First. The cobalt, iron, and copper solutions are prepared as definite volumetric solutions of half-normal strength. Second. The blending of these in any proportion to a constant volume insures the widest possible range of tints. It was found that in blends of the neutral (unacidulated) half-normal solution precipitation occurred within a few days after preparation; so, in subsequent work, acidulated solutions of half-normal strength were used, the solvent being a diluted hydrochloric acid made from 25 Cc. of concentrated (30 per cent.) acid and 975 Cc. of water. This modified the tints of the original solutions and the blends, but not enough to seriously affect the value of the fluids as standard colors, which were thus made practically permanent. This line of colors, which we called the " Co-Fe-Cu" tints, was fully described in the Druggists' Circular for March, 1914, and in that article it was frankly stated that the value of these tints was lessened through the fact that the cobalt solution did not furnish a sufficiently pure red color. In short, while the "Co-Fe-Cu" blends furnished an ideal line of green, yellow, and orange tints, they lacked in deep reds and dark blues. During the past year we have been able to pursue the problem to a successful issue by preparing a line of ammoniacal fluids which furnish most of the necessary tints. This line, which we call the "Co-Cro-Cu" fluids, was made up as follows: Red tenth-normal ammoniacal cobalt solution was prepared by dissolving 13.5 grammes of roseo-cobaltic-chloride (CoCl, 5NH.. H2O) in enough of a fluid consisting of 100 Cc. of 28 per cent. ammonia water and 900 Cc. of water to make a litre of finished product.* Since this strength of acid was used throughout our work, since it was made up empirically, and since its approximate strength is about 1 in 100, we will in future speak of it as 1 per cent. hydrochloric acid. The tenth-normal and fiftieth-normal solutions here described are based on the absolute amount of metal contained in each. As cobalt has the atomic weight 58.97 and a valence of 2, we assume that a tenth-normal cobalt solution will contain 2.9485 Gm. of the metal per litre and that the fiftiethnormal will contain one-fifth of this amount. As copper has the atomic weight 63.57 and the valence is 2, we assume that a tenth-normal copper solution will contain 3.1785 Gm. of the metal to the litre and that the fiftieth-normal will contain one fifth of this amount. As chromium has the atomic weight 52.00 The solution thus prepared was then assayed gravimetrically and so diluted that each cubic centimetre of fluid contained 0.0029485 Gm. of metallic cobalt. Experience showed, however, that empiric solutions prepared by dissolving 13.43 Gm. of the roseocobaltic chloride in a diluted ammonia of the strength just stated did not vary sufficiently from the standard in color intensity to make assay necessary in average work. If the commercial roseo-cobaltic chloride is not obtainable, it can be prepared as follows: Dissolve 12 Gm. of cobalt chloride (CoCl26H2O) in about 100 Cc. of water. After filtration, add 100 Cc. of 28 per cent. ammonia, shake well, and keep in a warm place (between 50° to 70° C.) six or eight hours, with occasional shaking. Let stand over night at ordinary room temperature. Warm again at same temperature as before and gradually dilute with water, a small portion at a time in such a way that the temperature of the liquid will not go below 50°, until it has been diluted to 1 litre. We now have an empiric tenth-normal ammoniacal cobalt chloride of a bright red color. This solution shows a yellow tinge when examined closely and therefore cannot be used as a standard. But from it the roseo-cobaltic salt can be made by the addition of an excess of concentrated hydrochloric acid and then boiling for several minutes until a precipitate forms. The precipitation can be hastened by chilling with running water. The precipitate collected on a filter is washed with 1 per cent. hydrochloric acid and is then dried in oven at about 100° C. Yellow tenth-normal ammoniacal chromium solution was made by dissolving 3 Gm. of ammonium dichromate ((NH4)2Cr2O7) in 50 Cc. of 1 per cent. hydrochloric acid, then adding 50 Cc. of 28 per cent. ammonia water, and finally diluting with enough water to make one litre. The solution was then assayed gravimetrically and so diluted that each cubic centimetre of fluid contained 0.0008666 Gm. of chromium. It was found, however, that by making an empiric solution of 2.1 Gm. of ammonium dichromate and presumably has the valence 6 in the dichromate, then a tenth-normal solution will contain 0.8666 Gm. of chromium to the litre and a fiftieth-normal I will contain one-fifth of this amount. This conception of fiftieth-normal chromium is quite different from the commonly-accepted term, tenth fiftieth or hundredth-normal dichromate solution used in the oxidizing titrations; in which case, as is well known, the titre of the solution is based on its oxygenproducing capacity, rather than its absolute chromium content. dissolved in I per cent. hydrochloric acid and then adding 28 per cent. ammonia water and diluting with water to 1 litre the color value was the same as that of the standardized solution. Blue tenth-normal ammoniacal copper solution was prepared by diluting 200 Cc. of standard assayed half-normal copper sulphate (containing 62.43 Gm. of CuSO45H2O to the litre) to tenthnormal alkaline fluid, by adding 100 Cc. of 28 per cent. ammonia water and then making up to 1 litre with water. Each cubic centimetre of this solution contained 0.0031785 Gm. of metallic copper. As with the other fluids, a carefully-prepared empiric solution. gave the same color value as the assayed standard. Assayed standard half-normal solutions of both copper sulphate and copper chloride were prepared. Each was diluted to tenth-normal ammoniacal copper in the manner just outlined and the two blue colors compared. They were practically identical both in tint and intensity of color. In making empiric copper solution, however, only perfect crystals of chemically pure copper sulphate were used. Further data concerning these three standard solutions will be found below. At this place we will merely point out that for ordinary purposes of comparison we found the blends prepared from these tenth-normal ammoniacal solutions were too dark. We therefore diluted them down with water to equal fiftiethnormal solutions, and from these latter we prepared the " Co-CroCu" blends. COMPARISON OF STANDARD TINTS. When starting on the investigation of which this paper is a part, the first difficulty encountered was the lack of an accurate and available basis of comparison for fluid colors. Of course, the solar spectrum itself was the first standard thought of, and one of us, with Dr. H. H. Schaefer, spent some weeks during 1913 attempting to match our colored fluids against this natural standard. But like other investigators, we found that the very brilliancy of the spectrum colors was a drawback, and that our colored fluids, attractive though they were, paled before the glory of a spectrum segment. We still hope, however, that we may yet be able to devise some scheme whereby our colors can be expressed in terms of solar tints. Finding, for the time being, this ideal plan futile, we turned to that interesting color apparatus, the Lovibond tintometer, which, as was explained in the Druggists' Circular article cited above, possessed the advantages of simplicity, permanency, and availability. Another of its advantages for our work was that the colors of the glass slides used in the tintometer were not superior to the tints of our standard colored fluids. In the article just referred to, the Lovibond readings of the acidulated half-normal (red) cobalt, (yellow) ferric iron, and (blue) copper solutions were given, as well as similar readings of the 88 blends already mentioned. During the past winter Lovibond readings of the "Co-CroCu" blends have been made, and the results are tabulated below: LOVIBOND READINGS OF THE TENTH-NORMAL STANDARD AMMONIACAL SOLUTIONS IN A 4-INCH CELL. NOTE. In this and all other Lovibond readings in this paper, the red Lovibond glasses were No. 200, the yellow were No. 510, and the blue were 1180. FIFTIETH-NORMAL STANDARD AMMONIACAL SOLUTION IN 2-INCH CELL. |