|
A Technical Guide for Wine Producers
|
|
RECENT ARTICLES | WYNBOER HOME
Effect of ozone on distillery wastewater treatment in wetlands
Du Plessis, K.1, Green, J .2, Sigge, G.2, Brits, T. 2 and Wooldridge, J. 1
1 ARC Infruitec-Nietvoorbij, Stellenbosch
Key words: chemical oxygen demand, constructed wetlands, distillery wastewater, ozone, polyphenol.
ABSTRACT
The discharge of wastewaters from wineries and distilleries is becoming increasingly restricted as pressures from environmental regulations increase (Van Schoor, 2005), and as awareness of the negative impacts of seasonal discharges of waters containing high nutrient and organic loadings into water courses spreads (Masi et al., 2002). Disposal of such waters by irrigation, as investigated by Mulidzi (2001), is also regulated, notably by the National Water Act (No. 36 of 1998).
Of the currently availably wastewater disposal strategies, none are entirely satisfactory, cost usually being the most significant constraint. Constructed wetlands have nevertheless been found to be an effective way of treating wastewaters from a number of industries (Hammer & Bastian, 1989; Gale et al., 1993; IWA Specialist Group, 2000; Ibekwe et al., 2003; Du Plessis, 2007 and others). However, constructed wetlands are easily overwhelmed, and should not be supplied with winery wastewater having chemical oxygen demands (COD, a measure of the oxygen needed to decompose the organic material in the wastewater) above 5000 mg O2 / litre if the death of plants growing near the wetland inlet is to be avoided (Shepherd et al., 2001). Distillery wastewater has a complex character and is difficult to treat (Yeoh, 1997; Sangave & Pandit, 2004), even where volumes are low. A medium-sized ethanol distillery producing 10 million litres of ethanol per year may generate up to 20 litres of wastewater per litre of ethanol produced (Van Haandel & Catunda, 1994). This means that treatment facilities must be able to handle high rates of throughput whilst retaining their effectiveness. According to Mulidzi (2005), one way in which effectiveness can be improved is to pre-treat the wastewater before it enters the wetland, the wetland thus becoming a secondary, as opposed to the sole treatment facility. Of the possible pre-treatments for distillery effluents, ozone, a powerful oxidising agent capable of increasing rate of biodegradation, appears to have potential. Studies by Benitez et al. (1999) and Martín et al. (2002), in which ozone was applied to winery and distillery wastewaters prior to anaerobic digestion, decreased COD whilst simultaneously increasing the rate at which anaerobic digestion proceeded in both effluent types.
Figure 1. The experimental wetland systems at the start (left image) and end (right image) of Trial 3. In each image, the control wetland is on the left and the ozone pre-treated wetland on the right. Note the greater height of the Phragmytes spp. in the ozone pre-treated wetland.
The primary aim of the present study was to investigate the effects of treating distillery wastewater with ozone before it entered the wetland on the quality of the water after it drained from the wetland. A secondary aim was to determine whether further benefits could be derived by subjecting the water that drained from the wetland to a second ozone treatment.
Two laboratory-scale artificial wetlands were constructed in a greenhouse at Nietvoorbij Research Farm, Stellenbosch. Each wetland consisted of a 0.06 cubic metre rectangular Perspex tank, with a base that sloped downwards at a low angle (1%) toward an outlet. Each tank was filled to 0.1 m from the top with irregularly fragmented dolomitic limestone gravel that had passed through a 25 mm square mesh grid. Nine Phragmytes spp. wetland plants (macrophytes) (IWA Specialist Group, 2000; Zingelwa, 2003) were planted in a regular grid pattern in each tank. Each tank, together with its gravel and plants constituted a single wetland system similar to the vertical flow design described by the IWA Specialist Group (2000). Using this pair of tanks, three trials were carried out consecutively. The purpose of these trials was to enable the effects of the ozone treatments to be tested at different combinations of COD and retention time (RT: the time taken for the wastewater to move through the wetland).
Trial 1:
The objective of Trial 1 was to determine the effect of ozone pre-treatment where COD was low and RT reasonably short.
