Photo-oxidation of p-arsanilic acid in acidic solutions: Kinetics and the identification of by-products and reaction pathways

Chemical Engineering Journal 243 (2014) 364–371

Contents lists available at ScienceDirect

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

Photo-oxidation of p-arsanilic acid in acidic solutions: Kinetics and the identification of by-products and reaction pathways Marianna Czaplicka a,⇑, Łukasz Bratek a, Katarzyna Jaworek a, Jan Bonarski b, Sylwia Pawlak b a b

Institute of Non Ferrous Metals, Sowinskiego 5 Str, Gliwice, Poland Institute of Metallurgy and Materials Sciences, Polish Academy of Sciences, Reymonta 25 Str, Krakow, Poland

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Pseudo-first order rate constants are

determined for each processes.  The degradation pathways are

proposed for ozonation and photoozonation.  The degradation pathways are proposed for oxidation and photooxidation by H2O2.  Formation of solid phases are observed.  Characterization of solid phases were done.

a r t i c l e

i n f o

Article history: Received 18 September 2013 Received in revised form 19 December 2013 Accepted 8 January 2014 Available online 17 January 2014 Keywords: Organoarsenic compounds Photodegradation Photo-oxidation Mechanism Solid phase

a b s t r a c t Arsenic compounds have been used extensively in agriculture for applications ranging from cotton herbicides to animal feed supplements. While studying the kinetics and mechanism of p-arsanilic acid degradation via oxidation, the photodegradation and photo-oxidation processes that occurred while using irradiation at 254 nm were assessed. Ozone and hydrogen peroxide were used as oxidants. The relationships among the process efficiency, the process conditions and the type of oxidiser were demonstrated. The rate constants for the decomposition based on pseudo first order kinetics were 0.76  10 3 min 1 during photodegradation, 27.85  10 3 min 1 during ozonation and 35.1  10 3 min 1 during photoozonation, 32.3  10 3 min 1 during oxidation with H2O2 and 36.4  10 3 min 1 during photo-oxidation with UV/H2O2. After identifying the major products, degradation mechanisms were proposed. During photodegradation, oxidation and photo-oxidation with H2O2, the generation of a solid phase composed primarily of As3O5(OH) was observed. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction Arsenic can occur naturally, but it may also be found in products used during various industrial processes, including leather and wood treatment, as well as in pesticides [1–3]. Anthropogenic arsenic contamination mainly arises from the production of metals and alloys, the refining of petroleum and the burning of fossil fuels and wastes. The concentration of arsenic in waters attributed to non-ferrous metallurgy may reach 16 g L 1 [4]. Depending on its ⇑ Corresponding author. Tel.: +48 23 238 0267. E-mail address: [email protected] (M. Czaplicka). 1385-8947/$ - see front matter Ó 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2014.01.016

redox potential (Eh) and pH, arsenic can take three forms in water. A high redox potential facilitates the generation of stable forms, such as H3AsO4, H2 AsO4 , HAsO24 at lower Eh values, the stable form is represented by metallic arsenic. Numerous examinations of natural waters have enabled the identification of 25 various arsenic compounds. In the natural environment, organic compounds containing arsenic are usually due to biological activity or the influence of copper metallurgy on the environment. Also, groundwater and soil contain organoarsenic species: monomethylarsenic acid (MMA), dimethylarsenic acid (DMA), trimethylarsine oxide and trimethyl arsine [5]. The presence of other organic compounds containing

