Sciencemadness Discussion Board

Analysis of rocks\minerals without HF

bob800 - 20-11-2012 at 16:54

I always thought it would be fun to do a qualitative analysis of various rock\minerals, with only classic "wet" chemical methods. However, it seems that acids like HCl or HNO<sub>3</sub> would leave silicate-based rocks unscathed... Are there any substances (other than HF or fluorides) that could be used to dissolve them?

vmelkon - 20-11-2012 at 17:55

Good question. I asked once why companies don't extract minerals from silicates. They tend to go for oxides, carbonates, hydroxides, chlorides.

Perhaps heating with carbon at high temperature would so something? Arc furnaces are used to reduce SiO2 to Si.

bob800 - 20-11-2012 at 18:38

Quote: Originally posted by vmelkon  
Perhaps heating with carbon at high temperature would so something? Arc furnaces are used to reduce SiO2 to Si.

Interesting idea... while I don't have access to an arc furnace, here's another use of high temperature fusion from Manual of Chemical Analysis of Rocks (available here).
Quote:

In a portionof 1 gram, silica, alumina,total iron as ferric oxide, manganese, lime, strontia, magnesia and titanium dioxide are determined. The powder is fused with five times its- weight of mixed sodium and potassium carbonates, the melt dissolved in hydrochloric acid and evaporated todryness,thus renderingthe silica insoluble. The silica is filtered off and in the filtrate alumina, ferric oxide,titanium dioxide and phosphoric anhydride are precipitated, first by sodium acetate if manganous oxide is to be determined,or by ammonia water if this is to be neglected. After filtration the precipitate is dissolved in nitric acid and reprecipitated by ammonia,and this repeated if there is much magnesia present. The precipitate is ignitecl and weighed,and then brought into solution by fusion with acid potassium sulphate. This is dissolved in water,the ferric iron reduced by hydrogen sulphide,the excess of this boiled off,and the total iron determined by titration with potassium permananganate. Titanium dioxide is determined in the same liquid by the colorimetric method, which consists in comparing the intensity of color of a known volume of the titrated fluid after oxidation by hydrogen peroxide,with that of a standard solution of titanium colored in the same way.


Would the fusion with carbonate require arc-furnace temperatures as well?

bahamuth - 20-11-2012 at 21:13

Really hot phosphoric acid eats through quartz/glass as butter, perhaps the silicates will suffer the same fate...

watson.fawkes - 21-11-2012 at 06:59

Quote: Originally posted by vmelkon  
Good question. I asked once why companies don't extract minerals from silicates. They tend to go for oxides, carbonates, hydroxides, chlorides.

Perhaps heating with carbon at high temperature would so something? Arc furnaces are used to reduce SiO2 to Si.
The resulting silicon is quite soluble in many molten metals. Ferrosilicon, for example, comes in any number of concentrations. I'm not sure if iron and silicon are fully miscible (there are, albeit, silicides formed), but if not it's pretty close.

bfesser - 21-11-2012 at 13:13

Quote: Originally posted by bob800  
I always thought it would be fun to do a qualitative analysis of various rock\minerals, with only classic "wet" chemical methods.
Likewise, I have thought about doing this for some time. In addition to the common analytes (Na, K, Ca, Mg, etc.), I'd like to quantitatively determine the Fe/Mn content of a certain group of carbonate/oxide-bearing rocks.
Quote: Originally posted by bob800  

Are there any substances (other than HF or fluorides) that could be used to dissolve them?
Perhaps concentrated or molten NaOH or KOH in a plastic beaker or Ni crucible?

If you think this would be fun, I could send you some samples taken from specimens in my rock/mineral collection that I've been trying to positively identify. Perhaps when I get around to doing the analyses myself, we could compare data.

