When a whipped butter deflates in the jar or turns back into a solid gritty puck
A whipped butter is a foam held by fat crystals, so it falls if whipped or stored warm. Whip temperatures, overrun figures and whether a rewhip will save it.
A whipped body butter is mostly air, and nothing holds that air in place except fat crystals. Fat crystals melt. Two separate faults then arrive as one complaint, a jar that deflated and a butter that set back into a gritty solid, and they take different rescues. This page tells you which one you have, in about ninety seconds, before you decide whether to remelt.
The bubbles are held by fat crystals alone, so the product has almost no heat tolerance. Whip at 18 to 22 C after a partial set, stop at 40 to 80 percent overrun, and keep it below 25 C. Between 27 and 30 C the crystal network fails, the air escapes, and the mass recrystallises slowly into a solid, gritty puck.
What is actually holding the air
Whipped cream is the product everyone has in mind, and it is the wrong model. Cream holds air with partially coalesced fat globules, adsorbed protein and a viscous aqueous phase. A whipped butter has no water, no protein, no gum and no surfactant. Two things stop a bubble escaping.
- Crystals sitting at the bubble surface. Solid fat particles adsorb at the air and oil interface and armour the bubble, the same particle stabilisation that keeps aerated shortenings standing up.
- A crystal network in the continuous oil phase. The bulk is an oleogel with a yield stress, and a bubble cannot rise through a material that will not flow under the small buoyancy force acting on it. That network is described under oleogels and wax networks and quantified under rheology and yield stress.
Both mechanisms are made of the same material, so they fail together and fast. The consequence worth carrying away: the temperature at which a whipped butter dies is not the melting point of the butter, but the lower temperature at which the solid fraction drops below what a foam needs. For shea that gap is around five degrees, and a warm shelf sits inside it.
The whip window: 18 to 22 C and a partial set
There are three states a fat mass can be in when the beater goes in, and only one of them aerates.
Liquid, above about 30 C. No crystals exist, so nothing armours a bubble and nothing holds the bulk. Air disperses and immediately recoalesces, and you end up with a dense block wearing a pale foamy skin that fools you for one day. This is the commonest reason a whipped butter was never whipped at all.
Fully set and cold, below about 15 C. The network is rigid, so the beater shears the mass into crumbs that never merge. People read this as graininess when it is nothing of the kind, and warming a spoonful to 20 C for half an hour and rewhipping it settles the question.
Partially set, 18 to 22 C. Enough crystals exist to armour bubbles, and enough liquid oil remains for the mass to flow around the beater. Aim for the consistency of very soft kitchen butter: opaque, matt, holding a spoon mark that does not slump. That is roughly 15 to 30 percent solid fat in a shea-based formula.
Getting there matters as much as arriving. Melt fully to 70 to 80 C and hold, as set out under pour temperatures, then cool fast and stirred to about 28 C before letting the mass drift into the whip window. A slow cool produces few large crystals, which means a weak network and a coarse texture, for the reasons given under crystallisation and cooling rate. A 500 g batch takes 25 to 45 minutes in a domestic fridge from 28 C, checked every ten. Then whip 5 to 10 minutes on high with a whisk attachment. A paddle folds, it does not aerate.
Do not add water, aloe juice, milk or hydrosol to make a whipped butter fluffier. It does aerate more easily, and it also turns an anhydrous product with no preservative into a microbiological problem with no preservative. The consequences are set out under water contamination in balm and do balms need preservatives.
Overrun: how much air the network can carry
Overrun is the food industry's term for aeration, and it is the number to record instead of whipping until it looks right. It is the volume of air added as a percentage of the original volume. Fill a container of known volume level with unwhipped set butter and weigh it, then do the same with whipped material, scooped in and levelled without pressing. Overrun is the difference in mass divided by the whipped mass, times one hundred.
| Overrun | Density (g/mL) | Feel | What happens over a month |
|---|---|---|---|
| 0%, unwhipped | 0.90 | Dense, waxy on the fingertip | Nothing. The reference. |
| 20 to 40% | 0.64-0.75 | Dense mousse, glossy, scoopable | Very stable, but it does not look whipped |
| 40 to 80% | 0.50-0.64 | Light, holds a soft peak | Stable at or below 25 C. The target band. |
| 80 to 120% | 0.41-0.50 | Very light, sharp peaks | Drains within days, weeps clear oil at the base |
| Above 120% | Below 0.41 | Candyfloss, no body | Falls under its own weight before it reaches a customer |
Above roughly 80 percent overrun the films of oil between bubbles are thin enough that gravity drains them faster than the network resists, and the drained oil appears as a clear pool at the base or a wet sheen on top. It looks identical to the beading described under balm sweating, but the fix is different: less air, not less liquid oil.
