Fine cracks in the plastic mean the terpenes are attacking the tube, not the balm
Limonene and other terpenes stress crack polystyrene and PET tubes. Which polymers resist it, a 30 day soak test and the packaging change that ends the problem.
Hairline cracks around a tube thread, a milky haze creeping up from the base, a cap that splits the first time someone twists it hard: these are all the same failure, and none of them is a fault in the balm. The plastic is being attacked by the volatile fraction of your fragrance. The fix is a different polymer, and it is a fix rather than a compromise, because the product inside is unchanged.
Fine cracks at the thread and base are environmental stress cracking. Terpenes such as limonene and menthol diffuse into the plastic and lower the stress needed to open a craze, so moulded-in strain that was harmless on day one becomes a crack by week six. Polystyrene, SAN and PET are vulnerable, polypropylene and HDPE are not. Repack, do not reformulate.
Crazing, cracking and haze are one process at three stages
A craze is not a crack. It is a thin layer of material that has pulled apart into microscopic voids bridged by drawn polymer fibrils, and because those voids scatter light, a crazed region turns white or silvery. That is the stress whitening you see when you flex a plastic part just past its limit. Under continued load the fibrils break, the voids join up, and the craze becomes a true crack that runs.
The haze is a related but separate symptom. When a small aggressive molecule dissolves into the surface of a clear polymer it swells and plasticises it, and light scatters from the disturbed surface layer, so a tube goes cloudy without a single crack in it. Cloudiness on the inside of a clear pack can also be balm that has bloomed onto the wall, which is a formulation matter rather than a packaging one, so check whether the haze wipes off with a fingernail before you blame the resin. It does not wipe off if the polymer is the problem.
The sequence for a fragranced balm in the wrong tube is consistent: haze first, usually within a fortnight in warm storage, then crazing at the highest-stress features, then cracks at four to twelve weeks, then the leaking tube a customer complains about. If you have found the leak before the crazing, look again with a bright light held behind the barrel.
Three ingredients, all of which have to be present
Environmental stress cracking needs stress, an active chemical, and time. Remove any one and it does not happen, which is why the same tube performs perfectly with an unfragranced balm and fails within a month with a citrus one.
Stress is mostly not applied by the user. It is frozen into the part during injection moulding, when polymer flows into a cool cavity, orients, and solidifies before it can relax. The residue of that shows up at predictable places: the gate where the melt entered, weld lines where two flow fronts met around a core, sharp internal corners, and the root of a moulded thread. A screw thread is a stress concentrator by geometry alone. Applied stress adds to it, from an over-torqued cap, a shrink sleeve, a tight retail display or the drop from a market table.
The chemical does not need to dissolve the polymer, and in most cases it cannot. It needs to be small enough to diffuse into the amorphous regions, and compatible enough to swell them, which lowers the local glass transition temperature and lets chains slide past each other under stresses that would otherwise be harmless. Nothing is consumed and no reaction occurs, so a solvent wipe that evaporates in a minute can still leave a part that cracks a week later.
Time is why this fault escapes a bench check. Diffusion into the wall is slow, so a tube filled on Friday looks flawless on Monday, and the customer finds the crack. Any test that runs shorter than about a fortnight will pass packs that fail in the market.
Why the terpenes and not the fats
The fatty part of a balm is chemically inert toward these plastics and, more to the point, physically too big to get in. A triglyceride is a molecule of around 850 to 900 g/mol with three long tails; it has almost no diffusivity in a solid polymer wall at room temperature. Limonene is 136 g/mol, mobile, and volatile enough to reach the plastic through the headspace as well as by direct contact. Menthol, camphor, eucalyptol, methyl salicylate, linalool and the light aromatic solvents carried in many fragrance oils are all in the same size class.
Chemical similarity finishes the argument. Polystyrene has a solubility parameter of roughly 18.6 MPa to the half power and limonene about 16.5 to 17, close enough that limonene does not merely craze polystyrene but dissolves it outright, which is exactly why it is used industrially to reduce waste expanded polystyrene. PET sits near 20.5, far enough away to avoid dissolution but well within the range where terpenes act as stress cracking agents.
Solubility parameters alone would predict that polypropylene, at roughly 16.3, should be the most vulnerable material of all, since it is the closest match to limonene. It is not, and the reason is crystallinity. Polypropylene is 40 to 60% crystalline and polyethylene 60 to 80%, and terpenes cannot enter a crystalline lamella. They swell only the amorphous fraction, which softens the wall slightly without giving a craze anywhere to open. Glassy amorphous polymers such as polystyrene, SAN and largely amorphous PET have no such protection.
