In my opinion, there is a kind of a mess with semantics. One could use the term ‘silica undersaturated’ for all systems which lack normative quartz, regardless of whether they contain normative nepheline, or not. Formally, this is not untrue because both nepheline-normative and olivine&opx-normative rocks are really undersaturated by SiO2 (from the point of view of a chemist). But I agree it would be better to use a uniform terminology.

Hi Petr,

how do you define an igneous rock with opx? Silica-undersaturated? Why an olivine-orthopyroxene-bearing (or -normative) rock should be defined as silica-undersaturated rock?

I continue to believe that the igneous world must be divided into three categories, not only two.

Hi Michele,

I do not oppose the three-fold division you mean, I am just trying to understand the reason for this terminological confusion. Perhaps, the terms 'nepheline-normative', 'ol&opx-normative', and 'quartz-normative' seem more appropriate than those which directly refer to silica saturation. The term 'critically saturated' describes the system at the threshold of free SiO2 phase stability; it is suitable for the Pl-CPx-OPx plane, but not for the Ol-Pl-CPx-Opx space?

silica saturation The concentration of silica (SiO2) in an igneous rock, relative to the concentration of other chemical constituents in the rock which combine with the silica to form silicate minerals. On this basis, three classes of igneous rock are recognized: (a) silica-oversaturated rocks (e.g. granite), in which there is more than enough silica to satisfy the requirements of all the major silicate minerals, the free silica appearing as quartz in the rock; (b) silica-saturated rocks (e.g. diorite), in which there is just enough silica present to satisfy the requirements of all the major silicate minerals, there being neither an excess nor deficiency of silica, resulting in a lack of both quartz and feldspathoid minerals in the rock; and (c) silica-undersaturated rocks (e.g. nepheline syenite), in which there is not enough silica present to satisfy the requirements of all the major silicate minerals, the silica deficiency being accommodated by the crystallization of feldspathoids (nepheline, leucite) in place of feldspar, the feldspathoids containing less silica in their structure than feldspars.

http://www.encyclopedia.com/doc/1O13-silicasaturation.html

Thanks Lala,

but your reply is quite obvious and generic.

I was asking something of deeper. Please read carefully my original message. What happens if you do not have large amounts of alkalies and the little Na is stored in cpx (e.g., in classical upper mantle rocks)?

michele

Comparing the degree of silica saturation may be somewhat confusing because it depends on three balances: (1) Opx-Ol, (2) Ab-Ne, (3) Ort-Le which may be, or may be not, coupled. For example, which rock is more silica undersaturated, one rock with 25 wt% Ol and no Ne (picrite), or another with 5 wt% Ne and 1% Ol (felsic nepheline syenite)? In the same way, which rock is more undersaturated, one with 25% Ne and no Le (nepheline syenite) , or another with 2% Na and 15% Le (synnyrite)? Depending of the whole rock composition all these combinations are posible.

This is why we devised a Silica-Saturation Index (SSI = (Q/[2(Ne + Ka )+ Ol])-1, where Q, Ka, and Ol are calculated as molar equivalents from decomposition of the CIPW normative Ort, Le, Ab, and Hy into their end-members and addition of these to the previously calculated Q, Ne, Ka, and Ol, if any. A SSI < 0, SSI = 0, and SSI > 0 means that the rock is silica undersaturated, just saturated, or oversaturated respectively. This permits comparing the degree of silica saturation in rocks of very different composition.

Bea et al.: Deep-seated kalsilite-bearing rocks of the Reguibat Rise, West African Craton: The Awsard massif, South Morocco. Earth Science Reviews (just accepted, pending of minor revisions)

Very interesting Fernando.

However I did not understood well the significance of the SSI. I think I have to wait until the final acceptance of your ESR paper.

First of all I never heard about synnyrites. Now I have realized (checking the LeMaitre et al. IUGS book) that it is a kalsilite-syenite.

I think that the presence of Na- or K-feldspathoids depends on the abundance of Na and K in the magma. If there is K available and there is normative nepheline but is no K-feldspathoid (leucite) the rock is not strongly silica-undersaturated. If there is normative nepheline AND leucite, the rock is strongly silica-undersaturated. If there is normative nephelline AND leucite+kalsilite the rock is even strongly silica-undersaturated. If there is normative nepheline AND ONLY kalsilite the rock has the maximum degree of silica undersaturation. Plagioclases and alkali feldspars may or may not be present in presence of these foids (classical concepts proposed in 1902 by Cross, Iddings, Pearson and Washington).

