File(s) under permanent embargo

Reason: Restricted by author. A copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library or by emailing

The geologic relevance of the liquid bismuth collector model at Stormont, TAS and Warrego, Tennant Creek, NT

posted on 14.02.2017, 00:55 by Cockerton, Amy Beth Donovan
The liquid bismuth collector model was first proposed as a mechanism by which bismuth in a molten state could remove gold from a hydrothermal fluid. This method was suggested to potentially explain the close correlation often observed between gold and bismuth. In this thesis, two different gold- and bismuth-rich deposits are studied to determine whether scavenging of gold by liquid bismuth could have played a role in the mineralisation of each deposit. The Stormont prospect in central-northern Tasmania has an ore mineral assemblage of native bismuth, bismuthinite, maldonite, galena, native gold and an unknown Bi-Te-S mineral hosted in a calc-silicate assemblage of grandite, hedenbergite, actinolite, magnetite, epidote, quartz, and calcite. Conditions of mineralisation at Stormont included low sulfur fugacity and mineralisation temperatures of >400-500°C. Here, native bismuth is closely spatially and geochemically related to gold, and high gold content is found almost exclusively with high bismuth content. Thus, it is concluded that the liquid bismuth collector model, played a role in the formation of this deposit. As the liquid bismuth collector model can scavenge gold from fluids of very dilute gold concentration, the model can produce a zone refinement of gold from interaction with different fluid fluxes of variable gold concentration with liquid bismuth droplets. The conditions of mineralisation at the Warrego deposit were found to contrast with those at Stormont. The ore mineral assemblage observed here includes chalcopyrite, bismuthinite and gold, amongst a gangue assemblage of magnetite, hematite, chlorite, pyrite, muscovite and quartz. Despite the moderate geochemical correlation between gold and bismuth, spatially and texturally there is no more association between them than with other ore and gangue minerals. Geochemical analysis revealed high gold concentrations with large variations in bismuth content. No native bismuth was found at Warrego, where bismuth exists mostly in bismuthinite. Bismuthinite is also associated with other sulfides such as chalcopyrite and pyrite in fractures within the host magnetite ironstone lode. Gold is associated with all ore and many gangue minerals. An estimated mineralisation temperature of -150° to 350°C is interpreted, based on fluid inclusion homogenisation temperatures. Precipitation was not likely temperature dependant. Thus, it is concluded that the higher sulfur content, and lower temperature conditions meant that liquid bismuth scavenging did not occur in the Warrego mineralising system. The Stormont prospect demonstrates that the liquid bismuth collector model is relevant in many ore-forming environments. The similarities and differences between Warrego and Stormont are important for outlining both ideal and detrimental conditions for liquid bismuth formation. Once formed, liquid bismuth will scavenge gold; however the conditions in which it forms must be initially primed and then maintained for effective gold scavenging. Important hydrothermal fluid characteristics include a relatively reduced environment, with low sulfur activity and temperatures above 271°C, which may be lowered to 241°C if gold is also present when native bismuth precipitates. This model may play an important part in either primary or secondary enrichment of gold, as unlike other deposition methods, gold can be severely under-saturated in the fluid, and yet still be concentrated by liquid bismuth.


Campus location


Principal supervisor

Andrew Tomkins

Year of Award


Department, School or Centre

Earth, Atmosphere and Environment

Degree Type



Faculty of Science