Optimization of continuous reactor at pilot scale for olive-oil mill wastewater treatment by Fenton-like process

Chemical Engineering Journal 220 (2013) 117–124

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Optimization of continuous reactor at pilot scale for olive-oil mill wastewater treatment by Fenton-like process Gassan Hodaifa a,⇑, J.M. Ochando-Pulido b, S. Rodriguez-Vives b, A. Martinez-Ferez b a b

Molecular Biology and Biochemical Engineering Department, Pablo de Olavide University, 14013 Seville, Spain Chemical Engineering Department, Granada University, 18071 Granada, Spain

h i g h l i g h t s " OMW reclamation from two-phase process was carried out in a CSTR by Fenton system. " The optimum [FeCl3]/[H2O2] ratio was in the range 0.026–0.058 w/w. 3+

" [Fe ] = 0.35–0.40 g dm

3

is necessary to achieve a higher COD removal (>97%).

" Water treated can be used for irrigation or discharged into urban wastewater system.

a r t i c l e

i n f o

Article history: Received 11 August 2012 Received in revised form 6 December 2012 Accepted 19 January 2013 Available online 30 January 2013 Keywords: Olive-oil mill wastewater CSTR Pilot plant Fenton reaction Flocculation

a b s t r a c t Reclamation of olive-oil mill wastewater (OMW) from two-phase extraction procedure was carried out in a continuous stirred tank reactor (CSTR) at a pilot plant scale by Fenton-like process. The effect of operating conditions such as pH, space–time, H2O2 and Fe(III) doses, as well as [FeCl3]/H2O2] ratio on the efficacy of Fenton’s process was investigated. It is demonstrated that Fenton’s process can effectively degrade organic matter in OMW. In the start-up stage, Fenton reaction reached steady state within 3 h. Oxidation of organic materials in OMW was pH dependent and the optimal pH was found to be 3. The optimum [FeCl3]/[H2O2] ratio was in the range 0.026–0.058 w/w, with Fe(III) concentration between 0.35 and 0.40 g dm3. The final values of COD and total phenols at the outlet of the pilot plant were close to 129 mg O2 L1 ([COD]initial = 4017 mg O2 dm3), and 0.5 mg dm3 ([total phenols]initial = 66.2 mg dm3), respectively. Finally, the produced water can be used for irrigation or discharged directly into the municipal wastewater system for ulterior tertiary treatment. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction Olive-oil mill wastewater (OMW) poses a serious environmental problem in the Mediterranean basin and concretely in the southern European countries like Spain, Italy, Greece and Portugal, where most part of the world olive oil is produced [1]. OMW effluents are well known by their seasonality and toxic character due to the presence of phenolic compounds, which are not suitable to be biologically managed [2]. Within this context, the application of chemical remediation strategies is required either to fulfil legislative requirements for direct disposal into the surroundings or, when economically wiser, to reduce toxicity and improve biodegradability to allow a posterior inexpensive bioprocess [3]. For these reasons, attention will be paid not only to chemical degradation parameters (such as chemical oxygen demand COD) but

⇑ Corresponding author. Tel.: +34 954 978 206; fax: +34 954 349 813. E-mail address: [email protected] (G. Hodaifa). 1385-8947/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2013.01.065

also to effluent’s toxicity and biodegradable character (phenoliccompounds) [4]. Advanced oxidation processes (AOPs) are known for their capability to mineralise a wide range of organic compounds. AOPs involve the generation of highly reactive radical species, mainly hydroxyl radical [5]. AOPs’ versatility is enhanced by the fact that there are many different ways of producing hydroxyl radicals. Heterogeneous photocatalysis using titanium dioxide (TiO2) and solar UV, possibly combined with hydrogen peroxide (H2O2), and homogeneous processes such as Fenton (Fe2+/H2O2) and photo-Fenton (Fe2+/H2O2/UV) reactions have been proved to be useful tools for the treatment of pesticide-containing wastewaters [6]. Among AOPs, Fenton and solar photo-Fenton are known within the scientific community for their effectiveness to treat wastewater with high phenols content, such as that from the cork manufacturing [7], olive oil [3,8], winery [9] and pulp and paper industries [10]. The advantages of the Fenton’s reagents are the high efficiency, simplicity, the lack of residues and capacity to treat many different compounds. In addition, it can be used as a pre-treatment stage be-

118

G. Hodaifa et al. / Chemical Engineering Journal 220 (2013) 117–124

fore the biological step in order to increase the biodegradability of the recalcitrant compounds and thus lower the toxicity of the wastewaters [11]. From the industrial point of view, Fenton’s process, operating at room pressure and temperature conditions, is economically preferable taking in regard avoidance of operational costs associated with the reactor’s heating up. On the other hand, simplicity in both equipment and operation has postulated Fenton’s reagent as one of the most economic alternatives for treating those effluents [12,13]. A wide variety of Fenton’s reagent applications have already been reported in scientific literature, such as textile wastewater [14], laboratory wastewater [15], industrial effluents [16], cosmetic wastewater [17], dye wastewater [18], pesticides [16], fermentation brine from green olives [19], pharmaceutical wastewater [20], cork cooking wastewater [7], pulp mill effluents [21], and phenol degradation [22]. In all cases, researcher’s pin-point for a percentage of residual hydrogen peroxide remaining unreacted at the outlet of the oxidation reactor or at the exit of the treatment plant. Moreover, the [catalyst]/[H2O2] ratio is a very important variable in the effectiveness of the final degradation of the process [11]. In the literature it can be found enough data on the degradation of organic matter by Fenton’s reagent, but all these works have been carried out in batch reactor just determining the main operating conditions affecting the chemical oxidation. To the best of the author’s knowledge, there are only a few studies available about Fenton’s reagent application in continuous reactors e.g., [23–25], and none were found about the degradation of OMW from continuous two-phase decanting processes nor the behavior of pilot-scale continuous reactors of Fenton’s process. In this work, OMW from the two-phase decanting process of olive-oil industry (wastewaters from olives and olive-oil washing) was treated in a continuous stirred tank reactor (CSTR) at a pilot plant scale by adding iron (III) chloride (FeCl3) and hydrogen peroxide (H2O2) into the reactor. The effects of reaction conditions (i.e., the presence of the catalyst, pH, Fe(III) and H2O2 doses or catalyst/H2O2 ratio, as well as reaction time) in the degradation of organic matter (COD) and phenolic compounds, together with residual H2O2 and final total iron were investigated. The optimization of continuous reactor operation and post flocculation using a lamellar settler were additionally addressed. 2. Experimental 2.1. Chemicals The different reagents used in this work were phenol (CAS Number 108-95-2), C6H5OH 99% (w/w) provided by Sigma– Aldrich, and hydrogen peroxide (CAS Number 7722-84-1), H2O2 (30% w/w), sodium hydroxide (CAS Number 1310-73-2), NaOH 98% (w/w), hydrochloric acid (CAS Number 7647-01-0), HCl 37% (w/w), iron (III) chloride (CAS Number 7705-08-0), FeCl36H2O, 30% (w/w) as catalyst, all them provided by Panreac S.A, and finally Nalco 77171 as flocculant (anionic polyelectrolytes oil-based) purchased from Nalco España S.A., Barcelona (Spain). 2.2. Wastewaters (OMW) In previous studies, Nieto et al. [3] had already indicated that, as a rule, olive-oil mill wastewater from olives washing does not exceed the permitted limits regarding its reuse for irrigation (Spanish environmental standards for use as irrigation water: pH = 6–9, suspended solids < 500 mg kg1, and COD < 1000 mg O2 dm3 [26]). Moreover, pollutants concentrations can be controlled by regulating the water volume, the interval time of use, or the amount of

