The University of Southampton
University of Southampton Institutional Repository

# The adsorptive bubble separation of trace metals from sea water

Espey, Quentin Ian (1989) The adsorptive bubble separation of trace metals from sea water. University of Southampton, Doctoral Thesis.

Record type: Thesis (Doctoral)

## Abstract

An adsubble separation column has been developed to investigate dissolved metals which could be extracted from sea water in situ. A theoretical model has been written to describe the removal, by batch foam fractionation, of surfactants (Sf) and colligands from sea water. The nuclides: 110mAg, 195Au, 51Cr, 59Fe, 54Mn, 109Cd, 65Zn, 57Co and 22Na were added to filtered 10m slope sea water. The Sf used were mainly cetyl and dodecyl trimethylammonium bromide (CTMAB and DTMAB) and sodium dodecyl sulphate (NaDS). The role of DOC and of up to 1 mmolℓ-1 added complexant, ethylenediamine tetra acetic acid (EDTA) were examined. Kinetic removal experiments were conducted at 0.6 ℓ min^-1 bubble rate (BR) for CTMAB and at 1 ℓmin-1 for NaDS and DTMAB with maximum concentrations of 169, 137 and 411 μmol ℓ-1. Standard conditions were a bubble path length (BPL) of 120 cm, a drainage length (DL) of 22 cm and a mean bubble diameter (D) of ca.650 μm. Sf removal kinetics were first order with Sf distribution coefficients (Kd) of 38, 18 and 20 (max) respectively. At low and high [Sf] to surface area ratios, and at low BPL, second and zero order removal, and diffusion effects occur respectively. The metal species recovered in the standard experiments (excluding DOC complexed Fe), were all by Sf attachment, then surface exchange of metal and counterion, with first order removal kinetics. Changes in physical parameters (high BR, DL and small D and BPL) lead to adsorbing colloid flotation. Surface exchange coefficients (Ke) were determined for some of the metal species and Sf. In these cases it is possible to predict the concentration at which 99% removal will occur. Ag was present as inorganic chlorides (recovered as Ag+DS- and AgCl2-CTMA+) and attached via natural dry' surfactants to organic colloids, also recovered with both Sf. The highest Ke value for Ag tracer, of 93000 was for Ag+DS- which underestimates that for natural Ag due to adsorption during storage. EDTA affects Ag recovery. The highest Ag tracer recovery percentage (R%) was 85% and the highest enrichment (E) was 302 (R% of 79%). Au was present as Au(I) chlorides and Au(III) associated with hydrophilic acids (ha). Au tracer was recovered as Au+ DS-, AuCl2-CTMA+, Au(OH)3ha-CTMA+ (Ke of 38600), as well as [AuOH2-EDTA]-CTMA+ with a combined Ke value for the last two of 116000. The highest R% for Au tracer was 94% and the highest E value, 346 (R% of 90%), the experimental limit. Recovery of natural Au will be a function of oxidation state distribution. Cr was present as Cr(VI) species and Cr(III) associated with ha. Cr tracer was recovered as CrO42- CTMA+, (Ke of 4900), forming particulates. Cr(III) could only be recovered slightly as [CrOH2 -EDTA]-DTMA+. A maximum recovery of 28% for Cr tracer was obtained with complete Cr(VI) recovery. The predominance of Cr(VI) in the sea will lead to higher Cr recoveries. Fe(III) was present as 65% hydroxy species, 27% oxyhydroxide colloids and 8% organic complexes. All were recovered with CTMAB. The minimum Ke (Fe colloid) was 17000 and the apparent' Ke (complexed Fe) was 4000. The excess hydroxy form leads to a maximum recovery of only 45%. Selectivity of Ag (relative to Fe) can be increased by a factor of 640 by variation of [Sf], [EDTA] and physical parameters. Mn was present as 68% mixed Mn(II/IV) colloids and 32% simple cations with a R% max for Mn of 9% (all colloids) with CTMAB. This increased to 40% by near complete recovery of Mn(II) as particulate [Mn-EDTA]-DTMA+ (Ke of 12000). For all metals R% is increased with longer chain Sf and selectively can be altered by variation of Sf charge and head group. Cd can be recovered with dodecylamine. Zn, Co, Rb and Li show no reaction. A classification scheme that includes the extent of hydrolysis, polarizing power and covalent character, and the metal R% tendencies suggest that the best metals for in situ extraction are fully hydrolysed ones (e.g. Au(III), Cr(VI), U(VI), Mo, V); and chloride complexed ones including Ag, Au(I), Hg and Pt. A prototype in situ extraction scheme has been proposed. These metals (and Fe) will take part in bubble scavenging reactions in the sea leading to transfer of metals from sea to air. For Ag, Au and Cr this process probably contributes substantially to atmospheric input. Both these categories, plus Fe, and Mn(II)-type metals (complexed by EDTA) could also be adsubble preconcentrated prior to analysis.

Published date: 1989

## Identifiers

Local EPrints ID: 461337
URI: http://eprints.soton.ac.uk/id/eprint/461337

## Catalogue record

Date deposited: 04 Jul 2022 18:43

## Contributors

Author: Quentin Ian Espey