M. Czaplicka et al. / Chemical Engineering Journal 243 (2014) 364–371

arsenic in the natural environment was also established, including phenylated arsenic compounds (e.g., 4-hydroxy-3-nitrophenylarsenic acid (roxarsone), 4-aminophenylarsenic acid (para-arsanilic acid)) [6]. Organoarsenic compounds typically exist in a pentavalent oxidation state and are introduced into the environment due to agricultural applications. Para-arsanilic acid (PPA) and roxarsone are used as animal feed additives for both pigs and chickens, while nitarsone and carbarsone are used to control blackhead disease in turkeys [7]. Numerous studies have covered the transformation of inorganic arsenic compounds during natural processes. Numerous oxidative methods have been employed to convert arsenite to arsenate under various conditions, such as with H2O2 [8,9], UV/H2O2, oxygen and ozone (O3), MnO2, Fenton’s reagent (Fe(II)/H2O2) or UV/Fe(III) [10–12]. However, each method has specific limitations and disadvantages. Photocatalytic processes for arsenic species are widely discussed in the literature, but few processes have been applied to the oxidation of organic arsenic species, such as hydrothermal treatments [13,14], TiO2 photocatalysis [13–17], and radiolysis [18]. TiO2 photocatalysis is an effective method for oxidising of organic arsenic species, such as MMA, DMA, and phenylarsonic acid [19–23]. Due to the use of PAA as an animal feed additive, particularly in the USA [7] and its presence in the process stream that enters the environment [4], determining the conditions of the degradation of PAA is important. The main objective of this study was to determine the kinetics and mechanisms of photodegradation, specifically the oxidation and photo-oxidation of p-arsanilic acid in an acidic environment. Ozone and hydrogen peroxide were used as oxidising agents. The efficiency of the processes depend the on process conditions and type of oxidiser. During photodegradation, oxidation and photo-oxidation with hydrogen peroxide, a solid phase mostly composed of As3O5(OH) was generated. The results indicated that the AOP processes in an acidic environment can be used to remove arsenic from water as an insoluble precipitate.

365

2.1. Analytical procedures The para-arsanilic acid (PPA) was determined using HPLC with a Perkin Elmer Series 200 DAD detector (Perkin Elmer, USA) and an Aquapore OD-300 (7 lm, 250  4.6 mm) column (Perkin Elmer, USA) in a reversed phase system. The isocratic elution was performed using 50 mM pH 2.6 phosphate buffer (potassium dihydrogenphosphate and water with the pH adjusted with phosphoric acid to pH 2.6) as a mobile phase. The determination was made at k = 260 nm wavelength. The oxidation and photo-oxidation products of PPA were determined using gas chromatography coupled with a mass detector (GC/MS) with Perkin Elmer equipment. Solid phase microextractions were used to separate the analytes from the solutions by applying fibres with varying polarities: 65 lm polydimethylsiloxane divinylbenzene (PDMS/DVB), 75 lm Carboxen (CAR), 85 lm polyacrylate (PA) [24,25]. The process products were identified based on the retention time and mass spectrum of the compounds. A spectrophotometric method was applied to determine the ammonium ions. The arsenic in solid phase was determined after mineralisation in a mixture of nitric(V) and sulphuric(VI) acids using atomic absorption spectroscopy with atomisation in a graphite furnace (GFAAS) through a direct method. 2.2. X-ray diffraction analysis of solid phase The X-ray phase analysis of the solids was based on X-ray diffraction patterns recorded with a Bruker D2 Phaser (Bruker Cooperation, USA) diffractometer equipped with an X Flash detector and energy-filtered radiation. In the experiments Cu Ka-serie radiation was applied. The phases were identified with EVA software delivered by Bruker Comp. and the ICDD crystallographic database (PDF + 4 package). Based on the experimental patterns, the volume fractions of the identified phases were estimated.

