1.4 Nanofibre characterisation

1.4.1 Electron microscopy

Microscopy allows us to magnify a sample and to inspect features which are invisible to the naked eye. While traditional light microscopy can be of use in the preliminary characterisation of nanofibres, allowing us to confirm the successful formation of fibres (or not), their nanoscale diameters (particularly when below 200 nm) mean that a detailed inspection of the surface morphology or diameter quantification is not possible using this technique. A typical light microscopy image of some fibres of diameter ca. 1 µm is given in Figure 1.1(c). The presence of fibres is clear, but even with these relatively large fibres their surfaces cannot clearly be resolved.

Light microscopy allows us to magnify samples by up to 1000-fold (but 100-fold is much more common); to gain more insight, electron microscopy is required. This permits magnification of 100,000-fold to 200,000-fold and nm-scale resolution. Both scanning and transition electron microscopy (SEM and TEM) are used in nanofibre characterisation, with the former being employed to study the fibre surfaces and the latter the interior structure. SEM uses the reflectance of an electron beam from the sample, and requires the sample under investigation to have a conductive surface. This is achieved for polymer fibres by coating them with a thin layer (5–20 nm) of metal, typically gold. SEM allows us to study thicker samples than TEM, since for the latter the electrons need to pass through the sample. However, only with TEM can we explore the interior structure of the materials. Typical SEM and TEM images of nanofibres are shown in Figure 1.6.

1.4.2 X-ray diffraction

X-ray diffraction (XRD) is a widely used technique in solid-state materials characterisation. It relies on the diffraction of a beam of X-rays by a regularly arranged lattice array of atoms or molecules. Because the distance between atoms or molecules in the solid state is comparable to the wavelength of an X-ray, the particular manner in which the components are arranged will lead to a distinct series of diffraction peaks (we term these Bragg reflections). These form an XRD pattern, a plot of intensity vs. diffraction angle (2θ). Thus, a crystalline material will have a particular diffraction pattern which can be used as a diagnostic tool to identify it. Different APIs will have different patterns, as will different polymorphs or pseudopolymorphs of a given drug, and through XRD all of these physical forms can be identified.

We can also use XRD to quantify the amounts of phases present in a mixture, and to identify the amorphous form (since there is no regular arrangement of atoms or molecules here, there is no diffraction and no reflections are observed in the pattern). Some exemplar diffraction patterns are depicted in Figure 1.7. In the context of nanofibre drug delivery systems, we employ XRD to determine the physical form of the API in the formulation (most commonly, to discern if it is crystalline or amorphous).

1.4.3 Thermal methods

Differential scanning calorimetry (DSC) is a method by which we can determine the temperature at which a transition happens and the energy associated with it. A sample is loaded into a small metal container (the pan), heated at a particular rate and the energy flow is measured during heating. The technique uses a sample pan and a reference pan (which is empty), and both are heated under the same conditions. The amount of energy (heat flow) required to heat the pan loaded with the sample is compared with the reference pan. Phase transitions such as melting (where a solid material is liquefied) and recrystallisation (the formation of a solid from a molten material) can easily be visualised. This is because, when an API melts, heat must be supplied to break the intermolecular bonds between the drug molecules. This is termed an endothermic transition, and requires the DSC instrument to apply extra energy both to heat the sample and break these bonds. Similarly, if a substance crystallises, intermolecular bonds are formed, and thus energy is given out. These events are manifested as signals in the DSC trace, as illustrated in Figure 1.8.

When working with amorphous systems, there exists another very important event which can be detected by DSC: the glass transition temperature (T_g). This is the temperature at which an amorphous material changes from being brittle to being rubber-like, and is manifested as a change in the baseline of the DSC thermogram (Figure 1.8(c)). Below the T_g, a polymer has low molecular mobility and thus is very brittle and easy to snap. Above T_g, molecular mobility is much higher, and so the material is rubbery and flexible.

DSC is widely used in the characterisation of drug-loaded nanofibres because it allows us to determine what physical form the drug takes (crystalline or amorphous; if crystalline, which polymorph or pseudopolymorph?). We can also use it to observe the presence of water or other solvents in the formulations. If water is present in a hydrate (i.e. there is water inside the unit cell of the crystalline structure), then we will usually see an endotherm at around 100ºC followed by an exotherm. These events arise because energy needs to be put into the material to drive away the water, and then the drug molecules will rearrange themselves to form an anhydrous crystal, giving out energy when intermolecular bonds are formed. If water is simply absorbed to the surface or adsorbed inside the body of the fibres, then a broad endotherm will be visible below 100ºC, with no subsequent exotherm. Similar considerations apply to other solvents; for instance, ethanol boils at 78ºC and so an ethanolate will usually show an endotherm/exotherm pattern at this temperature, while adsorbed or absorbed ethanol will only lead to an endotherm, which will occur at lower temperatures.

