Desiccant Packaging for Glass Bottles: Cap or Sachet

Which Products Genuinely Need a Desiccant and Which Do Not

A desiccant does one job. It holds the humidity inside a closed pack below a threshold for a defined period, by taking up water vapour faster than the product can absorb it. Everything else that is claimed for moisture control follows from that single function, and any pack where that function is not the limiting factor does not need a desiccant, however attractive the phrase looks on a label.

The products that genuinely need one share two properties: their degradation is driven by moisture, and their critical moisture level or water activity is low enough that ordinary headspace humidity will reach it inside the shelf life. That combination is common in dry powders and in solids with a large surface area. Instant drink powders, spray-dried flavour and colour powders, ground and instant tea, instant coffee blends, protein and collagen powders, powdered vitamin and mineral premixes, herbal extract powders, dry seasoning blends, effervescent tablets and granules, dry bakery mixes and hard capsules holding a hygroscopic active all sit in this group, because a powder presents a large surface to the headspace air and a small amount of absorbed water changes its texture, its flow, its colour or its potency.

The products that do not need one are equally worth naming, because a desiccant on the wrong pack adds cost and reduces the closure’s sealing reliability for nothing. Liquids, syrups and gels do not need moisture control, since the pack already contains water and the headspace humidity is set by the product itself. Products with a high water activity cannot be rescued by a desiccant, because the desiccant is exhausted within days and the product continues to degrade at its own moisture level. Products whose limiting failure is oxidation rather than hydrolysis, such as oils, fried snacks and some fatty powders, need a different countermeasure altogether: moisture control does nothing about rancidity and can even be irrelevant to it, and the answer there is usually an oxygen absorber, a better oxygen barrier or an antioxidant system. Alcoholic and high-solvent products sit outside the subject for the same reason. And solids whose critical moisture is far above any realistic headspace level, including products that retain a substantial residual moisture by design, are not improved by a desiccant at all.

Two nuances decide more projects than the product list does. The first is that a desiccant holds a dry product dry; it does not dry a wet one. If the product enters the pack above its critical moisture, the bed can only delay the inevitable and will be consumed within the first days of shelf life, and the correct fix is upstream, in the drying stage or the filling room, not in the cap. The second is that moisture control and the closure are one system. A bed behind a closure that leaks will be spent by vapour arriving from outside the pack, and the money spent on a larger bed would have been better spent on the sealing land. The reverse is also true: a modest bed behind a genuinely tight closure can hold a long shelf life comfortably. Both of these points come back repeatedly in the sizing section below.

What Can Be Chosen on the Closure Side Before You Commit to a Mould

Almost everything that determines whether a desiccant pack works is decided before the first sample is moulded, which is why the choice list deserves to be written down and closed off rather than discovered during sampling. On the closure side there is more freedom than most buyers expect. The shell format can be a screw closure, a lug closure or a disc-top, and the format decides how much internal volume is available for a bed. The shell material, its wall thickness and the internal ribs or ledges decide where the bed can be retained. The bed’s location can be central under the top panel or annular around the inside of the shell. The breathable membrane that closes the bed can be specified by its pore size, its air permeability, its thickness and the method by which it is welded or bonded to the shell. The closure height and its internal clearance can be treated as adjustable, within the limits of the finish. The colour and the decoration can be positioned so that the closure still reads normally on a shelf.

One decision above all the others should be taken first: does the pack keep a compression liner as well as the membrane, or does the membrane replace it. A closure that has both a breathable membrane over the bed and a hermetic compression liner sealing the finish will not work as a desiccant closure at all, because the liner blocks the very vapour path the bed depends on. This mistake is common when a desiccant bed is added to an existing closure specification, and it produces a pack that looks correct, tests as if it were sealed, and delivers none of the intended protection. The vapour path has to be traced from the headspace through the membrane to the bed, and it has to stay open in every direction along that route.

