Not every mark on marble is a stain. The ICOMOS-ISCS glossary, the international classification of stone deterioration, places stains among chromatic alterations and dulling among losses of material. Two distinct families. A loss of material cannot be cleaned: it is repolished. Aggressive cleaning makes it worse.
This page deals with remedial treatment, stain family by stain family. Routine cleaning is covered in how to clean marble day to day. The routine for each type of surface is covered in marble care and maintenance. Here, we deal only with marks that are already there.
Key points
- Before choosing a product, decide between stain and dulling. The ICOMOS-ISCS glossary classes them in two different deterioration families, and only the first can be cleaned.
- Six families of stain, six logics: grease, organic, metallic, white mark, biological, dye. The white mark itself splits into four mechanisms that the market confuses.
- The poultice is the only extraction method described step by step by an institution: 7 parts glycerine, 1 part sodium citrate, 6 parts warm water, left for 48 hours (U.S. General Services Administration).
- Three chemical families, three risks. Sodium citrate complexes iron without acidity, hydrosulphite reduces it (30 minutes per application). Oxalic acid, for its part, was excluded from a conservation laboratory’s tests because its pH was too low.
- No institutional source sets a point of no return in time. The documented criterion is depth: an iron stain diffuses to about 1 mm at most in a Carrara marble.
Stain or dulling: diagnosis comes before the product
Getting the family wrong costs you the surface. A stain is, according to the ICOMOS-ISCS glossary, an “area of limited extent showing a chromatic alteration and having an unpleasant appearance”. Foreign matter has coloured the stone. Dulling is the name the same glossary gives to the increase in roughness of an originally polished surface. The glossary places it among the features induced by loss of material: the stone itself has gone, locally.
The glossary also explains why this loss shows up as light. It defines discolouration as an increase in lightness caused, among other things, “by loss of polish, generally very superficial”. It specifies that “dark marbles and those of bright colours often undergo discolouration”. A surface that has become rough diffuses light instead of reflecting it. That is the whole mechanism of the white mark on black marble, which is almost never a deposit.
The distinction is made in raking light. A dulled area stays matt and flush with the surface, a deposit stands proud. This is a reasoned observation based on the ICOMOS definitions, not a standardised test.
The six families of stain in one table
This table tells you, before you open a bottle, what is reversible. The four “white mark” rows are four distinct deterioration patterns in the ICOMOS-ISCS glossary. Everyday language lumps them all together as “white haze”.
| Family | Origin | What you see | Treatment route | Reversibility |
|---|---|---|---|---|
| Greasy (oil, butter, cosmetics) | Grease deposited on the surface, which then migrates into the pore network | Area darker than the stone, diffuse outline, which darkens over time | Extraction by poultice with a white absorbent filler | Good as long as migration remains shallow; darkening may persist |
| Organic (wine, coffee, tea, fruit) | Food or plant dye in aqueous solution | Pinkish-brown to dark brown halo, outline often sharp | Extraction by poultice, agent chosen according to the dye | Varies with the dye and the depth reached |
| Metallic (rust) | Iron taken into solution, carried by water, then precipitated as oxyhydroxides in the stone | Orange to brown halo, sometimes arranged in concentric rings | Complexing poultice, or reducing agent for stubborn stains (GSA procedure) | Depends on the depth and stability of the mineral formed, not on time |
| White mark : dulling | Loss of polish after contact with an acid | Matt area on a shiny background, recessed or flush, following the outline of the spilt liquid | Repolishing | No cleaning changes anything: it is a loss of material |
| White mark : efflorescence | Soluble salts from the pore network, drawn up then evaporated at the surface | Clusters of white, powdery, loosely bound crystals that do not adhere | Dry removal, then treatment of the moisture source | Good at the surface, but recurs as long as water circulates |
| White mark : encrustation | Precipitation from solutions that have percolated through the stone | Compact, hard and strongly adherent surface layer | Specialist intervention | Restricted: ICOMOS warns that removal may be accompanied by detachment of stone |
| White mark : film | Foreign product: wax, adhesive, treatment residue | Thin, homogeneous, milky covering that follows the surface | Stripper suited to the nature of the film | Good: the film adheres but does not penetrate the stone |
| Biological (mould, algae) | Colonisation by organisms where moisture persists | Grey, green or black patches, often in joints and poorly ventilated corners | Detergent with pH 7 to 10, then removal of the cause of moisture | Good at the surface; certain to recur if moisture remains |
| Ink and dyes | Pigment or dye, often carried by a solvent | Line or patch the colour of the product, sharp outline | Extraction by poultice, agent dictated by the data sheet of the staining product | Variable; risk of spreading if the solvent is poorly chosen |
The reversibility column summarises the mechanisms described by the ICOMOS-ISCS glossary and by the GSA poultice procedure. It is not test data, and no standard measures it.
