Deaeration in Emulsion Making: Why Air Ruins Fill Weight and Texture

Air is the cheapest contaminant in a cosmetic plant & the one most often ignored, because it is invisible in the vessel & visible only later, in a jar that looks under-filled, a cream that has dulled from glossy to matte, a pump that spits on the first stroke, or a batch whose declared quantity no longer matches what the customer receives. Nothing about the formula is wrong. The batch was made with air in it, the air was counted as product at the filler, & the consequences arrive in the warehouse rather than in the laboratory.

This article sets out where air enters an emulsion, what it does to fill weight & declared quantity, why it dulls texture & accelerates oxidation, what vacuum deaeration actually removes, how blade geometry & addition sequence decide how much air is entrained in the first place, why deaeration gets harder as a batch thickens, & which in-process checks catch the problem before filling. It is written for the brand owner & the buyer who sign the specification, not for the process engineer who will set the vacuum. The emulsion types behind this are described in our emulsion formulation note, & the filling side is covered in our cream filling & jar handling guide.

Where air enters an emulsion

An emulsion picks up air at every stage where a surface is broken or a powder is wetted. Charging the vessel draws air in with the raw materials, particularly with a powder that is dumped rather than sucked in. A high-shear mixer run with the head too near the surface draws air down & disperses it as fine bubbles. A vortex that reaches the blade pulls air in continuously & beats it into the bulk. Even a well-run batch entrains some air during the emulsion point itself, simply because two immiscible phases are being forced into each other.

The transfer steps matter as much as the mixing. Pumping a finished bulk from the vessel to a holding tank, dropping it through a long hose, or transferring it by gravity from a height all reintroduce air that the mixing stage worked to remove. A batch that was properly deaerated in the vessel can arrive at the filler with a fresh population of bubbles, which is why the transfer route belongs in the process description rather than in the operator's judgement.

Powder wetting deserves a specific mention, because it is the largest single source in many formulations. A thickener, a clay or a pigment has air held between its particles & on its surface, & that air is released into the bulk as the powder wets out. Adding a powder quickly to the surface of a stirred liquid makes a raft that traps air beneath it; adding it slowly under good surface movement, or pre-dispersing it in a portion of the oil phase, reduces the load substantially.

What entrained air does to fill weight & declared quantity

Fill weight is the first place aeration shows up, & it is the place where it costs money rather than reputation. A filler dosing by volume delivers a fixed volume, & a fixed volume of an aerated bulk weighs less than the same volume of a deaerated one. The unit therefore meets the volume check at the head & falls below the weight check at the checkweigher, or, worse, passes both because the tolerance is wide enough to absorb the difference & ships light.

The consequence for a product sold by declared quantity is a compliance question rather than a quality one. A pack that declares a net quantity has to contain at least that quantity on average across a batch, & a batch filled by volume from an aerated bulk will tend towards the bottom of the distribution. The factory catches this at the weigh check & either rejects the batch or tops it up, both of which cost more than deaerating would have.

The second consequence is settlement. An aerated bulk continues to release bubbles after filling, & as the bubbles rise & break the level in the jar drops. A jar that looked correctly filled on the line looks short on the shelf two weeks later, & the customer does not distinguish between a jar that was filled light & one that settled. This is the mechanism behind a large share of the looks under-filled complaints a brand receives, & it is entirely a process problem.

Texture, appearance & the visible signs of aeration

Air changes how light moves through a product, which is why an aerated cream looks different from the same cream deaerated. A well-deaerated emulsion has a gloss, a depth & a clean break; an aerated one looks flatter, duller & slightly grey, because the bubble population scatters light at the surface. A brand that approved a glossy golden sample will reject a dull production batch on appearance alone, without anybody being able to name the cause.

The tactile signs are equally specific. An aerated cream feels lighter & less rich on the skin, spreads faster & seems to disappear sooner, & a gel can feel whipped rather than dense. On a pump, entrained air shows as a spitting or a spluttering first stroke, because the chamber is compressing bubbles rather than moving product. On a jar, it shows as a pinholed surface & a rim of dried product where bubbles have broken at the edge.

The trap for the brand is that these changes are qualitative & easy to argue about. The factory says the bulk meets specification, & it does, because viscosity & pH & appearance were measured on a sample that may itself have been taken from the top of the vessel. The control is a retained reference & a defined appearance check on a filled unit, compared under the same light against the golden sample, rather than a description in a specification.

Oxidation: the shelf-life cost of dissolved oxygen

Air in a bulk is not only bubbles; it is oxygen in solution & oxygen held at the interface with every bubble. That oxygen is available to the parts of a formula that oxidise: unsaturated oils, some vitamins & their derivatives, some botanical extracts, fragrance components, & certain colour systems. What oxidation produces is well known to any formulator: odour change, colour drift, a loss of activity in the material the brand is selling, & a shortening of the usable shelf life.

The relationship is not linear, because oxidation is catalytic once it starts. A small initial oxygen load can be consumed by an antioxidant system without visible consequence, while a larger load exhausts the antioxidant reserve & then proceeds through the oil phase. This is why deaeration is not a cosmetic improvement but part of the stability position of a formula containing oxidisable materials, & why it belongs in the process description for those formulas specifically.

