Why Beer Bottles Are Brown:
The Optical Engineering Hiding in Glass Packaging
From My Food Engineering Notebook – Series #5
Glass is the packaging material we trust most and question least. Going back through my ITU packaging notes, I realized “glass is inert” was a sentence I had memorized and never examined. So let’s examine it, starting with what glass actually is.
The definition in my slides is ASTM’s: an amorphous, inorganic product of fusion that has been cooled to a rigid condition without crystallizing. Physically it has a random atomic structure built from silicon-oxygen tetrahedra, and it behaves like a supercooled, extremely viscous liquid: as it cools, viscosity rises reversibly until it acts, for all practical purposes, as a solid. That amorphous structure is responsible for both its stiffness and its brittleness, and the second half of that sentence is the theme of this whole post.
Soda-lime glass accounts for nearly 90% of all glass produced. Typical container composition is 68–73% silica, with small amounts of alumina (1.5–2.0%) and iron oxides (0.05–0.25%). Each ingredient has a job. Sodium and potassium carbonates lower the fusion temperature and viscosity of silica. Calcium and magnesium carbonates (limestone, dolomite) act as stabilizers so the glass doesn’t dissolve in water. Alumina increases hardness and durability, boron reduces sodium leaching, and lead gives clarity and brilliance. Replace the alkali with boric oxide and you get borosilicate glass, the ovenware type.
The process is worth following step by step, because every stage exists to manage a problem. Raw materials plus cullet (15–50% of the batch, either factory scrap or consumer-recycled glass) are weighed, mixed and charged into a furnace at about 1500°C, producing molten glass that is chemically homogeneous and free of gas bubbles. The melt is then cooled to around 1100°C, carried through the forehearth, and cut into a gob at roughly 1100°C. Forming is done by blow-and-blow, wide-mouth press-and-blow, or narrow-neck press-and-blow (NNPB). NNPB is the newer route for lightweight bottles: a plunger presses the gob in the blank mold, giving more uniform wall thickness, better strength, and weight savings of up to 30%. Over the last 50 years, container weight has dropped by roughly 25–50%, which is a large part of why glass stayed competitive.
Then comes annealing. Containers cool unevenly during forming and handling, which locks in residual stress, so they pass through a lehr where temperature is raised to about 540°C, held for a few minutes, and cooled in a controlled way. A poorly annealed container can fail if its internal tension is high or the inner surface is bruised.
Get Ruveyda Yaylaçeşme’s stories in your inbox
Join Medium for free to get updates from this writer.
Next, surface treatment, which I now think is the most underrated step. At the hot end (~550°C), vapor containing tin or titanium tetrachloride contacts the outer surface and forms a thin film that prevents surface damage, strengthens the surface, and improves adhesion of the next coating. At the cold end (~100°C), an organic material (waxes, stearates, silicone, oleic acid, polyethylene) is sprayed on to lower the coefficient of friction so bottles slide along the filling line instead of scratching each other. This is not cosmetic, and here’s why.
Glass usually breaks under tensile stress, and the tensile strength of the surface decides when. My slide puts it as an equation: tensile strength plus stress concentration equals fracture. A bruise or contact with any hard body creates tiny cracks, invisible to the naked eye, and those flaws concentrate stress. So container strength is governed by surface condition, not by bulk material. That is why wall-thinned one-trip bottles depend so heavily on scuff-resistant tin treatments, and why shrink sleeves (thin foamed PS, or uniaxially oriented PVC or PS) exist: they prevent the surface contact that creates flaws in the first place.
The stresses a container sees in life are diverse: impact from filling lines, sideways squeezing from rollers, vertical loading from stacking, residual manufacturing stress, internal pressure from carbonated drinks (relevant to sparkling wine too), and thermal stress from hot filling. Fracture patterns are actually readable: impact failures start at the outer surface, pressure failures at the inner surface, and with practice the dominant load can be identified from the break pattern.
One correction to how I framed glass earlier: glass is not thermally robust. My notes are explicit that glass has the least resistance to temperature change of the common packaging materials. Sudden cooling puts tension on the outer surface and compression inside; sudden heating does the reverse. The stresses are temporary, but if they exceed the surface’s strength during that window, the container breaks. ASTM C149 tests this by immersing containers in a 63°C water bath for 5 minutes, transferring them to 21°C water, and counting breakages. That matters for washing, pasteurization and hot-fill lines.
Now the optical part. In silicate glass, transmission is intrinsically limited by silica absorption around 150 nm in the UV and 6000 nm in the IR, and everything in between is controlled with additives: metal oxides, sulfides or selenides. Amber comes from sodium sulfide, green from chromium and iron oxides, blue from cobalt and copper oxides, and a colorless but UV-absorbing glass can be made with cerium or titanium oxide. The functional definition of a light-resistant container is precise: one that passes no more than 10% of incident radiation at any wavelength between 290 and 450 nm through the average sidewall. That band is chosen because those wavelengths induce oxidation in foods. On the transmission curves, flint glass sits above 80–90% across the visible range, while amber stays low through the UV and violet-blue and only rises past roughly 500 nm. Green shows transmission windows that make it a weak protector.
The reason this band matters, from the wider literature rather than my slides, is photosensitized reactions. In beer, riboflavin absorbs blue and near-UV light and drives the breakdown of hop iso-alpha acids into radicals, ultimately forming 3-methyl-2-butene-1-thiol, the compound behind the “lightstruck” skunky off-flavor. A brown bottle is a filter tuned to remove exactly the wavelengths that reaction needs. Milk, vitamin C and carotenoids follow similar logic.
Two last notes, both regulatory. Glass itself is a near-zero-migration wall, but the closure liner or gasket is usually a polymer system with its own additives, and decoration or printing needs its own assessment, so “glassy equals no migration risk” describes the wall, not the package. And on sustainability: cullet lowers melting energy (a general literature figure, roughly 2–3% saving per 10% cullet), but the 1500°C melt and the weight in transport are the counterweights, which is why lightweighting has driven so much of the industry’s engineering.
Tomorrow: metal, coatings, and the sulfur chemistry that can turn the inside of a can black.