Overrun also explains the jar that arrived half empty. A 200 mL jar filled level at 60 percent overrun holds about 112 g. Collapsed, that 112 g occupies about 125 mL, so the customer opens a jar that is 37 percent headspace. Nothing leaked and nothing was short-filled. For fill weights across a range of jar sizes, use the batch calculator.
Why it falls between 27 and 30 C
Solid fat content is the proportion of a fat that is crystalline at a given temperature, measured by pulsed nuclear magnetic resonance. It governs everything a whipped butter does, and it falls away sharply at exactly the temperature of a shelf in a warm room.
| Temperature | Shea butter | Cocoa butter | Consequence for the foam |
|---|---|---|---|
| 10 C | 45-55% | 80-85% | Too rigid to aerate. The beater tears it up. |
| 18 to 22 C | 25-40% | 75-80% | The whip window for shea. Cocoa butter is far too hard. |
| 25 C | 18-28% | 65-75% | Softening but intact. A warm room is survivable. |
| 27 to 30 C | 7-18% | 40-50% | Network fails, bubbles coalesce and rise, foam collapses. |
| 32 to 35 C | 0-5% | 0-5% | Liquid or nearly so. Crude shea slips at about 32 C. |
Two things follow. The collapse temperature of a plain shea and oil whip lands at 27 to 30 C, several degrees below the slip point of the butter, because a foam needs more solid fat to stand up than a poured balm needs to feel firm. And that band overlaps skin temperature, a sunny windowsill, a car parcel shelf and a summer delivery van, which is why the fault is seasonal. The broader case is on balm melting in the heat.
Collapse does not reverse on its own. Putting the jar back in the fridge sets the fat again but does not put the air back, because the bubbles have already merged and escaped. That is why every rescue below involves a beater.
The second fault: the gritty puck
A collapsed butter does not stay soft. It sits near 30 C with most of its crystals molten, then cools slowly inside an insulating jar in still air. That is the textbook recipe for few nucleation sites and large crystals. Where shea is the main hard fat, the stearin fraction segregates and transforms to the beta form over the following days, giving a solid mass that feels sandy. The mechanism is on grainy shea butter and the vocabulary under fat crystal polymorphism.
A jar that got warm in July usually arrives with both faults stacked. Losing the air is a mechanical problem a whisk undoes. Coarse crystals are a thermal problem only a full remelt undoes. Rewhip a grainy puck and you get grainy foam.
Diagnose it in ninety seconds
Take a heaped teaspoon from the middle of an affected jar, not the top. Leave it at 20 C for ten minutes, then beat it hard with a fork for a minute. If it lightens visibly, the crystals are fine and only the foam failed. If it stays dense or gritty, the crystals have coarsened and beating will not help. That test picks the rescue.
| What you are looking at | Mechanism | Confirm by | Do this |
|---|---|---|---|
| Jar half empty, surface smooth and glossy, product still soft and smooth | Foam drained and collapsed, crystals unchanged | Teaspoon rewhip at 20 C: it lightens | Rechill the batch and rewhip |
| Solid puck, sandy on the fingertip, appeared after a warm week | Collapse, then slow recrystallisation of the stearin fraction | Rub a smear hard: the grit vanishes at skin temperature | Full remelt, hold, fast stirred cool, rewhip |
| Clear oil pooled at the base, product above it still light | Drainage from an over-aerated foam | Weigh a level 100 mL: under about 45 g means overrun above 100 percent | Rewhip to a lower overrun, or add wax |
| Hard lumps that survive rubbing and warming | Undissolved wax, or wax that set before the whip | Warm a sample to 45 C: shea grain clears, wax persists | Remelt fully and strain, per undissolved wax specks |
| Never rose at all, stayed dense and shiny from the start | Whipped from liquid, or from a mass above about 26 C | Your temperature record, not the bowl | Cool to 18 to 22 C and whip again |
| Crumbly, tears into lumps, will not smooth out | Whipped too cold, below roughly 15 C | Stand a sample 30 minutes at 20 C and rewhip | Warm the mass into the window and rewhip |
| Wet film, speckles, sour or musty smell | Water got in, from condensation or from fingers | Look along the jar wall for a film or coloured spots | Discard. See mould in balm |
Rescue one: rechill and rewhip
Use this when the teaspoon test lightened. Nothing is wrong with the fat, only with the air.
- Combine the batch. Scrape every jar into one bowl. Rescuing jar by jar guarantees they will not match, which is batch to batch inconsistency by another route.
- Bring the mass to 18 to 22 C. If it is warmer, 20 to 30 minutes in the fridge with a check at ten. If it is fridge cold, let it stand until a spoon meets mild resistance. Use a probe, not a fingertip.
