Pharma glass defects - 2. Stones

Hello everyone – this post is a short entry into my new “pharma glass defect of the week” series. I previously mentioned that I would start by focusing on critical defects. However, I also noted that the classification of a defect could also depend on several factors, including but not limited to size and location. Today’s defect is one such example.

A “stone” is an opaque inclusion within a glass container. There are multiple potential sources of stone defects, including un-melted raw materials, small fragments of refractory material from the glass-melting furnace, and/or transformation of the glass-forming melt by crystallization. Despite these multiple mechanisms, stones all have one common aspect – they generally occur during the initial melting step, prior to the formation of a glass container. Let’s cover each mechanism in a little more detail.

Mechanism #1: Un-melted raw materials

My ever-growing list of future posts includes a series devoted to the life cycle of a pharmaceutical glass container, including raw materials, melting, container manufacture, etc. For now, I’ll just say that any glass suitable for parenteral packaging will contain silica (SiO₂) as a principal component, and quartz sand is the raw material of choice for introducing silica into the mixture of materials known as a "batch" (see Footnote 1). Silica is a challenging material to melt on its own, and so we add other raw materials to the batch to act as “fluxes” – i.e., components that facilitate the formation of a molten liquid at lower temperatures. This molten liquid helps to dissolve the grains of sand to form a compositionally homogenous liquid over time. However, this process is not always completely effective, and so un-melted material can be carried through the furnace and become embedded within the resulting glass.

Mechanism #2: Refractory material

The furnaces used to melt pharmaceutical grade borosilicate and aluminosilicate glasses are constructed from ceramic refractory bricks that are designed to withstand high temperatures. Despite their good performance, all refractories will react with their molten contents over time. This can result in small fragments of refractory material being shed into the melt. These fragments ideally melt over time to form a homogeneous liquid – failure to completely dissolve produces a stone.

Mechanism #3: Crystallization

I covered something called the V-T diagram in my prior post on residual stress in glass containers. I noted how a liquid can cool to form a crystalline solid or a glassy material (see Footnote 2). However, this isn’t a simple either/or proposition. A molten liquid can also form a material containing a mixture of glassy and crystalline phases (see Footnote 2). For example, Figure 1 shows a piece of jewelry fabricated from cut and polished “snowflake obsidian”, a naturally-formed volcanic glass containing crystalline inclusions of cristobalite (a high temperature crystalline phase of silica – see Footnote 3). Changes to the composition of the molten liquid during processing can also induce crystallization. For example, the loss of volatile components at the surface of the glass-forming melts can create a silica-enriched skin that encourages the crystallization of cristobalite-containing “scum”.

A combination of characterization tools (optical microscopy, electron microscopy, energy-dispersive X-ray analysis, etc.), an understanding of materials science fundamentals, and detailed process knowledge is often needed to determine the root cause of stone defects in glass. Identification of the specific mechanism leading to stone defects in glass is an important first step in determining an appropriate corrective action. For example, stones resulting from un-melted raw material can occur because of changes in the particle size distribution of the raw materials, disruptions in the temperature profile of the furnace, and altered flow behavior of the melt (among other things).

I should also note that stones don’t necessarily have to occur in isolation.  For example, it wouldn’t be surprising to see “knots” in glass that also contains stones. A knot is simply a glassy inclusion of differing composition embedded within a matrix of the parent glass. Knots are often the result of a stone that has fused into a high viscosity liquid without yet fully blending into the surrounding melt.  You might also see small bubbles surrounding the stone depending upon the chemistry of the material comprising the inclusion and its reaction upon dissolving within the melt.

Finally, circling back to the point of this post – why do we care about stones in pharmaceutical glass containers and how are they classified according to defect lexicons such as PDA TR 43 (see Footnote 4)? I’m going to focus on tubular glass vials since that’s my general focus on a day-to-day basis (see Footnote 5). A stone is considered a “Critical” defect if it compromises seal integrity. This could occur because the continuity of the flange sealing surface is disrupted. A difference in the thermal expansion coefficients of a sufficiently large stone  and the surrounding glass matrix can also generate enough stress to cause cracking, which is itself another Critical defect that we will cover in a future post. A stone defect is considered “Minor” if it’s fully embedded within the vial body and ≥1 mm in size; a similarly embedded stone <1 mm is considered an “Acceptable imperfection”.

Questions? – please leave a comment below or feel free to directly contact me.

Footnotes

1.  A few points here – first, quartz is not the only raw material that is used to contribute silica to the composition of these glasses. For example, naturally occurring feldspathic and feldspathoid minerals such as albite, nepheline, orthoclase, etc. can also be used to introduce silica along with varying ratios of other elements such as sodium, potassium, and aluminum.

2. Glass-ceramic materials are a technologically important class of materials that leverages this idea, but in a more controlled manner. Glass-ceramics are generally formed by quenching a liquid to form a glass, which is subsequently re-heated to create crystalline inclusions of a known size and composition. Glass-ceramics are all around us in a variety of settings – dental ceramics, cooktops, and cookware, to name a few.

3. It’s not uncommon for a compound to exhibit multiple crystalline structures, also known as “polymorphs”. The stability of a given polymorph depends on conditions such as temperature and pressure. While the chemical composition of each polymorph is identical, the varying structures will result in materials with different properties. For example, cocoa butter (the fat component of chocolate) can crystallize into various polymorphs that impact the visual appearance, melting characteristics, and mouth-feel of chocolate confectionary.

4. PDA Technical Report No. 43 (Revised 2023) – Identification and classification of nonconformities in moulded and tubular glass containers for pharmaceutical manufacturing.

5. I'm primarily focused on defects within primary glass packaging, although the mention of converted tubular vials brings up a related point. I would expect a stone defect in a tubular glass vial to have also been present within the original glass tube used as an input material in the conversion process. With that said, I can't rule out the possibility that a (small) stone within a piece of glass tubing gets "erased" during the conversion process if it happens to experience enough heating, although I rate this as being pretty unlikely to occur.