2CH2NH3 Phosphatidylethanolamine
+ _ - CH2CH(NH3) COO Fosfatidylserine
OH OH
Phosphatidylinositol
HO OH OH
Cationic amphiphiles
CH CH2 + CH3 3 n _ N Br CH CH CH3 2 m 3 Dialkyldimethylammonium bromides
Anionic amphiphiles
CH3 CH2 O O + n Na P _ CH CH O O 3 2 m Sodium dialkylphosphate
3.2 Preparation of vesicles
3.2.1. Preparation of vesicles by sonication A simple way of preparing vesicles is by the following procedure. Dissolve about 10 mg of amphiphile in 1 ml of CH2Cl2 in a small test tube. The solvent is evaporated by slowly blowing nitrogen gas over (not through) the solution while slowly rotating the tube. This should be done in such a way that a thin film of amphiphile is formed on the inner wall of the tube. Next, about 1.5 ml of water is added. The thin film is dispersed in the water by ultrasonic treatment with a pen during 5 minutes. This should be done at a temperature above the phase transition temperature (see later in this Chapter), because only then the best packed bilayers are formed. Usually a temperature of about 70ºC is taken. A slightly cloudy dispersion is obtained that consists of vesicles.
- 30 - 3.2.2. Preparation of vesicles by the alcohol injection method. Vesicles can also be prepared by injecting a concentrated solution of amphiphiles in alcohol into a vigorously stirred aqueous solution at a temperature above the phase transition temperature.
3.2.3. The preparation of giant vesicles. a. 1 mg of amphiphile is dissolved in 1 ml of CH2Cl2. This is evaporated as described under 1. To the so-obtained thin film in the tube is added 5 ml of water. Now the test tube is placed in a water bath of about 70ºC. Slowly, the film of amphiphiles becomes hydrated and is released from the wall. Heating is continued for another 4 hours and the test tube is taken from the water and shaken gently for a few seconds. By this action, the big sphere is transformed into a collection of giant vesicles. b. In another method, an injection syringe with a blunt needle is filled with water. The tip of the needle is submersed for a short time in a solution of lipid in an organic solvent that is immiscible with water. After that the needle is placed in water and a giant vesicles is blown.
An illustration from our own work is a giant vesicle that is obtained from a mixture of amphotrope 1 and sodium dodecylsulphate (SDS) in a 1:1 molar ratio in water at 60ºC. This is a ion-pair amphiphile as discussed in Chapter 3. Under a microscope the structure of the giant amphiphile can be easily be observed, despite the very thin wall of the vesicle.
+ _ NC O (CH2)12 N(CH3)3 Br 1
Figure 3: Light microscopic recording of a giant vesicle of the ion-pair amphiphile 1:SDS (1:1).
- 31 - 3.3 Visualisation of vesicles by electron microscopy.
Small unilamellar and multilamellar vesicles can not be visualised by light microscopy. The resolution of this method lies in the order of magnitude of the wavelength of the used light; i.e. about 1 µm. If we want to see smaller vesicles electron microscopy is the method of choice. Because electrons can only travel a short distance in air, an electron microscope works under high vacuum conditions. This also means that the sample to be investigated has to be brought under these conditions. This gives problems, because under these conditions water will evaporate rapidly. Another aspect is that organic materials give no or only very little contrast, due to the absence of heavy atoms. Therefore, the sample has to be pre-treated. Several methods are available for this. A very simple and often used method is “negative staining”. In the case of vesicles, this means that a vesicle dispersion is brought into contact with a contrast-enhancing agent (“staining agent”). For vesicles with a negative surface charge, uranyl acetate is often used. The uranyl cations bind to the vesicle wall and thus afford the contrast when the sample is placed in an electron beam. For vesicles 2- with positive surface charges, tungsten salts (WO4 ) can be used. Because the vesicle itself is not stained, but its surface and surroundings, this method is called negative staining. Because of the high-vacuum conditions, the solvent (water) is completely gone. This can sometimes have consequences for the observed morphology, which may not be exactly the same as present in solution. An example of an electron microscopic image of a vesicle dispersion is given below.
Figure 4: Electron microscopy images of some negatively stained vesicles.
Other methods for sample preparation are the freeze fracture method and cryo-electron microscopy. With these methods, much better images can be obtained. With the freeze-fracture method, a drop of the vesicle dispersion is quickly frozen in liquid nitrogen. The frozen dispersion is broken with a knife. If the fracture reaches a vesicle, the fracture plane follows the lipid layer, because the weakest spot is the middle of the bilayer. So both the inner and outer lipid layers can be seen. A
- 32 - replica is made of the fracture by evaporating a 20Å thin layer of platinum and carbon onto it. Then a thicker layer of about 200Å is applied as a support layer. After melting of the original, this replica is investigated by electron microscopy. Cryo-electron microscopy is a relative new method of visualising vesicles. With this method, a drop of sample is also cooled rapidly, usually in liquid ethane at -170ºC; the ethane is kept at this temperature by external cooling with liquid nitrogen. The sample is present on a support material, which is a copper grid covered with a polymer foil (formvar) with a carbon coating. The frozen sample is kept at a temperature below -100ºC in a sample holder that is placed in the electron microscope. This sample is now investigated with a relatively weak intensity electron beam.
3.4 Phase transitions in amphiphilic bilayers
Vesicles from phospholipids or from synthetic amphiphiles exhibit a phase transition at a certain temperature. This can be measured as a small heat effect. At this temperature, a co-operative change takes place with regard to the ordering of the molecules in the bilayer. At a temperature below the phase transition temperature Tc the hydrocarbon chains of the amphiphiles are in a so-called ordered gel phase, in which the conformation of the hydrocarbon chains is predominantly all-trans.
Tc
Figure 5: Schematic representation of the transition from the gel to the liquid crystalline phase.
Above the Tc the chains are more disordered with more gauche conformations. This is now called a disordered liquid crystalline phase. When the phase transition temperature is passed upon heating, the effective length of the amphiphilic molecules will decrease. This will influence the thickness of the bilayer membrane. If the tilt angle of the lipid molecules does not change, the layer thickness decreases. For vesicles obtained from phospholipids it was found that Tc depends on the size of the vesicles as well as on the length of the hydrocarbon chains. The smaller the vesicles, the larger the curvature and the individual molecules will pack less well in the bilayer structure. This will decrease the Tc. The critical diameter is around 70 nm. For larger vesicles, the Tc will not change much anymore. With increasing chain length, Tc will also increase, because longer chains pack better than short ones. This is also the reason why Tc decreases with the introduction of branches in the hydrocarbon chains. Several techniques are available for measuring the phase transition temperature. An important method is DSC. A DSC cup is filled with a vesicle dispersion (usually 20-50 µl of a 1% dispersion) and now the temperature is raised. Upon passing the phase transition
- 33 - temperature, extra heat will be required which can be measured. Therefore, this method yields both the phase transition temperature and the associated enthalpy change. The degree of reversibility of the transition can also be determined with this technique by cooling the sample again. At the same temperature, one would expect a similar but now exothermic heat effect. Usually this transition is not found at the same temperature as upon heating, but at a much lower temperature. This points to a supercooling of the liquid crystalline phase. DSC is also a useful technique to determine if vesicles are really formed and not small droplets of dispersed amphiphile. By freezing the sample, the ice crystals will disrupt the vesicles. When the sample is now heated again, the phase transitions that were recorded during the first heating will now be absent. If dispersed droplets were present, the initially found transitions will also be present in the second heating curve.
Another method to determine the phase transition is by NMR spectroscopy. Below Tc, the line- 1 widths in the H-NMR spectrum will be much larger than above Tc, due to the lower mobility of the
CH2-groups in the bilayer below Tc. The phase transition temperature can also be determined by fluorescence spectroscopy when a fluorescent probe is added to the vesicle bilayer. A much-used fluorophore is diphenylhexatriene (2).
2
This elongated molecule will orient parallel to the alkyl chains in the bilayer. When this molecule is excited by polarised light and the molecule is almost immobile in the bilayer, the fluorescence will also be polarised. If movement is possible, a part of the polarisation will get lost. A change in the liquidity of the bilayer thus has an influence on the fluorescence depolarisation of the trapped probe molecule.
3.5 Trapping of guest molecules in a vesicle compartment
When during the preparation of vesicles guest molecules are added, these compounds will partly be present inside a compartment surrounded by the bilayer. Trapped and non-trapped molecules can be separated by gel permeation chromatography (GPC). This chromatographic technique separates on particle size. Small particles among which are the non-trapped molecules will diffuse in the pores of the stationary phase (column material), so their elution from the column is retarded. The larger vesicles, however, choose the shortest route and are eluted first. In this way one obtains a sample with vesicles containing trapped guest molecules. When such guest molecules have special properties, this can be used to further characterise the system. A well-known example is carboxyfluorescein (CF).
- 34 - HO O O
Carboxyfluorescein (CF) COOH
HOOC
CF is a fluorescent compound, so it can already be detected in very low concentrations. If CF is present in vesicles in a sufficiently high concentration that self-quenching occurs the CF in vesicles will not be fluorescent. Dilution of the vesicle dispersion will not influence the CF fluorescence, because the local concentration of CF does not change. If however CF is able to pass the bilayer membrane the molecules will be present in a very diluted environment, where quenching does not take place anymore. This means that by recording the fluorescence as a function of time, the transport of CF through the membrane can be measured. It was found for example that vesicles prepared below the Tc let pass CF much faster than vesicles prepared above Tc. Apparently less well packed bilayer structures are obtained in the gel phase than in the liquid crystalline phase. Trapping guest molecules in vesicle water compartments is an important research area related to controlled drug release. In a following paragraph (3.6), this will be discussed further.
3.6 Fusion of vesicles
Fusion of bilayer membranes is an important phenomenon in countless biological processes: it plays for example an essential role in the fertilisation where two cells merge and combine their contents. A practical use lies in the possibility to use vesicles in controlled drug release. A suitable method to investigate vesicle fusion is by electron microscopy. With this method, it is also well possible to distinguish the fusion process from vesicle aggregation. Aggregation is the first step in the fusion process, because close contacts between the vesicles is established. At the contact place, probably structures are formed that resemble (reversed) micelles and/or hexagonal phases. Vesicles dispersed in water do not spontaneously tend to aggregate or undergo fusion. This is because the water mantle prevents an intimate contact between the vesicles. Two possibilities can be proposed how this first step towards fusion can be realised. The first one is that the water mantle around the vesicle is removed. This is how fusion under the influence of polyethylene glycol (PEG) works. How it exactly proceeds is not known. It is presumed that small, dehydrated contacts are formed between the vesicles, because one PEG binds to more than one vesicle and it forms its own water mantle at the expense of those of the vesicles. This leads to mechanical tension in the bilayer, which causes its destabilisation. The cause of the mechanical tension is the change in curvature tension of the membranes. This mechanism is sometimes called the stalker mechanism.
polyethylene glycol HO O O OH O n
- 35 - Addition of calcium ions to negatively charged vesicles can also induce fusion. This is often a two step process. Below a critical calcium concentration only aggregation occurs. The divalent calcium ions form a bridge between the negatively charged bilayer vesicles. Addition of these calcium ions influences the membrane properties. This can be measured by determining the Tc. This Tc increases. When initially the vesicle bilayer is in the liquid crystalline phase, addition of calcium will increase the Tc.
Figure 6: Schematic representation of vesicle fusion with reversed micelles as intermediate structures.
The concentration of calcium that raises Tc to room temperature (or more precisely the temperature of the experimental conditions) is called the critical calcium concentration. At this concentration locally crystallisation of the lipid molecules at the contact surface occurs. This induces mechanical tension in the membrane and destabilisation of the bilayer. This mechanism is also called the adhesion mechanism and probably proceeds via reversed micelle-like structures. The difference between both fusion mechanisms lies in the cause for the origin of the mechanical tension, as a result of which the bilayer is destabilised.
3.7 Solubilisation
Compounds with an apolar or partly apolar character can dissolve in bilayer membranes. In cell membranes, this plays an important role in the functions of the membrane, like transport, recognition, etc. The interaction of some compounds with membranes will be discussed next.
1) Cholesterol Cholesterol can be taken up in different concentrations by the vesicle bilayer membrane. This solubilisation process is quite complex. Cholesterol seeks a position in the bilayer with its polar hydroxyl group near the polar-apolar interface. Being a guest molecule, cholesterol has a preference for the liquid crystalline phase. However, the rigid cholesterol skeleton limits the conformational freedom of the neighbouring amphiphiles and thus benefits trans conformations above gauche ones; - 36 - it prefers well-ordered chains as neighbours. These contradicting demands lead to a local phase separation, with a cholesterol-rich liquid crystalline phase, which is separated from the rest of the membrane.
CH3 CH3
HO cholesterol
2) Cholic acids Cholic acid and some derivatives, known as bile acids or bile salts, are examples of amphiphiles that can form micelles, although they do not have the classical polar head group - apolar alkyl tail. Because of its rigid shape, one side of the molecule is polar and the other side is apolar. They are called facial amphiphiles. Cholic acids can not penetrate the bilayer. Due its polar head group and its three hydroxyl groups at one side of the steroid skeleton, it can bind to the surface of the bilayer membrane. This causes a disturbance of the packing of the amphiphilic molecules of the bilayer. When the concentration of cholic acid is high enough the whole bilayer will be broken down and the lipid molecules will form mixed micelles together with the cholic acid molecules. The lipid molecules solubilise in the cholic acid micelles. This process of solubilisation also plays an important role in digestion of fats. Cholic acid is responsible for the transport of mixed micelles with fat molecules (mainly triglycerides). When these mixed micelles arrive at the intestine wall, the solubilized fat molecules will be solubilized by the intestine cell wall membranes. Because cholic acid is much more polar, it will not be taken up by these cell membranes. One can now ask why cholic acid does not solubilise the walls of the gall bladder. The answer is that cholic acid is present here as mixed micelles together with lecithin and cholesterol, which decrease its solubilizing power. When these mixed micelles become oversaturated with cholesterol, they can crystallise as the well-known gallstones. O
CH3 CH3 12 OH
H 7 OH OH 3 OH Cholic acid
3) Solubilisation of micelles by surfactants Above the CMC (Critical Micellar Concentration), surfactant micelles are well capable of solubilizing hydrophobic molecules, which can lead to mixed micelles if the molecules also contain polar parts. Vesicle-forming amphiphiles can also be dissolved by micelle-forming surfactants. This is used to break down biomembranes that contain membrane-bound proteins. This technique is used to isolate and purify these proteins. The reverse process is also possible. By dialysis of mixed surfactants containing both surfactants and phospholipids, the surfactant can be removed selectively and vesicles, with or without proteins are formed again. This is an important method to reconstitute
- 37 - all kinds of membrane-bound proteins in a new membrane environment with a well-known composition. All kinds of tailor-made functional membranes can thus be prepared. Other methods usually fail, because membrane-bound proteins often denature quickly when they are not surrounded by amphiphiles that simulate the bilayer membrane.
Question: Make use of the concept of surfactant parameter S in order to explain that ultimately mixed micelles are formed when solubilizing vesicles with surfactants.
