Suspensions
Susan A. Barker
Chapter contents
Solid particle–liquid vehicle interactions
The ‘electrical double layer’ theory
Factors affecting the electrical double layer
Particle movement in suspensions
Controlling particulate movement in suspensions
What is the desired sedimentation pattern?
Dispersibility issues – surface wetting
General suspension formulation considerations
Key points
Introduction
Suspensions are probably one of the most challenging pharmaceutical formulations that students and formulators are likely to encounter. Many of the issues relating to solution formulation development apply equally to suspension formulations, but there are several additional considerations relating to the solid component of the system and the interface between the solid and the liquid components. The intention of this chapter is to present these issues and discuss them in context, so that the reader will have a fundamental understanding of the science behind suspension formulation as well as the more patient-focused aspects.
The most important consideration in suspension formulation development is the interaction between the solid particles and the liquid vehicle, so this aspect will be considered in detail first in this chapter. Secondly, other considerations relating to the solid component will be discussed and finally aesthetic, patient-focused and practical matters will be reviewed.
Definition of a suspension
It is appropriate here to review the definition of a solution before going on to consider the definition of a suspension. Solutions are discussed in detail in Chapters 2 and 3. They can be formed from various combinations of phases. However, in pharmacy, the word ‘solution’, without further description, is generally understood to refer to a liquid system. A solution is a one-phase system where at equilibrium all the ingredients are dispersed evenly at a molecular level. Solutions are therefore optically clear as there are no solid particles remaining to disperse the light. In the context of a pharmaceutical formulation, the drug (solute) is added as a solid to the vehicle (solvent, most commonly water) and dissolves completely to give the formulation (a solution). Other formulation components, such as preservatives, buffers, flavours, etc., are added as required.
A pharmaceutical suspension is also a liquid system. However, in this case, the solid material (usually the drug) does not dissolve in the vehicle to any appreciable extent, but remains as solid particles which are distributed throughout the vehicle. Technically, the term suspension describes a dispersion of a solid material (the dispersed phase) in a liquid (the continuous phase) without reference to the particle size of the solid material. However, the particle size of the solid material can affect both its physical and chemical behaviour, so a distinction is usually made between a colloid or colloidal suspension with a particle size range of up to about 1 µm (see Chapter 5), and a ‘coarse dispersion’ with larger particles. Unfortunately, pharmaceutical suspensions fall across the borderline between colloidal and coarse dispersions, with solid particles generally in the range of 0.1 to 10 µm. Suspensions are not optically clear and will appear cloudy unless the size of the particles is within the colloidal range.
In the ideal suspension, the particles of the solid material are mono-dispersed spheres and are evenly suspended in three dimensions throughout the vehicle and remain so even after prolonged periods of time. Here, every dose from the suspension will contain the same amount of drug and will give the same clinical effect to the patient; such reproducibility is absolutely vital for all formulations, but is difficult to achieve with suspensions.
Solid particle–liquid vehicle interactions
Solid particle–liquid vehicle interactions determine the behaviour of suspensions and hence it is vital to have a working knowledge of them. Water is by far the most common vehicle used for pharmaceutical suspensions, due to its lack of toxicity, hence it is the interaction between water and the particles which is important. The behaviour of an individual particle will be discussed first, then the interactions between multiple particles. These issues are discussed in Chapter 5 in the context of dispersed systems in general and Chapter 4 discusses solid–liquid interfaces. The emphasis in this chapter is on pharmaceutical suspensions and the importance of particle charge and particle-particle interactions in successful suspension formulation.
The ‘electrical double layer’ theory
A solid material will not dissolve in a liquid vehicle unless it has some chemical similarity. Thus, drugs which are hydrophobic will not dissolve very well in water. Many modern drugs show very low aqueous solubility, because they are designed to fit into hydrophobic biological receptors, so although they may be very efficient once at their site(s) of action, they present a real challenge to the formulator, and many may be developed as suspension formulations.
