Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Treatment of spent wash water derived from shredded lettuce processing using a combination of electrocoagulation and germicidal ultraviolet light

Treatment of spent wash water derived from shredded lettuce processing using a combination of... Objective: Water recycling is a significant part of an overall water management system. The current study evaluated electrocoagulation, used in combination with ultraviolet light (at 254 nm), to reduce the organic content and enhance the microbiological quality, of wash water derived from shredded lettuce processing. Method: The composition of spent wash water derived from a commercial lettuce processing operation was used to prepare a simulated solution to be applied to validate the water recycling system. The simulated spent wash water was subjected to an electrocoagulation process followed by filtration and a tertiary ultraviolet (254 nm) treatment. The efficacy of the recycling treatment to decrease turbidity (nephelometric turbidity units, biological oxygen demand (BOD), chemical oxygen demand (COD) and decrease in introduced bacterial numbers. Results: Spent wash water sampled from a commercial processing line was found to be colloidal in nature (78 ± 26 NTU) with low total solids content (544 ± 87 mg/L), BOD (230 ± 53 mg/L) and COD (309 ± 53 mg/L). An electrocogaultion process performed for 10 min using 3.48 A/m current density at pH 6.5 and conductivity of >100 µS/cm supported an 87% removal of turbidity, 38% reduction in BOD along with 49% decrease in COD. The electrocoagulation process was also found to reduce the levels of Escherichia coli, Salmonella and Listeria monocytogenes by 1–2 log cfu. The tertiary UV treatment of water derived from the electro coagulation process, supported further reduction in model pathogens, although it was noted that the D values for inactivation were in the order of 2 2 1.01–1.60 mJ/cm , which compares to 0.22–0.31 mJ/cm in saline. The apparent increase in bacterial resistance to ultraviolet was likely due to the UV absorbing low molecular weight constituents within wash water that provided protection against inactivation. Conclusion: In conclusion, the study demonstrated the feasibility of applying electrocoagulation and UV to rapidly treat spent lettuce wash water to facilitate in-process recycling within shredded lettuce processing operations. Key words: Biological oxygen demand (BOD); Chemical oxygen demand (COD); Electrocoagulation; Fresh produce; Lettuce; UV; Washing; Water recycling. © The Author 2017. Published by Oxford University Press on behalf of Zhejiang University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com 148 K. Khalid D. Alharbi et al., 2017, Vol. 1, No. 2 ions. In the course of electrolysis, dissolution of the plates occurs gen- Introduction erating aluminium hydroxide, or polymers, that act to coagulate solids Ever since the introduction of processed leafy greens in the late 1980s that are raised to the surface via the generation of hydrogen gas at the to the produce sector, the demand for bagged salads has grown at a cathode (Butler et  al., 2011). The direct electro-oxidation of water rate of 10% each year and now represents a $51.4 billion industry in constituents reduce the COD and BOD although reductions in micro- North America alone (Warriner et al., 2009). One of the largest sec- bial levels by the process have not been previously described. With tors in the fresh produce market continues to be lettuce-based salads regards to the latter, tertiary treatment is applied that can include the (Farragher et al., 2016). In preparing the product, heads of lettuce are addition of chemical sanitizers or physical methods such as UV-C ger- passed through a shredding unit, and then through an initial wash to micidal lamps (Manzocco et al., 2015). The benefits of UV-C are that remove soils followed by a biocidal wash to reduce microbial load- the process can support the inactivation of microbes, in addition to the ing (Warriner et al., 2009). Hypochlorite is the mostly used sanitizing photo-oxidation of reactive species within wash water. agent partly due to low cost, and also because of historical reasons In the following, the application of electrocoagulation in com- along with ease of monitoring using oxidation reduction potential bination with UV to treat spent wash water derived from shredded (ORP)-based systems (Barrera et al., 2012). Through various studies it lettuce processing has been evaluated. The work encompassed for- has been clearly demonstrated that the efficacy of washing for remov- mulating a simulation matrix based on characterizing spent wash ing pathogens, such as Escherichia coli O157:H7, Salmonella, and water derived from a commercial processing line. The simulation Listeria monocytogenes, is limited to 1–2 log cfu reductions (Aixia spent water was used to assess the efficacy of electrocoagulation to and Micallef, 2015). Of more concern is the potential for pathogens remove turbidity, solids content, BOD, COD, and microbial loading. to be redistributed between produce batches via cross-contamination The inactivation kinetics of model bacteria by UV in treated spent events (Allende et al., 2008; Lopez-Galvez et al., 2010). To this end, it wash water were then determined. has been found that maintaining free chlorine levels within the wash tank is a critical factor in minimizing cross-contamination events (Luo et al., 2011). However, even with wash processes using ORP feedback Experimental systems, the ability to maintain free chlorine levels is compromised by Spent leafy green water characterization and the accumulation of organics within the wash water that react with preparation of simulated spent irrigation water hypochlorite to form disinfection by-products that essentially seques- Spent leafy green wash water was collected from a commercial pro- ters antimicrobial action (Luo et al., 2011). This is especially relevant cessing facility on three separate occasions and characterized with with lettuce that contains high concentrations of latex and antioxi- respect to turbidity, solids content, BOD, and COD (Table  1). The dants, along with other constituents that could potentially interact processing line processed 2300 kg/h of shredded iceberg lettuce over with chlorine (Sessa et al., 2000; ShihChi et al., 2016). To counter the a period of up to 30 h. Sanitizer (50–200 ppm sodium hypochlorite) accumulation of organic matter within the wash tank, it is possible was automatically added by using a dual pH–ORP control system, to simply add increasing concentrations of hypochlorite to meet the the level of chlorine and pH in water were maintained in the range chlorine demand (ShihChi et  al., 2016). However, this approach is of 600–850 mV and pH 6.5 – 6.9, respectively. The wash tanks were unfeasible due to the generation of high levels of disinfection by-prod- replenished with fresh water at a rate of 20–30% per hour. ucts that affect flavour and odour taints on the product. A more com- To ensure consistency between experiments, a simulated spent wash mon approach is to reintroduce or recharge the wash tanks with fresh water was prepared based on the composition of commercial wash water at a rate of 10%–50% depending on the processor (Warriner water samples (Table 1). Specifically, the simulated water was prepared and Namvar, 2014). However, recharging tanks significantly increases by blending iceberg lettuce leaves (6.6  g/l) with 0.1  g/l bentonite in the usage of water and contributes to the estimated 40 l of water to distilled water. Large particulates were removed from the homogenate produce 1 kg of lettuce. Moreover, the larger volume of water usage by passing through a 20 µm pore size cellulose filter. The conductivity increases the amount of wastewater that requires treatment, which of the filtrate was adjusted between 100 and 1000 µ S/cm using NaCl. further elevates costs (Manzocco et al., 2015). A potential approach to reduce water usage is through recycling of spent water back into the process following removal of solids along Analytical techniques with reduction in the biological oxygen demand (BOD) and chemical COD of water samples were determined using HACH DBR 200 oxygen demand (COD) (Manzocco et  al., 2015). Technologies that Reactor (Hach Co., Loveland, CO, USA) for digestion and HACH recycle water can be based on a combination of physical, biological, DR 2800 (Hach Co., Loveland, CO, USA) for colorimetric determina- and chemical steps (Tee et al., 2016). Although biological treatments tion method according to Standard Method 5220D (APHA, AWWA, are highly efficient at reducing the BOD and COD, the process is time consuming and adds to the footprint along with processing costs Table 1. Characteristics of spent wash water derived from a shred- ded lettuce processing facility. Also