Which polymer to trust
| Polymer | Typical use | Glass transition | Terpene resistance | Weak point |
|---|---|---|---|---|
| Polypropylene (PP) | Twist-up barrels, screws, caps, jars | -10 to 0 C | Excellent, swells slightly at high loads | Brittle in the cold, see below |
| HDPE | Squeeze tubes, deodorant bodies, closures | About -110 C | Excellent against terpenes | Classic stress cracking with surfactants, not with oils |
| LDPE | Soft squeeze tubes, liners, plugs | About -110 C | Good, but permeable, so fragrance escapes | Scent loss and slow weight loss through the wall |
| PET | Clear jars and bottles | About 78 C | Poor with terpene-rich products | Crazes at the base and neck, hazes overall |
| Polystyrene (PS) | Cheap clear caps, clear jars, sample pots | About 100 C | Very poor, limonene dissolves it | Cracks at threads within weeks |
| SAN | Clear jars and cap overshells | About 105 C | Poor, better than PS but still styrenic | Crazing at weld lines and gates |
| Copolyesters (PETG, PCTG) | Clear jars sold as glass alternatives | 80 to 90 C | Varies widely by grade | Never assume, always test the grade you bought |
| Aluminium and tinplate | Tins, slider tins, aluminium tubes | Not applicable | Immune | Internal lacquer and seams, plus rust |
The practical reading is short. A twist-up lip balm tube in unpigmented polypropylene, which is what most of the trade supplies, is not at risk from anything in a normal balm. Risk enters when someone chooses a clear pack for looks, or when a cap is a different resin from the barrel, which is common and rarely disclosed. The trade-offs between formats are laid out in the packaging guide and, for the specific choice, under lip balm tube versus tin.
Where the crack is tells you what caused it
| Where the damage is | What it means | Next move |
|---|---|---|
| Fine circumferential cracks in the thread root | Moulded-in stress plus an active chemical, worsened by cap torque | Change resin, and check how hard the capping head is set |
| Whitening or splits around the base and the screw housing | Highest residual stress in the part, the usual first failure | Change resin, run the soak test on the replacement |
| A straight line up the barrel | A weld line where two flow fronts met, opened by the chemical | Moulding fault plus chemical attack, take it back to the supplier |
| Cracks only on the cap, barrel intact | Cap is a different, more vulnerable resin than the barrel | Ask for the resin of every component, not just the body |
| Crazing on the outside surface only, near print or a label | Solvent from an ink, adhesive or a cleaning wipe | Switch to water-based or UV-cured print, stop solvent wiping |
| Uniform haze, no cracks, product still sound | Absorption and swelling, the early stage of the same process | Treat as a fail, it will crack given time |
| One clean sharp fracture, no whitening, after a knock | Cold brittle fracture, a different failure entirely | See the section on cold cracking below |
The thirty day soak test
Compatibility is a property of the pack and the product together, so it can only be tested as a pair. This protocol answers the question with nine tubes and an oven.
- Fill three tubes with the finished product. Use production packs from the batch you will actually buy, filled the way you fill, because both the resin lot and the fill temperature matter.
- Fill three with the same base unfragranced, as the control that isolates the fragrance from everything else in the formula.
- Fill three with the base carrying twice the fragrance load. If the doubled arm fails and the production arm passes, you have found your margin, and you know how much headroom a supplier change would eat.
- Cap to your normal torque, weigh each tube, and record the weight. Weight loss over the month measures permeation through the wall and closure, which is a second and quieter packaging failure.
- Hold thirty days at 40 C in the dark, the standard accelerated condition described under accelerated ageing. Keep three more tubes of the production fill at room temperature as the real time arm.
- Cool to room temperature before you assess. Warm polymer is more ductile and hides crazes that reappear on cooling. Give it four hours.
- Weigh again, then flex the base about twenty degrees and look along the surface at a low angle. Stress whitening that was not there at the start is a fail, whether or not anything has cracked. Then uncap and note whether the cap needs noticeably more or less force than a fresh one.
Thirty days at 40 C is a screen, not a shelf life. It reliably separates a pack that will fail from one that will not, and it does not tell you that a passing pack survives two years, which is why a room temperature arm runs alongside and gets checked at six and twelve months. The wider twelve week protocol, covering liners, labels and closures as well as the barrel, is set out under packaging compatibility, and the results belong in your batch records with the resin lot number attached.