2. Materials and methods

3. Results and discussion

The studies were performed using a 8.9 mmol L 1 para-arsanilic acid (Supelco Analytical, USA) solution, corresponding to 4.4 mmol L 1 arsenic. The base solution was prepared with demineralised water, and the pH was corrected to approximately 2 with hydrochloric acid (1:1). The photodegradation and photo-oxidation studies were conducted with a photoreactor (Kendrolab laboratory UV ‘‘System 4’’ reactor, Heraeus, Germany) and a low-pressure mercury-discharge lamp (14 W) cooled by a water jacket emitting at 254 nm. The oxidising agents (ozone and hydrogen peroxide) were introduced into the reactor. A predefined dose of H2O2, 50 mL (18.5 g) per 1 L of the input solution with 4.4 mmol L 1 arsenic was added to the reactor in 10 equal batches at 5 min intervals. A Wofil Ozone Technology (Ozonia, Switzerland) O1 ozone generator with a 5 g O3 m 3 capacity, corresponding to 12 g O3 h 1, was supplied with oxygen from a cylinder and was used to provide 2.2 g O3 h 1 to the reaction. During photodegradation, ozonation, oxidation with hydrogen peroxide and the advanced oxidation processes (UV/O3 and UV/ H2O2) the pH, redox potential (Eh), conductance and temperature were continuously measured. Studies regarding the photodegradation, oxidation and photo-oxidation were conducted for 360 min. After each cycle, the reaction solutions were stored in the dark for 72 h. After a solid phase was generated, the solutions were filtered; the residue was dried at 50 °C. The duration used to age the solutions was determined using the kinetics of their sedimentation.

The efficiency of PPA decomposition by photolysis, oxidation, UV/O3 and UV/H2O2 processes in acidic environment depends on the conditions, the type of process and the oxidant. Of the examined processes, photodegradation is the least efficient for PPA decomposition; only 17.7% PPA was removed, and the constant of decomposition rate under pseudo first order kinetics was 0.76  10 3 min 1 (Table 1). The rate constants for ozonation and photo-ozonation were 27.89  10 3 min 1 and 32.3  10 3 min 1, respectively. In Fig. 1, the linear relations of ln(ct/co) versus t are presented. Using hydrogen peroxide as the oxidiser resulted in 100% decomposition of the arsenic compound after 180 min. The process ran at a constant rate of 32.3  10 3 min 1. Upon comparison between ozonation and the oxidation with hydrogen peroxide, hydrogen peroxide is a more efficient oxidiser for PPA than ozone under the examined conditions (Table 1). The slower rate of ozonation relative to the oxidation with H2O2 was due to the pH dependence of the ozone form. At pH 2, the ozone remains in its molecular state and can participate in the direct oxidation reactions. Hoigné and Bader [26] found out that under acidic conditions, the direct oxidation with molecular ozone is the most important. Direct oxidations with aqueous ozone are relatively slow compared to indirect oxidations with hydroxyl free radicals. Using a UV/H2O2 system resulted in 100% decomposition of PPA after 140 min, achieving a rate constant of 36.4  10 3 min 1. The rate of PPA photo-oxidation with H2O2 exceeded that of photo-ozonation.

366

M. Czaplicka et al. / Chemical Engineering Journal 243 (2014) 364–371

Table 1 Results of photodegradation, oxidation and photooxidation of PPA. Agent process UV O3 UV/O3 H2O2 UV/H2O2

Constant rate of decomposition (k min

1

)

3

0.76  10 27.85  10 3 35.08  10 3 32.3  10 3 36.4  10 3

Half-life t1/2 0.9  10 24.9 19.8 21.4 19.0

3

R2

Initial pH

Final pH

0.976 0.992 0.990 0.984 0.989

2.49 2.24 2.17 2.18 2.24

2.13 1.67 1.95 2.03 1.99

Fig. 1. The relationship ln(c/co) vs. time.

