Caesium was first described in 1861 by Gustav Robert Kirchhoff and Robert Wilhelm Bunsen. They examined mineral water from Dürkheim and, after separating calcium, strontium, magnesium and lithium, discovered two previously unknown lines in the blue spectral range. From their observations, they concluded that there must be another, previously unknown element in the mineral water examined, which they named caesium, after the Latin caesius for "sky blue", because of the blue spectral lines.
Bunsen also attempted to separate cesium from the other alkali metals in order to explore further properties of the element. To do this, he added a platinum chloride solution to the solution in order to precipitate potassium and the newly discovered heavier alkali metals rubidium and cesium as an insoluble hexachloridoplatinate. The potassium could be removed by boiling several times in a little water. To obtain the pure chlorides, the platinum was reduced to the element with hydrogen so that the now water-soluble cesium and rubidium chlorides could be leached out. The separation of cesium and rubidium was accomplished by taking advantage of the different solubility of the carbonates in absolute ethanol, in which cesium carbonate is soluble in contrast to the corresponding rubidium compound. Cesium chloride also served Bunsen and Kirchhoff for a first determination of the molar mass of the new element, for which they found the value of 123.35 g/mol.
The two researchers were unable to obtain elemental cesium because electrolysis of molten cesium chloride produced a blue compound instead of the metal, which they called subchloride, but which was probably a colloidal mixture of cesium and cesium chloride. Electrolysis of an aqueous solution with a mercury anode formed the easily decomposable cesium amalgam.
The preparation of elemental caesium was finally achieved in 1881 by Carl Theodor Setterberg, who avoided the problems with chloride by using caesium cyanide for the fused-salt electrolysis. Initially, the relatively high temperature required to melt the cesium cyanide was a problem, but he was able to reduce this by eutecticizing with barium cyanide.