Oil weeping out of the bottom of a tube and pooling inside the cap
A hot fill, a soft stick and a threaded elevator give oil a path out of the base. A tube leak test, the fill window that seals it and the wax change to make.
A leaking twist-up tube shows itself in two places: a greasy ring on whatever the tube was standing on, or a pool of oil inside the cap. They are different faults with different fixes, and confusing them wastes a reformulation on a problem that was created at the pour. This page separates them, gives you a test you can run tonight on three tubes and a bottle of sunflower oil, and sets out the fill window that closes the path.
Most base leaks are balm that wicked past the elevator disc into the screw thread while it was still liquid, poured above about 75 C. Fill at 65 to 72 C with the elevator wound fully down, never above 78 C, and cool upright and uncapped for an hour. Oil in the cap is usually sweating or transit heat, not a leak.
The tube has no seal, and it was never meant to
A standard twist-up lip balm tube is three injection-moulded polypropylene parts. The barrel carries an internal thread or a pair of guide slots. Inside it rides the elevator, a flat disc on a short post. Underneath sits a base that the user turns, keyed through an opening in the bottom of the barrel. That opening is not a defect. It is how the turning motion gets in, and every tube of this design has one.
Nothing in the assembly is a gasket. The elevator disc is a wiper, moulded a few hundredths of a millimetre proud of the bore so its rim drags on the wall. That interference is the only thing between the balm above it and the open base below. A wiper holds back a solid perfectly well. It does not hold back a thin liquid, and no tube maker claims it does. So a permanent path exists from the top of the stick, around the disc, down the thread helix and out through the base. Everything else on this page is about whether liquid ever reaches the underside of that disc, and about what happens when a set stick turns liquid again later.
Two consequences follow immediately. First, a tube is a worse container than a tin for any formula that is soft or oil-heavy, which is part of the trade-off set out in tube against tin. Second, the pour is a sealing operation as much as a filling one, and the sequence on filling lip balm tubes exists largely to keep liquid out of the mechanism.
Why 78 C leaks and 68 C does not
Two physical quantities decide whether balm gets into the thread: the pressure pushing it and the time it stays fluid.
The pressure is small and easy to calculate. A 60 mm column of molten balm at roughly 0.86 g/mL presses on the elevator disc with about 500 Pa, which is density times gravity times height. That is a fraction of a percent of atmospheric pressure, and on its own it would be harmless. It stops being harmless because it acts on the whole annulus around the disc and it acts continuously for as long as the balm is liquid.
The travel is described by the Washburn relation for liquid entering a narrow gap: penetration distance goes as the square root of surface tension times gap width times elapsed time, divided by viscosity. Three of those terms are out of your hands. Molten wax and oil blends sit around 25 to 33 mN/m in surface tension and the surface energy of polypropylene is about 29 to 31 mN/m, so the melt wets the plastic readily rather than beading off it. The gap is whatever the moulder's tolerances left. Viscosity is the term you set at the bench, and it is also the term that changes fastest, because it climbs steeply the moment the highest-melting wax in the formula begins to nucleate.
Penetration goes with the inverse square root of viscosity, so doubling the viscosity does not halve the leak, it cuts the distance by roughly 30%. The larger effect is on the time term. Pour ten degrees cooler and the balm gels within seconds of touching the disc instead of staying mobile for a minute, and a term that collapses toward zero beats a term that improves by a third. That is why the gap between a 78 C fill and a 68 C fill is not a marginal preference. The full set of windows, and what each end of them costs, is on pour temperatures.
Very hot balm does two things at once. It flows into the thread, and it warms a thin polypropylene wall enough to relax the moulded-in stress in it, which can distort the bore slightly around the disc. A tube that has been filled at 85 C may leak on the bench and then twist stiffly for the customer, because the same fill caused both. Treat 78 C as a ceiling with no upside above it.