For a same-week answer, borrow the bent strip principle from ISO 22088-3. Cut a strip from a spare barrel of the resin in question, bend it around something round to hold it under a fixed strain, wet the outside of the bend with the neat essential oil, and leave it at room temperature. A susceptible polymer whitens along the bend within hours to two days. It overstates severity, because neat oil is harsher than a balm, so treat a pass as encouraging and a fail as a reason to run the full thirty days.
How much fragrance is too much
There is no threshold that holds across every resin lot, but the pattern is consistent enough to plan around. Below roughly 0.3% of a citrus or mint oil, stress cracking failures in styrenic and PET packs are uncommon though not impossible over a two year date. Between about 0.3 and 1% they become sporadic and lot dependent, which is the worst case commercially because they pass your samples and fail in the market. Above about 1% they are the expected outcome. In polypropylene the same loads are unremarkable, and problems only appear at the concentrations used in solid perfume, where 10 to 25% fragrance will swell and soften even a PP pack over months.
Lip balms are usually protected by their own regulation, because IFRA Category 1 constituent limits keep most essential oil loads well under 1%, as set out under essential oils in balms. The high-terpene products in a typical range are the ones nobody thinks of as fragranced: a muscle rub or chest rub carrying menthol and camphor at several percent, a peppermint lip product pushed for the tingle, a moustache wax scented for effect. Those are the packs to test. The actives and their own limits are covered under menthol, camphor and actives.
Fragrance oils are not automatically safer than essential oils here. Many are diluted in solvents such as dipropylene glycol or benzyl benzoate, and some of those are more aggressive to styrenics than the aroma chemicals they carry. Ask the supplier for a packaging compatibility statement, and treat its absence as a reason to test rather than a reassurance. The composition question is covered under fragrance oils in balms.
Rescuing the batch
The balm is not the casualty here, and that changes the economics of the decision completely.
- Quarantine the whole lot, including packs that look sound. Stress cracking arrives at different times across a batch because moulded-in stress varies from cavity to cavity, so an intact-looking tube from the same delivery is a fail waiting to happen.
- Recover the product. Wind the sticks out and cut them off, or scrape jars, into one clean vessel. Keep anything that has leaked out of a cracked tube separate and discard it.
- Remelt and repack. Melt gently, hold only long enough to get a uniform liquid, and add essential oils, flavour and tocopherol at the end below about 50 to 60 C so you are not driving off what you paid for. Pour into polypropylene at the temperatures given under pour temperatures, following the technique under filling lip balm tubes.
- Restate the date honestly. The oils have aged and been through a second heat cycle, so the best-before runs from the original manufacture, not from the repack.
Do not sell packs that show stress whitening, even if none has cracked or leaked. A crazed wall is mechanically compromised and will open in a warm bag or a coat pocket. There is a second reason to repack rather than ship: a polymer that has absorbed enough terpene to craze has also had its additive package extracted into your product to some degree, and a food-contact grade resin is only assessed for migration under the conditions it was tested for. The exposure is likely small, the uncertainty is not worth carrying, and the tubes are the cheapest component in the product.
Cold cracking is a different failure with a different fix
A tube that arrives in January in two clean pieces, with no whitening and no haze, has not been stress cracked. It has fractured in the brittle regime. Polypropylene, the material this page has been recommending, is the one with the problem: its glass transition sits between roughly -10 and 0 C, so a PP part loses most of its impact strength in a cold van and shatters where the same part would deform in July.
The distinguishing signs are useful. Chemical stress cracking produces slow, whitened, often multiple cracks at the highest-stress features, and it appears after weeks in storage. Brittle fracture produces one clean sharp break with a glassy fracture surface, no whitening, and it happens at the moment of impact. If you can date the damage to a delivery rather than to a shelf, it is the cold.
The fixes are different too. Impact-modified polypropylene copolymer, which contains an ethylene phase, keeps useful toughness well below zero, and HDPE stays tough to around -100 C. Specify a copolymer grade if you ship into cold winters, add packaging that prevents a drop onto a doorstep, and ask the supplier for a brittleness temperature determined by ASTM D746 rather than accepting the word "durable". Winter transit damage sits alongside the other shipping faults collected under balm changed after shipping, and its summer counterpart under balm melting in the heat.
Prevention, and the questions to ask a supplier
- Specify polypropylene for barrels, screws and caps, and get the resin of every component in writing. A PP barrel with a PS cap fails at the cap.
