How to store soluble mineral specimens
Halite fails above 75.3% humidity, chalcanthite below 35% and pyrite above 60%. The only band that suits all three crystals in one collection is 35 to 45% RH.
There is no single safe humidity, and that is the whole problem. Halite deliquesces above 75.3% relative humidity at 25 °C, sylvite above 84.3%, chalcanthite crumbles to powder below 35%, and pyrite, an iron sulphide, reacts above 60%. The only band that suits a chalcanthite crystal and a pyrite together is 35 to 45% RH, which the published general recommendation of 45 to 60% sits outside.
In short
- The published general figure for a mixed geological store is 45 to 60% relative humidity. That is the right answer for a room of rocks and fossils and the wrong answer for a drawer containing both a chalcanthite and a pyrite.
- Chalcanthite sets a floor and pyrite sets a ceiling, and they nearly meet. Deep blue copper sulphate crystals lose water and crumble to a pale blue powder below 35% RH; iron sulphides should be kept at 45% RH or lower. The overlap is a 10-point window.
- Only three species in this site's 109-sheet reference set record water solubility at all — halite, sylvite and chalcanthite. The soluble group is tiny. The dehydrating group is larger and quieter, and it is the one that loses specimens without anybody noticing.
- Deliquescence points are calculable, not folklore. The saturated-salt relative humidities used here — 75.3% for sodium chloride and 84.3% for potassium chloride at 25 °C — come from the standard NBS polynomial fits and are the same numbers laboratories use to calibrate hygrometers.
- The answer is per-specimen microclimates, not one room. Sealed boxes with conditioned silica gel, one band per box, and a cheap hygrometer in each. Trying to run a whole cabinet at a single humidity guarantees losing one end of the collection.
| Species | Formula | Safe band | What goes wrong outside it | Source of the figure |
|---|---|---|---|---|
| Chalcanthite | CuSO4·5H2O | Above 35% RH | Loses water below 35% and crumbles to a pale blue powder | NatSCA leaflet |
| Pyrite and other iron sulphides | FeS2 | 45% RH or lower | Above 60% reacts with atmospheric water to make a weak sulphuric acid | NatSCA leaflet |
| Halite | NaCl | Below 75.3% RH | Deliquesces — takes up water and dissolves in its own brine | Greenspan, NBS 1977 |
| Sylvite | KCl | Below 84.3% RH | Deliquesces | Greenspan, NBS 1977 |
| Autunite | Ca(UO2)2(PO4)2·11H2O | Stable, middling | Dehydrates to meta-autunite: −9.3% mass, −18% volume, and it flakes | Handbook of Mineralogy |
| Torbernite | Cu(UO2)2(PO4)2·12H2O | Stable, middling | Dehydrates to metatorbernite | Handbook of Mineralogy |
| Heulandite and the platy zeolites | (Ca,Na2)Al2Si7O18·6H2O | Stable, middling; avoid heat | Dehydrates; cracking and loss of transparency | Handbook of Mineralogy |
| A mixed geological store | — | 45 to 60% RH | The published general figure, and wrong for both ends of this table | NatSCA leaflet |
The conflict table, and the ten-point window it leaves
Put the published figures on one axis and the conflict is immediate.
From the Care and Conservation of Geological Specimens leaflet: a dry environment damages specimens, and deep blue copper sulphate crystals will lose water and crumble to a pale blue powder in relative humidity (RH) below 35%. Pyrite-containing specimens should not be stored in conditions above 60% RH and preferably should be stored in conditions of 45% RH or lower. A mixed geological store should be held between 45% and 60% RH.
From Greenspan's 1977 NBS tables of humidity fixed points: a saturated sodium chloride solution sits at 75.3% RH at 25 °C and saturated potassium chloride at 84.3%. Those are the deliquescence points — above them, halite and sylvite take water from the air and begin to dissolve in it.
Now intersect them. Chalcanthite needs above 35%. Pyrite wants 45% or below. The band that satisfies both is 35 to 45% RH — ten percentage points wide. Halite and sylvite are comfortable anywhere in it, because their thresholds are far higher.
