Water tends to leach metals near the surface and deposit them deeper through precipitation in two layers. The upper layer is usually made of metal carbonates, oxides, and occasionally pure or native metals, while the second and deeper layer is metal sulfides. The following excerpt is summarized above.

Source: "Secondary Enrichment of Ore Deposits" by Samuel Franklin Emmons, which was published in the American Institue of Mining Engineers, Transaction 30, (1901), p. 177-201.

The paper does not go into detailed chemistry for the most part but mentions that the chemistry varies from one morphology of soil/rock to another. The author postulates a thermodynamic reason along the lines of reduction potential that facilitates this stratification:

The actual changes observed by me in a great body of pyrite ($FeS_2$) carrying galena (PbS or Lead(II)Sulfide) in a limestone ($CaCO_3$)country-rock, which had undergone partial decomposition from the periphery inwards, are as follows: the original fresh pyrite or marcasite crystals are first disintegrated and slightly pitted on the surface, then changed to melanterite or hydrated ferrous sulfate and the galena becomes anglesite. In the outer or more fully oxidized zone, the iron-vitriol has changed in part to yellow, basic sulfate; in part to limonite with a separation of native sulfur.

The theoretical changes that are assumed to take place by the action of waters carrying oxygen or oxidizing agents are: first, an alteration of the iron sulfide ($FeS$) to ferrous sulfate (Iron(II)Sulfate or $FeSO_4$) with the formation of sulfuretted hydrogen (Hydrogen Sulfide? $H_2S$) and sulfur which may have oxidized to sulfuric ($H_2SO_4$)or sulfurous acid ($H_2SO_3$). By further oxidation the ferrous sulfate will become, in part at least, ferric sulfate (Iron(III)Sulfate or $Fe_2(SO_4)_3$), and this in its turn will react upon the remaining ferrous sulfate, or upon the sulfides, and form more ferrous sulfate or sulfates of the other metals which are present. By this cycle of reactions, a supply of both ferric and ferrous sulfates would seem to be provided in the oxidized zone, but the extending downwards of the ferric salts would decrease as the supply of oxygen in the waters became less abundant. It may be assumed that the sulfates of the metals thus formed would be transported for greater or lower distances, generally in proportion to their solubility, the iron sulfates being the most soluble; next, those of copper and zinc; silver sulfate is less soluble and also more readily decomposed, while lead sulfate is extremely insoluble.

  • $\begingroup$ ""The paper does not go into the chemistry of this and only mentions that the chemistry varies from one morphology of soil/rock to another."" This was a rather wise statement by the author, who obviously did not know a lot about chemistry. $\endgroup$
    – Georg
    Aug 25, 2013 at 19:55
  • $\begingroup$ Asked this question on a Geology forum. $\endgroup$
    – Dale
    Aug 28, 2013 at 3:29

2 Answers 2


Why Do Carbonates, Oxides, and Pure Metals Precipitate Before Metal Sulfides?

They don't. I cannot get to the particular paper you read, but I can provide both a logical and a quantitative argument that the metal sulfides precipitate first. The thermodynamics of this situation do not involve reduction potentials so much as simple solubilities. Metal sulfides are much less soluble in water than metal carbonates.

From your description, I imagine a system like the following image. There are two strata of sedimentation: an upper layer of metal oxides and a lower layer of metal sulfides.

Sedimentation strata diagram showing two strata: a metal sulfide layer on the bottom and a metal oxide and carbonate layer on the top

There are three reasonable explanations why the sulfides are on the bottom.

  1. The sulfides did not precipitate from the water; they were always there. The carbonates and oxides deposited on top of them.
  2. The carbonates and oxides were deposited first, and the sulfides were deposited second, settling through the carbonates and oxides to form a deeper layer.
  3. The sulfides were deposited first, and the carbonates and oxides were deposited second.

Explanation 1 is very reasonable, but counterproductive. We will ignore it.