An apparently odd characteristic of such hydrophobic drugs is that once dispersed as solid particles in an aqueous environment, they will acquire a charge. This is counter-intuitive, as hydrophobic materials will generally repel water; if they could ionize, they would be expected to show a reasonable degree of aqueous solubility and would not remain as solid particles within the aqueous dispersion. However, the charge produced is not as a result of ionization of the drug, rather it is due to the ionization of water:
(26.1)
The liberated protons will then be solvated by intact water molecules to form hydronium ions (H3O+) and larger structures with more water molecules. One result of this is that they are generally less mobile than the hydroxide ions (OH−) produced from the initial ionization reaction shown in Equation 26.1. Some of the hydroxide ions will then collect on solid surfaces within the aqueous dispersion and give rise to an apparent negative charge on these surfaces. Overall, within the system, electrical neutrality must be maintained, so there will be a gradation of charge from a high negative charge on the surface of the particle down to no charge (overall neutrality) in the bulk vehicle. This gradation of charge occurs in two stages, giving rise to two ‘layers’ of charge surrounding the particle, hence the term electrical double layer.
Figure 26.1 illustrates a single solid particle in water, showing a negatively charged surface arising from the OH− ions. The innermost layer or ‘halo’ around the particle has a predominance of positively-charged ions. The outer layer or halo also has a predominance of positively-charged ions, but to a lesser extent than the inner layer. Finally the bulk vehicle has no overall net charge. Charges in the inner layer are held tightly to the particle and this is therefore known as the fixed layer, whereas charges within the outer layer are more mobile and can move away from the solid surface and hence this is denoted the ‘diffuse layer’. The two layers, fixed and diffuse, are also known as the Stern and Gouy-Chapman layers, respectively.
The rigidity with which the charges are held to the particle affects the intensity of charge at any point between the charged surface and the bulk external liquid. Throughout the fixed layer, there is a linear decrease of overall charge from the particle surface (high) to the edge of the fixed layer (lower). From this point until the edge of the diffuse layer, i.e. the beginning of the bulk liquid, there is an exponential decrease of charge. This phenomenon is discussed in Chapter 5.
Factors affecting the electrical double layer
The addition of formulation excipients can change the behaviour of a solid particle in a suspension, by affecting either the fixed layer or the diffuse layer, or both. Materials which can ionize, for example sodium chloride, will increase the amount of mobile charges available in the system. At low to medium concentrations, for example 0.01 M, such charges are generally located only within the diffuse layer, as shown in Figure 26.2a, and therefore will not affect the surface potential ψ0 or the Stern potential ψδ (see Chapter 5 for explanations). The increase in the number of individual charges within the diffuse layer will result in easier neutralization of the remaining charge from the particle (ψδ) and hence will lead to a thinning of the diffuse layer. Mathematically, this is because the distance over which ψδ becomes ψδ/e, i.e. 1/κ (the ‘Debye-Hückel length’, Chapter 5), is smaller. It should be evident from the above discussion, that an increased concentration of the additional charges would be expected to lead to a greater reduction of 1/κ. In fact, the relationship is a square root one, in that 1/κ is inversely proportional to the square root of the ionic strength of the medium. This is the same as saying that κ (the ‘Debye-Hückel length parameter’, Chapter 5) is directly proportional to the square root of the ionic strength of the medium. Care must be taken, therefore, to consider the chemistry of dissolved ionic materials: the ionic strength of a calcium chloride (CaCl2) solution is higher than that of a sodium chloride (NaCl) solution of the same molar concentration. The effects of these two solutions on the electrical double layer will consequently be different.
Higher concentrations of ionic materials, for example 0.1 M, will not just result in a greater effect on the diffuse layer, but some of the charges will migrate through into the fixed layer and become adsorbed onto the surface of the particle itself, shown in Figure 26.2b. In this case, the charge on the particle surface will decrease, which will have the automatic effect of lowering the Stern potential and a secondary effect of reducing the zeta potential, as the charge reduction across the diffuse layer will begin at a different value.
Surfactants added into the system at concentrations below their critical micelle concentration (cmc) will localize on the surface of the particles, as shown in Figure 26.2c. At concentrations above the cmc, surfactant micelles will be formed, with a central hydrophobic core into which the hydrophobic drug may dissolve (Chapters 5 and 24). To avoid this, it is necessary to ensure that the surfactant concentration remains below the cmc. The addition of the surfactant to the surface of the particle will change the particulate charge, certainly in its magnitude but possibly also in its sign. The effects will be dependent on the chemistry of the surfactant itself, i.e. whether it is cationic, anionic or non-ionic. Such charge modification will affect the fixed layer directly, rather than the diffuse layer, and a variation in the surface charge will naturally lead to an alteration of the Stern potential. As described above, this will then have a secondary effect on the zeta potential, as the charge decay across the diffuse layer will start from a different value. Ionic surfactants can also release ionic components into the medium (e.g. Na+ ions from sodium lauryl sulphate), which will then have their own direct effects on the diffuse layer. The overall effect of addition of surfactants will need to be considered on a case-by-case basis, based on their chemistry.
