Gently does it: SIMS of drug-polymer composites separates spectra of individual components

Controlled drug release systems and SIMS

The great demand for controlled release drugs has been matched by analytical developments which can characterise the various drug delivery systems. One of the more popular methods is time-of-flight secondary ion mass spectrometry (TOF-SIMS) in which a beam of primary ions is directed onto the surface and the secondary ions released are trapped and analysed.

When organic compounds are the target compounds, the secondary ions tend to be highly fragmented or rearranged, although this does depend on the experimental parameters such as the primary ion energy and the beam density. So, only a small proportion of emitted ions are related directly to the structure of the compound.

For drug delivery systems, which contain a polymer matrix as well as the drug, the resulting mass spectra can be highly complex, requiring considerable expertise to interpret.

The early use of SIMS tended to use atomic ions in the primary beam and small libraries of mass spectra of organic compounds were produced. Latterly, these have been replaced by ionic clusters, which produce spectra that are largely incomparable to those in the libraries.

The combination of complicated mass spectra and the inability to match with library spectra encouraged two scientists at the National Physical Laboratory at Teddington in the UK to develop an alternative procedure that simplifies the SIMS mass spectra.

G-SIMS hits the spot

Ian Gilmore and Martin Seah named their new process gentle SIMS, although it has more to do with processing of the spectra than their generation. The duo postulated that secondary ions emitted near the point of impact on the surface would be more fragmented than those emitted further away, due to the difference in energy densities, so generating a population of intact secondary ions and their fragments. At lower primary ion beam energies, the proportion of intact ions would increase.

This hypothesis was proven experimentally and led to the development of an equation describing the mass spectrum, incorporating an extrapolation factor known as the g index. This factor could be used to separate the secondary ions in the spectra on the basis of fragmentation energy, in the same fashion as chromatography, to produce simplified spectra in which the dominant ions were more easily related to the structures of the analytes.

At higher g index values, the G-SIMS spectra comprise a large proportion of fragments that are easily generated, such as molecular ions. As the g index decreases, the proportion of fragmentation increases.

Now, in conjunction with scientists from the University of Nottingham, the technique has been applied to two model drug-polymer systems. Codeine or bupivacaine were dispersed in poly(lactic acid) which is a widely used biodegradable polymer that has already been characterised by TOF-SIMS and G-SIMS.

Initial studies of the composites by X-ray photoelectron spectroscopy demonstrated that the drugs were distributed vertically throughout the sample, while atomic force microscopy showed that the surface was smooth with no indications of phase separation.

The SIMS spectra were collected for two types of atomic primary ion, caesium and argon monatomic ions, which were rastered across the surfaces of the samples in a grid. The resultant peak intensities were used to plot a thermal colour intensity map for each g value, similar to a chromatogram, which the researchers named a g-ogram.

For each mass peak, coloured intensity peaks were generated which characterised the peak. Those which began dark then lightened with increasing g index values originated from processes requiring low energy input having little fragmentation. Conversely, those changing from light to dark as the g value increased were related to processes requiring higher energy and producing greater levels of fragmentation.

For each drug-polymer matrix, two classes of peaks could be observed, corresponding to peaks from the drug or the polymer. They were separated by a horizontal line in the g-ogram, at g index values of 4 and 6 for bupivacaine and codeine, respectively, easily distinguishing the peaks of each component.

The peaks derived in this way for poly(lactic acid) and for each drug corresponded to known SIMS mass spectral peaks, although there was some overlap between peaks from the polymer and the drug across the g-ograms. Nevertheless, the separation was considered to have "worked excellently, considering there is no a priori information." The good separation was observed for drug polymer ratios ranging from 1:10 to 1:40.

One major disadvantage of this approach is the inability to separate the spectra of two drugs in a polymer matrix, but there remain many single drug-polymer combinations for which it is applicable.

In practice, peak processing can be carried out in real time, collecting and interpreting the spectra during data acquisition, due to the advent of a binary ion source.

The researchers declared that this spectral simplification method would also work with other types of imaging mass spectrometry, such as MALDI imaging and could be extended to the analysis of impurities or additives in polymers and other materials.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.