CR Scientific

 Newsletter Archive - Favorite Articles

You may notice the end-notes don't necessarily start at "1".  The following article is an excerpt from a back issue of our newsletter.

Qualitative Analysis of a Jade-Like Rock
from Northwestern Washington

by C. Thorsten

Recently the author obtained a green, mottled rock which had come from a beach in the state of Washington.  The collector who had originally found it stated that in many years of walking up and down that beach, he'd found no other rocks like it.  Having very little familiarity with northwestern Washington and being therefore unable to recognize its mineral assemblages simply by looking at them, the author decided to resort to some fairly simple qualitative tests.
Figure 1 shows a slab cut from the rock.

A slice of the unidentified beach rock

Figure 1.  A slab cut from the beach rock.
Note the greenish, almost teal coloration.

1.  Visual and Mechanical Properties:
Simple scratch tests, concentrating on the greenish mineral grains, suggested a hardness of 6 to 7;  this measure was difficult to obtain precisely, even when studying the surface of the specimen with a magnifier.  There was some doubt whether a hardness sample really did make a scratch, or whether it was simply a polished streak that looked like a scratch.  The stone was able to scratch orthoclase (H=6) definitely, but only after repeated attempts.  Orthoclase, on the other hand, was not able to scratch the unidentified stone at all.  Finally, the stone was completely unable to scratch quartz (H=7).  Therefore, it seeemed reasonable to assign H=6.5 to the sample.  
The unidentified stone had a high overall toughness (resistance to crushing), as evidenced by its behavior under the hammer.  
Minerals make up any given rock, but a rock is not always made of a single mineral;  sometimes, different minerals are in grains too small to tell apart.  In the case of this unidentified beach rock, the zones of green and of white seemed to be of a different composition from the darker (brown) regions, which appeared to be alteration products of the green / white mineral or minerals.  These dark, mostly brown regions seemed to exist along grain interstices and old micro-fractures where water could have exerted its effect through the ages.
The brown mineral was examined at 10x and 30x.  It had a greasy or waxy luster; the overall appearance was reminiscent of something from the serpentine or chlorite groups.  A sharp needle was used under the microscope in an attempt to scratch some of this material, which proved to be very soft: estimated hardness perhaps 2, no more than 3.  It smeared in very much the same manner as talc or serpentinized alteration minerals that one might find in certain marble or skarn deposits.
The green mineral was harder, around 6.5 as stated before.  Inspection at 10x and 30x suggested it was not at all amorphous;  rather, it seemed to be a fairly typical, massive-form silicate made of interlocking crystal grains.  Though the material didn't exhibit cleavage, massive-form minerals often behave this way.  Cleavage, therefore, is not an especially useful property unless one is dealing with definite crystals. 
The green mineral did exhibit vaguely splintery fracture habit which could be seen clearly with and even without magnification.  This habit was entirely different from anything the author had ever observed in chalcedony (i.e., chert, jasper, or agate).  Figure 2 doesn't adequately capture the play of light that gave away the individual crystal grains, nor does it show off the splintery fracture, but these were visible in real life.
photo taken with Mini-VID USB camera
Figure 2. The green mineral at 10x magnification

2. Fusibility:  A chip consisting mostly of the greenish material fused only along thin edges in the propane torch flame (somewhere between 1600 and 1900C at the hottest portion). These edges melted fairly quickly, but continued heating did not melt the thicker portions at all. The flame assumed a bright sodium-yellow with a fine but persistent border of red to red-orange (Ca or possibly Sr). The mineral's green color disappeared, leaving an ivory to pale yellow color.

3. Closed Tube Test:  A short piece of 5 mm borosilicate glass tubing, sealed off at one end by heating, served as the closed tube. The powdered mineral sample did give off some water upon prolonged heating. No other sublimates were observed. 

4. Solubility:  A fragment placed in concentrated HCl did not seem to dissolve at all, but the solution turned from colorless to strong yellow.
Aqueous ammonia was added to this yellow solution, and a considerable amount of rust-brown precipitate quickly formed. The yellow color disappeared. When the mineral chip was taken out of the liquid and washed, it was evident that the brown, serpentinaceous mineral had dissolved in the HCl but the green mineral had been mostly unaffected.

5. Bead Tests:  Perhaps to the relief of some readers, but to the disappointment of others who were hoping the CR Scientific Newsletter's apparent monomania for element 25 might never end, the bead tests did not suggest the presence of manganese in the sample. Furthermore, there were noted differences this time between the "hot" and "cold" states of the beads.

