Wastewater disinfection with PAA and UV combined treatment: a pilot plant study

Water Research 37 (2003) 2365–2371

Wastewater disinfection with PAA and UV combined treatment: a pilot plant study Cecilia Caretti*, Claudio Lubello Dipartimento di Ingegneria Civile, Universita" degli Studi di Firenze, via S.Marta 3, Firenze 50139, Italy Received 22 April 2002; accepted 20 December 2002

Abstract This study is part of a larger research project on Advanced Treatments for wastewater reuse in agriculture. Because of Italy’s strict microbiological limits on unrestricted wastewater reuse in agriculture (2 MPN/100 ml Total Coliforms), a very high degree of disinfection is necessary. The objective of this study is to proceed in validating, with a pilot plant experimentation, previous laboratory results on the disinfection efficacy of the synergic combined treatment between ultraviolet irradiation (UV) and peracetic acid (PAA). The research has been carried out through a 5 month on-site experimental study in a pilot plant, considering four different solutions: PAA addition, UV irradiation, addition of PAA upstream the UV device (PAA+UV) and addition of PAA downstream the UV device (UV+PAA). In the investigated experimental conditions (2–8 ppm of PAA with 10–30 min contact time; 100–300 mJ/cm2 UV), it has been impossible to meet the microbiological limits through an exclusive use of UV irradiation or PAA. The disinfection efficacy enhances by using the UV+PAA treatment, but a much higher efficacy gain occurs by using the PAA+UV treatment. In this latter case, the higher efficiency is recognized as being brought about by the formation of free radicals due to the photolysis of the PAA when in presence of the UV rays. A preliminary cost analysis has been carried out in order to highlight the more economically advantageous solution which guarantees compliance to the strict limits. r 2003 Elsevier Science Ltd. All rights reserved. Keywords: Wastewater; Disinfection; Peracetic acid; Ultraviolet irradiation; Advanced oxidation processes; Agricultural reuse

1. Introduction The advanced oxidation processes (AOPs) are based on two main steps—the generation of free radicals and the oxidation reactions of the polluting molecules caused by these radicals: þpollutants

AOP-d OH ƒƒƒƒƒ! CO2 þ H2 O þ inorganic ions: In literature, the principal applications of these processes refer to the oxidation of organic compounds, dissolved inorganic compounds and other pollutants *Corresponding author. Tel.: +39-0554796435; fax: +39055495333. E-mail address: [email protected]fi.it (C. Caretti).

that are toxic and/or refractory to biological treatments [1–3,22]. Few references were found, however, regarding their use for wastewater disinfection [4,5]. The hydroxyl radical is primarily responsible for the efficiency of the AOPs. Its high oxidation potential in fact guarantees a rapid degradation of the polluting molecules. In both natural and drinking water, the average life-span of the dOH radical is very short (about 10 ms); however, sufficient for the radical to perform its oxidizing action and promote the formation of a chain reaction capable of producing other free radicals [6,7]. Since the disinfection process of PAA seems to take place according to radical type reactions, we have tried to research if and how the UV could accelerate such reactions.

0043-1354/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0043-1354(03)00025-3

C. Caretti, C. Lubello / Water Research 37 (2003) 2365–2371

2366

Contrary to other AOPs (UV/H2O2, O3/H2O2, O3/ UV, TiO2/UV), numerous bibliographic references for the combined treatment between peracetic acid (PAA) and UV do not exist [8]. It is possible to go back to the peroxide acids theory to show how the action of UV produces a homolytic rupture in the O–O bond of the PAA molecule, with the subsequent formation of the hydroxyl radical: þhv

CH3 CO3 H ƒ! CH3 COd2 þ d OH: The CH3COd2 molecule rapidly declines forming CHd3 and CO2, while the molecule of PAA can subsequently react with the dOH radicals produced, according to the following reactions of addition and subtr action of a labile hydrogen: CH3 CO3 H þ d OH -CH3 CO4 H2 -CH3 CO2 H þ d OOH;

effective synergy between the two treatments and the considerable increase of the effectiveness of the combined treatment as compared to the single treatments considered separately [12]. Moreover, the combined treatment of PAA and UV proved to be more effective than the combined treatment of hydrogen peroxide (H2O2) and UV [13]. This data has also been confirmed by spectroscopic analysis, which showed a remarkable increase in the production of free radicals when moving from the combined treatment of H2O2 and UV to the combined treatment of PAA and UV [14,15]. The aim of this study is to proceed in validating previous laboratory results, with a pilot plant experimentation, paying particular attention to the technical and economic analysis of the process. The tests were carried out at the Central Wastewater Treatment Plant (WWTP) of Pistoia (Central Italy).