Before Trial 1 commenced, both wetlands were operated for an initial period of 27 days on raw wastewater with an initial (inflow) COD of 2200 mg O2 / litre. This wastewater was obtained in bulk from the final settlement dam at Distell’s Goudini distillery near Worcester. The purpose of this pre-trial period, during which new wastewater was pumped into the wetlands each day at a rate adjusted to allow a total retention time (RT) within the wetland of nine days, after which the wastewater drained under gravity into collector tanks, was to allow the wetlands to acclimatise (or stabilise). Thereafter, one randomly selected wetland was designated as the control, and the other as the pre-ozonated wetland. After the 27-day acclimatisation period the control continued to receive none ozone treated wastewater with a COD at the inflow of 2200 mg O2 / litre at an RT of nine days. In contrast, the ozone pre-treated wetland received wastewater (also with an inflow COD of 2200 mg O2 / litre at an RT of nine days) that had been pre-treated with ozone by passing the wastewater through a venturi system into which ozone from a 4.8 g / hour generator was introduced at a flow rate of 4 litres / minute. The ozonated wastewater was collected in plastic drums, frozen, stored and thawed out as required.
After the 72-day evaluation (trial) period, freshly drained samples of the control and ozone pre-treated wastewaters were analysed to determine COD, alkalinity, phosphate and polyphenol concentrations, as well as conductivity, solids and colour. Wastewater colour (absorbance) was measured at 254 nm and 475 nm. Absorbance at these wavelengths is indicative of the presence of the precursors of such toxic substances as humic acids and dopachrome. Alkalinity, total solids (TS) and total suspended solids (TSS) were monitored according to standard methods (APHA, 1998). Conductivity was measured with a standard conductivity meter, whilst COD and phosphorous (PO43-) were determined colourimetrically. Total polyphenol content was determined using the Folin-Ciocalteau method (Singleton & Rossi, 1965). Each analysis was repeated three times.
Trial 2:
The purpose of Trial 2 was to determine the effect of ozone pre-treatment on wetland performance at a higher COD, but the same RT, as in Trial 1. In addition, a post-wetland ozone treatment was added.
As in Trial 1 the wetlands were subjected to a 27-day acclimatisation period. During this period the COD of the wastewater increased step-wise from 3750 mg/L in days 1 to 9, to 5500 mg /L in days 10 to 18 and, finally, to 7100 mg/L in days 19 to 27. As in Trial 1, the wastewaters were collected in bulk from Distell’s Goudini distillery, though at different times, ozonated, and stored until required. Also as in Trial 1, the RT was nine days and the trial period 72 days. During this period the control and pre-ozonated wetlands received wastewater with an inflow COD of 7100 mg O2 / litre (3.23-fold the COD in Trial 1).
Figure 2. Effect of passage through the wetland, and of post-wetland ozonation on ozone pre-treated distillery wastewater. Left, wastewater prior to entry into the wetland. Centre, the same wastewater after it had passed through the wetland. Right, wastewater after it had passed through the wetland and undergone post-wetland ozonation.
Samples of the wastewater that drained from the control and ozone pre-treated wetlands on Day 72 were collected and divided into two. One set of subsamples were analysed as in Trial 1. The second set of subsamples received ozone at a concentration of 400 mg / L, then analysed.
Trial 3
Trial 3 was carried out to determine the effect of ozone pre-treatment on wetland performance at the same COD, but a longer RT, than in Trial 2. The trial procedure used in Trial 3 was the same as in Trial 2, excepting that the rate of inflow was reduced so that one pore volume was replaced over a 12-day period (as opposed to the nine day period used in Trials 1 and 2). Because of the extended RT, the acclimatisation period was increased to 36 days, to allow time for three changes of wastewater to take place, each taking 12 days.
The parameters used in the three trials are summarized in Table 1.
Table 1. Summary of trials and treatment parameters. Each trial consisted of an ozone pre-treated and a control (not ozone pre-treated) wetland treatment. In all cases acclimatization was carried out using wastewater that was not pre-treated with ozone.