A second decision concerns the outer format, and it is easy to defer and expensive to forget. Because a breathable closure is not hermetic, the filled pack usually needs an outer barrier for transport and storage, and the closures themselves usually need to arrive in a sealed pouch. The pouch quantity matters as much as the pouch material: if one pouch holds more closures than a shift can use, the remainder is exposed to the filling room for the rest of the day and arrives at the capper already partly spent. The useful number to agree is closures per pouch, matched to a realistic consumption rate, together with a rule on what happens to an opened pouch at the end of a shift.

Finally, the venting and cleaning requirements of the closure should be settled early. The bed has to be protected from liquid and from dust, the membrane must not be damaged by a wash or a rinse, and the closure must survive whatever cleaning the line applies to closures. A desiccant closure that cannot be cleaned is a closure that must be opened inside the zone or in a controlled area, and that is a line decision rather than a packaging decision.

Three Placement Options and How the Choice Follows the Fill Weight

There are three places to put a desiccant, and the choice between them is decided mostly by geometry and by how much risk the market will tolerate, not by capacity alone. The first is an integrated desiccant closure, where the bed sits inside the shell behind a breathable membrane. The second is a loose sachet or a canister dropped into the bottle before capping. The third is a separate compartment, which in practice means a stopper or a canister pressed into the neck, or a pack deliberately built with a divided interior.

The integrated closure option keeps the desiccant with the pack, removes any possibility of the sachet being mistaken for the product or swallowed, needs no station on the filling line, and gives the consumer a clean experience. Its constraint is space. The bed can only be as large as the inside of the shell allows, which puts a ceiling on the total capacity, and that ceiling interacts directly with the headspace volume. A small fill in a tall bottle creates a large headspace full of air to be dried and to be kept dry, while a large fill in the same bottle leaves very little. A cap-mounted bed therefore suits a pack where the fill is generous relative to the container, or where the shelf life is moderate, or where the closure’s barrier is good enough that the ingress term is small.

The loose sachet or drop-in canister offers far more capacity per pack, because the bed is no longer limited by the inside of the closure, and it is usually the cheapest way to buy capacity. What it introduces is a foreign body, with everything that follows. The sachet can be ingested, it can be a choking hazard, it can block the neck, and it can be damaged during filling so that desiccant dust enters the product. In markets where the warning requirement is strict, or where the consumer is unlikely to read one, this risk has to be managed by design rather than by advice, and in some product categories the loose sachet is simply not an acceptable answer. It also adds a handling step at the filler, where the sachet must be dispensed reliably and checked for, and where a missed sachet is a silent failure that no downstream inspection will catch.

The separate compartment route, usually a canister or a stopper seated in the neck, sits between the two. It offers good capacity, it is visible and easy to explain, it is simple to remove, and it can double as a physical stopper that limits spillage. Its limits are the neck diameter, since a narrow neck may not accept a canister at all, and the fact that it is a separate part with its own tolerance and its own insertion force, which has to be compatible with the finish and with the closing torque. Where the neck is wide enough and the consumer is expected to pour from it, this is often the best compromise between protection and consumer handling.

The fill weight and the headspace are the variables that decide between the three more often than the shelf life does. The useful calculation at the concept stage is the ratio of headspace volume to fill weight. Where that ratio is high, a cap-mounted bed is fighting a large volume of air, and either the fill should grow or the placement should change. Where the ratio is low, the cap route becomes viable and the closure geometry is the remaining constraint. The mouth diameter then rules out the canister for narrow-neck formats, which is the point at which narrow-neck packs and cap-mounted beds naturally pair with each other.

How a Desiccant Cap Differs From a Plain Lined Cap at the Sealing Face

A plain lined closure is a compression barrier. The liner is squeezed between the shell and the sealing land of the finish, contact pressure does the work, and the vapour barrier is created by the liner material being pressed flat against a defined ring of glass. Torque, liner thickness and liner compressibility together decide how good that barrier is, and a leak appears when the compression is lost anywhere around the ring.

A desiccant closure works the other way round. It is deliberately permeable to water vapour, because the bed behind the membrane can only control the headspace if vapour can reach it. The membrane, or the vented path through the shell, is a controlled opening that passes vapour and blocks liquid and particulates. The consequence is that the closure is a liquid seal with an intentional vapour path, rather than a barrier with no vapour path at all. The difference is not a matter of degree; it changes what the finish, the torque window and the test plan are for.