CH3 CO3 H þ d OH -CH3 COd þ O2 þ H2 O: 2. Materials and methods The presence of hydrogen peroxide within the commercial product of the PAA contributes not only to the formation of new PAA as soon as it is consumed, but also to the formation of new hydroxyl radicals. This study is part of a larger research project aimed at studying the possibility of reusing municipal and industrial wastewater for irrigation purposes. In order to carry out this study, three experimental plants have so far been built at three different Wastewater Treatment Plants in Tuscany (Central Italy): (1) to evaluate the suitability of municipal wastewater for the irrigation of ornamental species; (2) to evaluate the suitability of municipal wastewater for the irrigation of fruits and vegetables; and (3) to study the possibility of reusing mixed municipal and industrial wastewater for the irrigation of ornamental species [9]. The different nature of the sewage treated by the plants and the different type of species irrigated makes the three individual experiments carried out both independent and particular. What these experiments have in common is the necessity to individualize an efficient disinfecting system of the secondary effluent, so that the result of the experiment is guaranteed from not only an agronomic but also a strictly sanitary point of view [10,11]. The particular attention paid to this aspect was necessary because of the fact that the irrigation is carried out in urbanized areas and, above all, due to the strict nature of microbiological limits that the Italian legislation requires for wastewater agricultural reuse (maximum concentration of 2 MPN/100 ml Total Coliforms for unrestricted irrigation). In previous laboratory studies, the possibility of using the combined treatment of PAA and ultraviolet irradiation (UV) to disinfect secondary wastewater effluents has been evaluated. The analyses carried out showed the

2.1. Experimental set-up In order to evaluate the synergy between PAA and UV, we have set up a pilot plant (flow rate ranging from 2 to 16 m3/h), fed by Pistoia’s WWTP secondary effluent. A scheme of the pilot plant is presented in Fig. 1. During the experimentation, we evaluated the effect of different solutions which can be summarized as follows: 1. disinfection with PAA; 2. disinfection with UV; 3. addition of PAA upstream the UV device (PAA+UV); 4. addition of PAA downstream the UV device (UV+PAA). In the third case, the contact time of PAA is negligible and equal to the detention time in the UV reactor, varying between 3 and 12 s according to the entrance capacity. However, when the PAA is introduced after the treatment with UV rays, we allowed for a contact time which, based on what was observed in the preliminary laboratory tests, has been calculated at 30 min. The pre-treatment of the secondary effluent before disinfection consists of two rapid filters (OFSY 30Culligan srl) whose technical characteristics are summarized in Table 1. PAA was added during the experimentation with doses ranging from 1 to 8 mg/l. In order to favour the mixing between PAA and water inside the pipes, a narrowing valve was positioned immediately after the point where the disinfectant was introduced.

C. Caretti, C. Lubello / Water Research 37 (2003) 2365–2371

PAA tank

Doser

2367

pumps

To the sewer Air compressor Secondary effluent Filter

Filter

Storage

UV

tank

To the irrigation system

To the sewer Backwashing Fig. 1. Pilot plant scheme.

Table 1 Technical characteristics of the double-stage filter OFSY 30 WGR Technical characteristics Exercise capacity (m3/h) Operating pressure (bar) Duration of wash cycle (min) Water volume for washing (m3) Size (mm3)

8–10 1.8–7 15–20 B6 1600  1130  1770

Table 2 Weight percentage composition Oxistrong 5 Peracetic acid Hydrogen peroxide Acetic acid Water and stabilizers

Table 3 Technical characteristics of the UV device Technical characteristics of Modello M86 Number of lamps Type of lamps Emission of lamps (W real UV ) Reactor diameter (mm) Reactor length (mm) Net volume of contact chamber (l) Hydraulic pressure (bar) Electric absorption (kWh)