shown are the characteristics (Bouallagui et  al., 2005). Therefore, physical and chemical methods of a simulated spent shredded lettuce wash water prepared from a are preferred for a continuous or semi-continuous process in which blend of lettuce leaves, bentonite, and NaCl. the water can be returned into the processing line. Here, a coagula- tion step is performed to aggregate the solids to facilitate filtration, Parameter Commercial spent Simulation spent although residual coagulant agent such as alum can be a problem due wash water wash water to toxic residue (Warriner and Namvar, 2014). Electrocoagulation has Turbidity (NTU) 78 ± 26 50 ± 13 been applied to treat a range of wastewater types from domestic sew- pH 6.9 6.9 age through to water derived from winery operations although it has Total solids (mg/l) 544 ± 87 560 ± 48 not been evaluated for treating spent leafy green wash water (Butler Total soluble solids (mg/l) 104 ± 98 55 ± 12 et  al., 2011). The electrocoagulation chamber typically consists of BOD (mg/l) 230 ± 53 164 ± 105 sacrificial aluminium plate electrodes that are polarized at a defined COD (mg/l) 309 ± 67 285 ± 92 voltage. Oxidation reactions occur at the anode that can lead to the Conductivity (µS/cm) 785 ± 102 100–1000 production of hypochlorite depending on the concentration of chloride Treatment of spent wash water, 2017, Vol. 1, No. 2 149 WEF, 1989). BOD was determined in accordance with method 5210 VI ⋅⋅ t Energy consumption = (2) (APHA, AWWA, WEF, 1989). Total soluble solids (TSS) was quanti- fied by using filtration (1.5  µm pore size glass microfibre filter; Cole Parmer, Montreal, Canada) followed by drying of the retentate at where V is the voltage, I is current (A), t is the run time (in s), and 105°C in accordance with method 2540D (APHA, AWWA, WEF, ν is the volume of liquid treated (m ) (Chopra and Sharma, 2014). 1989). Total solids (TS) was determined using method 5210 (APHA, The energy consumption was calculated in Watts (W) and the final AWWA, WEF, 1989). Turbidity was measured by a turbidity meter amount is reported in kW/h/m (Chopra and Sharma, 2014). (Oakton T-100; Cole Parmer) with the pH and conductivity of sam- 3+ The rate of Al dissolution was calculated using Faraday’s law: ples measured using an Extech DO610 meter (Cole Parmer). It ⋅⋅ M Bacteria cultivation and enumeration Rate of electrode dissolution = (3) zF ⋅⋅ ν The bacteria used in the study were E.  coli P36, which were origi- nally isolated from spinach that expressed kanamycin resistance. Salmonella typhimurium WG49 and L.  monocytogenes were used; where I is the current (in A), t is treatment time (in s), M is the both of which were obtained from the American Type Culture molecular weight of the Al electrode (26.98 g/mol), z is the number Collection (ATCC). The bacteria were individually cultivated in of electrons transferred (for Al it is 3), F is Faraday’s constant (96485 200 ml of Tryptic Soy Broth (TSB; Difco. Sparks, MD, USA) at 37°C C/mol) and ν is the volume of water treated (in m ) (Chopra and for 24  h with the cells being harvested by centrifugation (5000  g, Sharma, 2014). 10 min, 4°C). The cells were washed once in saline and then resus- pended to a final cell density of 8 log cfu/ml. The cell suspensions Ultraviolet collimated beam were stored at 4°C until required, but for no longer than 4 days. Spent wash water (500 ml) from the electrocoagulation process was Escherichia coli P36 was enumerated by plating onto TSA con- filtered through a 20  µm pore size cellulose filter and inoculated with taining 50 µg/ml kanamycin that were then incubated for 24  h at the test bacteria to a final cell density of 7 log cfu/ml. UV treat- 37°C. Salmonella was enumerated on XLD incubated at 37°C for ment was performed using a collimated beam apparatus described 24 h with modified Oxford agar being used to enumerate L. mono- by Bolton and Linden (2003). The system consisted of a 30 W low cytogenes at 30°C for 48 h. pressure mercury vapour UV lamp emitting at 254  nm (Trojan Technologies Inc., London, Canada). The sample (5 ml) was placed Electrocoagulation process in a Petri dish (50  ×  35  mm; Kimax, Kimble Chase, Vineland, NJ, An electrolysis cell was constructed using 6 aluminium plates (7.5 cm USA), then transferred to a holding position on a magnetic stirrer × 7.5 cm, 0.3 cm thickness; effective working area of 570 cm ) with under the collimating tube that uniformly illuminated the test area. an intra-electrode spacing of 2 mm. Each of the three anode plates The sample was illuminated for a designated time and then removed faced a cathode with electrical connections being sealed with silica. for microbiological analysis to determine the number of survivors. The electrodes were polarized at a defined voltage using an Extech A new sample was used for each UV dose treatment to avoid changes DC Power Supply (Cole Parmer) with stirring being achieved via a in volume. magnetic stirrer rotating at 120 rpm. The plate electrodes were sub- Total UV dose (E ) was calculated using the following equation: ave merged in 500 ml volumes of the test sample and voltage applied for the appropriate treatment time. Samples (25 ml) were removed and EE =×× PF WF ×× DF RE (4) ave o passed through a 20 µm pore size cellulose filter and the turbidity of the filtrate determined using a turbidity meter. Turbidity removal effi- where E is the average of six measurements of UV intensity of ciency was determined for all samples using the following equation: the test area measured using a radiometer (Model UVX Digital Radiometer; UV Inc., Upland, Canada). RE is the reflection factor, () C − C PF is the Petri dish factor, WF is the water quality factor, and DF is (1) Turbidity removal efficiency = × 100 C the divergence factor. where C represents the initial turbidity and C is the final turbidity, Experimental design and statistics measured after electrolysis. Each experiment was repeated at least three times and statistically Trials were performed to determine the reduction in model bacte- analysed using ANOVAR and Tukey’s test (S-Plus, Insightful Corp., ria during the electrocoagulation process. Here, simulated spent wash NY, USA). In all cases, the significance level was set at P ≤ 0.05. water was prepared as described in Table 1 with conductivity adjusted to 500 µS/cm. Volumes (500 ml) of the simulated spent wash water was inoculated with E. coli, Salmonella, or L. monocytogenes to a final cell Results density of 5 log cfu/ml. Aliquots (1 ml) were removed to determine the Electrocoagulation treatment initial cell count and the remainder of the sample was electrocoagu- A simulated spent lettuce wash water matrix was prepared to ensure lated for 10 min using an applied potential of 4 V. Upon completion of consistency between the different experiments. The composition of the electrocoagulation process, the flocculated solids were removed by the simulated wastewater were based on that of a commercial shred- filtration through a 20  µm pore size cellulose membrane. Samples were ded lettuce processing line at the midpoint of a typical processing taken of the filtrate to enumerate the number of survivors. activity. The water was characterized by having relatively low solids with corresponding low BOD and COD (Table 1). In preparing the Energy and material consumption simulation spent wash water, the main focus was placed on the sol- The energy consumption was determined using the following ids content given this contributed to turbidity of the solution. The equation: total solids of the simulation wash water the other parameters were 150 K. Khalid D. Alharbi et al., 2017, Vol. 1, No. 2 within the approximate range of that derived from commercial pro- numbers did not correlate to the applied voltage (Table 2). Listeria cess water (Table 1). was also reduced by the electrocoagulation process but by a signifi- It was found that, when simulated spent wash water was sub- cantly (P < 0.05) lower amount compared to E. coli or Salmonella. jected to electrocoagulation treatment, the extent of turbidity reduc- tion was dependent on the treatment time and applied voltage. The UV treatment of filtered electrocoagulated water highest rate of turbidity removal was observed at 8 V although the The UV-assisted inactivation of E.  coli, Salmonella, and Listeria solids removal was insignificantly different (P > 0.05) to 6 V follow- was determined with the model bacteria suspended in spent lettuce ing a 10 min treatment (Figure 1). In contrast, when lower voltages water that had been subjected to electrocoagulation treatment (6 V, were applied, the corresponding turbidity decrease was lower within 10 min, pH 6, 500 µS/cm), and then filtered. The D value for each of the 10 min timeframe. the bacteria was determined from the inactivation curves and com- The extent of turbidity removal was dependent on the pH of the