washed olives per water load in the washing machines. Meanwhile, in the case of water from olive-oil washing, reference values are widely exceeded, with the exception of suspended solids load. The phenolic content in olives-washing wastewater was practically zero, as it could only come from over-ripe, broken, or crushed olives. On the other hand, in the olive-oil-washing wastewater, coming from vertical centrifugation, the contact between the olive-oil and the water enhances phenolic compounds transfer from the oil phase to the water phase, in which they are more soluble, according to their distribution coefficient. Pollutants concentrations found were consistent with those expected in OMW deriving from two-phase decanting process, in contrast with OMW from three-phase decanting process, which is more concentrated [27,28]. OMW used in this research was collected from several olive-oil mills (operating with the two-phase decanting process) in the provinces of Granada (Spain), and the samples from olives and olive-oil washing wastewaters were mixed in the laboratory in 1:1 v/v) proportion to regulate the initial value of COD (4017 mg O2 dm3, Table 1) in the oxidation reactor to get a final treated water at the end of the process complying with the requirements for water irrigation [26]. 2.3. Description of the pilot plant The pilot plant used in this work was located on the premises of the Department of Chemical Engineering at the University of Granada. The process undertaken at the plant consisted of the following: 1. Chemical oxidation was carried out in a continuous stirred tank reactor (CSTR). The reactor (vertical cylindrical vessel with flat bottom) was 0.16 m in diameter and its overall height was 0.38 m. The working volume and the overall volume of the reactor were 7.0 dm3 and 7.4 dm3, respectively. The static liquid height was 0.36 m. 2. Neutralization was carried out in a stirred tank similar to the reactor described above (with 7 dm3 capacity). 3. Solid–liquid separation was conducted in a lamellar settler with 22 dm3 capacity. 4. Six storage tanks: (i) for OMW (32 dm3 capacity), (ii) for the oxidant reagent (3 dm3 capacity), (iii) for the catalyst (3 dm3 capacity), (iv) for the neutralizing agent (3 dm3), (v) for the preparation of the flocculant provided with agitation system (1 dm3 capacity), and (vii) for the final treated water (50 dm3 capacity). 5. Nine peristaltic pumps. 6. Pneumatic level sensors. 7. Three stirring system formed by overhead stirrers installed in the oxidation reactor, neutralization tank, and flocculant storage tank. The stirring equipment operating at 60 rpm consists of three impellers. A three-bladed mixing propeller was placed above

Table 1 Characterization of olive-oil mill wastewater (mixture of olives and olive-oil washing wastewaters, 1:1 v/v) from two-phase process used in the oxidation experiments. Parameter

Average value

pH Electric conductivity (lS cm1) [Fe]Total (mg dm3) Total phenols (mg dm3) COD (mg O2 dm3)

6.1 1410.0 7.9 66.2 4017.0

G. Hodaifa et al. / Chemical Engineering Journal 220 (2013) 117–124

the bottom of the reactor, and others two turbines with four blades inclined 45° above were installed in the shaft. Each impeller was 0.05 m in diameter, the vertical distance between the impellers was 0.12 m and the lower impeller was located 0.05 m from the bottom of the reactor. This stirring system ensures perfect mixture in the reactor with the molecular oxygen generated by Fenton’s reaction. 8. Programmable logic controller (PLC) Once the oxidation and neutralization tanks were fed with wastewater to be treated (OMW), the reactants (oxidant and catalyst), neutralizer (sodium hydroxide) and flocculant, as appropriate, were pumped by means of peristaltic pumps. H2O2 (5% w/v) and FeCl3 (8% v/v) were used as oxidant and catalyst in Fenton’s reaction, respectively. Neutralizing agent (NaOH 1 M) for pH adjustment and flocculant (1 mg dm3 of Nalco 77171) were added into the neutralization tank [3,29,30]. The facility is provided with a programmable logic controller, whereby the corresponding dosages in the tanks are adjusted, as well as the agitation and the liquid level therein. OMW passes from the reaction unit to the neutralization-flocculation unit by overflow and so does it from this latter to the sedimentation settler and then to the final treated water storage tank. The beginning of the experiments on continuous mode was carried out after having reached steady state (maximum degradation degree of organic matter) in the batch step (start-up stage). Experiments were performed continuously for periods of 10 h, taking samples regularly to assess the ability of the depuration process operating in continuous. The basic variables of the pilot plant operation are: pH, [FeCl3]/[H2O2] ratio, and temperature. 2.4. Procedure For operation of the CSTR, once the wastewater entered the reactor, an established amount of ferric chloride was added continuously and then the hydrogen peroxide solution, keeping agitation at 60 rpm [3]. For all experiments of this work, the reaction in the first step was conducted during three hours (start-up of continuously-operated chemical reactor). The end of start-up stage of the CSTR was considered as zero time to begin continuous operation (second step), with a mean residence time in the reactor of four hours, as by this way of operation the residual oxidant concentration in the reactor is highly enhanced and therefore the ability to degrade the organic matter is promoted. Moreover, all experiments were extended until ten hours, making sure that the mixture had reached constant concentrations. The space–time, s (h), in the CSTR is equal to the mean residence time. Space time in the continuous reactor was calculated from the following equation: 3

