Effect of pH on pentachlorophenol degradation in ... - CiteSeerX [PDF]

Chemical Engineering Journal 168 (2011) 1209–1216

Contents lists available at ScienceDirect

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

Effect of pH on pentachlorophenol degradation in irradiated iron/oxalate systems Qing Lan a,b , Hong Liu a , Fang-bai Li b,∗ , Feng Zeng a,∗∗ , Cheng-shuai Liu b a b

School of Chemistry and Chemical Engineering, Sun Yat-sen University, Guangzhou, Guangdong 510275, China Guangdong Key Laboratory of Agricultural Environment Pollution Integrated Control, Guangdong Institute of Eco-Environmental and Soil Sciences, Guangzhou 510650, China

a r t i c l e

i n f o

Article history: Received 21 November 2010 Received in revised form 4 February 2011 Accepted 7 February 2011 Keywords: Pentachlorophenol Oxalic acid Hematite Goethite pH

a b s t r a c t This study investigated the photodegradation of pentachlorophenol (PCP) in two types of iron oxide/oxalate systems as a function of pH. Two iron oxides of hematite (␣-Fe2 O3 ) and goethite (␣-FeOOH) were selected. The experiments conducted at pH values 3.5, 5.0, and 7.0 showed that PCP photodegradation proceeded rapidly at 3.5, and slowed down with the increase in pH. To account for the effect of pH on the degradation kinetics, the surface charge of iron oxides, the adsorption of PCP and oxalic acid onto the iron oxides at different pH values were researched. The presence of oxalic acid gave the iron oxides a negative charge at pH values above 3, which caused the decrease of PCP adsorption with increasing pH value. A higher PCP adsorption in low pH may contribute to PCP photodegradation. On the other hand, iron oxides adsorbed oxalic acid to form photoactive Fe(III)–oxalate complexes. Compared with ␣-FeOOH, ␣-Fe2 O3 had a much stronger ability to adsorb oxalic acid and showed better photocatalytic activity. Furthermore, the detection results of H2 O2 during the process of photodegradation showed that a low pH favored the H2 O2 generation. The combination of oxalic acid photodegradation and intermediates during the process of PCP photodegradation indicated that at an initial pH of 3.5, the PCP photodegradation occurred mainly via the attack by • OH, and that at initial pH values 5.0 and 7.0, a direct photolysis mechanism was dominant. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Pentachlorophenol (PCP) is the most toxic of chlorophenols (CPs) and has been widely used as a wood preservative, fungicide, insecticide, and herbicide. Its environmental stability is concomitant with its ubiquity in the surface water, soil, and groundwater [1]. Unfortunately, the biodegradation of PCP lasts long and is far from complete decomposition [2]. The reduction of PCP by zerovalence iron or Fe(II)-based technology occurs quite slowly [3,4]. Although PCP has absorption below 370 nm [5], its direct photolysis suffers from the formation of more toxic products, such as polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans [6]. Recently, Yin et al. reported the reductive dechlorination of PCP by photocatalysis with Ti-doped ␤-Bi2 O3 and the combination of photocatalysis and laccase catalysis [7,8]. On the other hand, as a kind of advanced oxidation technology (AOT), the indirect photolysis of homogeneous photo-Fenton reaction can effectively degrade PCP by the attack of hydroxyl radicals (• OH) [5,9]. In fact, • OH can be generated from irradiated Fe–oxalate systems belonging to other indirect photolytic approaches [10–16]. In

∗ Corresponding author. Tel.: +86 20 8702 4721; fax: +86 20 8702 4123. ∗∗ Corresponding author. Tel.: +86 20 8411 4133; fax: +86 20 8411 2245. E-mail addresses: [email protected], [email protected] (F.-b. Li), [email protected] (F. Zeng). 1385-8947/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.cej.2011.02.017

irradiated Fe–oxalate systems, the photo-Fenton-like reaction can be generated. Firstly, the photo-reduction of Fe(III)–oxalate complexes yields Fe(II) and oxalate radicals (C2 O4 )•− , which induces the formation of active species including H2 O2 , carbon-centered radicals (CO2 )•− , superoxide radicals (O2 •− ), and hydroperoxyl radicals (• OOH) [12–14]. Secondly, H2 O2 reacts with Fe(II) to generate • OH by Fenton reaction. Simultaneously, the iron redox cycling ensures the continued production of H2 O2 and • OH [11,17–19]. Undoubtedly, the in situ formation of H2 O2 is economically attractive in practical applications. Oxalic acid, which is secreted by plant roots [15] or formed by the incomplete combustion of hydrocarbons [13,14], is also ubiquitous in soil, water, and the atmosphere. Compared to refractory toxic organic compounds, oxalic acid, as a short chain carboxylic acid, is easily biodegradable [20,21]. Furthermore, heterogeneous photo-Fe–oxalate systems do not easily bring about the secondary contamination of abundant dissolved Fe ions, and are easily recovered in practical treatment. This study selected two iron oxides (IO), hematite (␣-Fe2 O3 ) and goethite (␣-FeOOH), and investigated the degradation of PCP in the two heterogeneous irradiated Fe–oxalate systems. We focused on the explanation of the effects of pH on PCP photodegradation because pH is an important factor affecting the activity of photo-Fenton systems [21,22]. In such heterogeneous systems, the chemical adsorption of oxalic acid as a critical process occurs on the surface of iron oxide leading to the formation of Fe–oxalate complexes. Previous reports have indicated that pH significantly affects

1210

Q. Lan et al. / Chemical Engineering Journal 168 (2011) 1209–1216

Relative Intensity (a.u.)

3000

(110) 4.172

α−FeOOH

(111) 2.447 (130) 2.688

2500

2000

(221) 1.719

(120) 3.388

1500 (012) 3.675

(104) 2.701

20

30

(116) (214) (024) 1.692 1.486 1.841 (300) (018) 1.453 1.596

(113) 2.203

40

2.3. Photochemical experiments α−Fe2O3

(110) 2.518

1000

(151) 1.562

ο

50

tures were ascertained by X-ray powder diffraction (XRD) (Rigaku D/Max-III, Japan). The XRD results showed that pure ␣-Fe2 O3 and ␣-FeOOH were obtained by comparing with standard XRD photographs. In addition, using the Brunauer–Emmett–Teller (BET) method (Micromeritics ASAP-2010 M, USA), their specific surface areas were measured to be 29.4 and 32.3 m2 g−1 , respectively, indicating a close specific surface area between the two iron oxides.

