Dissertation PhD--University of Stellenbosch, 2000. / ENGLISH ABSTRACT: Adsorption tests were conducted with four different ion-exchange resins to determine
the equilibrium adsorption of a range of precious and base metals. The adsorption
characteristics were determined for synthetic single metal, as well as for multicomponent
and base metal solutions. The effect of the el- concentration on the
equilibrium adsorption was established for three different Hel concentrations in the
above solutions. From the ion-exchange characteristics determined, a selective
adsorption sequence is proposed for the separation of precious and base metals.
Pure platinum, palladium and gold were dissolved in aqua regia and diluted to 2000
ppm (as metal) in 4M Hel. Ruthenium, rhodium and iridium were dissolved from
pure salts in Hel. A 2000 ppm base metal solution was prepared by dissolving all the
required components, including precious metals, to match an in-plant industrial basemetals
solution composition. For each precious metal the equilibrium adsorption was
determined for a couple of solution concentrations. Data points for adsorption curves
were established by varying the amount of resin added to the test solution of a specific
concentration. The equilibrium solution concentrations were determined by Iep
analysis after 24 hours of exposure, using the bottle-roll technique.
The experimental results obtained indicate a possible process route for the separation
of precious metals with ion-exchange resin. The XAD7 resin is highly selective for
gold from mixed solutions containing precious and base metals. It is also evident that,
with the gold removed from the solution, the A22 resin adsorbs only palladium.
IR200 resin adsorbs only the base metals from the solution. With all other precious
metals removed from the solution (platinum and ruthenium must be extracted by other
means), iridium can be adsorbed from the solution by IRA900 resin which is highly
selective for iridium over rhodium. For all of the anion resins, XAD7, IRA900, and
A22, the chloride concentration of the solution did not have a big effect on the
adsorption capacity. However, the adsorption of base metals on IR200 is sensitive to
chloride concentration, with a rapid reduction in adsorption at higher chloride
concentrations. Statistical models were developed for the adsorption of each of the precious metals,
as well as for the base metal solution. All adsorption data, obtained for a resin
(typically 250 equilibrium data points), was used in the development of the model.
The SPSS statistical software package was used to develop linear regression models.
The interaction between all the input parameters, e.g. the interaction of gold and
chloride ions, was modelled by specifying the product of the gold and the chloride
concentrations as an input variable. The variables that determine the adsorption
quantities were identified.
High solution concentrations of the target adsorption component increase the
adsorption quantity. It has been established that a higher platinum concentration
increases the adsorption quantity of gold on XAD7 resin. However, the adsorption
quantity is reduced at higher ruthenium concentrations. The adsorption quantity of
iridium on IRA900 is reduced with increased rhodium concentration. The adsorption
quantity of palladium on A22 is increased by the presence of rhodium and decreased
by larger concentrations of iridium and platinum. The adsorption of base metals on
IR200 is decreased at higher acid concentrations. Higher concentrations of gold in the
base metal solution also decrease the adsorption quantity of base metals. The model
predicted adsorption of each component compares well with the actual measured
values.
In batch adsorption tests the counter ion is not removed from the resin. The resin
capacity for a specific ion concentration could therefore not be determined. As such,
the adsorption models are only valid for the initial part of the ion-exchange process.
The effect of kinetics on the adsorption was not determined. / AFRIKAANSE OPSOMMING: Adsorpsietoetse is gedoen met vier verskillende ioonuitruilharse om die ewewig
adsorpsie van edelmetale en basismetale te bepaal. Die adsorpsie karakteristieke is
bepaal vir sintetiese enkelmetaal-, meermetaal-, en basismetaaloplossings. Die effek
van die Cl konsentrasie op die ewewigadsorpsie is bepaal vir drie
soutsuurkonsentrasies in al die bogenoemde oplossings. 'n Prosesvloeidiagram vir die
selektiewe adsorpsie van edelmetale en basismetale met behulp van die vier
verskillende ioonuitruilharse word voorgestel.