The rate constants for photo-oxidations with ozone and H2O2 are higher than the processes with only those oxidisers, revealing a synergistic effect with irradiation. The significant impact of radiation on the process rate is particularly evident when comparing ozonation with photo-ozonation. In acidic media, ozone selectively attacks organic species, usually with small rate constants. When the solution is irradiated using UV light (UV/O3 system), the pollutants degrade faster and more non-selectively because they react with primarily hydroxyl radicals. Therefore, photo-ozonation involves slower, more selective oxidation reactions involving molecular ozone and less selective indirect oxidation reactions. The halflife of PPA degradation at pH 2 were 0.9  10 3, 24.9, 19.8, 21.4 and 19.0 min for photodegradation, ozonation, UV/O3, oxidation by H2O2 and UV/H2O2, respectively (Table 1). According to the half-life for PAA (Table 1), the photodegradation was not as fast as the oxidation processes. Numerous reports have presented the half-life for As(III) with various oxidising agents at different pH [8,9,27]; however, scarce data are available regarding the half-life for PAA or As(V) during oxidation and photo-oxidation processes. For example, Lescano et al. [27] obtained a 3.5 s half-life for the As(III) photo-oxidation process at initial pH values from 5.6 to 6.7 and 253.7 nm with 20 mg/L H2O2. The significantly different half-lives for the photo-oxidation of As(III) and PAA most likely arose from the differences in the degree of oxidation for As (As in PAA exists in the oxidation state V), the pH and the form of the investigated compound. In all processes, a decrease in pH was observed. Oxidising conditions (Eh > 200 mV) are dominant in the reaction solution for all of the discussed processes; however, the observed changes in the redox potential in the solution after the reaction over time varied between the processes. That parameter decreased during photodegradation for the first hour, remaining constant for the remaining time. The nearly constant value of Eh (approximately 550 mV) is a characteristic feature during oxidations and photo-oxidation with H2O2; during ozonation and photo-ozonation, significant changes in Eh over time were observed. During the first 10 min, a 75-mV decrease Eh was observed, and in the remaining time, a strong increase of that parameter was registered (Fig. 2). It should be noted, however, that changes of Eh were larger in photo-ozonation than in ozonation. In aqueous envi-

Fig. 2. Redox potential vs. time.

ronments, ozone can react according to two mechanisms: (i) direct reaction and (ii) ozone decomposition to generate the .OH that participates in oxidation reactions. The relationship between Eh increases and the time needed for ozonation and photo-ozonation are most likely caused by the stabilisation of O3 in the reaction environment. When pH < 3, ozone does not decompose, allowing hydroxyl radical generation to occur. Consequently, the increased redox potential of the solutions during ozonation and photo-ozonation results from the dissolved ozone in the reaction solutions and the products of the oxidation. Conductivity was the second physicochemical parameter that was continuously measured during the processes. That parameter increased over time during the photodegradation, ozonation and UV/O3 processes (Fig. 3) because easily dissociated intermediate products of primary and secondary reactions formed. The increased conductivity strictly depends on the type of process; the change is larger during ozonation and UV/O3 processes than photodegradation. Using H2O2 with and without light did not increase the conductivity.

Fig. 3. Conductivity vs. time.

367

M. Czaplicka et al. / Chemical Engineering Journal 243 (2014) 364–371

A precipitated solid phase was observed for photodegradation, oxidation and photo-oxidation with hydrogen peroxide, but not ozonation or photo-ozonation. The formation of a residue that was insoluble in the acidic environment indicates diverse mechanisms for PPA decomposition exist, depending on the type of the oxidant.

NH3+

NH3 HO hv

+

As

OH

pH =2,5, H2O OH

3.1. Product identification and reaction mechanism To identify the reaction products, usually gas and liquid chromatography methods coupled with a mass spectrometer, NMR [4,5,13] and theoretical methods [28] are used. A qualitative analysis of the solutions after the reaction using GC/MS showed products with different molecular structures, depending on the PPA degradation process. 3.2. Photodegradation During the PPA photodegradation, aniline was identified as the main degradation product in the post-reaction solutions. A curve representing the kinetics of aniline formation is shown in Fig. 4. One possible mechanism for the PPA photodegradation is based on the photolytic reductive cleavage of the As–C bond, leading to the generation of aniline and inorganic arsenic compounds. The strongly acidic environment hinders any further photodegradation of aniline, explaining the lack of any other process products. The amine group becomes protonated at low pHs, deactivating the ring. Tang et al. [29] established that in acidic environments with exposure to 254 nm radiation, only 5% of the compound undergoes photodegradation. Concurrently, the formation of a solid phase was observed during the process. Fig. 5 shows the proposed reaction pathway. 3.3. Ozonation and photo-ozonation The decomposition of ozone in water depends strongly on the pH of the water. The reaction under acidic conditions is dominated by the action of molecular ozone. Ozone molecules can react in several ways with organic compounds: utilising ozone as a dipole (Criegee mechanism), an electrophilic reagent or a nucleophilic reagent. Molecular ozone can react with aromatic compounds at both the aromatic ring and the substituents. The reactivity of ozone is very limited (standard reduction potential E0(O3/H2O) = +2.07 V). At pH < 3, hydroxyl radicals do not affect the decomposition of ozone. The mechanism of the ozone-mediated oxidation reaction strongly depends on the number of acidic hydrogen atoms on the

