Please see the following paper

Article REE minerals at the Songwe Hill carbonatite, Malawi: HREE-en...

Best Regards Rachid Benaouda

thank you Osama for the paper suggestion

Yttrium anomalies are caused by either equilibrium with F-bearing fluids, or processes at low temperatures. This starts at around 700 °C, and the magnitude of the Y anomaly gets very strong at 400 °C and lower.

The magnitude and direction depends on what exactly you're analysing (whole rock, specific minerals, etc).

In general. late stage hydrothermal alteration by F-rich fluids is extremely common in carbonatites. I would expect Y anomalies to be the rule, not the exception.

Thank you Michael for your answer. The remouval of Y by F-bearing fluids is also proposed in previous papers e.g. Bühn 2008 or Brom-Fendley et al. 2017 as suggested by Ousama, but I dont have any experimental data for that. You poited out, Y anomaly occurs at 700 °C and gets stronger at 400 °C. I would be very grateful, if you could provide experimental studies that show these values?. In addition, I didn't find any evidence from the field samples that may show the fractionation of Y by Y-bearing mineral phases, like fluorite or xenotime.

Thans again

Hi all,

Indeed, as Michael mentioned, the most likely cause of an Y anomaly (relative to Ho) is the effect of F-bearing fluids. I suggest this as the cause of the Y/Ho anomalies I observed at Songwe Hill, and this is supported by textural and (more circumstantial) fluid inclusion evidence. However, I wouldn't say that the work is definitive, and there could well be other explanations.

With regards to experimental work, I typically rely on the following paper: LOGES, A., MIGDISOV, A. A., WAGNER, T., WILLIAMS-JONES, A. E., AND MARKL, G. 2013. An experimental study of the aqueous solubility and speciation of Y (III) fluoride at temperatures up to 250 C. Geochimica et Cosmochimica Acta 123:403–415.

Cheers

Sam

Hi Sam,

thank you for your reply. I have recently read the interessting experimental work of Loges et al. 2013 in addition to your precious papers. Michael pointed out the influence of temperatures between 400 °C and 700°C on Y anomaly, therefor I am also interessted to this work.

All the best

Rachid

So there isn't much experimental work on the topic beyond the Loges et al 2013 paper.

The evidence for that is somewhat indirect. The decoupling between Y and Ho (and the other REE) is not the cause, it's the result of differing behaviour of Y in general at low temperatures. This is manifested by the anomalous Y behaviour in fluids, and it can be also seen in mineral-mineral partitioning (which is independent of fluid composition).

At high temperature mineral solubility and partitioning experiments (in all kinds of melts and minerals), let's say 900C and higher, you don't see the Y-Ho decoupling. You do see it in low temperature experimental work (eg Loges). There aren't any studies that I know of exploring that temperature range in between.

However, working on natural samples I can see that in carbonatitic mineral assemblages that likely formed at ~800C you can just barely see the Y anomaly, and it becomes more prominent once you look at assemblages that formed at lower T.

Another problem, particularly for cases where the anomaly is weak, is how you define it? Some people use Y/Ho (assuming the non-anomalous chondritic value is around 26 to 28), others place Y between Ho and either Er or Dy and calculate the anomaly using the root of the geometric mean of the neighbours (just like you do Eu anomalies). This also depends on the slope of the REE pattern, because different slopes will lead to different values for the anomaly, when there isn't a difference. This also assumes that there was no anomaly to begin with, which may not be always the case.

In the meanwhile, until somewhat properly calibrates it, I think it's best to use Y anomalies cautiously as a relative measure within a single locality and take comparisons with other localities with a grain of salt.

many thanks Michael for your very informative response. It seems very plausible. Owing to that, the occurrence of negative Y anomaly in some carbonatite is likely not a primary signature, but may be the result of post magmatic processes and fluid reworking, as carbonatites are very succeptible to fluid alteration.

1. What do you mean by "Y anomaly in carbonatite"? Is that in the minerals? In the whole rock? What does the whole rock represent (probably not the liquid)?

2. What do you mean by "primary"? Igneous? There was just a paper out showing that some igneous Ce anomalies are inherited from Ce-anomalous subducted sediments. If you can inherit Ce anomalies into "primary" magmas, why couldn't you do this with carbonatites? It all depends on how much of the Y/REE signature in the carbonatitic magma is derived from the subducted sediment, or the (presumably non-anomalous) mantle that it interacts with on the way up.

Y anomalies have to be interpreted in context within a locality, and making blanket statements based on Y anomalies or comparing to other localities should be done with care. For example, I'm working on a locality where the igneous minerals show chondrite-like Y/Ho values, and later hydrothermal-looking minerals also show increasingly anomalous Y/Ho values. In my case, it works and makes sense. But this becomes clear only when taken together with other lines of evidence. I would caution against using Y anomalies alone as an indicator for anything.

Hi Michael,

1. I ment, negative Y anomaly occurs in all whole rock samples of the carbonatite I am working on.

2. Yes, I mean by primary carbonatite, the carbonatite that is formed directly from a mantle derived carbonatitic magma (igneous).

In your previous answer, you suggested that Y-Ho decoupling is unlikely to occur in high temperatures, but it does in low temperatures (Loges et al. 2013). Therefore, I though, the occurrence of unusual negative Y anomaly in whole rock carbonatite might be the result of post magmatic processes, that may affect primary carbonatite, Unless if the carbonatitic magma was derived from a Y depleted source in the mantle, which is improbable.

I agree with you that negative Y anomaly alone should not be taken as evidence for that, but there is many evidences that show that this carbonatite have experienced intensive hydrothermal influence.

Hi guys,

I am kind of skeptical about de-coupling of Y and REE by F-bearing fluids. At least at T

Hi Artas,

thank you very much for your advice, which will be seriously considered. Indeed, as you showed in your previous and precious papers (e.g. 2009, 2012, 2013, 2014...) the very low solubility of REE -fluoride at T°

Hi All, Decoupling of Y and REE could be related to any phase separation, and does not necessarily have to be F-related (though in some systems that might help). Y is a lighter than REE of a similar ionic radius, therefore always plots off Onuma parabolae defined by REE in mineral-melt element-partitioning diagrams. 

Olin and Wolff (2010, CTMP 160) give a theoretical framework for this mass-fractionation effect, with analogy to stable isotope systems. Here, 'mass fractionation between two otherwise identical ions may be crudely approximated by treating the metal-oxide bonds as a diatomic molecule that behaves as a simple harmonic oscillator... the heavy molecule has a lower vibrational frequency than the lighter molecule and hence is more energetically stable'.

Heavier atoms (REE) therefore favour the phase with stronger M-O bonds. Additionally, much like stable isotope systems, mass fractionation should be more pronounced at lower temperatures, consistent with Michael Anenburg's observation. 

Thanks Charlie for your contribution. The mass-controlled fractionation seems not to have a significant impact on decoupling twins during magma fractionation or hydrothermal moblisation, as it is not usual to see in most magmatic rocks, although it is already proposed to explain some occuring fractionation of other twins like Zr-Hf and Nb-Ta (Hui et al. 2011 and Nui et al. 2012). I will surely read the paper of Olin and Wolff you suggested to get more info about this topic.

Thank you again