Each family, its cause and its reversibility
Grease stains: oil, butter, cosmetics
A grease stain darkens the stone because it gets into it. The grease fills the pore network and changes the refractive index of the surface. It keeps migrating after the spill, which is why a halo seems to spread and darken for several days. The treatment route is extraction, not surface dissolution. These stains are the most frequent cause of complaint on marble worktops. The source is permanent there.
Organic stains: wine, coffee, tea, fruit
Wine and coffee pose two problems in one. The dye stains. The acidity of the drink can dull the same area. After the dye has been extracted, a matt mark sometimes remains that is no longer a cleaning matter. In other words, an organic halo treated successfully can reveal dulling previously masked by the colour.
Metallic stains: rust
Rust is not deposited; it migrates. The iron goes into solution, then water carries it into the pores. It then oxidises and precipitates as insoluble oxyhydroxides inside the stone (Reale et al., Molecules, 2023). Wiping the surface therefore has no effect. And the iron can come from the stone itself as much as from an object placed on it. The difference between deposited rust and native rust governs the whole treatment.
White marks: four mechanisms not to be confused
Only one of the four white marks can be removed without risk. The film, of foreign origin, comes off with a suitable stripper. Efflorescence can be removed but returns if the water source remains, and it can crystallise beneath the surface. The ICOMOS glossary calls that case subflorescence and notes that it often leads to loss of stone through scaling. Encrustation adheres strongly, to the point that removing it can tear off material. Dulling, finally, cannot be cleaned.
Biological stains: mould and algae
A biological stain is a symptom of moisture, not of a lack of cleanliness. The bleach reflex should be avoided here. The producer Levantina explicitly rules out bleach, ammonia and descalers on this type of stone. It allows only detergents with a pH of 7 to 10. Treating the colonisation without treating the ventilation or the leak guarantees it will come back.
Ink and dyes
Ink is treated by extraction, with a specific risk: spreading. The solvent that mobilises the pigment can also carry it sideways through the pores and enlarge the affected area. No institutional source publishes a formulation by type of ink. The only reliable information remains the data sheet of the staining product and that of the solvent being considered.
The poultice, the only extraction method documented step by step
A poultice est une pâte formée d’une charge absorbante blanche et d’un agent actif. On l’applique sur la tache et on la maintient humide, pour extraire la substance vers l’extérieur de la pierre au lieu de la repousser vers l’intérieur. La seule procédure institutionnelle publiquement disponible est celle de la U.S. General Services Administration, écrite pour les taches de rouille sur calcaire et marbre.
It gives a formulation, fillers and times. The standard poultice combines 7 parts glycerine, 1 part sodium citrate and 6 parts warm water. The glycerine keeps the paste moist. The permitted fillers are white and absorbent: moulding plaster, untreated white flour, tissue paper, kitchen roll, powdered chalk, talc, fuller’s earth, acid-free paper pulp, whiting. The application time is about 48 hours.