Leak, sweat or melt: three faults that look alike
Before changing anything, work out which one you have. The single most useful discriminator is timing: a fill leak appears within hours of pouring, sweating appears over days to weeks, and heat damage appears after a journey or a warm week.
| What you see | Mechanism | Confirm by | First fix |
|---|---|---|---|
| Thin clear film on the base within hours of filling | Balm wicked past the disc into the thread while still liquid | Check whether the first and hottest tubes of the run are the worst affected | Chill to full set, wipe, drop the pour to 66 to 70 C |
| Greasy ring under a tube that is weeks old, stick intact | Syneresis: liquid oil above what the wax network can hold, migrating down the barrel | Wipe clean, hold at 30 C for 24 hours and see whether it returns | Raise the structurant 1 to 2 points |
| Oil pooled in the cap, dome of the stick dull or dished | Capped before the stick was cold, so the warm oil film transferred to the cap | Leave three tubes uncapped for two hours and compare | Cap at room temperature, one to two hours after pouring |
| Whole stick slumped or leaning, oil throughout the packaging | Sustained heat above the softening range of the wax network | Ask what the parcel's journey was, not what the formula is | Reformulate the melting profile, then fix the logistics |
| Leakage from one carton only, all units affected | Transit heat plus a parcel lying flat | Compare a retained sample kept on your own shelf | Dispatch policy and packing orientation |
| Oil in the cap threads on a heavily flavoured batch | Low-viscosity flavour and ester carriers migrating out of the network first | Make a control batch without the flavour phase | Cut the mobile phase or bind it with a second wax |
If the middle two rows describe your batch, this is not the page you need. The mechanism, the arithmetic and the industry fix for oil leaving a set balm are on balm sweating, and the case where the stick itself gives way is on balm melting in the heat.
The tubes themselves are not all the same
Tube quality is the variable makers most often ignore and suppliers least often specify. Four things differ, and all four bear on leaking.
Bore roundness and disc interference. The wiper only works if the disc is round, the bore is round, and the difference between them is small and consistent. A worn or poorly cooled tool gives an oval bore, and an oval bore leaks on two sides while sealing on the other two. This is also the single largest source of unit-to-unit variation within one bag of tubes.
Flash on the disc rim. A thin skirt of excess plastic left at the parting line breaks the contact between rim and wall into a series of channels. You can often see it by winding a fresh tube fully up in good light.
Wall thickness. Typical commercial barrels run about 0.6 to 1.0 mm. A thin wall flexes when the tube is gripped, momentarily opening the annulus, and it relaxes more readily under a hot fill.
Base construction. Some designs use a one-piece base that snaps into the barrel skirt, others a two-piece base with a separate turning collar. The two-piece designs have an extra joint at exactly the place where escaping oil arrives, and they are more often the economy option.
Polymer choice matters too, though less for leaking than for cracking. Polypropylene is the normal barrel material and it tolerates terpene-rich formulas reasonably. Polystyrene and PET caps and barrels do not, and the fine cracks they develop are a separate fault covered on tubes cracking or crazing. If you are choosing between components rather than diagnosing a batch, start from the packaging guide and run the compatibility protocol on packaging compatibility before committing to a thousand units.
The three-tube oil test
This separates a tube problem from a fill problem in 24 hours, and it costs three tubes.
- Take three tubes from the bag you are actually using. Not a sample the supplier sent separately. Wind every elevator fully down and check it is seated.
- Fill each with about 4.5 g of a plain liquid oil at room temperature. Sunflower or fractionated coconut is ideal. Do not warm it, and do not use your balm: the point is to present the tube with a liquid it can never seal against.
- Cap two, leave one uncapped, and stand all three upright on white paper. Put the paper somewhere that holds about 30 C, which is a warm windowsill or the top of a fridge in most houses.
- Read at one hour, at 24 hours and at 72 hours. Mark the outline of each tube on the paper so you can tell a stain from a smear you made handling it.
- Repeat with one tube lying flat if you post your balm. That is the transit case, and it fails differently.
| Result | What it means | What to do |
|---|---|---|
| Stain inside one hour | The wiper is not making contact at all on at least one tube | Reject the batch of tubes, and ask the supplier for the moulding lot |
| Clean at one hour, ring by 24 hours | Normal for this test. The tube is marginal against a free liquid | Usable, but only with a cool fill and a firm formula |
| Paper still dry at 72 hours | The hardware is sound | Any leak you have is your pour temperature or your formula |
| Upright dry, horizontal tube stains | Head pressure on the barrel wall rather than the disc | Ship and store upright, and firm the stick for transit |
| Capped tubes wet inside the cap, base dry | Not a leak at all: oil is climbing the barrel and pooling at the cool end | Treat as sweating, not as a tube fault |
Rescuing the batch on your bench
Remelting is safe. Waxes and refined vegetable oils are not chemically changed by a second pass at 70 to 80 C, so a rescued batch is not a compromised batch. The two situations where it is not safe are a batch that already smells rancid, where you will simply produce smooth rancid balm, and a batch carrying botanical material or a heat-sensitive active that will not survive a second melt. Everything else can be reprocessed, and the arithmetic for the wax change is below.
