Delamination is defined as separation into constituent thin layers (lamellae).  Glass delamination, which can have several causes, could then be defined as a separation of surface glass, as from a vial, into thin layers resulting in a flaky appearance.  Pitting could be related to, or even be the initial stages of glass delamination; data will be provided that supports a relationship.  The presence of glass delamination is a serious concern to the pharmaceutical industry, not only because suspended glass flakes in any pharmaceutical product poses an obvious health hazard, but because of regulatory quality control standards, and medical profession requirements.    

There are, of course, many ways of addressing the problem of glass delamination in pharmaceutical vials, including looking into the relationship of chemical nature of vial contents to degree of delamination and conditions of vial manufacture.  Specialized lighting and imaging methods are other research approaches.  The results from the present microscopical study, although confined to the vials of only three different manufacturers, indicate support for the theory that pitting may be the precursor or early indicator of full-scale glass delamination.

As already suggested, glass delamination in pharmaceutical vials can occur during manufacturing of the vial, e.g. fusion at the neck or base, where delamination and/or pitting are frequently found; or by adverse reaction to the vial contents, e.g. etching due to an acidic or basic solution, or one with a high sodium/salt content.  Solution-filled vials often show delamination at the fill line.  For analytical purposes, therefore, such suspect vials should be left filled and not sent empty.  From the standpoint of the microscopist this point must be emphasized.  In the past we have been obliged to speculate a fill-line delamination in an empty vial based on previous experience with fill-line delamination problems.  Filled vials are vastly preferred by analytical microscopists because such fill lines can be recorded, and, most importantly, the contents can be filtered under cleanroom conditions, in order to concentrate and study the glass delamination.  Figure 1, for example, is a reflected-light photograph of delaminated glass recovered on a polycarbonate membrane filter; the brown colored flakes seen in the photo are thinner.

McCrone Associates received a project from a pharmaceutical company that requested comparison in the amount of glass delamination from three different glass vial manufacturers.  The vials were received empty, but had previously contained the product (except for one reference vial); the product was described as strongly basic.  All vials were subjected to the same temperatures and time studies documented and provided by the client.  Figure 2 shows the areas of a glass vial that will be referred to when delamination and pitting are found.  All photomicrographs were taken in reflected light.

click image to enlarge (646K) FIGURE 1

click image to enlarge (283K) FIGURE 2

For confidentiality purposes, the manufacturers will be referred to as Company A, B, and C.    In addition to determining the amount of glass delamination, further analysis was requested for Company B only, due to an unusual vertical-striation appearance. The first set of three vials represented all three manufacturers and were identified as having been kept at 55˚C for a two week period.  The second set of vials represented Company B and were identified as having been kept at 40˚C, 30˚C, 25˚C, and 4˚C respectively for a four week period.  The third set was Company B only, and identified as having been kept at 40˚C, 30˚C, 25˚C, and 4˚C respectively for an eight week period.  In addition to these client supplied vials, we requested for our own research purposes vials from Companies A and C that had been kept at 40˚C, 30˚C, 25˚C, 4˚C for a four week period. 

The findings of the first study revealed pitting in vials from all three companies (pitting can be seen in the flakes in Figure 1).  Company A’s pitting was at the base of the vial; Company C’s was at the base, and fine pitting was found at the neck; Company B’s pitting was found throughout the vial.  Delamination was definitely present in the vials from Company A and C; both delamination bands were common to what was believed to have been the fill line.  Company C’s delamination is closer to the base of the vial, because the vial was larger than A and B; it could be assumed the vial is filled with the same amount of solution.  Company B had two horizontal rings of defects that were not positively identified as delamination.  Something not commonly seen was vertical striations down the entire vial, and brown residue within the striations.  Analysis will provide results of these findings.

Figures 3 and 4 show delamination and pitting for Company A’s vials.  There is a dark pink/brown residue below the delamination band; and there is pitting at the base of the vial which could be a thinner area of delamination.  Scanning electron microscope (SEM) images show delamination (Figure 5) and pitting (Figure 6); a flake and a piece flaking off can be seen in Figure 5:  pitting can be visualized going beneath the surface (Figure 6).

click image to enlarge (290K) FIGURE 3

click image to enlarge (342K) FIGURE 4

click image to enlarge (214K) FIGURE 5

click image to enlarge (235K) FIGURE 6

The pink/brown residue seen in company A’s vials was isolated for infrared spectrometry (IR), along with delamination flakes, to see if there was a similarity between them so as to be able to conclude whether or not the residue could be delamination.  X-ray photoemission spectrometry (XPS) was also used to compare the delamination and residue.  All IR spectra show the broad band at ~1100λ which is characteristic of silica, suggesting glass.  We can conclude from this information that the delamination and residue are similar.

Figure 7 shows the striations and pitting from company B’s strange striations.  Figures 8 and 9 show the striations and the brown residue within the striations.  Figure 10 shows that the “defects” of company B’s vials are very similar to the delamination bands of the other companies vials.  They look slightly different, and were at the very neck region of the vial, possibly the fill line.  No IR spectrum of this delamination was obtained because a representative sample could not be removed.

click image to enlarge (173K) FIGURE 7

click image to enlarge (115K) FIGURE 8

click image to enlarge (130K) FIGURE 9

click image to enlarge (437K) FIGURE 10