REPROCESSING SLIMES TAILINGS FROM A TUNGSTEN MINE

Citation
D. Clemente et al., REPROCESSING SLIMES TAILINGS FROM A TUNGSTEN MINE, Minerals engineering, 6(8-10), 1993, pp. 831-839
Citations number
8
Categorie Soggetti
Engineering, Chemical","Metallurgy & Mining",Mineralogy
Journal title
ISSN journal
08926875
Volume
6
Issue
8-10
Year of publication
1993
Pages
831 - 839
Database
ISI
SICI code
0892-6875(1993)6:8-10<831:RSTFAT>2.0.ZU;2-Q
Abstract
This paper describes the development of a new process route for ultraf ine wolframite recovery, to enable the efficient reprocessing of histo rical mine tailings as well as current plant slimes tailings from the Beralt Tin and Wolfram operations at Minas da Panasqueira in Portugal. Much of the wolframite in these tailings is below 25 microns in size and is associated with a complex mixture of other minerals including z inc, copper and other sulphides, tourmaline, siderite, ferromagnetic a nd other strongly magnetic material. The paper reports on flotation, m agnetic separation and gravity concentration testwork, undertaken at b oth laboratory and pilot scale on the historical mine tailings and on the current slimes tailings of the plant. Froth flotation of the Panas queira wolframite has been investigated over a number of years with ve ry inconsistent results. This work highlighed further inconsistencies between the current tailings (which floated) and the old dam tailings (which did not). High intensity magnetic separation (at low to moderat e fields) could not recover the finest wolframite and could not achiev e the differential needed between the wolframite and siderite and the other strongly magnetic minerals present. At higher fields, there was a substantial recovery of the magnetic silicates and consequently poor weight rejection. Gravity separation testwork centred on the Multi-Gr avity Separator with some work also being carried out using the Kelsey jig and Duplex separators. Results indicated that a three stage proce ss with intermediate sulphide flotation could potentially produce a wo lframite concentrate of 50% to 55% WO3, at a reasonable recovery. Conc entrate cleaning options were logistically very difficult to investiga te, one tonne of dam material only yielding a few kilogrammes of conce ntrate for testwork. In order to provide a thorough investigation, the limited testwork was backed up by extensive computer simulation. Comp uter simulation also played an integral part in developing the overall process flowsheet and in optimising this flowsheet to maximise recove ries.