s¼

Vðvolume of reactor; dm Þ 3

1

Q ðvolumetric flow rate; dm h Þ

ð1Þ

During the experiment, samples were taken at regular intervals (0.5 h) during the 10 h of operation. To evaluate the ability of the Fenton-like process regarding OMW treatment, it is necessary to know the influence of the variables that control the oxidation process: catalyst, pH, s, [FeCl3]/ [H2O2] ratio, and temperature (in this work, temperature was considered as a variable imposed by the external environment, hence it is equal to ambient temperature). Four sets of experiments were carried out. In the first set, the effect of the catalyst on the oxidation was studied. With this purpose, two experiments were performed, without and with catalyst ([Fe3+] = 0.04 g dm3) in the reactor. In both experiments the

119

operating conditions were the following: hydrogen peroxide, [H2O2]initial = 30 g dm3, s = 4 h, pH unadjusted, and ambient temperature. In the second set, the influence of pH was addressed. In these experiments the operating conditions were: [Fe3+]initial = 0.04 g dm3, [H2O2]initial = 20 g dm3, s = 4 h, ambient temperature and the pH was varied as follows: unadjusted, 3 and 7. In the third set, the space time was studied, for which two different s values were used. For this series the operating conditions were [Fe3+]initial = 0.2 g dm3, [H2O2]initial = 20 g dm3, and s values used were 3 and 4 h. In the fourth set, the [FeCl3]/ [H2O2] ratio influence was investigated. With this intention, the [FeCl3]/[H2O2] ratio was varied using the following values: 0.006, 0.026, 0.029, 0.039, 0.058, and 0.26 w/w. These values were selected based on the study by Nieto et al. [3] on the best [FeCl3/ H2O2] ratio (the values were varied between 0.1 and 1.5 w/w) in a batch reactor, in which decreasing final COD and total phenols values in the treated OMW with increasing [FeCl3]/[H2O2] ratio from 0.1 to 0.25 was highlighted. Finally, in order to optimize the use of the catalyst and the oxidizing agent, the evolution of the following parameters was determined along the experiment: pH, H2O2 residual, phenolic compounds and COD. 2.5. Analytical methods The OMW used in this investigation was characterized by measuring pH, electric conductivity, total iron concentration, total phenols, chemical oxygen demand (COD) and residual hydrogen peroxide concentration. The pH-values were measured using a CRISON pH-meter, LPG 21 model. Electric conductivity was assessed with a CRISON conductivity meter, GLP31 model. All iron ions were reduced to iron ions (II) which, in a thioglycolate medium with a derivative of triazine, formed a reddish-purple complex that was determined photometrically at 565 nm [31,32]. Total phenols and phenol derivatives reacted with a thiazol derivative, giving a purple azo dye, which was determined photometrically at 475 nm [31,32]. Chemical oxygen demand (COD) was measured by the photometric determination (620 nm) of the concentration of chromium (III) after 2 h of oxidation with potassium dichromate/sulphuric acid/silver sulfate at 421 K [33]. At the time of the determination of COD, manganese oxide was used to remove residual hydrogen peroxide which remained unreacted in the samples taken from the reactor [34]. Concentration of residual hydrogen peroxide was determined using the colorimetric method, by means of titanium sulfate, in virtue of its simplicity and accurate measurement. Titanium sulfate reacted with the H2O2 present in the solution, forming a yellow complex with a maximum absorbance around 410 nm [35]. Finally, all the experiments were made at least in duplicate and the analytical methods were applied at least in triplicate. The calculation and statistical methods used are available in the OriginPro 8.0 program. 3. Results and discussion 3.1. Effect of catalyst on OMW degradation In the first experiment, proposed to evaluate the oxidizing power of hydrogen peroxide in the absence of catalyst in continuous operation mode, as happened in the start-up stage, the percentage of COD removal was found to be less than 67%, whereas

120

G. Hodaifa et al. / Chemical Engineering Journal 220 (2013) 117–124

in the presence of small amounts of catalyst ([Fe3+] = 0.04 g dm3) this value exceeded that percentage (74%). In contrast, phenolic compounds were observed to maintain the same reduction (99%), something that happened in all continuous experiments, being these compounds the first ones to be degraded in Fenton’s reaction mechanism. In this sense, it should be recalled the fact that Fenton’s oxidation of OMW organic matter can be represented by two-step first order kinetic model [36,37], that is oxidation upon two first-order sequential steps: k1st

step

COD þ H2 O2 ! Products Products þ H2 O2

k2nd

ð2Þ

step

! CO2 þ H2 O þ inorganic salts

pH

Unadjusted 3 7

[H2O2]initial (g dm3)

[Fe3+]initial (g dm3)

20 20 20

0.04 0.04 0.04

Total phenolsfinal (mg dm3)

CODfinal (mg O2 dm3)

0.5 ± 0.001 0.2 ± 0.002 0.7 ± 0.002

527 ± 10 802 ± 13 910 ± 18

In terms of percentage of total phenols removal, this value was above 99% in all cases and at both stages of the CSTR.

The effect of the initial pH on the degradation of OMW throughout Fenton’s reaction in CSTR was addressed. Fig. 1 shows the effect of the initial pH in the removal percentages of total phenolic compounds and COD at the outlet of CSTR. In this work, higher removal percentage of COD (72%) was observed within Fenton’s oxidation when adjusting pH to a value about 3, or left free without adjustment, but at pH = 7 this percentage dropped to 67%. This behavior was observed by other authors, as it is usually accepted that acidic pH levels near 3 are usually optimum for Fenton’s oxidation [23,24,38]. At pH above 3.5 performances significantly decreased, mainly because the dissolved fraction of iron species decreases [39]. Actually, at high pH values Fe(III) precipitates, therefore decreasing the concentration of dissolved Fe(III). Consequently, the concentration of Fe(II) species also decreases because iron (III) hydroxides are much less reactive than dissolved Fe(III) species towards H2O2. The process performance is therefore affected because a smaller steady-state concentration of hydroxyl radicals is attained. In the case of unadjusted pH, the pH of the mixture in the reactor decreases reaching values close to 3, due to the use of FeCl3 as catalyst, salt of a strong acid, and also owed to the mineralization of the phenolic compounds into simpler compounds, including acids [2]. Similar results (Table 2) were obtained at the end of start-up stage of CSTR (%[COD]removal > 80% when pH = 3 or unadjusted, and %[COD]removal < 78% at pH = 7).

