The Curious Case of Columbium
Peter E. Childs

Introduction
Current periodicals on metals published in the U.S.A. have advertisements for "Columbium", symbol Cb, and its alloys. However, you will not find this metal in a European catalogue of metals and their compounds, or in the IUPAC-aproved Periodic Table - instead you will find the element niobium, symbol Nb, with the same physical and chemical properties. And thereby hangs a tale.

A hatchet job on Hatchett
Very few chemists today have heard of Charles Hatchett (born 2/1/1765 and died 10/3/1847 in London). In his day Charles Hatchett had a reputation as a skilled chemical analyst. He was the son of a wealthy coach builder in London, who built coaches for royalty and thus he was brought up in luxury. We know hardly anything of his education or how he came to his knowledge of chemistry. His father wanted Charles to follow in his profession and was disappointed when he chose science instead. He is said to have offered his son £3,000 (an enormous sum of money then) and a seat in Parliament if he would give up chemistry. Quite a temptation! However, he made Charles a generous allowance so that he was free to pursue his chemical interests.

In 1786 Charles Hatchett married Elizabeth Collick and in 1800 started a small chemical manufacturing business near Chiswick in London. He employed a young man called William Thomas Brande and taught him chemistry and mineralogy. Brande was eventually to become a Professor at the Royal Institution in London. In 1818 William Brande married Hatchett's second daughter Anna. Hatchett was a friend and neighbour of the Limerick chemist Peter Woulfe, famous for the Woulfe's bottle.

Charles Hatchett's important scientific work was done in the period 1796 to 1806. He was elected a Fellow of the Royal Society in 1797, which is a measure of the esteem in which he was held by his fellow scientists. In 1801 he described in a paper his analysis of a mineral called columbite, named after the location where it had been found in North America. This mineral sample from Massachusetts had lain in the British Museum since 1753. He described the mineral as ".. a heavy black stone with golden streaks ... from Mr. Winthrop": John Winthrop was the first Governor of Connecticut, the source of the mineral. Hatchett showed that the mineral contained a new element and he called it columbium and the mineral columbite, after its place of origin. This year (2001) marks the bicentenary of his discovery.

Niobium (Columbium) always occurs with tantalum because of the similarity in their atomic size. The minerals columbite and tantalite contain both elements, differing only in their proportion: columbite contains more niobium and tantalite more tantalum. In 1802 Ekeberg discovered a new metal in a rare earth mineral called yttrotantalite, and called it tantalum because like Tantalus in the Greek myths who could not drink, this new element would not react with acids. Ekeberg wrote: "This metal I call tantalum .. partly in allusion to its incapacity, when immersed in acid, to absorb any and be saturated."

Both columbite and tantalite were analysed by William Wollaston in 1809 (after Hatchett had effectively given up science and taken over his father's coach-building business following the latter's death). Wollaston was confused by the similarity in the physical and chemical properties of the two elements and he thought they were the same i.e. he stated that Hatchett's columbium and Ekeberg's tantalum were in fact the same element and consequently the two elements were confused until 1844. The result was that due to Wollaston's influence, Hatchett's claims and name for the new element were disregarded, although Ekeberg's name survived attached to both elements. Two elements for the price of one!

The matter was not resolved until 1844, not long before Hatchett died, when Heinrich Rose 'rediscovered' columbium, but he now called it niobium after the Greek nymph Niobe, who was the daughter of Tantalus - thus recognising the close relationship between the elements tantalum and niobium (columbium). Hatchett had already shown that the oxide of niobium, Nb2O5, and the corresponding oxide of tantalum, Ta2O5, had different properties. Wollaston's reputation had ensured that his erroneous views prevailed over Hatchett's, especially as Hatchett had given up science by then. Even Rose's researches failed to clarify the chemistry of niobium and tantalum, because of their complexity and similarity to each other.