Three practical consequences follow. First, torque changes meaning. On a plain lined cap, torque is a proxy for compression and therefore for barrier performance. On a desiccant cap, torque holds the closure on and keeps the liquid seal closed, but it no longer defines the vapour barrier, and an over-tightened cap can crush the bead that seals the membrane to the shell, while an under-tightened cap can leave a liquid path at the finish. The torque window therefore has to be set from the closure design and verified on the pack, rather than inherited from a plain cap. Second, the seal between the membrane and the shell becomes a critical joint. It is usually a welded or bonded ring, and it has to be continuous, because a pin hole in that joint is a permanent, invisible leak into the headspace. Third, the finish land matters more, not less. On a breathable closure, a defect on the sealing land does not simply admit a little moisture against a compression liner; it becomes a direct channel from the room into the headspace, and the bed pays for it.

The outward-facing consequence is often the one that is missed at the design stage. Because the vapour path is open by design, the bed is in contact with the outside atmosphere through the closure, and an unprotected pack stored in a humid place will spend its capacity on the room rather than on the product. This is why a desiccant pack is usually shipped inside an outer barrier, either a foil pouch around the finished pack or a sealed liner inside the carton, in the same way that a loose sachet is shipped inside its own foil wrapper. Treating that outer barrier as optional is the most reliable way to make a correctly sized bed perform like an undersized one.

Silica Gel, Molecular Sieve or Clay: Reading the Sorption Behaviour

The three desiccant families used in closures and canisters behave differently across the humidity range, and choosing between them is a question about where on that range the pack has to sit rather than a question about which one is better. A number quoted without a relative humidity is not a capacity, it is a maximum, and a maximum is exactly the number that flatters whichever material is being sold.

Silica gel works by physical adsorption and its capacity climbs steeply as relative humidity rises. It is at its most useful in the middle and upper part of the humidity range, where a moderate amount of material holds a large amount of water, and it is the natural choice when the requirement is to keep the headspace in a moderate band rather than to drive it very low. It is widely available in food and pharmaceutical grades and it is usually the lowest-cost of the three in the form used in closures. Its limitation is the same curve read backwards: at very low humidity its holding capacity falls away, so it is a poor choice where the threshold the product must stay below is itself very low.

Molecular sieve, usually a type 3A or 4A zeolite, adsorbs strongly even at low humidity, which means it can hold a headspace genuinely dry. That property is what it is bought for. It is more expensive than silica gel, it needs a higher temperature to regenerate, and its strong affinity is not selective: it will also take up other small polar molecules, including some aroma compounds, which is why a switch to molecular sieve can change the smell of a product without any change to the product itself. It is the right answer where the product’s critical humidity is low or the shelf life is long, and it is the wrong answer where the target is moderate and cost matters.

Clay desiccants, based on montmorillonite or bentonite, sit at the cost-driven end. Their capacity at a given humidity is lower than silica gel, their uptake is slower, and their dust and friability have to be managed, but they are cheap and adequate where the target humidity band is not tight and the shelf life is moderate. They are common in bulk dry goods and in applications where the pack merely needs to avoid a spike rather than to hold a low humidity throughout.

Two further properties decide real projects. The first is kinetics, which matters when the exposure is short and severe, for example a pack closed in a humid filling room and then sealed into an outer barrier: a fast material is more useful there than a high-capacity slow one. The second is the starting condition of the bed itself. A closure that has been stored in an open cardboard box in a humid warehouse, or a pouch that was opened on Monday and used on Friday, has already spent part of its capacity before the pack is closed, and no capacity figure on a datasheet accounts for that. The control is procedural rather than technical: sealed pouches, a defined pouch size, and a rule about the filling room.