The times change radically with the agent, and that is the point consumer pages leave out. Sodium hydrosulphite, a reducing agent, is left in place for only 30 minutes per application. A commercial rust remover is limited to 24 hours at most. The GSA also warns that “this treatment may cause some etching of polished surfaces”. It adds that this etching “can be corrected by repolishing”. A poultice is therefore never a trivial step on a polished finish.
What makes a stain worse instead of removing it
The substrate is soluble in acid, and that is where most recipes fail. Marble and limestone are made of calcium carbonate. Oxalic acid is often presented as the stone stain remover. Yet a conservation laboratory withdrew it from its tests “because the pH is too low and damages the stone surface” (Cezar, 1998). The same publication notes that “the solubility of carbonate increases down to pH 5, beyond which irreversible decomposition occurs”.
Commercial rust removers deserve the same caution. The Natural Stone Institute writes that these removers “contain traces of hydrofluoric acid”. It adds that products based on lemon or vinegar can dull or etch calcareous stones. No institutional source publishes the composition of rust removers sold in France. Only the product’s data sheet and safety data sheet give that information.
Bicarbonate, for its part, poses a mechanical rather than a chemical problem. It is alkaline and does not attack calcium carbonate. The risk is abrasion. The Natural Stone Institute points out that scouring powders contain abrasives liable to scratch some stones. And calcite sits at 3 on the Mohs scale (Earth Sciences Museum, University of Waterloo). A poorly rinsed salt residue can also recrystallise as efflorescence.
Depth decides, not time
No institutional source sets a time beyond which a stain becomes permanent. The GSA states none. What is measured is depth. On an artificially stained Carrara marble, iron diffusion into the calcitic matrix reaches about 1 mm at most. On a historic sample, calcium becomes the predominant element again only at about 1.7 mm (Reale et al., Molecules, 2023). The same study observes that the iron evolves into more stable phases, which are therefore harder to extract.
The practical consequence is simple. A poultice works on the first few tenths of a millimetre. What has gone deeper does not come back up by surface chemistry. Acting early is still useful, not because a countdown has started, but because the depth reached increases with contact time.
This sensitivity can in fact be measured before purchase. The NF EN 16301 standard defines a test for determining the resistance of natural stone to accidental staining. A test value can therefore be requested from a supplier, batch by batch, just like water absorption. This also applies to marble tiles intended for a wet room.
Frequently asked questions
How can you tell whether it is a stain or a permanent mark?
By checking whether material has been added or removed. A stain adds colour. A loss of material removes polish and leaves a matt area following the outline of the spilt liquid. The first calls for extraction, the second for repolishing. The ICOMOS-ISCS glossary places them in distinct deterioration families.
After how long does a stain become irreversible?
No institutional source gives a duration. The times in circulation are not substantiated. The documented criterion is the depth reached and the stability of the mineral formed. For an iron stain, measured diffusion does not exceed about 1 mm in a Carrara marble (Reale et al., 2023).
Is bicarbonate safe on marble?
No, but not for the reason often given. It is alkaline and does not attack calcite. The risk is scratching a surface whose key mineral sits at 3 on the Mohs scale. Add to that a salt residue liable to recrystallise. Vinegar, by contrast, acts chemically. It is the reagent used to identify a limestone.
A stain that disappears when the stone is wet: is it serious?
That is typical behaviour for a dulled area. When wet, the roughness is optically filled and the mark fades, then reappears as it dries. No product corrects this; only repolishing does. The details of the diagnosis and the three levels of severity are in the article on white marks on black marble.
Key points
Identifying the family before acting avoids most of the damage. The six families of stain share neither their mechanism, nor their treatment route, nor their degree of reversibility. And the white mark alone hides four of them. The poultice is the only extraction method described by an institution. That same institution warns that it can dull a polished finish.
Two marks deserve a separate diagnosis, because they are the most often mistreated: the white mark and rust. Continue with the diagnosis of white marks on black marble, or with the difference between deposited rust and native rust. To stop having to remove stains, prevention lies in the care routine, surface by surface.