The Swedish chemist Blomstrand (1866) and the Swiss chemist Marignac (1866) finally sorted out the chemistry of the two elements. It was Blomstrand who first isolated metallic niobium by reducing niobium(V) chloride with hydrogen. Roscoe obtained a better yield by reducing the trichloride, which he had discovered. The first pure samples of both niobium and tantalum were not made until 1907 when W. von Bolton reduced K2NbOF5 and K2TaOF5 with sodium metal. It had taken over 100 years to go from the discovery to the isolation of niobium (columbium).

Confusion compounded
Both names for element 41 are still in use. The official IUPAC name is niobium (Nb), adopted as late as 1950, but the North Americans always called it columbium and in the metals industry they still do. Why this name should have persisted in this way across the Atlantic is a mystery, as is the persistence in using it despite international agreement to call it niobium. The fact that it was named after its America and its American origin is probably sufficient explanation.

Hatchett's reputation in Great Britain and Europe was as a mineral analyst, but he also applied his skills to the analysis of shell, bone and dental enamel. He also did work in natural product chemistry and Thomas Thomson lamented his loss to chemistry as a result of the "baneful effects of wealth and business cares".

No doubt his absorption is his business affairs ensured that he hadn't the time or the inclination to answer Wollaston's criticisms of his work and to defend his discovery of columbium. If he had done so, and shown that tantalum and columbium were different elements as was done later, then we would all be calling element no.41 columbium and not niobium.

The interesting aspect of this whole affair is the persistence of this name in North America even today, despite the fact that Hatchett wasn't an American! For example, the Mineral Commodity Summaries from the U.S. Department of the interior refers to the metal as Columbium (Niobium).

Extraction and sources of niobium
Niobium and tantalum always occur together, in varying proportions, in the columbite-tantalite minerals of formula (Fe,Mn)M2O6 (M=Nb,Ta), and in pyrochlore. Niobium has a crustal abundance of 20 ppm. and tantalum 1.7 ppm.

Alkali fusion followed by digestion with acids is used to dissolve out impurities in the minerals leaving Nb and Ta undissolved. The problem is then to separate the two similar elements. M.C. Marignac developed a method in 1866, which remained in use for a century. He treated the mixture with dilute hydrofluoric acid: tantalum forms the less soluble K2TaF7 and can be separated form the more soluble K3NbOF5.3H2O. The modern process uses solvent extraction to separate the metals. Tantalum is extracted from dilute HF solution by methylethylketone (MEK) and increasing the acidity allows niobium to be extracted separately. Addition of water strips out the metals from MEK solution. Both metals are then converted to their pentoxides, M2O5, which are reduced with either sodium or carbon to the metals. The carbon reduction is done at over 2000°C in vacuum and the product is vacuum refined by electron beam melting. An alternative process uses the electrolysis of molten fluorides.

Production figures
In 1999 the world mine production of niobium was 18,500 tons with Brazil (16,000 tons) and Canada (2,300 tons) being the major sources. World mine production of tantalum is much smaller at 473 tons in 1999, Australia (350 tons), Brazil (60 tons) and Canada (60 tons) being the major sources.

Uses of niobium
Niobium's main use is in stainless steels as ferroniobium (ferrocolumbium). Niobium/zirconium wire is used in superconducting magnets, such as those used in NMR machines and body scanners.

References
Dictionary of Scientific Biography, New York: Charles Scribners, 1972
C.Hatchett, Phil. Trans. Royal Society 92 49-66 1802
E.M.Weeks, J.Chem.Ed., 15 153-158 1938
(reprinted in Discovery of the Elements, Chemical Education Publishing Company, 7th. edition 1968)
J.Partington. A History of Chemistry, Vol. III London: Macmillan, 1972 p 705-706 J.Partington. A History of Chemistry, Vol.IV, London:Macmillan 1972, pp.188-190
U.S. Mineral Commodity Surveys,
http://minerals.usgs.gov/minerals/pubs/mcs/2000/mcs2000.pdf

This article was previously published in Chem13News. It is republished in CinA! to mark the bicentenary of Hatchett's original discovery and naming of niobium as columbium. The production statistics have been updated to 1999.


"All great scientists have, in a certain sense, been great artists; the man with no imagination may collect facts, but he cannot make great discoveries."
Karl Pearson
The Grammar of Science