Blended beds are the fourth option and are sometimes proposed as a way to cover a wider humidity band with one closure. A blend does behave as a weighted average of its components, but it also adds a variable: the ratio has to be justified from the product’s sorption curve rather than from a supplier’s general recommendation, and the mixture has to be characterised rather than assumed. A blend is worth considering where a pack genuinely has to be protected through an initial very dry period and then across a longer, more moderate period, and where the product’s aroma has been tested against the stronger of the two materials. Where a single well-understood material already holds the target, a blend adds cost and a question mark for no gain.

Sizing the Charge: Working Back From Moisture Content and Shelf Life

Sizing a desiccant is a subtraction problem before it is a materials problem. The bed has to hold everything that pushes the pack towards its moisture threshold, and there are four of those loads. The first is the water that must be removed from the product to bring it below its critical moisture or water activity, which is where the product’s own sorption data comes in. The second is the water that arrives from outside through the closure and the container over the shelf life, which is the product of the closure’s moisture vapour transmission rate and the humidity difference it sees, integrated over the months in the field. The third is the water already sitting in the headspace air at the moment the closure is applied, which is small in absolute terms but not zero, and which is larger when the filling room is humid or the headspace is large. The fourth is the water that desorbs from the container wall, the closure and the membrane after closure, which is negligible for glass but not for the closure assembly and any residual process moisture.

Add the four, multiply by a safety factor, and divide by the working capacity of the chosen desiccant at the humidity the bed will actually experience at the end of its life. Two rules make that arithmetic behave. Work in units of water, in milligrams, rather than in grams of desiccant, because the conversion between the two is where the errors accumulate. And use the working capacity at the end of life, at the highest humidity the pack will see, rather than the headline maximum, because a bed is only useful while it still has somewhere to put the next milligram.

The single most useful fact about the arithmetic is which term dominates. For a pack with a long shelf life, the ingress term almost always beats the other three, often by a wide margin, because the product only has so much water to shed once while the closure admits water every day. That has a direct commercial consequence: the highest-leverage decision in a moisture-control project is the barrier performance of the closure, and the desiccant is the second-order decision that follows it. A larger bed behind a poor closure buys a few months and raises the cost of every pack; a moderate bed behind a good closure can carry a long shelf life comfortably.

The inputs for that calculation have to be sourced rather than assumed. The product’s moisture sorption behaviour, ideally as an isotherm, should come from the product laboratory or from a sorption analysis, and it defines the critical threshold and the load that has to be removed. The closure’s moisture vapour transmission rate should come from a measurement on the actual closure, or from the closure supplier on a comparable design, since published figures for a generic closure shape are not a substitute. The intended storage humidity, the fill weight and the headspace volume come from the project. Without the isotherm and the closure measurement, the sizing is a guess with a safety factor attached, and the safety factor usually turns out to be the most expensive component of the pack.

Two practical consequences follow for the filling operation, and both are cheaper to fix before the specification is frozen than afterwards. First, the headspace should be as dry as practical at closing, because humid air in the headspace is a load the bed has to carry and because dry air or a nitrogen flush changes that load predictably. Second, the bottle should be dry, and the finish in particular should be dry, because residual rinse water is the most common single reason a correctly sized bed appears to have failed. Neither of those is a desiccant decision, but both are part of the system that the desiccant sizing assumes.

What a Desiccant Closure Asks of the Neck Finish

The full load is a collaboration between the fill weight the brand specifies and the headspace it leaves, but the concentrated load is the desiccant itself, and the list below is the one to work through before the finish is agreed. A desiccant closure is usually taller and often heavier than the plain closure it replaces, so thread engagement length, finish height and the internal clearance of the shell have to be designed together rather than sequentially. A finish that was dimensioned for a shallow plain cap will not necessarily carry a taller desiccant shell with the same thread engagement, and a shortened engagement is a liquid seal risk at torque.

The sealing land needs two properties that a plain cap does not demand at the same level. It needs to be wide enough and flat enough for the closure’s sealing bead to sit on, and it needs to be free of any channel, step or mould joint running across it. On a breathable closure, a discontinuity across the land is not a small leak against a compression liner; it is a permanent path from the room into the headspace, and the bed pays for it continuously. Blistering near the neck seam, a visible joint line and a land that is not concentric with the thread all fall into the same category, and they should be treated as barrier defects rather than as appearance issues.

