Why Cosmetic Bottles Leak, Crack or Craze

Last updated: 1 September 2026.

Quick answer: Almost every bottle failure traces back to one of four causes: a dimensional mismatch between bottle and closure, a material that is incompatible with the formulation, a process defect in the molding or sealing step, or damage introduced during filling and transport. Which of those it is determines who pays — and the answer is usually discoverable in an afternoon if you keep the failed samples and the production records. The reason this matters commercially is that a leaking bottle does not cost you a bottle; it costs you a filled product, a retailer relationship and, in the worst case, a recall.

What are the failure modes, and what does each one look like?

Failure	What you see	Most common cause
Leak at the neck	Product around the closure, wet cartons	Neck finish out of tolerance, or closure torque / liner mismatch
Leak through the wall	Product weeping through the body	Insufficient wall thickness, or pinholes from contamination
Cracking under load	Brittle fracture, often at the shoulder or base	Wrong resin grade, or environmental stress cracking
Crazing / stress whitening	Milky haze where the plastic was stressed	Material stressed beyond its limit by formulation or assembly
Paneling	Bottle walls suck inwards	Hot filling into a bottle not designed for it, or vacuum from cooling
Pump or sprayer failure	No output, or leaking at the crimp	Closure specification mismatch, or assembly torque
Base blow-out	Rupture at the base	Drop impact on a cold or over-thin bottle

Each of these has a different owner. The mistake most buyers make is treating all of them as “supplier quality problems”, which delays the fix and often points the blame at the wrong party.

Why does a bottle leak at the neck?

This is the most common complaint and, in our experience, the least often the bottle’s fault.

A closure seals against a neck finish through three surfaces working together: the sealing surface (the top of the neck or the bore), the thread engagement, and the liner or wad inside the closure. A leak appears when any one of them is out of specification:

Neck finish out of tolerance — if the bottle’s neck is at the edge of the tolerance band, a closure that is also at its edge will not seal even though both parts are technically “in spec”.
Torque — under-torqued closures back off in transit; over-torqued closures distort the sealing surface or strip threads. Torque is an application parameter set on your filling line, not a property of the bottle.
Liner compatibility — an induction-seal liner, a foam liner and a cone liner behave differently against different formulations. Choosing the liner without testing against the actual product is a common and avoidable error.
Thread and neck standard mismatch — a 24/410 closure on a 24/415 neck looks fine and does not seal properly. Neck finish standards are specified in how to choose a cosmetic bottle manufacturer.

If the leak is at the neck, measure the neck finish, measure the closure, and check the application torque before concluding anything about the bottle body.

What is environmental stress cracking, and why does it matter for HDPE?

Environmental stress cracking — usually shortened to ESC or ESCR — is the failure mode that most often surprises buyers, because the bottle looks fine when it leaves the factory.

The mechanism: a plastic under modest mechanical stress, in contact with a particular chemical, develops cracks that would not occur from either the stress or the chemical alone. In HDPE bottles, surfactants, oils and certain preservative systems are classic agents. The bottle survives filling, sits on a shelf, and fails weeks later.

Three things reduce the risk, and all three belong in the specification:

Resin grade with adequate stress-crack resistance — grades are not interchangeable, and the grade is a cost line, which is one reason two quotes for the same bottle differ.
Wall thickness and design that limit stress concentration — sharp internal corners and thin shoulders concentrate stress.
Compatibility testing with the real formulation — the only reliable way to know. This is covered in detail in what a resin COA and migration test should show.

If a bottle fails weeks after filling with no obvious cause, suspect ESC before suspecting the molding.

When is it a material problem rather than a molding problem?

The distinction matters because the fixes are completely different.

Material problems tend to present as:

Failure appearing across an entire lot, not randomly
Cracking or crazing in the same location on multiple bottles
A clear correlation with a specific formulation or filling temperature

Molding process problems tend to present as:

Random distribution within a lot
Visible defects — short shots, flash, weld-line weakness, contamination
Failure concentrated at a specific point in the production run

Transport and handling problems tend to present as:

Damage concentrated at the outside of pallets or the bottom cartons
Drop or compression damage with an obvious impact point
Failures that correlate with a specific shipment rather than a production batch

Keep the failed samples and note the carton and pallet position. That information alone often identifies the cause.

Who is responsible — and how do you establish it without a fight?

Responsibility follows the cause, and the way to establish the cause is evidence rather than assertion.

