Glass Packaging Shelf Life: What the Pack Decides

In a glass pack the bottle or jar almost never sets the shelf life. Glass is inert and impermeable, so the date code is decided by what is fitted to it or sealed inside it: the closure and its liner, how much light gets through, the decoration, and the gas left above the fill. Packaging can adjust and verify each of those, but it cannot remove a sensitivity that belongs to the formulation, so the first job is to find which factor gives way first on your filled pack.

What to establish before changing anything

An empty container does not age in any way that matters, so asking how long a glass bottle lasts leads nowhere. The question worth asking is which packaging factor shifts first under the storage and display conditions this product will really see, and how that shift would be detected. If an investigation starts with the glass body, it has started in the wrong place.

Three inputs are needed to answer it:

• the content and how sensitive it is;
• the shelf life you are aiming for;
• the conditions along the route, meaning warehouse temperature, temperature in the delivery van and the shop, light on the shelf, and how long the pack stays open once the customer is using it.

Next, settle the direction of the failure. Either something is leaving the pack (moisture, aroma, carbonation, or volatiles moving into the closure) or something is getting in (oxygen, moisture, or light that sets off reactions in the product). The closure system governs both more than the glass does, yet the remedies can pull in opposite directions, so a fix proposed before the direction is known is a guess.

Two neighbouring questions are handled elsewhere. Which documents must travel with a glass pack, and what a declaration has to contain, is covered in our guide to the food contact declaration for glass bottles. Which glass colour blocks which part of the spectrum, and how to write that into a specification, is covered under amber and brown glass. Here the pack is assumed to exist already.

Five packaging factors and who owns each

Most shelf life results in glass come down to five factors. The table sets out how each one works, what packaging can change, how the change is checked, and where the packaging side's responsibility stops. That last column is where disputes begin: no pack can rescue an unstable formulation, and no reformulation can make up for a leaking closure.

These factors interact. Darker glass lowers the light load, and a lower light load slows the rate at which the product uses up oxygen that came in past the closure, so both changes work against one and the same failure. A change to one factor can also open a problem in another. For both reasons a change is always verified on the complete pack, never on a component alone.

Closure, liner and added seals

After filling, nothing but the closure stands between the product and the air. It is also the component most often chosen on cost instead of performance, which makes it the place where a packaging change usually pays back most.

The sealing interface

Each finish seals in its own way, and each has one quantity to watch. A thread finish needs the liner pressed onto the sealing surface at the right load, so the torque window and the conditions under which torque is measured are what count. A lug finish depends on vacuum pulling the lid onto its gasket, so the vacuum level after filling is the figure to track. A crown depends on the crimp gripping the bead.

An unevenly seated seal is the hard case. It looks fine on inspection and then fails in storage, and the symptom is a gradual drift in the product, not a visible leak.

The liner

Liner polymers and films differ widely, both in how much oxygen and moisture they let through and in how well they stand up to a given product. With an oxygen-sensitive content, a low-permeability liner or a construction containing foil will move the result further than anything done to the glass. With a chemically aggressive content the priority reverses: choose for resistance first and treat permeability as the second criterion.

Product interaction tends to show up in the liner before it shows anywhere else. For an unusual product, expose the liner material to it directly before the closure specification is frozen.

An induction seal or inner film

When a standard closure falls short of the barrier the target needs, an induction seal or an inner film can be added. The closure then works as a mechanical seal and a barrier seal together. This alters the pack, the filling process and the way the customer opens the product, so it needs validation on the filling line, not a decision made at a desk.

Light protection through colour, wall thickness and coverage

Light is the factor buyers underrate most, because it works on the product all the time and leaves no visible sign until the damage is done. A pack that holds steady in a dark warehouse may fail in a sunlit shop window with nothing wrong with the container. Two packs sharing a shape and a colour can also perform differently, since transmission follows wall thickness and the share of the surface that is covered.