60

70

2θ ( ) Fig. 1. The XRD patterns and d-spacing values of ␣-Fe2 O3 and ␣-FeOOH.

the adsorption of oxalic acid onto iron oxides [17,23]. Mazellier and Sulzberger reported an optimum pH of 4 for oxalate adsorption onto ␣-FeOOH [17], whereas Zhang et al. showed that the adsorption of oxalate onto ␣-Fe2 O3 decreased as the pH rose [23]. During the adsorption process, the non-reductive/reductive dissolution of iron oxide takes place and light irradiation greatly enhances the reductive dissolution of Fe(III)–oxalate complexes [24–26]. In addition, various iron oxides demonstrate different dissolution properties due to their different thermodynamic stabilities [27]. Although low pH favors the photodegradation of organic compounds in Fe–oxalate systems [15,16], increase in pH leads to the conversion of the dominant Fe(III) complex to Fe(C2 O4 )2 − and Fe(C2 O4 )3 3− , which exhibit higher efficiency in the photoreaction [10,12]. However, increased pH accelerates the oxidation of Fe(II) and leads to a lack of Fe(II), impeding the iron redox cycling. Thus, the effects of pH are rather complex in the studied heterogeneous systems and understanding the effects of pH on the photodegradation of PCP is important. In the present study, the photodegradation kinetics of PCP, the adsorption of PCP and oxalic acid, the reduction and dissolution of iron oxides, the in situ formation of H2 O2 and the intermediates products were studied in detail as a function of pH. The findings would serve to facilitate the understanding of PCP degradation in the heterogeneous irradiated Fe–oxalate systems and aid in developing an engineered system to purify water and wastewater containing PCP. 2. Experiments and methods 2.1. Chemicals Deionized water (18.2 m cm) from an ultrapure water system (Easy Pure’II RF/UV, USA) was used in all experiments. PCP (98%) was purchased from Aldrich, USA. Tetrachloro-pbenzoquinone (99%), tetrachloro-o-benzoquinone (97%), methanol [high-performance liquid chromatography (HPLC) grade], and hexane (HPLC grade) were obtained from Acros, Belgium. Other analytical grade chemicals were purchased from the Guangzhou Chemical Co., China. All chemicals were used as received except acetic anhydride, which was redistilled for gas chromatography with mass spectrometry (GC/MS) analysis. 2.2. Preparation and characterization of iron oxides Two iron oxides were synthesized according to previously reported procedures [15,16]. As shown in Fig. 1, their crystal struc-

Aqueous PCP stock solution (0.075 mM) was prepared as previously described [19]. In all photodegradation experiments, aqueous suspensions contained PCP at an initial concentration of 0.0375 mM, oxalic acid at an initial concentration of 1.2 mM, and iron oxide at a dose of 0.40 g L−1 . All experiments were carried out in a photochemical reactor previously described in detail [15] using an 8 W LZC-UV lamp (Luzchem Research, Inc.) with an emission peak at 365 nm as the light source. The experiments were performed at 30 ◦ C kept by a thermostatic water bath. Prior to irradiation, adsorption/desorption equilibrium was established in the aqueous PCP suspension in the dark for 30 min. The suspension was continuously stirred by a magnetic stirrer and was bubbled with air. The samples were taken at preset time intervals and then centrifuged at 4500 rpm for 25 min, and further filtered through a 0.45 ␮m filter for the HPLC, ion chromatography (IC), and GC/MS analyses. 2.4. Adsorption experiments The adsorption experiments of PCP or oxalic acid at different pH values were conducted in Erlenmeyer flasks in the dark by equilibrating 50 mL aliquots containing iron oxide (0.4 g L−1 ) and various concentrations of PCP or oxalic acid. The suspension was stirred in a thermostatic shaker at 180 rpm at constant temperature (30 ± 1 ◦ C) for 24 h. After the suspension was filtered through a 0.45 ␮m filter, the filtrate was analyzed for concentrations of PCP or oxalic acid. The 0.45 ␮m filter with a 13 mm diameter was obtained from MEMBRANA (Germany). The pH scope of its application is in the range of 1–14. The amount of adsorbed PCP or oxalic acid was obtained based on the mass balance in the solution before and after the adsorption process. 2.5. Analysis PCP was detected by HPLC (Waters 1525/2487) with an Xterra C18 reverse-phase column (250 mm × 4.6 mm, 5 ␮m, Waters, USA). The mobile phase contained a mixture of 1% aqueous acetic acid and methanol (20:80 = v/v) at a flow rate of 1.0 mL min−1 . The column temperature and the UV detection were set to 35 ◦ C and 295 nm, respectively. The concentrations of oxalic acid were determined by IC (Dionex ICS-90) with a Dionex IonPac AS14A column (250 mm × 4.6 mm, Dionex, USA). The mobile phase is the aqueous 1.0 mM NaHCO3 –8.0 mM Na2 CO3 solution and the flow rate is 1.0 mL min−1 . Zeta-potential measurements were conducted using a zeta potential analyzer (ZetaPlus2002, Brookhaven Inc., USA). Prior to measurement, suspensions containing 0.40 g L−1 iron oxide at different pH levels were equilibrated in the dark in a thermostatic shaker at 180 rpm at constant temperature (30 ± 1 ◦ C) for 24 h. During analysis, the equilibrated slurry was placed into the electrophoresis cell, which was thoroughly washed and rinsed with ultrapure water followed by rinsing with the suspension to be measured. Dissolved Fe(II) was colorimetrically measured by the ferrozine method. The total dissolved iron was determined in the same way after adding 10% OHNH3 Cl to reduce all Fe(III)(aq) to Fe(II)(aq) [15].