3.3. Effect of space–time on OMW degradation Fenton-like process for OMW degradation consisted in two steps: in the first step, the COD values are reduced practically instantly, and in the second step, COD values are slowly diminished until equilibrium [36,37], which implies that space–time in the reactor is a key parameter to be considered. The study of the space time was carried out by performing experiments with two different space–time values (3 and 4 h), and with the following initial operating conditions: [H2O2] = 20 g dm3, [Fe3+] = 0.2 g dm3, pH unadjusted, and ambient temperature. The results obtained suggest that there is little difference between operating with a space–time of 3 h or 4 h, since the percentages of organic matter removal are similar (80.3% and 81.1%, respectively) and the residual oxidant in the reactor as well (5194 and 5110 mg dm3, respectively). Similar results were obtained by Zhang et al. [25] but in the case of landfill leachate treatment by Fenton’s reagent in a continuous stirred tank reactor. By reducing the space–time in the reactor (Fig. 2), it can be noted that the profile of the evolution of the oxidant concentration is similar for both the space–times used, reaching a concentration peak (at 2 h) and then stabilized until the end of the experiment. From the foregoing, it can be concluded that operating upon space–time equal to 3 h increases the productivity of the plant. However, it is necessary to remember that operating with smaller

%Total phenolsremoval

100

Initial conditions of the experiments

ð3Þ

3.2. Influence of initial pH on OMW degradation

110

Table 2 Effect of pH on the total phenolic content and COD values determined at the end of the startup stage of CSTR. Common operation conditions: [COD]initial = 4017 mg O2 dm3, [total phenols]initial = 66.2 mg dm3, space–time = 4 h, and ambient temperature.

12000

τ = 4 h,

70

τ=3h

10000

H2O2, mg dm

-3

60

% COD

50 40 30

8000 6000 4000

20 2000 10 0 0 Unadjusted

3

7

pH Fig. 1. pH effect in the removal percentages of COD and total phenols from OMW using Fenton’s system at the outlet of CSTR. Operation conditions: [COD]ini3 , [total phenols]initial = 66.2 mg dm3, [H2O2]initial = 20 tial = 4017 mg O2 dm g dm3, [Fe3+]initial = 0.04 g dm3, s = 4 h, and ambient temperature.

0

2

4

6

8

10

12

t, h Fig. 2. Comparison between the evolution profiles of [H2O2] concentration in CSTR at different space–times (3 h or 4 h). Common operation conditions: [COD]ini3 , [H2O2]initial = 20 g dm3, [Fe3+]initial = 0.2 g dm3, pH unadtial = 4017 mg O2 dm justed, and ambient temperature.

121

G. Hodaifa et al. / Chemical Engineering Journal 220 (2013) 117–124 Table 3 Percentages of COD and total phenols removal determined at the outlet of CSTR. Common operating conditions: space–time 4 h, pH unadjusted, and ambient temperature. Initial conditions of the experiments [H2O2] (g dm3)

[Fe3+] (g dm3)

[FeCl3]/[H2O2] (w/w)

20.0 45.0 20.0 15.0 17.6 45.0

0.04 0.40 0.20 0.20 0.35 4.00

0.006 0.026 0.029 0.039 0.058 0.260

-3

[ H2O2], mg dm

Previously, it has been shown that the presence of catalyst in the reactor improves the degradation degree of OMW in comparison with the simple action of the oxidant. The relationship between the catalyst and the oxidizing agent in the reactor is crucial to achieve a high degree of degradation, additionally to the viability of the process on an industrial scale. In Fenton’s process, operating costs (represented by the costs of the catalyst and the oxidizing reagent used) are the main costs when selecting and implementing this procedure for wastewater treatment. With this regard, experiments were performed to achieve optimal, so as to accomplish a sufficient degradation degree for reuse of the produced water as well as to avoid leaving too much residual reactants in the treated water with no contribution in the process. Table 3 shows the operating conditions of the experiments performed to determine the optimum reactants ratio. Fig. 3 shows the evolution with time of the net change in oxidant concentration (DH2O2 = H2O2 residual in CSRT-H2O2 at the end of startup step) and COD concentration (DCOD = COD in CSTR–COD at the end of startup step) in the reactor from the very first moment in which it starts to operate continuously. In general, this figure shows that the net change in the values of DH2O2 and DCOD decreases with increasing the catalyst/oxidant ratio. In other words, the degradation of organic matter and in particular of phenolic compounds (data not shown) show a sharp decrease in their values within the reactor, practically upon introducing the wastewater stream (OMW) into the reactor. This corresponds to the first stage of Fenton-like reaction, Eq. (1), which is achieved by further degradation of the organic matter, in an infinitesimal time [37]. However, this observation is not fulfilled (for both DH2O2 and DCOD values) in the case of catalyst/oxidant ratios corresponding to 0.026, 0.029 and 0.039. This fact can be justified taking into account the iron concentrations ([Fe3+] = 400, 200, and 200 mg dm3, respectively) used in each [FeCl3]/[H2O2] ratio applied. In Fig. 4, the influence of Fe(III) concentration in the net change values of H2O2 and COD is studied. With this goal, experiments where catalyst/oxidant ratio were set to remain constant (equal to 0.058) were carried out, varying Fe(III) concentrations as follows: 286, 351, and 400 mg dm3. The importance of the Fe(III) concentration in COD removal as well as in residual hydrogen peroxide concentration in the reactor is clearly evidenced (Fig. 4). This implies that the relationship between the catalyst and the oxidant concentrations is not sufficient to optimize the process, but the optimization of the Fe(III) concentration once identified the optimum ratio of catalyst and oxidant must be considered, too.