pared with values obtained in saline. It was found that the D values spent wash water (Figure  2). The highest decrease in turbidity was for the different bacteria in saline was varied between 0.22 and 0.28 observed in the neutral to alkali regions, provided the conductivity 2 mJ/cm but were significantly (P < 0.05) higher when performed in of the solution was 500 µS or higher. When the conductivity of solu- electrocoagulated spent water-treated samples (Table 3). The results tion was 110 µS, the optimum pH for achieving maximal turbidity indicate that the apparent UV resistance of the model bacteria is removal was pH 6.5, but the electrocoagulation process was signifi- higher in treated spent wash water compared with saline. It was cantly lower in the alkali or more acidic samples (Figure 2). also noted that the UV transmission of the spent wash water was The current density, aluminium loss from electrodes, and energy 10 ± 0.4% with a measured turbidity of 1.6 ± 0.27 nephelometric consumption increased with voltage (Table 2). However, the reduc- turbidity units (NTU). tion in BOD and COD supported by electrocoagulation performed at 4 vs 8 V was not significantly different (P > 0.05) between the Discussion two voltages (Table  2). The electrocoagulation process reduced the levels of E.  coli and Salmonella although the extent of decrease in The measured parameters of spent wash water derived from com- mercial processing were lower than those reported by Weng et  al. (2016). The authors reported the COD of spent lettuce wash water was in the order of 25 000 mg/l and solids content in the order of 690 mg/l. Yet, it should be noted that the spent wash water described by Weng et  al. (2016) was generated under laboratory conditions by submerging shredded lettuce in distilled water and consequently not representative of commercial conditions. However, it is likely that the spent wash water characteristics would vary between shredded lettuce processing lines depending on factors such as water:product ratio along with the rate in which tanks are recharged with fresh water. The efficacy of the electrocoagulation treatment process to reduce turbidity of simulated spent water was dependent on the applied voltage, pH, and conductivity. The influence of voltage on the electrocoagulation process was related to the current density 3+ generated along with dissemination of Al from the anode, in addi- tion to the generated hydroxide and hydrogen. The current densities required to support an 87% reduction in turbidity were significantly Figure 1. Effect of treatment time and applied voltage on turbidity removal of lower for spent wash water from shredded lettuce wash water when simulated spent lettuce wash water by electrocoagulation treatment. compared with values reported for other processing waste water types. For example, wastewater derived from wineries required cur- rent densities in the order of 400 A/m (Kara et al., 2013) with that derived from potato processing 150 A/m (Kobya et  al., 2006). In other electrocoagulation applications related to wastewater derived from food or related processing, current densities required to achieve high turbidity removal range from 70 to 500 A/m (Drogui et  al., 2008; Kara et al., 2013). Yet, it should be noted that the aforemen- tioned waste waters typically have high solids content along with CODs and BODs compared with shredded lettuce spent wash water used in the current study. Yet, comparable current densities for treat- ing wastewater with low solids content such as cereal processing or water derived from container wash station (Butler et  al., 2011; Kara, 2013). In practical terms, the low current density requirement to achieve turbidity removal means an overall lower energy require- ment and extended working life of electrodes, in addition to lower maintenance requirements. Figure 2. The effect of initial pH and conductivity of simulated spent lettuce The efficacy of turbidity removal was influenced by pH and the wash water by electrocoagulation process using an applied voltage of 6 V and treatment time of 10 min. conductivity of the spent wash water sample. The pH effect is related Treatment of spent wash water, 2017, Vol. 1, No. 2 151 Table 2. Effect of applied potential on the current density, energy consumption, electrode erosion, BOD, COD, and reduction in bacterial numbers. The electrocoagulation process was performed in simulated spent wash water as described in Table 1 with the conductivity set to 500 µS/cm. The electrocoagulation process was performed for 10 min and samples taken away for analysis following filtration. Voltage Current Energy Al dissolution BOD COD Log count reduction (V) density consumption (g/m ) (mg/l) (mg/l) 2 3 Escherichia Salmonella Listeria (A/m ) (kW/h/m ) coli monocytogenes a a a a a 2 1.74 ± 0.30 0.07 11.2 ± 3 Not determined Not determined 1.99 ± 0.06 1.92 ± 0.74 1.08 ± 0.51 b b a a a a a 4 3.48 ± 0.50 0.27 22.4 ± 5 28.7 ± 5.6 46.2 ± 3.9 1.81 ± 0.36 1.87 ± 0.33 1.12 ± 0.23 c c a a a a a 6 5.22 ± 0.18 0.60 33.6 ± 15 28.1 ± 6.8 49.0 ± 4.7 1.98 ± 0.70 1.90 ± 0.20 1.00 ± 0.28 d d a a a a a 8 6.96 ± 0.75 1.07 44.7 ± 12 31.8 ± 7.9 44.6 ± 3.7 1.86 ± 0.58 2.10 ± 0.60 1.15 ± 0.40 Values within columns followed by the same letter are not significantly different. be attributed to flocculation of solids with direct electro-oxidation Table 3. D values (dose to support 1 log reduction) for the UV inacti- playing a lesser role. vation of Escherichia coli, Salmonella, and Listeria monocytogenes Electrocoagulation is not specifically designed as an antimicro- in saline or water derived from simulated spent wash water that has been passed through the electrocoagulation process. bial step unless a salt such as NaCl is supplemented into the waste stream to support the generation of hypochlorite as with electro- Bacterium D value (mJ/cm ) lyzed water (Park et  al., 2008). The salt concentration to support the formation of electrolyzed water is in the order of 0.5% w/v, Saline Electrocoagulated which is far in excess of concentrations encountered in spent lettuce Sspent Wwashwater wash water. Nevertheless, a 1–2 log reduction of the model bacteria Aa Ba L. monocytogenes 0.31 ± 0.06 1.01 ± 0.28 was observed by the electrocoagulation process. If it unclear at this Aa Bb E. coli P36 0.22 ± 0.10 1.60 ± 0.15 time whether the reduction was by direct inactivation by generated Aa Ba Salmonella 0.28 ± 0.05 1.14 ± 0.40 hypochlorite or from being entrapment within the coagulated floc material. Nevertheless, it was evident that the electrocoagulation Values within columns followed by the same lower case letter are not sig- process could contribute to reducing the microbial burden of spent nificantly different. Values within rows followed by the same capital letter are wash water. not significantly different. In the current study, UV was applied as a tertiary treatment to reduce counts of model pathogens to reduce risks of cross-contam- ination in wash tanks. When the inactivation kinetics for the model to aluminium chemistry that, under acidic conditions, forms with 3+ bacteria was performed in water, the D values obtained were in the the monomer (Al ) being the predominant species and polymeric same order as those values published by others (Guerrero-Beltran form dominating at more neutral to alkali pH (Palacios et al., 2016). and Barbosa-Canovas, 2004; Gabriel and Nakano, 2009). However, The polymeric form of aluminium has the property of higher charge in the treated spent lettuce wash water, the D values were increased. neutralization and stabilizing flocs, compared with the monomer, This can be attributed to the UV absorbing constituents associated thereby improving solids removal (Palacios et  al., 2016). It is also with the water that were not removed by electrocoagulation. Indeed, be noted that the charge carried by plant constituents (for exam- the UV transmission of the electrocoagulated water was 10% despite ple, humic acid) would be net negative at neutral to alkali pH that the low turbidity. A similar finding has been reported for UV treat- would also facilitate the flocculation process via interaction with the ment of plant derived wastewater whereby low-molecular-weight polymeric aluminium coagulant (Haynes and Mokolobate, 2001; constituents had a protective towards microbes during treatment Palacios et al., 2016). (Antonelli et al., 2008). Although the presence of UV absorbing con- The effect of pH on the turbidity removal was also influenced stituents can be viewed as a disadvantage of UV, it is possible to sup- by the conductivity of the spent wash water. This can be attributed plement water with hydrogen peroxide to promote an AOP reaction to the potential drop that ultimately decreases the charge density to enhance lethality of UV (Guimaraes et  al., 2016). However, in generated at the given voltage. In practical terms, the