Pharma glass compositions - Part 6. Trace impurities

Hello everyone – the first 5 parts of this series focused on the components that are intentionally added to glass formulations to achieve specific properties and performance (see Footnote 1) – but what about the unintentional components?? It’s not necessarily a topic that is broadly discussed, but you will find references to impurities in brochures for pharma glass tubing and containers.? For example, I pulled up publicly available technical data for 51 expansion borosilicate glass from the three largest suppliers of pharma grade tubing and found the following statements:

·?????? Compliant with “Heavy metal requirements set forth in EU Packaging, Packaging Waste Directive 94/62/EC, and CONEG Toxics in Packaging regulations”

·?????? “The heavy metal content for the elements lead, cadmium, mercury and hexavalent chromium is below 100 ppm.”

·?????? “Contents of Pb, Cd, Hg, Cr?? is below the 100 ppm limit value stated by the US Toxics in Packaging Clearing House (TPCH) and European Parliament and Council Directive Article 11 on 94/62/EC of 10. Dec. 1994 on packaging and packaging waste with updates 2001/171/EC and 2006/340/EC.”

As you can see, these statements are focused on heavy metals that have toxicological implications for the storage of drug products.? However, these are not the only impurities that can appear in a pharma glass.

For simplicity, let’s focus on sand (principally quartz having the nominal composition SiO?) since it is the majority raw material used in the manufacturing of pharmaceutical glass containers.? A number of trace impurity elements can be found in quartz, including aluminum, titanium, iron, sodium, potassium, magnesium, etc. I literally mean impurities in quartz here – these elements can substitute for positions within the quartz crystal structure that would normally be occupied by silicon ions (Si??).? For example, assume that an aluminum ion (Al3?) occupies what would normally be a silicon site in the structure of quartz.? The ionic charge difference between aluminum and silicon (3+ and 4+, respectively) means that an additional cation such as sodium is needed to balance things out (see Footnotes 2 and 3).? While it’s formally correct to say that elements such as aluminum and sodium are an impurity in quartz, they’re perhaps not great examples since most glasses suitable for parenteral packaging also contain these elements as intentional components that are introduced via other raw materials (the one exception being fused quartz glass, which is intended to be pure SiO2).? We might therefore consider elements such as iron and titanium to be the “real” impurities here (see Footnote 4).

Another source of impurities is non-quartz minerals physically mixed with sand, including a number of iron-containing compounds (e.g. hematite and magnetite), aluminosilicates (e.g., sillimanite, muscovite, and kyanite), zircon, and much more.? The suppliers of glass-making sands work to remove these tramp contaminants through a variety of processes such as washing, magnetic separation, and flotation.? Once again, it is impossible to achieve 100% purity, and some level of trace contamination will be present.? A related source of impurities are small-scale inclusions, meaning a mineral phase(s) is found physically embedded within a grain of quartz through a variety of processes.? While inclusions may be an annoyance if all you want is pure sand, they can also produce some visually stunning results at a large scale – see for Figure 1 for an example.

That’s enough about quartz – let’s zoom back out to impurities within the resulting glass.? We have to accept that some level of impurities will unavoidably be present in any pharma glass container.? Instead, we focus on controlling impurity levels through proper raw material selection combined with monitoring glass composition to ensure impurities remain stable (see Footnote 5).? Some of those impurity levels have to be in compliance with applicable requirements, particularly for heavy metals having increased toxicological risk as noted near the beginning of this post.? We also have to consider how much of those elemental impurities might come out of the glass and into the drug product (see Footnote 6).

Finally, simply demonstrating that our impurity levels are under control and in compliance doesn’t necessarily mean that those same impurities will never have an observable consequence for a drug product.? For example, Mould et al. recently hypothesized that ppb levels of iron leached from clear borosilicate glass vials was a contributing factor to the degradative oxidation of polysorbate 80, an important excipient used to inhibit aggregation in protein-based therapeutics (see Footnote 7).

Questions or comments? – please leave them below or feel free to directly contact me.

?

Footnotes

1.? The prior posts in this series include: 1) network formers, 2) network modifiers, 3) network intermediates, 4) fining agents, and 5) modern pharma glass formulations.

2.?????? Refer to my prior post on network intermediates for further discussion on this topic.

3.?????? You might be wondering why aluminum would choose to make this substitution – one answer is “thermodynamics”.? The more mechanically minded among you might think about thermodynamics as a tool for designing engines and pumps, but it can also be very successfully applied to predicting the behavior of materials.? For example, we can think about the free energy of mixing in “solid solutions”, meaning a crystalline solid solution of nominal composition (e.g., SiO?) that can accommodate some amount of other elemental substitutions without forming a new compound.? In short, it turns out that allowing these substitutions is more energetically favorable.? You can sort of turn this argument on its head as an explanation for why it is essentially impossible to produce a 100% pure material at bulk scale.

4.?????? Once again, this statement is relative – iron and titanium can also be intentional components in typical amber glasses used to protect photosensitive drugs from degrading when exposed to ultraviolet light.? I should also note that the refractory bricks lining a glass melting brick could also be considered a sort of impurity in the final glass.? However, these bricks tend to also be made from elements that are already intentional components of the pharma glass.

5.?????? I’ll be following this up with one or more posts describing how we typically perform compositional analysis of pharma glasses.

6.?????? Refer to my prior post on fining agents for further discussion on elemental impurities as they relate to the E&L profile of pharma glass.

7.?????? Reference: Mould R, Sargent Jr PW, Huang Y, Fields AL, Zhang L, Herbert FC, Stewart SL, Wang T (2023).? Impact of primary container closure system on PS80 oxidation and the mechanistic understanding.