8 Low pressure 20 163 792 13.62 8–12 0.8

2.2. Wastewater characteristics

5 28 8 59

The commercial product used was Oxistrong 5 (Ausimont Spa) whose weight percentage composition is listed in Table 2. The technical characteristics of the UV device used (Model M86 of the System line-Montagna Spa) are reported in Table 3. During the experimentation, the UV dosages ranged from about 100 mJ/cm2 to about 300 mJ/cm2. The water storage takes place in a horizontal zinc tank with a 10,000 l capacity. Having to remain for periods ranging from 2 to 3 h inside the storage tank before being sent to the irrigators or to filters backwashing, the water which, after disinfection, presents a bacterial content respecting the strict legislation regarding wastewater reuse for irrigation, risks being subject to bacterial regrowth. In order to avoid this phenomenon, 2 mg/l of PAA with a bacteriostatic function is dosed within the tank.

The principal chemical–physical and microbiological characteristics of the secondary effluent are summarized in Table 4. 2.3. Analytical methods Regarding the choice of microorganisms to be analysed, we took into account the legislation for wastewater reuse in agriculture and the vast bibliography regarding the subject. The methods of analysis used are the Multiple Tube Technique for Coliforms and Streptococci, the Membrane Filter Technique for Escherichia coli and Pseudomonas aeruginosa and the Pour Plate Method for the Etherotrophic Count. The indicator bacteria and the analytical methods used are summarized in Table 5. Because of the different detection limits of the analytical methods, in this study we made the absence of microorganisms correspond to the results of o1.1 MPN/100 ml for the Multiple Tube Technique and to the result of 0 CFU/100 ml for the other methods. 2.4. Experimental runs The experimental study has been carried out for 5 months and samples have been taken twice a week. For

2368

C. Caretti, C. Lubello / Water Research 37 (2003) 2365–2371

Table 4 Chemical–physical and microbiological characteristics of the effluent of Pistoia’s WWTP Parameter

Max.

Min.

Avg.

No. det.

pH Conductivity at 20 C (mS/cm2) Total solids (mg/l) Turbidity (NTU) COD (mg/l O2) BOD5 (mg/l O2) Calcium (mg/l Ca) Magnesium (mg/l Mg) Carbonates (mg/l CO3) Bicarbonates (mg/l HCO3) Alkalinity (mg/l CaCO3) Total phosphorous (mg/l P) Total ammonia (mg/l NH4) Nitrous nitrogen (mg/l N–NO2) Nitric nitrogen (mg/l N–NO3) Chlorides (mg/l Cl) Sodium (mg/l Na) Potassium (mg/l K) Iron (mg/l Fe) Manganese (mg/l Mn) Total Coliforms (MPN/100 ml) Faecal Coliforms (MPN/100 ml) Faecal Streptococci (MPN/100 ml) Escherichia coli (CFU/100 ml) Pseudomonas aeruginosa (CFU/100 ml) Heterotrophic plate count 22 C (CFU/ml) Heterotrophic plate count 37 C (CFU/ml)

8.1 779 30.0 2.9 75 36 96 39 o0.1 238 200 6.1 1.7 0.42 9.6 119 110 18 0.14 0.03 1.1  106a 1.1  106a 110,000 108,000 35,000 280,000 197,000

7.3 572 1.0 1.2 23 10 60 12 o0.1 209 35 0.7 0.5 0.01 5.2 46 95 12 0.06 o0.02 15,000 15,000 700 9600 0 7400 7600

7.7 709 7.4 1.9 37 22 80 20 o0.1 224 142 2.1 0.9 0.21 14.2 84 100 15 0.09 o0.02 307,000 25,700 24,000 36,500 8500 67,000 50,000

23 23 36 20 23 16 19 13 16 16 16 20 20 44 17 21 15 15 15 15 32 32 31 32 31 31 31

a

106 is 6 Logs unit.

Table 5 Indicator bacteria and analytical methods Indicator bacteria

Cultural media

Heterotrophic plate count 22 C Standard plate count agar Heterotrophic plate count 36 C Standard plate count agar E. coli T.B.X medium Pseudomonas aeruginosa Cetrimide agar Staphylococcus aureus Mannitol salt agar Total Coliforms Lauryl tryptose broth Faecal Coliforms Lauryl tryptose broth Faecal Streptococci Azide dextrose broth

Temp. ( C) Time (h) Cultural media

Temp. ( C) Time (h)

22 37 44 42 36 36 36 36

— — — — — 36 44 36

each experimental run we took three samples and we considered the average value of the three results. All the samples were processed within 48 h of sampling.