3.8 Applications of vesicles
3.8.1 Photodynamic therapy Photodynamic therapy is a method that for localising and treating tumours by using photoactive compounds. Upon irradiation with light, these compounds can produce cytotoxic compounds. The selectivity of the therapy is based on the ability of the photoactive compound to concentrate more in tumour cells than in normal cells. This can be accomplished in different ways; for example, the coupling of the photoactive compound (S) to a lipoprotein that contains a specific receptor for the tumour cells. For the therapy, it is also necessary that oxygen is dissolved in the neighbourhood of the tumour cells and that light of the correct wavelength and intensity is present. The mode of action is than as follows: Irradiation and absorption of light brings the photoactive compound in a singlet- exited state. Via intersystem crossing, this state is transformed in the triplet state. This triplet state can now react further by two different routes. In the Type I process the triplet state reacts further · with a H-donor (LH), which results in the formation of a lipid radical (L ) and the radical anion of · the photo-active compound (S -) together with H+. Both products lead ultimately to the formation of the lipid hydroperoxydes (LOOH). In the Type II process, the triplet state reacts directly with oxygen. Via energy transfer, the photoactive compound goes to the ground state and singlet oxygen 1 is produced ( O2). Singlet oxygen subsequently also reacts with lipid molecules (LH) to lipid hydroperoxydes (LOOH).
Figure 7: Type I en II reactions in which lipid hydroperoxydes are formed.
The hydrophobic photoactive compounds can be applied through vesicles. Due to its hydrophobic character the compound dissolves in the bilayer and not in the water compartment. An example of a
- 38 - photoactive compound is zinc phthalocyanine that is used in the photodynamic treatment of 1 tumours. The sensibilizing ability is probably caused through the formation of singlet oxygen ( O2). Singlet oxygen subsequently reacts with the double bonds in lipid molecules of the membranes forming hydroperoxydes. This destroys the function of the membrane, ultimately killing the cell.
N N N
N Zn N
N N N
Zinc phthalocyanine (ZnPc)
3.8.2 DNA-transfection Molecules, especially polar and macromolecular ones, can not easily pass the lipid cell membrane. In nature, special proteins carry out that function for specific molecules. This makes it difficult to bring large molecules, like DNA, in cells on purpose as would be desirable for gene therapy. It was found that complexes of (negatively charged) DNA with cationic vesicle forming surfactants were able to pass cell membranes through a mechanism that is not yet completely elucidated and understood. This process is called transfection. An amphiphile that is often used is DOTMA. In addition, a combination with dioleylphosphatidylethanolamine (DOPE) in a 1:1 molar ratio is often used. This mixture is commercially available under the name lipofectin. At the moment, several other candidates are investigated that are more effective and less toxic to the cell.
H2C O C18H35
CH O C18H35 DOTMA + – H2C N(CH3)3 Br
This transfection method can be used in gene therapy, where genes (pieces of DNA) are brought in the selected cell, after which they start to synthesise the coded peptides. This is an important process, because many diseases can be reduced to defect or missing genes. With the aid of gene therapy, it is also possible to block the synthesis of unwanted or excess peptides. This is done by adding a piece of single strand DNA to the cell that is complementary to the part to be blocked (anti sense technology).
3.8.3 Controlled drug release By administering drugs packaged in vesicles their activity can possibly be enhanced, because the distribution in the body can be modified by this method, preferably in such a way that a high concentration is obtained where the drug is needed. This is called “targeting”. Apart from this,
- 39 - vesicles are also able to release the drug in a controlled manner. This is caused by a gradual transport of the enclosed compound through the vesicle wall. A serious problem is that the used vesicles are often quickly recognised by the body as an alien particle and transported to the liver and spleen, where they are broken down and thus can not reach their destination. By using amphiphiles with oligoethyleneoxyde chains coupled to their head groups vesicles can be prepared with extended lifetimes in the blood stream. These are so-called stealth amphiphiles. An example of such an amphiphiles is given below.
O
C17 C O CH2 - C17 C O CH O O
O H2C O P O CH2CH2NH C O CH2CH2O CH3 O n n=16-113
The oligoethyleneoxyde chains at the vesicle surface probably inhibit the contact with other membrane walls. This can be regarded as sterically stabilised vesicles. This can however also be a problem for the release of the drug. Due to their long retention time in the blood stream, the vesicles are able to reach places where small leaks are present in the blood vessels, where they disintegrate in the small gap and release their content. Leaking vessels are much more abundant in tumours than in other parts of the body. This leads to a locally 10-30 times higher concentration of drugs than would have been possible without these “drug carriers”.
3.9 Literature 1. M. Rosoff, Ed., Vesicles, Surfactant Science Series, volume 62, Marcel Dekker, New York, 1996.
- 40 - Chapter 4: Monolayers
4.1 Introduction Water-soluble amphiphiles have the tendency to concentrate at the air - water interface due to their amphiphilic character. The hydrophilic head group is in the water and the apolar tail points in the air. This is already happening at very low concentrations. When the concentration increases, the concentration of amphiphiles at the interface increases, but also the chance that the amphiphiles form aggregates like micelles or bilayers in the water phase. Water insoluble amphiphiles do not form aggregates in water, but when present at the water surface, they can order at the water - air interface. Because the polar head groups like to stay in contact with water, they preferentially form a monolayer at the water surface. The amount of material needed to cover a surface with a layer that is one molecule thick is very small.
Question: Calculate the surface that can be covered with 1 ml of for example a vegetable oil (a triglyceride) with a thickness of one molecule. Assume that the length of the molecule is about 2 nm (20 Å) and that the molecules are oriented perpendicular to the surface.
This experiment was already performed in 1765 by Benjamin Franklin, which made him the first person to describe the formation of monomolecular layers, although he probably did not realise that. The triglycerides he used are not good amphiphiles. There is hardly repulsion between the head groups and the tails tend to stick together, which makes the monolayer in fact only a thin coherent film, with not much ordering. With amphiphiles with a more polar head group the repulsion between the head groups is often stronger than the mutual van der Waals attraction between the apolar tails. In those cases, the amphiphiles are present as isolated molecules at the water surface. This situation resembles a 2-dimensional gas state. Because of the presence of these amphiphiles at the surface, the surface tension is lowered. This can easily be measured, and one can easily determine the relation between the surface tension and the concentration of an amphiphile at the water surface. In practice, this experiment is performed with a Langmuir trough.
- 41 - 4.2 Langmuir trough
The Langmuir trough consists of a shallow tank, covered with Teflon, which is water repellent, and that is filled with ultrapure water (see Figure 1). The available surface can be diminished by a movable barrier that can be pushed over the surface. With a second floating barrier containing a pressure sensor, the surface tension can be measured (the horizontal pressure equals the length of the barrier l times the surface tension π). Another method to measure the change in surface tension is with the so-called Wilhelmy plate. This is a piece of paper or metal that partly hangs in the water. The weight of the water meniscus the hangs onto this plate depends on the surface tension and changes in this weight can easily be measured.
Figure 1 Langmuir trough used to study insoluble monolayers.
In a typical experiment, a known amount of amphiphile that is dissolved in a volatile organic solvent is added drop by drop on a clean water surface with a syringe (for example 50 µl of a 1% solution). The solvent evaporates and the amphiphilic molecules spread on the surface. When the surface is made smaller the available area for the molecules decreases and at a certain point the molecules will start to touch each other. The pressure rises. From these data a so-called π-A curve can be constructed (with a computer program), in which the surface tension π is plotted against the available area A per molecule.
4.3 Phase behaviour
At high available areas, the molecules behave as a two-dimensional gas. The pressure is very low, i.e. π = kT/A when it behaves as an ideal two-dimensional gas (k (Boltzmann constant) = ~ 1.4·10-23 J/K). Compare with an ideal gas, for which PV = nRT. By decreasing the area the molecules start to interact and touch each other and as a result can in principle undergo several (2-dimensional) phase transitions; from gas-like to liquid-like to solid (crystalline)-like. Often more than one liquid-like phase transition is found, those are sometimes called liquid-expanded and liquid-condensed. Probably there is an analogy with the various possible smectic liquid crystalline phases. The compressibility differs from phase to phase and decreases upon going from gas-like to crystalline-
- 42 - like. At phase transitions, temporarily two phases co-exist. A general picture of a π-A isotherm can look like shown in Figure 2. Usually, the pressure changes that occur before the pure liquid expanded phase is present can not be seen due to the low experimental pressure. In addition, crystallisation is not always reached.
Figure 2: General -A curve for insoluble monolayers at the air-water interface. G = gas-like; LE = liquid- expanded; T = transition; LC = liquid-condensed; S = solid.
Upon increasing the pressure, the molecules come in closer contact, the space available for the tails becomes increasingly smaller, and the tails orient perpendicular to the surface.
Water
gas-like film state liquid expanded film state
Water
liquid condensed film state condensed solid film state
Figure 3: Changes in the ordering of the tails of the amphiphiles upon decreasing the available surface. When going from a) to d), the area decreases and consequently several phases are passed.
The room between the tails decreases and in the liquid condensed or in the crystalline phase the obtained area is a good measure for the cross section of one molecule. When the most compressed state is reached the pressure increases sharply upon further reduction of the area and the monolayer breaks at some point. This is often seen as a sharp reduction of the pressure. If this “collapse”
- 43 - already takes place in the liquid phase, one often obtains a triple layer, that upon further compression can even give 5 or 7-layers. A hint that this is happening can often be seen from the fact that a plateau in the curve is found, whose end is at 1/3 of the beginning of the plateau.
Figure 4: π - A isotherm from which the formation of a triple layer can be deduced and a model for such a triple layer.
4.4 Relation between structure and monolayer behaviour.
With a full compression in the monolayer, the cross-sectional area of the molecule is found. For carboxylic acids, for example a cross-section of about 18-20 Å2 is found independent of the length of the molecule.
n-Hexatriacontanoic acid (C35H71COOH)
Stearic acid (C17H35COOH)
Isostearic acid ((CH3)2CH(CH2)14COOH)
Molecular cross sectional areas Figure 5 Influence of branching in an alkyl tail on the π - A curve.
- 44 - This is the cross-section of the alkyl tail in an all-trans conformation. Upon branching of the alkyl tail with one methyl group the cross section already increases to for example 32 Å2 for isostearic acid (see Figure 5). Due to the irregularity in the apolar tail, good packing is not possible and thus the pressure at which collapse occurs strongly decreases. For a more voluminous elongated molecule like cholesterol, a cross-section of about 50Å2 is found. These values are quite realistic as judged from molecular models.
cholesterol HO When an amphiphile contains a polar group in its apolar tail, an increase in surface pressure is often already observed at relatively high areas. This is caused by the fact that the compound has now initially two (or more) contact points with the water surface. Examples are:
12-hydroxystearic acid
Figure 6 π - A isotherms of two compounds that have multiple contact points with water.
For the compounds 1-n,m there are initially even three contact points with water. In the flat part of the curve, a transition takes place where the weakest contact with water is broken and the molecules orient perpendicular to the water surface. At about 40 Å2 this condition is reached.
4.5 Visualisation of monolayers
Additional information on the behaviour of monolayers on a water surface can for example be obtained with a Brewster angle microscope (BAM). This technique is based on the following principle. Light that shines on a water surface is partly refracted and partly reflected (see Figure 7). When linearly polarised light, whose polarisation plane is parallel with the water surface, shines on the surface under the so-called Brewster angle no reflection takes place. For this angle the equation tanϕ = n / no holds, in which ϕ is the angle between the incoming beam and the normal, n is the refractive index of the liquid (water) and no is the refractive index of air. Question: Calculate the refractive index for the air-water interface (n = 1.333).
- 45 -
laser CCD ϕ
Figure 7 Principle of the Brewster microscope and a picture of its actual design.
When the surface is examined at the reflection angle, while polarised laser light shines on the surface at angle ϕ , a dark image is seen. If the refractive index of the surface changes, because a monolayer is present, some reflection of light occurs. With the help of a magnifying glass and a sensitive camera (CCD), domains of material present at the surface can be made visible. The intensity of the reflected light depends on the difference in refractive index between water and the material and on the thickness of the layer. In the region where liquid expanded and liquid condensed phases coexist domains of both phases can readily be observed. The collapse of monolayers can also be seen. In the region of the curve where no pressure is measured, one can examine if the molecules spread homogeneously on the surface or if island-like clusters of molecules are present in which the molecules aggregate. Some examples of BAM pictures are seen in Figure 8.
Figure 8 BAM-pictures of pentadecanoic acid (left) and dioctadecyldimethylammonium bromide (right)
Sometimes circular domains are visible for liquid phases, like in a liquid condensed phase in a liquid expanded phase of pentadecanoic acid (left) or a dendritic structure for a more solid phase of dioctadecyldimethylammonium bromide (right).
- 46 - 4.6 Molecular recognition at monolayers
Besides on pure water, monolayers can also be formed on water to which compounds have been added like salts, metal ions but also small organic molecules, polymers, proteins, etc. In this way, the interaction between these compounds and monolayers can be studied. An important reason for investigating interactions at monolayers is the fact that a monolayer strongly resembles a bilayer as it is present in vesicles and cell membranes. It is sometimes easier to do research on such a system than on actual bilayer vesicles or cell membranes in solution. We will discuss three examples.
4.6.1 Chiral recognition of taste compounds The enantiomers S(+)-carvone (present in caraway seeds and dill) and R(-)-carvone (present in spearmint) give a different taste (and smell). This difference is caused by a different interaction with our taste receptors (although 10% of the people can not tell the difference). These receptors are membrane-bound proteins that are present in special sensory cells. The amphiphiles that constitute the membrane are however also chiral. When these L-phospholipids, e.g. L-dipalmitoylphos- phatidylcholine (L-DPPC) are introduced as a monolayer on a subphase containing 5 mM S- or R- carvone, then it is found that the monolayer with R-carvone in the subphase has a larger area at the same pressure. Although the chiral carbon atom of the phospholipid is not directly involved in binding, the difference in binding with S- or R-carvone is clear. In addition, from other measurements, for example the behaviour at different temperatures, it is clear that R-carvone binds stronger to L-DPPC monolayers than S-carvone. This type of observations suggests that besides the stereochemistry of the membrane proteins also the stereochemistry of the phospholipid component of the membrane-bound protein can play a role in taste and smell perception.
O O C H O O N 15 31 * O C15H31 O O O P O O S(+)-carvone R(-)-carvone DPPC
4.6.2 Interactions with chiral amino acids Chiral interactions have also been observed between amphiphiles with a chiral amino acid head group. The π-A curves of racemic and enantiomeric pure stearoylserine are different. The area necessary for the racemate is clearly larger. Thus, poorer packing is possible for these compounds. It means that the molecules prefer a homochiral packing. This is found for strong (electrostatic) interactions between 2 of the 3 groups that are in contact with water (see Figure 9).
- 47 - O O H racemic C H π H3C O N 17 35 * H OH enantiomer pure stearoylserinemethyl ester A
Figure 9 Top: π - A curves of racemic and enantiomer pure stearoylserine methyl ester. Bottom: Model for favourable homochiral interactions; strong interactions between 2 of the 3 non-equal groups (A+ and B-) gives the best packing for enantiomers (b) and favourable heterochiral interactions; strong interactions between equal groups gives the best packing for the racemate (c).