Following are the observed results:

Borax - R.F.- Yellow (hot); Nearly Colorless with a faint aqua tinge (cold)
Borax - O.F. - Bright Yellow (hot); Colorless (cold)

Salt of Phosphorus - R.F. - Colorless (hot); Colorless (cold)
Salt of Phosphorus - O.F. - Bright Yellow (hot); Colorless (cold)

Sodium Carbonate - R.F. - Muddy Yellow-Brown (hot); Pale Green (cold)
Sodium Carbonate - O.F. - Muddy Yellow Brown (hot); Dirty Yellow (cold)

6. Chromate Flux:  Powdered sample was fused on the carbon block with chromate flux (1 part KHSO4, 1 part K2CrO4, 2 parts sulfur; as per Smith, 1953). The coating near the assay was pale white;  same color, hot or cold.

7. Iodide Flux: (1 part KHSO4, 1 part KI, 2 parts sulfur; as per Smith, 1953) The coating near the assay was dense white, fading to pale white at some distance from the center. This outer coating was volatile in the reducing flame, but the inner coating was very stable. There were some tiny but definite yellow spots also noted. The coating looked the same, hot or cold.

8. Aqueous Chemical Tests
A.  To prepare the sample for dissolving, it was mixed with an equal volume of sodium carbonate and fused on the carbon block. The fused mass was allowed to cool for about 20-30 minutes and then crushed into powder.
B.  The resulting powder was placed in a micro beaker and covered with dilute HCl.  Profuse bubbling occurred.  A flocculent and highly insoluble residue now floated in the liquid, eventually settling to the bottom. The solution remained colorless, and there were still undissolved lumps at the bottom.
C.  In order to dissolve the remaining sample, concentrated (37%) HCl was carefully added to the micro beaker. The bubbling became severe, and the solution quickly turned bright yellow with the slightest influence of green. The color remained this way.
D.   The flocculent material was allowed to settle. The yellow solution was siphoned off carefully with a dropper for further tests.
E.  A drop of this solution was added to a spot plate containing potassium iodide. This caused it to turn a deep and pure shade of yellow (as opposed to having a hint of greenish, which the HCl solution had before this). Adding strong NaOH to this destroyed the yellow color completely.
F.  About 1/2 mL of the same solution (from 8D) was placed in a small test tube. The addition of strong sodium hydroxide caused a cloudy, white precipitate to form. Excess NaOH failed to make this precipitate redissolve, as far as could be seen. A separate aliquot of solution was tested with aqueous ammonia instead of NaOH; this also caused a white precipitate which turned light brown on standing. Excess ammonia did not dissolve the precipitate.
G.  Another aliquot of solution from 8D was placed in a clean test tube and brought up, using NaOH, to a pH just below neutral. The addition of sodium ammonium phosphate (salt of phosphorus) caused a cloudy white precipitate, presumably MgNH4PO4 • 6H2O.
H.  A drop of the solution from 8D was added to a spot plate well containing potassium chromate. The solution immediately turned bright orange and cloudy; more HCl made the cloudiness disappear, but the orange color persisted. Bringing the pH back up to around 3-4 (by adding NaOH) made the orange compound precipitate out again.
I.  About 1/2 mL of the solution from 8D was placed in another small test tube. Adding ammonium sulfide solution caused a cloudy to almost gelatinous precipitate that was dark gray to black. It is not certain whether this was one compound or two; around the edges the material was white, suggesting maybe a hydroxide mixed with a sulfide ((NH4)2S can precipitate both if the right ions are present). Boiling the suspended precipitate down in a crucible caused it to turn rust-brown in the final stages of heating. When it was cool, this residue was taken up in a few drops of dilute HCl.  Addition of sodium hexacyanoferrate (II) caused the color to turn deep green to blue-green (this was not the typical deep blue of Prussian Blue; it's possible the color was caused by the the similar iron hexacyanoferrate compound known as Prussian Green / Berlin Green, which could form under the right conditions).
J.  Another 1/2 mL of the solution from 8D was placed in a clean test tube and brought to just below pH 7 with NaOH (the point at which the hydroxide precipitate redissolved with difficulty). Saturated aqueous oxalic acid was then added and the solution boiled for several minutes. No precipitate formed at first, but the next day there was a white, crystalline precipitate at the bottom.  This was washed several times in distilled water, placed in a crucible and covered with conc. HNO3, and evaporated to dryness (fume hood!!) to destroy the oxalate.  The residue was then taken up in HCl and made alkaline with aqueous ammonia. There was no precipitate noted, suggesting a lack of Th, Sc, and the rare-earth elements but not ruling out Ca, Ba, or Sr (as per Chapter VII, Procedure 4 in Smith, 1953)
K.  Another 1/2 mL of the solution from 8D was tested with metallic zinc. No color change was observed, even after boiling. This suggests the absence of Ti.
L.  Just to be sure of no Mn, Co, or Ni, an aliquot of solution from 8D was made alkaline with ammonia. In the presence of NH4Cl (as would be the case when neutralizing HCl with ammonia solution), these three will not come out of solution until boiled with H2O2 (Smith, 1953).  This treatment caused no observable changes in the test solution.