3. Results and discussion 3.1. PAA Table 6 shows the results of disinfection with PAA. The table indicates the average results, expressed as Log

72 72 24 48 48 48 48 48

— — — — — Brilliant green bile broth Brilliant green bile broth Ethyil violet azide broth

— — — — — 48 48 48

inactivation, obtained during the months of experimentation. Even though the inactivation levels obtained at the highest dosages are quite elevated (above 3 Log for the Total Coliforms) they are not, however, sufficient to ensure compliance with Italian legislation for wastewater reuse for irrigation. In fact, given the extreme variability of the concentration of Total Coliforms present in the secondary effluent (4–6 Logs, see Table 4), a disinfection system must be perfected which is capable of guaranteeing inactivations up to 6 Log.

C. Caretti, C. Lubello / Water Research 37 (2003) 2365–2371

2369

Table 6 PAA disinfection efficiency PAA dosage (ppm)

2 2 2 4 4 4 8 8 8

Contact time Average log inactivation (min) Total Faecal Coliforms Coliforms

Faecal Strept.

E. coli

Pseudomonas Heter. plate aeruginosa count 22 C

Heter. plate count 36 C

10 20 30 10 20 30 10 20 30

2.74 5.26 5.34 5.35 4.75 5.09 6 5.39 4.93

2.49 3.85 3.71 3.37 3.37 3.78 4.02 3.59 4.42

2.37 2.47 2.13 2.58 2.6 2.72 3.16 3.28 2.94

1.42 1.73 1.54 1.54 1.88 1.99 2.11 2.02 2.46

1.48 2.11 2.52 1.66 2.77 3.12 2.65 3.77 3.99

1.83 2.53 2.78 2.35 3.10 3.24 3.21 3.83 4.21

1.42 2.33 2.07 1.35 1.96 1.93 2.11 2.24 2.34

Table 7 UV disinfection efficiency UV dosage (mJ/cm2)

110 165 192 220 330

Average log inactivation Total Coliforms

Faecal Coliforms

Faecal Strept. E. coli

Pseudomonas aeruginosa

Heter. plate count 22 C

Heter. plate count 36 C

3.39 3.59 4.06 3.84 3.56

3.39 3.83 3.95 4.89 T.I.

3.27 4.52 T.I. T.I. T.I.

1.35 1.37 3.28 1.55 2.19

3.04 1.52 1.85 1.72 2.02

2.59 1.48 1.99 1.83 2.44

3.78 4.09 4.76 4.02 T.I.

T.I.: Total inactivation of the microorganisms.

By using PAA as the only disinfectant system, the problem of not reaching the Italian limit for wastewater reuse for unrestricted irrigation remains unsolved, according to the literature data, unless we resort to a drastic increase in the dosages [16–19]. We have verified that a dosage of 500 ppm of PAA with 30 min of contact time guarantees total inactivation of the bacteria. 3.2. UV Analogous results have been obtained when using only UV treatment. Table 7 shows the average results, expressed as Log inactivation, obtained during the months of experimentation. The high UV doses enable to inactivate only 3.39–4.06 Logs of Total Coliforms. Total inactivation of both Faecal Coliforms and E. coli was reached under the highest UV dose of 330 mJ/cm2. In the experimental conditions studied, according to the literature data, UV treatment proved to be insufficient in reaching the levels defined by Italian legislation for wastewater reuse for irrigation [20,21]. Furthermore, the lack of a guarantee regarding a bacteriostatic effect makes the addition of a chemical treatment necessary.