4.6.3 Biotine - streptavidine interactions As is true for any receptor: the more binding places the more precise the fit of the receptor with the substrate and the stronger the binding. Very strong binding occurs between certain proteins like enzymes and their natural substrates and substrate inhibitors. One of the strongest bindings known is the binding between streptavidine and biotine.
O OH O O H O N S O N NH H O S Biotine containing lipid O
When a monolayer is made of an amphiphile that contains a biotin fragment in its head group and streptavidine is added to the subphase, a 2-D domain of streptavidine molecules under the monolayer is formed very rapidly. Streptavidine has 4 binding sites for biotin, of which 2 are present on each side of the molecule. This means that the streptavidine monolayer can now act as a matrix for the formation of subsequent monolayers of other molecules that are brought in the subphase and that contain a biotin fragment. In the example (Figure 10), biotinylated FAB is used.
- 48 -
Figure 10 Construction of a 2-D domain of proteins at a lipid monolayer.
4.7 Applications of monolayers
One of the simplest applications of monolayers on water is the prevention of evaporation of the water. Furthermore, the reduced surface tension makes it impossible for certain insects to “hang” on the surface. This can be used for the fight against mosquito larva and malaria. As seen in the previous paragraph, monolayers are in certain aspects good models for bilayer membranes and are often used as model systems to study processes in bilayers. Monolayers form a very clean surface at which crystallisation can be studied and also be controlled. We have discussed the 2-D crystallisation of protein molecules, but the crystallisation of minerals can also occur in a special way at monolayers. With the aid of the Langmuir-Blodgett technique, it is possible to transfer monolayers to a solid substrate. It is also very well possible to transfer more monolayers after one another for the construction of a well-defined and ordered multilayer system.
4.8 Literature 1. R. H. Tredgold, Order in thin organic films, Cambridge University Press, 1994 2. J.-H. Fuhrop en J. Köning, Membranes and molecular assemblies: the synkinetic approach, Royal Society of Chemistry, 1994.
- 49 - 4.9 Intermezzo: AMPHOTROPES From Liquid Crystals 13, 1993, 57-69
The supramolecular self-organisation of an amphotropic cholesterol derivative
Micelles, liposomes and liquid-crystalline phases
by GERO DECHER and HELMUT RINGSDORF
The cholesterol derivative tetraethoxycholesteryl semisuccinate is both a mesogen and an amphiphile. This combination of both molecular prerequisites permits two types of supramolecular self-organization: the formation of a thermotropic liquid-crystalline phase and of various forms of aggregates in contact with water or other solvents. Depending on the pH of the aqueous medium the compound self organizes in micelles or liposomes. At high concentrations lyotropic liquid- crystalline phases are obtained. The formation of liposomes and lyotropic phases is not restricted to water as a solvent but can also be induced in pure organic media such as water-free diethyleneglycol. Due to the broad range of supra- molecular structures that depend both on molecular shape and on amphiphilic properties we propose to call the title compound a model for amphotropic phase behavior.
1. Introduction In life science, the self-organisation of matter in supramolecular assemblies is one of the basic prerequisites for all biological processes. Cellular liquid crystals are necessary tools for the chemical, electrical and mechanical functions of life [1]. Compartmentation of cells and subcellular organelles is enabled by the self-organisation of amphiphilic lipid molecules in an aqueous surrounding. The driving force of this self-assembly, the hydrophobic effect, is also responsible for the formation of lyotropic liquid-crystalline phases and micelles, which are themselves important biological structures. In materials science, self-organisation plays an important role in the field of thermotropic liquid crystals. They are interesting and potential candidates in new areas such as non-linear optics, optoelectronics, information storage or self- reinforcing plastics. The molecular basis for this type of self-organisation are mesogens, which are composed of rigid, form-anisotropic (for example rod- or disc-like) molecules. Despite the fact that supramolecular self-organisation is one obvious connection between the two areas life sciences and materials sciences, the gap between the fields has not really been bridged [2]. Early classifications even termed some mesogenic molecules containing classical amphiphilic functional groups such as the carboxyl group and aliphatic tails 'non-amphiphilic' [3]. Even now, there exist only a few compounds, which are able to self-organise, both via their amphiphilic and mesogenic subunits. In this connection we present a simple derivative of cholesterol, tetraethoxy- cholesteryl semisuccinate 1, which is able to self-organise in a broad range of supramolecular structures (see figure 1) depending on both its amphiphilic properties (micelles, monomolecular layers, bilayer membranes, lyotropic liquid- crystalline phases) and its shape anisotropy and mesogenic structure (thermotropic liquid crystalline phases).
- 50 - O HO O O O O O O Structural formula of tetraethoxycholesterolsemisuccinate 1
The formation of lyotropic liquid-crystalline phases and liposomes is not restricted to water as a solvent and occurs also in water-free diethyleneglycol (DEG). Therefore, the title compound is a model for the ability to self-organise via two entirely different routes. Such a combination of mesogenicity (thermotropism) and amphiphilicity (lyotropism) in a single molecule could be termed amphotropism [4] as already mentioned in a review that includes a preliminary report on 1 [2].
2. Materials and methods Tetraethoxycholesteryl semisuccinate 1 was synthesised by a standard esterification from 1.288 g of tetraethoxycholesterol [5] using a 1-5-fold excess of succinic anhydride and triethylamine as base in 30 ml of CH2Cl2 as solvent. Unreacted succinic anhydride was quenched with diethyl amine after a reaction time of 2 hours. Another 50 ml of CH2Cl2 were added and the amines were extracted three times into 20 ml of an aqueous solution containing 0.1 mol of HCl and 0.1 mol of NaCl. The organic phase was dried with magnesium sulphate and the solvent was distilled off. The residue was chromatographed on silica gel with a solvent mixture of ethyl acetate, petroleum ether and tetrahydrofuran (2:2:1) containing 2 vol.% glacial acetic acid. The product fractions were collected, the solvent distilled off and the product dried at 10-2 torr and a temperature of 363 K. The material was re-dissolved in hexane and filtered through a 1.0 m Teflon filter. Finally, the hexane was distilled off and the product dried again for two hours at 10-2 torr and a temperature of 393 K. 0.992 g of pure tetraethoxycholesteryl semisuccinate C39H6608 (MW 662.95) were obtained as a colourless grease-like material corresponding to a yield of 65 per cent.
Figure 1. Schematic of molecular structure and Figure 2. DSC heating and cooling scans of pure 1 supramolecular aggregates of 1. (heating/cooling rate 10 K/min
- 51 - 3 Results and discussion This cholesterol derivative 1 is able to self-organise in a broad range of supramolecular structures as schematically depicted in figure 1. The liquid-crystalline state of matter was discovered slightly over 100 years ago with the synthesis of some cholesterol fatty acid esters [6,7]. While the function of cholesterol in the living body, as well as its role as an amphiphile in model membrane systems, have been extensively investigated [8], it was only recently discovered that some derivatives of cholesterol are able to form bilayer membranes and liposomes by self-organisation of the pure compounds in water [5,9,10]. The phase behaviour of liquid crystals is induced either by temperature (thermotropism) or by the influence of a given solvent (lyotropism). The vast majority of liquid crystals belongs to either one of these groups though some exceptions have been reported previously (see, for example, [10-17]). There is, however, still an argument as to whether the thermotropic phase of some substances is truly dependent on their mesogenic properties or whether it is a pseudo- thermotropic lyotropic phase [11]. In the case of compound 1, the observed liquid-crystalline phase is a true thermotropic phase, since other cholesterol derivatives like cholesteryl myristate, a classic non-amphiphilic thermotropic liquid crystal, also exhibit the same liquid-crystalline phase (SmA). The phase behaviour of tetraethoxycholesteryl semisuccinate 1 was analysed by calorimetry, X-ray diffraction and polarisation microscopy. Figure 2 shows the heating and cooling DSC scans of pure 1. It exhibits a thermotropic liquid-crystalline phase ranging from 245 K to 386 K. On cooling the compound turns into a glass at 245 K instead of crystallising, a behaviour normally only observed with polymeric liquid crystals. Attempts to prepare these so-called ordered glasses from non- polymeric liquid crystals by rapid cooling are not new, but in most systems, these glasses undergo recrystallisation upon heating (see, for example, [18-20]). The cholesterol derivative 1 shows a completely reversible glass transition (with preservation of order in the glassy state), which is independent of the cooling rate. Attempts to induce crystallisation by prolonged annealing at temperatures slightly above the glass transition temperature did not produce any significant changes in the DSC scan. The enthalpy of the phase transition at 386 K is 5.1 kJ mol -1 and is in the range expected for a smectic to isotropic transition. X-ray diffraction supports the view that the observed mesophase belongs to the bilayer-smectic type. The upper trace of figure 3 (A) shows five orders of reflection for the smectic layer spacing of 5.35 nm. From space filling models, a molecular length of 3.74 nm is calculated assuming a maximum extended conformation. Since the layer spacing is larger than one but smaller than two molecules, the smectic layers must be bilayers in which two molecules are arranged either in an interdigitated way (see figure 3 (C)) or highly tilted or we have to assume a non-extended conformation. If we assume an interdigitated structure the observed layer spacing is matched when both the hydrophilic and hydrophobic side-groups are interdigitated as shown in figure 3 (C). A smectic phase is also in agreement with phase-contrast microscopy. Figure 4 (A) shows the characteristic texture, smectic fans and a homeotropic background, of 1 at room temperature after rapid cooling from the isotropic state. Cooling the isotropic melt to 385 K results in the formation of batonnets (see figure 4 (B)). If the compound is cooled slowly between two hydrophilic glass plates, separated by less than approximately 50µm, only the homeotropic texture is observed. Upon shearing, birefringence is observed immediately, but the texture returns to homeotropic within minutes, since the carboxylic terminal groups of 1 anchor strongly to the glass thus favouring homeotropic alignment.
- 52 -
Figure 3. (A) X-ray diffractogram of 1 at 296 K (reflected intensity versus scattering angle). Upper trace thermotropic liquid crystal
phase, lower trace lyotropic phase containing 30 wt% water (the
symbol •10 at 2θ = 2.5º denotes a tenfold increase in detector sensitivity). (B) Ball and stick representation of 1 indicating the estimated length of the molecule and its sub-units. (C) Schematic
of the arrangement of the molecules in the bilayer smectic A phase
assuming an extended conformation of the molecules and a non-
tilted, interdigitated ordering that matches the observed layer
spacing
Figure 4. Birefringent textures of liquid crystal phases. (A) and (B) correspond to the thermotropic phase, (C) and (D) to lyotropic phases. (A) Smectic fans at 303 K after cooling from the isotropic melt. (B) Batonnets at 385 K after cooling from the isotropic melt. (C) Oily streaks of a sample with 30 wt.% water. (D) Oily streaks of a sample with 30 wt.% diethyleneglycol. - 53 - Due to its amphiphilic structure 1 can also self-organise in water or protic organic solvents using the driving force of the hydrophobic effect. As already shown in figure 1 micelles, monomolecular layers at the air/water interface and liposomes are formed. Depending on the pH of the aqueous medium the terminal carboxylic group allows for protonation/deprotonation and thus regulates hydrophilicy, solubility and aggregation behaviour. At pH 9.3 (equivalence point), the sodium salt of compound 1 shows a critical micelle concentration of 1.6x10-3mol l-1 (figure 5(A)). At higher concentrations, the micellar phase shows birefringence, indicating the formation of a lyotropic phase.
Figure 5. Determination of molecular aggregation with water. (A) Determination of the critical micelle concentration at pH
9.3 (surface tension not scaled with respect to pure water). (B) Surface pressure versus area isotherm of a monolayer of 1 on pure water at pH 5.5.
A lamellar lyotropic phase can also be obtained directly by extensively mixing pure 1 with water or diethyleneglycol. Figures 4 (C) and (D) show the textures of the lyotropic lamellar phases of 1 with 30 wt.% of these two solvents. Interestingly, 30 wt.% water does not significantly increase the bilayer repeat distance as seen by X- ray diffraction (see lower trace of figure 3 (A)). This seems to indicate that the water is incorporated in the hydrophilic head group region and does not separate the layers at this concentration. At pH 5-6 1 is water insoluble and shows a different aggregation behaviour. At the air/water interface 1 forms stable monomolecular layers (see figure 5 (B)) that can be compressed into a tightly packed state. However, attempts to deposit multilayers of 1 onto solid supports by the Langmuir-Blodgett technique, either from pure water or from a
CdCl2 containing subphase, proved unsuccessful. Swelling of a thin film of 1 in contact with water or water-free diethylene glycol leads to the formation of stacks of bilayer membranes and giant liposomes (see figures 6 (A) and (D) as documented by phase contrast microscopy. In addition figure 6(D) shows an electron micrograph (freeze etching technique) of a multilamellar liposome of an identical preparation in water. The spherically closed structure of the lipid bilayer was confirmed by a dye entrapment experiment. Liposomes were prepared by injecting a pentane solution of 1 into an aqueous solution of pyrenetetrasulfonic acid tetrasodium salt at 323 K. After removal of the non-encapsulated dye by gel permeation chromatography, the liposomal fractions were investigated by phase contrast (see figure 6 (C)) and fluorescence microscopy. It was confirmed that the dye molecules were completely entrapped within the liposomes inner compartments.
- 54 -
Figure 6. Characterisation of liposomal aggregates. (A) , (B) and (C) are optical phase-contrast micrographs. (A) After swelling of a thin film of 1 in excess water. (B) After swelling of a thin film of 1 in excess diethyleneglycol. (C) Liposome preparation by pentane injection method. (D) Freeze-fracture electron micrograph of a multilamellar liposome after swelling of a thin film of 1 in water.
4. Summary and conclusions The broad range of supramolecular aggregates that are formed by tetraethoxycholesterol semisuccinate 1 and described here is easily visualised based on its molecular structure. The formation of thermotropic mesophases is well known from similar ethers of cholesterol. All experimental data point to a bilayer smectic A phase. The lyotropic phase behaviour is strongly influenced by the carboxylic acid group. At low pH the carboxylic acid group is protonated and the compound is water-insoluble. Therefore, it can be spread at the air/water-interface and compressed to a tightly packed monolayer. When bulk material is brought into contact with water, the solvent penetrates into the bilayers and gives rise to a lamellar lyotropic structure. When more water is added, the bilayers finally separate and form spherically closed membranes (liposornes). At high pH the carboxylic acid forms a salt, rendering the compound water-soluble and resulting in the formation of micellar aggregates. The lyotropic phases are not restricted to water as a solvent. A lamellar phase and the formation of liposomes are also observed in diethyleneglycol. Tetraetboxycholesteryl semisuccinate is one example for a molecule, which achieves its tailor-made physical properties from the combined properties of its sub-units. In other words, its designed molecular architecture leads to the intended formation of various supramolecular assemblies. In the life sciences it is easily visualised that the complexity of the biological functions is related to the size of the functional entity, being either macromolecules (proteins or DNA), aggregates of macromolecules or aggregates of macromolecules and many small molecules (see, for example, [21,22]).
