Fast contact copper extraction
Tamminen, Jussi (2022-12-19)
Väitöskirja
Tamminen, Jussi
19.12.2022
Lappeenranta-Lahti University of Technology LUT
Acta Universitatis Lappeenrantaensis
School of Engineering Science
School of Engineering Science, Kemiantekniikka
Kaikki oikeudet pidätetään.
Julkaisun pysyvä osoite on
https://urn.fi/URN:ISBN:978-952-335-911-6
https://urn.fi/URN:ISBN:978-952-335-911-6
Tiivistelmä
The intensification of copper solvent extraction was examined in this thesis. Extraction with hydroxyoximes is used in industry for copper separation. The reaction between hydroxyoxime and copper is a two-phase interfacial complexation reaction in which aqueous copper cations exchange with hydrogen ions bound to hydroxyoxime. As extraction is an interfacial reaction, its rate is dependent on the area between phases, that is, an interfacial area (A). Moreover, the rate depends on the stagnant interfacial boundary layer thickness, or diffusion path length (l di f). These variables depend on droplet size, which in turn depends on the mixing conditions in the reactor. Intensified mixing will lead to intensified extraction.
Copper extraction kinetic studies were conducted in different contactors. The mass transfer into a single droplet was examined using a concentration determination method based on image analysis. The image analysis method was also used to determine mass transfer into individual droplets during breakage and coalescence experiments. The mass transfer was found to increase during droplet breakage but not during coalescence. The increased mass transfer in breakage was due to an increase in interfacial area and mixing, as the breakage of rising droplet after collision with a blade is a more violent process than coalescence. The coalescing droplets were stationary, and the interfacial area decreased, which led to a constant mass transfer during coalescence.
The other devices used in the studies were a conventional stirred tank and rotor-stator devices with intensified mixing. The rotor-stator devices were used in both batch and continuous-flow reactors. The main difference between these reactors and earlier intensified extractors, such as AKUFVE1 or high speed stirring in Morton flask, is the use of a rotor-stator mixer, which leads to intensified mixing in a smaller equipment volume. The kinetic data were modeled using a reactor models developed for copper extraction. The extraction is a two-phase reaction; its equilibrium is dependent on conditions in both phases. This is indicated by a decrease in the extraction equilibrium constant as a function of aqueous phase ionic strength. In a similar manner, the increase in diluent solubility parameter leads to a decrease in equilibrium constant.
The kinetic constant of copper extraction increases as a function of mixing intensity. In order to characterize mixing conditions in extraction, the droplet sizes and mixing power were measured. Measurements were made in conventional stirred tank and rotor-stator mixed continuous flow reactor and the droplet sizes were correlated with mixing power. The starting point was a single droplet extraction without mechanical stirring, which naturally yielded the lowest kinetic constant value. As expected the rate increased in a conventional stirred tank as the impeller speed increased, whereas other variables, such as feed concentrations were kept constant. The increase in impeller speed enhanced extraction both in batch and in continuous-flow rotor-stator mixers. Short residence times are required for extraction in rotor-stator reactors because of high mixing intensities. The kinetic reaction constant (k) data of all stirred reactors were dependent on specific mixing power input (P/m) to exponent 0.625. The specific mixing power input was varied in three orders of magnitude when using LIX 984 extractant. The mixing intensity was varied here in a much wider range than in typical mass transfer measurement studies. A comparison of data with other copper extractants revealed that the correlation had reasonable agreement with the kinetic data over five orders of magnitude of P/m.
The kinetic constants (k) determined in different devices (single droplet, stirred tank, and rotor-stator reactors) had a good correlation with A2/l di f. The interfacial area and diffusion path length (l di f) depend on droplet size and, on mixing power. Area and diffusion path length cannot vary independently in stirred reactors. Kinetic constant dependence on interfacial area and diffusion path length illustrates the interfacial nature of solvent extraction.
Copper extraction kinetic studies were conducted in different contactors. The mass transfer into a single droplet was examined using a concentration determination method based on image analysis. The image analysis method was also used to determine mass transfer into individual droplets during breakage and coalescence experiments. The mass transfer was found to increase during droplet breakage but not during coalescence. The increased mass transfer in breakage was due to an increase in interfacial area and mixing, as the breakage of rising droplet after collision with a blade is a more violent process than coalescence. The coalescing droplets were stationary, and the interfacial area decreased, which led to a constant mass transfer during coalescence.
The other devices used in the studies were a conventional stirred tank and rotor-stator devices with intensified mixing. The rotor-stator devices were used in both batch and continuous-flow reactors. The main difference between these reactors and earlier intensified extractors, such as AKUFVE1 or high speed stirring in Morton flask, is the use of a rotor-stator mixer, which leads to intensified mixing in a smaller equipment volume. The kinetic data were modeled using a reactor models developed for copper extraction. The extraction is a two-phase reaction; its equilibrium is dependent on conditions in both phases. This is indicated by a decrease in the extraction equilibrium constant as a function of aqueous phase ionic strength. In a similar manner, the increase in diluent solubility parameter leads to a decrease in equilibrium constant.
The kinetic constant of copper extraction increases as a function of mixing intensity. In order to characterize mixing conditions in extraction, the droplet sizes and mixing power were measured. Measurements were made in conventional stirred tank and rotor-stator mixed continuous flow reactor and the droplet sizes were correlated with mixing power. The starting point was a single droplet extraction without mechanical stirring, which naturally yielded the lowest kinetic constant value. As expected the rate increased in a conventional stirred tank as the impeller speed increased, whereas other variables, such as feed concentrations were kept constant. The increase in impeller speed enhanced extraction both in batch and in continuous-flow rotor-stator mixers. Short residence times are required for extraction in rotor-stator reactors because of high mixing intensities. The kinetic reaction constant (k) data of all stirred reactors were dependent on specific mixing power input (P/m) to exponent 0.625. The specific mixing power input was varied in three orders of magnitude when using LIX 984 extractant. The mixing intensity was varied here in a much wider range than in typical mass transfer measurement studies. A comparison of data with other copper extractants revealed that the correlation had reasonable agreement with the kinetic data over five orders of magnitude of P/m.
The kinetic constants (k) determined in different devices (single droplet, stirred tank, and rotor-stator reactors) had a good correlation with A2/l di f. The interfacial area and diffusion path length (l di f) depend on droplet size and, on mixing power. Area and diffusion path length cannot vary independently in stirred reactors. Kinetic constant dependence on interfacial area and diffusion path length illustrates the interfacial nature of solvent extraction.
Kokoelmat
- Väitöskirjat [1099]