Page images

In the analyses by Jannasch and Kalb the following examples are very striking:

[blocks in formation]

It would hardly be wise to dismiss these variations as due to errors, especially when the summation of the analyses is good and the analysts are known to be trustworthy. Such errors on the part of either Jannasch or Riggs would be almost incredible, and I am therefore inclined to believe that the analyses are good, and that we should seek a cause for the variations. In my scheme of formulation the bivalent group

of atoms=Al- BO, occurs. Replace this in part by the similar gronps

-Al-OH and Al-F, and the variations are accounted for. This supposition satisfies the analyses completely, and covers the ground. It is in accord with all the evidence, even though its validity may not be definitely proved. By its application to the discussion of the analyses the divergencies between the calculated composition and the composition as found can be notably diminished.

But although the formulæ which I have adopted serve to express the composition of all tourmalines, they sủill leave room for alternatives. Penfield and Foote, as well as myself, assume that tourmaline is a mixed salt containing distinct boric and silicic radicles. Future investigation may prove that it is really derived from a complex boro-silicic acid, as yet unknown; and the same conception may be true of other species, such as axinite, datolite, danburite, cappelenite, etc. A series of borosilicic acids is theoretically conceivable; and until this question bas been considered, the constitution of all the minerals above mentioned must be regarded as unsettled.





In the analysis of a number of highly titaniferous magnetites containing chromium, phosphorus, and vanadium, the satisfactory separation of all these bodies in a form fit for separate determinations became a serious problem. The method of T. Fischer -digestion of the precipi. tated lead salts with a strong solution of potassium carbonate-appears to offer the long-needed satisfactory quantitative separation of arsenic, phosphorus, chromium, tungsten, and molybdenum from vanadium, the normal lead meta-vanadate remaining quite unattacked, according to the author, while the other lead salts are wholly decomposed, but the applicability of this method to the separation of the minute amounts often found in rocks and ores has not been tested. The time required is considerable, hence it was desirable to devise a more rapid way for determining both chromium and vanadium without resorting to this separation. That this object has been measurably achieved, with cer. tain limitations as to vanadium, the work thus far done seems to indicate. The present paper will deal only with the rapid estimation of chromium either in absence or presence of any or all of the elements above mentioned.

In view of the high coloring power of the chromates, it is surprising that so little use has been made of this property as the basis for a quantitative method for the estimation of chromium. A search through some of the more important textbooks has revealed no reference to such a method, although L. de Koningh? has successfully applied it in the analysis of articles of food. Yet the results attainable by colori. metric comparisons of dilute alkaline solutions of unknown strength with those of a known standard leave little to be desired in point of quantitative accuracy.

"Inaugural Dissertation, Rostock, 1894.
2 Nederl. Týdsch. voor Pharmacie, Chemie, on Toxicologie, 1889, p. 257.

As with colorimetric methods in general this one gives better results with small than with large percentages of chromium, yet it can be applied in the latter cases with very fairly satisfactory rosults by making a larger number of consecutive comparisons with the same solution.


The chromium is brought into a measured volume of solution as monochromate rendered alkaline by sodium carbonate, and the whole or a portion of this solution is then compared with a definite amount of a somewhat stronger standard, likewise made alkaline with sodium carbonate. The latter is diluted with water till both seem to be exactly alike in color, when a simple calculation gives the amount of chromium sought. The actual comparison takes little time, and any number of repetitions can be made if desired in order to secure greater accuracy from the mean of a large number of observations. The preparation of the solution to be tested offers nothing novel, but certain precautions have to be observed therein as well as in the color comparisons which will be touched upon later.


Two standard solutions were prepared by dissolving 0.25525 and 0.5105 gram potassium chromate in 1 liter of water made alkaline by a little sodium carbonate, each cubic centimeter then corresponding respectively to 0.1 and 0.2 milligram chromic oxide, in which latter form chromium is usually reported in rocks and ores. Definite amounts of one of the standards were then diluted with varying amounts of water in a tall, square glass vessel with exactly parallel sides. Into an exact duplicate of this vessel 5 cubic centimeters or more of the standard were introduced from a burette and diluted with water from another burette till exact agreement seemed to be reached on looking through the glasses horizontally.