Thread geometry deserves separate attention because of re-closing. A desiccant pack is normally opened and closed repeatedly over its period of use, and each re-closing has to reproduce the liquid seal without an operator checking it. A thread with a clean lead-in and a consistent start helps the consumer find the thread the first time, and a pitch that has been co-designed with the shell reduces the chance of cross-threading, which is the failure that both breaks the seal and damages the finish. The finish height should also leave enough engagement that a partially closed cap still holds the closure on, since a cap that has been left loose is the normal condition of a pack in use.

The last requirement is dryness at closing, and it is the one that most often conflicts with production. A wet finish wets the membrane, and a wet membrane slows or blocks the vapour path that the whole design depends on, which means a pack that was correctly sized can behave as if it had no bed at all. High-speed filling, foaming products and vertical filling valves all tend to put product near the mouth, so the fill method, the fill rate and the headspace have to be reviewed as part of the moisture-control project rather than left to the line. The interaction between the extra cap height and the capping head, the cap chute and the feeder also belongs in that review, since a taller and heavier closure changes head travel and can change the torque window.

Contact Rules When the Desiccant Shares Space With the Product

There are two fundamentally different situations here, and they carry different requirements. In the first, the desiccant is sealed behind a membrane inside the closure and is not intended to contact the product at all; the membrane is the separation and its integrity is the control. In the second, the desiccant sits openly in the same space as the product, as a loose sachet or an unsealed canister, and direct contact is possible. The second case carries the stricter material and labelling requirements, and in some product categories it is disallowed outright.

Which framework applies depends on the product’s regulatory status and the destination market rather than on the closure. For a food or beverage product, the relevant references are normally the food-contact material regulations of the market, including the FDA food-contact requirements in the United States and the framework of EU 1935/2004 and EU 10/2011 in Europe, together with any national scheme that sits on top. For a pharmaceutical or nutraceutical pack, the relevant references normally include the USP chapters that deal with packaging materials and with the moisture vapour transmission performance of containers, alongside the market’s own drug packaging rules. We name these because they are the frameworks a buyer will be asked about, and we do not give conclusions on them: whether a particular desiccant grade, membrane or closure qualifies is a determination for the brand’s regulatory route, and it depends on the product, the market and whether the desiccant is enclosed or openly present. The practical step is to ask the desiccant and closure suppliers for the grade’s declared regulatory status and to have the brand’s regulatory function confirm the classification before the mould is cut.

Three technical points deserve a place beside the regulatory question. The first is dust. Friable materials shed fines, and a canister or a sachet that is not dust-tight can release those fines into a powder product where they are impossible to see and impossible to separate. A membrane that retains fines, or a genuinely dust-tight sachet, is part of the specification rather than an accessory. The second is aroma. A strong desiccant is not selective, and it can adsorb the volatile fraction of a flavoured product over the shelf life, so a change from silica gel to molecular sieve can change how a product smells even when nothing else has changed. Where aroma is part of the product’s value, the bed choice should be tested against the finished product rather than selected on capacity alone. The third is the warning requirement. Where a desiccant is loose inside a pack, the label normally has to instruct the consumer not to eat it and to keep it out of reach of children, and in some markets that instruction is mandatory and its wording is fixed.

Who Signs Off What: Design Freeze, Sample and Qualification Stages

A desiccant project runs through six stages, and each one ends in a decision that the next stage assumes. Running them out of order is what produces packs that pass a loading test and fail in the market, so the sequence is worth treating as a gate list rather than as a description.

The first stage defines the requirement. The product’s critical moisture content or water activity, the target shelf life, the intended storage humidity and the sensitivity to oxygen as well as to moisture are all written down here, and the outcome of the stage is a decision about whether moisture control is needed at all. This is the stage where a project most often discovers that the product’s limiting failure is oxidation, which changes the countermeasure entirely. The second stage chooses the placement: closure, loose sachet or neck compartment, weighed against available capacity, foreign body and ingestion risk, line complexity and market rules. The outcome is a frozen placement decision and a closure format concept, and this is the point at which the commitment to a mould becomes real.