In all three trials the wastewater COD decreased during its passage through the wetland (Table 2). Relative to the controls, ozone pre-treatment improved (reduced) COD by 11% in Trial 1, 6% in Trial 2 and 1% in Trial 3. Since it is likely that the ozone content of the wastewater declined rapidly during storage (ozone content was not measured at the wetland inflow), the observed improvements in post-wetland COD must have been due to oxidation which took place soon after the ozone was introduced. If this were the case then the ozone pre-treated wetland received wastewater in which the organic compounds had already been partly degraded. That the ozone-induced decrease in post wetland COD in Trial 2 was less than in Trial 1 probably indicated that the 9-day RT was too short to permit the Trial 2 wastewater, with its 323% greater inflow COD, to be processed with the same efficiency as in Trial 1. The highest percentage reductions in COD during wetland transit were observed in Trial 3. That the reduction in COD was greater in Trial 3 than in Trial 2, even though the inflow COD was the same in both trials, can only have been due to the 33% (3-day) longer RT used in Trial 3. Since ozone pre-treatment only resulted in a 1% improvement in COD over the control in Trial 3, it seems possibly that ozone pre-treatment can be dispensed with where 7100 mg O2 / litre COD wastewater is allowed a 12-day RT, although this possibility can only be applied with confidence to a wetland of the type used in this trial.
Table 2. Percentage change in wastewater characteristics due to passage through an artificial wetland, and the effects of withholding (control) or applying an ozone pre-treatment.
Wastewater phosphate concentrations also decreased during passage through the wetlands and, in all three trials, decreased to a greater extent in the ozone pre-treated than in the control treatments. Differences due to ozone pre-treatment decreased from 24% to 9% to 7% in trials 1, 2 and 3, respectively. Ozone pre-treatment improved polyphenol elimination by 9% in Trial 1, and by 4% in Trial 2, compared to the control. Despite the longer RT used in Trial 3, polyphenols were reduced less effectively in Trial 3 than in Trial 2. Total suspended solids concentrations in the three trials followed much the same pattern as COD, except that the effects of ozone pre-treatment were more marked (35%, 13% and 6%, respectively, in Trials 1, 2 and 3), also relative to the control.
In both the control and ozone pre-treatments, colour at 475 nm improved to a greater extent in Trial 3 than in Trial 2 which in turn showed a better improvement than that observed in Trial 1. The colour difference between control and ozone pre-treated wastewaters decreased from 24% in Trial 1 to 13% in Trial 2 to 7% in Trial 3, following much the same response pattern as was observed for COD, phosphates and polyphenols. Colour at 254 nm was less effectively reduced during passage through the wetland than at 475 nm, particularly in the control.
The effects of the treatments on TSS (that part of the TS that can be removed by filtration, as compared with the dissolved solids which pass through a filter, but appear as residue after a sample of wastewater has been slowly evaporated to dryness) was greater, and more consistent than were their effects on TS. Passage through the wetlands decreased TSS in all three trials, and in the control, as well as in the ozonated treatments. The extent to which TSS was lower in the ozone pre-treated than the control treatment was greatest in Trial 1 (35%), intermediate in Trial 2 (13%) and low in Trial 3 (6%).
The total (suspended + dissolved) solids (TS) content of the control wastewater was barely affected by passage through the wetland in Trial 1, though some reduction was observed in the ozone pre-treatment. In trials 2 and 3, in which the inflow COD was much higher than in Trial 1, limited reductions in TS were also observed in the control wastewaters following their passage through the wetlands. In contrast the effect of ozone pre-treatment in Trial 2 and Trial 3 was to bring about an increase in TS content. This increase was ascribed to partial breakdown of the organic residues coating the roots and gravel under the oxidizing influence of the ozone pre-treated wastewater, and release of the resultant fragments into the wastewater flow. That the treatments were generally able to bring about appreciable decreases in TSS, but only promoted relatively small decreases in TS, is explicable if it is assumed that the treatments converted solids from the suspended to the dissolved form, without bringing about their complete elimination from the TS category. This assumption is in line with the findings of Verma et al. (2007) who, while investigating the biodegradation of wastewater sludge, found that concentrations of dissolved solids increased linearly with total solids.
