Cracked and crumbly balm: surface cracks, crumbling and snapping sticks
Surface cracks come from cooling stress, crumbliness from too much hard wax. A crack type table, percentage fixes and cocoa butter brittleness.
Cracking and crumbling look like one problem and are two. A crack is a stress fracture in a balm that is otherwise sound, and it is almost always a cooling fault. Crumbliness is a property of the material itself, and it is almost always a formula fault. Telling them apart before you change anything saves a lot of wasted test batches.
Cracks come from cooling: the surface sets, the interior keeps contracting, and the shrinkage stress fractures the rigid skin. Crumbliness comes from formula: too much hard wax and too little liquid oil or soft butter to plasticise the network. Cool slower and more evenly for cracks, cut hard wax for crumble.
Crack type, cause and fix
| What you see | Cause | Confirm by | Fix |
|---|---|---|---|
| Single crack radiating from the centre of a tin | Shrinkage stress pulling on a set skin | There is usually a dip at the crack's origin | Pour 4 to 5 C cooler, cool at room temperature |
| Fine crazing over the whole surface | Very fast surface cooling, often a draught or a fridge | Only the top 1 mm is affected | Cool covered by a box, not chilled. Reflow the surface |
| Crack around the perimeter where balm meets the tin | Contraction away from a cold metal wall | Often accompanied by pale wet spots | Warm the containers before filling |
| Balm breaks into chunks when scooped | Brittle network: too much hard wax for the oil | A chunk snaps cleanly rather than deforming | Cut hard wax 2 to 3 points, raise liquid oil |
| Powdery, dry crumbs on the fingertip | High carnauba or stearic acid, low plasticiser | Check the hard wax total against 1 to 3% | Reduce carnauba, add a viscous oil |
| Stick snaps off in the tube | Hollow column, or a brittle formula, or both | Cut a tube lengthways | Hollow: see tunnelling and dips. Solid: soften the formula |
| White or grey streaks and patches on a dark balm | Fat bloom from partial melting and recrystallisation | The marks melt away under a fingertip and resist water | Storage control, full melt and controlled cooling |
Why surfaces crack
A balm poured at 68 C into a 20 C tin has a steep temperature gradient across a few millimetres of depth. The top and the walls set first, forming a rigid shell whose dimensions are now fixed. The interior is still liquid and still contracting, because solid fat is denser than melt. That contraction pulls the shell inward. The shell is a brittle crystal network with essentially no ability to stretch, so it relieves the stress by fracturing, usually at the thinnest or most stressed point, which is the centre of a dip.
Everything that steepens the gradient makes it worse: a draught across the bench, a cold stainless surface under the tins, a fridge or freezer used to speed setting, and pouring hotter than necessary, which leaves more contraction to happen after the skin forms. Cracking and sink holes therefore travel together, and the fix is largely shared: pour in the correct window, cool evenly, and cover the tray loosely with a cardboard box to slow the surface without chilling anything. Pour windows for each format are in pour temperatures.
Set poured containers on wood, card or a folded towel, never directly on a cold metal or stone bench. The base is the largest cooled surface in a tin and it drives most of the gradient.
Why balms crumble
Crumbliness is ductility, or the lack of it. A balm that deforms under a fingertip has a crystal network with enough liquid oil between the crystals to let them slide past one another. A balm that fractures instead has a network so densely packed with rigid, high melting crystals that there is no lubricating phase to allow that movement. The liquid oil is not filler, it is the plasticiser that makes the solid behave like a solid rather than like a biscuit.
Three formula patterns produce it. First, high total hard wax, above roughly 20% in a product that is not a solid perfume. Second, a high melting wax used at a normal wax percentage: carnauba melts at 82 to 86 C and builds a rigid network, which is why it belongs at 1 to 3% of a formula rather than as a main structurant, and why candelilla at 68 to 73 C is more forgiving in the same role. Third, a hard fat total that has crept up while the liquid oil phase shrank, which happens easily when butters are added to fix softness. Compare your numbers against the windows in lip balm ratios: in most balms the liquid oils should be 45 to 70% of the total, weight percent of a formula totalling 100.