Q. Lan et al. / Chemical Engineering Journal 168 (2011) 1209–1216

A

Ct / C0 of PCP

B

α−Fe2O3

1.0

α−FeOOH

0.8 0.6 0.4 0.2 0.0 0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

Ct / C0 of OX

1.0

0.8

0.6

0.4

7

6

pH

The dissolved Fe(III) content can be obtained by subtracting the dissolved Fe(II) from the total dissolved iron. The H2 O2 in aqueous solution was measured using an H2 O2 analyzer (Lovibond-ET8600, Germany), in which a Lovibond reagent was used to react with H2 O2 in a 10 mL vessel to form a colored solution and the concentration of H2 O2 was then photometrically determined at 528 nm with a detection limit of 0.05 mg L−1 . The Lovibond reagent is a kit of H2 O2 analyzer and is a patented tablet. During the PCP photodegradation experiments, a Lovibond reagent was first added into the vessel before the sample was taken from the photoreactor. At each time interval, 10 mL of reaction solution was sampled and immediately filtered through a 0.45 ␮m filter before being placed into the vessel and the concentration of H2 O2 in the filtrate was measured at once. As this procedure was completed within 1 min, the validity of H2 O2 concentration was ensured. Following the analytical procedure described by Oturan et al. [28], two PCP degradation intermediates, tetrachloro-pbenzoquinone and tetrachloro-o-benzoquinone, were identified by comparing their retention times in HPLC with internal standards. The mobile phase and flow rate were the same as noted above for the PCP analysis, but the ratio of 1% acetic acid in water to methanol was 25:75 (v/v). The column temperature and the detector wavelength were set to 30 ◦ C and 280 nm, respectively. A GC/MS (Thermo Trace-DSQ-2000, USA) system with electron ionization and an Agilent silicon capillary column (0.25 mm Ø × 30 m, Agilent, USA) were also used to identify the intermediate products of PCP. The samples were pretreated by extraction as previously described [5]. Additionally, an LC-10A system (Shimadzu) with IC-A3 column was used to determine the other intermediates, including HCOOH and CH3 COOH. The mobile phase was a mixture of 2.5 mM phthalic acid and 2.4 mM Tris. The flow rate and the column temperature were set to 1.2 mL min−1 and 40 ◦ C, respectively.

1211

IO alone / pH 3.5 UV alone / pH 3.5 UV / IO / pH 3.5 UV / OX / pH 3.5 UV / OX / IO / pH 3.5 UV / OX / IO / pH 5.0 UV / OX / IO / pH 7.0

5

3. Results and discussion

4

3.1. Photodegradation of PCP

3 0

• OH

20

40

60

0

20

40

60

Reaction time (min) Fig. 2. The variations of PCP and oxalic acid (OX) concentrations as well as pH over the photoreaction time of 1 h in different iron oxide (IO) systems.

0.04

α−Fe2O3

R = 0.96

α−FeOOH

0.03 -1

k ( min )

In the irradiated Fe–oxalate systems, H2 O2 and can be formed at much lower concentrations of Fe(III) (1 ␮M) and oxalate (5 ␮M) [13,14]. The optimal initial oxalic acid concentration of 1.2 mM for the heterogeneous photo-Fe–oxalate systems has been explored in our previous research [18] and was used in this study. Fig. 2 presents the profiles of the PCP and oxalic acid concentrations, as well as the pH during the experiments. PCP was not removed under the condition of using only iron oxide. Under UV and UV/OX conditions after 1 h, PCP was directly photodegraded by 54.6 ± 1.70% and 59.6 ± 3.60%, respectively, which are higher than those under UV/IO conditions. It might because the turbidity of iron oxides inhibited UV penetration in UV/IO systems. Under the UV/OX/IO condition at an initial pH of 3.5, the PCP photodegradation was dramatically enhanced that should owe to the attack of • OH formed by the Fenton reaction. However, under the UV/OX/IO condition at initial pH values of 5.0 and 7.0, the PCP removal was only slightly higher than that under UV/IO systems. Overall, the photodegradation rate of PCP was higher in the ␣-Fe2 O3 system than in the ␣-FeOOH system. Increasing pH is known to accelerate the direct photolysis of PCP under UV in homogeneous systems [29]. However, in heterogeneous UV/OX/IO systems, the increase in initial pH was detrimental to PCP degradation. The photodegradation of PCP in UV/OX/IO systems approximately followed first-order reaction kinetics with different first-order rate constants (k) as presented in Fig. 3. Compared with the rate at the initial pH of 3.5, the photodegradation PCP at initial pH values of 5.0 and 7.0 decreased obviously and

R = 0.79 R = 0.97 0.02

R = 0.99 R = 0.97

R = 0.98

0.01 3

4

5

6

7

pH Fig. 3. The dependence of the first-order rate constants (k) of PCP photodegradation on the initial pH value, R is the correlation coefficient of fitting.

1212

Q. Lan et al. / Chemical Engineering Journal 168 (2011) 1209–1216

80

A

B

α-Fe2O3

α -FeOOH

90

-1

Cis / PCP (mmol kg )

Zeta potential (mv)

IO alone IO / OX

40

0

A

B

α−Fe2O3

α− FeOOH -1

0.10 g L IO

-1

75

0.10 g L IO

60 45 30

pH 3.5 pH 5.0 pH 7.0

15

-40

0 2

4

6

8

2

4

pH

6

8

10

0

4

8

12

16

0

4

8

12

Equilibrium concentration of PCP (μΜ)

pH

Fig. 4. The zeta-potentials of iron oxides in the absence and presence of oxalic acid (1.2 mM) as a function of pH.