- Stop the travel. Put the whole tray in a fridge at 4 to 6 C for 20 to 30 minutes. Once the balm is fully set, nothing more moves into the thread, and you are diagnosing a fixed situation rather than a worsening one.
- Wipe every base dry. A paper towel, not alcohol. Solvent on the base carries oil further into the mechanism. Save the alcohol wipe for the outer barrel where a label goes.
- Stand them uncapped on clean paper for 24 hours at room temperature. This is the sorting step. Tubes that stay clean are finished product. Tubes that stain again have balm sitting in the thread and will keep weeping.
- Decant the failures. Wind each stick fully up, cut it off level with the rim, then warm the remainder to about 40 C and tip it out. Do not reuse those tubes: the thread is full of set balm and cleaning it properly takes longer than the tube is worth.
- Add 1 to 2 points of wax before repouring. For a 100 g batch at 12% wax, adding x grams to reach 14% means (12 + x) divided by (100 + x) equals 0.14, so x is 2.3 g. The batch becomes 102.3 g at 14.0% wax. If you would rather hold the batch weight, remove 2 g of liquid oil and add 2 g of wax instead. The wax ratio calculator does this for any starting point.
- Repour at 66 to 70 C into fresh tubes. Elevators fully down, spout 5 to 10 mm above the rim, one continuous stream, fill 1 to 2 mm below the rim rather than to the brim.
- Cool upright and uncapped for at least an hour, then cure 24 to 48 hours before boxing. Record the pour temperature against the batch code so the next investigation has data. The format is on batch records.
Pour two tubes at the top of your window and two at the bottom, from the same jug, and mark them. Stand all four on paper at 30 C for 24 hours. If only the hot pair stains, you have proved the mechanism in a single afternoon and you never have to argue about it again.
The fill window, band by band
| Pour band | C | At the disc | Net result |
|---|---|---|---|
| Slurry | Below 62 | Melt too thick to reach the annulus, leaving a void under the stick | No leak now, a hollow the user finds when winding down |
| Cool | 62-65 | Fills fully in a wide bore, marginal in a narrow one | Safe against leaking, some risk of flow lines |
| Target | 65-72 | Wets the disc, gels before it travels far down the thread | The working window |
| Margin | 72-78 | Mobile for tens of seconds, reaches the first turn of the thread | Tolerable in a good tube, leaks in a loose one |
| Over | Above 78 | Free-running past the disc, plus stress relaxation in a thin barrel | Base leak, stiff twist, or both |
Three habits at the bench matter as much as the number. Wind every elevator fully down before pouring, because a disc sitting 2 mm up costs fill weight and leaves a void. Warm cold tubes to 30 to 35 C rather than raising the pour temperature, since that removes the cold wall without adding superheat. And do not go back to top up a tube whose surface has already set, because a top-up disc that shears off later is the fault described on tunnelling and dips.
Where most complaints actually originate
A stick that is sound at 22 C is not the same object at 40 C. The yield stress of a wax network falls long before anything melts, so a balm that behaves as a solid on your shelf behaves as a very stiff paste in a delivery van, and the weight of the stick above the disc is then enough to press material into the annulus. The relationship between temperature and the strength of that network is set out on rheology and yield stress.
Add orientation. A parcel lying flat puts the length of the stick against the barrel wall and points the free surface at the cap. That is why the classic customer complaint is oil in the cap rather than oil on the base, and why the same batch that survives on your own shelf fails after a journey. Standard test practice reflects this: conditioning and transport thermal protocols such as ASTM D4332 and ISTA 7D exist precisely because ambient laboratory storage does not represent distribution.
Nothing at the fill bench fixes this. The levers are formulation and logistics: raise the melting profile with a point or two of a high-melting wax from the comparison on waxes compared, dispatch early in the week so nothing sits in a depot over a weekend, and be honest on the listing about summer deliveries. The full dispatch policy is on shipping in hot weather, and what a journey does to a product that arrives intact but different is on balm changed after shipping.