In the following tables are recorded all observations without regard to the sequence in which they were made. No greater pains were taken to get exact agreement of color than are ordinarily observed in our routine titanium estimations, which are carried out in a precisely similar manner, so that the results may be taken to represent everyday work without extreme precautions. In only two cases can the observations be considered really bad, viz, the third comparison of No. 6 and the first of No. 16.


(Ten cubic centimeters standard represent 1 milligram chromic oxide.)

[blocks in formation]
[blocks in formation]

Per cent.


[blocks in formation]



Mg8. 0.98 1.92 4.06 4.05 4. 17 4.11 4. 13 4.05



[merged small][ocr errors]
[blocks in formation]



[blocks in formation]
[blocks in formation]
[blocks in formation]
[blocks in formation]
[blocks in formation]

Mgs. Mgs. 1 -0.02 2

.08 4 + .06 4 + .05 4 +.17

+.11 4 + .13 4

+ .05 4

.10 4

.04 4

+ .11 4

It .32 4 It .04 4

H.04 5 +.06 5

.00 6

. 19

+.005 7.5


+ .105
7.5 +.11


.006 1

+.02 1.42 + .01 1. 42 + .05 1.5 .02 1.6 .06 1.6 + .03 3. 19 . 23 3. 19 .09 6. 205 + .45 6. 205 + .285

44.00 21.00 14. 60 14. 25 13.95 14. 30 14.20 14. 65 15. 65 15. 25 14.30 13. 15 14. 75 29.50

9.75 10.00 7. 20 6.65 6. 60 3. 15 4.70 10. 15 19.75 10.60 18. 60 35. 70 16.55 13.30 24. 30

8.10 14. 20

7.00 12.00

40 25


98.0 96.0 101.5 101. 2 104.4 102.9 103.3 101.4 97.5 99.0 102.9 108. 0 101.0 101.0 101. 2 100.0

96.7 100.1 100.2 101.4 101,5

99.2 100.6 102.0 100,7 103,5 98.7 96. 2 101.9 92.8 97.2 100.7 104. 6

[blocks in formation]
[blocks in formation]
[blocks in formation]

1. 006
1. 02
1. 63
2. 96
3. 10
6. 25


11 26.9

[blocks in formation]



[merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small]

1.0 2.0 1.0 1.0 2.0 1.05 2.05 2.0 4.0

[blocks in formation]

Mean percentage found, 100.5.

a Color in this dilution too faint.
b Limit of dilution for clear distinction of color in a thickness of 3.3 cm.


(Varying amounts of standard No. 2 (1 c. c. = 0.2 mg. Cr2O3) diluted till of the same concentration as

standard No. 1.)

[blocks in formation]

Mean percentage found, 99.3; grand mean, 100.36.

The first table and the grand mean show an apparent personal tendency toward slightly high results, though it is possible that this is due to a slight difference in the iuternal dimensions of the two glasses, the same one always having been used for the standard solution. If this is so, a long series of tests with glasses reversed should give a general mean slightly below 100.

TESTING THE METHOD ON ORES AND Rocks. In order to prove the value of the method in rock analysis, varying amounts of the standard solution were evaporated in a large crucible with 5 grams of an iron ore carrying phosphorus and vanadium, and fused with 20 grams of sodium carbonate and 3 grams sodium nitrate. The aqueous extract, after reduction of manganese by methyl or ethyl alcohol, was nearly neutralized by nitric acid and evaporated to secure approximate separation of silica and alumina. As a precautionary measure, since a little chromium is usually carried down, the precipi. tate was ignited, silica was removeil by hydrofluoric and sulphuric acids, the residue was fused with sodium carbonate, and alumina again separated as before. To the combined filtrate was added mercurous nitrate, and the slightly washed copious precipitate of phosphate, chromate, vanadate, and carbonate of mercury was ignited with the paper in a platinum crucible, which can be done without much fear of loss or of injury to the crucible. The residue was then fused with a little sodium carbovate, extracted with water, filtered into a graduated flask and made up to 50 or 100 cubic centimeters, according to the intensity of the color, and compared with the standard. A similar operation was carried out with a silicate rock. Table III shows the results.

« ՆախորդըՇարունակել »