A Shallower Understanding of Glass Delamination Made Possible by X-Ray Photoelectron Spectroscopy (XPS)

At McCrone Associates, we are frequently asked to assist our clients with solving glass delamination challenges. Glass delamination occurs when the interior surface of a glass vial succumbs to chemical attack from its contents. This damage typically begins with surface pitting and can eventually lead to the release and suspension of thin flakes, or lamellae, of the glass surface into the liquid. These flakes present a serious concern for pharmaceuticals in particular, due to the health hazards and quality issues that their presence raises. Using an array of microscopic and spectroscopic tools, the scope of the problem may be assessed and its root cause ultimately addressed. With our recent installation of a ThermoFisher Scientific Nexsa X-Ray Photoelectron Spectrometer, which features the Snap Map function for precise location of samples, a dual-mode ion beam sputtering system for more precisely controlled decontamination of delicate lamellae, and a tunable electron flood for optimized charge neutralization, we can offer a more complete compositional profile of suspected delamination flakes than previously achieved.

The advantage offered by X-ray photoelectron spectroscopy (XPS) analysis for lamellae characterization is the exceptionally shallow information depth (on the order of a few nanometers) of the spectra. On the other hand, a scanning electron microscope (SEM), though it offers far superior imaging capabilities, features a significantly larger analytical information depth (on the order of a micrometer). As a result, while the SEM can acquire high-quality images of suspected glass delamination flakes, the energy dispersive X-ray spectra (EDS) acquired during imaging provides much more information about the substrate (usually a carbonaceous filter medium) than it does on the composition of the flake itself.

Although previous XPS spectrometers could, in principle, measure the composition of only the outer surface of lamellae, the challenges of locating a flake, selectively cleaning organic residue (such as active pharmaceutical ingredients or excipients adsorbed to the surface following filtration) from the flake surface without destroying it, and acquiring its spectrum without electrostatic surface charging rendering the data invalid, often proved insurmountable. In contrast, the Nexsa Snap Map routine enables rapid XPS scanning of a region of the substrate to locate the highest intensity of an element of interest; for glass delamination analysis, this can indicate the presence of a flake minimally covered with organic residue. Once a flake is located, an argon ion cluster beam can be used to gently clean the organic material from it without destroying the delicate flake underneath. Once cleaned, a properly tuned electron flood allows high-quality XPS spectra to be acquired, thus offering a more informative and actionable analysis of the flake than microscopy-based methods alone.

Thus, XPS analysis can be more fully integrated into a protocol for characterizing suspected glass delamination. Upon arrival of a drug vial, the fluid contents are filtered while the vial itself is analyzed for pitting and other topographic signs of chemical attack. The filter residues are observed under an optical microscope to identify possible glass flakes (Figure 1). In the SEM, higher magnification is possible (Figure 2), as well as EDS composition information (Figure 3), though at most only a few percent of the signal actually originates from the flake itself.

Once the high-quality images have been acquired, the sample can be moved into the Nexsa, where an integrated optical finder camera and the Snap Map routine are used to identify likely candidates for analysis (Figure 4). The ion cluster source cleans these locations, and utilization of the electron flood assures that well-resolved, stable photoelectron spectra are acquired from these non-conductive materials. Instead of generating spectra dominated by carbon (as is the case with SEM/EDS), the quantitative analysis of the XPS spectrum of a glass delamination flake reveals only a few percent carbon (Figure 5). This information, combined with that obtained by XPS depth profiling of the interior surface of the vial, can offer a more complete picture of the lamellae composition compared to the source glass and provide potential clues for identifying the mechanism of their formation.

Meanwhile, other structures that appear to be glass delamination can be positively identified as other materials (Figure 6), thereby informing changes in the remediation strategy.

We look forward to complementing our current multi-technique, in-depth approach to analyzing glass delamination, which includes optical and electron microscopies, infrared spectroscopy, and X-ray dispersive spectrometry, with the shallow understanding of a flake’s nanometers-thick composition made possible by XPS.