Pharma glass compositions - Part 5. Modern formulations

Hello everyone – Parts 1, 2, 3, and 4 of this series on pharma glass compositions were organized in a way that considers the impact of components on glass structure and properties. We’ll bring everything together in this post by reviewing the general families of glass compositions that are used in pharmaceutical packaging.? Let’s begin by reviewing some nomenclature.? The scheme that is used to name a glass compositional family is not nearly as formal as the system promoted by an organization such as IUPAC to name an organic compound.? For example, what does it mean to be a “borosilicate” glass? Yes, it contains boron and silicon (generally expressed in their oxide forms as B?O? and SiO?, respectively), but pharma grade borosilicates will also contain some level of Al?O?, alkali oxides, and alkaline earth oxides.? However, it’s unwieldy to write and talk about a “sodium-potassium-magnesium-calcium-alumino-borosilicate glass” – we just shorten it to “borosilicate glass” for convenience (see Footnote 1).? I mention all of this because the compositional families we’re about to discuss frequently share many components in common.? However, it’s the changing relative ratios of these components that justify a new designation.

Table 1 provides a summary of approximate compositional ranges of eight different modern pharmaceutical glass compositions (see Footnotes 2 and 3).? The term “modern” is relative – many of these glasses have roots that go back one or more centuries (see Footnote 4).? However, they do reflect glass formulations that are currently in use for pharmaceutical packaging.? Note that the last row of Table 1 simply answers the question of whether or not a fining agent is used (in most cases, the answer is “yes”).? The specific fining agent that is used for a given glass will depend on the manufacturer and its composition.

Starting at the right-hand side of Table 1, we have soda lime silicate (SLS) glass (both tubular and moulded), or what is also often called Type III glass.? It is typically used for storage of oral solids and liquids but is not generally considered suitable for parenteral drug products due to its moderate chemical resistance.? One of the primary compositional trends to notice within Table 1 is the sum of alkali and alkaline earth oxides, also known as network modifiers.? You’ll notice that the total alkali/alkaline earth content of the SLS glasses is greater than all of the remaining compositions, thereby disrupting the connectivity of the glass network and increasing susceptibility to corrosion. Dealkalization treatments can be applied to the inner surface of a SLS glass container to increase the apparent hydrolytic resistance, thereby creating a Type II glass container that may be considered more suitable for parenteral packaging.? However, it’s important to note that such treatments don’t necessarily improve the overall durability of the container under all conditions.

Let’s next consider the three columns labeled high, medium, and low borosilicates as a group.? All three compositions are capable of meeting the hydrolytic resistance criteria to be classified as a Type I glass container and be considered generally suitable for parenteral packaging (see Footnote 5).? A significant majority of the parenteral glass containers in use today are manufactured from one of these three compositions, although parenteral drugs are also certainly found in moulded borosilicate, particularly in larger containers (see also “Moulded Borosilicate” entry in Table 1).? The categories of high, medium, and low borosilicate also correspond to descriptors that are based on the approximate thermal expansion of these compositional families – 33 expansion, 51 expansion, and 70 expansion borosilicates, respectively (see Footnotes 6 and 7).? Once again, reviewing the network modifier content is instructive.? The total amount of network modifiers in the composition decreases in the order: high borosilicate, medium borosilicate, low borosilicate.? Given this information, I would expect the relative hydrolytic resistances of these glasses to generally decrease in the same order.

Aluminosilicate glasses are one of the most recent additions to the compositional families shown in Table 1 (Footnote 8).? Their primary distinction is the absence of the element boron, thereby avoiding an artifact of the tubular conversion process that can impact hydrolytic resistance and increase the risk profile for glass delamination.? You might notice the relatively high concentration of Al?O? in the aluminosilicate family compared to the other compositions.? This leads to the fairly natural assumption that aluminosilicates will display higher levels of aluminum when performing E&L studies.? This turns out to not be the case for aluminosilicates used for parenteral packaging. ?Durability cannot simply be described by applying Le Chatelier’s principle to the individual components of a glass for multiple reasons, including: 1) the difference in relative bond strengths amongst the various glass components and 2) the complex interplay among glass components in determining the structure (and hence the properties) of a glass.