% Total phenolsremoval

71.6 ± 1.9 93.4 ± 0.9 81.2 ± 0.7 76.4 ± 0.8 91.3 ± 1.6 91.2 ± 0.5

99.3 ± 0.2 99.0 ± 0.3 99.6 ± 0.2 99.3 ± 0.2 99.7 ± 0.2 99.8 ± 0.1

20000

15000

[FeCl3]/[H2O2] 0.006 0.026 0.029 0.039 0.058 0.26

10000

5000

0 0

(b)

2

4

6

8

10

12

14

t, h 800

600 -3

3.4. Effect of catalyst/oxidant ratio on OMW degradation

(a)

[ COD], mg O 2 dm

space–time values (s < smin = 3 h where you get the ultimate degradation) may produce instability in the efficiency of the reactor in the virtual case of errors (e.g., inflow of OMW to the reactor).

% CODremoval

400

[FeCl3]/[H2O2] 200

0.006 0.026 0.029 0.039 0.058 0.26

0

-200

-400 0

2

4

6

8

10

12

14

t, h Fig. 3. Behavior of the net change in the values of DH2O2 (DH2O2 = H2O2 residual in CSRT-H2O2 at the end of startup step) and DCOD (DCOD = COD in CSTR–COD at the end of startup step) with operation time in CSTR. Experiment conditions: [FeCl3]/ [H2O2] ratios studied were 0.006, 0.026, 0.029, 0.039, 0.058, and 0.26 w/w. Common operation conditions: [COD]initial = 4017 mg O2 dm3, pH unadjusted, space time = 4 h, and ambient temperature.

Raised this question, we represented all Fe(III) concentrations used in the whole sets of experiments (including data from experiments not included in this study) versus the percentages of COD and total phenols removal (Fig. 5). This figure points for a linear relationship (solid line in Fig. 5) between COD removal and Fe(III) concentrations in the range of 0–400 mg dm3 (corresponding to catalyst/oxidant ratio ranging from 0 to 0.058), and from this concentration on (until 4 g dm3) iron (III) does not exhibit any influence on the value of removed COD. In terms of total phenols

122

G. Hodaifa et al. / Chemical Engineering Journal 220 (2013) 117–124

% total phenolsremoval

(a)

6000 5000

100 95 90 85

3000 2000 1000

[FeCl3]/[H2O2] = 0.058

0

3+

3+

[Fe ] = 286,

3+

[Fe ] = 351,

[Fe ] = 400

80

% CODremoval and

[ H2O2], mg dm

-3

4000

105

75 70 65 0

0

(b)

2

4

6

8

10

12

500

1000

1500

14

2000

2500

3+

3000

3500

4000

-3

[Fe ], mg dm

t, h

Fig. 5. Influence of Fe(III) in the %[COD] and %[total phenols] removed from OMW treated by Fenton-like reaction. Experiment conditions: Fe(III) concentrations: 0– 4000 mg dm3, and H2O2 concentrations: 14,365–45,000 mg dm3. Common operation conditions: [COD]initial = 4017 mg O2 dm3, pH unadjusted, space time = 4 h, and ambient temperature.

400

200

7

100

6

0

5

3+

4

[FeCl3]/[H2O2] = 0.058

-100

3+

3+

[Fe ] = 286,

3+

[Fe ] = 351,

[Fe ] = 400

3 0

2

4

6

8

10

12

-3

[Fe ], mg dm 0 40 200 400 4000

pH

[ COD], mg O 2 dm

-3

300

14

t, h Fig. 4. Comparison between the behavior of the net change in the values of DH2O2 (DH2O2 = H2O2 residual in CSRT-H2O2 at the end of startup step) and DCOD (DCOD = COD in CSTR–COD at the end of startup step) at different Fe(III) concentrations and constant catalyst/oxidant ratio (0.058 w/w) in CSTR. Experiment conditions: Fe(III) concentrations: 286, 351, and 400 mg dm3, and H2O2 concentrations: 14365, 17600, and 20,000 mg dm3, respectively. Common operation conditions: [COD]initial = 4017 mg O2 dm3, pH unadjusted, space time = 4 h, and ambient temperature.

elimination percentages, there was a degradation of more than 99% in all cases (Fig. 5). The above results show that excess reagents in the reactor do not contribute satisfactorily to the COD removal from OMW. Thus, when the oxidizer is in excess, it passes through the CSTR without reacting, and high residual oxidant concentrations were registered in the water treated at the outlet of reactor (Fig. 3), whereas when high catalyst concentrations were attained in the reactor, decomposition reactions which do certainly not contribute to the oxidative degradation of organic matter were promoted ([Fe3+] > 0.058, Fig 5). The optimum ratio between the catalyst and the oxidizer was in the range 0.026–0.058, with a Fe(III) concentration between 0.35 and 0.40 g dm3 (Table 3), upon which higher degradation of present contaminants is reached, also implying a better use of the reagents placed in the reactor (Fig. 3). Moreover, it is worth pointing that at an industrial level acidification in the reactor to adjust the pH to 3 is not required (Fig. 6). The catalyst addition actually produces a lowering of pH to values next to 3 (when [FeCl3]/ [H2O2] > 0.01, [3]), which is the optimum pH value.

2 1 0 0

2

4

6

8

10

12

14

t, h Fig. 6. pH evolution in CSTR reactor determined at different Fe(III) concentrations. Experiment conditions: [FeCl3]/[H2O2] ratios studied were (j) 0, (s) 0.004, (N) 0.039, (h) 0.026, and (d) 0.26 w/w. Common operation conditions: [COD]ini3 , pH unadjusted, space time = 4 h at [FeCl3]/[H2O2] ratios = 0, tial = 4017 mg O2 dm 0.004, and 0.039 w/w and 3 h at [FeCl3]/[H2O2] ratios = 0.026 and 0.26 w/w, and ambient temperature.

Finally, note that the profile of residual hydrogen peroxide in the reactor is related to the iron (III) concentration, in a manner that when operating without catalyst, the residual hydrogen peroxide (unreacted) increases (Fig. 3a), and with the catalyst introduction this residual hydrogen peroxide begins to react, improving the degradation degree of the organic matter in OMW and thus achieving COD values lower than those obtained in the startup of the CSTR (Fig. 3b).

