Identification of Lander Blue and Distinguishing from Look Alikes, Using Geochemistry
Neil Ray, West Texas Analytical Laboratory, Geochemist & Mineralogist
Lander Blue is indeed the most coveted and rarest turquoise available on the market, with some material approaching over $400/ct. Lander Blue does have a distinct appearance about it, tight deep blue turquoise set against a golden webbed matrix with black chert. Unfortunately, there are some mines with high grade turquoise that mimics the appearance of Lander Blue. In fact, gem grade Hubei that was collected in the 1980’s almost looks identical to Lander Blue. When the Hubei turquoise hit the market, there were unscrupulous dealers passing it off as authentic Lander Blue. Much of this material became set in jewelry and sold to the unwary public, with its true origins seemingly lost in time. Such is the case with most vintage jewelry, as the origin of the turquoise remains ultimately unknown. I have seen many so-called Lander Blue pieces offered for sale online, and sometimes the seller is clueless to the authenticity and sometimes they are acting fraudulently. Most times they believe their piece is authentic because they purchased it as Lander Blue. It is best to purchase Lander Blue with a trail of known provenance and a guarantee on authenticity, however a lot of times that option is unavailable. People turn to social media for an opinion and self-gratification on their so-called Lander piece. Unfortunately, the identification of geological materials is very difficult to do from a photograph alone, especially turquoise with its many color and matrix variations. I have even seen some people claim that their Lander Blue piece is just a lower grade Lander, which is not correct by any means. I don’t believe any lower grade Lander Blue exists, considering the rarity of only 100 lbs that has ever been collected. All Lander Blue is seemingly high grade to gem grade, and mid-grade Lander if it did exist would certainly be rarer than the high grade. As such, if it lacks the high to gem grade appearance, then more than likely it is not Lander Blue. What if there was a reliable method to chemically verify Lander Blue from any look alike turquoise than simple observation?
Methods and Testing Procedures
When the discussion arises on using chemistry to identify the origin of turquoise it seems rather obvious to simply analyze a known specimen of turquoise and compare it to the analysis of an unknown for similarities. Unfortunately, it is not that simple, and identification can’t be made by simply stating that this specimen has for example the same amount of lead as this unknown specimen, therefore it has to be from this location. Turquoise is either formed by supergene processes or it is a product of hot hydrothermal metal rich fluids, yielding a complex variability even in material from the same mine and location. Some elements are considered mobile (ones that would vary) and some are considered immobile (reflective of the source or host rock). To further complicate matters some hydrothermally altered rocks show a great deal of variation within the same specimen. When performing a chemical analysis, the ideal method would be to grind the specimen to homogeneity prior to analysis, but such is not acceptable to precious materials, especially Lander Blue. The lack of homogeneity is overcome by comparing three elements at a time, rather than just one or two elements against each other. Geochemists use what is called a ternary diagram, which allows for three elements to be compared against each other at once. If one element varies then the rest will vary accordingly and the elemental ratios are the same, allowing for a distinct field or region for the specimen in question to be plotted in. Considering the many mines that exist and the complexity, several ternary diagrams are used to make the assessment of identification. Now that brings us to the next topic of discussion, what methods and procedures are used to analyze elemental chemistry. As I mentioned it would be ideal to grind the sample prior to analysis, but that is not a possibility. The most accurate and accepted method used for elemental analysis is inductively coupled plasma optical emission spectroscopy (ICP-OES). The specimen is ground and digested in a combination of acids, usually nitric and hydrofluoric and analyzed as a liquid for elemental analysis. Though destructive it offers the lowest detection limits and almost every element on the periodic table is available for analysis. Again, with Lander Blue and precious gemstones such a method is not an option, thus an alternative method known as x-ray fluorescence (XRF) is recommended. XRF is preferred as its non-destructive and will not harm the specimen in any way. The XRF analyzer sends x-rays that cause electrons to temporarily jump orbitals, which emit light at specified wavelengths that can be read by intensity to determine elemental composition. Again, it is completely harmless to the specimen and the most accepted method used for the analysis of precious gemstones and jewelry. XRF analysis has been employed at high end jewelry stores for years as an