Conclusions and Discussion
The rock's physical properties, taken together with the chemical test results, suggest a variety of jade;  the elements established with reasonable certainty are Si, Fe, Na, Ca, and Mg. The following is a list of the steps which gave the best evidence:
Silicon......step 8B
Iron......step 8I and step 5, especially the sodium carbonate bead. There might have been interfering ions which prevented the green color from coming out in the R.F. beads of borax and salt of phosphorus-- possibly by hindering the reduction of Fe+3 to Fe+2.
Sodium....... the yellow flame test
Calcium......step 8J and the red-orange flame test
Magnesium......steps 8F and 8G
Another element suggested in the tests was Lead, Pb (steps 7, 8E, 8H). The fact that NaOH caused the yellow precipitate in the KI test to disappear is actually not surprising:  first of all, PbI2 is soluble in alkali (check any suitable reference, such as the CRC Handbook of Chemistry and Physics);  second, lead forms an amphoteric hydroxide, meaning that it will dissolve in excess NaOH to form a colorless, complex ion.  Sorum (1953) indicates this is an anion with the formula Pb(OH)4-2. As for the formation of an orange precipitate rather than a yellow one in the chromate test (8H), recall that lead chromate can be either yellow or orange depending on what conditions formed it (example: crocoite).  The colors observed in the borax and salt of phosphorus bead tests (step 5) don't at all contradict the possibility of lead in the sample, either.
What additional tests might have been useful?  There are quite a few: specific gravity, optics, and quite a few more chemical tests; the possibilities are limited only by what equipment the experimenter has available.  Note that we didn't even test the flocculent residue from step 2 for W, Ta, or Nb, for example; nor did we attempt to precipitate the suspected Pb+2 ions using cold HCl, as one would do in a typical qualitative analysis scheme 6. The presence of Mo, V, and a few other metals could have been sought as well. Once again, the author performed enough tests to give at least a plausible idea of the subject's identity: a variety of jade, probably an amphibole and most likely nephrite 7, a mineral whose green color is caused by ferrous iron, and which may contain traces of lead and other elements that were present either as part of the primary mineral or as constituents of accessory minerals present in grains too small to pick out (but which found their way into the crushed sample and therefore the chemical tests).
Hopefully this has been an informative journey for our readers... it was certainly enlightening to this writer, who originally thought the rock was green jasper!


6 Instead, any Pb present would've remained in solution, thanks to the strong HCl added to the pulverized fusion (which also released heat).  To precipitate lead as the chloride, it requires cold, dilute HCl.  Back to article

7 While nephrite (actinolite) doesn't contain sodium in its formula, we've already explored the seawater possibility. Alternatively, our mineral could be one of the many, many amphiboles or pyroxenes similar in composition to actinolite but off by only a sodium atom here and there... the number of different silicate minerals that can possibly contain Fe, Ca, Mg, and Na is impressive.   Back to article

Works Cited:

Brush, George, and Penfield, Samuel. Determinative Mineralogy and Blowpipe Analysis, 16th ed. New York: John Wiley & Sons, 1926.

Smith, Orsino C. Identification and Qualitative Analysis of Minerals, 2nd ed. Princeton, New Jersey: D. Van Nostrand Co., 1953.

Sorum, C.H. Introduction to Semimicro Qualitative Analysis, 2nd ed. New York: Prentice-Hall, 1953.

While the information in this newsletter is thought to be accurate to the best of the authors' current knowledge, it is not guaranteed to be free of errors or to be suitable for any particular use. The procedures and experiments outlined within can be dangerous or even fatal if carried out improperly. If you choose to attempt any of them, you proceed entirely at your own risk.

This newsletter is copyright of CR Scientific LLC.  You may print out or make photocopies of this newsletter for educational and personal use  (Click here for Copyright and Fair Use details .)  Please email us if you find any errors or omissions or would simply like to make a suggestion.

General Terms of Use for this website

CR Scientific Catalog
Newsletter Index Page
Lab Glassware
Main Page