To verify the findings of DSC with regard to solvent inclusion, a widely used technique is thermogravimetric analysis (TGA). This measures the mass of a sample as a function of temperature. If a volatile component such as water or another solvent is present in a formulation it will be evaporated upon heating, leading to mass loss. TGA thus allows the amount of the volatile component to be quantified. The temperature at which this occurs (and the ratio of API:volatile component) can be used to verify whether the solvent is present in the unit cell of the crystal or is merely adsorbed or absorbed. The disadvantage of this technique is that, while weight changes can clearly be seen, it is not always possible to determine what may have caused the change leading to possible misinterpretation of results. The combination of TGA with a mass spectrometer or infrared (IR) spectrometer allows an evolved gas to be analysed and its identity confirmed.

1.4.4 Infrared spectroscopy

When molecules absorb IR radiation, the bonds in them vibrate (bend and stretch). Different types of bonds will require different amounts of energy to vibrate, and so the exact wavelength of IR absorbed to cause these transitions is correlated with the types of bonds present. IR spectroscopy thus provides information on the structure of a molecule, since the presence of, for example, C=O, -OH, or -NH₂ groups can be discerned. An IR spectrum is a plot of absorbance or transmittance (the two are related through the Beer–Lambert law, T = 10^–A) vs. the wavenumber (1/λ) of radiation absorbed. The typical IR frequencies of commonly encountered groups are summarised in Table 1.2.

Table 1.2 The infrared wavenumbers at which common functional groups are seen

For solid-state materials, the vibrations of intermolecular bonds between molecules can also be seen through IR spectroscopy. These are referred to as phonon vibrations and occur below 1000 cm^–1. Typically the pattern of phonon peaks is too complex to unravel fully. In the context of drug-loaded polymer fibres, IR is most commonly used to provide additional evidence on the physical form of the API in the formulation, and to seek insight into any intermolecular interactions present. If the API is amorphously distributed in the carrier, then there will be no or minimal API–API interactions, and thus the phonon vibrations of the raw API will not be visible in the IR spectrum. The formation of hydrogen bonds or other intermolecular interactions will result in the change in position or the disappearance of peaks. For instance, where an API has COOH groups, the characteristic C=O stretches will shift in position as a result of H-bonding.

By way of an example, IR spectra for fibres prepared from poly(vinyl pyrrolidone) (PVP) and indomethacin, together with the chemical structures, are presented in Figure 1.9. The full-length spectrum in Figure 1.9(a) demonstrates that indomethacin has many peaks below 1000 cm^–1, a number of which will be due to the phonon vibrations of this crystalline material. PVP, an amorphous polymer, has only a few peaks in this region, and most of the API phonon peaks are clearly absent in the spectrum of the drug/polymer fibres. This indicates that the indomethacin was amorphously distributed in the fibres, a hypothesis which was validated using DSC and XRD.3 The enlargement of the 1800–1500 cm^–1 (carboxylate) region in Figure 1.9(b) shows that indomethacin has two peaks in this region, at 1712 and 1689 cm^–1 arising from the COOH and amide groups (Figure 1.9(c)). PVP has a single band at 1654 cm^–1 from its C=O group. In the spectrum of the fibres, these various carboxylate peaks have merged into a single band at 1660 cm^–1, indicating the existence of intermolecular interactions between the API and polymer.

IR spectroscopy is thus very useful to provide insight into both physical form and intermolecular forces. Molecular modelling approaches, in which computational calculations are used to explore the possibility of interactions between species, can be helpful in confirming the interactions suggested from IR spectroscopy. At this point, we should briefly touch on Raman spectroscopy: this is a technique similar to IR spectroscopy, but whereas IR spectroscopy uses the absorption of light to generate a spectrum, Raman spectroscopy exploits the scattering of light by vibrating molecules. Both can be used to explore drug-loaded fibres and provide similar information on the stretching and bending of bonds.

1.4.5 Functional performance

1.4.5.1 Dissolution and permeability testing

One of the most crucial attributes of a formulation to be tested is its functional performance. For drug-loaded nanofibres we are typically most interested in the drug release profile, and the amount of drug which can permeate through biological membranes. Exactly how the drug release profile is measured will depend on the application intended: a formulation intended as an oral fast-dissolving system should not be subject to the same release test as one designed for extended release in the small intestine. More details of the precise tests which have been used will be given in subsequent chapters, but one very common assay is dissolution testing.