Packaging has three levers:

• Glass colour. Darker glass cuts transmission in the damaging band. This is the usual reason a product goes into amber and not flint.
• Glass path length. More glass in the direction the light arrives from cuts transmission again.
• Coverage. A full-wrap label or shrink sleeve takes away the exposure wherever it sits. On a pack whose decoration spans most of the body, a sleeve can contribute as much to stability as a colour change would.

The size of the gain cannot be taken for granted. Transmission changes with wavelength, glass composition and thickness, and the product's own sensitivity changes over that same band. So "darker" is not automatically "better". Find out which wavelengths harm the content, then measure the candidate pack across them.

Decoration as a shelf life factor

Decoration is seldom on the list when shelf life is discussed, and now and then it turns out to be the cause. There are three ways it can matter.

Migration. Inks, coatings and adhesives are designed for the outside of the container. When decoration sits near the fill, or a solvent-based system goes onto a thin-walled container, movement into the product has to be assessed, not waved through. The rules on permitted materials and documentation belong to the food contact declaration; the assessment itself is part of the shelf life work.

Optical effect. Dense print or a metallic coating alters how much light reaches the product behind it. A label may therefore protect, or it may behave as a selective filter. Once the product's light sensitivity is known, the coverage pattern is a technical variable as well as a visual one.

Appearance drift. Certain inks and coatings shift in colour or gloss under the very conditions that age the product, most of all when the product contains solvents or the pack stands in strong light. Nothing has gone wrong functionally, but a retailer will still treat it as a shelf life failure. It is found by setting the stored pack beside the sealed reference sample, since no measurement on the product will reveal it.

Where packaging responsibility stops

Shelf life investigations stall most often because two teams each hold half the problem. Writing the split down speeds everything up.

• Packaging side. Barrier performance of closure and liner; seal integrity and the window in which it holds; light transmission of container and decoration; the headspace volume the closure can take; and whether the packing system suits the storage and transport conditions. All of these are measured on the pack and can be changed without touching the recipe.
• Formulation side. How sensitive the content inherently is to oxygen, light, temperature and time; any inert gas used at filling; the water activity or pH that steers microbiological and chemical pathways; and how well the content tolerates the chosen liner and ink materials.
• Shared. Deciding which failure mode is the limiting one, setting the fill parameters that fix the headspace, and defining what failure looks like for this product.

The shared definition of failure is where projects go wrong most. It has to be set in sensory or analytical terms before the trial starts. Otherwise the study ends in a debate over whether the change that appeared is acceptable.

What accelerated aging can and cannot show

A true shelf life study lasts as long as the shelf life, so projects run accelerated aging for an early read. Filled packs are held under harsher conditions than intended storage, normally raised temperature and frequently controlled light and humidity, and the attributes a real-time study would track are measured. Standard methods exist, among them the ASTM accelerated aging approach and similar protocols from pharmaceutical and food packaging practice. The results are then used to estimate behaviour at ambient conditions.

Its strength is ranking. Put two closure systems through identical severe conditions; if one lasts markedly longer, that tells you something useful before anyone trusts the absolute values. No quicker route exists for choosing among candidate configurations.

Its weakness is the extrapolation. Heat can trigger chemistry that never happens at ambient temperature, soften a liner so the seal load changes, and shift the failure from one component to another. An early failure in an accelerated study points to a real problem. Survival is an encouraging signal and nothing firmer than that. Whether severe-condition results carry over to ambient depends on the particular pack and content, and the case has to be made from the mechanism.

In practice, accelerated work sets the direction and narrows the field, and a real-time study of the actual pack supports the date code. If the launch cannot wait, declare a shorter initial shelf life backed by the real-time data points already in hand and lengthen it as more arrive. Publishing an extrapolated figure as if it had been measured is the thing to avoid.

Making a shelf life claim that stands up

A claim is a commitment, and it holds only while it stays within what the data support. Four habits do most of the work.