effective method to verify both gemstones and precious metal content; however, not all XRF analyzers are created equal. When considering an instrument, it is best to use a XRF with a large detector capable of analyzing light elements such as aluminum and silicon. A smaller detector may have the limitations of less sensitivity, only allowing for a detection limit of 100 ppm, rather than 10 ppm or less. Another important consideration is the analyzer must be equipped with a geochem or mining mode, rather than just soil mode. Benchtop XRF analyzers can be user calibrated, however handheld models are factory calibrated with modes that are matrix dependent. The most common analyzer employs an alloy mode that scrap yards use to verify metal content and will not work on geological materials. One final consideration is most handheld models can be mounted in a stationary test stand, which is highly recommended not only for ergonomics but also for repeatability of measurement. XRF does have limitations though, in that it cannot typically detect elements lighter than magnesium, which means no sodium. Since sodium is such an important constituent of rocks, determination of it is critical. Additionally, aside from elemental iron, iron exists in both a 2+ and 3+ state, which is critical to the analysis of green iron bearing turquoise and associated minerals. The XRF will only determine total iron and cannot differentiate between iron 2+ and iron 3+. Software known as P3M was developed after 10 years of research that employs a complex algorithm to calculate sodium, differentiate iron 2+ from iron 3+, and even calculate lithium in any geological sample. The algorithm has been tested with over an 85% accuracy on samples ran by ICP and against published data from 50 peer reviewed journals on multiple rock types. P3M also uses the elemental analysis to calculate the mineralogy of 100 different mineral species, some of which are non-existent to turquoise, however P3M is also used on igneous, sedimentary, and metamorphic rocks. So, with an explanation of the testing method and procedure, as well as the P3M model and algorithm, the next question is what about the turquoise? To accurately identify Lander Blue, it is important to start with verifiable Lander Blue as a baseline reference to compare against unknown Lander Blue. Mike Ryan with Turquoise in America provided specimens of Lander Blue from his extensive Callais collection and to test against potential outliers a colleague of mine purchased two specimens of Lander Blue directly from Bob Brucia with NevadaGem for analysis. Finally, Mike Ryan provided another set of potential Lander Blue specimens. First, let’s examine the known Lander Blue and see how it compares against suspected Lander Blue.
Lander Blue of Verifiable Provenance
For the initial study on Lander Blue chemistry, four specimens of verified provenance from the Callais collection were selected for analysis. The specimens consist of a cabochon set in an 18kt gold pin and four cabochons representing some of the variability in color and matrix that Lander Blue visually displays. Specimen #1 has a slightly lighter blue coloration with less matrix, specimen #2 shows some variability in the size of the webbing, and specimen #3 shows much smaller turquoise (micro-webbing). Specimen #4 is a gem grade example highlighted nicely in an 18 kt mount with coral and diamond accents, this specimen displays the classic dark blue coloration and black matrix that Lander Blue is known for. This specimen was selected to identify the interference that jewelry mounted pieces pose in analysis, such as background gold or silver interference. All four specimens are considered very high grade to gem grade.
Callais collection: (Specimen #4) Lander Blue in 18 kt gold pin.
Two gem grade specimens were acquired directly from Bob Brucia for comparison to the Callais collection. These two specimens are visually similar to the Callais collection, except for the distinct presence of gold webbing in the matrix, two gem grade examples of Lander Blue.
The mineralogy of the four test specimens is outlined below with minerals that are not present omitted from the P3M calculation. The mineralogy is consistent for all four samples, with the micro-web sample #3 showing only a few minor differences. This specimen has lower quartz, no iron oxides and it is absent of secondary sulfide mineralization, except for proustite. Proustite is a rare silver arsenic sulfide, known as ruby silver for its deep red color, however it quickly oxidizes to black and at a low concentration of only 0.45% it would only be discernable under magnification. Proustite forms at low temperatures and is considered a supergene mineral, which will be discussed later. Specimen #4 has the highest concentration of silver, which isn’t surprising considering that it is set in a jewelry mount, however P3M calculations identified the silver content as alloy and didn’t allocate it to mineral species. The two specimens from Bob Brucia show considerably more quartz and no clay minerals. Specimen #5 also contains siderite, an iron carbonate and trace wolframite, which is a manganese iron tungstate.