- 55 - In this report, we have discussed one example in the area of materials science, where a similar correlation of functionality (aggregation behaviour) and molecular size holds true. The combination of mesogenicity and tuneable amphiphilicity in a single molecule requires a certain molecular size and leads to the observed amphotropism.
References [1] BROWN, G. H., and WOLKEN, J. J., 1979, Liquid Crystals and Biological Structures (Academic Press).
[2] RINGSDORF, H.,SCHLARB, B.,and VENZMER, J.,1988, Angew.Chem.,100,117;1988, Angew. Chem. Int. Ed., 27, 113.
[3] GRAY, G. W., 1974, Liquid Crystals and Plastic Crystals, vol. 1 edited by G. W. Gray and P. A. Winsor (Ellis Horwood
Ltd), p. 103.
[4] Amphotropism, amphotropic: a combination of two driving forces for the molecular self-organization. The expressions are derived from: amphi-, amph[o]- [from Greek: αµφι−, αµφ[ο]-] both, on both sides, of both kinds; and -tropic [from
Greek: τροπη, τροπς] that which turns, turn, direction. Amphotropic is also used for a subclass of murine retroviruses. There are three types of oncoviruses, ecotropic ones that affect only murine cells, xenotropic ones that affect only non- murine cells and amphotropic ones, that affect both. Additionally in a recent publication (FULLER, S., HOPWOOD, J.,
RAHMAN, A., SHINDE, N., TIDDY, G., ATTARD, G. S., HOWELL, O., and SPROSTON, S., 1992, Liq. Crystals, 12, 521) the term amphitropic has been used in this context.
[5] PATEL, K. R., LI, M. P., SCHUH, J. R., and BALDESCHWIELER, J. D., 1984, Biochim. biophys. Acta, 797, 20. [6] REINITER, F., 1888, Mh. Chem., 9, 421.
[7] LEHMANN, O., 1890, Z. Kristallogr., 18, 464. [8] GIBBONS, G. F., MITROPOULOS, K. A., and MYANT, N. B., 1982, The Biochemistry of Cholesterol (Elsevier
Biomedical Press).
[9] BROCKERHOFF, H., and RAMSAMMY, L. S., 1982, Biochim. biophys. Acta, 691, 227.
[10] ABID, S. K., and SHERRINGTON, D. C., 1987, Polym. Commun., 28, 16.
[11] MARKUS, M. A., and FINN, P. L., 1985, Molec. Crystals liq. Crystals Lett., 2, 159. [12] KELLER-GRiFFITH, R., RINGSDORF, H., and VIERENGEL, A., 1986, Coll. Polym. Sci., 264, 924.
[13] BODEN, N., BUSHBY, R. B., FERRIS, L., HARDY, C., and SIXI, F., 1986, Liq. Crystals, 1, 109. [14] LEHMANN, B., and FINKELMANN, H., 1986, Coll. Polym. Sci., 264, 189.
[15] BRUCE, D. W., DUNMUR, D. A., LALINDE, E., MAITLIS, P. M., and STYRING, P., Nature, Lond., 323, 791.
[16] BAZUIN, C. G., GUILLON, D., SKOULIOS, A., and NICOUD, J.-F., 1986, Liq. Crystals, 1, 181.
[17] ZIMMERMANN, H., POUPKO, L., LUZ, Z., and BILLARD, J., 1989, Liq. Crystals, 6, 151. [18] SORAI, M., YOSHOKA, H., and SUGA, H., 1984, Liquid Crystals and Ordered Fluids, vol. 4, edited by A. C. Griffin and J. F. Johnson (Plenum Press), p. 233.
[19] ROSTA, L., 1985, Molec. Crystals liq. Crystals, 127, 195.
[20] KUNIHISA, K. S., and SATOMI, Y., 1986, Molec. Crystals liq. Crystals, 141, 1.
[21] LUISI, P. L., 1983, Chimia, 37, 73.
[22] SRERE, P. A., 1984, Trends biochem. Sci., 9, 387.
- 56 - Chapter 5: Polymers
5.1 Introduction.
Polymers are the most important class of organic construction materials. Even in the Stone Age, (bio)polymers were used as construction material, e.g. wood, wool, cotton, horn, leather, etc. Nowadays all kinds of new synthetic polymers are also available, like polyethene, polystyrene, PVC, polyesters, butadiene rubber, to name only a few of the most well-known polymers. The mutual interactions between the individual polymer chains largely determine the material properties of these polymers. For this reason, a short treatment of polymers in this course is justified. Furthermore, some classes of polymers show special supramolecular ordering, like liquid crystalline polymers that will be treated in the chapter on liquid crystals. In the following paragraphs, some polymerisation reaction mechanisms will be treated and we will treat the properties of especially linear synthetic polymers in some more detail. This chapter can also be considered as an introduction to biopolymers like proteins that will be treated elsewhere in more detail.
5.2 Polymerisation reactions.
A polymer can be formed from its monomers in different ways. One can make a distinction between addition polymerisation and condensation polymerisation. In addition polymerisation the monomer contains a double bond or a ring system that is opened during the polymerisation reaction. In fact, the polymer is formed by a shift of the electrons (or bonds). In general, an initiator is necessary for this reaction to occur.
X = H; polyethene n X = CH3; polypropene X n X = COOCH3; polymethylacrylate X X = C6H5; polystyrene
X X= O; polyethyleneoxide n X = CH CH CH C(O)NH; Nylon 6 X n 2 2 2
Condensation polymerisation is characterised by the expulsion of a group, often water or HCl, or a reorganisation of atoms takes place, as in the formation of polyurethanes, in which among others an H-shift occurs. The products of these reactions are often carboxylic acid derivatives like esters or amides. Some well-known biopolymers like polysaccharides, proteins and DNA are in fact condensation polymers.
- 57 - Question: Which groups of the monomers are involved in the condensation polymerisation leading to polysaccharides, proteins and DNA. What is split off? O
H2N Cl NH2 + Cl O O H N + HCl Nylon 6,6 N H n O
HO N C O OH + O C N
O H N O N O 6,4-polyurethane H n O
5.3 Mechanisms for the addition polymerisation.
Addition polymerisations do not occur spontaneously. An initiator is needed. The reactions proceed as follows: first, an active initiator is formed from the initiator molecule, for example by bond breaking. The active initiator can be a radical or an ion (anion or cation). Then this active initiator reacts with a monomer M, that becomes then activated itself by this process.
R-I R* + I* (Initiation) I* + M IM*
After that, it can react further with another monomer, etc which keeps the polymer growing.
M IM* IMM* I(M) n* (Propagation)
Finally a so-called termination reaction will take place, after which the polymer stops growing. This termination process can occur through several mechanisms.
I(M) * + A-X n I(M)nA + X* (Transfer)
I(M)n* + I(M)m* I(M) n(M) mI (Recombination)
+ (Disproportionation) I(M)n* I(M)m* I(M)n + I(M)m
- 58 - In a termination according to a chain transfer reaction, a polymer chain will react with a molecule A-X to give a group A at the end of the polymer and a new active initiator X* is formed, that can again start as a beginning for a new polymer chain. In this reaction, the number of active initiator molecules does not change. In the recombination process, two active centres react with another to a non-reactive compound. In this process two active centres disappear and one longer chain is formed. With the disproportionation reaction, two active centres react with one another to give two non-reactive molecules, usually by transfer of an atom (see also next section).
5.3.1 Radical polymerisation. Examples of radical initiators are the following compounds. Some give radicals upon heating, others by irradiation with light (usually UV).
O O O ∆T 2 O O O• Benzoyl peroxide
CH3 CH3 CH ∆T 3 H C N N CH 3 3 2 H3C • + N2 CN CN CN 2,2'-Azobisisobutyronitril (AIBN)
O OH O HO hν + • • H H Benzoïne
A radical polymerisation can now proceed as follows:
Y + I I I Y Y Y n
Termination can occur by recombination of two polymer radicals or by a chain transfer reaction with a solvent molecule, monomer, polymer or initiator. It is also possible that a disproportionation reaction occurs.
Y Y Y H I I 2 Y I + n Y Y H n n H
- 59 - These termination processes give a large spread in polymer chain lengths. By using recently developed initiators and catalysts, it is possible to carry out radical polymerisations with a small spread in chain lengths (molecular weight). With these methods, one ensures that fast chain transfer reactions occur, that constantly generate new radicals at the ends of the polymer chains and make them disappear again in the form of a double bond. In the radical form, the chain grows but with a terminal double bond, it does not grow. However, it can be brought back in the growing (radical) form.
5.3.2 Ionic polymerisation Ionic polymerisations proceed as follows: first, an initiator molecule A-B is split in a cation and an anion under the reaction conditions.
A–B A + B
Depending on the nature of A+ and B- and the properties of the monomer a further reaction can occur: in an anionic polymerisation B- reacts with the monomer. If an electron-withdrawing group R (e.g. ester, cyano) is present, the formed anion is stabilised.
B + B C C B(M)n R R
This can than react further with other monomers (propagation). Termination through recombination of two polymer fragments will not occur in this case, because they will repulse one another because of their negative charges. After all monomers are consumed, the polymeric anions will remain in solution, because anions (and cations) are much more stable than radicals. After adding more monomer, the polymer chains will grow again. This is why these polymerisation reactions are called living polymerisation reactions. An important difference with radical polymerisations is that chain transfer reactions and spontaneous termination reactions do not occur. Therefore, these reactions give polymers with a much narrower molecular weight distribution than radical polymerisations. The reaction can finally be terminated by addition of for example water or an alcohol or an alkyl halogenide
B(M) + H O n 2 B(M)nH + OH
- 60 - Example of an anionic polymerisation:
Li + Y Li butyl lithium Y n M H2O + LiOH n n H Y Y Y Y
Example of an cationic polymerisation: For this type of reaction a Lewis acid is often used in combination with a protic co-catalyst like alcohol or water, for example:
+ H TiCl4 + CH3OH Cl4Ti O CH3
The ratio between these two catalysts is rather critical and depends on the equilibrium constant + under the reaction conditions. Enough H has to be generated, but an excess CH3OH terminates the reaction:
n M H + H R R
CH3OH H H OCH3 n n R R R R
5.4 Non-linear polymers
In the preceding section, we assumed that linear polymers are formed. With radical polymerisations there is a reasonable chance that side-reactions occur that can lead to branching of the polymers. A radical may seize a proton or another group from the polymer, leading to a new radical at a position somewhere in the middle of the polymer chain, from which a new chain may grow (= chain transfer).
- 61 - M M X X n X n X n
X m
Linear and slightly branched polymers are also called thermoplastics, because they become liquid upon heating. In the liquid phase they can be poured in a matrix, this upon cooling yields an object with a certain shape, for example a bucket. When so-called bifunctional monomers are added to the polymerisation mixture, crosslinks are formed between the polymer chains, which produces a network structure. When the number of crosslinks is low, this yields rubber-like materials. An example is polystyrene that can be crosslinked by adding divinylbenzene. With more crosslinks, the material becomes harder and less easy to deform.
+ 100 : 1
Styrene Divinylbenzene
Crosslinking can also be done in a later stage with preformed polymers that still contain reactive groups. This is done with the preparation of natural rubber. In this vulcanisation process, sulphur is added, which connects the unsaturated polymer chains with another into a network.
+ Sn Sn
Something similar is found in proteins, where two cystein residues can form a crosslink between two distant positions of a peptide chain.
- 62 - Obviously, crosslinked materials will not become liquid upon heating. Materials with many crosslinks have to be polymerised in the shape of the object. This is why these materials are often called thermosets. A rather old example is Bakelite (the first commercial polymer, called after the Flemish chemist Baekeland).
OH OH OH H
+ O (1 : 1.5) + H2O H
In this reaction phenol and formaldehyde are mixed in a 2 : 3 ratio and heated, yielding a highly crosslinked material. In practice, a prepolymerisation is performed with a lower concentration of formaldehyde. This prepolymer is then mixed with filler, dye and an additional amount of formaldehyde and brought in a matrix and heated under pressure, which gives the last crosslinks and the material becomes hard in its final correct shape. For modern materials, linear polymers are usually employed that still contain polymerisable groups. By later adding additional monomer and an initiator, crosslinks are formed, which cause the material to become hard. Examples are polyesters, which together with fillers like glass fibres form an important class of construction materials.
HO HO OH O O OH + O O O O
m O O O O O O + O O n
First a polycondensation of glycol with maleic acid is performed that yields rather short polymers (M ~3000). These are then dissolved in styrene, to which an initiator is added (for construction materials many other compounds may be present like fillers, weakeners, dyes, etc). After this mixture is brought in the correct shape and heated, radicals will be formed. This causes the styrene to polymerise and form crosslinks between the original polymer chains. Something similar happens with alkyd resins that are used in paint industry. In painting polymers that still contain double bonds are applied as a thin film. Under the influence of oxygen and metal ions radicals are formed that
- 63 - subsequently induce the crosslinking of the polymer chains and are responsible for the formation of the solid paint film.
5.5 Linear polymers from different monomers.
When two different monomers M1 and M2 are used under free radical conditions, we can write down four different reactions for the propagation reaction:
k M 11 1 + M1 M1M1 k M 12 1 + M2 M1M2 k M 21 2 + M1 M2M1 k22 M + M 2 2 M2M2
When we define the reactivity ratios as r1 = k11/k12 and r2 = k22/k21 then these ratios actually describe the difference in reactivity of a certain end group radical with monomer M1 or monomer
M2. When for example r1 is smaller than 1, it means that ~M1• reacts rather with M2 than with M1. Now we can distinguish different situations: a) When the product r1r2 is about 1, a random polymer is formed, which means that monomers M1 and M2 are randomly taken up in the polymer. When both r1 and r2 are also about 1 and the amounts of M1 and M2 in the feed are about equal, about equal amounts of M1 and M2 are build into the polymer. This is for example the case when two monomers are used that strongly resemble one another, like a mixture of methyl methacrylate and ethyl methacrylate. Also, when the values of r1 and r2 are not very alike, the polymerisation will be random. When r1 = 10 and r2 = 0.1, r1r2 will be
1 and the chance that M1 will be introduced to both ~M1• and ~M2• will be 10 times higher than M2. b) When the product of r1 and r2 approaches 0, i.e. that r1r2 ~0, then alternating copolymers are formed, because
~M1• reacts predominantly with M2 rather than with M1 and
~M2• reacts predominantly with M1 rather than with M2
An example is the reaction of maleic anhydride with styrene. In this case r1r2 is practically 0. This means that an almost perfectly alternating copolymer is formed, even when both monomers are not present in the feed in equal amounts.
- 64 - + n O O O O O O
c) When r1r2 > 1, that means when r1 and/or r2 >1, block copolymers are formed. When for example r1>1 this means that ~M1• prefers to react with M1 rather than with M2. When also r2 >1, this means that growth mostly occurs with the same monomer as the radical end group. Only sporadically a different monomer is attached to the growing polymer and the polymer now grows further with a monomer that is also the radical at the end of the polymer. Well-defined block copolymers are however usually prepared in a different manner, i.e. by a living (ionic) polymerisation. In this case, one monomer is reacted until it is all used up. Then another monomer is added, after which the polymer chains can grow again. In this way well defined diblock-, triblock-, etc copolymers can be made. This kind of polymers is interesting, because they can be used as surfactants. Polymers like [PEO]n[PPO]m and [PEO]n[PPO]m[PEO]o behave as micelle formers in water.