Fig. 5. The adsorption isotherms of PCP on iron oxides in the presence of 1.2 mM oxalic acid.

had a similar rate constant. It will be discussed in detail in Section 3.5.

be desorbed by centrifugation. As the samples were analyzed after centrifugation during the photochemical experiments, the removal of PCP should owe to the photoreaction rather than the adsorption. In addition, the adsorption isotherms of oxalic acid onto the iron oxides were measured, and the results are shown in Fig. 6. Clearly, the isotherms could be described by the Langmuir model as marked by the solid lines. Also, the adsorption of oxalic acid onto iron oxides depended on pH. With the increase in pH, the adsorption of oxalic acid increased. Although there were close specific surface areas between the two iron oxides, the adsorption ability of ␣-Fe2 O3 for oxalic acid was much higher than that of ␣FeOOH, which agrees with previous studies [30]. The adsorption of oxalic acid contributes to the photodegradation of PCP because the photoactive Fe(III)–oxalate complexes that leads the genera-

3.2. Adsorption of PCP and oxalic acid on the iron oxides To account for the difference in the kinetic rates of PCP photodegradation at different initial pH values, the relationships between surface charges of iron oxides and pH were investigated. Fig. 4 shows the zeta-potentials of iron oxides in the absence and presence of oxalic acid as a function of pH. In the absence of oxalic acid, the isoelectric point (iep) of iron oxide was at pH 7.4 for ␣Fe2 O3 and at pH 7 for ␣-FeOOH, both of which are in agreement with previously reported values [30]. However, the presence of oxalic acid caused a significant shift of iep to a lower pH value of 1.8 and 2.5 for ␣-Fe2 O3 and ␣-FeOOH, respectively. This could be attributed to the replacement of surface hydroxyl groups attached to the metallic centers with adsorbed oxalate anions. During the adsorption process, the Fe(III)–oxalate complexes are formed as follows [31]:

A

α−Fe2O3

B

α−Fe2O3

1500

−

Fe–H2 C2 O4 + OH

1000

500

-1

The shift in iep clearly indicated a specific adsorption of oxalic acid on iron oxides through the formation of a ligand bond, rather than by a purely electrostatic interaction. More importantly, the adsorption of oxalic acid gives the iron oxides a negative charge over a wide pH range of 3–9, which greatly influences the surface properties of iron oxides. Adsorption experiments of PCP in the presence of 1.2 mM oxalic acid at different pH values were also conducted. The results are depicted in Fig. 5. PCP with a concentration range of 0.00375–0.0375 mM was completely (100%) adsorbed on 0.40 g L−1 iron oxides at a pH of 3.5. To obtain a full picture of the isotherm, the adsorption isotherms of PCP by a less amount of iron oxides (0.10 g L−1 ) at a pH of 3.5 were specifically carried out. Overall, the Freundlich model could describe the adsorption of PCP well. Moreover, with the decrease in pH, the adsorption of PCP also increased. It should be related to the fact that ␣-Fe2 O3 and ␣-FeOOH are negatively charged at pH values above 3, as shown in Fig. 4. As the pKa value of PCP is 4.75 [32], 5.32%, 64.0%, and 99.4% of PCP exist as the dissociated species (PCP− ) at pH values 3.5, 5.0, and 7.0, respectively. It suggests that the repulsion between the surface of iron oxides and PCP rapidly increases with increasing pH. Considering that PCP photodegradation increases with decreasing pH in UV/OX/IO systems, the higher adsorption at lower pH might favor the PCP photodegradation in such heterogeneous systems. In the experimental concentration range, the adsorbed PCP could

Cis (mmol kg )

Fe–OH + H2 C2 O4 ↔

2000

0

OX alone 0.0

600

0.4

0.8

1.2

OX with 0.0375 mM PCP 1.6

2.0 0.0

C

α−FeOOH

0.4

0.8

1.2

1.6

2.0

D

α−FeOOH

500

pH 3.5 pH 5.0 pH 7.0 pH 9.0

400 300 200 100

OX with 0.0375 mM PCP

OX alone

0 0.0

0.4

0.8

1.2

1.6

2.00.0

0.4

0.8

1.2

1.6

2.0

Equilibrium concentration of oxalic acid (mM) Fig. 6. The adsorption isotherms of oxalic acid on iron oxides as a function of pH.

Q. Lan et al. / Chemical Engineering Journal 168 (2011) 1209–1216

3.3. Reduction and dissolution of Fe from the iron oxides To explore iron cycling during PCP photodegradation, the nonreductive/reductive dissolution of iron oxides was investigated. In the absence of oxalic acid, no dissolved Fe(III)/Fe(II) was found (data not shown). On the other hand, in the presence of oxalic acid, small amounts of dissolved Fe(III)/Fe(II) in the range of 0.01–0.10 mM were detected and were found to be pH dependent, as illustrated in Fig. 7. After 30 min of adsorption/desorption equilibration, dissolved Fe(III) was formed by the desorption of surface Fe(III)–oxalate complexes, and Fe(II) was generated by the slow reductive dissolution in the dark. In the early photoreaction stages, light irradiation promoted the reductive dissolution of Fe(III)–oxalate complexes [24]. The dissolved Fe(III) and Fe(II) all significantly grew under an initial pH of 3.5, whereas under initial pH values of 5.0 and 7.0, the dissolved Fe(II) only slightly increased. In the late photoreaction stages, the dissolved Fe(III)/Fe(II) all significantly declined, which might because the pH quickly jumped from

0.12

A

0.12

B α-Fe2O3 / pH 3.5 α-Fe2O3 / pH 5.0

0.09

Dissolved Fe(II) (mM)

0.09

Dissolved Fe(III) (mM)