The DLVO theory
Pharmaceutical suspensions are not composed of a single particle of drug suspended in a liquid medium, but rather of multiple particles; this leads to multiple particulate interactions. These interactions can, to some extent, be thought of as the interactions of the diffuse layers around individual particles and hence the electrical double layer provides the basis for understanding inter-particulate interactions. The DLVO theory describes these interactions.
The DLVO theory (Chapter 5 provides more detail) is concerned with predicting the stability of lyophobic (‘solvent-hating’) colloids and is relevant here because of the particle size of pharmaceutical suspensions. Essentially, it calculates the energies of attraction and repulsion between similar particles and predicts the overall energy of interaction. From this, deductions can be made as to the likely behaviour of the suspension, e.g. whether particles coalesce and settle, or remain evenly dispersed throughout the medium. This is arguably the most important question in pharmaceutical suspension formulation development, as a fundamental specification for such a formulation is dose reproducibility, which is most easily achieved from a system which remains well dispersed under all conditions.
To calculate the total energy of interaction, VT, between two particles, the values of VA and VR are summed, as shown in Equation 26.2. VR is the energy of electrical repulsion and by convention this carries a positive sign. VA is the energy of Van der Waals attractions and by convention is given a negative sign.
(26.2)
Figure 26.3 shows the values of VA, VR and VT for two similar particles suspended in a medium and interacting. Further detailed relationships involving VA and VR can be found in Chapter 5. It is important to note that the VA and VR curves shown in Figure 26.3 are not mirror images of each other.
The easiest way to consider what happens when two particles interact is to remember that the VT line gives the overall energy of interaction and that this will change depending on the distance between the two particles. There are three important zones, or values of VT, in the DLVO diagram: the primary minimum, the secondary minimum and the primary maximum, and the behaviour of the suspension will be dependent on which zone the particles are in. It must also be remembered that all particles will have some thermal energy and will show some movement, whether caused by Brownian motion, the effects of gravity or by external agitation.
The primary minimum
The ‘primary minimum’ zone is described as a ‘minimum’ because the total energy is calculated to be below zero (remember that repulsive energy is described as positive and attractive energy as negative). It is described as ‘primary’ because it is the largest negative deviation from zero. Particles in the primary minimum zone show a higher energy of attraction than repulsion and are therefore likely to move closer together. Imagine two particles are just far enough apart that the energy of attraction balances out the energy of repulsion, so that the overall energy of interaction is zero. Any movement of the particles which brings them closer together will result in an overall mathematical decrease in VT, i.e. VT is now attractive and the particles will continue to move closer together. As they do so, the strength of the overall attractive forces increases, moving the particles still closer together, resulting in a further increase in the attractive forces, and so on. The kinetic energy that the particles have (= kT, where k is the Boltzmann constant and T the temperature in Kelvin) is not high enough to overcome the attractive energy, VT and therefore the particles will eventually aggregate irreversibly. Particles will initially show ‘flocculation’, whereby the individual particles are loosely attracted to each other, but still act independently; subsequently they will demonstrate ‘coagulation’ where particles will collide and form larger particles. Such behaviour is undesirable for pharmaceutical suspensions as it will have serious negative effects on the reproducibility of dosing from the system. These changes are illustrated in panel A of Figure 26.4.