O O O O O n m n m o Examples of diblock and triblock copolymers
5.6 General properties of linear polymers.
5.6.1 Molecular weight A linear polymer is build up from a certain number of monomeric units. Usually the polymer chains have a different length. The average molecular weight can be given in different ways: a) The number averaged molecular weight. The total weight is divided by the number of molecules. This is the same as taking the sum of the weights of the different fractions with the same molecular weight (nkMk) and dividing this by the total number of molecules: p ƒnkMk 1 Mn = p ƒnk 1
- 65 - b) The weight averaged molecular weight. Of every fraction with the same molecular weight Mk, the weight is determined (wk = nkMk). The average molecular weight is now:
p ƒ wkMk 1 Mw = p ƒ wk 1
When a material consists of polymers with the same chain lengths, than Mn = Mw and their ratio equals 1. The larger the spread in molecular weight, the more Mn and Mw will differ from one
Mw another. The polydispersity index: PDI = therefore is a measure for the molecular weight Mn distribution.
Figure 1 Average number (Mn ) en weight average (Mw ) in a typical polymer sample.
The values of Mn and Mw can be determined with different techniques. Some are good for determining Mn others are good for determining Mw . An analysis technique that is often used is gel permeation (or size exclusion) chromatography, in which polymers are separated, based on their size. The longest chains elute first, the smallest last. For a good analysis, the eluent is at the same time analysed with viscosity measurements and light refraction. The experimental details and the theoretical backgrounds of these techniques are rather complex and will not be treated here.
5.6.2 Phase behaviour Like many other compounds, polymers can exist in the solid phase and in the liquid phase. Some polymers crystallise below a certain temperature and thus exhibit a normal melting point where the crystalline phase transforms in the liquid phase. However, many polymers do not crystallise easily. Sometimes they can be crystallised by keeping them 10 – 30 °C below their melting point for hours,
- 66 - days or weeks. This temperature is low enough for crystallisation and high enough to ensure that the molecules are mobile enough to orient themselves and adapt at a crystal surface. Even then, crystallisation will not take place for 100%, because there are always irregularities, like end groups, branching points, etc. Other polymers will never crystallise. At low temperatures, they will be present as a glass. In a glass, the movements of the polymer backbone will be absent; that means rotations about the C-C bonds do not take place anymore, but the backbones will also not move anymore with respect to one another. The stopping of the motions of different elements in a polymer chain does not have to occur at the same temperature. For example in PMMA (polymethyl methacrylate) the glass transition temperature is at about 110°C, but the movements of the methyl ester groups do not stop until at ~20 °C. This influences the material properties and consequently the material becomes more brittle below this temperature.
n PMMA CH3 O O
The glass transition temperature depends on the flexibility of the polymer backbone and the interactions between the different chains. This is represented in the following figure where the influence of chain flexibility, interchain distance and chain interactions on the glass transition temperature of some polymers with a polyethylene backbone is given.
250 200 N
150 -CN 100 -Cl Tg 50 -CH3 0 -CH3 -CH3 -CH2CH3 -50 -CH2CH2CH3 -100 -CH2CH2CH2CH3 -H -150
Figure 2 Influence of chain flexibility (column 1), interchain interaction (column 2) and interchain distance (column 3) on the glass transition temperature for some poly-ene polymers is given (the groups in the figure are the side groups of the polyethylene polymer).
It can be seen in this figure that the larger the side group, the higher Tg, because this hinders the possibilities for rotation of the backbone (first column). Furthermore, it can be seen that Tg increases in the series of side groups that are about equally large, CH3, Cl and CN, this is caused by - 67 - the increase in polarity of the side groups and therefore the mutual interactions. When the size of the flexible alkyl side group increases, the interactions between the backbones decrease and Tg decreases. This is sometimes called a plasticizing effect. Above the glass transition temperature, the rubber state exists. The backbones are still almost immobile with respect to one another, because they are fixed by mutual entanglements. By exerting pressure on the material, it exhibits elastic behaviour, which means it has the tendency to revert to its former shape. Upon further increasing of the temperature, the number of fixation points decreases and the material starts to behave as a true (viscous) liquid. The viscosity of a polymer is much higher than of low molecular weight compounds and increases with molecular weight. Viscosity measurement is therefore one of the most useful methods for determining the molecular weight of a polymer. Polymers also increase the viscosity of a solution in which polymers are dissolved. This is caused by the possibility of entanglement of the polymer chains. It can easily be seen that this depends on the concentration of the polymer and on the length of the polymer.
5.7 Literature
1 A. E. Schouten and A. K. van der Vegt, Plastics, Delta Press, Amerongen (or as a Prisma pocket, Spectrum), 1987. (Dutch) 2 P. C. Hiemenz, Polymer chemistry; the basic concepts, Dekker, 1984. 3 G. Odian, Principles of polymerization, Wiley Interscience, 1991 4 A. A. Collyer, Liquid crystal polymers: From structures to applications, Elsevier, 1992 5 C. B. McArdle, Side chain liquid crystal polymers, Blackie, 1989.
- 68 - Chapter 6: Liquid Crystals
6.1 Phases and ordering Many solid materials that we know consist of crystals (sugar, salt, granite, and metal). In these materials, the molecules (or atoms or ions) have a fixed position with respect to one another in a grid. Under the influence of temperature, the atoms of the molecules only vibrate around their fixed position. For larger organic molecules, their orientation is also fixed. So one speaks of a positional and orientational ordering. Molecules form crystals in order to optimise their mutual interactions, like Vanderwaals interactions, dipole and hydrogen bond interactions. This is counteracted by thermal movements. Above a certain temperature these vibrations become so strong that the mutual interactions weaken and the fixed positions with respect to one another can no longer be maintained. The molecules start to move randomly and the positional and orientational ordering gets lost. The mutual interactions are still strong enough to minimise the empty space between the molecules (vacuum). A liquid is formed in which the molecules take up a certain volume of space. Upon further heating, the movements of the molecules become so strong that all interactions between the molecules get lost and the molecules completely fill the available space. We now call this a gas. At still much higher temperatures (at which organic molecules are already completely decomposed into atoms), a separate phase exists, in which the electrons are no longer bound to the nuclei of the atoms. This phase consists of cations and electrons and is called a plasma. Gas and plasma phases will not be discussed here. In the remainder, we will concentrate on solids and liquids. As we have seen, the molecules in a crystal have a fixed position and orientation in the crystal grid. The molecules are not mobile and they are highly ordered with respect to one another. In a liquid, the molecules are mobile and there is no fixed positional and orientational ordering. In the following scheme, the mobility is plotted versus against ordering.
crystal
mesoglass mesophase r e d r o glass liquid
mobility
Scheme I Phase behaviour as a function of order and mobility.
In this scheme, some phases are present that we have not discussed yet. They will be shortly treated here and in the remainder of the chapter they will be discussed more extensively.
- 69 - The glass phase is probably known already. As can be seen in the scheme, the molecules in a glass have no mobility with respect to one another. It is therefore a solid. On the other hand, the molecules are not ordered and thus it can be considered in this respect as an immobilised liquid. Examples are windowpanes, glazing on pottery, icing on food, some plastics. When a liquid is cooled it will generally crystallise. However, crystallisation is not an easy process, because it requires reorganisation of the molecules in the crystallising liquid. It is greatly enhanced by the presence of crystallisation nuclei. Compounds that do not crystallise easily can often be supercooled considerably before crystallisation occurs. If the liquid is cooled fast, there is a possibility that no crystallisation nuclei are formed. When the temperature drops below the so- called glass transition temperature, the mobility of the molecules disappears and they can no longer reorient, which is necessary for crystallisation. Now we are talking about a glass. Generally, a glass is easily formed when the molecules do not easily fit in a crystalline packing or when the viscosity of a liquid is high. Both conditions are often met by a polymer and polymeric materials are often present as glasses. Also simple compounds like water and alcohols can be obtained as a glass provided the cooling is fast enough. A very general rule is that the glass transition temperature is 2 about /3 Tm, in which Tm is the melting point of the compound in Kelvin. Therefore, it is not a thermodynamically stable phase. Even so, pane glass and plastics can remain in this state for ages. The other phase that is present in the scheme is the mesophase. Literally, it means middle phase. This phase is also called liquid crystalline phase and this name already indicates that we deal with a phase that has both liquid and crystal properties. From the scheme, it can be seen that the molecules move with respect to one another, but that the are also ordered. Although mesophases are not very well known, they are certainly not rare. An example of a compound that shows liquid crystalline behaviour is cholesteryl myristate, a compound that is present in cell membranes and in the deposit in blood vessels that causes diseases of the heart and blood vessels.
Cholesteryl myristate
O
O
At room temperature, the pure compound is a crystalline solid. At 71ºC it melts and forms a turbid, viscous liquid that transforms into a normal "clear" liquid at 85ºC. Between 71 and 85ºC a liquid crystalline phase is present, in which the elongated molecules move more or less parallel with respect to each other. So there is an orientation ordering. With this type of molecules, the mesophase is present in a certain temperature range. This is why these phases are also called thermotropic liquid crystalline. The strong mutual interactions between molecules in crystals can also be broken by dissolving the compound in a solvent. Often the dissolved molecules will move independently through the solution, but some molecules have a tendency to come together and interact with one another to
- 70 - form ordered aggregates. These phases (ordered aggregates in a solvent) are called lyotropic liquid crystalline phases (lyo = liquid). Their phase behaviour depends strongly on the concentration in the solvent (mostly water). Molecules that form these phases often have a strongly polar past (e.g. an ionic group) and an apolar part (e.g. an alkyl chain). Depending on concentration and molecular structure very different kinds of ordered structures can be formed. Examples are soaps and phospholipids. Biopolymers like DNA and proteins often also can form very ordered structures in water.
6.2 Types of liquid crystals
Liquid crystals are compounds that exhibit a liquid crystalline phase. The liquid crystalline phase is a middle phase (also called mesophase) between a liquid and a crystalline solid that is present in a certain temperature range. In the liquid crystalline phase the molecules move around, but have some ordering with respect to one another. This ordering gives special properties to the materials and is essential for applications of these compounds in devices like displays.
Isotropic Nematic Smectic
Figure 1. Phases in which molecules can be present. The two on the right side are examples of common mesophases or liquid crystalline phases.
The reason that some molecules form liquid crystalline phases has to do with their shape. It is not difficult to imagine that rod-like or disk-like molecules will order different from sphere-like or egg-like particles. Depending on the shape of the molecules three different types of liquid crystals are distinguished, that can each give different liquid crystalline phases. 1. Calamitic liquid crystals, consisting of rod-like molecules. 2. Discotic liquid crystals, consisting of disk-like molecules. 3. Sanidic liquid crystals, consisting of board-like molecules, having a long shape, but with a clear difference in thickness in the other two dimensions. In the following paragraphs, the properties of especially calamitic liquid crystals will be treated in more detail.
- 71 - R R
O O O N N N H R N R O N N N R H R N N N
NC R = -CH 2OC12H25
calamitic discotic R R
O OC5H11 O O O OC5H11 O C H O O 5 11 O C H O 5 11 O sanidic
Figure 2 Examples of calamitic, discotic and sanidic liquid crystals.
6.3 General description of liquid crystalline phases.
An example of a liquid crystalline compound is O 4-methoxybenzylidene-4’-butylaniline (MBBA). N Below 22 ºC this compound is crystalline. After MBBA melting the compound is present in a liquid crystalline phase and at 47 ºC it transforms in a true liquid phase (also called isotropic phase). From this, we see that the liquid crystalline phase, like other phases, is present in a discrete temperature range. The molecules in this phase are oriented approximately parallel to one another and on average point in the same direction. This average direction is called the director ( n ). However, individual molecules can have a rather large deviation from parallelity. A quantitative measure for this parallelity is the order parameter S:
S = 1/2<3cos2θ - 1>.
This order parameter is 1 for a perfect ordering and 0 for a normal isotropic liquid. The molecules in the liquid crystalline phase can undergo rotations, translations and conformational changes. In general, clusters of millions or billions of molecules exist in which the orientation of the molecules is on average the same (the size of the clusters is in the order of µm's). In other clusters however the preferred average orientation can be completely different. Because the molecules within such a cluster have a preferred orientation, the physical properties in such a cluster are not the same in very direction. This is called anisotropy. The anisotropy is found for many physical properties like viscosity, electrical and thermal conductivity, etc, but is especially - 72 - important for the optical properties. The refractive index is different for the different directions under which a cluster can be viewed. This is also the reason that a liquid crystalline material that contains these clusters appears turbid. Besides the property that liquid crystal molecules orient with respect to one another, they can also orient themselves along a surface (parallel or perpendicular) or under the influence of a magnetic or electric field. These properties are of utmost importance for all kinds of applications of these compounds.
Figure 3 Orientation of a liquid crystal in an electric field (left) or along grooves on a surface (right). These grooves can be obtained by rubbing a surface covered with a thin polymer layer with a cloth in one direction. The grooves have to be very narrow (~ 1 nm).
6.4 Methods for investigating liquid crystals.
6.4.1 Polarisation microscopy Because of the optical anisotropy, liquid crystalline phases can be examined with the help of a so-called polarisation microscope. A sample is placed under a microscope as a thin film between glass slides. The film has a thickness in the order of a few µm, which is still thousands of molecules thick. The sample can be heated with a small oven. Above and below the sample two polarisation filters are present that have their polarisation direction twisted at an angle of 90º with respect to one another (see Fig 4).
Figure 4 Principle of the polarisation microscope: The first polarisation filter lets only pass light that has the same polarisation direction as the filter (vertical in the figure). When the second filter is placed perpendicular to the first one, no light passes the second filter. When the polarisation plane is rotated slightly by a sample between the filters, there
is a horizontal component that can pass the second filter.
The first filter lets only pass light that is polarised in one direction. Because the second filter lets only pass light that is polarised in the direction perpendicular to the first, normally no light - 73 - passes this pair of filters. However, if the polarisation direction of the light is changed by the medium between the filters some light gets through the second filter. Because the liquid crystalline is anisotropic the refractive indices in the different directions are not the same and this can cause the polarisation plane of the light to rotate. This is the reason that for many liquid crystals some light passes the second polarisation filter. Depending on the orientation of the liquid crystal domains with respect to the propagation direction of the light different images or textures can be observed. Different phases show different textures that are often characteristic for a particular liquid crystalline phase and can be used to identify the phase. The texture is caused by the orientation of the liquid crystalline domains with respect to the polarisation plane and the boundaries between these domains, where defects in the ordering occur. With this technique, it is also very well possible to determine the phase transition temperatures of the liquid crystalline phases, because the temperature of the oven (hot-stage) can be changed in a constant programmable manner.
