Atomic Weights of Titanium, History

The earliest accepted values for the atomic weight of titanium were inaccurate on account of imperfect analytical methods. Thus Dalton, in 1808, gave the value Ti = 40 (O = 7), whilst Berzelius in 1813 suggested the number 1801.0 (O = 100) or 288.16 (O = 16). That the atomic weight of titanium must be of the order of 48 is evident from the following considerations:

1. The specific heat of the metal between 0° C. and 440° C. ranges from 0.112 to 0.162. Assuming a mean atomic heat of 6.4, according to the law of Dulong and Petit, the atomic weight of titanium must lie between 57 and 40.
2. Rutile is isomorphous with cassiterite (SnO₂); and fluotitanates are isomorphous with fluosilicates, M₂'SiF₆, fluozirconates, M₂'ZrF₆, and fluostannates, M₂'SnF₆. From Mitscherlich's Law of Isomorphism it follows that the formula for rutile is TiO₂, whilst the general formula for fluotitanates is M₂'TiF₆. Now analysis shows that each of these gram-molecules contains approximately 48 grams of titanium. This latter weight, therefore, is approximately the atomic weight of the element.
3. There is a space in the fourth group of the Periodic Table for an element of atomic weight lying between 44 and 51. The chemical and physical properties of titanium indicate that this is the element required satisfactorily to fill the gap. Judging by comparison of the differences between the atomic weights of other adjacent elements in the table, the atomic weight of titanium should approximate to 48.

The analytical investigations on which the atomic weight of titanium is based fall into two categories: (i) the work of Berzelius, H. Rose, Mosander, Pierre, and Demoly from 1813 to 1849; (ii) the work of Thorpe, 1883-1885.

Except for the early work of Berzelius, and that of H. Rose in 1823, who prepared the disulphide TiS₂, and converted it into the dioxide TiO₂, the processes adopted have consisted in hydrolysing the pure tetrachloride or tetrabromide with water or alkali, and estimating the titanic oxide and the hydrochloric or hydrobromic acid produced. The titanic oxide was estimated by Thorpe by decomposing the tetra-halide with water in a platinum or porcelain dish, according to the reaction

TiX₄ + 3H₂O ⇔ H₂TiO₃ + 4HX,

evaporating with ammonia, and then igniting and weighing the residue. The halogen hydracid was estimated by hydrolysing the titanium halide, allowing the solution to stand until it was clear, then either precipitating nearly all the halogen with a known weight of pure silver dissolved in nitric acid, and finishing the estimation by titrating with centinormal silver solution, or by adding excess of silver solution and weighing the silver halide precipitated.

The results of Thorpe alone are accepted in computing the atomic weight of titanium. They are as follows:

Ratio	Atomic Weight of Titanium
TiCl4:4Ag = 0.43997:1	48.03
TiCl4:4AgCl = 0.33119:1	48.06
TiCl4:TiO2 = 2.3712:1	48.09
TiBr4:4Ag = 0.85234:1	48.14
TiBr4:4AgBr = 0.48965:1	48.16
TiBr4:TiO2 = 4.5931:1	48.06
Mean value	48.09

The mean value adopted by the International Committee on Atomic Weights for 1917 is 48.1.