conductivity practice, the summation of microbial reduction by electrocoagula- of spent wash water derived from shredded lettuce processing was tion, UV, and hypochlorite added to the wash tank would be suf- low and variable. Yet, even at 100 µS/cm, it was possible to reduce ficient to negate food safety risks presented by cross-contamination turbidity by >80% at pH 6.5 which is around the pH of most com- of pathogens during the wash process. mercial produce wash tanks maintained at when using hypochlorite as a sanitizing agent (Barrera et al., 2012). The reduction of BOD and COD is frequently used as a key met- Conclusion ric in defining the success of an electrocoagulation treatment. In the current study, the BOD and COD were within acceptable limits prior The study demonstrates that a combination of electrocoagulation to electrocoagulation treatment but further reduced. The decrease in followed by UV treatment can be applied as a rapid method for COD and BOD was achieved through solids removal, polymeriza- removing turbidity, solids content, and microbial loading of spent tion of reactive monomers, and/or direct electro-oxidation (Prajapati wash water. A  notable feature of the electrocoagulation process is and Chaudhari, 2015). It is possible that all three mechanisms were the low current density required to achieve a rapid reduction in occurring in treating spent wash water although it was interesting turbidity. Although the reduction of COD and BOD was moderate, to note that the extent of reduction was independent of the cur- it should be noted that both were relatively low in the wastewater rent density. This suggests that the decrease in BOD and COD could stream. It was also demonstrated that electrocoagulation can reduce 152 K. Khalid D. Alharbi et al., 2017, Vol. 1, No. 2 Guimaraes, J. R., Guadagnini, R. A., Bueno Franco, R. M., dos Santos, L. the levels of bacteria of concern in fresh produce processing, thereby U. (2016). Inactivation of Clostridium perfringens, total coliforms, and adding the lethality of the UV-based tertiary treatment. The recycling Escherichia coli by UV/H O in wastewater treatment plant effluent. Jour - 2 2 system proposed can form part of a water management system to nal of Advanced Oxidation Technologies, 19: 93–97. reduce usage in shredded lettuce processing. Haynes, R. J., Mokolobate, M. S. (2001). Amelioration of Al toxicity and P deficiency in acid soils by additions of organic residues: a critical review of the phenomenon and the mechanisms involved. Nutrient Cycling in Acknowledgements Agroecosystems, 59: 47–63. The authors thank the Ontario Ministry of Agriculture, Food and Rural Kara, S. (2013). Treatment of transport container washing wastewater by Affairs for financially supporting the research under the New Directions electrocoagulation. Environmental Progress & Sustainable Energy, 32: Program (project # 051057). They also thank the participating lettuce proces- 249–256. sor for their help and assistance. Kara, S., Gurbulak, E., Eyvaz, M., Yuksel, E. (2013). Treatment of winery wastewater by electrocoagulation process. Desalination and Water Treat- Conflict of interest statement. None declared. ment, 51: 5421–5429. Kobya, M., Hiz, H., Senturk, E., Aydiner, C., Demirbas, E. (2006). Treatment References of potato chips manufacturing wastewater by electrocoagulation. Desali- Aixia, X., Micallef, S. (2015). Impact of different washing methods on epi- nation, 190: 201–211. phytic bacterial communities on organic lettuce. Journal of Food Protec- Lopez-Galvez, F., Gil, M. I., Truchado, P., Selma, M. V., Allende, A. (2010). tion, 78: 250. Cross-contamination of fresh-cut lettuce after a short-term exposure dur- Allegretti et  al. (2016). Recurrent Clostridium difficile infection associates ing pre-washing cannot be controlled after subsequent washing with chlo- with distinct bile acid and microbiome profiles. Alimentary Pharmacology rine dioxide or sodium hypochlorite. Food Microbiology, 27: 199–204. & Therapeutics, 43: 1142–1153. Luo, Y. et al. (2011). Determination of free chlorine concentrations needed to Allende, A., Selma, M. V., Lopez-Galvez, F., Villaescusa, R., Gil, M. I. (2008). prevent Escherichia coli O157:H7 cross-contamination during fresh-cut Impact of wash water quality on sensory and microbial quality, includ- produce wash. Journal of Food Protection, 74: 352–358. ing Escherichia coli cross-contamination, of fresh-cut escarole. Journal of Manzocco, L. et al. (2015). Efficient management of the water resource in the Food Protection, 71: 2514–2518. fresh-cut industry: Current status and perspectives. Trends in Food Science American Public Health Association (APHA). (1998). Standard Methods & Technology, 46: 286–294. for Examination of Water and Wastewater. 17th edn., Washington, DC: Palacios, R. J. S., Kim, D. G., Ko, S. O. (2016). Humic acid removal by electro- American Public Health Association. coagulation: characterization of aluminum species and humic acid. Desali- Antonelli, M., Mezzanotte, V., Nurizzo, C. (2008). Wastewater disinfection by nation and Water Treatment, 57: 10969–10979. UV irradiation: short and long-term efficiency. Environmental Engineering Park, E. J., Alexander, E., Taylor, G. A., Costa, R., Kang, D. H. (2008). Effect Science, 25: 363–373. of electrolyzed water for reduction of foodborne pathogens on lettuce and Barrera, M. J., Blenkinsop, R., Warriner, K. (2012). The effect of different process- spinach. Journal of Food Science, 73: M268–M272. ing parameters on the efficacy of commercial post-harvest washing of mini- Prajapati, A. K., Chaudhari, P. K. (2015). Physicochemical treatment of distill- mally processed spinach and shredded lettuce. Food Control, 25: 745–751. ery wastewater—a review. Chemical Engineering Communications, 202: Bolton, J. R., Linden, K. G. (2003). Standardization of methods for fluence 1098–1117. (UV dose) determination in bench-scale UV experiments. Journal of Envi- Sessa, R. A., Bennett, M. H., Lewis, M. J., Mansfield, J. W., Beale, M. H. ronmental Engineering-Asce, 129: 209–215. (2000). Metabolite profiling of sesquiterpene lactones from Lactuca spe- Bouallagui, H., Touhami, Y., Cheikh, R. B., Hamdi, M. (2005). Bioreactor cies—major latex components are novel oxalate and sulfate conjugates of performance in anaerobic digestion of fruit and vegetable wastes. Process lactucin and its derivatives. Journal of Biological Chemistry, 275: 26877– Biochemistry, 40: 989–995. 26884. Butler, E., Hung, Y. T., Yeh, R. Y. L., Al Ahmad, M. S. (2011). Electrocoagula- ShihChi, W., Yaguang, L., Jie, L., Bin, Z., Jacangelo, J. G., Schwab, K. J. (2016). tion in Wastewater Treatment. Water, 3: 495–525. Assessment and speciation of chlorine demand in fresh-cut produce wash Chopra, A. K., Sharma, A. K. (2014). Disinfection of biologically treated water. Food Control, 60: 543–551. municipal wastewater using electrochemical process. Separation Science Tee, P. F. et  al. (2016). Review on hybrid energy systems for wastewater and Technology, 49: 2613–2619. treatment and bio-energy production. Renewable & Sustainable Energy Drogui, P., Asselin, M., Brar, S. K., Benmoussa, H., Blais, J. F. (2008). Electro- Reviews, 54: 235–246. chemical removal of pollutants from agro-industry wastewaters. Separa- Warriner, K. Namvar, A. (2014). Postharvest washing as a critical control point tion and Purification Technology, 61: 301–310. in fresh produce processing: alternative sanitizers and wash technologies. Farragher, T., Wang, W. C., Worsley, A. (2016). The associations of vegetable Global safety of fresh produce: A handbook of best practice, innovative consumption with food mavenism, personal values, food knowledge and commercial solutions and case studies. UK: Woodhead. demographic factors. Appetite, 97: 29–36. Warriner, K., Huber, A., Namvar, A., Fan, W., Dunfield, K. (2009). Recent Gabriel, A. A., Nakano, H. (2009). Inactivation of Salmonella, E.  coli and advances in the microbial safety of fresh fruits and vegetables. In Taylor, S. Listeria monocytogenes in phosphate-buffered saline and apple juice by L. (ed.) Advances in Food and Nutrition Research. Vol. 57. San Diego, CA: ultraviolet and heat treatments. Food Control, 20: 443–446. Elsevier Academic Press Inc., 155–208. Guerrero-Beltran, J. A., Barbosa-Canovas, G. V. (2004). Review: advantages Weng, S., Yaguang, L., Jie, L., Bin, Z., Jacangelo, J. G., Schwab, K. J. (2016). and limitations on processing foods by UV light. Food Science and Tech- Assessment and speciation of chlorine demand in fresh-cut produce wash nology International, 10: 137–147. water. Food Control, 60: 543–551. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Food Quality and Safety Oxford University Press

Treatment of spent wash water derived from shredded lettuce processing using a combination of electrocoagulation and germicidal ultraviolet light

Loading next page...