| Specimen #1 | Specimen #2 | Specimen #3 | Specimen #4 | Specimen #5 | Specimen #6 |
Quartz | 5.67 | 6.21 | 2.50 | 7.26 | 16.68 | 15.09 |
Orthoclase | 1.83 | 0.98 | 3.73 | 1.96 | 1.47 | 1.11 |
Magnesiochromite | 0.05 | 0.02 | ----- | 0.02 | ----- | ----- |
Chromite | ----- | ----- | 0.01 | ----- | 0.01 | 0.03 |
Magnetite | ----- | ----- | 0.01 | ----- | 0.79 | ----- |
Hematite | 0.74 | 0.38 | ----- | 0.23 | 0.81 | 0.56 |
Rutile | 0.95 | 0.01 | ----- | 0.03 | ----- | 0.17 |
Turquoise | 83.56 | 87.22 | 85.12 | 79.06 | 78.01 | 81.52 |
Ankerite | 0.49 | ----- | ----- | 0.93 | 1.03 | 0.56 |
Siderite | ----- | ----- | ----- | ----- | 0.43 | ----- |
Rhodochrosite | 0.04 | 0.01 | ----- | 0.03 | 0.02 | 0.02 |
Azurite/Malachite | 0.27 | 0.31 | 1.16 | 0.57 | 0.12 | 0.28 |
Halite | 1.62 | 0.31 | 1.01 | 0.96 | 0.01 | 0.10 |
Fluorite | 0.02 | 0.01 | 0.02 | 0.02 | 0.01 | 0.01 |
Pyrite | 0.63 | 0.58 | 0.41 | 1.80 | 0.07 | 0.35 |
Illite/Clays | 3.91 | 3.78 | 5.46 | 6.87 | ----- | ----- |
Chalcopyrite | 0.01 | 0.02 | ----- | 0.02 | 0.01 | 0.01 |
Bornite | 0.04 | 0.05 | ----- | 0.03 | 0.03 | 0.05 |
Proustite | ----- | ----- | 0.45 | ----- | ----- | ----- |
Arsenopyrite | 0.11 | 0.04 | ----- | 0.21 | 0.12 | 0.14 |
Wolframite | ----- | ----- | ----- | ----- | 0.01 | ----- |
Scheelite | 0.06 | 0.07 | 0.10 | ----- | 0.01 | ----- |
Limonite/Goethite |
|
|
|
| 0.36 |
|
Pyrolusite | ----- | ----- | 0.02 | ----- | ----- | ----- |
The results of the XRF analysis show a consistent chemistry, with specimens #5 and #6 having considerably more silicon, reflecting a higher quartz content. It is also noted that Lander Blue has higher concentrations of zinc, ranging from 1000 ppm to 5000 ppm in the specimens that was analyzed. Though the zinc content is considerably high, it is not high enough to yield the mineral faustite, zinc is actively replacing copper within the turquoise. As such, Lander Blue can be considered zincian turquoise and the zinc likely acts as a chromophore that enhances the deep blue coloration. The water content in the turquoise structure may be coupled with the zinc to influence the coloration, however more research is needed on zinc acting as a blue chromophore. The most interesting aspect of Lander Blue chemistry is the uranium content, which is 27 ppm to 59 ppm in the samples analyzed in the dataset. Though 50 ppm (0.005%) is considered low and is certainly not radioactive by any means, it offers a unique allocation towards the identification of Lander Blue. Most turquoise has uranium concentrations that are only 10 ppm, some can be much higher and as high as 100 ppm. This assessment allows for simplistic screening of specimens for Lander Blue origin, as if the unknown in question only has 10 ppm uranium or it is significantly higher than 60 ppm, than it is most certainly not Lander Blue. It should be noted that a uranium concentration that falls within this range also doesn’t guarantee that the specimen in question is Lander Blue, as a lot of other mines such as Indian Mountain and even some Hubei material have similar uranium concentrations. It can only be said that exceptionally low or exceptionally high uranium is the basis for assessment of none Lander Blue and the use of ternary diagrams will be needed to properly identify the specimen, which will be discussed later.
| Specimen #1 | Specimen #2 | Specimen #3 | Specimen #4 | Specimen #5 | Specimen #6 |
SiO2 | 6.15 | 7.51 | 5.02 | 7.58 | 18.58 | 16.57 |
Al2O3 | 27.42 | 39.11 | 21.32 | 31.42 | 29.36 | 27.99 |
MgO | 1.69 | 1.71 | 1.45 | 2.12 | 1.80 | 1.83 |
CaO | 0.76 | 0.32 | 0.61 | 0.16 | 0.48 | 0.64 |
K2O | 0.25 | 0.17 | 0.54 | 0.26 | 0.24 | 0.17 |
Na2O | 0.11 | 0.12 | 0.12 | 0.12 | 0.33 | 0.11 |
P2O5 | 20.30 | 30.00 | 21.14 | 23.83 | 32.18 | 31.62 |
TiO2 | 0.78 | 0.01 | 0.12 | 0.02 | 0.08 | 0.17 |
MnO | 0.02 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 |
FeO | 0.21 | 0.14 | 0.83 | 0.07 | 0.64 | 0.26 |
Fe2O3 | 2.00 | 1.27 | 0.01 | 0.59 | 3.71 | 1.70 |
S | 0.30 | 0.34 | 0.22 | 0.78 | 0.06 | 0.20 |
Cl | 0.71 | 0.16 | 0.46 | 0.40 | 0.01 | 0.05 |
|
|
|
|
|
|
|
Trace Elements (ppm wt.)