My main problem was, however, referring to the alkali-poor systems (e.g., picrites or peridotites). A picrite (olivine-rich rock) with no CIPW-nepheline but with CIPW-plagioclase is certainly less silica undersaturated (it is NOT silica-undersaturated) compared with a rock with low olivine and 5% CIPW nepheline. A nepheline-syenite is certainly less silica-understaturated than a kalsilite-syenite. A kalsilite-syenite (i.e., alkali feldspar- and kalsilite-bearing rock) is certainly less silica-undersaturated than a kamafugite or a true leucitite (officially a rock without alkali feldspars or plagioclases).

According to CIPW calculation scheme, we cannot have kalsilite-bearing syenites. A synnyrite, being a type of a syenite, must contain alkali feldspar. According to CIPW norm, a rock with alkali feldspar cannot contain kalsilite. It depends what scheme do you follow.

What is the SSI value of a rock with no Q, Ne and Ka (e.g., a peridotite)? Can you use this parameter only on CIPW grounds, without taking into consideration the true modal analysis of a rock? In what case can you obtain a SSI value = 0?

michele

Hi Michele,

What is important in the silica saturation concept is the kind of mineral-melt equilibria that a given composition allows to realize. However in multiphase-multicomponent systems, like silicate melts, equilibria are strongly dependent on pressure. In the simple Mg2SiO4-SiO2 system there is also an intermediate compound (Enstatite) whose melting behavior at high P is incongruent to become congruent at low P. The consequence is well known: the same ol+opx rich peridotitic source may produce, on account of the melting degree, basalts whose low pressure cotectic paths are controlled by a thermal barrier.

Resuming silica saturation is a dynamic concept that explains some peculiar behavior of silicate melts.

Fernando,

The simple measure of silica undersaturation is just an amount of normative foids and olivine, isn’t it? We just have to keep in mind that one Ne molecule requires 2 SiO2 to be ‘neutralized’, while Ol and Lc take one. For example, a Ne-free rock with 25% Ol is more undersaturated than that with 5% Ne and 1% Ol. Is it necessary to add a special coefficient like SSI?

Hi Pietro,

of course the olivine-enstatite relations are well known aspects for more than 40 years. My question was more detailed: Can I consider an assemblage made up of olivine+orthopyroxene (fo-rich + en-rich) as silica-saturated rock? It is not a matter of what may happen to such a rock if it experiences partial melting.

One of our colleagues believe that the presence of olivine means silica-undersaturation of the rock, but I disagree with him.

Petr:

according to your definition, the presence of olivine (without nepheline or other foids) automatically means that a rock is silica-undersaturated. I do not believe that this is correct.

I also do not believe that a rock with 25% olivine is more undersaturated than a rock with 1% CIPW-ol and 5% CIPW-ne. This because I can have a mantle rock with more than 25% olivine, but also with 10-20% enstatite. Do you believe that a rock with 10-20% enstatite is more silica-undersaturated than a rock with 5% nepheline?

michele

Any equilibrium depends on pressure. If you use the Silica saturation concept for classification purposes you have as a hidden assumption of the CIPW normative calculation that are related to a low-pressure cryistallization sequence. The contraddiction arises just from pretending to make some mineral appear at a pressure at which it cannot be present in the phase diagram...

Hi Michele,

Let's clarify what do we understand under the term 'saturation'. Wikipedia tells: "In physical chemistry, saturation is the point at which a solution of a substance can dissolve no more of that substance and additional amounts of it will appear as a separate phase ". In other words, rocks which contain free SiO2 are silica-oversaturated (under equilibrium conditions, of course), and rocks which comprise Ol and /or foids are silica-undersaturated, regardless of whether Ne or Ol is present. The borderline case, the critical saturation, is that which corresponds to the Opx-Pl-Cpx plane of Yoder and Tilley's tetrahedron.