Finally we have fused quartz (a.k.a. fused silica), a glass that is nominally comprised of 100% SiO? and another new addition to the pharma glass packaging space (see Footnote 8).? In reality, it will contain some level of trace contamination that causes the purity to be slightly less than 100% (see Footnote 9).? It’s another example of how Le Chatelier’s principle doesn’t apply in a simple way to the chemical durability of glass.? Here we have a glass that is essentially pure SiO?.? Does this mean it generates the highest levels of silicon in an extraction/leachable study?? The opposite result is actually expected in practice – fused quartz is one of the most chemically durable glasses owing to the strength of silicon-oxygen bonds and the maximized connectivity of the glass network (refer back to my post on network formers for more information).? The trade off with fused quartz is ease of manufacturing.? The lack of network modifiers means that significantly higher temperatures are required to achieve a viscosity that allows forming of fused quartz into tubing and containers.

Questions or comments?? Please leave them below or directly contact me.

?

Footnotes

1.?????? And if you’re paying close attention, you’ll notice that I didn’t mention fining agents.? This is fairly typical practice.

2.?????? The values shown in Table 1 are being provided as ranges since there is some variation in composition among the various manufacturers of pharma grade glasses.? Just remember that the values that for a given glass composition should sum to 100 when being specified on a percentage basis.? Stated differently, 100 grams of a glass having a composition of 15Na?O-10CaO-75SiO? on a weight percentage basis should nominally contain 15 grams of Na?O, 10 grams of CaO, and 75 grams of SiO? (I’m ignoring the use of fining agents and the presence of impurities in this example).? I should also mention that while it’s practically useful to specify compositions on a weight percentage basis, it is often better to express glass compositions on a mole percentage basis to relate composition to properties.? This is particularly true when a glass is composed of oxides having drastically different molecular weights.? Going into more detail on this point is beyond the scope of this post – just know that it can be important to understand what basis is being used (weight percent or mole percent) when specifying a glass composition.

3. The first version of this article was published in June 2024.? I chose to slightly revise Table 1 to be consistent with recommendations recently made by a PDA expert committee in response to revisions proposed to USP <1660> in September 2024.? The committee included representatives from multiple major suppliers of pharma glass packaging.? As a result, we were able to pull together a better data set that more accurately captured the full range of commercially available glass compositions.

4.?????? A detailed review of the historical evolution of glass compositions used in pharmaceutical packaging is beyond the scope of this post.? If you’re interested in learning more, I would encourage you to look up the following article.? Reference: Schaut RA and Weeks WP (2017).? Historical review of glasses used for parenteral packaging. PDA Journal of Pharmaceutical Science and Technology, 71: 279-296.

5.?????? Emphasis is on the phrase “…are capable of…”. ?All of these glasses can be sufficiently durable as tubing, but a poorly controlled conversion process can turn them into vials that fail to meet the hydrolytic resistance criteria required for Type I status.

6.?????? I should emphasize that these are approximate thermal expansions that have been adopted as popular jargon to classify glass compositions.? For example, if you look up the technical data sheets for what might generally be called 51 expansion glasses, you will find some small amount of variation of the actual thermal expansion.

7.?????? 70 expansion borosilicate glasses for parenteral packaging are primarily found within the Asian market.? My understanding is that they are gradually being phased out in favor of 51 expansion borosilicates, although I don’t have a clear line of sight to if or when this transition may fully take place.

8.?????? Both aluminosilicate and fused quartz have recently been classified as Type I glasses due to October 2023 revisions to USP <660>.? Prior to that time, Type I glasses had to be a based on a borosilicate composition.? This is not the case for EP (3.2.1) that still includes the borosilicate requirement.

9.?????? This statement about trace contamination is true for any glass.? For example, you may notice a slight colored tint when looking down a piece of pharma grade glass tubing.? This coloration is the result of small amounts of transition metals such as iron that are naturally present in the glass as a trace contaminant.? In the case of fused quartz, we would also consider elements such as sodium, potassium, etc. to be trace contaminants – these same elements are often majority components of the other compositional families shown in Table 1.