The third stage sizes the bed. The desiccant family, the grade and the nominal quantity are chosen against the calculation described above, and the assumptions, the safety factor and the source of each input are recorded in writing. The fourth stage co-designs the finish and the closure geometry, then samples. Sampling is where the mechanical questions are answered rather than the moisture questions: whether the closure fits and torques correctly, whether the membrane is intact and its shell joint continuous, whether the bed is retained under vibration and impact, whether the closure feeds and caps on the actual machine, and whether the finish arrives dry in production. The fifth stage qualifies the pack, which means storing the real product in the real pack at the intended and accelerated conditions and measuring the headspace humidity and the product’s moisture over time. The sixth and last stage freezes the specification and the procedural controls: the pouch size, the rule for an opened pouch, the incoming inspection that prevents desiccant closures from sitting in a humid room, and the filling-room conditions the sizing calculation assumed.

The question the qualification study answers is not whether the desiccant works. It is whether the bed still has usable capacity at the end of the shelf life under the worst conditions the pack will meet, which is a different and much harder test. A study that shows the headspace staying low for the first three months demonstrates almost nothing, because the first three months are the easy part; the informative measurement is the last one, taken after the pack has spent its full life at the humidity the distribution route actually delivers.

Reading the Capacity and Shelf Life Target Table

The table below is the working document for this page. It is organised by capacity and shelf life target rather than by product category, because the target is the input an engineer can defend and the product name is not. Read it as a direction-setting tool: it indicates which placement becomes practical at a given combination of target and fill, which desiccant route follows, and what has to be proved before the design is scaled. The loading column deliberately describes an approach rather than a quantity, because the quantity depends on the product’s own sorption data and on a measured closure transmission rate.

What Happens to Protection After the First Opening

A closed pack and an opened pack are two different systems, and the difference is not a gradual degradation of the same performance. While the pack is closed, the bed is working against a headspace whose only significant supply of new moisture is the closure’s transmission rate, which is slow. Once the pack is opened, the same bed is working against a room, and the supply of new moisture is set by the room rather than by the closure. The remaining capacity is the only thing standing between the product and that room, and because the bed was sized with a safety factor for the closed case, the remaining capacity is spent quickly wherever the air is humid.

That has three design consequences. The first is that the closure must re-close in a way that restores the liquid seal, because a pack that cannot be closed properly after opening is exposed continuously rather than intermittently. The second is that a period after opening is a real number which depends on the room, not on the pack, so if the product will not tolerate a short open life, the answer is usually to split the pack into smaller units, so that the consumer opens a small quantity at a time and finishes it, rather than to fit a larger bed. The third is that the labelling has to set the expectation honestly: keep the pack closed, keep it out of humid rooms such as bathrooms, re-close after each use, and do not treat the desiccant as a rechargeable system.

Where the desiccant is a separate object inside the pack, the sachet or canister count can be matched to the intended period of use rather than to the shelf life. A pack intended to be consumed within a month after opening can carry a bed sized for the closed shelf life plus that month, and the spent bed is discarded with the pack. That is a more honest design than a single large bed that is expected to cover an undefined open period, and it also gives the consumer a visible cue that protection has been used up.

Humidity During Transport and Storage Is Half the System

Every capacity calculation for a desiccant pack assumes a humidity outside the pack. The distribution route does not respect that assumption. A container that crosses a warm, humid route and then arrives in a cold warehouse passes through saturation and condensation, and the result is not merely humid air but liquid water on surfaces. A fibreboard carton that has absorbed moisture releases it slowly over the following weeks, and a pallet that is moved from a cold store into a warm room condenses water on every cold surface it contains. Against that, a bed sized for a closed pack in a temperate warehouse is being asked to do two jobs at once.