To quantify the degree of silica undersaturation, we have to calculate the amount of pure SiO2 to add to reduce the system to the Opx-Pl-Cpx plane. Taking into account that one Ne molecule requires 2 SiO2 to produce albite, and Ol takes only one SiO2 to produce Opx, their ability to 'absorb' SiO2 is 2:1. That is, 25% mol. Ol consume 25% mol. SiO2, and 5% Ne consume 10% SiO2. Weight proportions are somewhat different to molar ones, but not critically.

Please correct me if I am wrong.

Petr

Hi Pietro and Petr, thanks for the comments.

In any case I repeat my question: is an olivine-orthopyroxene-dominated assemblage to be considered as a silica-undersaturated rock? I continue to believe no.

A rock with modal or CIPW-normative ol-opx cannot be considered silica-undersaturated but is critically silica-saturated. At the same time, the only presence of opx (without quartz) is not sufficient to define a rock as silica-oversaturated. I also note that CIPW composition only with opx and without quartz or olivine are impossible.

I am not worrying about the equilibrium of an olivine-orthopyroxene rock at high or low pressures or the composition of a melt derived from this rock. I am dealing with a typical anhydrous mantle peridotite (virtually Na-K-free). Typical lherzolites (but also harzburgites and ol-websterites should be defined critically silica-saturated rocks, not silica-undersaturated rocks.

michele

Hi Michele,

The SSI simply provides a method to calculate the excess/defect of silica with respect to the CIPW saturated mineralogy independently of what causes the silica undersaturation. The reason for devising such a parameter was to compare the silica activity of rocks with 10 wt.% K2O and 4 wt.% Na2O (nepheline syenites) with respect to others with 17 wt.% K2O and 0.1 wt.% Na2O (synnyrites). The former have CIPW Ort + Ab + Ne. The later Ort + Ne + Le. The reason for the disappearance of albite in the later is the very low Na2O contents. Your reasoning is OK if the Na/K ratio does not change very much, but not for rocks with very low Na2O contents

Cheers

F

See the attached graphs with some experimental results at different P and projections of a collection of different primary melts issuing from both a peridotitic and a piroxenitic mantle...

Hi Pietro,

I cannot download the plots to inspectionate them carefully. could you attach them as downloadable files?

For Fernando:

According to your SSI, a rock with olivine and hypersthene, (and, obviously, no quartz nor nepheline) is silica-undersaturated. The Earths' mantle would be constituted by silica-undersaturated mineral assemblage. I am sorry, but I disagree with this definition. The Earth's mantle is made up of critically silica-saturated rocks, able to produce, after partial melting silica-oversaturated to -undersaturated melts, depending on the degree of melting and depth of melt extraction.

Before dealing with relatively are or very rare lithologies (e.g., synnyrites) I would know the definition of silica-saturation for the by far much more common rock types on Earth (peridotites).michele

Magma chemistry reflects itself in mineralogy. When ever the magma has enough silica to form silica rich mienrals (Qz, Opx, feldspars) then the magma is saturated, otherwise it will form silica poor minerals (Ol, feldsathoids) and it is underaturated. Why do we need to complicate this issue?

Hi Sobhi,

this is not an unuseful consideration at all. Peridotites are by far the most common rock type on Earth. If we do not know how to classify them in terms of silica saturation it is a sign of our poor knowledge of the matter.

What happens if a rock has, at the same moment opx and ol (as in peridotites and websterites)? You say that if a rock has opx is silica saturated, while if a rock has olivine it is silica undersaturated....

Thanks Mike,  If we consider the basalt tetrahedra, the Ol-Opx -Plag plane lies within the tholleiite which is is considered a saturated type of magma. So Ol-Opx assemblage would be considered saturated, while olivine alone or Ol+ feldspathoid are indicators for undersaturation

Ok, this is what I believe. But there are many other persons that believe that the presence of modal or normative olivine in a rock means silica-undersaturated conditions. The presence of olivine cannot be considered as an evidence of silica poornes if it is associated with orthopyroxene (and, obviously, no foids nor quartz).

Forsterite -Opx-Q or foide-feldspar-Q can show saturation of Si in the melts

Hi Michele,

I think you're completely right. I think those rocks with ol + opx should be named as Si-saturated (critically). 

Yalcin