The interaction with the pack matters too. A jar is opened repeatedly & has a large headspace, so a cream in a jar will meet fresh oxygen during use regardless of how well it was deaerated; an airless pump limits that. Deaeration therefore protects the period before opening & the first weeks of use, which is the period the stability programme & the period-after-opening statement are both concerned with.

Vacuum mixing: what it removes & what it does not

Vacuum deaeration works by lowering the pressure above the bulk, which expands the bubbles, makes them buoyant & drives them to the surface where they break. It is effective on the coarse & medium bubble population & progressively less effective on very fine bubbles, because a small bubble has a high internal pressure & a low rise velocity, & it may simply sit in the bulk for as long as the vacuum is applied. Time under vacuum is what removes the fine fraction, not the depth of the vacuum.

The limit that surprises people is viscosity. As the bulk thickens, the rise velocity of a bubble falls, & a bubble that would rise & break in a thin emulsion in a minute may take hours in a thick cream at the same vacuum. This is why deaeration is usually carried out at a defined point in the process, while the bulk is still warm & still at a lower viscosity, rather than at the end when the structure has set & the air is trapped.

Vacuum does not remove dissolved oxygen efficiently on its own, & it does not remove air that is bound to a powder surface or stabilised by a surfactant film at the interface. For those, the answer is upstream: reduce the air introduced, use a defoamer where the formula allows one, & in sensitive formulas consider an oxygen-scavenging approach or a nitrogen blanket on the vessel. The vacuum is the last step in a chain, not the whole solution.

Mixing geometry, blade type & the vortex problem

The cheapest air to remove is the air never entrained, & that is decided by the mixer. A blade run fast enough, or mounted close enough to the surface, creates a vortex that reaches the blade & pulls air into the bulk continuously. A blade mounted too high in a batch that is too small for it does the same. Baffles exist to break the vortex & convert rotation into shear, & a vessel running without them on a low-viscosity batch is generating bubbles rather than dispersion.

The blade choice is a trade-off rather than a preference. A high-shear rotor-stator gives a fine droplet size quickly, which is what sets the texture & the stability of an emulsion, & it is also the most efficient air entrainer in the plant when it is run with the head near the surface. A slow anchor or a counter-rotating system entrains far less air & gives a coarser droplet size. The usual resolution is to use high shear for the emulsion point only & to complete the batch under low shear.

The operating discipline is a set of recorded parameters rather than a skill. Speed, blade immersion, batch level relative to the blade, whether baffles are fitted, & the duration of the high-shear phase all belong in the batch record, because they are the variables that decide how much air the vacuum stage has to remove. A process that leaves these to the operator produces batches with different air loads & therefore different fill weights.

Addition sequence & the point where air is hardest to remove

Order of addition decides both the droplet size & the air load, & the two are not always aligned. Adding the water phase to the oil phase slowly, under moderate shear, with the phases at the temperature the process specifies, gives a predictable emulsion with a modest air load. Adding a powder thickener while the bulk is thin lets the air escape as the powder wets; adding it after the structure has built traps it.

The sequence point that matters most is the one where the viscosity rises. Once a polymer has hydrated or a wax system has begun to set, the bulk stops releasing bubbles, & anything entrained after that point stays in the product. The practical rule is to complete every high-air operation before the structure builds, to deaerate while warm & thin, & to add the volatile, the fragrance & the heat-sensitive active at the defined cooling point with gentle movement rather than with shear.

The second sequence point is the post-addition of a thickener or a neutraliser used to adjust viscosity at the end of a batch. It is a legitimate correction, & it is also an air-entraining step performed at the worst possible moment. Where a correction is needed, the batch should be given a further deaeration period afterwards, & the correction & the extra deaeration should both appear in the batch record so the next batch can be made correctly the first time.

Viscosity, temperature & why deaeration gets harder as the batch thickens

Bubble rise follows a relationship in which velocity falls with viscosity & rises with the square of the bubble radius, which is a formal way of saying that a thick cold bulk holds small bubbles almost indefinitely. The practical consequence is a window: deaeration is cheap while the batch is warm & thin & expensive once it has cooled & thickened. A process that cools first & deaerates second has chosen the expensive route.

Temperature has a second effect through the surfactant & the wax system. A warm bulk has a lower interfacial viscosity & a more mobile interface, so bubbles coalesce & rise more readily. As the batch cools towards a wax setting point or a gel point, the interface stiffens & the bubbles are held in place. The same batch deaerated at two different temperatures can differ measurably, which is why the temperature at the deaeration step is a process parameter rather than a note.

The limit on how warm is the formula itself. Heat-sensitive actives, volatile components & some fragrance materials are lost or degraded at elevated temperature, so the deaeration window is bounded above by the formula & below by the viscosity. Finding that window is a formulation exercise as much as a process one, & it is one of the reasons a formula that behaves beautifully at laboratory scale can be difficult at production scale.