And the published general recommendation of 45 to 60% touches that window at exactly one point, its upper end. A store run to the general figure is a store run at the top of the acceptable band for iron sulphides and above it for most of the range. That is not a criticism of the guidance, which is written for whole collections including fossils, bone and shale — sub-fossil specimens, the same leaflet notes, should not be stored below 45% RH. It is an argument that a mineral collection cannot be run as one environment.
Only three species here are actually soluble, and that is not the problem
Method. On 20 September 2026 the Handbook of Mineralogy sheet for every species this site covers — 109 sheets after removing duplicates, mirror copies and non-species files — was searched for any statement of solubility in water.
Three matched: halite, sylvite and chalcanthite. That is it. Out of 109 species, three carry an explicit water-solubility note.
The temptation is to conclude that soluble specimens are a small problem. The opposite is true, and the reason is that solubility is visible and dehydration is not. A halite that has been in a damp room is obviously wet; nobody misses it. A specimen that has quietly changed hydration state looks the same until it falls apart.
The same search turned up the quiet group: autunite and torbernite dehydrate to meta-autunite and metatorbernite, and heulandite is recorded as dehydrating. The arithmetic for the autunite case is on is autunite dangerous — a 9.3% loss of mass, a 10% gain in density and an 18% loss of volume, which is why dehydrated autunite flakes. The pharmacosiderite page describes a species whose formula is written 6-7 H2O precisely because the occupancy varies with the air around it.
So the rule that follows is not about solubility at all: any species whose formula contains water is a species whose condition depends on the air in the box.
What to actually do, per specimen
1. Sort by band, not by species. Three groups: keep dry (iron sulphides, 40 to 45%), keep middling (hydrated species, 40 to 50% and stable), keep below deliquescence (halite, sylvite, anything soluble — under 60% is plenty). Most collections need two boxes, not thirty.
2. Use sealed clear boxes with conditioned silica gel. Silica gel conditioned to a target humidity holds that humidity rather than drying everything out, which matters because uncontrolled desiccant is how chalcanthite is lost. A clear lidded box also lets you look without opening, which the Care and Conservation of Geological Specimens leaflet recommends for exactly that reason.
3. Put a cheap hygrometer in each box and read it. An inexpensive digital hygrometer is accurate enough to tell 40% from 60%, which is the whole decision. Calibrate it if you want to be careful: a sealed jar with saturated sodium chloride reads 75.3% at 25 °C, which is the same number in this article's table, used the other way round.
4. Do not chase stability you cannot hold. The leaflet warns that rapid or severe fluctuation — greater than 10% in an hour — splits specimens. A cupboard in an interior room beats a display cabinet on an outside wall, and both beat an elaborate system that gets switched off.
5. Keep the label out of the specimen's microclimate. Acid from a decaying sulphide attacks paper, and a specimen that survives without its label has lost most of what made it worth keeping. Labels and cataloguing are on how to label and catalogue a collection, and the damp-house case is on storing a collection in a damp house.
Where the numbers come from, and how far to trust them
Two of the four figures in this article are measurements from the physical chemistry literature and two are conservation recommendations, and they are not the same kind of number.
75.3% and 84.3% are thermodynamic. They are the equilibrium relative humidities above saturated solutions of sodium and potassium chloride at 25 °C, fitted from measurement, and they are used as humidity calibration standards. They move with temperature — a few tenths of a percent per degree — and otherwise they are fixed properties.
35%, 45% and 60% are practice. They come from a conservation leaflet written for collections staff, and they encode judgement about margins and about what is achievable in a real store. They are not thresholds at which something switches on. Treat them as the centre of a soft band rather than a cliff edge, which is how the people who published them intend them.
One honest gap. The exact relative humidity at which chalcanthite's transition from the pentahydrate takes place is a number we did not find in a source we could verify, so this page uses the conservation figure of 35% and does not invent a thermodynamic one. That distinction — reporting the practice figure and declining to manufacture a physical constant — is the difference between a page you can check and one you cannot.
We sell nothing and hold no stock; the only commercial route on this site is the wanted list, and the rest of this section is indexed at collecting guides.