tion of active species (H2 O2 and • OH) are formed by the oxalic acid adsorption process [17,19]. The higher photodegradation rates of PCP in the ␣-Fe2 O3 system than in the ␣-FeOOH system might be a reflection of this contribution. On the other hand, for the individual iron oxides, the increasing trend of oxalic acid adsorption with pH appeared not to favor PCP photodegradation, which might because higher pH impeded the iron cycling as discussed in Section 3.5. The effect of pH on the adsorption of oxalic can be explained by the type of Fe(III)–oxalate complexes and by the speciation of oxalic acid. In the presence of Fe(III), four possible Fe(III)–oxalate complexes, namely FeHC2 O4 2+ , FeC2 O4 + , Fe(C2 O4 )2 − , and Fe(C2 O4 )3 3− , can be formed with overall dissociation constants (kd ) of 2.95 × 10−10 , 3.98 × 10−10 , 6.31 × 10−17 , and 3 × 10−21 , respectively [33]. Obviously, based on the values of kd , it can be deduced that the stability of Fe(III)–oxalate complexes largely increase with increasing coordination number of C2 O4 2− as follows: Fe(C2 O4 )3 3−  Fe(C2 O4 )2 −  FeC2 O4 + ≈ FeHC2 O4 2+ . While the fraction of C2 O4 2− in the solution just increases with increasing pH, and at pH >6.0, C2 O4 2− becomes the sole species. It suggests that, at high pH, there are more C2 O4 2− to form more stable Fe(III)–oxalate complexes that are difficult to desorb. And only C2 O4 2− is present in the suspensions at pH values 7.0 and 9.0, so the amounts of Fe(III)–oxalate formed on the same iron oxide were close. Furthermore, the interactions among PCP, oxalic acid, and iron oxides were complicated, as shown in Fig. 6. Although PCP concentration (0.0375 mM) was much less than that of oxalic acid (0.1–2.0 mM), its effects on the adsorption of oxalic acid were significant. In the ␣-Fe2 O3 system, the presence of PCP hampered the adsorption of oxalic acid at a pH of 3.5, but was promoted at pH values of 5.0, 7.0, and 9.0. In the ␣-FeOOH system, the existence of PCP notably impeded the adsorption of oxalic acid at all studied pH values. The discrepancies may be considered to involve the different surface functional groups of ␣-Fe2 O3 and ␣-FeOOH. The surface hydroxyl groups arising from adsorption of water or from the structure are the functional groups of iron oxides and varied largely in type, proportion, and density for different iron oxides [30]. Factors such as the pH, the coordination of the adsorbate, and the crystal morphology of iron oxides also greatly influence the number of functional groups corresponding to the maximum uptake of the adsorbed species and the activity of iron oxides [30]. In further research, more techniques, such as IR spectroscopy and extended X-ray absorption fine structure (EXAFS), should be used to explore the interaction mechanisms among oxalic acid, iron oxides, and PCP. In this study, our results showed that more oxalic acid can be adsorbed on ␣-Fe2 O3 than on ␣-FeOOH in the absence and presence of PCP, which may contribute to the PCP photodegradation in the UV/OX/␣-Fe2 O3 system.

1213

0.06

0.03

0.00

α-Fe2O3 / pH 7.0 α-FeOOH / pH 3.5 α-FeOOH / pH 5.0 α-FeOOH / pH 7.0

0.06

0.03

0.00 0

20

40

60

0

20

40

60

Reaction time (min) Fig. 7. The variation of dissolved Fe during PCP photodegradation under an initial concentration of 1.2 mM oxalic acid. (A) Fe(III) and (B) Fe(II).

the initial pH values of 3.5 and 5.0 up to 6.0 and 7.0, as shown in Fig. 2. In summary, as the initial pH is increased, the amounts of dissolved Fe(III)/Fe(II) obviously decreased, which clearly suggested that lower pH values facilitate the dissolution of iron oxides. In addition, it was found that the dissolved Fe(III)/Fe(II) was higher in the ␣-FeOOH system than in the ␣-Fe2 O3 system, which is reasonable because ␣-Fe2 O3 has a higher thermodynamic stability compared with ␣-FeOOH. The results are consistent with the previous study [27]. However, the higher photodegradation rates of PCP were obtained in the ␣-Fe2 O3 system, rather than in the ␣FeOOH system. It implies that the dissolved Fe(III)/Fe(II) species is not the critical factor to determine the photocatalytic activity of iron oxide in UV/OX/IO systems. Zhao’s group also reported that in the UV/H2 O2 /IO system, different iron oxides with almost the same amounts of dissolved Fe varied in photoactivity [34]. As such, the adsorbed Fe(III)/Fe(II) species and/or the surface structure of iron oxides may play important roles because iron cycling simultaneously occurs on the surface of the iron oxide and in the bulk solution. Unfortunately, various structures and different crystallinities of iron oxides make it difficult to find an identical extraction method to quantify the adsorbed Fe(III)/Fe(II) properly on the surface of the iron oxide. In further research, the exact interface mechanism on the surface of iron oxides needs to be explored. 3.4. Formation of hydrogen peroxide In the Fe(III)–oxalate system, H2 O2 is a very important active species because the most reactive oxidant, • OH, is acquired only through the reaction of H2 O2 with Fe(II). In this system, H2 O2 can be generated in the following ways [10,12]: FeII + O2 •− + 2H+ → FeIII + H2 O2 II

Fe

+ • OOH + H+

• OOH + O •− 2

→ Fe

III

+ H2 O2

+

+ H → O2 + H2 O2

• OOH + • OOH

→ O2 + H2 O2

k = 107 –108 M−1 s−1 −1 −1

5

k = 10 M

s

(2) −1 −1

7

k = 9.7 × 10 M 5

−1 −1

k = 8.3 × 10 M

s

(1)

s

(3) (4)

Simultaneously, H2 O2 is consumed by the following reactions [10,12]: FeII + H2 O2 → FeIII + OH− + • OH

k = 53 M−1 s−1

(5)

FeII (C2 O4 ) + H2 O2 → FeIII (C2 O4 )+ + OH− + • OH k = 3.1 × 104 M−1 s−1

(6)

1214

Q. Lan et al. / Chemical Engineering Journal 168 (2011) 1209–1216

A

B

α−Fe2O3

α−FeOOH

-1

Concentration of H2O2 (mg L )

3.0

2.4

UV / IO / OX / pH 3.5 UV / IO / OX / pH 5.0 UV / IO / OX / pH 7.0 UV / IO / pH 3.5

1.8

1.2

0.6

0.0 0

20

40

60

0

20

40

60

Reaction time (min) Fig. 8. The formation of H2 O2 during PCP photodegradation at various initial pH values in different iron oxide (IO) systems.