Measuring aeration: density, appearance & in-process checks

Aeration is measurable, & the cheapest measurement is density. A bulk with a known formulation has an expected density at a stated temperature, & a sample drawn from the vessel & compared against that value gives a direct read on how much air is present. The method needs care: the sample must be drawn without itself entraining air, measured at a stated temperature, & compared against a reference generated on a properly deaerated batch rather than against a theoretical value.

The production check that matters is at the filler, because that is where the air becomes a customer problem. A checkweigher with a stated frequency & a rejection rule catches a drifting fill weight, & a weight trend rather than an individual reading is what reveals aeration, since the effect is systematic across the batch rather than random. A line that records weight but never reviews the trend will fill an entire aerated batch inside tolerance.

The appearance check is the one a brand should insist on. A filled unit from the start, the middle & the end of the run, held overnight & compared against the golden sample under the same light, catches settlement, surface pinholing & dullness. It costs three jars & a day, & it is the check that connects the process to the thing the customer actually sees.

Filling, settling & what happens after the vessel

Filling is the last point at which air can be removed or added. A piston filler with a bottom-up fill & a controlled stroke introduces little; a gravity fill from a height, a long unrestricted hose, or a hopper that is allowed to run low & draw from a vortex all reintroduce it. The hopper level is worth a specific rule, because a hopper that runs down to the outlet is the most common source of aeration in an otherwise well-run batch.

Settling after filling is the reason a batch can pass every check on the line & still be short on the shelf. The bubbles that were not removed rise over hours or days, the level drops, & the customer sees a jar with a gap at the shoulder. The remedy at the factory is a settling period before capping or a post-fill check after a stated interval; the remedy at the specification stage is to set the nominal fill with settlement in mind.

The last point is the headspace in the pack & the way the closure is applied. A jar filled to the brim with an aerated bulk has nowhere for the bubbles to go except out over the rim, which produces the dried ring customers notice. A pack with a designed headspace absorbs the settlement visually, & a closure applied promptly limits the surface drying. Both are packaging decisions, & both interact with the process rather than sitting alongside it.

Aeration sources at a glance

The table below sets the main sources of entrained air against how the air gets in, what it causes in the finished unit, & the control that works at each point. The rows describe ordinary process practice rather than fixed values, & the right control for a specific product depends on the formula, the vessel & the filler being used.

The controls in the table are cumulative rather than alternative, because a batch that entrains air at four points will not be saved by a good vacuum stage at the end. The sequence that works is to minimise entrainment in the vessel, deaerate at the point where the bulk is warm & thin, protect the bulk through transfer, & fill with a route & a hopper discipline that does not undo the work. Each step is cheap on its own & the batch is only as good as the weakest one.

What to put in the brief

Deaeration is a process specification & it should be written as one rather than assumed. Name the appearance you are buying, because dullness & settlement are the symptoms you will actually see. Name the fill control you expect, because a declared quantity is a compliance matter. Name the pack, because the headspace & the closure decide how visible the settlement is. Then ask how the factory controls each of the sources above.

• The appearance standard: gloss, colour & break, with a retained golden sample as the reference.
• The fill control: by weight or by volume, the tolerance, & the checkweigher frequency.
• The declared quantity & the average rule that applies in your market.
• The pack & closure, with the headspace you can accept visually.
• The viscosity window at the fill temperature, since it sets bubble release.
• Whether the formula contains oxidisable materials that make deaeration critical.
• The settling check you expect: interval, sample points & comparison method.
• The batch record parameters you want named: shear time, vacuum time & temperature.

Then ask for the mixing parameters to be recorded rather than left to the operator. A batch record that names the blade, the speed, the immersion, the shear duration, the vacuum duration & the temperature at deaeration is what makes a later comparison possible when a batch comes out dull or light. Without those numbers the investigation has nothing to work from & the answer becomes we will be more careful next time.

Where our capability fits

We handle aeration as a process specification rather than as an operator skill. For an emulsion project we run, the shear phase is limited & the head is submerged, powders are added under controlled surface movement while the bulk is thin, deaeration is carried out at the point in the cooling phase where the viscosity still permits bubble release, & the transfer route to the filler is defined so the work is not undone afterwards. We record the parameters in the batch record, we check fill weight at a stated frequency & review the trend, & we hold start, middle & end units overnight for a settling & appearance comparison against the golden sample.

We work with brands on private label, OEM/ODM contract manufacturing & full custom formula development, & the deaeration window is established during development so the production process has something to hold to. Our lines include a pro-xylane soothing facial cream in a jar & a blue hydrating peptide cream jar, filled with the same fill-weight control, settling check & retention discipline that an air-sensitive emulsion needs.

Tell us the emulsion you are making, the pack you intend to fill & the appearance standard you are buying, & we will come back with the deaeration point, the fill control & the settling check we would run. Start at our inquiry page, email capsulebiotech@163.com, or message the right team on WhatsApp: Capsule Series (capsules only) +86 158-7655-5319, Cosmetics Series (serum, hair care, body care, fragrance) +86 158-7655-5319.

General manufacturing information, not regulatory or legal advice. Fill control, declared quantity rules & the average quantity duties that apply to a prepacked product vary by market, & the position for a specific product & pack should be confirmed with the competent authority or a qualified regulatory adviser before it is placed on sale.