6.4.2 Differential Scanning Calorimetry (DSC) When a material undergoes a phase transition, like melting or crystallisation (or when it undergoes a reaction) heat is taken up or released. Many of these processes simply occur by a change of temperature, i.e. heating or cooling. An important and powerful technique to study especially phase transitions in a qualitative and quantitative manner is Differential Scanning Calorimetry or short DSC. With DSC apparatus a sample of the material and a reference material that does not exhibit a phase transition are placed in two different small ovens and simultaneously heated or cooled at a constant rate. Usually, the sample is weighted and placed in a closed aluminium pan and the reference is an empty pan. The energy that is needed to heat both samples or that is liberated when the samples cool are compared and the difference is plotted against temperature. Depending on the actual temperature and the programmed temperature more or less electrical current (= heat) is sent to both ovens.
sample reference
Controller and PC
heating ovens Figure 5 Schematics of a DSC.
In order to heat a material without a phase transition at a constant rate the following amount of heat has to be added: mA x CpA, in which mA is the mass and CpA is the heat capacity of material A. (Heat capacity is the amount of Joules that is required to increase the temperature of 1 gram (or 1 mole) of material with 1°C). The difference in heat that is needed to heat both ovens is mA x CpA - - 74 - mR x CpR; this is a constant value if CpA and CpR are independent of temperature, which is usually the case. This means that a straight line is found when the difference in heat flow (in J/s = Watt) is plotted against the temperature. When a phase transition occurs, for example when a compound melts, extra heat has to be added to the oven with the material, in this case melting energy (= mA x
∆HA).
Figure 6 Peak in a DSC thermogram caused by the melting of a compound (left) and an example of a thermogram (heat vs. T; note that in this case the arrow of the heat flow points down!) of a liquid crystalline compound (right). The transition from smectic-A to nematic has a very low heat effect in this case. In general the transition from crystalline to liquid (crystalline) has the highest heat effect.
When the peak is integrated against time, the enthalpy of the phase transition is obtained. However, the heat flow is usually plotted against temperature in order to simply see the temperature at which the phase transition takes place. Now when another, for example two times higher heating rate is used, the heat for the melting has to be added in a shorter time. Because the heat flow is measured in Watt (J/s), the peak is about two times larger in the corresponding plot of heat flow versus temperature. Therefore, the peak height in DSC is proportional with the amount of sample and with the heating rate. Besides melting of a compound (crystalline to isotropic), heat effects also occur at other phase transitions, like crystalline to liquid crystalline, liquid crystalline to liquid crystalline, etc. Sometimes even crystalline to crystalline transitions are found. The strongest change in enthalpy is usually found for the crystalline to liquid (crystalline) transition. Upon cooling the phase transitions are reversed. In addition, the heat effects are reversed: heat is released at a phase transition. The phase transition temperatures for the liquid crystalline transitions that are found upon heating or cooling are usually close together. For crystallisation often a supercooling phenomenon is found, that means that the compound can be cooled considerably below the melting point before crystallisation occurs. This phenomenon is often found for polymers and in these cases a transition to a glass is often found. At a glass transition, we do not find a peak in the DSC, but a change in - 75 - height of the base line. This means that the heat capacity of the material changes. This can be understood as follows: the temperature of a material is determined by the degree of movement of the atoms and molecules (translation, rotation and vibration). By adding heat, the atoms, groups of atoms and molecules move more vigorous and the temperature increases. For a material in the glassy state, the possibilities for movement are limited (translation is not possible anymore), therefore less heat is necessary for increasing the temperature of a material. This means that at the transition from glass to a liquid the heat capacity of a material increases.
Figure 7 Example of a DSC- thermogram that shows a glass transition at ~ 35ºC and a liquid > –
crystalline to isotropic transition at ~ w o
l 160ºC. Upon heating and cooling the f
t
a transitions are found at e
H approximately the same temperatures and the heat effects 20 40 60 80 100 120 140 160 are about the same (∆H at 160 °C Temperature (°C) and ∆Cp at 35 °C).
At a phase transition, e.g. melting, the mobility of molecules increases drastically, that means the ordaining decreases and the entropy increases. Because the enthalpy of a transition can be determined from the DSC-thermogram and because ∆G at a phase transition is 0, this means that ∆H = T∆S. from this formula the entropy change of the phase transition can be determined. When ∆S is divided by the gas constant R, a dimensionless unit ∆S/R is obtained, this is a measure for the order change due to the phase transition. Upon heating, we usually find endothermic peaks, because heat is required for a phase transition. Upon cooling, we find exothermic peaks. Sometimes however, exothermic peaks are observed upon heating. This can be caused by a chemical reaction that takes place in the material, but more often, this is caused by the crystallisation of a supercooled liquid. The reason is that crystallisation can be strongly suppressed by low temperatures due to the low mobility of the molecules. When the temperature is raised, the mobility of the molecules increases and crystallisation takes place after all. There are also some compounds that can exist in different crystal forms. When the low-melting form melts upon heating, it can crystallise in the high-melting form, which melts again at a higher temperature.
- 76 - 6.4.3 Röntgen reflection analysis This technique is very suitable for investigating ordered systems like crystals, but also liquid crystals and multilayer systems. The principle of Röntgen reflection is given by Bragg's law, which states that constructive reflection of electromagnetic radiation occurs when:
d = n λ / 2 sin ϕ , in which d is the layer thickness, n is an integer, λ is the wavelength of the radiation and ϕ is the reflection angle.
ϕ
d ϕ
d sinϕ Figure 8 Reflection of electromagnetic radiation on a layered surface. In the left hand figure α should be read as ϕ.
This law describes when constructive reflection at a layered system occurs. When this condition is met, the difference in path length (= 2dsinϕ) u pon reflection at different layers equals a integer number of waves of the radiation ( = nλ). The phase of the radiation reflected from different layers is the same, and thus constructive reflection occurs. When this condition is not met, extinction of radiation occurs. When the number of layers increases, the angle at which reflection occurs becomes more precise determined; only when this condition is exactly met reflected radiation can be measured. With this method information about layer thickness can be obtained. It is good to realise that Bragg's law applies for all kinds of electromagnetic radiation.
Röntgen radiation
The wavelength of the röntgen radiation that is often used is from Cu-Kα radiation (λ = 1.54 Å), but other wavelengths can also be generated with other metals. The radiation is generated in a röntgen tube, where under the influence of a high tension, for example 40 kV, electrons from the cathode impact on the anode. These fast electrons can, among others, hit and remove electrons from the inner electron shells (K-shell or s-electrons). The resulting hole is filled up from electrons from one of the outer shells, usually the L-shell (p-electrons). The energy that is released is emitted as radiation. Because this energy is exactly determined, the röntgen radiation that is emitted is also of an exact wavelength. With the aid of filters, radiation of other wavelengths is largely filtered away.
- 77 - Reflection from grids Ordered molecules (or atoms) in crystals form a 3-dimensional grid. Through these grid points various planes can be drawn. Reflection on these planes can occur under the right conditions of the angle and wavelength of light. In the figure some possible planes through a 2-dimensional grid are drawn.
Figure 9 Different possible planes in a hexagonal grid
For a 3-dimensional grid many different planes are possible, each with their own angles and layer distances. When a single crystal of a compound, in which the ordering in the whole crystal is the same, is placed in an X-ray beam, reflections occur with varying intensity under different angles. The intensity of such a reflection depends on the electron density in the layers from which reflection occurs and this depends again on how heavy the atoms in such a plane are. By exactly measuring the reflections in a shell around the crystal, and with the aid of computer programs, the repeating distances, the dimensions of the unit cells and the distances between the different atoms can be determined. Ultimately, the positions of all atoms in a unit cell can be determined and thus the exact molecular structure. The possibility to determine the exact molecular structures of large and complex molecules like proteins, DNA-fragments, etc, has increased strongly in recent years.
6.5 Description of some common liquid crystalline phases.
6.5.1 Nematic phase The liquid crystalline phase that resembles the normal isotropic phase most closely is the nematic phase. Rod-like molecules like MBBA form such a phase in which the molecules can move freely with respect to one another, but in which only an average preferred orientation of the long axis is present.
Figure 10 Representation of a nematic ordering (left) and two typical textures observed with polarization
microscopy (right).
- 78 -
With polarisation microscopy this phase often exhibits a thread-like texture (Figure 10; rightmost figure), hence its name (nema (Greek) = thread). This texture is caused by defect lines. Around a defect line in the nematic phase, the molecules are ordered radially. When a defect line is parallel with the propagation direction of the light (do not confuse this with the polarisation direction) a so-called Schlieren texture (Figure 8; middle) is seen with dark points, from which 2 or 4 light and dark bands sprout.
Figure 11 Defect lines in nematic phases.
Nematic liquid crystals tend to lie flat on a surface. When the surface is treated with certain polymers and rubbed in one direction, fine grooves are formed on the surface in which elongated molecules are accommodated (see figure 3). With this method it is possible to orient all molecules over a large area in the same direction. Because the molecules in the bulk will align with these surface molecules it is possible to obtain a rather well ordered liquid crystalline film. With röntgen reflection only a diffuse reflection peak is found that corresponds with an average distance of 4 to 5 Å. This is the distance between the parallel alkyl and aromatic groups.
6.5.2 Chiral nematic phase When chiral molecules form a nematic phase, they are no longer on average parallel oriented with respect to one another, but twisted slightly. The consequence is that macroscopically a helix structure is present in the phase. Depending on the nature of the molecules either a left- or a right-handed helix can be present. The pitch of the helix (i.e. the distance after which the molecules have the same average orientation) is usually between a few hundred to several thousands of nm large. This means that for a certain wavelength of light (which could be visible light), a layer structure is present of which the layer distance corresponds with the wavelength of that light in the medium. In that case, light of that wavelength is reflected from the material. For thicker sample often pearl of mother-like colours are seen. For thin films between two glass slides the first molecules are often parallel to the Figure 12 Ordering in a glass-slides and thus the helix goes from one glass slide to the other. chiral nematic phase.
- 79 - These films clearly reflect light of a single color and can in principle be used as a color filter. Light that falls on a chiral nematic phase is split into two components; left-handed and right- handed circularly polarized light. A right-handed helix reflects the right-handed polarized component and lets the left-handed component pass. The reverse is also true. The following simple relation exists between the wavelength of the reflected light (λ) and the pitch (p) in which n is the average refractive index. λ=n.p,
Fig. 13 The selective reflection wavelength can easily be measured by placing a thin film in an UV- Vis spectrometer. At the reflection wavelength half the amount of light is transmitted (i.e. with a circular polarization opposed to the sense of the helix). This gives a peak comparable to absorption. With circularly polarized light the sense of the helix can be determined. This is also possible with a circular dichroism measurement.
Addition of a small amount of a chiral compound that is not liquid crystalline by itself to an achiral nematic phase will induce the chiral nematic state. In many cases and for low concentrations of additive the inverse of the pitch (p-1) is proportional with the mole fraction (x) of the chiral additive:
p-1 =P.x
P is called the helical twisting power (HTP). This depends on both the nature of the chiral additive, the nematic matrix and the temperature.
6.5.4 Smectic phases When in a liquid crystalline phase there is not only orientational ordering, but also positional ordering in the form of a layer structure, these phases are called smectic. The molecules are still able to rotate and move (translate) with respect to one another, but translation from one layer to another is limited. Due to this higher ordering, smectic phases are more viscous than nematic phases. Many different smectic phases are known that are indicated with different letters. In general the ordering (and complexity) of the phase increases with the place in the alphabet. Compounds can often have different liquid crystalline phases depending on the temperature. Upon cooling from the liquid crystalline phase the ordering will generally increase upon going to lower temperatures. If
- 80 - there is a nematic phase it will occur at a higher temperature than the smectic phases. Now we will discuss some common smectic phases.
In the smectic-A phase (SmA), the long axis of the molecules will on average be perpendicular to the layer structure.
Figure 14 Structure of a smectic-A phase (left) and a typical polarisation microscopy texture (right)
When this smectic phase is moved (sheared) between two glass slides, the smectic layers will often become oriented parallel to the slides. In areas with such an ordering, there will be no texture visible with polarisation microscopy, because the refractive index of such a sample in the x- and the y-axis will be the same. Because light passes the sample in the z-direction, the polarisation plane will not be rotated and light can not pass the second filter. An ordering with the smectic layers perpendicular to the propagation direction of light is called homeotropic. For a non-homeotropic ordering of the SmA-phase, a typical fan-structure is often found (Figure 14). With X-ray reflection, we will see besides a diffuse band at a large angle, a sharp reflection at small angle. This latter reflection corresponds with the distance between the smectic layers.
Figure 15 a)Röntgen pattern for a smectic-A phase. The ring around the centre (θ is small; d is large) comes from the layer structure and the diffuse ring (θ is large; d is small) comes from the repeating distance between alkyl groups and aromatic groups. b) X-ray of an ordered sample
with the smectic layers vertical.
In the smectic-C (SmC) phase, the long axes of the molecules lie at a tilt angle with the layer structure. In a homeotropic ordering (in which the layers are parallel to the glass slides), Schlieren-like textures can be seen under a polarisation microscope. This is because the refractive indices in the x- and y-direction are now different from one another due to the tilt angle of the molecules.
- 81 -
Figure 16 Structure of a Sm-C phase (left), fan-shaped (middle) and typical Schlieren texture (right).
Figure 17 Schematic representation of the different liquid crystalline smectic phases. Triangles and arrows indicate the tilt directions. The open ovals on the bottom rows indicate the preferred orientations of the molecules perpendicular to their long axis. This can for example play a role in aromatic compounds. The difference between for example hexatic B and crystal B lies in the difference in long range ordering; the so-called correlation length for hexatic B is much smaller than in crystal.
In the other smectic phases there is also an additional hexagonal or pseudo-hexagonal ordering present within the layer structure. The consequence is that the diffuse ring in the X-ray pictures become sharper or become split-up in several sharp peaks because the short-range ordering (4 a 5 Å) becomes better. A good overview of the possible orderings in smectic liquid crystalline phases can be found in figure 17. Variables are whether or not a hexagonal ordering is present in the layers, whether or not a tilt angle is present or whether or not an additional rotational ordering (herringbone) ordering is present. The phases with the highest orderings can hardly be distinguished from crystals, also because they are almost not fluid anymore. Because calamitic - 82 - liquid crystals are not always symmetric, there appear to be several sub-types of ordering, for example within a SmA-phase. In literature one can therefore often find further subdivisions of the different liquid crystalline phases.
Figure 18 Some smectic phases represented in a different way. From left to right: SmB, Sml or SmF, SmE.
6.5.5 Chiral smectic phases Chiral compounds can also form smectic phases. When chiral molecules form a smectic-A phase, the phase itself is not specially ordered, but when there is a transition of this compound from chiral nematic to SmA, an interesting behaviour often takes place. Because the molecules have to go from a helix structure to a layer structure, the helix often unwinds close to the transition of the chiral nematic to the SmA-phase. Due to this unwinding, the pitch increases and thus the selective reflection wavelength also increases in a rather small temperature range. This has been applied in sensitive temperature measurements of surfaces. It is for example possible to detect inflammations by applying a thin layer of liquid crystalline material on the skin. The chiral nematic - smectic A transition has to be around 37°C and the normal selective reflection wavelength is around 400 nm (blue). This means that between about 40°C and 37°C the selective reflection increases strongly. This allows one to detect areas that are warmer and could be inflamed. In the chiral smectic-C phase (SmC*) there is also a helix structure, like in the chiral nematic phase. However, here the subsequent layers with tilted molecules are slightly rotated with respect to each other. Due to the macroscopic helix, chiral smectic-C phases can also selectively reflect light. The orientation of the tilt can be influenced by an electric field, and thus these phases can be used in displays that can theoretically be switched much faster than nematic displays. Other chiral tilted smectic phases have a similar ordering.
