P.K. Holt, G.W. Barton and C.A. Mitchell* Department of Chemical Engineering, University of Sydney, NSW 2006, Australia (E-mail:
[email protected]) * Institute for Sustainable Futures, UTS, PO Box 123, Broadway, NSW 2007, Australia Abstract Electrocoagulation removes pollutant material from water by a combination of coagulant delivered from a sacrificial aluminium anode and hydrogen bubbles evolved at an inert cathode. Rates of clay particle flotation and settling were experimentally determined in a 7 L batch reactor over a range of currents (0.25–2.0 A) and pollutant loadings (0.1–1.7 g/L). Sedimentation and flotation are the dominant removal mechanism at low and high currents, respectively. This shift in separation mode can be explained by analysing the reactor in terms of a published dissolved air flotation model. Keywords Batch electrocoagulation; DAF model; water treatment
Introduction
Electrocoagulation has long been accepted as an ideal technology to upgrade water quality and has been successfully applied to a wide range of pollutants in an even wider range of reactor designs (Vik et al., 1984). However, while there is agreement about the technical viability, the interactions between the various underlying mechanisms are less well understood, with the result that electrocoagulation must still be categorised as essentially empirical and heuristic. The batch process for a clay pollutant is summarised in Figure 1. Coagulant is generated from a sacrificial aluminium anode while hydrogen is simultaneously evolved at the inert cathode. The aluminium cation hydrolyses, with dynamic changes in the aqueous speciation of aluminium dictating both the coagulant’s availability and physical form. Initially the aluminium cations contribute to charge neutralisation of the pollutant particles as the isoelectric point is attained. Here a sorption coagulation mechanism occurs resulting in the formation of loose aggregates. As time progresses, further aluminium cation addition results in amorphous aluminium hydroxide precipitation that contributes to pollutant aggregation via an enmeshment mechanism (sweep coagulation) (Holt et al., 2002). The final stage is pollutant removal – where the coagulated aggregates interact with bubbles and are floated to the surface or settle to the bottom of the reactor. This paper focuses on the separation phase for a batch reactor removing clay particles from water. The paper aims to explain a curious shift between flotation and sedimentation pollutant removal that was observed to occur over a quite modest applied current range.
Water Science and Technology Vol 50 No 12 pp 177–184 © IWA Publishing 2004
Deciphering the science behind electrocoagulation to remove suspended clay particles from water
Experimental methods
Electrocoagulation runs were carried out using a 7 L Perspex batch reactor as described previously by the authors (Holt et al., 2002). For each run a dynamic mass balance was conducted over the reactor allowing the construction of a pollutant concentration profile as a function of time. Turbidity, zeta potential and pH were measured off-line using a Merck Turbiquant 1500T (tungsten lamp), Malvern Zetasizer and a calibrated pH meter, respectively. All experiments were conducted at ambient temperature (nominally 20°C).
177
Electrochemistry
Coagulation
Flotation
P.K. Holt et al.
Aluminium
Stable floc
Flotation (k1) Coagulant and hydrogen generation
Mixing, contact patterns
Aqueous speciation & hydrolysis
Al3+
(Al3+)
Settling (k2) Anode dissolution and H2 generation Diffusion limited
Coagulation Aggregation Reaction limited Sweep coagulation mechanism aggregation
Sludge Pollutant removal Flotation
i
t
t
i
Reaction limited (charge transfer)
Diffusion limited aggregation
Sorption coagulation
Settling
INHERENTLY INTERCONNECTED
f(I,t)
Figure 1 Mechanistic summary of electrocoagulation
The electrode arrangement consisted of five stainless steel cathodes interspersed with four aluminium anodes, with brass rods used to connect the parallel plate electrodes. The current was held constant for each run. Runs investigated the effect of current over the range 0.25 to 2.0A and initial pollutant loading from 0.1 to 1.7 g/L. To minimise solution resistance, conductivity was increased using sodium chloride (0.20 g/L). The “pollutant” used was a commercial potter’s clay, comprising kaolinite (67%), quartz (25%), illite/mica (3%), feldspar (3%) and other trace components (2%), as characterised by X-ray diffraction analysis. Clay density was measured at 1.8 g cm–3 on a dry basis while the particle size distribution (PSD) was measured using a Malvern Mastersizer S with coarser particles (above 53 µm) separated using sieves. Experimental results Pollutant characterisation
The mass fraction at each sieve size is summarised in Table 1. The bulk of the mass (69%) exists as silt and clay (particles smaller than 53 µm). The size distribution for particles smaller than 53 µm is shown in Figure 2. For this size fraction, the d50 was 2.5 µm. Dynamic mass balance results
The cumulative mass collected at the surface as a percentage of total mass removed for each current is shown in Figure 3 for a series of runs at an initial pollutant (clay) loading of 0.8 g/L. Table 1 Sieve fraction greater than 53 µm Size fraction
178
>1.00 mm 710 µm–1.0 mm 500–710 µm 212–500 µm 53–212 µm