Conductivity (electrolyte or salt content) was not reduced by passage through the wetland in Trials 1 and 3 (the results for Trial 2 were inconclusive). Salt removal in wetlands is dependent on root uptake. The failure of the Phragmytes species to reduce the wastewater salt content in Trials 1 and 3 may either be due to inability on the part of the Phragmytes species concerned to remove the salts in the forms in which they were present in the wastewater, or to poor root function stemming from inhibited growth.
Alkalinity levels tended to increase as the wastewater passed through the wetland, notably in the control treatments. These increases were attributed to breakdown of the dolomitic limestone as the wastewater passed through the pores in the gravel.
During Trial 3 the plants in the wetland which received ozone pre-treated wastewater grew appreciably better than those in the control wetland (Figure 1), reaching average heights of 118 cm and 95 cm, respectively. Since both wetlands were subjected to similar conditions of temperature, light intensity and flow rate, the difference in growth can only be ascribed to neutralization by the ozone pre-treatment process of certain of the phyto-inhibitory (toxic) components of the wastewater and, perhaps, their conversion into more readily biodegradable forms. The positive effects of ozone pre-treatment were sufficient to over-ride or mask any negative effects on the biological system operating within the wetland that may have been caused by the presence of any residual ozone in the wastewater. Because wetlands rely heavily on their macro plant components (IWA Specialist Group, 2000; Zingelwa, 2003), pre-treatments that improves the health and growth potential of the wetland plants should lead to greater removal efficiencies in the long term.
In the control and ozone pre-treated wastewaters that passed through the wetlands in Trials 2 and 3, post wetland ozonation increased the effectiveness with which COD, polyphenols, TSS and colour were controlled, relative to their concentrations in the post wetland wastewaters that did not receive the supplementary ozone treatment. In contrast, post ozonation increased TS, probably by converting suspended solids into the dissolved component of the TS. Wastewater parameters after passage through the wetland and the post-ozonation treatment did not show a consistent response to the ozone status (ie. control or ozone pre-treated) of the wastewater that entered the wetland. With the exception of polyphenols and colour (Figure 2), the benefits conferred by post-ozonation were small. Thus, although post wetland ozonation contributes to wastewater quality improvement, it is not a substitute for ozone pre-treatment.
Table 3. Percentage change in wastewater parameters due to the application of a post-ozonation treatment to the control and ozone pre-treated wastewaters after they had passed through the wetlands in Trial 2 and Trial 3.
Conclusions
A laboratory scale wetland trial showed that pre-treatment of the wastewater with ozone may have the potential to improve the quality of the wastewater that drains from wetlands. Criteria that proved amenable to ozone pre-treatment were COD, polyphenols and TSS contents, and colour, notably at 475 nm. Retention time within the wetland system is a critical factor and must be adjusted to suite the COD of the wastewater flowing into the system, high COD wastewaters requiring longer RT’s than low COD wastewaters. Conceivably, if RT’s could be sufficiently extended, the benefits associated with ozone pre-treatment could decrease, possibly to the point where such pre-treatment may no longer be considered necessary. Where an additional, post-wetland ozone treatment was applied, further improvements in waste water quality were obtained. Although ozone may thus improve the effectiveness of distillery and, by inference, winery wastewater purification in artificial wetlands, the question of whether ozonation will be viable in full-scale commercial settings is likely to be one of cost effectiveness; cost effectiveness being dependent on the efficiency of the apparatus used. Since data from a laboratory scale trial can not be reliably extrapolated to predict what will happen in a full scale wetland, and since investigation of the methods by which the efficiency of the ozonation process could be determined lay outside the scope of this trial, it is recommended that a further study should be carried out. The aim of this new study should be to compare ozone application systems and application rates and determine how ozone use in conjunction with wetland systems can be rendered most cost effective.
For more information contact Keith Du Plessis on (021) 809 3158 or at duplessisk@arc.agric.za.
References
APHA, 1998 (20th ed). Standard methods for the examination of mater and wastewater. American Public Health Association, Washington DC, USA. pp. 4 - 106.
|
|
Wynboer is incorporated in WineLand, magazine of the SA wine producers.
Visit our sister sites:
|
| UP |
COPYRIGHT (C) 2000 WineLand |