- 83 - Figure 19 Comparison between a non-chiral smectic-C, a chiral smectic-C and a chiral nematic (cholesteric) structure. Other tilted smectic phases behave similarly to smectic-C phases.
6.5.6 Discotic liquid crystalline phases These phases are found for molecules with a flat disk-shaped core to which flexible alkyl tails are attached. Some examples of molecules with a discotic liquid crystalline phase are represented in Figure 20.
R R R R
R R
R = -OOC-C7H15 : C 80 D 83 I R R N N N R R H O O2CC7H15 N N
C7H15CO2 O2CC7H15 R H R N N N R = -CH OC H : C 78 D 264 I C7H15CO2 O2CC7H15 2 12 25
H15C7CO2 O R = -O(CH2)6O(C2H4O)3CH3 : C -20 D 160 C 107 D 127 I R R
Figure 20. Some examples of discotic liquid crystals and their phase behaviour (C = crystalline; D = discotic; I = isotropic). The last compound is liquid crystalline at room temperature and comes from research in Wageningen on organic solar cells (J. Mater. Chem., 7, 615 (1997).
The ordering can be as represented in Figure 21. The ordering in the first picture is called discotic nematic. Higher orderings are called columnar phases. As can be seen in this figure, more types of ordering and phases are possible, just like in smectic phases for calamitic liquid crystals. Chiral discotic phases are also possible.
- 84 -
Figure 21 Structure models for a discotic nematic liquid crystalline phase (left) and different columnar phases (right).
6.6 Structure - property relations
As we have seen, rod-like molecules give liquid crystalline phases that differ from those of disc- like molecules. Calamitic liquid crystals are in general composed of the following structural elements: a rigid central piece (core), consisting of several aliphatic or aromatic rings A (often connected by bridging groups Z), to which flexible end groups R are attached. The structures are therefore often build-up as follows: R-(A-Z)n-A-R, in which n = 0, 1, 2, etc.
A Z R
CnH2n+1
O CnH2n+1 N O CnH2n+1 N N O N N O O CnH2n+1 O O Me CN Me
NO2
When the rigid core becomes longer or the alkyl chains, usually a better ordering of the liquid crystalline phase is observed. This means a stronger tendency to form smectic-A phases or smectic phases with a higher ordering. This can for example be seen in the following series of compounds, where elongation of the alkyl chain results in a change from a nematic phase into a smectic phase.
- 85 -
O
H2n+1Cn N N CnH2n+1
Figure 22 Transition temperatures in a homologous series of di-alkoxyazoxybenzenes. With longer alkyl chains, a smectic phase is formed instead of a nematic phase. The squares indicate the melting points.
In this figure we also see another effect that can sometimes be observed, that is that the physical properties vary as a function of the parity of the flexible alkyl group. The nematic to isotropic transition temperatures of the compounds with an even number of carbon atoms in the tails are generally slightly higher than for the compounds with an odd number of carbon atoms in the tail.
250 T/°C 225 O NC O (CH 2)n O 200
175 Figure 23 Example of a series of twin liquid crystals 150 and their transition temperatures. Both the smectic to 125 nematic (middle curve) and the nematic to isotropic 100 transition temperatures (top curve) clearly exhibit an 75 1 2 3 4 5 6 7 8 9 10 11 odd - even effect. n
This odd-even effect is much clearer in a special class of liquid crystalline compounds, the so-called twin (or dimer) liquid crystals. An example of such compounds from the research in Wageningen is shown in Figure 23, together with its phase transition temperatures (Receuil des Traveaux Chimiques des Pays-Bas, 113, 524 (1994); Liquid Crystals, 18, 843 (1995)).
- 86 - The reason for this behavior lies in the fact that the rigid parts are more or less parallel when they are connected with an even spacer. Thus, they have a better ordering in the liquid crystalline phase and thus become isotropic at a higher temperature than the compounds with an odd spacer.
even odd Figure 24 Models for the ordering in a twin liquid crystal that explains the alternating behavior.
6.7 Applications of liquid crystals
Applications of liquid crystals are largely based on the fact that the refractive index of the materials is anisotropic, and the fact that the orientation of the molecular aggregates can easily be changed under the influence of an electrical field. We shall briefly discuss the application in LC-displays (LCD).
The first discovered and still widely used displays, for example in calculators and wristwatches, are based on the so-called "twisted nematic" effect. A cell of this type looks as follows:
Figure 25. Model for a twisted nematic liquid crystalline display (TN)-LCD. On the left is the off state and on the right the on state. E = transparent electrodes with polymer alignment coating; G = glass; P = polarises; LC = nematic liquid crystal.
In the off state (without voltage over the cell) the liquid crystals are parallel to the surfaces that are coated with a polymer film. Because the surfaces are rubbed in directions that are perpendicular to one another and because the molecules align with the micro grooves parallel with the rubbing direction, there is a quarter helix in the bulk upon going from one glass plate to the other. On the outside the glass plates are covered with a polarization filter that lets light pass
- 87 - in a direction parallel to the rubbing direction. Light that falls on the cell is polarized and the polarization direction is rotated 90° by the quarter helix and the light leaves the cell through the second polarization filter. For many applications a mirror is placed below the second filter in such a way that light passes the cell again. When a voltage is applied over the cell, with the help of transparent electrodes (or a small part of it; called a pixel) all molecules will align parallel with the electric field. Now the polarization plane of light that passes the first filter is not rotated anymore and thus can not pass the second filter. This pixel will now appear black. The liquid crystalline material has to meet several conditions. It has to be liquid crystalline in the correct temperature range (in practice between -10 en +50 C) and the viscosity has to be rather low in order to be able to rapidly switch the orientation of the molecules. Furthermore, the voltage where the orientation of the molecules changes has to be as low as possible (the present day displays have switching voltages of 1 to 3 V). Furthermore the compounds have to be chemically and photochemically stable. Usually mixtures of compounds are used that have a large dipole moment and a low molecular weight. An example of an important component in these mixtures is the compound shown here. At the moment, liquid crystals are also being used in larger and more complicated displays like in computer monitors and televisions. The advantages are clear: cathode ray tubes as they are used in conventional monitors are heavy, they need high tensions which leads to the formation of ozone, the large volume of vacuum can lead to implosions when damaged but most importantly, they have a high energy consumption. In order to use liquid crystals for these displays several problems have or had to be solved. The most important ones are the response time, which has to be fast enough for fluid motion (>50 Hz) and the number of pixels has to be very high in order to get a high information density (typically 1 to 2 million pixels). This also means that every pixel needs a set of three electrodes (one for each basic colour). Furthermore, the intensity must be adjustable (grey values) and the angle under which a good image can be seen (viewing angle) should be large. Several of these problems have been solved, but some still need improvement. In the NC industry, people are still looking for better technology and also for liquid crystalline materials with better properties. Besides nematic materials that are normally used, compounds with a chiral smectic-C phase are promising, due to their fast response time compared to nematic materials.
- 88 - 6.8 Liquid crystalline polymers.
An important class of polymers in which additional ordering is present are the liquid crystalline polymers. These polymers combine the ordering behaviour of low molecular weight liquid crystals with the robustness and favourable materials properties of polymers. Besides interesting optical properties, caused by the anisotropy in these materials, these polymers often also exhibit other good materials properties, like a good impact resistance and high tensile strength. Two important types of liquid crystalline polymer can be distinguished. In main chain liquid crystalline polymers (MC-LCP), the heads and tails of the monomers are connected to each other. In side chain liquid crystalline polymers (SC-LCP), the polymerisation is on one end of the monomer, which results in a polymer backbone to which the mesogenic groups are attached.
Main chain liquid crystalline polymer (MCLCP)
Side chain liquid crystalline polymer (SCLCP)
Figure 26 Schematic drawings of main chain and sidechain liquid crystalline polymers
6.8.1 Main chain liquid crystalline polymers. These are often synthesized by a condensation reaction. Often ester or amide connections between the monomeric units are present. For example:
H2N NH2 + HOOC COOH H3C CH3
NH N HO OH H O O n Bisphenol-A Kevlar, Twaron
A commercial example are the aramides (Kevlar, Twaron) made by a reaction of 1,4- diaminobenzene with 1,4-benzenedicarboxylic acid in which besides the polymer also water is formed. Because the mesogenic groups like to be parallel by π-stacking and hydrogen bonding, the main chains will also be ordered parallel. This results in materials with a high tensile strength and a high glass transition temperature Tg. For a number of applications this is important. Sometimes the Tg is too high for the processing of these materials. In those cases, introducing - 89 - asymmetric or non-linear units in the polymer chain, like bisphenol-A or similar compounds can lower Tg. By simultaneously spinning these materials and stretching the fibers, the polymer chains become very well parallel ordered and strongly anisotropic fibers are formed that have very favorable mechanical properties. In non-liquid crystalline materials the alignment of the main chains is more difficult, because it does not occur spontaneously and consequently it is more difficult to turn them into strong materials. When the rigid aromatic parts are connected by flexible spacers, materials are formed that have lower phase transition temperatures. They exhibit a strong odd - even effect in their properties as a function of spacer parity. Twin (or dimer) liquid crystals (discussed earlier) can thus be seen as (simple) model compounds for these polymers.
HO OH + HOOC (CH2)m COOH
O O (CH2)m C O O n
These MC-LCP materials can be applied as strong and tough fibres. They are also applied as so- called "alignment layers" for liquid crystalline displays. A thin film of material is applied to a surface and rubbed in one direction. This causes the polymer chains to also orient in one direction and this becomes a good attachment place for the monomeric liquid crystals in the bulk that preferentially orient themselves with respect to the substrate.
6.8.2 Side chain liquid crystalline polymers. These are usually prepared by providing a monomeric unit with a polymerisable group and subsequently perform the polymerisation reaction. An alternative is to attach the mesogenic groups to an existing polymer chain. Polymerisable groups that are often used are acrylates, methacrylates, epoxides and alkenes. The group R in the examples shown here can be rather arbitrary, in this case a mesogenic group, usually containing an alkyl spacer in order to separate the polymer backbone from the mesogenic groups.
O O R R O R n O R n O O O R O O R R n O R n
- 90 -
One method to modify existing polymers with liquid crystalline side-chains is the method used to prepare polysiloxanes. A Si-H group is reacted with an alkene under the influence of UV or a platinum catalyst, yielding polymers with direct Si-C bonds.
H C Si H 3 R H3C Si + R O O H C Si H 3 H3C Si R O O n n
An example of a side chain liquid crystalline polymer is the following polyacrylate, in which the different structural elements are represented.
n N C N O O O H
These elements are the polymeric backbone, the spacer and the mesogenic group which all have a certain influence on the properties of the material. The stiffness of the backbone is particularly important for the glass transition temperature (Tg). A rigid backbone will lead to materials with a high T g, sometimes so high that their liquid properties are not important for the applications at room temperature, although their ordering properties might be. Examples of polymers with rigid backbones are the following polymers that were prepared by a so-called alternating co-polymerisation.
n O O + O N O N R R Polymers with a flexible backbone are for example polysiloxanes and polyacrylates, which can have Tg's below 0°C. The length of the flexible spacer is of utmost importance for the liquid crystalline properties of the polymers. When the spacer is long enough, the mesogenic groups can move and order with the other mesogenic groups almost independently from the backbone. When the decoupling is large enough, usually with more than 6 flexible CH2-units, the best possible ordering of the mesogenic groups is found. It is for example often found that with a certain backbone and
- 91 - mesogenic group a short spacer gives nematic phases, while a longer spacer gives smectic phases. Of course the mesogenic group is important for the liquid crystalline properties. Factors like dipole moment, length-diameter ratio, presence of chirality, etc, strongly determine the liquid crystalline properties of the polymers, just like in the monomers. The polymerisation reaction yields polymers with a different degree of polymerisation. Often polymers are formed with an average length of tens to many hundreds of monomeric units. For a polymerisation degree of less than about 15 there is a strong dependence of the properties on the polymerisation degree, for higher degrees the properties are less dependent on the polymer length. Polymers can form the same liquid crystalline phases as low molecular weight liquid crystals. Because the different parts of these polymers have a different polarity, there is a tendency for microphase separation, which means that parts that have the same polarity have the tendency to seek each other and to aggregate. In smectic layers, we see therefore often that the polymeric backbone, the spacers and the mesogenic groups lie in different layers. For polymers there is often an ordering found that can be called interdigitated (double comb configuration). The degree of interdigitation can vary strongly, depending on for example dipole moment of the rigid cores, length of spacers and alkyl end groups and cross sectional area of the cores and alkyl groups.
Figure 27 Different degrees of interdigitation in the smectic layers of polymers.
Side chain liquid crystalline polymers have an application in optical layers. A recent interesting example is the application as a polarizing filter for LCD-screens.
6.9 An example of research on liquid crystalline polymers
- 92 - 6.9.1 Liquid crystalline polymers with a smectic-B ordering. In this phase, the mesogenic groups are present in layers, in which there is also a hexagonal ordering within the layers. In a sample domains are present that have a random orientation in space. When such a sample is placed in an X-ray beam, circular reflection patterns are formed.
Figure 28 Circular X-ray reflection patterns formed because of a SmB-ordering.
When the intensity of the reflection of the following polymer is plotted as is a function of the angle, a profile is obtained that given in the following figure:
(CH2)9 O OCH3
n O O O
Figure 29 Rontgen reflection intensity plotted as a function of the angle.
Reflections are found for angles 2ϕ of 2.7°, 5.4° and 19°. This means that the layer thickness of the smectic layer is about 32A and the mutual distance between the mesogenic groups is about 4.5Å. When the ordering becomes smectic-A, a similar pattern is found, only the peak at 19° (4.5 Å) becomes much broader (diffuse). This is because the ordering within the layer decreases, but the average distance between the mesogenic groups remains the same (~4.5 Å).
When instead of a non-oriented sample a sample is placed in the beam that is oriented, for example by pulling a thread, we see the following pattern. Now we see that the ordering of the layers is perpendicular to each other, which is in agreement with a smectic-B ordering.
Figure 30 Pattern for an oriented smectic-B sample
- 93 -
6.9.2 A multilayer on a surface. When a thin multilayer of the above mentioned polymer is spread on a silicon surface, and this surface is studied with X-ray reflectivity under low angles, the following pattern can be obtained:
Figure 31 X-ray reflection pattern for the non- Figure 32 Electron-density profile calculated ordered film (lower curve) and for the film with a from the reflection curve and the layer ordering (upper curve). corresponding model for the ordering of the polymer.