 
/lp/oxford-university-press/treatment-of-spent-wash-water-derived-from-shredded-lettuce-processing-EBSmKTT45z

References (31)

Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of Zhejiang University Press.
ISSN
2399-1399
eISSN
2399-1402
DOI
10.1093/fqsafe/fyx012
Publisher site
See Article on Publisher Site

Abstract

Objective: Water recycling is a significant part of an overall water management system. The current study evaluated electrocoagulation, used in combination with ultraviolet light (at 254 nm), to reduce the organic content and enhance the microbiological quality, of wash water derived from shredded lettuce processing. Method: The composition of spent wash water derived from a commercial lettuce processing operation was used to prepare a simulated solution to be applied to validate the water recycling system. The simulated spent wash water was subjected to an electrocoagulation process followed by filtration and a tertiary ultraviolet (254 nm) treatment. The efficacy of the recycling treatment to decrease turbidity (nephelometric turbidity units, biological oxygen demand (BOD), chemical oxygen demand (COD) and decrease in introduced bacterial numbers. Results: Spent wash water sampled from a commercial processing line was found to be colloidal in nature (78 ± 26 NTU) with low total solids content (544 ± 87 mg/L), BOD (230 ± 53 mg/L) and COD (309 ± 53 mg/L). An electrocogaultion process performed for 10 min using 3.48 A/m current density at pH 6.5 and conductivity of >100 µS/cm supported an 87% removal of turbidity, 38% reduction in BOD along with 49% decrease in COD. The electrocoagulation process was also found to reduce the levels of Escherichia coli, Salmonella and Listeria monocytogenes by 1–2 log cfu. The tertiary UV treatment of water derived from the electro coagulation process, supported further reduction in model pathogens, although it was noted that the D values for inactivation were in the order of 2 2 1.01–1.60 mJ/cm , which compares to 0.22–0.31 mJ/cm in saline. The apparent increase in bacterial resistance to ultraviolet was likely due to the UV absorbing low molecular weight constituents within wash water that provided protection against inactivation. Conclusion: In conclusion, the study demonstrated the feasibility of applying electrocoagulation and UV to rapidly treat spent lettuce wash water to facilitate in-process recycling within shredded lettuce processing operations. Key words: Biological oxygen demand (BOD); Chemical oxygen demand (COD); Electrocoagulation; Fresh produce; Lettuce; UV; Washing; Water recycling. © The Author 2017. Published by Oxford University Press on behalf of Zhejiang University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com 148 K. Khalid D. Alharbi et al., 2017, Vol. 1, No. 2 ions. In the course of electrolysis, dissolution of the plates occurs gen- Introduction erating aluminium hydroxide, or polymers, that act to coagulate solids Ever since the introduction of processed leafy greens in the late 1980s that are raised to the surface via the generation of hydrogen gas at the to the produce sector, the demand for bagged salads has grown at a cathode (Butler et  al., 2011). The direct electro-oxidation of water rate of 10% each year and now represents a $51.4 billion industry in constituents reduce the COD and BOD although reductions in micro- North America alone (Warriner et al., 2009). One of the largest sec- bial levels by the process have not been previously described. With tors in the fresh produce market continues to be lettuce-based salads regards to the latter, tertiary treatment is applied that can include the (Farragher et al., 2016). In preparing the product, heads of lettuce are addition of chemical sanitizers or physical methods such as UV-C ger- passed through a shredding unit, and then through an initial wash to micidal lamps (Manzocco et al., 2015). The benefits of UV-C are that remove soils followed by a biocidal wash to reduce microbial load- the process can support the inactivation of microbes, in addition to the ing (Warriner et al., 2009). Hypochlorite is the mostly used sanitizing photo-oxidation of reactive species within wash water. agent partly due to low cost, and also because of historical reasons In the following, the application of electrocoagulation in com- along with ease of monitoring using oxidation reduction potential bination with UV to treat spent wash water derived from shredded (ORP)-based systems (Barrera et al., 2012). Through various studies it lettuce processing has been evaluated. The work encompassed for- has been clearly demonstrated that the efficacy of washing for remov- mulating a simulation matrix based on characterizing spent wash ing pathogens, such as Escherichia coli O157:H7, Salmonella, and water derived from a commercial processing line. The simulation Listeria monocytogenes, is limited to 1–2 log cfu reductions (Aixia spent water was used to assess the efficacy of electrocoagulation to and Micallef, 2015). Of more concern is the potential for pathogens remove turbidity, solids content, BOD, COD, and microbial loading. to be redistributed between produce batches via cross-contamination The inactivation kinetics of model bacteria by UV in treated spent events (Allende et al., 2008; Lopez-Galvez et al., 2010). To this end, it wash water were then determined. has been found that maintaining free chlorine levels within the wash tank is a critical factor in minimizing cross-contamination events (Luo et al., 2011). However, even with wash processes using ORP feedback Experimental systems, the ability to maintain free chlorine levels is compromised by Spent leafy green water characterization and the accumulation of organics within the wash water that react with preparation of simulated spent irrigation water hypochlorite to form disinfection by-products that essentially seques- Spent leafy green wash water was collected from a commercial pro- ters antimicrobial action (Luo et al., 2011). This is especially relevant cessing facility on three separate occasions and characterized with with lettuce that contains high concentrations of latex and antioxi- respect to turbidity, solids content, BOD, and COD (Table  1). The dants, along with other constituents that could potentially interact processing line processed 2300 kg/h of shredded iceberg lettuce over with chlorine (Sessa et al., 2000; ShihChi et al., 2016). To counter the a period of up to 30 h. Sanitizer (50–200 ppm sodium hypochlorite) accumulation of organic matter within the wash tank, it is possible was automatically added by using a dual pH–ORP control system, to simply add increasing concentrations of hypochlorite to meet the the level of chlorine and pH in water were maintained in the range chlorine demand (ShihChi et  al., 2016). However, this approach is of 600–850 mV and pH 6.5 – 6.9, respectively. The wash tanks were unfeasible due to the generation of high levels of disinfection by-prod- replenished with fresh water at a rate of 20–30% per hour. ucts that affect flavour and odour taints on the product. A more com- To ensure consistency between experiments, a simulated spent wash mon approach is to reintroduce or recharge the wash tanks with fresh water was prepared based on the composition of commercial wash water at a rate of 10%–50% depending on the processor (Warriner water samples (Table 1). Specifically, the simulated water was prepared and Namvar, 2014). However, recharging tanks significantly increases by blending iceberg lettuce leaves (6.6  g/l) with 0.1  g/l bentonite in the usage of water and contributes to the estimated 40 l of water to distilled water. Large particulates were removed from the homogenate produce 1 kg of lettuce. Moreover, the larger volume of water usage by passing through a 20 µm pore size cellulose filter. The conductivity increases the amount of wastewater that requires treatment, which of the filtrate was adjusted between 100 and 1000 µ S/cm using NaCl. further elevates costs (Manzocco et al., 2015). A potential approach to reduce water usage is through recycling of spent water back into the process following removal of solids along Analytical techniques with reduction in the biological oxygen demand (BOD) and chemical COD of water samples were determined using HACH DBR 200 oxygen demand (COD) (Manzocco et  al., 2015). Technologies that Reactor (Hach Co., Loveland, CO, USA) for digestion and HACH recycle water can be based on a combination of physical, biological, DR 2800 (Hach Co., Loveland, CO, USA) for colorimetric determina- and chemical steps (Tee et al., 2016). Although biological treatments tion method according to Standard Method 5220D (APHA, AWWA, are highly efficient at reducing the BOD and COD, the process is time consuming and adds to the footprint along with processing costs Table 1. Characteristics of spent wash water