• OH + H

2 O2

→ H2

O + • OOH

7

−1 −1

k = 3 × 10 M

s

(7)

Thus, H2 O2 concentration depends on both generation and consumption rates. To some extent, H2 O2 concentration can be used as an indicator to evaluate the activity of systems [18,19]. In the absence of oxalic acid, no H2 O2 was detected within 60 min, as shown in Fig. 8. On the other hand, in the presence of oxalic acid, the variation of H2 O2 shows a pattern of rising and then falling, and its concentration was largely dependent on pH. With the increase of initial pH, the in situ formation of H2 O2 diminished in the two iron oxides systems. At an initial pH of 3.5, the concentration of H2 O2 was the highest. At an initial pH of 5, the concentration of H2 O2 dropped markedly. At an initial pH of 7, the H2 O2 concentration was detected to be quite small. Additionally, more H2 O2 was formed in situ in the ␣-Fe2 O3 system than in the ␣-FeOOH system. These results all coincide with the kinetics of PCP photodegradation, as described in Section 3.1. 3.5. pH-dependent photodegradation mechanism of PCP With the results above, the difference in the rates of PCP photodegradation at different initial pH values could be understood. First, the presence of oxalic acid significantly influences the surface charge of iron oxides and gives the iron oxides a negative charge at pH values above 3, which then affects the PCP adsorption. A low pH is beneficial to PCP adsorption in the presence of oxalic acid and might result in a more rapid PCP degradation. Second, iron oxide adsorbs oxalic acid to form the photoactive Fe(III)–oxalate complexes that contribute the photodegradation of PCP. Third, pH influences the generation of H2 O2 . Low pH favors the formation of more H2 O2 in situ in the Fe(III)–oxalate system, further accelerating the PCP photodegradation. Moreover, Fig. 1 shows that the changing inflection point of PCP concentration, oxalic acid concentration, and the pH value all occurred at 25 min in the UV/OX/␣-Fe2 O3 system and 10 min in the UV/OX/␣-FeOOH system at initial pH 3.5. Obviously, this same trend implies that the same dominant reaction mechanism was present at an initial pH of 3.5 in UV/OX/IO systems. In the control system of UV/OX, the degradation of oxalic acid followed a different pattern and the increase of pH was less. Specifically, in UV/OX/IO systems, the concentration of oxalic acid decreased significantly at an initial pH of 3.5, whereas at the other initial pH

values, the concentration changed slowly. At the same time, the pH values increased sharply during the photoreaction at the initial pH of 3.5, whereas at the initial pH values of 5.0 and 7.0, the pH varied much more slowly. These results indicate that another dominant mechanism is present at initial pH values 5.0 and 7.0 in UV/OX/IO systems. In fact, the degradation rate of oxalic acid could reflect the amount of • OH in the UV/OX/IO systems because oxalic acid can be effectively degraded by • OH at a high rate of 106 –107 M−1 s−1 [12]. A more rapid degradation of oxalic acid corresponded to a higher amount of • OH. As indicated by the rapid decrease in oxalic acid concentration, the attack of • OH is mainly responsible for both the degradation of PCP and oxalic acid at the initial pH of 3.5. In contrast, at the initial pH values of 5.0 and 7.0, the ability of iron oxide systems to generate • OH weakens as indicated by the slow degradation of oxalic acid, and thus the PCP degradation is mainly be attributed to direct photolysis. This case is in agreement with the generation of H2 O2 during PCP photodegradation. Besides, with a rising initial pH, the direct photolysis of PCP could be accelerated [29], although the amount of H2 O2 and • OH drops. Ultimately, PCP was photodegraded at a similar rate at the initial pH values of 5.0 and 7.0. On the other hand, the formation of • OH in the systems is strongly dependent on pH. According to the method by Panias et al. [33], at different initial pH values with an initial oxalic acid concentration of 1.2 mM, Fe(III) as Fe(C2 O4 )3 3− and Fe(II) as Fe(C2 O4 )2 2− were the dominant species, both of which are highly active. The former has the higher quantum yield of photolysis [11,17] and the latter can react with H2 O2 to form • OH at much faster rates than the free Fe2+ in the solution [10,12]. However, at higher pH values, the presence of highly active Fe–oxalate complexes did not accelerate PCP photodegradation or H2 O2 formation, as shown in Figs. 2 and 8, respectively. Similarly, Mazellier and Sulzberger reported that in the photo/OX/␣-FeOOH system, the degradation rate of diuron sharply dropped at pH values above 4.0, although the more stable and active species of Fe(C2 O4 )3 3− , Fe(C2 O4 )2 − , and Fe(C2 O4 )2 2− were the dominant species in those cases [17]. They pointed out that the pH dependence of the efficiency of Fe(II) detachment from the crystal lattice of ␣-FeOOH is the primary cause [17]. We consider the hindrance of higher pH to iron cycling to be an important reason. At high pH values (>4.0), the oxidation of Fe(II) becomes much faster than at low pH. It means that a high pH is helpful to a rapid diminishing of Fe(II), thereby slowing down the iron

Q. Lan et al. / Chemical Engineering Journal 168 (2011) 1209–1216

8

GC/MS

25

Cl

Cl

2.0

20

OH

3

Cl

Cl

Cl

Cl OH

Cl

O

Cl

Cl Cl Cl

Cl

Cl

Cl

O

Cl Cl

Cl

10 Cl

Cl

OH

CH3 C OH O

5

H

0.5

Cl

Cl

O

1.0

O

OH

2

OH Cl

5

-4

OH

15

O

1.5

mv (10 )

5

Au (10 )

5

Intensity (10 )

O

4

OH

O

Cl Cl

6

IC

HPLC

OH Cl

7

1215

C

OH

Cl

1

0 0

0.0 10

15

20

25

3

6

9

12

15

3

6

9

12

15

Retention time (min)