We can distinguish two layer systems: one layer of organic material between the silicon and the air with a certain thickness that again consists of a number of separate thinner layers. For both constructive reflection occurs when sin
Question: Calculate the polymer layer thickness and the total film thickness from the reflection peaks (A = 1.54 Å)
6.10 Amphotropics (surfactants with thermotropic liquid crystalline properties).
In this chapter and also in the chapter on surfactants we have seen that individual molecules of compounds with special structural characteristics can form aggregates. We can distinguish between thermotropic liquid crystals that can form nematic, smectic or discotic phases and lyotropic liquid crystals that can form micellar, lamellar, hexagonal, etc, phases, depending amongst others on the
- 94 - amount of water present. Although in both cases aggregation occurs, there are large differences in both the structures of the compounds and in the types of aggregates they form. The phases that resemble each other closest are the lamellar phase for lyotropic and smectic phase for thermotropic liquid crystals.
thermotrope lyotrope (amphiphile)
amphotrope
Figure 33 Schematic representation of the structural elements of an amphotropic compound.
However, it is very well possible to combine the structural elements of thermotropic and lyotropic liquid crystals. The resulting molecules that have a polar head group and a rigid unit in their apolar tail often have both lyotropic and thermotropic liquid crystalline properties. These molecules are also called "amphotropes".
Examples of some amphotropes are given below:
H2n+1Cn N O–(CH2)m N Br I
NC O–(CH2)n N Br II
NC O–(CH2)n III N Br
NC O–(CH2)n
O IV HO O O O O O O
The last compound (IV) is a thermotropic liquid crystal (SmA) from -28°C to 113°C. Furthermore, this compound shows, depending on pH and concentration, several lyotropic liquid crystalline
- 95 - phases, like micelle or vesicle (see also the paper at the end of the chapter on surfactants). Compound II forms micelles in water at low concentration and lamellar phases at high concentrations. No phases are found that have a high curvature like hexagonal or cubic phases. It is remarkable that usually a smectic-A phase is found for the thermotropic liquid crystalline phases of these compounds. If parts with the same polarity (and other properties) prefer to be close together, the following types of ordering are possible:
A B
Figure 34 Possible orderings in layered structures of amphotropes.
These orderings have indeed been found for both the thermotropic and lyotropic phases of compounds I. When the end group length and the spacer length are varied, the compounds with long spacers and short tails will have a preference for ordering A. Those with short spacers and long tails prefer ordering B. In both cases, the space will be filled most efficiently. Compounds of this type prefer to form bilayer or lamellar structures (vesicles) in water, even when only one tail is attached to the head group. When the total length of the tail is relatively short or when the end group is short or absent, the amphotropes rather form micelles (for example compound II). Amphotropes with two tails (in which one or two tails contain a rigid unit; e.g. III) prefer to form bilayer structures just like other two-tailed amphiphiles. In bilayer structures of amphotropes with aromatic groups a special phenomenon occurs. The aromatic groups are close together due to stacking; this caused the π-electrons to interact with one another. Therefore, a splitting of the excited state occurs in two levels of which one is allowed. For parallel transition dipole moments, this leads to a shift of the absorption wavelength in the UV -Vis spectrum.
2 N −1 µ 2 ∆ν= (1− 3cos 2 α) hc N r 3
- 96 - The shift depends on the magnitude of the transition dipole moment µ, the distance between the molecules r, the number of molecules N in a certain stack and the angle α between the transition dipole moments (usually the long axis of the chromophore) and the plane of the layer. This effect is known as "exciton coupling". It follows from the formula that for parallel chromophores a blue shift occurs; this type of aggregate is called H-aggregate. When α becomes smaller than 55° a red shift occurs and the aggregates are called J-aggregates.
E+ Em S1 D
E-
So 90° 54.7° 0°
monomer H-aggregate monomer J-aggregate α
Figure 35 Schematic representation of the exciton splitting of the excited state due to aggregation. Depending on the orientation of the transition dipole moments a blue- or a red shift in the absorption spectrum can occur.
An example is the following two-tailed amphotrope that forms vesicles in water:
NC O–C12H24 N Br
4-CN H25C12
40.8 Å
Figure 36 UV absorption spectrum of compound 4- Figure 37 Schematic representation of the CN as a monomer in ethanol (dotted line) and as bilayers of 4-CN vesicles in water (solid line) - 97 - These shifts can amount to several tens of nm. These shifts can be used to follow changes in the bilayer membranes, like solubilisation by micelle-forming surfactants and surfactant exchange. This kind of modification of surfactants offers the possibility to study membrane processes with spectroscopic techniques. In biochemistry for example amphiphiles containing fluorescent chromophores can be used to study membrane mobility. Besides calamitic amphotropes, there are also discotic amphotropes. Interesting examples are the following polyhydroxy compounds that can exhibit thermotropic and lyotropic liquid crystalline behaviour. The phase behaviour strongly depends on the number of alkyl tails that is attached to the central ring system. With one alkyl tail, a smectic layer system is found. With more alkyl tails, the behaviour goes to discotic phases. An increase of the number of alkyl tails decreases the number of molecules that is required to form a disk. This behaviour is also found for sugar molecules containing more or less alkyl tails.
Figure 38 Phase behaviour of amphotropic polyhydroxy compounds as function of the number of alkyl tails.
6.11 Literature
1. H. Stegemeyer, Liquid Crystals, Springer Verlag, 1993. 2. G. Vertogen en W. H. de Jeu, Thermotropic Liquid Crystals, Fundamentals, Springer Verlag, 1987 3. P. J. Collings, Liquid Crystals. Nature's Delicate Phase of Matter, Adam Hilger, Bristol, 1990 4. D. Demus et. al. Handbook of Liquid Crystals, 1998 5. Website: http://plc.cwru.edu/
- 98 - 6.12 Intermezzo: SPECIAL PLASTICS
Instead of using a standard telephone, we will talk to each another via a plastic screen that we put on our head like glasses. Instead of watching the offered television programs we will watch news, sports and movies at the time we want to. Personal computers will no longer be big and placed in a study, but flat and be present in different parts of the house, sometimes hidden, sometimes as design objects. Everywhere in the house there are possibilities to obtain information, for example a digital paper at the breakfast table or a visual make up advice in the bathroom projected directly on your image on the mirror.
Monomers are the building blocks of polymer structures and their precise positioning can be compared to building with bricks. Like architects, chemists are trying to build monumental works of art. With a sophisticated combination of monomer design and polymerisation processes, they are able to produce smart plastics with special optical properties. Whether futuristic image given above will become real is not clear yet, but it is certain that in the near future smart plastic materials will play an important role. Investigators from Philips Research and the section Polymer Chemistry and Technology of the Technical University of Eindhoven have taken up the challenge to design plastics for the new information society.
Manipulation with light. Many processes in information and communication technology are based on the processing and manipulation of light. Sending light pulses through glass fibbers (and maybe in the near future at a local level through plastic fibbers) offers a very efficient system for digital data transport. The possibilities for optically reading and writing data with laser technology is rapidly expanding. Displaying data with flat screens is currently optimised. Ultimately every pixel will be controlled through a minuscule light switch or transmitter. It is expected that optical data processors can make computers several orders of magnitude faster. The development of new specialty polymers leads to important innovations and improvements in optical systems. In these special polymers the structures of the molecules is carefully designed and the molecules are carefully positioned, partially through a self-assembling process. It is this combination of molecular design and molecular positioning that determines the interaction with light: how light passes through the material and what happens with the light. Control of the molecular organisation can lead to functional plastic products; products that are important in for example information displays and other devices. The complexity of the molecular organisation in a polymer is also controlled by the polymerisation process. Five polymerisation processes (see frames 1 to 5) show how the ultimate properties of a polymer are determined by the choice of monomer and the method of polymerisation.
- 99 - Disorderly network with surface structure. An often-used polymerization method is photoreplication. A liquid mixture of a bifunctional monomer and a photo initiator are applied between a template and a binding surface. After polymerization and removal of the template the surface of the hard plastic is a negative of the template. The obtained material is very suited for optical applications like complex lenses. An application is the fabrication of lenses for projection television. The complex lens system consists of a fresnellens and a series of cylinder lenses at the side of the image forming optics and at the viewer side it consists of an array of light refracting lenticular lenses that are responsible for the precise light distribution an viewing angle.
Control at the nanometer level. The five examples show that by a combination of molecular self organisation (in this case by liquid crystallinity) and imposed organisation (photo induced diffusion, photo replication, etc), structures can be obtained whose dimensions can be controlled from large to very small. In the first example surface structures are realised that are still visible with the naked eye, while in the last example a helix-like aggregate is created with a typical dimension of 300 molecules in a second much larger superstructure. Even within a single turn of the helix, the positioning of the molecules is further structured. Technologies are being developed that will make this type of structure control possible on the nanometer level. Together with a carefully designed molecular structure it are these dimensions that are required to manipulate light in devices like refracting polarises, plastic wave guides and abnormal lens shapes. This type of structure control will probably also be used to develop new light switches and recording media.
- 100 -
Ordered molecular structure. When the monomers from the previous example can be ordered before the polymerisation process, the resulting network is strong enough to retain the ordering permanently. A way to achieve this is by making the monomers liquid crystalline. In most cases, the molecular ordering in the polymer network will be the same as in the monomers. This means that plastic sheets can be made consisting of ordered calamitic mesogens. Because the molecules in these films are aligned, the interaction with Network with a helical structure. light depends on the angle the light makes with the When a chiral centre is present in the liquid crystalline ordering direction of the network. If the refractive index is monomeric unit, new possibilities become available. When measured with polarised light parallel to the average the monomer mixture contains only one of both direction of the long axes of the molecules, one usually enantiomers, the long axis of the molecules will no longer finds a much larger value than when measured be parallel in the nematic phase, but constitute a helix. perpendicular to this direction. This so-called optical When monomers of this type are polymerised, a polymer anisotropy can be used in many optical devices. network is formed in which the helical ordering is An example of an application is a polarisation splitting preserved. The pitch of the helix (the distance over which a element for a projection display, based on image formation complete rotation of 3600 takes place) is 150 nm, which with a liquid crystalline reflective light switch. corresponds with about 300 molecules. This so-called polarising beam splitter (PBS) consists of Now the interesting thing is that these chiral molecules can two prisms, coupled by a film of the oriented polymer. The be mixed with non-chiral liquid crystals, so it is possible to refractive index of the glass of the prisms is the same as choose a pitch of the helix between 150 nm and infinite. the extraordinary index of the plastic layer. This means With a large pitch, so that only a part of the 3600 rotation is that an incoming beam of polarised light, with its vibration completed over the thickness of the thin film, the films can direction parallel to the long axis of the molecules in the be used as compensation films for "supertwisted nematic" polymer network, does not optically observe a transition (STN) displays, often used in laptop computers. It is also between the glass and the polymer and is not refracted. interesting to adjust the pitch of the helix to a value that Then the beam arrives at the electro-optical switch, based corresponds to a wavelength in the visible spectrum. on a liquid crystal that does or does not change the Because of Bragg reflection, light with a wavelength that polarisation direction of the light. When the polarisation equals the pitch of the helix times the refractive index will direction is not changed, the reflected beam passes the be reflected, while light with other wavelengths will pass PBS again without refraction and continues in the direction the film. This gives these liquid crystalline materials vivid of the source. If the polarisation direction is changed, the and bright colours. These materials are used in some light will encounter a lower refractive index at the types of wine thermometers. boundary between glass and polymer in the PBS and will A complication is that only half the amount of light is be completely refracted, for example in the direction of a reflected, because only light with a circular polarisation projection screen. Because a liquid crystal panel consists direction that follows the helix is reflected. This means that of many pixels, this is a very elegant way to produce an not only separation of colours occurs, but also separation image. of polarisation. The resulting polarisation selective colour filters can be used for reflective liquid crystal displays, for energy-effective transmissive liquid crystal displays and for colour separation in liquid crystal projection television.
- 101 -
Helix gradient with a discontinuous rotation. In the previous example, a network was created with a Helical network with a gradient in the molecular pitch. helical ordering and gradual pitch change. In the previous example, it was shown how the pitch of the If the rotation within a turn of the spiral is locally faster molecular helix can be adjusted accurately by mixing a chiral in a certain direction, a very special type of staircase is monomer with a non-chiral liquid crystalline monomer. An formed. This is depicted above. Relatively more additional element of self-organisation can be added to this mesogenic units have their long axes in the plane of system. It is possible to impose a superstructure on the helix, in the paper than perpendicular to it. which the pitch of the helix gradually changes upon going from The method that has been devised for such an one surface of the film to the other. This is done by a process ordering is based on a photo-initiator with mesogenic based on diffusion. During the polymerisation, the chiral properties that orders in the helix. This molecule monomer is forced to one side of the film, whereas the non- mainly absorbs UV-light with a polarisation direction chiral monomer is forced to the other side. This will result in a parallel to the long axis. This means that when the film gradual change in the pitch. For this process, a different of monomers is irradiated with polarised UV-light, the reactivity of the monomers is necessary: the faster reacting initiators that are parallel to the polarisation direction monomer is built in first at the top of the film. This is because the are preferentially activated and start the polymerisation intensity of the UV-light that is used to induce the polymerisation reaction. This will locally lead to faster reaction of the reaction is highest. This causes the less reactive monomer to chiral molecules and diffusion of the less-reactive non- diffuse deeper in the film. This leads to the concentration chiral molecules. These will react later in the process gradient and thus the gradient in helix pitch. and give rise to the less twisted parts of the helix. The structure that is formed in fixed form in the film is presented This film has the property that it acts as an optical schematically. In this case, the topside is rich in chiral monomer, retarder and it will generate linearly polarised light from which was the most reactive. The pitch is here the smallest. The circularly polarised light. bottom-side is richer in non-chiral monomer and has a larger When this ordering is included in a film with a gradual pitch. The change in pitch is more or less linear. change in pitch a film is obtained that transmits one Special about this film is that now light from the whole visible part of the light that falls on it as linearly polarised light spectrum can be reflected, because any wavelength and reflects the other half as circularly polarised light. corresponds to a part of the helix with a fitting pitch. The This type of layer is not yet applied in devices, but it polarisation separating ability of this film is not only present for a certainly has potential. special wavelength but for all the visible light. Consequently, white light is separated in left-handed and tight-handed circularly These examples are illustrations of molecular polarised light by this film. One part is reflected and the other architectures with increasing complexity, in which in part transmitted. the last example three molecular structures of different However, for most display applications linearly polarised light is nature and dimensions are superimposed. It is an required. Conventional polarisation filters transmit less than 50% illustration of state of the art control over molecular light, because half is absorbed. Linearly polarised light can also ordering that leads to modern materials with functional easily be obtained in high yield from circularly polarised light with properties a plastic sheet of polymer with a uniaxial orientation, obtained for example as described in example 2. This film has a certain thickness, which makes it act as a quarter wave plate. The advantage of this method is that now polarised light is produced without absorption (i.e. loss) of light with the wrong polarisation direction. Light with the wrong polarisation direction can be reflected in the optical system behind the display, which produces unpolarised light. This can again be split by the polarisation filter, etc. This process is repeated until all the light has passed the polarisation filter or is lost by absorption. The result is a display in which the brightness is increased with 70%, which can-be translated in lower energy consumption and thus a longer lifetime of the battery in portable applications.
- 102 -