derived from a shred- ded lettuce processing facility. Also shown are the characteristics (Bouallagui et  al., 2005). Therefore, physical and chemical methods of a simulated spent shredded lettuce wash water prepared from a are preferred for a continuous or semi-continuous process in which blend of lettuce leaves, bentonite, and NaCl. the water can be returned into the processing line. Here, a coagula- tion step is performed to aggregate the solids to facilitate filtration, Parameter Commercial spent Simulation spent although residual coagulant agent such as alum can be a problem due wash water wash water to toxic residue (Warriner and Namvar, 2014). Electrocoagulation has Turbidity (NTU) 78 ± 26 50 ± 13 been applied to treat a range of wastewater types from domestic sew- pH 6.9 6.9 age through to water derived from winery operations although it has Total solids (mg/l) 544 ± 87 560 ± 48 not been evaluated for treating spent leafy green wash water (Butler Total soluble solids (mg/l) 104 ± 98 55 ± 12 et  al., 2011). The electrocoagulation chamber typically consists of BOD (mg/l) 230 ± 53 164 ± 105 sacrificial aluminium plate electrodes that are polarized at a defined COD (mg/l) 309 ± 67 285 ± 92 voltage. Oxidation reactions occur at the anode that can lead to the Conductivity (µS/cm) 785 ± 102 100–1000 production of hypochlorite depending on the concentration of chloride Treatment of spent wash water, 2017, Vol. 1, No. 2 149 WEF, 1989). BOD was determined in accordance with method 5210 VI ⋅⋅ t Energy consumption = (2) (APHA, AWWA, WEF, 1989). Total soluble solids (TSS) was quanti- fied by using filtration (1.5  µm pore size glass microfibre filter; Cole Parmer, Montreal, Canada) followed by drying of the retentate at where V is the voltage, I is current (A), t is the run time (in s), and 105°C in accordance with method 2540D (APHA, AWWA, WEF, ν is the volume of liquid treated (m ) (Chopra and Sharma, 2014). 1989). Total solids (TS) was determined using method 5210 (APHA, The energy consumption was calculated in Watts (W) and the final AWWA, WEF, 1989). Turbidity was measured by a turbidity meter amount is reported in kW/h/m (Chopra and Sharma, 2014). (Oakton T-100; Cole Parmer) with the pH and conductivity of sam- 3+ The rate of Al dissolution was calculated using Faraday’s law: ples measured using an Extech DO610 meter (Cole Parmer). It ⋅⋅ M Bacteria cultivation and enumeration Rate of electrode dissolution = (3) zF ⋅⋅ ν The bacteria used in the study were E.  coli P36, which were origi- nally isolated from spinach that expressed kanamycin resistance. Salmonella typhimurium WG49 and L.  monocytogenes were used; where I is the current (in A), t is treatment time (in s), M is the both of which were obtained from the American Type Culture molecular weight of the Al electrode (26.98 g/mol), z is the number Collection (ATCC). The bacteria were individually cultivated in of electrons transferred (for Al it is 3), F is Faraday’s constant (96485 200 ml of Tryptic Soy Broth (TSB; Difco. Sparks, MD, USA) at 37°C C/mol) and ν is the volume of water treated (in m ) (Chopra and for 24  h with the cells being harvested by centrifugation (5000  g, Sharma, 2014). 10 min, 4°C). The cells were washed once in saline and then resus- pended to a final cell density of 8 log cfu/ml. The cell suspensions Ultraviolet collimated beam were stored at 4°C until required, but for no longer than 4 days. Spent wash water (500 ml) from the electrocoagulation process was Escherichia coli P36 was enumerated by plating onto TSA con- filtered through a 20  µm pore size cellulose filter and inoculated with taining 50 µg/ml kanamycin that were then incubated for 24  h at the test bacteria to a final cell density of 7 log cfu/ml. UV treat- 37°C. Salmonella was enumerated on XLD incubated at 37°C for ment was performed using a collimated beam apparatus described 24 h with modified Oxford agar being used to enumerate L. mono- by Bolton and Linden (2003). The system consisted of a 30 W low cytogenes at 30°C for 48 h. pressure mercury vapour UV lamp emitting at 254  nm (Trojan Technologies Inc., London, Canada). The sample (5 ml) was placed Electrocoagulation process in a Petri dish (50  ×  35  mm; Kimax, Kimble Chase, Vineland, NJ, An electrolysis cell was constructed using 6 aluminium plates (7.5 cm USA), then transferred to a holding position on a magnetic stirrer × 7.5 cm, 0.3 cm thickness; effective working area of 570 cm ) with under the collimating tube that uniformly illuminated the test area. an intra-electrode spacing of 2 mm. Each of the three anode plates The sample was illuminated for a designated time and then removed faced a cathode with electrical connections being sealed with silica. for microbiological analysis to determine the number of survivors. The electrodes were polarized at a defined voltage using an Extech A new sample was used for each UV dose treatment to avoid changes DC Power Supply (Cole Parmer) with stirring being achieved via a in volume. magnetic stirrer rotating at 120 rpm. The plate electrodes were sub- Total UV dose (E ) was calculated using the following equation: ave merged in 500 ml volumes of the test sample and voltage applied for the appropriate treatment time. Samples (25 ml) were removed and EE =×× PF WF ×× DF RE (4) ave o passed through a 20 µm pore size cellulose filter and the turbidity of the filtrate determined using a turbidity meter. Turbidity removal effi- where E is the average of six measurements of UV intensity of ciency was determined for all samples using the following equation: the test area measured using a radiometer (Model UVX Digital Radiometer; UV Inc., Upland, Canada). RE is the reflection factor, () C − C PF is the Petri dish factor, WF is the water quality factor, and DF is (1) Turbidity removal efficiency = × 100 C the divergence factor. where C represents the initial turbidity and C is the final turbidity, Experimental design and statistics measured after electrolysis. Each experiment was repeated at least three times and statistically Trials were performed to determine the reduction in model bacte- analysed using ANOVAR and Tukey’s test (S-Plus, Insightful Corp., ria during the electrocoagulation process. Here, simulated spent wash NY, USA). In all cases, the significance level was set at P ≤ 0.05. water was prepared as described in Table 1 with conductivity adjusted to 500 µS/cm. Volumes (500 ml) of the simulated spent wash water was inoculated with E. coli, Salmonella, or L. monocytogenes to a final cell Results density of 5 log cfu/ml. Aliquots (1 ml) were removed to determine the Electrocoagulation treatment initial cell count and the remainder of the sample was electrocoagu- A simulated spent lettuce wash water matrix was prepared to ensure lated for 10 min using an applied potential of 4 V. Upon completion of consistency between the different experiments. The composition of the electrocoagulation process, the flocculated solids were removed by the simulated wastewater were based on that of a commercial shred- filtration through a 20  µm pore size cellulose membrane. Samples were ded lettuce processing line at the midpoint of a typical processing taken of the filtrate to enumerate the number of survivors. activity. The water was characterized by having relatively low solids with corresponding low BOD and COD (Table 1). In preparing the Energy and material consumption simulation spent wash water, the main focus was placed on the sol- The energy consumption was determined using the following ids content given this contributed to turbidity of the solution. The equation: total solids of the simulation wash water the other parameters were 150 K. Khalid D. Alharbi et al., 2017, Vol. 1, No. 2 within the approximate range of that derived from commercial pro- numbers did not correlate to the applied voltage (Table 2). Listeria cess water (Table 1). was also reduced by the electrocoagulation process but by a signifi- It was found that, when simulated spent wash water was sub- cantly (P < 0.05) lower amount compared to E. coli or Salmonella. jected to electrocoagulation treatment, the extent of turbidity reduc- tion was dependent on the treatment time and applied voltage. The UV treatment of filtered electrocoagulated water highest rate of turbidity removal was observed at 8 V although the The UV-assisted inactivation of E.  coli, Salmonella, and Listeria solids removal was insignificantly different (P > 0.05) to 6 V follow- was determined with the model bacteria suspended in spent lettuce ing a 10 min treatment (Figure 1). In contrast, when lower voltages water that had been subjected to electrocoagulation treatment (6 V, were applied, the corresponding turbidity decrease was lower within 10 min, pH 6, 500 µS/cm), and then filtered. The D value for each of the 10 min timeframe. the bacteria was determined from the