3.6. Intermediates from PCP degradation To confirm the difference in the mechanisms above, the intermediates during PCP photodegradation were measured. At initial pH 3.5, six intermediates were identified, namely tetrachlorocatechol (TeCC) and 2,3,5,6-tetrachloro-1,4-hydroquinone (TeCHQ) by GC/MS, tetrachloro-o-benzoquinone (o-chloranil) and tetrachlorop-benzoquinone (p-chloranil) by HPLC, and HCOOH and CH3 COOH by IC. At the initial pH values 5.0 and 7.0, four intermediates of TeCC, TeCHQ, o-chloranil, and p-chloranil were detected, but HCOOH and CH3 COOH were not found. The chromatograms of the detected intermediates are shown in Fig. 9. Among these intermediates, TeCC, TeCHQ, and p-chloranil have been reported by Benitez et al. as the intermediates of PCP during direct photolysis [29]. These results imply that the mineralization of PCP at an initial pH of 3.5 was higher than at initial pH values of 5.0 and 7.0, and the formation of HCOOH and CH3 COOH at low initial pH resulted from the • OH attack. The intermediate p-chloranil can be quantified due to its relative stabilization. As shown in Fig. 10, in the ␣-Fe2 O3 system, the accumulated concentration of p-chloranil initially increased and then decreased, suggesting an obvious degradation process. In contrast, in the ␣-FeOOH system, the concentration of p-chloranil increased quickly at the initial stage and further increased slowly until a plateau. It appears that the degradation rate of p-chloranil was slower than its generation rate, which should be attributed to the insufficiency of active species in the ␣-FeOOH system. The direct photolysis of PCP might form more toxic intermediates, including polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans [6]. These were not found in this study, which may be due to their amounts being lower than the limits of detection. In the study of Fukushima and Tatsumi, humic acids could effectively repress the production of octachloro-dibenzo-p-dioxin during PCP degradation in the photo-Fenton system, which can be attributed to the radical coupling between phenoxy radicals formed by the attack of • OH radicals to PCP and phenolic moieties in

-1

redox cycling process. As a result, the generation of active species, such as H2 O2 and • OH, is inhibited at higher initial pH. Therefore, at pH 3.5, the mechanism of • OH attack is dominant for the PCP photodegradation in the Fe(II)–oxalate system, whereas direct photolysis becomes dominant at the higher pH values of 5.0 and 7.0.

Concentration of P-Chloranil ( mg L )

Fig. 9. The chromatograms of the detected intermediates by GC/MS, HPLC and IC.

1.2

A

B

α−Fe2O3

α−FeOOH

0.9

0.6

0.3 UV / IO / OX / pH 3.5 UV / IO / OX / pH 5.0 UV / IO / OX / pH 7.0

0.0 0

20

40

60

0

20

40

60

Reaction time (min) Fig. 10. The accumulated concentrations of p-chloranil during PCP photodegradation at various initial pH values. (A) ␣-Fe2 O3 and (B) ␣-FeOOH.

humic acids [9]. Future investigations should pay close attention to the detection of polychlorinated products because even very small amounts can cause tremendous hazards. Anyway, the involvement of HCOOH and CH3 COOH at pH 3.5 and their absence at pH values 5 and 7 helps confirm the different mechanisms working in the Fe(III)–oxalate systems at pH 3.5 and at pH values 5.0 and 7.0. 4. Conclusions PCP can be effectively photodegraded in iron oxide/oxalate systems containing hematite (␣-Fe2 O3 ) and goethite (␣-FeOOH). The different initial pH values led to different rates of PCP photodegradation. The PCP photodegradation at pH 3.5 proceeded more rapidly than at pH values 5.0 and 7.0. The low pH not only enhanced PCP adsorption in the presence of oxalic acid, but also enhanced H2 O2 formation. Thus, the low pH leads to a dominant • OH attack mechanism. In contrast, a high pH leads to a dominant direct photolysis mechanism. The intermediates involved during photodegradation at pH 3.5, including HCOOH and CH3 COOH, that are not detected during photodegradation at pH values 5.0 and 7.0 confirmed the pH-dependence of PCP photodegradation. Therefore, pH plays an important role in PCP photodegradation via irradiated iron oxide/oxalate systems. With better understanding of such systems,

1216

Q. Lan et al. / Chemical Engineering Journal 168 (2011) 1209–1216

the findings in this study serve as a reminder that great care should be taken in any further development of engineered systems to treat water and wastewater containing PCP. Acknowledgments We are particularly grateful to Prof. X.Z Li for many valuable suggestions and comments. This work was supported by the National Science Foundation of China (Nos. 40771105, 50978260, and 21077136), China Postdoctoral Science Foundation Funded Project (No. 20100470954), Guangdong Natural Science Foundation (No. 10451027501005705) and Youth Teacher Initial Funding of Sun Yat-sen University (No. 31000-3126170). References [1] D.F. Goerlitz, D.E. Troutman, E.M. Godsy, B.J. Franks, Migration of woodpreserving chemicals in contaminated groundwater in a sand aquifer at Pensacola, Florida, Environ. Sci. Technol. 19 (1985) 955–961. [2] J.G. Mueller, D.P. Middaugh, S.E. Lantz, P.J. Chapman, Biodegradation of creosote and pentachlorophenol in contaminated groundwater: chemical and biological assessment, Appl. Environ. Microbiol. 57 (1991) 1277–1285. [3] H.K. Young, R.C. Elizabeth, Dechlorination of pentachlorophenol by zero valent iron and modified zero valent irons, Environ. Sci. Technol. 34 (2000) 2014–2017. [4] F.B. Li, X.G. Wang, Y.T. Li, C.S. Liu, F. Zeng, L.J. Zhang, M.D. Hao, H.D. Ruan, Enhancement of the reductive transformation of pentachlorophenol by polycarboxylic acids at the iron oxide–water interface, J. Colloid Interf. Sci. 321 (2008) 332–341. [5] M. Fukushima, K. Tatsumi, K. Morimoto, Influence of iron(III) and humic acid on the photodegradation of pentachlorophenol, Environ. Toxicol. Chem. 19 (2000) 1711–1716. [6] S. Vollmuth, A. Zajc, R. Niessner, Formation of polychlorinated dibenzop-dioxins and polychlorinated dibenzofurans during the photolysis of pentachlorophenol-containing water, Environ. Sci. Technol. 28 (1994) 1145–1149. [7] L. Yin, J. Niu, Z. Shen, J. Chen, Mechanism of reductive decomposition of pentachlorophenol by Ti-doped ␤-Bi2 O3 under visible light irradiation, Environ. Sci. Technol. 44 (2010) 5581–5586. [8] L. Yin, Z. Shen, J. Niu, J. Chen, Y. Duan, Degradation of pentachlorophenol and 2,4-dichlorophenol by sequential visible-light driven photocatalysis and laccase catalysis, Environ. Sci. Technol. 44 (2010) 9117–9122. [9] M. Fukushima, K. Tatsumi, Degradation pathways of pentachlorophenol by photo-Fenton systems in the presence of iron(III), humic Acid, and hydrogen peroxide, Environ. Sci. Technol. 35 (2001) 1771–1778. [10] M.E. Balmer, B. Sulzberger, Atrazine degradation in irradiated iron/oxalate systems: effects of pH and oxalate, Environ. Sci. Technol. 33 (1999) 2418–2424. [11] B.C. Faust, R.G. Zepp, Photochemistry of aqueous iron(III)–polycarboxylate complexes: roles in the chemistry of atmospheric and surface waters, Environ. Sci. Technol. 27 (1993) 2517–2522. [12] J.S. Jeong, J.Y. Yoon, pH effect on OH radical production in photo/ferrioxalate system, Water Res. 39 (2005) 2893–2900. [13] Y.G. Zuo, J. Hoigné, Evidence for photochemical formation of H2 O2 and oxidation of SO2 in authentic fog water, Science 260 (1993) 71–73.