Cause	Usually responsible	Evidence that settles it
Neck finish out of tolerance	Bottle manufacturer	Dimensional report on retained samples
Closure not matched to neck	Whoever specified the closure	Neck and closure specifications
Application torque wrong	Filler	Torque records from the capping head
Material incompatible with formula	Whoever selected the resin without testing	Compatibility test results
Liner incompatible with product	Closure or liner specifier	Liner specification and test data
Transit damage	Carrier or packing spec	Carton condition, pallet position, drop-test data
Design cannot survive the fill	Product designer	Fill temperature and vacuum behaviour

Two things prevent most disputes before they start. First, agree the specification in writing — neck finish standard, gram weight, resin grade, torque range, drop-test criteria. Second, retain samples from every production lot. A factory running ISO 9001:2015 process control will have lot records and retained samples; a factory that cannot produce them is telling you something important about how the next problem will be handled.

The vetting questions that predict this behaviour are in 7 red flags when sourcing cosmetic bottles from China.

How do you prevent failures rather than litigate them?

Four steps, in order of cost-effectiveness:

Test the real formulation in the real bottle. Not water, not a proxy — the actual product. This is the highest-value hour you can spend, and it is covered step by step in the resin and migration documents to request.
Specify and verify the neck finish and torque range, then measure torque on the line at the start, middle and end of each run.
Agree a drop-test and compression standard before the first shipment, and test to it. This catches the packing spec problems that cause most transit damage.
Run a pilot shipment before the full order. A single pallet through the real distribution channel will find problems while they are still cheap.

None of these add meaningful cost. All of them are cheaper than a single failed production run.

Does the material choice change the failure profile?

Yes, substantially. Material selection is the first risk decision, not the last.

HDPE — tough and chemically resistant to many formulations, but susceptible to environmental stress cracking with certain surfactants and oils, and opaque by default.
PET — excellent clarity and good barrier, with a different compatibility profile and a lower tolerance for hot filling without design accommodation.
PETG — clarity with better impact behaviour than PET in some applications, at a higher material cost.
PP — good chemical resistance and temperature tolerance, commonly used for closures and jars.

The choice is a compatibility decision as much as a marketing one. The full comparison is in plastic vs glass cosmetic bottles, and the regulatory dimension is covered in FDA cosmetic packaging rules.

The decision, in one paragraph

When a bottle fails, resist the reflex to call the supplier. Identify the failure mode first — leak at the neck, weep through the wall, stress cracking, paneling, transit damage — because each one has a different owner and a different fix. Specify the interface (neck finish, closure, liner, torque range) in writing, test your real formulation in the real bottle before the production order, retain samples from every lot, and you will resolve the occasional failure in an afternoon instead of an argument. For the specification discipline that prevents most of these problems, see how to choose a cosmetic plastic bottle manufacturer in China.

Frequently asked questions

My bottles leaked after filling but passed the factory’s leak test. Why?

Factory leak tests are usually performed on empty bottles with air or vacuum. Your product’s surface tension, viscosity and chemical behaviour are different, and application torque is applied by your line, not the factory. Test filled bottles.

Is a leak always the bottle’s fault?

No. In our experience neck-finish and closure-interface issues — including torque and liner selection — account for the majority of leak complaints, and several of those are set by the filler rather than the bottle maker.

How long should I wait before I can rule out stress cracking?

Stress cracking can appear weeks after filling. A compatibility test using elevated temperature is designed to accelerate it; see accelerated migration testing and what it proves.

What is stress whitening, and is it dangerous?

It is visible micro-crazing where the material has been stressed past its limit. It is both a cosmetic defect and a warning — the same area is a likely crack initiation point.

Can I claim against the supplier for a failed shipment?

Only if the failure is attributable to a specification the supplier did not meet. That is why written specifications and retained samples matter: they turn a claim into a measurement.

Do heavier bottles fail less?

Generally yes, up to a point — more material means more margin against impact, paneling and stress. Gram weight is the most direct lever, and the most direct cost line. See why two quotes differ by 30%.

Next step

Tell us your formulation type, fill temperature and closure specification, and we will recommend a resin and neck finish combination that has run the same application before — and quote to a written spec you can hold us to.

Shijin Packaging — factory-direct cosmetic and daily-chemical plastic bottles since 2003, operating 30+ automatic blow-molding lines and 10+ injection-molding machines across a 15,000 m² facility in Huizhou, Guangdong, with daily capacity of approximately 200,000 pieces.

Website: https://shijinpackaging.com
Full catalogue: https://shijinpackaging.com/catalog/
Email: sales@shijinpackaging.com