Suiwer platinum, palladium en goud is opgelos in koningswater en verdun na 2000
dpm (uitgedruk as metaal) in 4 molaar HCl. Rutenium-, rodium- en
iridiumoplossings is verkry deur die oplos van suiwer edelmetaalsoute in HCl. 'n
Basismetaaloplossing van 2000 dpm is voorberei deur die individuele komponente,
wat die edelmetale ingesluit het, in die regte verhouding te kombineer om die
samestelling van 'n industriële basismetaaloplossing te verkry. Vir elke edelmetaal is
die ewewigadsorpsie bepaal vir tipies twee konsentrasies van edelmetaaloplossings.
Verskillende data punte by 'n spesifieke edelmetaalkonsentrasie is bepaal deur die
hoeveelheid hars wat by die toets oplossing gevoeg word te varieer. Die
oplossingkonsentrasies by ewewig is bepaal deur IGP analise na 24 uur blootstelling
met die roterende botteltegniek.
Die resultate wat verkry is dui op 'n moontlike ioonuitruiigebaseerde proses vir die
skeiding van edelmetale. Die XAD hars is selektief vir goudadsorpsie uit gemengde
oplossings wat al die edelmetale, asook basismetale bevat. Dit is ook bevind dat
indien goud uit die gemengde oplossing verwyder word, die A22 hars slegs
palladium adsorbeer uit die oplossing. Die IR200 hars adsorbeer slegs basismetale.
Wanneer al die ander edelmetale uit die oplossing verwyder is (platinum en rutenium
moet met alternatiewe prosesse verwyder word) kan iridium geadsorbeer word met
!RA900 hars vanuit 'n iridium- en rodium- gemengde oplossing. Vir al drie die
anioonharse wat getoets is, naamlik, XAD7, !RA900 en A22, het die
chloriedkonsentrasie nie 'n groot effek op die adsorpsie gehad nie. Die hoeveelheid basismetale wat op IR200 hars geadsorbeer word is egter baie sensitief vir die
chloriedkonsentrasie, met 'n vinmge afname In adsorpsie by hoër
chloriedkonsentrasies.
Statistiese modelle is ontwikkel vir elke hars en vir elk van die edelmetale, asook die
basismetaaloplossing. Al die eksperimentele data wat vir elke hars verkry is, tipies
250 ewewigspunte, is gebruik in die ontwikkeling van lineêre regressie modelle vir
die primêre absorberende spesie op die hars.
Hoër konsentrasies van die teiken adsorpsie komponent verhoog die adsorpsie
daarvan. Hoër platinumkonsentrasies verhoog die adsorpsie van goud op XAD7,
maar die teenwoordigheid van rutenium verlaag adsorpsie. Die adsorpsie van iridium
op IRA900 word verlaag met hoër rodiumkonsentrasies. Die adsorpsie van palladium
op A22 verhoog met die teenwoordigheid van rodium, maar neem af met hoër
konsentrasies van iridium en platinum. Die adsorpsie van basismetale op IR200 neem
af by hoër suurkonsentrasies. 'n Hoër goudkonsentrasie verlaag ook die adsorpsie van
basismetale. Die gemodelleerde adsorpsie hoeveelhede vergelyk goed met die gemete
waardes.
Aangesien die uitgeruilde ioon nie verwyder word uit die toetsoplossing nie, kon die
harskapasiteit vir 'n spesifieke ioonkonsentrasie nie bepaal word nie. Die modelle is
derhalwe slegs getoets vir die aanvanklike deel van die ewewigsdata en die kinetika
van adsorpsie is nie in ag geneem nie.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/51936 |
| Date | 12 1900 |
| Creators | Els, Ellis Raymond |
| Contributors | Lorenzen, L., Aldrich, C., Stellenbosch University. Faculty of Engineering. Dept. of Process Engineering. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
| Format | 413 p. : ill. |
| Rights | Stellenbosch University |
Page generated in 0.0032 seconds