[14] Y.G. Zuo, J. Hoigne, Formation of hydrogen peroxide and depletion of oxalic acid in atmospheric water by photolysis of iron(III)-oxalato complexes, Environ. Sci. Technol. 26 (1992) 1014–1022. [15] F.B. Li, X.Z. Li, C.S. Liu, X.M. Li, T.X. Liu, Effect of oxalate on photodegradation of bisphenol A at the interface of different iron oxides, Ind. Eng. Chem. Res. 46 (2007) 781–787. [16] C.S. Liu, F.B. Li, X.M. Li, G. Zhang, Y.Q. Kuang, The effect of iron oxides and oxalate on the photodegradation of 2-mercaptobenzothiazole, J. Mol. Catal. A: Chem. 252 (2006) 40–48. [17] P. Mazellier, B. Sulzberger, Diuron degradation in irradiated, heterogeneous iron/oxalate systems: the rate-determining step, Environ. Sci. Technol. 35 (2001) 3314–3320. [18] Q. Lan, F.B. Li, C.X. Sun, C.S. Liu, X.Z. Li, Heterogeneous photodegradation of pentachlorophenol and iron cycling with goethite, hematite and oxalate under UVA illumination, J. Hazard. Mater. 174 (2010) 64–70. [19] Q. Lan, F.B. Li, C.S. Liu, X.Z. Li, Heterogeneous photodegradation of pentachlorophenol with maghemite and oxalate under UVA illumination, Environ. Sci. Technol. 42 (2008) 7918–7923. [20] E. Espejo, E. Agosin, Production and degradation of oxalic acid by brown rot Fungi, Appl. Environ. Microbiol. 57 (1991) 1980–1986. ˜ F. Torrades, L.A. Pérez-estrada, I. Oller, S. Malato, M.I. Mal[21] J. GarcÍa-Montano, donado, J. Peral, Degradation pathways of the commercial reactive azo dye Procion Red H-E7B under solar-assisted photo-Fenton reaction, Environ. Sci. Technol. 42 (2008) 6663–6670. [22] A.W. Vermilyea, B.M. Voelker, Photo-Fenton reaction at near neutral pH, Environ. Sci. Technol. 43 (2009) 6927–6933. [23] Y. Zhang, N. Kallay, E. Matijevic, Interactions of metal hydrous oxides with chelating agents. VII. Hematite–oxalic and citric acid systems, Langmuir 1 (1985) 201–206. [24] D. Panias, M. Taxiarchou, I. Paspaliaris, A. Kontopoulos, Mechanisms of dissolution of iron oxides in aqueous oxalic acid solutions, Hydrometallurgy 42 (1996) 257–265. [25] M.I. Litter, E.C. Baumgartner, G.A. Urrutla, M.A. Blesa, Photodissolution of iron oxides. 3. Interplay of photochemical and thermal processes in maghemite/carboxylic acid systems, Environ. Sci. Technol. 25 (1991) 1907–1913. [26] C. Siffert, B. Sulzberger, Light-induced dissolution of hematite in the presence of oxalate: a case study, Langmuir 7 (1991) 1627–1634. [27] B. Sulzberger, H. Laubscher, Reactivity of various types of iron(III) (hydr)oxides towards light-induced dissolution, Mar. Chem. 50 (1995) 103–115. [28] M.A. Oturan, N. Oturan, C. Lahitte, S. Trevin, Production of hydroxyl radicals by electrochemically assisted Fenton’s reagent: application to the mineralization of an organic micropollutant, pentachlorophenol, J. Electroanal. Chem. 507 (2001) 96–102. [29] F.J. Benitez, J.L. Acero, F.J. Real, J. García, Kinetics of photodegradation and ozonation of pentachlorophenol, Chemosphere 51 (2003) 651–662. [30] R.M. Cornell, U. Schwertmann, The Iron Oxides: Structure, Properties, Reactions, Occurrences and Uses, third ed., Wiley-VCH, 2003. [31] R.M. Cornell, P.W. Schindler, Infrared study of the adsorption of hydroxycarboxylic acids on ␣-FeOOH and amorphous Fe(III)hydroxide, Colloid Polym. Sci. 258 (1980) 1171–1175. [32] W.Y. Shiu, K.C. Ma, D. Varhanickova, D. Mackay, Chlorophenols and alkylphenols: a review and correlation of environmentally relevant properties and fate in an evaluative environment, Chemosphere 29 (1994) 1155–1224. [33] D. Panias, M. Taxiarchou, I. Douni, I. Paspaliaris, A. Kontopoulos, Thermodynamic analysis of the reactions of iron oxides: dissolution in oxalic acid, Can. Metall. Q. 35 (1996) 363–373. [34] J. He, W.H. Ma, J.J. He, J.C. Zhao, J.C. Yu, Photooxidation of azo dye in aqueous dispersions of H2 O2 /␣-FeOOH, Appl. Catal. B: Environ. 39 (2002) 211–220.