Explanation 2 is the one supposed by the OP in the question, and perhaps it was also provided in the paper. Let us examine this idea a little. In order for this to happen, the metal sulfides would have to be denser than the metal carbonates and oxides to settle through. All it takes the counterexample of iron to see that this is not always the case. Iron (II) sulfide is denser than iron (II) carbonate but less dense than iron (II) oxide. - Iron (II) oxide, $\ce{FeO}$, 5.745 g/mL - Iron (II) carbonate, $\ce{FeCO3}$, 3.961 g/mL - Iron (II) sulfide, $\ce{FeS}$, 4.84 g/mL

Additionally, if the sulfides were settling through, the layers would not be so well stratified - the boundary between them would be indistinct.

Sedimentation strata diagram showing three strata: a metal sulfide layer on the bottom, a metal oxide and carbonate layer on the top, and a mixed layer with all three in the middle.

Metal sulfides are much less soluble in water than metal carbonates.

That leaves with Explanation 3, which is the correct one. Metal sulfides are almost universally less soluble that metal carbonates. We'll get to where the oxides come from later. Both metal carbonates and metal sulfides are nearly insoluble in water, so the minute amounts present are hard to quantify in the "grams of solute per liter of solution" kind of way. The saturated solution may only have few parts per billion of the compound. Instead, we talk in terms of the solubility equilibrium, in particular, the solubility product (constant), $k_{sp}$, the equilibrium constant associated with the combination dissolution and dissociation.

We can determine the concentration from the $k_{sp}$. Consider barium carbonate $k_{sp}=5.1\times10^{-9} \text{ M}^3$. The solubility equilibrium is:

$$\ce{BaCO3(s) <=>Ba^{2+}(aq) + CO3^{2-}(aq)}$$

And the equilibrium constant is (the concentration of solid barium hydroxide is constant and is thus part of $k_{sp}=k_{obs}$. $$k_{sp} = [\ce{Ba^{2+}}][\ce{CO3^{2-}}]=5.1 \times 10^{-9} \text{ M}^3$$

Since $[\ce{CO3^{2-}}]=[\ce{Ba^{2+}}]$ (mass balance): $$[\ce{Ba^{2+}}][\ce{CO3^{2-}}]=[\ce{Ba^{2+}}]^2$$ $$[\ce{Ba^{2+}}]^2=5.1\times 10^{-9} \text{ M}^3$$ $$[\ce{Ba^{2+}}]= 7.14\times 10^{-5} \text{ M}=>0.0141 \text{ grams per liter}$$

Now, let's look at carbonates and sulfides. All $k_{sp}$ came from here.


  • Carbonate $\ce{CdCO3}$, $k_{sp}=5.2\times 10^{-12} \text{ M}^2$, $[\ce{Cd^{2+}}]=2.28\times 10^{-6} \text{ M}$
  • Sulfide $\ce{CdS}$, $k_{sp}=8\times 10^{-27} \text{ M}^2$, $[\ce{Cd^{2+}}]=8.94\times 10^{-14} \text{ M}$

Cadmium sulfide is less soluble than cadmium carbonate by a factor $10^8$ (100 million). The same behavior exists for cobalt, copper, iron, lead, manganese, nickel, silver, and zinc (all metals for which data for both the carbonate and sulfide are listed).

Why might this paper suppose that the sulfides are deposited second when the sulfides are much less soluble than the carbonates?

The authors of the paper probably did not know that (and perhaps few others did either). That paper was published in 1901. Svante Arrhenius presented his thesis on the behavior of salts in water in 1884. It was not well thought of initially. He received a Nobel Prize for the work in 1903 (after the paper was published), but there is little reason to believe the authors (mining engineers) would have been taught controversial and cutting edge physical chemistry in whatever minimal introductory chemistry courses they had taken (if any). I may mention recent Nobel work in chemistry in my intro courses, but I certainly don't teach recent Nobel work in chemistry in into courses.

Where do the metal oxides come from?

Metal oxides form either by decomposition of the carbonates or by oxidation of pure metals (which probably occur from the decomposition of the sulfides):

Decomposition of sulfides:

$$\ce{FeS(s) <=> Fe(s) + S(S)}$$

Oxidation of metal:

$$\ce{6Fe(s) + 3O2(aq) -> 3Fe2O3(s)}$$ $$\ce{2Fe(s) + 3H2O(l) -> Fe2O3(s) + 3H2} \text{ (acid is helpful here)}$$

Decomposition of the carbonate. Since the carbon dioxide would be carried away by the water, this equilibrium is continually pushed toward the oxide. $$\ce{FeCO3(s) <=> FeO(s) + CO2(aq)}$$


In marine environment, reducing conditions generally occur a short distance below the depositional interface and sulfur is available in solution and organic matter. The result is reduction of iron, first to mono-sulfides and then to pyrite (see Chap. 19). p.403.