3.3. Combined treatment between PAA and UV In order to test the disinfection efficacy of the combined treatment between PAA and UV, we considered two different solutions: addition of PAA upstream the UV device (PAA+UV) and addition of PAA downstream the UV device (UV+PAA). The pilot plant results confirmed the results found in a previous laboratory study [12]: thanks to the synergy between PAA and UV, the PAA+UV treatment, which is based on radical reactions, is more efficient than the UV+PAA treatment. The proof of the effective action resulting from the development of radical reactions comes from the results of the tests carried out (Table 8). In fact, it resulted that the introduction of PAA before the UV guarantees, in equal dosages, significantly better results as compared to the case in which the disinfectant is introduced after the physical treatment. In fact, regarding this last case, the effectiveness of the combination of the two treatments simply results from the summation of the effects of the single disinfectants by direct action. However, when the PAA is introduced before the UV, the resulting effect goes beyond simple

C. Caretti, C. Lubello / Water Research 37 (2003) 2365–2371

2370

Table 8 Efficiency of disinfection with PAA and UV PAAa (ppm)

2 2 4 0 0 0

UV (mJ/ cm2)

165 192 192 165 192 192

PAAb (ppm)

0 0 0 2 2 4

Average log inactivation Total Colif. Faecal Colif.

Faecal Strept.

E. coli

Pseud. Heter. plate Heter. plate Aeruginosa count 22 C count 36 C

T.I. T.I. T.I. 4.57 3.43 5.6

T.I. T.I. T.I. T.I. T.I. T.I.

4.28 T.I. T.I. 4.28 T.I. T.I.

2.02 2.65 T.I. 2.32 1.56 T.I.

T.I. T.I. T.I. 4.03 3.43 T.I.

1.87 2.11 3.3 2.13 1.46 2.63

1.92 2.31 3.36 2.1 1.64 2.62

T.I.: Total inactivation of the microorganisms. a PAA introduced before the UV reactor. b PAA introduced after the UV reactor.

Fig. 2. Total Coliforms inactivation and isocost curves varying with dosages (costs expressed in Euro).

summation, which confirms a real synergy between the two treatments. The results obtained are compared in Table 8. Based on these results, the configuration which provides for the addition of PAA before the UV has been chosen in the final scheme of the pilot plant. We have also proceeded in extending the analytical research on different dosage configurations. The results obtained by the final pilot plant configuration, ranging the dosages of both PAA and UV, are as reported in Fig. 2. The figure refers to the Total Coliforms and represents the averages of the results obtained during the 5 month period of experimentation. As above mentioned, due to the high concentration of Total Coliforms in the secondary effluent and to its high variability, in order to ensure compliance with Italian

legislation for unrestricted wastewater reuse for irrigation, a disinfection system must be perfected which guarantees inactivations up to 6 Log. For the graphic representation we made correspond to 6 Log all inactivations X6 Log. The isocost curves, which have been superimposed on the inactivation graph, have been calculated so that we can compare the different solutions which guarantee compliance to Italian legislation for wastewater reuse for unrestricted irrigation. These curves have been calculated considering a continuous (24 h a day) running of the plant, no longer on a pilot scale, but on a real scale (capacity of 300 m3/ h). Costs include the disinfection system with PAA and UV (equipment, chemicals, energy) and the

C. Caretti, C. Lubello / Water Research 37 (2003) 2365–2371

addition of 2 ppm of PAA as a bacteriostatic agent in the storage tank.

4. Conclusions According to our results, neither the treatment with only PAA (under economically acceptable working conditions), nor the treatment with only UV is capable of guaranteeing the disinfection level required by Italian legislation for wastewater reuse in agriculture. Considering the combined treatment of the two disinfectant agents, the effectiveness increases. As confirmation of the actual synergy between the two treatments, the inactivation levels obtained with equal dosages are inferior when the PAA is added after the UV reactor as compared to the case in which it is added before. In this latter case, the higher efficiency is recognized as being brought about by the formation of free radicals due to the photolysis of the PAA when in presence of the UV rays. Considering a secondary effluent concentration of about 6 Logs MPN/100 ml Total Coliforms, in the final scheme of the pilot plant, the more economically competitive solution, which guarantees compliance to the strict limits of Italian legislation for wastewater reuse for unrestricted irrigation, is by the disinfection with 2 ppm of PAA and 192 mJ/cm2 UV. The results of our study can be considered as preliminary results and further research is being conducted in order to verify the effect of such a disinfection system on other pathogenic agents (Protozoa and spores), bacteria reactivation and the possible risk of unwanted by-products formation.