inactivation curves and com- The extent of turbidity removal was dependent on the pH of the pared with values obtained in saline. It was found that the D values spent wash water (Figure  2). The highest decrease in turbidity was for the different bacteria in saline was varied between 0.22 and 0.28 observed in the neutral to alkali regions, provided the conductivity 2 mJ/cm but were significantly (P < 0.05) higher when performed in of the solution was 500 µS or higher. When the conductivity of solu- electrocoagulated spent water-treated samples (Table 3). The results tion was 110 µS, the optimum pH for achieving maximal turbidity indicate that the apparent UV resistance of the model bacteria is removal was pH 6.5, but the electrocoagulation process was signifi- higher in treated spent wash water compared with saline. It was cantly lower in the alkali or more acidic samples (Figure 2). also noted that the UV transmission of the spent wash water was The current density, aluminium loss from electrodes, and energy 10 ± 0.4% with a measured turbidity of 1.6 ± 0.27 nephelometric consumption increased with voltage (Table 2). However, the reduc- turbidity units (NTU). tion in BOD and COD supported by electrocoagulation performed at 4 vs 8 V was not significantly different (P > 0.05) between the Discussion two voltages (Table  2). The electrocoagulation process reduced the levels of E.  coli and Salmonella although the extent of decrease in The measured parameters of spent wash water derived from com- mercial processing were lower than those reported by Weng et  al. (2016). The authors reported the COD of spent lettuce wash water was in the order of 25 000 mg/l and solids content in the order of 690 mg/l. Yet, it should be noted that the spent wash water described by Weng et  al. (2016) was generated under laboratory conditions by submerging shredded lettuce in distilled water and consequently not representative of commercial conditions. However, it is likely that the spent wash water characteristics would vary between shredded lettuce processing lines depending on factors such as water:product ratio along with the rate in which tanks are recharged with fresh water. The efficacy of the electrocoagulation treatment process to reduce turbidity of simulated spent water was dependent on the applied voltage, pH, and conductivity. The influence of voltage on the electrocoagulation process was related to the current density 3+ generated along with dissemination of Al from the anode, in addi- tion to the generated hydroxide and hydrogen. The current densities required to support an 87% reduction in turbidity were significantly Figure 1. Effect of treatment time and applied voltage on turbidity removal of lower for spent wash water from shredded lettuce wash water when simulated spent lettuce wash water by electrocoagulation treatment. compared with values reported for other processing waste water types. For example, wastewater derived from wineries required cur- rent densities in the order of 400 A/m (Kara et al., 2013) with that derived from potato processing 150 A/m (Kobya et  al., 2006). In other electrocoagulation applications related to wastewater derived from food or related processing, current densities required to achieve high turbidity removal range from 70 to 500 A/m (Drogui et  al., 2008; Kara et al., 2013). Yet, it should be noted that the aforemen- tioned waste waters typically have high solids content along with CODs and BODs compared with shredded lettuce spent wash water used in the current study. Yet, comparable current densities for treat- ing wastewater with low solids content such as cereal processing or water derived from container wash station (Butler et  al., 2011; Kara, 2013). In practical terms, the low current density requirement to achieve turbidity removal means an overall lower energy require- ment and extended working life of electrodes, in addition to lower maintenance requirements. Figure 2. The effect of initial pH and conductivity of simulated spent lettuce The efficacy of turbidity removal was influenced by pH and the wash water by electrocoagulation process using an applied voltage of 6 V and treatment time of 10 min. conductivity of the spent wash water sample. The pH effect is related Treatment of spent wash water, 2017, Vol. 1, No. 2 151 Table 2. Effect of applied potential on the current density, energy consumption, electrode erosion, BOD, COD, and reduction in bacterial numbers. The electrocoagulation process was performed in simulated spent wash water as described in Table 1 with the conductivity set to 500 µS/cm. The electrocoagulation process was performed for 10 min and samples taken away for analysis following filtration. Voltage Current Energy Al dissolution BOD COD Log count reduction (V) density consumption (g/m ) (mg/l) (mg/l) 2 3 Escherichia Salmonella Listeria (A/m ) (kW/h/m ) coli monocytogenes a a a a a 2 1.74 ± 0.30 0.07 11.2 ± 3 Not determined Not determined 1.99 ± 0.06 1.92 ± 0.74 1.08 ± 0.51 b b a a a a a 4 3.48 ± 0.50 0.27 22.4 ± 5 28.7 ± 5.6 46.2 ± 3.9 1.81 ± 0.36 1.87 ± 0.33 1.12 ± 0.23 c c a a a a a 6 5.22 ± 0.18 0.60 33.6 ± 15 28.1 ± 6.8 49.0 ± 4.7 1.98 ± 0.70 1.90 ± 0.20 1.00 ± 0.28 d d a a a a a 8 6.96 ± 0.75 1.07 44.7 ± 12 31.8 ± 7.9 44.6 ± 3.7 1.86 ± 0.58 2.10 ± 0.60 1.15 ± 0.40 Values within columns followed by the same letter are not significantly different. be attributed to flocculation of solids with direct electro-oxidation Table 3. D values (dose to support 1 log reduction) for the UV inacti- playing a lesser role. vation of Escherichia coli, Salmonella, and Listeria monocytogenes Electrocoagulation is not specifically designed as an antimicro- in saline or water derived from simulated spent wash water that has been passed through the electrocoagulation process. bial step unless a salt such as NaCl is supplemented into the waste stream to support the generation of hypochlorite as with electro- Bacterium D value (mJ/cm ) lyzed water (Park et  al., 2008). The salt concentration to support the formation of electrolyzed water is in the order of 0.5% w/v, Saline Electrocoagulated which is far in excess of concentrations encountered in spent lettuce Sspent Wwashwater wash water. Nevertheless, a 1–2 log reduction of the model bacteria Aa Ba L. monocytogenes 0.31 ± 0.06 1.01 ± 0.28 was observed by the electrocoagulation process. If it unclear at this Aa Bb E. coli P36 0.22 ± 0.10 1.60 ± 0.15 time whether the reduction was by direct inactivation by generated Aa Ba Salmonella 0.28 ± 0.05 1.14 ± 0.40 hypochlorite or from being entrapment within the coagulated floc material. Nevertheless, it was evident that the electrocoagulation Values within columns followed by the same lower case letter are not sig- process could contribute to reducing the microbial burden of spent nificantly different. Values within rows followed by the same capital letter are wash water. not significantly different. In the current study, UV was applied as a tertiary treatment to reduce counts of model pathogens to reduce risks of cross-contam- ination in wash tanks. When the inactivation kinetics for the model to aluminium chemistry that, under acidic conditions, forms with 3+ bacteria was performed in water, the D values obtained were in the the monomer (Al ) being the predominant species and polymeric same order as those values published by others (Guerrero-Beltran form dominating at more neutral to alkali pH (Palacios et al., 2016). and Barbosa-Canovas, 2004; Gabriel and Nakano, 2009). However, The polymeric form of aluminium has the property of higher charge in the treated spent lettuce wash water, the D values were increased. neutralization and stabilizing flocs, compared with the monomer, This can be attributed to the UV absorbing constituents associated thereby improving solids removal (Palacios et  al., 2016). It is also with the water that were not removed by electrocoagulation. Indeed, be noted that the charge carried by plant constituents (for exam- the UV transmission of the electrocoagulated water was 10% despite ple, humic acid) would be net negative at neutral to alkali pH that the low turbidity. A similar finding has been reported for UV treat- would also facilitate the flocculation process via interaction with the ment of plant derived wastewater whereby low-molecular-weight polymeric aluminium coagulant (Haynes and Mokolobate, 2001; constituents had a protective towards microbes during treatment Palacios et al., 2016). (Antonelli et al., 2008). Although the presence of UV absorbing con- The effect of pH on the turbidity removal was also influenced stituents can be viewed as a disadvantage of UV, it is possible to sup- by the conductivity of the spent wash water. This can be attributed plement water with hydrogen peroxide to promote an AOP reaction to the potential drop that ultimately decreases the charge density