|
|
|
|
|
|
|
Strontium | 2649 | 35 | 220 | 151 | 99 | 396 |
Barium | 2180 | 1787 | 2300 | 1769 | 1050 | 2311 |
Rubidium | 13 | 6 | 12 | 1 | 14 | 11 |
Zirconium | 11 | 20 | 6 | 11 | 5 | 4 |
Molybdenum | 28 | 34 | 27 | 17 | 32 | 19 |
Vanadium | 122 | 149 | 290 | 184 | 722 | 837 |
Nickel | 29 | 29 | 27 | 192 | 39 | 37 |
Copper | 53366 | 68293 | 56895 | 48244 | 58290 | 58333 |
Zinc | 3644 | 1093 | 2053 | 2149 | 4789 | 2830 |
Chromium | 293 | 175 | 295 | 134 | 265 | 778 |
Lead | 7 | 5 | 7 | 0 | 7 | 7 |
Arsenic | 422 | 193 | 614 | 754 | 547 | 601 |
Tungsten | 310 | 458 | 529 | 0 | 102 | 86 |
Antimony | 36 | 17 | 31 | 43 | 47 | 45 |
Tin | 12 | 23 | 118 | 86 | 21 | 29 |
Silver | 205 | 22 | 41 | 1264 | 56 | 61 |
Bismuth | 7 | 5 | 6 | 17 | 4 | 4 |
Niobium | 9 | 3 | 4 | 4 | 4 | 2 |
Uranium | 34 | 42 | 45 | 27 | 59 | 38 |
Thorium | 3 | 2 | 3 | 0 | 4 | 4 |
Lithium | 0.20 | 0.03 | 0.03 | 0.09 | 0.69 | 1.44 |
Suspected Lander Blue
Mike Ryan provided ten specimens of suspected Lander Blue for analysis, again all considered very high to gem grade. Two specimens are particularly interesting in that they show more compact and larger webbing, and some show the prominent gold webbing that is very similar to specimens #5 and #6 from Bob Brucia.
Suspected Lander Blue: Top from left to right (Specimens A-E), middle from left to right (Specimens G-H), and Bottom (Specimens I-J).
Specimens A & C have the highest quartz content and are similar to specimens #5 & #6, which show the prominent gold webbing. Specimen C is particularly interesting in that it shows significant cassiterite, which is a tin oxide. From initial examination of the dataset it was evident that cassiterite may be part of the hydrothermal mineralization sequence of the Lander Blue deposit, however upon further examination the cassiterite is a relict of the polishing process. P3M’s algorithm has the unique ability to distinguish between enhancements and treatments, as well as surface alterations. The algorithm looks specifically for potassium and rubidium ratios. It’s widely accepted that high potassium is an indication of Zachery treatment. High potassium doesn’t necessarily indicate Zachery treatment, as its also an indication of wax polish. Early polishing techniques often employed paraffin wax as a finish, since paraffin wax is a solid an emulsion is made using potassium hydroxide to achieve a liquid consistency. The wax has most certainly been removed from the surface from years of natural wear, however remnants of the potassium remain. Moreover, tin oxide was likely used to polish the specimen with the wax holding remnants of it in place within the natural porosity of the surface. Specimen C is a very rare example of early lapidary work of Lander Blue. Specimens B & G have the highest clay content, where the clay likely occurs within grain boundaries. Illite is a common mineral formed from weathering of aluminum silicates, mainly feldspar and indeed both specimens B & G contain orthoclase. Conversely, chlorite is a clay mineral formed as a product of hydrothermal alteration and both specimens C & H show minor chlorite. Specimen H has the lowest quartz content of all the suspected Lander Blue analyzed, coupled with minor albite feldspar. However, specimen F has the highest concentration of feldspar, producing the greatest visual intricate gold webbing of all the suspected specimens that were analyzed. Specimen D has the highest iron concentration, occurring as limonite, hematite, and siderite, which yields more of a brown matrix in comparison to the rest of the specimens that were analyzed. Specimens I and J have comparable mineralogy to the other suspected Lander Blue specimens. However, specimen E is considerably different from all of the suspected specimens, it has the highest turquoise content of all the specimens analyzed, exceeding both the specimens of known provenance and the specimens in the suspected dataset. Additionally, it has the highest concentration of arsenopyrite, with an arsenic concentration that is two times higher than all the specimens analyzed. Moreover, it has the lowest zinc and uranium concentrations, coupled with the highest chromium concentration of all the specimens analyzed. Such deviations indicate that specimen E cannot be classified as Lander Blue.