References [1] Braun AM, Oliveros E. How to evaluate photochemical methods for water treatment. Water Sci Technol 1997;35(4):17–23. [2] Luck F, Djafer M, Karpel Vel Leitner N, Gambert B, Legube B. Destruction of pollutants in industrial rinse waters by advanced oxidation processes. Water Sci Technol 1997;35(4):287–92. [3] Bolton JR. Use of the UV/O3 and UV/H2O2 advanced oxidation technologies for the treatment of contaminated waters. In: Proceedings of the 15th Ozone World Congress Conference, London, United Kingdom, 11–15 September 2001. [4] Kleiser G, Frimmel FH. Reduction of bacterial growth after oxidation by using AOP. Water Supply 1998; 16(3–4):209–10. [5] Feitz AJ, Aplin R, Waite DT. Advanced oxidation processes and potential applications. Water 1999;26(1):34–5. [6] Crittenden JC, Hu S, Hand DW, Green SA. A kinetic model for H2O2/UV process in a completely mixed batch reactor. Water Res 1999;33(10):2315–28.

2371

[7] Hoign!e J. Inter-calibration of OH radical sources and water quality parameters. Water Sci Technol 1997;35(4): 1–8. [8] Rajala-Mustonen RL, Toivola PS, Heinonen-Tanski H. Effects of peracetic acid and UV irradiation on the inactivation of coliphages in wastewater. Water Sci Technol 1997;35(11–12):237–41. [9] Gori R, Lubello C. Pilot plant for reclaimed wastewater reuse in nurseries. Water Sci Technol 2000;42(1–2): 221–6. [10] Mujeriego R, Asano T. The role of advanced treatment in wastewater reclamation and reuse. Water Sci Technol 1999;40(4–5):1–9. [11] Shuval HI, Yekutiel P, Fattal B. An epidemiological model of the potential health risk associated with various pathogens in wastewater irrigation. Water Sci Technol 1986;18(10):191–8. [12] Lubello C, Caretti C. Wastewater disinfection with peracetic acid and UV. In: Proceedings of the International Symposium on Sanitary and Environmental Engineering, Trento, Italy, 18–23 September 2000. [13] Lubello C, Caretti C, Gori R. Comparison between PAA/ UV and H2O2/UV disinfection for wastewater reuse. In: Proceedings of the International Conference on Innovations in Conventional and Advanced Water Treatment Processes Conference, Amsterdam, The Netherlands, 26–29 September 2000. [14] Roubaud V, Sankarapandi S, Kuppusamy P. Quantitative measurements of superoxide generation using the spin trap 5,5 dimethyl-1-pyrroline-N-oxide (DMPO). Anal Biochem 1997;247:404–11. [15] Bianchini R, Calucci L, Lubello C, Pinzino C. Intermediate free radicals in the oxidation of wastewaters. Res Chem Intermediates, 2002;28(2,3). [16] Baldry MGC, French MS, Slater D. The activity of peracetic acid on sewage indicator bacteria and viruses. Water Sci Technol 1991;24(2):353–7. [17] Lazarova V, Janex ML, Fiksdal L, Oberg C. Advanced wastewater disinfection technologies: short and long term efficiency. Water Sci Technol 1998;38(12): 109–17. [18] Langlais B, Triballeau S. Tests on the disinfection of a secondary effluent using peracetic acid. In: Proceedings of the Water Management in Coastal Areas Conference, Montpellier, France, 31 March–2 April 1992. [19] Liberti L, Notarnicola M. Advanced treatment and disinfection for municipal wastewater reuse in agriculture. In: Proceedings of the Advanced Wastewater Treatment, Recycling and Reuse Conference, Milano, Italy, 14–16 September 1998. [20] Lyn AD, Chiu K, BlatchleyIII ER. Numerical modeling of flow and disinfection in UV disinfection channels. J Environ Eng 1999;125(1):17–26. [21] Moreno B, Goni F, Fernandez O, Martinez JA, Astigarraga M. The disinfection of wastewater by ultraviolet light. Water Sci Technol 1997;35(11–12):233–5. [22] Von Sonntag C, Dowideit P, Fang X, Martens R, Pan X, Schuchmann MN, Schuchmann HP. The fate of peroxyl radicals in aqueous solution. Water Sci Technol 1997;35(4):9–15.