Iron minerals are forming today in several environments:

(a) tidal flats and bog lakes

(b) oceanic shelves in tropical climates

(c) deep basins with restricted circulation, such as fjords and the Black Sea

The formation of pyrite in muds has been the subject of several theoretical, experimental, and field studies. Sulfur isotope analysis has helped to throw light on the origin of the sulfur. These studies have been summarized by Berner (1970) and his conclusions are given below. He deduced that the fulfur in the pyrite of modern muds comes from two sources (a) organic matter and (b) sulfate dissolved in seawater. Although organic matter forms up to 10% of some modern marine muds, this carbonaceous material contains only 1% sulfur. Yet the muds frequently contain more than 1% pyrite and so it is clear that an additional source of sulfur is required. The other source is the bacterial reduction of dissolved sulfate, which has been shown by sulfur isotope studies to be supplied by diffusion into the sediment from the overlying seawater. Where dissolved $H_2S$ exists, pyrite is generally the thermodynamically stable form of iron; pyrite is not formed directly however.

  • The first iron compounds to appear are black, finely disseminated iron mono-sulfides.

They can tribute much of the black color generally seen just beneath the surface of tidal flats and organic-rich lakes and stagnant basins. These compounds, which are thermodynamically meta-stable, are observed to change to pyrite a few inches below the sediment-water interface. Berner has presented evidence indicating that

  • the formation of pyrite takes place mainly by the reaction of mono-sulfides with elemental sulfur produced from the oxidation of $H_2S$ either inorganically or through the action of sulfur-oxidizing bacteria.

  • If the bottom waters contain dissolved oxygen, elemental sulfur is produced by the reaction of FeS and $H_2S$ with dissolved oxygen stirred into the sediment by storms, currents, and burrowing organisms.

The fact that free $H_2S$ is present in a sediment, or even in the deeper parts of basins with restricted circulation, does not necessarily mean that iron sulfides will be the only iron minerals present. In the Black Sea deeper waters have a slightly higher salinity (22 ppt) than the surface waters (18 ppt). The result of the density stratification has been to prevent mixing between the surface, oxygenated waters, and the waters below a depth of 150m with the result that the waters from 150m to depths in excess of 200 m contains dissolved $H_2S$. Muds in the deeper parts of teh Black Sea contain 3 to 5% total iron, on a carbonate-free basis. Normally 30 to 50% of this iron is present as iron sulfide. Not all the iron is present as pyrite because over 95% of the iron supplied to the Black Sea is in the form of very fine clastic particles (magnetite, hematite, chlorite), which are not converted to sulfides as they settle through the deeper waters despite the high content of $H_2S$ in these waters. Once buried below the sediment interface, the conversion of the iron compounds to sulfides is limited by the availability of organic matter, sulfur, and perhaps also by slow reaction rates. This situation occurs in spite of the high content of organic matter, which reaches values in excess of 5% organic carbon. (To me this is confusing because "organic matter" first causes pyrite and then doesn't... perhaps the word "normally" was meant to apply to pyrite formation and the latter to this special case of increasing salinity at depth.)

The "bog iron ores" of modern swamps and lakes represent one of the few examples of modern iron-rich sediments. They are found in pororly drained, recently glaciated regions or areas of older iron-rich rocks. Two general groups have been recognized: lake ores and bog ores. Lake ores consist of oolitic or pisolitic grains, cemented together into discs up to a foot or more across. They are commonly found in water only a few feet deep, at the border of the lake. Bog ores form thin layers of earthy-to-pisolitic iron oxides at the surface or below several meters of peat. Both types appear to be formed by the migration of acidic organic-rich groundwater. Precipitation of ferric iron takes place where the migrating soil or swamp-derived waters enter a more oxidizing, less acidic environment.


Blatt, Harvey, Gerard V. Middleton, and Raymond C. Murray. Origin of Sedimentary Rocks. Englewood Cliffs, N.J: Prentice-Hall, 1972. Print.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.