| Specimen #A | Specimen #B | Specimen #C | Specimen #D | Specimen #E |
Quartz | 21.26 | 5.03 | 22.54 | 9.47 | 0.79 |
Orthoclase | 3.48 | 3.87 | 0.01 | 0.97 | 0.96 |
Chromite | ----- | 0.02 | 0.01 | 0.01 | 0.04 |
Ilmenite | 0.16 | ----- | ----- | ----- | ----- |
Magnetite | ----- | 0.02 | 1.47 | ----- | 0.24 |
Hematite | 0.96 | ----- | 0.01 | 3.22 | 1.56 |
Rutile | 0.05 | ----- | ----- | ----- | ----- |
Turquoise | 69.99 | 81.26 | 72.81 | 79.60 | 94.75 |
Ankerite | 1.68 | 0.47 | ----- | 1.32 | ----- |
Siderite | ----- | 0.47 | ----- | 1.32 | ----- |
Rhodochrosite | 0.05 | ----- | 0.02 | ----- | 0.02 |
Azurite/Malachite | 0.22 | 1.04 | 0.40 | ----- | 0.05 |
Halite | 1.18 | 0.61 | 0.42 | 0.53 | 0.36 |
Fluorite | 0.01 | 0.02 | 0.02 | ----- | 0.01 |
Pyrite | 0.91 | 0.62 | 0.59 | 0.32 | 0.49 |
Chlorite/Serpentine | ----- | ----- | 1.04 | ----- | ----- |
Illite/Clays | ----- | 6.19 | ----- | ----- | ----- |
Chalcopyrite | ----- | ----- | 0.02 | ----- | 0.01 |
Bornite | ----- | ----- | 0.05 | ----- | 0.06 |
Proustite | ----- | 0.37 | ----- | ----- | ----- |
Arsenopyrite | ----- | ----- | 0.11 | 0.28 | 0.66 |
Cassiterite | ----- | ----- | 0.33 | 0.03 | ----- |
Scheelite | 0.05 | ----- | 0.06 | 0.02 | ----- |
Linonite/Goethite | ----- | ----- | 0.09 | 2.90 | ----- |
Pyrolusite | ----- | 0.01 | ----- | 0.01 | ----- |
| Specimen #F | Specimen #G | Specimen #H | Specimen #I | Specimen #J |
Quartz | 7.33 | 3.56 | 0.36 | 2.23 | 7.39 |
Albite | 1.48 | ----- | 0.56 | ----- | ----- |
Orthoclase | 9.17 | 1.88 | 3.58 | 1.33 | 1.54 |
Magnesiochromite | 0.02 | 0.04 | 0.02 | ----- | 0.04 |
Ilmenite | ----- | ----- | ----- | 0.03 | ----- |
Magnetite | ----- | ----- | ----- | 0.22 | ----- |
Hematite | 2.29 | 0.76 | 0.27 | 1.35 | 0.79 |
Rutile | 0.05 | 0.94 | 0.04 | ----- | 0.02 |
Turquoise | 78.00 | 86.67 | 90.22 | 92.00 | 87.56 |
Ankerite | ----- | 0.54 | ----- | 0.71 | 0.95 |
Rhodochrosite | ----- | 0.02 | 0.03 | 0.03 | 0.02 |
Azurite/Malachite | ----- | 0.23 | 0.56 | 0.11 | 0.27 |
Halite | 0.52 | 1.35 | 1.45 | 0.98 | 0.67 |
Fluorite | ----- | 0.02 | 0.02 | 0.02 | 0.02 |
Pyrite | 0.40 | 0.49 | 1.38 | 0.48 | 0.47 |
Chlorite/Serpentine | ----- | ----- | 1.19 | ----- | ----- |
Illite/Clays | 0.30 | 3.26 | ----- | ----- | ----- |
Chalcopyrite | ----- | 0.02 | 0.03 | 0.01 | 0.01 |