- Carnauba: 1 to 3% of the formula. Above that, expect drag and crumble.
- Liquid oils: keep at or above 45% in a stick, 55% in a pot balm.
- Cocoa butter: brittle above roughly 20 to 25% of the formula, easier at 10 to 15%.
- A plasticising move: swap 5 points of hard wax for 5 points of a viscous oil or a soft ester.
- Cooling: room temperature setting, no draught, no fridge for hard formulas.
Cocoa butter brittleness and bloom
Cocoa butter deserves its own section because it is brittle for a reason that has nothing to do with how much of it you used. It is strongly polymorphic: the same triglycerides pack into at least six recognised crystal forms with different melting behaviour and different mechanical character. The stable high forms are dense, well-ordered and hard, and they snap.
| Form | Greek | Packing | Classic point | Modern range |
|---|---|---|---|---|
| I | sub-alpha | 2L | 17.3 C | 13.0-18.0 C |
| II | alpha | 2L | 23.3 C | 17.1-24.0 C |
| III | beta prime 2 | 2L | 25.5 C | 22.4-28.0 C |
| IV | beta prime 1 | 2L | 27.5 C | 21.0-33.0 C |
| V | beta 2 | 3L | 33.8 C | 30.0-34.5 C |
| VI | beta 1 | 3L | 36.3 C | 33.5-38.5 C |
Publish the classic six as the conventional reference set, then be honest about what they are. Modern work reports overlapping ranges rather than points, forms III and IV overlap heavily, some authors treat V and VI as subphases of beta rather than separate forms, and form III may be a mixture of II and IV. So "form V melts at 33.8 C" is the midpoint of a range, and the six numbers are a convention rather than something you could reproduce by measuring an arbitrary sample.
Bloom is the visible consequence. The popular explanation, that bloom is simply the form V to form VI transition, is too simple: all chocolates that bloom have lost form V and the bloom crystals are form VI, but not all material in form VI shows visible bloom, and a 2025 study found a sample confirmed as form V by X-ray scattering that bloomed anyway. Bloom crystals are roughly 2 to 10 micrometres long and 10 to 50 nanometres thick. Storage below 18 C inhibits bloom, 18 to 30 C is the risk band, and above 32 to 34 C partial melting drives recrystallisation directly. Sugar bloom has no analogue in an anhydrous balm, because there is neither sugar nor water, so any bloom on a balm is fat bloom or grain.
You cannot temper shea to form V. The nomenclature belongs to cocoa butter. Shea's stable end state is plain beta and the goal is many small beta crystals, produced by a nucleation burst rather than by seeding. Chocolate tempering carries 2 to 5% surviving form V seed under controlled shear, which the cosmetic version of the advice always omits. The shea version of this problem is covered in grainy shea butter.
Formula fixes, as percentage moves
| Change | Size of move | Trade-off |
|---|---|---|
| Cut carnauba, replace with beeswax | -2 points carnauba, +3 beeswax | Slightly softer, much less crumble and drag |
| Raise the liquid oil phase | +5 points | Softer product, watch for weeping |
| Add a viscous, cling-forming oil | 5 to 15% | Adds gloss and tack as well as ductility |
| Swap part of the cocoa butter for a softer butter | -10 points cocoa | Less snap, less structure, may need 1 point more wax |
| Add cetyl alcohol | 2 to 4% | Body without brittleness, faint waxy feel |
Work one change at a time and let each test batch stand for at least 48 hours before judging it, because the crystal network is still transforming for days after the pour. The relationship between crystal size, network strength and how a balm feels is set out in balm texture science, and the melt ranges you need to choose a replacement structurant are in the wax comparison. Natural fats and waxes have melting ranges, not points, which is precisely why two batches of the same recipe can differ in snap.
Process fixes: cooling control
- Melt fully and hold. 70 to 80 C in a water bath. Partial melting is the direct route to bloom in a cocoa butter formula, because surviving crystals seed the wrong forms.