- Never wipe a filled or empty pack with isopropanol or a solvent cleaner. The solvent evaporates and the damage does not.
- Use water-based or UV-cured inks and adhesives, and avoid tight shrink sleeves on a resin you have not tested. Related label failures are covered under labels peeling off tins.
- Set the capping torque to the closure supplier's figure and check it, rather than tightening by feel.
- Store empty packs away from solvents, and never next to opened essential oils. Vapour phase attack is real.
- Fill at the lowest temperature that gives a clean pour. A hot fill puts a thermal stress into the wall on top of the moulded-in stress.
- Retest whenever the supplier, the resin lot or the fragrance changes, which is part of what sourcing ingredients properly involves.
The decision rule, and what is left uncertain
If the product carries any appreciable citrus, mint or camphoraceous load, use polypropylene or HDPE and the question is closed. If you want a clear pack badly enough to accept the risk, run the thirty day soak test on the exact grade with the exact formula, keep a room temperature arm running for a year, and expect to repeat it whenever anything upstream changes. If a pack has already crazed, repack it. The balm is fine, the tubes are the cheapest thing in the bill of materials, and no amount of reformulating the balm will save a resin that was the wrong choice.
Two honest limits. The percentages in this page are trade experience rather than a published test series, because compatibility data is generated privately by packaging suppliers and rarely released; the polymer ranking is well established, the exact thresholds are not, and yours will differ. And a passing thirty day screen is evidence about failure, not a guarantee of shelf life, since stress cracking is a statistical process driven by a distribution of moulded-in stresses that varies between cavities and between lots. That is why the answer to "is this pack compatible" is always a test you ran on the pack you bought, not a table.
Frequently asked questions
Why did my peppermint lip balm crack the tube but the plain one did not?
Menthol and the terpenes in peppermint oil are small, mobile molecules that diffuse into the plastic, swell its amorphous regions and lower the stress needed to start a craze. The fats in the balm are far too large to do that. So the same tube, the same moulding stress and the same shelf give a failure with one product and none with the other.
Is balm from a cracked tube still safe to use?
The balm itself is chemically unaffected by the cracking, and remelting it is safe. The caution is that a polymer which has absorbed enough fragrance to craze has also been in intimate contact with your product for weeks under conditions its food-contact assessment did not cover. Recover the product, discard anything that leaked out, and repack into polypropylene.
Which plastic should a lip balm tube be?
Polypropylene, for the barrel, the internal screw, the elevator and the cap. It resists terpenes because it is 40 to 60% crystalline and the crystalline regions are impenetrable to them. HDPE is equally resistant and is used for squeeze formats. Polystyrene, SAN and PET craze and crack with fragranced products and should be treated as clear-pack luxuries that need testing.
How long does a packaging compatibility test need to run?
Thirty days at 40 C in the dark is the minimum useful screen, with a room temperature arm running alongside for at least a year. Anything shorter passes packs that fail in the market, because diffusion into the wall is slow. A bent strip test with neat oil gives a same-week indication but overstates severity.
Can I stop the cracking by reducing the essential oil?
Sometimes, and it is the wrong lever. Failures become sporadic rather than absent somewhere below about 1% in a vulnerable resin, which means they pass your samples and appear in customers' bags. Changing the tube to polypropylene removes the mechanism completely and costs less than reformulating a scent you already sell.
My tubes cracked in transit in winter with no whitening at all. Same problem?
No, that is brittle fracture. Polypropylene has a glass transition between roughly -10 and 0 C, so it loses impact strength in a cold van and breaks cleanly on impact rather than crazing slowly on a shelf. Look for one sharp break and a glassy fracture surface. The fix is an impact-modified copolymer grade and better transit protection.
Sources and further reading
- ASTM International, ASTM D1693, Standard test method for environmental stress-cracking of ethylene plastics, West Conshohocken.
- International Organization for Standardization, ISO 22088, Plastics: determination of resistance to environmental stress cracking, in particular Part 3, bent strip method.
- ASTM International, ASTM D543, Standard practices for evaluating the resistance of plastics to chemical reagents.
- ASTM International, ASTM D746, Standard test method for brittleness temperature of plastics and elastomers by impact.
- European Union, Regulation (EC) No 1223/2009, Annex I Part A, cosmetic product safety report, which requires information on the relevant characteristics of the packaging material, including purity and stability.
- Wright, D. C., Environmental Stress Cracking of Plastics, Rapra Technology, Shawbury, 1996.
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.