The discovery of the second “spectral element” occurred in the studies of a rare mineral, lepidolite (called also lilalite because of its lilac colour). For the first time a detailed chemical analysis of lepidolite was performed by M. Klaproth at the end of the 18th century. But the experienced analyst did not discover alkalis in the mineral. Doubting his own results, Klaproth decided to repeat the analysis and this time (1797) he found the following components: 54.5% silicon dioxide, 38.25% aluminium oxide, 4% potassium oxide, and 0.75% manganese oxide. The missing 2.5 per cent Klaproth ascribed to the loss of water contained in the mineral. However, no matter what ingenious techniques the chemist tried, he could not determine the content of the two most important components: lithium (it had not been discovered yet by that time) and fluorine; thus, the nature of lepidolite remained obscure.
At the beginning of 1861 a sample of this mineral from Saxony fell into the hands of R. Bunsen and G. Kirchhoff, who separated alkaline components form it and precipitated potassium in the form of chloroplatinate. After a thorough washing the precipitate was subjected to spectral analysis. On February 23, 1861, the chemists reported the existence of a new alkali metal in lepidolite to the Berlin Academy of Sciences. The Scientists asserted that magnificent dark red colour of the line of the new metal gave them every reason to name the element “rubidium” and assign to it the symbol Rb from the Latin word rubidus, which meant a deep red colour. Then Bunsen and Kirchhoff discovered rubidium in the same mineral spring water in which cesium was found a year before. The rubidium content turned out to be only slightly higher than that of cesium. Metallic rubidium was prepared by R. Bunsen in 1863.