It all depends on the geology of the area of study. But generally you may consider Al, Fe and Si. However, we should know about the mineralogy of the area of study

Dear Mr Mahar,

show me please your heavy mineral suite you are dealing with and the goal of your study and I can tell you what elements warrant being analyzed by EMPA. In the present state it would only end up in a listing of elements and still remain a matter of conjecture.

Best regards

Harald G. Dill

As Karbassi says it depends on the area, but do not limit yourself to Al, Fe and Si. All the 10 major elements are present. I notice that you are particularly interested in the trace element component, so these will not help you very much. We have started looking at density separated beach sands using EDX, we don't have EPMA, and the range of minerals, and therefore elements, is quite large. We are using automated image analysis of the entire slide to locate the grains, each of which is then automatically analysed for chemistry using whole particle scanning. Admittedly these are very fast scans only suitable for a qualitative identification of which mineral phases are present and not suitable for full quant. But now we have a map of where the various minerals are it is easy to relocate examples and do full quants on selected grains.

As for trace elements, our rapid scans of a single beach sand showed  Zr, Ba, Yb, Hf, Sr. For a fuller elemental suite I would also consider looking for Cr (chromite), Sn (cassiterite), Y (xenotime), Ce, La, Th (monazite).  

There are plenty of papers out there that go into this in more detail.

Dear Mr. Mahar,

please find attached some papers on transparent and opaque heavy minerals from the alluvial fan through the beach placers which you can analyze using SEM-EDX or EMPA. If you have special target areas with PGE, Au or REE heavy minerals let me know this. The reprints are a only a selection showing you what you might expect in stream sediments, placers or, in general, in clastic sediments.

Best regards

Harald G. Dill

Doing trace elements by EPMA takes a lot of counting time. Although you may ultimately be most interested in the provenance of the economical minerals like zircon and rutile, these do not show too much variation in the oxides that you can most easily measure by EPMA. You may consider whether accompanying minerals may not yield more information (garnet, spinel, ilmenite).

If you have the appropriate EPMA, you could try and do chemical U-Th-Pb dating of the monazites.

As said by those above, the trace element geochemistry of those heavy minerals will depend on the source area and reflect the composition and evolution of that source area. Beach processes will sort and concentrate the heavy minerals, but do not influence significantly in their trace element geochemistry. That is, the crucial part to understand the provenance and source region history of the sand deposits is not the environment (beach) but the composition of the sand: classic sandstone petrology with provenance analysis will provide you the basic information. Focusing just on heavy minerals will make you loose easily accesible information recorded by sandstone composition.

Zircons are rich in Zr and always Hf.  The thorium content in zircons is a bit variable but usually several thousand ppm.  Uranium follows the crustal relationship with thorium, so is close to Th*0.25.  Monazite is a rare earth phosphate that is rich in Th and U.  If you have an ECS or LithoScanner geochemical logging data, you might use Gd to spot it.  Rutile (and anatase) are in our experience pretty much devoid of elemental substitutions and are closely approximated by TiO2.

Dear Mr. Mahar,

a method not available in every department but speeding up  the workflow during analysis of heavy minerals in particulate sections is MLA (mineral liberation analysis). Maybe you have access to an instituation conducting this method.

Regards

H.G.Dill

My best bet is the knowledge of the regional geology. All three minerals are common occurrences in granites and are found concentrated, associated  in beach sands. What other minerals come associated to them? Ilmenites?

Regards, Z.Silva

Thanks everybody for your suggestions.

Dear Harald, I am working on three type of sands, beach sands along coastal areas of Karachi, dune sands in the southeastern desert and alluvial deposits along the Indus River.

Beach sands are mainly composed of quartz, zircons, magnetite, rutile with some ilmenite and garnets. However, I am planning to perform detailed mineralogical analyses. I have only access to EDS. the dunal sand contains 5-10% heavies however, the fraction of particular minerals is not known yet. the objective is bifold 1) to evaluate the REE potential of these sand deposits and 2) provenance and source region history of these deposits. alluvial deposits from the present day flood plans along the Indus River  are collected to compare with the ancient beach deposit at the coast. I am particularly interested to use  geochemistry of these accessory phases, for example,  some authors have used rutile (Cr, Ta, Nb etc) for provenance and source region interpretations. Your suggestions Please

analysis of major ,minor trace and REE in ilmenite, Zircon , garnet, monazite would provide valuable information on provenaceand grade of beacdeposits for effective mining. We are looking for collborative workrs in undertaking the studies in indian coast.