Dissolution, the transition of drug molecules from the solid state into solution, is often the rate-limiting step to drug absorption in the body, and thus an understanding of this can give us an idea of how a formulation will perform in vivo. Most often, dissolution testing seeks to mimic the transit of the drug through the GI tract following oral administration. There are well-established tests stipulated in pharmacopoeias which allow for quality control between batches to be maintained, and go some way towards mimicking in vivo events. Typically, a formulation is immersed for 2 h in 0.1 M HCl solution (pH 1.0) to simulate the pH of the stomach, and then (if required) in a pH 6.8 phosphate buffer representative of the lower parts of the GI tract. Often this is achieved by starting with 750 mL of the acidic solution and after 2 h adding to that a sodium phosphate solution to raise the pH to 6.8.

The US Pharmacopoeia (USP) lists four different types of dissolution test, of which the most commonly used is probably method II, the paddle method. The apparatus required for this experiment is given in Figure 1.10. In USPII experiments, a formulation is stirred at 50 rpm and 37ºC at the required pH. At periodic intervals, an aliquot is removed and the concentration of drug determined using a technique such as ultraviolet (UV) spectroscopy or high-performance liquid chromatography (HPLC). We then construct a plot of percentage release vs. time, which helps us to understand how and where the drug will be released in the body. These tests are not truly representative of what happens in the body – they replace the complex physiology of the GI tract with a litre of a buffer, and do not take into account the precise make-up of the gastric fluids in the body – but because the tests are standardised they permit easy comparison of different formulations. Further, the pharmacopoeias stipulate certain requirements for different dosage forms (e.g. for immediate release, the British Pharmacopoeia states that 80% of the drug should be released within 15 min in the stomach), and dissolution testing allows new materials to be assessed against these standards.

An API, if given orally, must both dissolve into solution from a solid dosage form and also permeate through the biological membranes in the intestine to enter the bloodstream and have systemic activity. Thus, a second commonly performed test is a permeation test. These vary in nature depending on the exact formulation being developed, but all involve investigating the passage of the drug through a biological membrane (or an artificial mimic thereof). Often, porcine intestines are obtained and cut into sections for permeation tests. The formulation is placed in a donor chamber, which is separated from a receptor chamber by the biomembrane; a schematic illustration of this is given in Figure 1.11. A Franz cell is often used for these experiments, but alternative apparatus is also available. Samples are taken periodically from the receptor component to quantify the amount of drug which has passed through the membrane.

1.4.5.2 API quantification

Both the assays described above in section 1.4.5.1 require the amount of API in solution to be quantified. This is achieved using UV-visible spectroscopy or HPLC in the vast majority of cases. UV spectroscopy relies on the fact that APIs usually possess a chromophore (a system of alternating single and double bonds which absorb light in the UV-visible window). The amount of light absorbed is directly proportional to concentration, which means that once a suitable calibration curve has been plotted, UV spectroscopy allows the rapid determination of drug content. Much contemporary dissolution apparatus comes with inline UV monitoring, permitting automated determination of the drug release profile.

In some cases – such as for the simultaneous monitoring of two APIs being released from the same formulation – it can be advantageous to use HPLC. This typically uses a UV detector to quantify the drug concentration, but rather than a sample being directly placed in a UV spectrometer, in HPLC the analyte is first loaded on to a stationary phase (the column), a solid material to which the various components of the sample being analysed will adhere with different strengths. A mobile phase (solvent) is flowed through the column, and depending on the strength of interactions between the API and the column the drug will be freed from the stationary phase at a particular time. A UV detector positioned at the end of the column records the concentration of drug exiting the column as a function of time. HPLC is beneficial because it allows two substances with similar UV spectra to be separated and quantified, and can also lead to more sensitive and accurate detection.

1.4.6 Stability studies

The stability of a formulation is of paramount importance. Inevitably, there will be some degradation of a product upon storage, and it is necessary to understand the rate at which this happens and to set acceptable limits to allow a shelf-life to be stipulated. Degradation could take many forms, including chemical degradation of the active ingredient, crystallisation of an amorphous material or microbial contamination. Pharmaceutical scientists often make use of accelerated ageing studies to gain rapid insight into these processes and their rates. In such experiments, a formulation is exposed to stress conditions comprising elevated temperatures and relative humidities (RH). These will typically increase the rate at which degradation processes occur – for instance, the amorphous form will convert to a crystalline material more rapidly at 80ºC and 80% RH than at room temperature and 50% RH.

The formulation is monitored over a period of time, and the extent of the degradation process at each time is determined using analytical techniques such as those detailed above. This permits the rate of the process to be determined, and the Arrhenius equation (k = Ae^–Ea/RT) can be used to back-calculate the rate at room temperature, and thus estimate the ultimate shelf-life of the material. This method is far from perfect, because changing the temperature and humidity conditions can change the processes which occur as well as their rates, but it is widely used in the early stages of formulation development to obtain data rapidly. Ultimately the final dosage form will need to be studied under normal storage conditions, but such experiments can take years and are thus not practicable in the early development stages.