3.5. Neutralization and solid–liquid separation by lamellar settler After the CSTR, the stream to be further treated (treated OMW) passes through the neutralization–flocculation unit, where the pH is adjusted by a PLC using NaOH to target values within the range 6–7, and the flocculant (1 mg dm3 of Nalco 77171) is as well

123

G. Hodaifa et al. / Chemical Engineering Journal 220 (2013) 117–124

Table 4 Characterization of the treated water (OMW treated) at the outlet of the pilot plant. Common operating conditions: space–time 4 h, pH unadjusted, and ambient temperature. Initial experimental operating conditions 3+

Average values obtained at the outlet of the pilot plant

[H2O2]initial (mg dm3)

[Fe ]initial (mg dm3)

[FeCl3]/ [H2O2] (w/w)

[H2O2]residual (mg dm3)

[Total Fe]residual (mg dm3)

CODfinal (mg O2 dm3)

Total phenolsfinal (mg dm3)

30,000 30,000 20,000 20,000 15,000 17,322 45,000

0 40 40 200 200 346 400

0 0.004 0.006 0.029 0.039 0.058 0.026

10615 ± 8 5007 ± 2 5764 ± 3 2849 ± 1 3666 ± 2 1129 ± 1 1334 ± 1

2 ± 0.01 9 ± 0.10 10 ± 0.5 14 ± 0.2 11 ± 0.3 8 ± 0.2 7 ± 0.1

1355 ± 22 912 ± 10 904 ± 8.2 450 ± 3.1 825 ± 4.1 129 ± 0.8 198 ± 1.2

0.9 ± 0.009 0.2 ± 0.001 0.5 ± 0.003 0.4 ± 0.002 0.3 ± 0.001 0.5 ± 0.004 0.6 ± 0.004

pumped. Selection and optimization of the flocculant load have been previously determined in our former works [30]. OMW then passes by overflow to a lamellar settler. Upon neutralization and flocculation units the following tasks were achieved: 1 – partial decomposition of residual oxidant when pH passed from 3 to 6–7 [40], and 2 – formation of catalyst aggregates which can be separated in the settling unit, thereby reducing its concentration to minimum values, and also slightly decreasing COD values. The concentration of phenolic compounds is not altered, having low values in the order of 1 mg dm3 at the outlet of CSTR. Table 4 shows the average values of the different parameters measured in the current output of the lamellar sedimentation. This table shows that increasing the concentration of the catalyst implies a decrease in the concentration of residual hydrogen peroxide and in the final COD values registered. However, no influence of the catalyst was detected in the final totals phenols or in the final total iron concentrations in the treated water. The final COD values obtained at the output of the plant are lower than those registered at the CSTR outlet. To sum up, Fenton’s treatment process conducted in this plant shows a degradation of organic matter and total phenols of OMW in the order of 97% and over 99% (COD = 129 mg O2 dm3 and total phenols = 0.5 mg dm3, Table 4), respectively. These results show that treatment of OMW process using the Fenton-like reaction is able to mineralize virtually all organic material and the phenolic compounds, providing transparent and odor-free final treated water. Advanced oxidation processes (AOPs) show great potential in treating chlorinated organic pollutants like PCBs. These AOPs generate extremely reactive radicals that oxidize organic matter, leading to carbon dioxide, water and other oxidation products without the transference of pollutants from one phase to another, as in adsorption or membrane separation [41]. Fenton’s reagent can be employed to treat a variety of industrial wastes containing a wide range of organic compounds like phenols (especially chlorophenols), formaldehyde, pesticides, wood preservatives, plastic additives, and rubber chemicals. The process may be applied to wastewaters, sludges, and contaminated soils with a reduction of toxicity, an improvement of biodegradability, as well as odor and color removal [42]. In previous work, we have analyzed more than 20 industrial wastewaters from olives and olive-oil washing produced by different olive-related industries. Pesticide residues (Simazine = 0.84 lg dm3, Diuron = 16.1 lg dm3, and p,p’-dichlorodiphenyldichloroethylene (pp-DDE) = 0.09 lg dm3) were only detected in some olive-washing wastewaters; all these pesticides were removed completely from the effluent treated by the Fenton-like reaction [3]. Oller et al. [43] indicated that the problem of toxic pollutants present in the environment must be tackled not only by determining pollutants using chemical oxygen demand measurement, etc., but also by biological assays (toxicity tests). In this sense, Mert

et al. [44] have shown that the treatment of wastewater samples (pH 5.2, 115 g O2 dm3, 32 g suspended solids dm3, 5.6 g phenols dm3) from an olive mill plant operating with the continuous three-phase decanting process with a daily olive processing capacity of 30 tons in Bursa City (Turkey) by Fenton (FeSO47H2O–H2O2) and Fenton-like (FeCl36H2O–H2O2) processes allows considerable inhibitory removed effect (COD > 80% and total-phenol > 85% removal) from untreated OMW. Toxicity/biodegradability tests used in this case were activated sludge inhibition tests. On the other hand, wastewaters used in this work proceed from an olive mill plant which operated using the continuous two-phase decanting processes and thus have lower values of organic load (4 g O2 dm3) and total phenols (0.0662 g dm3) (Table 1) than that of three-phase decanting process [44]. Also noteworthy, that Hodaifa et al. [28] have shown that the microalgae Scendesmus obliquus can grow in undiluted raw wastewater from two-phase decanting processes. However, in the case of raw wastewater from three-phase decanting processes the maximum concentration allowing the growth of the microalga at the same specific growth rate is 20% (v/v). This justifies the almost complete removal of growth inhibitors (organic matter, phenol compounds, and pesticides) by the Fenton-like process, allowing safely reuse in irrigation or discharge directly into the municipal wastewater system for ulterior tertiary treatment. Finally, AOPs can be suitably used for the treatment of industrial wastewater with relatively small COD (<5.0 g dm3), since higher COD values would require the consumption of too large amounts of expensive reactants [45]. Also, in this case the final water obtained meets the Spanish standard requirements for use in irrigation (COD < 1000 mg O2 dm3 [26]). Other oxidation techniques have been used to treat olive-mill wastewater, such as ozonation, electrochemical technology [13], solar photocatalytic systems [8], biological treatment [46], and combined bioremediation [47]. In all cases, the economic cost is very high, and the olive-oil industry in its current state (small factories) is not ready to assume these high costs (mostly energy costs and investment).