Bornite | ----- | 0.06 | 0.06 | 0.05 | 0.05 |
Arsenopyrite | 0.29 | 0.11 | 0.14 | 0.45 | 0.16 |
Acanthite | ----- | ----- | 0.03 | ----- | ----- |
Cassiterite | 0.13 | ----- | ----- | ----- | ----- |
Scheelite | ----- | 0.05 | 0.06 | ----- | 0.04 |
Pyrolusite | 0.02 | ----- | ----- | ----- | ----- |
| Specimen #A | Specimen #B | Specimen #C | Specimen #D | Specimen #E |
SiO2 | 19.92 | 7.81 | 13.66 | 8.44 | 2.18 |
Al2O3 | 18.26 | 26.57 | 17.67 | 23.52 | 28.64 |
MgO | 1.53 | 1.55 | 1.40 | 1.87 | 1.79 |
CaO | 0.76 | 0.53 | 2.77 | 0.87 | 0.69 |
K2O | 0.47 | 0.59 | 2.54 | 0.15 | 0.14 |
Na2O | 0.35 | 0.12 | 3.36 | 0.02 | 0.78 |
P2O5 | 16.60 | 22.26 | 18.32 | 19.18 | 25.46 |
TiO2 | 0.10 | 0.10 | 0.05 | 0.03 | 0.01 |
MnO | 0.02 | 0.01 | 0.01 | 0.01 | 0.01 |
FeO | 0.47 | 1.40 | 0.93 | 1.39 | 0.49 |
Fe2O3 | 2.53 | 0.01 | 1.30 | 12.56 | 4.96 |
S | 0.38 | 0.34 | 0.28 | 0.19 | 0.26 |
Cl | 0.50 | 0.30 | 0.18 | 0.27 | 0.17 |
|
|
|
|
|
|
Trace Elements (ppm wt.)
|
|
|
|
|
|
Strontium | 1230 | 228 | 463 | 386 | 472 |
Barium | 1413 | 1033 | 1943 | 1306 | 810 |
Rubidium | 17 | 19 | 20 | 17 | 6 |
Zirconium | 9 | 9 | 5 | 6 | 6 |
Molybdenum | 36 | 30 | 36 | 399 | 26 |
Vanadium | 268 | 267 | 142 | 958 | 380 |
Nickel | 25 | 25 | 26 | 52 | 40 |
Copper | 43211 | 57559 | 45399 | 57918 | 65245 |
Zinc | 1421 | 4868 | 1822 | 2965 | 759 |
Chromium | 226 | 439 | 260 | 345 | 875 |
Lead | 115 | 22 | 7 | 203 | 26 |
Arsenic | 443 | 530 | 397 | 1205 | 2677 |
Tungsten | 243 | 88 | 303 | 112 | 80 |
Antimony | 11 | 42 | 25 | 94 | 18 |
Tin | 70 | 128 | 2068 | 221 | 16 |
Silver | 33 | 37 | 28 | 20 | 8 |
Bismuth | 6 | 6 | 6 | 7 | 6 |
Niobium | 14 | 5 | 6 | 8 | 3 |
Uranium | 33 | 49 | 77 | 60 | 17 |
Thorium | 3 | 2 | 3 | 0 | 3 |
Lithium | 0.42 | 0.07 | 0.00 | 0.02 | 0.00 |
| Specimen #F | Specimen #G | Specimen #H | Specimen #I | Specimen #J | |
SiO2 | 12.15 | 4.72 | 4.85 | 3.79 | 8.27 | |
Al2O3 | 20.60 | 26.96 | 25.73 | 24.60 | 23.23 | |
MgO | 1.73 | 1.63 | 1.69 | 1.79 | 1.60 | |
CaO | 1.37 | 0.74 | 0.87 | 0.49 | 0.58 | |
K2O | 1.32 | 0.28 | 0.46 | 0.18 | 0.21 | |
Na2O | 1.46 | 0.11 | 2.08 | 0.33 | 0.11 | |
P2O5 | 18.99 | 21.62 | 21.68 | 22.94 | 21.63 | |
TiO2 | 0.04 | 0.83 | 0.03 | 0.01 | 0.01 | |
MnO | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | |
FeO | 0.00 | 0.22 | 0.49 | 0.38 | 0.24 |