- Whip 3 to 5 minutes on high. A rewhip needs less time than the original because much of the network is already fine grained. Stop at a soft peak.
- Check the overrun before filling. Weigh a level 100 mL: 50 to 64 g is the target band.
- Scoop into cool jars, never pour. Filling warm reintroduces the problem you just fixed.
- Hold a jar at 30 C for 48 hours. If it survives, it will survive a shelf. If not, go to rescue two and add a structurant.
Do not add liquid oil at this stage to make it creamier. Liquid oil lowers solid fat content, which is the one thing already in short supply.
Rescue two: remelt, and yes that is safe
Remelting frightens people who have learned that reheating a cream breaks it. That rule belongs to emulsions. A whipped butter has no water phase and no emulsifier, so there is nothing to break. Melting erases every crystal and lets you build the network again. The only real losses are volatile, which is why the heat sensitive phase goes back in at the end.
- Melt the lot to 70 to 80 C and hold 20 to 30 minutes. Beta form shea stearin clears at about 41 to 43 C, so a melt that merely looks liquid at 40 C leaves seed crystals and the grain returns within a week.
- Cool fast with stirring to about 28 C. A cold water bath and a spatula. Agitation multiplies nucleation sites, which is how you get many small crystals rather than few large ones.
- Add the heat sensitive phase below 50 to 60 C. Tocopherol, flavour and any essential oils, weighed in advance so the addition takes seconds.
- Drop to 18 to 22 C and hold until partially set. Fridge with checks, or a cool room.
- Whip 5 to 10 minutes and stop at 40 to 80 percent overrun. Longer is not better. Friction alone lifts a kilogram batch by several degrees in a stand mixer, so past a point you are warming the product you are trying to stiffen.
- Fill, cap and retain a sample for ten days. A same day check tells you nothing about the beta transformation that produces grain.
Before remelting a whole batch, run 100 g through the full sequence with 4 percent added beeswax and compare it with 100 g reprocessed unchanged. Hold both at 30 C for two days. Two small test pots cost less than a second failed production run.
Prevention: four levers, in order of effect
A pure butter and oil whip has no heat resistance to give. Every durable whipped butter on a shop shelf contains a structurant, and makers who cannot work out why theirs collapses have usually left one out on the grounds that it was not natural enough.
| Lever | Working rate | What it does | What it costs you |
|---|---|---|---|
| Beeswax | 3-6% | Raises the collapse temperature several degrees and stiffens the continuous phase | Above about 6 percent it is too firm to aerate well, and drag becomes obvious. Not vegan. |
| Candelilla wax | 2-4% | Same effect, vegan, roughly half to two thirds the beeswax rate for equal firmness | Less cushion, a short brittle texture at the top of the range |
| Stearic acid | 3-5% | Adds fine, high melting crystals and lifts solid fat content through the 25 to 32 C danger band | A drier, slightly draggy slip and a whiter product |
| Harder butter in the hard fat share | 10-25% of the hard fat | Kokum or cocoa butter lift the whole solid fat curve | Cocoa butter brings its own graininess and a brittle set |
| Liquid oil ceiling | Keep below 30-35% | Every point of liquid oil removed is solid fat gained at 30 C | Below about 15 percent the whip goes short and crumbly |
The two waxes are set side by side under waxes compared, and the swap rate between them is in the wax substitution calculator. Stearic acid is the least understood lever and the most useful in a hot climate, being a single fatty acid rather than a triglyceride and quick to crystallise: detail on stearic acid. Splitting a mostly shea hard fat with a more uniform butter cuts grain risk at the same time, as argued in butters compared.
One popular lever does not work. Cornflour, arrowroot and tapioca at 2 to 5 percent genuinely cut the greasy afterfeel, but they sit in the oil phase and add nothing to the crystal network, so a starched butter collapses at exactly the same temperature as an unstarched one. Trade-offs on starches and silica.
Packaging and the room it lives in
A whipped butter is packaging sensitive in a way a poured balm is not. Wide, shallow jars beat deep narrow ones twice over: less product weight presses on the bottom layer, so the foam compresses less under its own head, and a customer can scoop rather than dig. Fill to about 90 percent, because whipped material settles in the first week and an overfilled jar smears its lid.
Metal tins are a poor choice: they conduct heat straight into the product from a warm hand, they seal badly against the oil a drained foam releases, and a slip lid pressed onto a proud fill crushes the foam it touches. Glass and thick-walled polypropylene are the sensible options, and closures are covered under the packaging guide and packaging compatibility.
Transport finishes the job. Parcels sit in vans that reach 40 C and more, and no formulation survives that as a foam. Ship in cool weather, tell customers not to leave the jar in a car, and read shipping in hot weather before a July launch. For hot climates year round the honest answer is often a poured body balm instead, and the two are compared under body balm versus body butter.