to enhance lethality of UV (Guimaraes et  al., 2016). However, in generated at the given voltage. In practical terms, the conductivity practice, the summation of microbial reduction by electrocoagula- of spent wash water derived from shredded lettuce processing was tion, UV, and hypochlorite added to the wash tank would be suf- low and variable. Yet, even at 100 µS/cm, it was possible to reduce ficient to negate food safety risks presented by cross-contamination turbidity by >80% at pH 6.5 which is around the pH of most com- of pathogens during the wash process. mercial produce wash tanks maintained at when using hypochlorite as a sanitizing agent (Barrera et al., 2012). The reduction of BOD and COD is frequently used as a key met- Conclusion ric in defining the success of an electrocoagulation treatment. In the current study, the BOD and COD were within acceptable limits prior The study demonstrates that a combination of electrocoagulation to electrocoagulation treatment but further reduced. The decrease in followed by UV treatment can be applied as a rapid method for COD and BOD was achieved through solids removal, polymeriza- removing turbidity, solids content, and microbial loading of spent tion of reactive monomers, and/or direct electro-oxidation (Prajapati wash water. A  notable feature of the electrocoagulation process is and Chaudhari, 2015). It is possible that all three mechanisms were the low current density required to achieve a rapid reduction in occurring in treating spent wash water although it was interesting turbidity. Although the reduction of COD and BOD was moderate, to note that the extent of reduction was independent of the cur- it should be noted that both were relatively low in the wastewater rent density. This suggests that the decrease in BOD and COD could stream. It was also demonstrated that electrocoagulation can reduce 152 K. Khalid D. Alharbi et al., 2017, Vol. 1, No. 2 Guimaraes, J. R., Guadagnini, R. A., Bueno Franco, R. M., dos Santos, L. the levels of bacteria of concern in fresh produce processing, thereby U. (2016). Inactivation of Clostridium perfringens, total coliforms, and adding the lethality of the UV-based tertiary treatment. The recycling Escherichia coli by UV/H O in wastewater treatment plant effluent. Jour - 2 2 system proposed can form part of a water management system to nal of Advanced Oxidation Technologies, 19: 93–97. reduce usage in shredded lettuce processing. Haynes, R. J., Mokolobate, M. S. (2001). Amelioration of Al toxicity and P deficiency in acid soils by additions of organic residues: a critical review of the phenomenon and the mechanisms involved. Nutrient Cycling in Acknowledgements Agroecosystems, 59: 47–63. The authors thank the Ontario Ministry of Agriculture, Food and Rural Kara, S. (2013). Treatment of transport container washing wastewater by Affairs for financially supporting the research under the New Directions electrocoagulation. Environmental Progress & Sustainable Energy, 32: Program (project # 051057). They also thank the participating lettuce proces- 249–256. sor for their help and assistance. Kara, S., Gurbulak, E., Eyvaz, M., Yuksel, E. (2013). Treatment of winery wastewater by electrocoagulation process. Desalination and Water Treat- Conflict of interest statement. None declared. ment, 51: 5421–5429. Kobya, M., Hiz, H., Senturk, E., Aydiner, C., Demirbas, E. (2006). Treatment References of potato chips manufacturing wastewater by electrocoagulation. Desali- Aixia, X., Micallef, S. (2015). Impact of different washing methods on epi- nation, 190: 201–211. phytic bacterial communities on organic lettuce. Journal of Food Protec- Lopez-Galvez, F., Gil, M. I., Truchado, P., Selma, M. V., Allende, A. (2010). tion, 78: 250. Cross-contamination of fresh-cut lettuce after a short-term exposure dur- Allegretti et  al. (2016). Recurrent Clostridium difficile infection associates ing pre-washing cannot be controlled after subsequent washing with chlo- with distinct bile acid and microbiome profiles. Alimentary Pharmacology rine dioxide or sodium hypochlorite. Food Microbiology, 27: 199–204. & Therapeutics, 43: 1142–1153. Luo, Y. et al. (2011). Determination of free chlorine concentrations needed to Allende, A., Selma, M. V., Lopez-Galvez, F., Villaescusa, R., Gil, M. I. (2008). prevent Escherichia coli O157:H7 cross-contamination during fresh-cut Impact of wash water quality on sensory and microbial quality, includ- produce wash. Journal of Food Protection, 74: 352–358. ing Escherichia coli cross-contamination, of fresh-cut escarole. Journal of Manzocco, L. et al. (2015). Efficient management of the water resource in the Food Protection, 71: 2514–2518. fresh-cut industry: Current status and perspectives. Trends in Food Science American Public Health Association (APHA). (1998). Standard Methods & Technology, 46: 286–294. for Examination of Water and Wastewater. 17th edn., Washington, DC: Palacios, R. J. S., Kim, D. G., Ko, S. O. (2016). Humic acid removal by electro- American Public Health Association. coagulation: characterization of aluminum species and humic acid. Desali- Antonelli, M., Mezzanotte, V., Nurizzo, C. (2008). Wastewater disinfection by nation and Water Treatment, 57: 10969–10979. UV irradiation: short and long-term efficiency. Environmental Engineering Park, E. J., Alexander, E., Taylor, G. A., Costa, R., Kang, D. H. (2008). Effect Science, 25: 363–373. of electrolyzed water for reduction of foodborne pathogens on lettuce and Barrera, M. J., Blenkinsop, R., Warriner, K. (2012). The effect of different process- spinach. Journal of Food Science, 73: M268–M272. ing parameters on the efficacy of commercial post-harvest washing of mini- Prajapati, A. K., Chaudhari, P. K. (2015). Physicochemical treatment of distill- mally processed spinach and shredded lettuce. Food Control, 25: 745–751. ery wastewater—a review. Chemical Engineering Communications, 202: Bolton, J. R., Linden, K. G. (2003). Standardization of methods for fluence 1098–1117. (UV dose) determination in bench-scale UV experiments. Journal of Envi- Sessa, R. A., Bennett, M. H., Lewis, M. J., Mansfield, J. W., Beale, M. H. ronmental Engineering-Asce, 129: 209–215. (2000). Metabolite profiling of sesquiterpene lactones from Lactuca spe- Bouallagui, H., Touhami, Y., Cheikh, R. B., Hamdi, M. (2005). Bioreactor cies—major latex components are novel oxalate and sulfate conjugates of performance in anaerobic digestion of fruit and vegetable wastes. Process lactucin and its derivatives. Journal of Biological Chemistry, 275: 26877– Biochemistry, 40: 989–995. 26884. Butler, E., Hung, Y. T., Yeh, R. Y. L., Al Ahmad, M. S. (2011). Electrocoagula- ShihChi, W., Yaguang, L., Jie, L., Bin, Z., Jacangelo, J. G., Schwab, K. J. (2016). tion in Wastewater Treatment. Water, 3: 495–525. Assessment and speciation of chlorine demand in fresh-cut produce wash Chopra, A. K., Sharma, A. K. (2014). Disinfection of biologically treated water. Food Control, 60: 543–551. municipal wastewater using electrochemical process. Separation Science Tee, P. F. et  al. (2016). Review on hybrid energy systems for wastewater and Technology, 49: 2613–2619. treatment and bio-energy production. Renewable & Sustainable Energy Drogui, P., Asselin, M., Brar, S. K., Benmoussa, H., Blais, J. F. (2008). Electro- Reviews, 54: 235–246. chemical removal of pollutants from agro-industry wastewaters. Separa- Warriner, K. Namvar, A. (2014). Postharvest washing as a critical control point tion and Purification Technology, 61: 301–310. in fresh produce processing: alternative sanitizers and wash technologies. Farragher, T., Wang, W. C., Worsley, A. (2016). The associations of vegetable Global safety of fresh produce: A handbook of best practice, innovative consumption with food mavenism, personal values, food knowledge and commercial solutions and case studies. UK: Woodhead. demographic factors. Appetite, 97: 29–36. Warriner, K., Huber, A., Namvar, A., Fan, W., Dunfield, K. (2009). Recent Gabriel, A. A., Nakano, H. (2009). Inactivation of Salmonella, E.  coli and advances in the microbial safety of fresh fruits and vegetables. In Taylor, S. Listeria monocytogenes in phosphate-buffered saline and apple juice by L. (ed.) Advances in Food and Nutrition Research. Vol. 57. San Diego, CA: ultraviolet and heat treatments. Food Control, 20: 443–446. Elsevier Academic Press Inc., 155–208. Guerrero-Beltran, J. A., Barbosa-Canovas, G. V. (2004). Review: advantages Weng, S., Yaguang, L., Jie, L., Bin, Z., Jacangelo, J. G., Schwab, K. J. (2016). and limitations on processing foods by UV light. Food Science and Tech- Assessment and speciation of chlorine demand in fresh-cut produce wash nology International, 10: 137–147. water. Food Control, 60: 543–551.

Journal

Food Quality and SafetyOxford University Press

Published: May 1, 2017

There are no references for this article.