1. Attach the conditions. Give the temperature range, the light exposure and, where it matters, the humidity, and say what the interval is counted from, for example the filling date and not the dispatch date. Without these the figure is only a number.
2. Name the basis. Say whether the figure comes from real-time data, from real-time and accelerated data combined, or from extrapolation. Buyers ask this more and more, and naming the basis is far more convincing than asserting a duration.
3. Keep the pack apart from the promise. A pack can be specified, tested and documented. Storage after dispatch is outside the supplier's control. Records of what the pack was designed and verified against therefore carry more weight than the number, above all when the product passes through several distribution steps.
4. Write the failure criterion first. Shelf life is the time until a defined attribute passes a defined limit. With no written criterion the interval has no definition, and the figure gets reopened the first time two people disagree over whether a change is noticeable. A known endpoint also makes the storage trial quicker to interpret.

The seal is the attribute most commonly used as the leading indicator that a pack is approaching its limit. How it is measured and compared is explained in our guide to torque testing for bottle caps.

Assumptions that undermine a claim

• Treating glass as protection. Glass keeps out oxygen and moisture, but it is transparent and does not seal itself. An inert container is not a protected product, and confusing the two is the most frequent mistake in this field.
• Borrowing a figure. A shelf life measured for one content in one pack does not carry over to a different content, closure or colour, because the limiting mechanism changes with the combination. A number taken from a neighbouring project is a hypothesis awaiting a test.
• Testing the bottle and not the pack. An empty bottle sitting in a warehouse says nothing about shelf life. Measure filled, sealed, packed units under the conditions the product will face.

When the investigation becomes a specification

The investigation ends once the limiting factor is identified and a configuration is selected. From then on the task is to fix and record the pack: container and colour, closure and liner, the sealing window and its measurement method, the decoration system, the fill height, and the storage conditions against which the whole assembly was verified.

The pack also becomes one line in a sourcing programme at this stage. The questions turn to the container, whether it can be had in the colour and finish required, and how the specification stays unchanged from one repeat order to the next. Our glass bottle and jar collections are where that container work starts.

To begin on a shelf life target with us, send the content and its known sensitivities, the target and the conditions it assumes, and the channel conditions: storage temperature, time on display and expected light exposure. From those we can suggest the packaging-side changes most likely to shift the limiting factor, with a verification point for each.

Frequently asked questions

Does an empty glass bottle have a shelf life of its own?

Not in any practical sense. What needs answering is how long the filled pack stays within specification, and the closure, liner, seal, light exposure, decoration and headspace decide that, not the glass body.

Which factor usually runs out first?

Usually the closure system and liner. The order depends on the content, though. A light-sensitive product behind a sound seal may be limited by transmission, and a product that attacks the liner may be limited by that interaction ahead of both. Identify the limiting factor by measuring the actual pack; a general rule will not do it.

Will a different closure really lengthen shelf life?

Yes, and it is often the biggest packaging lever on offer. Lower permeability in closure and liner, together with a seal confirmed as fully seated, can outweigh any container change. Adding an induction seal or inner film goes further, at the cost of a changed filling process and opening experience that must be proven on the line.

How much difference does darker glass make?

No general figure applies, because the answer turns on which wavelengths affect the content and on the glass thickness in the light path. Measure the candidate packs across the relevant band and compare them on that. Where a label or sleeve covers most of the body, it removes exposure there whatever the glass colour.

How do accelerated results relate to the final figure?

They screen and rank options; they do not replace real-time data. When a supplier answers a shelf life enquiry, the reply should say how the figure was arrived at and should not pass off an extrapolation as a measurement.

Is the packaging supplier or the brand responsible for shelf life?

Both, with the split put in writing. Packaging answers for closure barrier, seal integrity, light transmission of container and decoration, and tolerable headspace. Formulation answers for the content's inherent sensitivity, the purge gas and the fill parameters. Both sides agree the definition of failure before a storage trial begins.

How should the shelf life be declared on the pack?

Together with the storage conditions it assumes, the point it is counted from, and the basis it rests on. When a declaration is challenged, the pack's supporting documentation generally settles the matter, not the number printed on the label.