- Pour in the window, not above it. Tins 60 to 68 C, tubes 65 to 72 C. Every degree of overheat is contraction that happens after the skin forms.
- Warm the containers. Around 30 C. It removes most of the wall-contact gradient that causes perimeter cracks and wet spots.
- Cool at room temperature, protected. No draught, no fridge, no freezer for hard formulas. A loosely inverted cardboard box over the tray is enough.
- Lid only when fully set and at room temperature. Lidding warm traps a warm headspace and slows the surface, which encourages both bloom and a soft top.
- Store cool and steady. Below 18 C inhibits bloom in cocoa-rich products, and 18 to 30 C is the band where it develops.
When it is not worth saving
A cracked tin is worth reflowing and a bloomed one is worth remelting, since neither fault involves any chemical change and the material is intact. A genuinely crumbly batch is different: you cannot fix a brittle formula by reheating it, because the problem is the ratio and not the process, so a remelt without added oil returns exactly the same product. Adding oil means recalculating and expanding the batch, which is often more work than pouring a corrected recipe from scratch. Write off small brittle batches, keep them for rough hand use where crumble matters less, and put the effort into the next formula. The exception is any batch where the crumble is accompanied by a stale smell, because that indicates oxidation, and no amount of reformulating rescues oil that has already gone.
Frequently asked questions
Why does my balm crack on top as it cools?
Because the surface set before the interior finished contracting. The skin is rigid, the material beneath is still shrinking, and the stress has to go somewhere. Draughts, cold tins, a cold bench and pouring very hot all sharpen the temperature gradient that causes it. Cool more slowly and evenly, and pour nearer the bottom of the pour window.
Why is my balm crumbly rather than smooth?
Crumbliness is a brittle crystal network: plenty of hard wax and not enough liquid oil or soft butter to plasticise it. Carnauba is the usual suspect because it is very hard and sets into a rigid, high melting network. Keep carnauba at 1 to 3% of the formula and make sure the liquid oil phase is at least half the total.
Why does my lip balm snap off in the tube?
Usually one of two things. Either the column is hollow, so the wall is thinner than it looks, or the formula is brittle and has no ductility at room temperature. Cut a tube lengthways to tell them apart. A hollow column is a pouring fault, a solid column that snaps cleanly is a formula fault.
What are the white streaks on my cocoa butter balm?
Fat bloom: recrystallised fat at the surface, typically crystals a few micrometres long. It is a cosmetic fault, not spoilage, and it comes from partial melting and slow recrystallisation, often after warm storage. There is no sugar bloom in an anhydrous balm because there is no sugar and no water for it.
Can I temper shea butter to form V like chocolate?
No, that is a category error. Forms I to VI are cocoa butter nomenclature. Shea's stable end state is plain beta, and the aim with shea is many small beta crystals rather than a specific metastable form. Chocolate tempering also deliberately keeps a few percent of surviving form V seed under controlled shear, which the cosmetic version of the advice omits.
Does a cracked balm still work?
Yes. A surface crack in a tin is cosmetic and the balm underneath is unchanged. It matters for anything you sell, and it often travels with other cooling faults such as sink holes, so treat it as a signal to review the pour and cooling steps rather than as damage in itself.
Sources and further reading
- Wille, R. L. and Lutton, E. S., Polymorphism of cocoa butter, Journal of the American Oil Chemists' Society, 43:491-496, 1966.
- Ghazani, S. M. and Marangoni, A. G., Molecular origins of polymorphism in cocoa butter, Annual Review of Food Science and Technology, 12:567, 2021.
- Stobbs, J. A. and colleagues, Polymorphic transitions in cocoa butter, Crystal Growth and Design, 25(9), 2025.
- Lonchampt, P. and Hartel, R. W., Fat bloom in chocolate and compound coatings, European Journal of Lipid Science and Technology, 106:241, 2004.
- Sato, K., Crystallization behaviour of fats and lipids: a review, Chemical Engineering Science, 2001.
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.