The primary maximum
The naming of the ‘primary maximum’ zone follows the same conventions as for the primary minimum. The primary maximum zone is described as a ‘maximum’ because the total energy is calculated to be above zero (using the convention of repulsive energy being positive and attractive energy being negative). It is described as ‘primary’ because it is the largest positive deviation from zero. Particles in the primary maximum zone show a higher energy of repulsion than attraction and are therefore likely to remain separate or ‘deflocculated’. This is illustrated in panel B of Figure 26.4. At first sight, this would appear to be a good formulation strategy for pharmaceutical suspensions, as if the particles can be forced into the primary maximum zone then they should remain independent and hence dosing would be expected to be reproducible. This is true when the kinetic energy of the particles is less than VT and they are, if anything, more likely to move away from each other, which will have the effect of decreasing the magnitude of VT but maintaining an overall repulsive effect. However, if the kinetic energy of the particles is high enough, for example if the temperature is increased, then this can overcome the energy barrier imposed by VT with the result that the particles can then move closer together. In this case, VT will initially decrease but remain repulsive, so the particles will still exist as independent entities. However, the magnitude of the difference between VT and the particles’ kinetic energy is now greater and therefore they are likely to move even closer together. At some point, the particles will be sufficiently close so that the overall energy of interaction becomes negative, i.e. it is now predominantly attractive, and the particles enter the primary minimum zone with the consequences described above. In summary, therefore, formulating pharmaceutical suspensions so that the particles are in the primary maximum zone can be considered to be risky.
The secondary minimum
Panel C of Figure 26.4, shows the behaviour within the secondary minimum zone. As its name suggests, the ‘secondary minimum’ gives rise to an overall attractive energy of interaction between particles, but of a lower magnitude than that seen in the primary minimum. The particles here show an overall limited attraction to each other and behave as ‘floccules’, loose aggregates of individual particles. Depending on the kinetic energy of the particles, their behaviour will vary slightly. If the kinetic energy is less than the VT, then the particles will move closer together under the influence of the VT, but will not collide and coalesce as the VT is still relatively weak. As the particles move further together, the attractive forces will reach their highest point (although not as strong as in the primary minimum zone) then decrease and overall VT becomes weakly repulsive, which will have the effect of forcing the particles apart. At this stage, VT once again dominates over the kinetic energy and the particles will be attracted weakly to each other. In essence, the particles are maintained in their flocculated state, that is they still exist as individual particles, but are loosely grouped together in floccules. If, however, the kinetic energy of the particles is greater than the VT, then the particles will be able to move further apart. As they do this, the overall VT will become less attractive and ultimately will become, to all practical purposes, zero. In this case, the particles will behave independently, will not flocculate and will not coalesce. In either case (kinetic energy greater to or less than VT), coalescence and coagulation of particles is minimal, and hence this is usually the desired strategy for developing pharmaceutical suspensions.
Controlling particulate behaviour in suspensions
From the discussion above, it can be seen that the behaviour of particles in suspension is complex, even when only two individual interacting particles are considered; the behaviour ultimately being dependent on the relative contribution of the repulsive and attractive energies at any separation distance. Examining the equations that govern these two aspects (Eqns 5.24 and 5.25, respectively, repeated here for convenience) can give some clues as to which factors can be manipulated during suspension formulation to alter the behaviour of the particles and which factors cannot be altered.
(5.24)
(5.25)
A, the Hamaker constant (Eqn 5.25).
This factor is constant for each combination of particle and medium. As the particulate material within a pharmaceutical suspension is the drug, then the formulator has no opportunity to change the physicochemical nature of the particles. Although theoretically the medium may be altered, which will then change the Hamaker constant, most pharmaceutical suspensions, certainly those intended for oral drug delivery, are aqueous. Hence, the two components in the suspension which contribute to the Hamaker constant (the drug and water) are fixed, and this factor is, in effect, non-modifiable.
ε, the permittivity of the medium (Eqn 5.24).
The permittivity of the medium is related to its polarity, so therefore varying the medium will have a direct effect on the repulsive energy between particles in the system. Water is the most common medium for pharmaceutical suspensions and addition of dissolved solids, such as electrolytes, to water will have a relatively minor effect on its permittivity, compared to the effect of changing from water to, for example, oil. Overall, therefore, for the purposes of pharmaceutical suspensions, the permittivity can be considered to be that of water and will have limited variability.
H, distance between particles (Eqns 5.24 and 5.25).
The distance between particles can be considered to be both a cause and effect of the balance between the attractive and repulsive energies of the system, as discussed in the previous section: particles very far apart will have very limited interaction and particles located close to each other will be attracted or repelled depending on exactly how far apart they are, and may move in response to the dominant VT