The decision rule, and what it cannot settle
Four outcomes, four different owners of the problem. If a tube leaks plain oil at 30 C, it is the tube, and no fill temperature will rescue it. If it leaks within hours of pouring and the hottest tubes of the run are worst, it is the pour. If it weeps weeks later at room temperature with the stick intact, it is the formula's oil binding capacity and wax is the lever, which usually also means the stick was on the soft side to begin with, as diagnosed on balm too hard or too soft. If it only fails after a journey, it is heat, and the answer is a different melting profile rather than a different technique.
What this page cannot decide for you is which supplier's tubes hold. Moulding tolerances are not published, they vary between lots from the same supplier, and no standard exists for the leak resistance of a cosmetic twist-up tube. The only reliable knowledge is the test you run yourself, on the bag in front of you, with the balm you actually make, at the temperature you actually ship to. Run it once per supplier and once per new formula, and keep the paper.
Frequently asked questions
Why is oil leaking out of the bottom of my lip balm tubes?
Almost always because the balm was poured hot enough to stay liquid after it passed the elevator disc, so it ran into the screw thread underneath and out through the opening in the base. Pour at 65 to 72 C with the elevator wound fully down, and treat 78 C as an absolute ceiling. A loose or oval disc fit in a cheap tube makes the same fill leak when a better tube would not.
Is a leaking tube of lip balm still safe to use?
Yes. The oil that escapes is the same oil that was in the product, and nothing about the leak makes the balm unsafe. What you lose is fill weight, presentation and, if you are selling it, the customer's confidence. The exception is a batch that smells sharp or waxy, which points at oxidation rather than a leak and is a reason to discard it.
How do I tell a leak from sweating?
Timing and location. A fill leak appears within hours of pouring, comes from the base, and is worst on the tubes you poured first when the melt was hottest. Sweating appears over days or weeks as beads on the stick or a film that creeps up the barrel into the cap, and it comes back after you wipe it away. Sweating is fixed with wax, leaking with temperature.
Can I remelt lip balm that leaked and pour it again?
Yes. Waxes and refined oils are unchanged by a second melt at 70 to 80 C, so a reprocessed batch is not a degraded one. Wind the sticks up, cut them off, warm the remainder to about 40 C and tip it out, then remelt the lot together. Do not reuse the tubes that leaked, because their threads are full of set balm.
What temperature should I fill lip balm tubes at?
65 to 72 C for a standard oval twist-up tube, measured with a probe in the melt rather than an infrared reading off the surface. Below about 62 C the melt is a slurry and will not fill around the elevator disc. Above 78 C it stays liquid long enough to run into the thread, and can also relax the moulded stress in a thin barrel wall.
Why is there oil inside the cap but the tube looks fine?
Two common causes. Either the tubes were capped while still warm, so the oil film on the dome transferred to the cap, or oil is migrating out of the stick and collecting at the coolest point in the pack. Cap at room temperature an hour or two after pouring. If it comes back on cold tubes, treat it as sweating and raise the structurant by one to two points.
Will adding more wax stop tubes leaking?
It stops the slow kind. One to two points more wax raises the oil binding capacity of the network, which addresses weeping weeks later and improves survival in transit. It does nothing about a fill leak, because at 78 C the extra wax is molten too. Fix the pour temperature first, then use wax to deal with what is left.
Sources and further reading
- Washburn, E. W., The dynamics of capillary flow, Physical Review, 17(3):273-283, 1921.
- ASTM International, ASTM D4332: Standard Practice for Conditioning Containers, Packages, or Packaging Components for Testing, West Conshohocken, PA.
- ASTM International, ASTM D4169: Standard Practice for Performance Testing of Shipping Units, West Conshohocken, PA.
- International Safe Transit Association, ISTA 7D: Temperature Test for Transport Packaging, East Lansing, MI.
- International Organization for Standardization, ISO 22716: Cosmetics, Good Manufacturing Practices (GMP), Guidelines on Good Manufacturing Practices, 2007.
- European Parliament and Council, Regulation (EC) No 1223/2009 on cosmetic products, EUR-Lex.
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.