The countermeasures are system-level rather than closure-level. Where the product requires it, the finished pack is shipped inside an outer barrier, either a foil pouch around the individual pack or a sealed liner inside the carton, and that outer barrier is what keeps the distribution humidity away from the bed. Where the carton itself is the barrier, carton-level desiccant can be used, and for the sea leg a container desiccant is a normal part of the shipping specification. Procedurally, the filled packs should be kept off the dock and out of the yard, warehouse humidity should be monitored rather than assumed, and the desiccant closures themselves should be stored in their sealed pouches until the moment of use. A pouch left open in a humid filling room for a week has already spent part of the capacity that the sizing calculation counted on, and nothing on the pack will show it.

Temperature cycling deserves to be treated as the primary mechanism rather than as a secondary one, because condensation delivers water in liquid form, and liquid water will exhaust a bed far faster than vapour at the same average humidity. This is also where the moisture question touches the rest of the packaging specification, since condensation inside a closed carton affects labels, closures and the carton itself, and the carton and pallet arrangements are set out on the packing specification page. What is worth stating here is that the boundary between the two subjects is not a gap: the desiccant page decides how much water the pack may admit, and the packing and storage pages decide how much water the pack will actually meet.

Finally, the storage and handling of the closure is part of the humidity system. Desiccant closures should arrive and be stored in their sealed pouches, be issued in quantities matched to a shift, and be protected from the filling room until they are used. Documentation for the closures should include the grade of desiccant, the nominal quantity, the pouch quantity and the conditions the material was dried and packed under, so that incoming inspection can confirm that a delivery matches what the sizing assumed. The documentation that accompanies a batch is dealt with on glass bottle quality certificate, and the desiccant closure should be covered by the same discipline rather than by an informal arrangement with the closure supplier.

Which Closure and Packaging Questions Belong on a Different Page

This page covers moisture control and the desiccant scheme, and only that. The choice of closure format, the finish family, the application method and the way a closure is selected for a product belong on the closure guide linked at the top, which is the hub for this cluster, and that page is the right starting point where the open question is which closure to use rather than how to control moisture inside the pack.

The ordinary compression liner is the subject of the cap liners page linked at the top, including liner materials, thickness and the torque behaviour that compresses a liner into a seal. The desiccant closure is deliberately separated from that family here because the two work in opposite directions: one blocks vapour by compression, and the other admits vapour by design. Where a pack needs both a conventional seal and moisture control, the two must be reconciled deliberately, and the most common error is a closure that does both and therefore achieves neither.

The carton, pallet, liner and packaging specification is covered on the packing specification page linked above, and the warehouse and storage conditions are covered on the storage conditions page linked in the same place. Both matter to a moisture-control project because the humidity a pack actually meets is decided there, but neither is a desiccant subject. On the same hub, the test protocol for child-resistant closures is a separate regulatory subject with its own page, and it is worth noting only because a desiccant closure and a child-resistant closure are sometimes needed on the same pack, in which case the two designs have to be combined rather than chosen between. And the counterpart project page on aseptic filling, linked at the top, covers the environment at the moment of closing; between the two pages, the pack’s condition is covered from the fill through the shelf life.

Frequently Asked Questions About Moisture Control in Glass Bottles

Does every powder product need a desiccant?

No, and adding one where it is not needed costs money and usually weakens the closure. The test is whether moisture actually drives the product’s failure mode and whether the critical moisture or water activity is low enough that ordinary headspace humidity will reach it within the shelf life. Products whose limiting problem is oxidation need a different countermeasure, and products that enter the pack with a high residual moisture cannot be rescued by a bed at all. The cheapest and most reliable way to settle the question is a sorption isotherm on the product, read against the humidity the pack will actually meet in distribution.

Is a desiccant cap better than a sachet dropped into the bottle?

They solve the same problem with different constraints, and the choice is usually decided by risk rather than by capacity. A cap-mounted bed removes a foreign body from the pack, needs no extra station on the line and gives a clean consumer experience, but it is limited by the space inside the shell, so it suits packs where the fill is generous relative to the headspace. A sachet or canister offers much more capacity per pack at lower cost, but it introduces an object that can be ingested, can block the neck and can be damaged during filling, and in some markets and categories that is unacceptable. Where the neck is wide enough, a canister in the neck is often the middle path.

How much desiccant does a bottle actually need?