4. Conclusions The present study on the effectiveness of Fenton’s reagent in the treatment of OMW from continuous two-phase decanting processes was performed not only considering the CSTR stage, but also necessarily examining the complete set of all components of the plant to determine the actual effectiveness of the process. Control of all materials used in the process is critical to attain optimized operating conditions, and to minimize the amount of the reagents untapped (residual iron, and H2O2 unreacted). The effectiveness regarding OMW treatment by using Fenton’s reagent in a CSTR at pilot plant level to reduce the content of organic matter in the treated effluent has been demonstrated. High conversion levels were dependent on the operating conditions,

124

G. Hodaifa et al. / Chemical Engineering Journal 220 (2013) 117–124

i.e., pH, space–time in the CSTR, reagents dosage, and catalyst/oxidant ratio. Optimum operation values were: pH = 3 achieved by the addition of the catalyst (without acid addition when [FeCl3]/ [H2O2] > 0.01), smin = 3 h, [FeCl3]/[H2O2] ratio from 0.026 to 0.058 w/w upon Fe(III) concentration ranging between 0.35 and 0.40 g dm3, and ambient temperature. These conditions allow the removal of 97% of organic matter and the 99% of phenolic compounds load present in OMW. Finally, this study offers a solution for reducing the environmental effects of wastewaters generated in two-phase extraction process of olive-oil industry. The final water treated at the outlet of the plant can be used directly for irrigation or straightly discharged into the municipal wastewater system for posterior tertiary treatment. Acknowledgements We are grateful to the Ministry of Science and Technology for financial support through Project PPQ2003-07873 ‘‘Treatment of olive-oil mill wastewater for its reuse in agriculture irrigation’’, and the project CTQ 2007-66178 ‘‘Depuration process of the olive-oil mill wastewaters by Fenton like treatment and later purification through biosorption’’. References [1] C. Gomec, E. Erdim, I. Turan, A. Aydin, I. Ozturk, Advanced oxidation treatment of physico-chemically pre-treated olive mill industry effluent, J. Environ. Sci. Health Part B 46 (2007) 741–747. [2] R.C. Martins, R.M. Quinta-Ferreira, Remediation of phenolic wastewaters by advanced oxidation processes (AOPs) at ambient conditions: comparative studies, Chem. Eng. Sci. 66 (2011) 3243–3250. [3] L.M. Nieto, G. Hodaifa, S.R. Vives, J.A. Casares, S.B. Driss, R. Grueso, Treatment of olive-mill wastewater from a two-phase process by chemical oxidation on an industrial scale, Water Sci. Technol. 59 (10) (2009) 2017–2027. [4] J.A. Zazo, J.A. Casas, A.F. Mohedano, J.J. Rodriguez, Semicontinuous Fenton oxidation of phenol in aqueous solution: a kinetic study, Water Res. 43 (2009) 4063–4069. [5] O. Legrini, E. Oliveros, A.M. Braun, Photochemical processes for water treatment, Chem. Rev. 93 (1993) 671–698. [6] F.C. Moreira, V.J.P. Vilar, A.C.C. Ferreira, F.R.A. dos Santos, M. Dezotti, M.A. Sousa, C. Goncalves, R.A.R. Boaventura, M.F. Alpendurada, Treatment of a pesticide-containing wastewater using combined biological and solar-driven AOPs at pilot scale, Chem. Eng. J. 209 (2012) 429–441. [7] A.M.A. Pintor, V.J.P. Vilar, R.A.R. Boaventura, Decontamination of cork wastewaters by solar-photo-Fenton process using cork bleaching wastewater as H2O2 source, Sol. Energy 85 (2011) 579–587. [8] W. Gernjak, M.I. Maldonado, S. Malato, J. Cáceres, T. Krutzler, A. Glaser, R. Bauer, Pilot-plant treatment of olive mill wastewater (OMW) by solar TiO2 photocatalysis and solar photo-Fenton, Sol. Energy 77 (2004) 567–572. [9] M.S. Lucas, R. Mosteo, M.I. Maldonado, S. Malato, J.A. Peres, Solar photochemical treatment of winery wastewater in a CPC reactor, J. Agric. Food Chem. 57 (2009) 11242–11248. [10] A.M. Amat, A. Arques, F. López, M.A. Miranda, Solar photo-catalysis to remove paper mill wastewater pollutants, Sol. Energy 79 (2005) 393–401. [11] B. Bianco, I. De Michelis, F. Vegliò, Fenton treatment of complex industrial wastewater: optimization of process conditions by surface response method, J. Hazard. Mater. 186 (2011) 1733–1738. [12] R. Andreozzi, V. Caprio, A. Insola, R. Marotta, Advanced oxidation processes (AOPs) for water purification and recovery, Catal. Today 53 (1) (1999) 51–59. [13] P. Cañizares, R. Paz, C. Sáez, M.A. Rodrigo, Costs of the electrochemical oxidation of wastewaters: a comparison with ozonation and Fenton oxidation processes, J. Environ. Manage. 90 (2009) 410–420. [14] I. Arslan-Alaton, Degradation of a commercial textile biocide with advanced oxidation processes and ozone, J. Environ. Manage. 82 (2007) 145–154. [15] C.T. Benatti, C.R.G. Tavares, T.A. Guedes, Optimization of Fenton’s oxidation of chemical laboratory wastewaters using the response surface methodology, J. Environ. Manage. 80 (2006) 66–74. [16] M.I. Badawy, M.Y. Ghalyb, T.A. Gad-Allah, Advanced oxidation processes for the removal of organophosphorus pesticides from wastewater, Desalination 194 (2006) 166–175. [17] P. Bautista, A.F. Mohedano, M.A. Gilarranz, J.A. Casas, J.J. Rodriguez, Application of Fenton oxidation to cosmetic wastewater treatment, J. Hazard. Mater. 143 (2007) 128–134. [18] J.H. Ramirez, C.A. Costa, L.M. Madeira, Experimental design to optimize the degradation of the synthetic dye Orange II using Fenton’s reagent, Catal. Today 107–108 (2005) 68–76.