Some samples I plan to work with:
<ul type="circle"><li><a href="http://en.wikipedia.org/wiki/Bentonite" target="_blank">bentonite</a> <img src="../scipics/_wiki.png" /> from Platteville Fm. (weathered ash from Appalachian volcano?)</li>
<li>Mt. St. Helens ash from <a href="http://en.wikipedia.org/wiki/1980_eruption_of_Mount_St._Helens" target="_blank">May 18, 1980 eruption</a> <img src="../scipics/_wiki.png" /></li>
<li>Fe-bearing ore (carbonate?)</li>
<li><a href="http://en.wikipedia.org/wiki/Niobrara_Formation" target="_blank">Niobrara chalk</a> <img src="../scipics/_wiki.png" /></li>
<li>various limestone (CaCO<sub>3</sub>;) dolostone (MgCO<sub>3</sub>;) &mdash; to determine Ca/Mg ratio</li></ul>

[Edited on 7/9/13 by bfesser]

chemrox - 24-11-2012 at 18:20

You don't want to use HF? I don't blame you. Total digestions are usually done with HCl/HNO3 but the minerals are generally sulfides. Query the usgs library. There are many analytical wet-chem procedures that have been developed and written up over the years. Maybe peroxy acids can get your silicates into solution. I'm speculating and I dislike getting answers that are speculation. On the other hand your answers are there if you spend the library time. Good luck.

bob800 - 25-11-2012 at 09:17

Quote: Originally posted by bfesser  
If you think this would be fun, I could send you some samples taken from specimens in my rock/mineral collection that I've been trying to positively identify. Perhaps when I get around to doing the analyses myself, we could compare data.

U2U sent.

UnintentionalChaos - 25-11-2012 at 09:45

Quote: Originally posted by bfesser  

<li>Mt. St. Helens ash from <a href="http://en.wikipedia.org/wiki/1980_eruption_of_Mount_St._Helens" target="_blank">May 18, 1980 eruption</a> <img src="../scipics/_wiki.png" /></li>


Lithium metaborate fusion and flame AA. :cool: At least, that's what we did in instrumentation lab.

See also "Mount St. Helens Ash from the 18 May 1980 Eruption: Chemical, Physical, Mineralogical, and Biological Properties. Science. 1980, 4461, 1116-1125"

I used to have (maybe still do but I can't find it) a copy of this paper.

<!-- bfesser_edit_tag -->[<a href="u2u.php?action=send&username=bfesser">bfesser</a>: fixed broken image(s)]

[Edited on 7/9/13 by bfesser]

bob800 - 25-11-2012 at 10:07

Thanks to all who replied! I did some googling and it appears that much can be done with a block of charcoal, a blowpipe, and some borax and carbonate. See: http://archive.org/details/manualdetermina01penfgoog

feacetech - 25-11-2012 at 13:11

I use mixed HCl/HNO3 or HClO4 on fine ground phosphate rocks and serpintine rocks to digest for quantative analysis.

Poppy - 25-11-2012 at 14:55

Even lemon juice can dissolve some rocks, though some quartz looking parts are apparently left behind. Thus you could theorically extract the "useful" part of of it, not sure if the silicon bearing compounds are chemically bound to the other minerals

daragh8008 - 5-12-2012 at 16:34

Molten NaOH is great for silicon as mentioned above. It's the only thing I've seen eat a 1cmX1cm silicon wafer in 15 seconds. Don't let the 1ft jet flame put you off! Although silicon oxides are a lot less energetic.

bfesser - 20-12-2012 at 11:38

While browsing the <a href="http://www.usgs.gov/" target="_blank">USGS</a> website, I found two geochemistry PDFs. They may not be particularly helpful for the chemistry, but are worth mentioning.

<strong><a href="http://pubs.usgs.gov/of/2011/1193/OF11-1193.pdf" target="_blank">Inorganic Chemical Analysis of Environmental Materials—A Lecture Series</a></strong>
<strong><a href="http://pubs.usgs.gov/of/2011/1187/pdf/ofr2011-1187.pdf" target="blank">Quality Assurance and Quality Control of Geochemical Data: A Primer for the Research Scientist</a></strong>