It is a subtraction, not a rule of thumb. Add the water that has to be removed from the product to reach the target, the water that enters through the closure over the shelf life, the water already in the headspace air at closing, and the water that desorbs from the closure assembly afterwards. Multiply by a safety factor, then divide by the working capacity of the chosen material at the humidity the bed will see at the end of its life rather than at its maximum. Work in milligrams of water rather than grams of desiccant. For a long shelf life the ingress term usually dominates everything else, which is why the closure’s barrier is the first thing to improve and the desiccant is the second.

Can silica gel and molecular sieve be used in the same closure?

They can be blended, and blends are sometimes used to cover a wider humidity band, but a blend adds a variable and should be chosen for a stated reason rather than as a hedge. The two materials behave differently across the range: silica gel is strong in the middle and upper humidity range, while molecular sieve holds a genuinely dry headspace. A blend is worth considering where the pack has to be protected during an initial dry period and then across a longer moderate period, and where the product’s aroma tolerance has been checked, since a stronger material can strip volatiles and change the smell of the finished product.

Will a desiccant cap fit my existing neck finish?

It will fit if the finish and the closure geometry are designed together, which is not the same as fitting the cap onto an existing finish. A desiccant closure is normally taller and often heavier than a plain cap, so thread engagement, finish height and the internal clearance of the shell all interact. The sealing land also has to be wide enough and flat enough for the closure’s bead, and free of any channel or mould joint across it, because on a breathable closure a channel is a permanent route from the room into the headspace. The extra height also has to be checked against the capping head travel, the cap chute and the feeder.

Does the desiccant have to be food grade?

It has to be of a grade permitted for the use it is put to, and which scheme applies depends on the product’s regulatory status, the destination market and whether the desiccant is sealed behind a membrane or openly present in the pack. The frameworks a buyer is normally asked about include the FDA food-contact requirements, the EU framework on food-contact materials, and the USP chapters dealing with packaging and moisture vapour transmission for pharmaceutical packs. We do not give conclusions on classification. The practical step is to obtain the grade’s declared status from the desiccant and closure suppliers, and to have the brand’s own regulatory function confirm the classification before the tooling is cut.

How long does protection last after the pack is opened?

It depends almost entirely on the room, not on the pack, because once the pack is open the bed is working against ambient air rather than against the slow transmission rate of the closure. The bed was sized with a safety factor for the closed case, so the remaining capacity is spent quickly wherever the air is humid, and the open life is short relative to the shelf life. The design answers are to make sure the closure re-closes properly, to split the pack into smaller units so a unit is finished soon after it is opened, and to state a realistic period after opening on the label. Fitting a larger bed does not extend the open life in any useful proportion.

Starting a Desiccant Project: The Three Inputs That Matter

A moisture-control project can be moved forward a long way on three inputs. The first is the product’s moisture behaviour: its current moisture content or water activity, the threshold above which it degrades, and ideally a sorption isotherm showing how much water it takes up at each humidity level. Where the isotherm does not exist, that is the first item to commission, because it is the input that cannot be substituted by anything else. The second is the pack geometry: the fill weight, the intended headspace, the container volume and the neck finish, all of which together decide the placement that is even possible.

The third is the commercial and distribution frame: the target shelf life, the storage and transport conditions the pack will actually meet, the destination markets and their labelling rules for a desiccant, and the annual volume together with how it is split into batches. With the three, a reply can set out whether moisture control is needed at all, which placement fits the geometry, which desiccant family and grade the target points to, how the sizing should be calculated and which measurements are still missing, what the finish and the closure have to be designed for, and what the qualification study has to measure at the end of the shelf life rather than at the beginning.

Where a project is already past the tooling stage, the same three inputs still help, because the most common finding at that point is not that the bed is the wrong material but that the closure barrier, the filling-room humidity or the product’s moisture at packing is setting the limit. Those are cheaper to identify from the data than from a complaint. We do not publish indicative MOQ, unit pricing, tooling cost or lead time for desiccant closures; those follow the closure format, the desiccant grade and the pouch arrangement, and a direction on the design can be given far earlier than a number can be given on the price.