[19] F.J. Rivas, F.J. Beltran, O. Gimeno, P. Alvarez, Optimization of Fenton’s reagent usage as a pre-treatment for fermentation brines, J. Hazard. Mater. B96 (2003) 277–290. [20] H. Tekin, O. Bilkay, S.S. Ataberk, T.H. Balta, I.H. Ceribasi, F.D. Sanin, F.B. Dilek, U. Yetis, Use of Fenton oxidation to improve the biodegradability of a pharmaceutical wastewater, J. Hazard. Mater. B136 (2006) 258–265. [21] E.C. Catalkaya, F. Kargi, Color, TOC and AOX removals from pulp mill effluent by advanced oxidation processes: a comparative study, J. Hazard. Mater. 139 (2007) 244–253. [22] A. Lopez, G. Mascolo, A. Detomaso, G. Lovecchio, G. Villani, Temperature activated degradation (mineralization) of 4-chloro-3-methyl phenol by Fenton’s reagent, Chemosphere 59 (2005) 397–403. [23] J.H. Ramirez, F.M. Duarte, F.G. Martins, C.A. Costa, L.M. Madeira, Modelling of the synthetic dye Orange II degradation using Fenton’s reagent: from batch to continuous reactor operation, Chem. Eng. J. 148 (2009) 394–404. [24] F.J. Rivas, V. Navarrete, F.J. Beltran, J.F. Garcia-Araya, Simazine Fenton’s oxidation in a continuous reactor, Appl. Catal. B: Environ. 48 (2004) 249–258. [25] H. Zhang, H.J. Choi, C.-P. Huang, Treatment of landfill leachate by Fenton’s reagent in a continuous stirred tank reactor, J. Hazard. Mater. B136 (2006) 618–623. [26] Resolution of Guadalquivir River Basin president, Scattering washing waters from olive-oil mill, Ministry of Environment, Seville, Spain, June 9, 2006. [27] M.J. López, J. Moreno, A. Ramos-Cormenzana, The effect of olive mill wastewaters variability on xanthan production, J. Appl. Microbiol. 90 (2001) 829–835. [28] G. Hodaifa, M.E. Martínez, S. Sánchez, Use of industrial wastewater from oliveoil extraction for biomass production of Scenedesmus obliquus, Bioresour. Technol. 99 (5) (2008) 1111–1117. [29] L. Nieto, G. Hodaifa, S.R. Vives, J.A. Casares, Industrial plant for olive mill wastewater from two-phase treatment by chemical oxidation, J. Environ. Eng. 136 (11) (2010) 1309–1313. [30] L. Nieto, G. Hodaifa, S.R. Vives, J.A. Casares, J. Ochando, Flocculation– sedimentation combined with chemical oxidation process, Clean–Soil Air Water 39 (10) (2011) 949–955. [31] ISO 8466-1, Water quality-Calibration and evaluation of analytical methods and estimation of performance characteristics – Part 1: Statistical evaluation of the linear calibration function, Genève, Switzerland, 1990. [32] DIN 38402 A51, German standard methods for the examination of water, waste water and sludge; general information (group A); calibration of analytical methods, evaluation of analytical results and linear calibration functions used to determine the performance characteristics of analytical methods (A 51), 1986. [33] DIN 38409-41, German Standard Methods for Examination of Water, Waste Water and Sludge; Summary Action and Material Characteristic Parameters (group H); Determination of the chemical oxygen demand (COD) in the range over 15 mg/l (H41), 1980. [34] A. Balcioglu, I. Arslan, Treatment of textile industry wastewater by enhanced photocatalytic oxidation reaction, J. Adv. Oxid. Technol. 4 (2) (1999) 189–195. [35] G.M. Eisenberg, Colorimetric determination of hydrogen peroxide, Ind. Eng. Chem. 15 (1943) 327–328. [36] R.C. Martins, R.J.G. Lopes, R.M. Quinta-Ferreira, Lumped kinetic models for single ozonation of phenolic effluents, Chem. Eng. J. 165 (2010) 678–685. [37] L. Nieto, G. Hodaifa, S.R. Vives, J.A. Casares, J. Ochando, Degradation of organic matter in olive-oil mill wastewater through homogeneous Fenton-like reaction, Chem. Eng. J. 173 (2011) 503–510. [38] E. Neyens, J. Baeyens, A review of classic Fenton’s peroxidation as an advanced oxidation technique, J. Hazard. Mater. 98 (2003) 33–50. [39] M. Pera-Titus, V. Garcia-Molina, M.A. Baños, J. Gimenez, S. Esplugas, Degradation of chlorophenols by means of advanced oxidation processes: a general review, Appl. Catal. B: Environ. 47 (2004) 219–256. [40] M.I. Badawy, M.E.M. Ali, Fenton’s peroxidation and coagulation processes for the treatment of combined industrial and domestic wastewater, J. Hazard. Mater. B136 (2006) 961–966. [41] C. Rodrigues, L. Madeira, R. Boaventura, Treatment of textile effluent by chemical (Fenton’s reagent) and biological (sequencing batch reactor) oxidation, J. Hazard. Mater. 172 (2009) 1551–1559. [42] M. Pera-Titus, V. García-Molina, M.A. Baños, J. Giménez, S. Esplugas, Degradation of chlorophenols by means of advanced oxidation processes: a general review, Appl. Catal. B: Environ. 47 (2004) 219–256. [43] I. Oller, S. Malato, J.A. Sánchez-Pérez, Combination of advanced oxidation processes and biological treatments for wastewater decontamination—a review, Sci. Total Environ. 409 (2011) 4141–4166. [44] B.K. Mert, T. Yonar, M.Y. Kilic, K. Kestio_glu, Pre-treatment studies on olive oil mill effluent using physicochemical, Fenton and Fenton like oxidations processes, J. Hazard. Mater. 174 (2010) 122–128. [45] R. Andreozzi, V. Caprio, A. Insola, R. Marotta, Advanced oxidation processes (AOPs) for water purification and recovery, Catal. Today 53 (1999) 51–59. [46] F.J. Benítez, I.L. Acero, T. González, J. García, Organic matter removal from wastewaters of the black olive industry by chemical and biological procedures, Process. Biochem. 37 (2001) 257–265. [47] S. Khoufi, F. Aloui, S. Sayadi, Treatment of olive mill wastewater by combined process electro-Fenton reaction and anaerobic digestion, Water Res. 40 (2006) 2007–2016.