What is not established, and when to stop
Be clear about the status of the numbers above. There is no published cosmetic literature on whipped butter stability as such. The mechanism is borrowed from aerated food science, where particle stabilised foams in fat continuous systems are well described, and the solid fat figures come from fractionation and confectionery work on the same butters. The overrun bands and the 27 to 30 C collapse figure are working values consistent with that physics, not measurements from a controlled study, and they move with provenance and formula. Run your own 30 C hold.
The decision rule is short. If the teaspoon test lightens, rewhip the batch and sell it. If it stays gritty, remelt, and use the remelt to add 3 to 5 percent wax or stearic acid rather than reproducing the formula that just failed. If it smells sharp or waxy it is oxidising, and a rescue only gives you smooth rancid butter, so read rancid balm and bin it. If water got in, bin it without deliberation. And if a formula has failed a 30 C hold twice, stop rescuing and change the formula: collapse is a design fault far more often than a technique fault.
Frequently asked questions
Why did my whipped body butter go flat in the jar?
The air was held by fat crystals, and those crystals melted. A shea-based whip loses its structure somewhere between 27 and 30 C, well below the 32 C slip point of the butter itself, because a foam needs more solid fat to stand up than a firm balm does. Once bubbles have merged and escaped, cooling the jar sets the fat again but does not put the air back.
Can I rewhip a collapsed body butter?
Often, yes. Take a teaspoon from the middle of the jar, bring it to about 20 C, and beat it with a fork for a minute. If it lightens, the crystals are still fine and you can rechill the whole batch to 18 to 22 C and rewhip it for three to five minutes. If it stays dense or feels sandy, the crystals have coarsened and only a full remelt will fix it.
Is it safe to melt a whipped butter down and start again?
Yes. A whipped butter is anhydrous, with no water phase and no emulsifier, so there is no emulsion to break. Melting erases every crystal and lets you rebuild the structure from scratch, and nothing is chemically degraded by it. Add flavour, essential oils and tocopherol back below 50 to 60 C, because those are the only things a remelt actually costs you.
What temperature should I whip body butter at?
18 to 22 C, after a partial set. The mass should look like very soft butter and hold a spoon mark without slumping, which is roughly 15 to 30 percent solid fat. Whipping from a liquid melt cannot work, because there are no crystals to hold bubbles, and whipping from a fridge-hard block just tears it into crumbs that never come together.
Why did my whipped butter turn gritty after it collapsed?
Because it recrystallised slowly. A collapsed butter sits near 30 C with most of its crystals molten, then cools gradually inside an insulating jar in still air, which produces few nucleation sites and large crystals. Where shea is the main hard fat, the high-melting stearin fraction segregates and transforms to the beta form over the following days, and those aggregates are what your fingertip detects.
How much wax should I add to stop a whipped butter melting?
Three to six percent beeswax, or two to four percent candelilla for a vegan version, as weight percent of the whole formula. Stearic acid at three to five percent does a similar job and raises solid fat content across the 25 to 32 C band more efficiently, at the cost of a drier slip. Above about six percent wax the mass becomes too stiff to aerate properly.
Why is oil pooling at the bottom of my whipped butter?
Almost always over-whipping. Past roughly 80 percent overrun, the films of oil separating the bubbles are thin enough that gravity drains them faster than the crystal network can hold them, and the drained oil collects at the base. Weigh a level 100 mL of the product: under about 45 g means you aerated it too far. Whip less, or add a structurant.
Sources and further reading
- American Oil Chemists' Society, Official Method Cd 16b-93, Solid Fat Content by Low-Resolution Nuclear Magnetic Resonance, Direct Method, AOCS, Urbana.
- International Union of Pure and Applied Chemistry, Standard Method 2.150, Solid Content Determination in Fats by NMR, in Standard Methods for the Analysis of Oils, Fats and Derivatives, 7th edition.
- Campbell, G. M. and Mougeot, E., Creation and characterisation of aerated food products, Trends in Food Science and Technology, 10:283-296, 1999.
- Rousseau, D., Fat crystals and emulsion stability, a review, Food Research International, 33:3-14, 2000.
- Abdel-Razek, A. G. and colleagues, Characterisation of crude shea butter and its stearin and olein fractions, Foods, 12(8):1626, 2023.
- Ray, J. and colleagues, Polymorphic behaviour of shea stearin, European Journal of Lipid Science and Technology, 115:1094, 2013.
- Gunstone, F. D., The Chemistry of Oils and Fats: Sources, Composition, Properties and Uses, Blackwell Publishing, Oxford, 2004.
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.