HO

As

O

OH Fig. 5. Pathway proposed for photodegradation of PPA.

nitrogen atom and the activity of the carbon atom(s) surrounding the amino group, respectively. The 1-chloro-4 nitrobenzene was the primary product identified for both the PPA ozonation and UV/O3 process. The amino group of PPA under acidic conditions remains protonated, forming an electron-withdrawing group. The amino group (–NH2) is electron donating and therefore activates the aromatic ring by increasing its electronic density. Aromatic compounds with more delocalised electrons are more reactive toward ozone. The oxidation of the –NH2 group to form NO2 is the result of a selective reaction with molecular ozone. Similar results were reported by Sarasa et al. [30]: the selective reaction between molecular ozone and –NH2 under acidic conditions forms a nitro group. The chlorine atom present in the product molecule has been substituted for arsenic, indicating As–C bond cleavage and a reaction between the intermediate product and the chloride ions introduced while correcting the pH level. Cl ions can quench hydroxyl radicals, generating chlorinated products, while Claq can facilitate the formation of eaq under UV-irradiation. The lack of other photo-ozonation products, which are formed during indirect oxidation, suggested that direct oxidation is the primary process for ozonation and photo-ozonation in acidic solutions. Because it is soluble in the aqueous environment, H3AsO4 forms; the high redox potential favours the stability of H3AsO4. The proposed reaction mechanism explains the increase in the conductivity over time and the decrease in pH after the reaction compared to the initial pH (Table 1). The constants for the PPA decomposition rate indicate (Table 1) that photo-ozonation is faster than ozonation. The irradiation mainly influences the rate of reaction, as confirmed by the presence of the same products in the ozonation and in UV/O3 processes. The proposed reaction pathway leading to formation of the major ozonation product is presented in Fig. 6. 3.4. Oxidation and photo-oxidation with H2O2

Fig. 4. The kinetics curves of formation of aniline.

The following products were identified in the oxidation and photo-oxidation reactions with H2O2 and PPA: p-nitrophenol, aminophenol, aniline, nitrobenzene, phenol, p-benzoquinone, hydroquinone and NH3. The presence of p-nitrophenol indicates the cleavage of the As–C bond, the addition of a hydroxyl radical to the ring and the simultaneous oxidation of the amine group. The As–C bond cleavage via the addition of a hydroxyl group forms aminophenol. Fig. 7 presents a proposed decomposition pathway for PPA during the primary reactions. The other products identified are generated by the secondary reactions during the oxidation and photo-oxidation of p-nitrophenol. Similar results were reported by Di Paola et al. [31], Hori et al. [32], Alif et al. [33] regarding the photodegradation of nitrophenols. The nitrogroup in in nitroaromatics can be eliminated easily, favouring the electrophilic substitution of hydroxyl radicals at the para position relative to the hydroxyl group [33]. The OH radicals may also react directly at the position with the nitrogroup,

368

M. Czaplicka et al. / Chemical Engineering Journal 243 (2014) 364–371

Fig. 6. Pathway proposed for ozonation of PPA.

NO2 OH + NH 3

+

As

OH

OH hv; pH = 2; H2O2

HO

As

OH

O NH2

The aminophenol decomposes via hydroxyl radicals to generate benzoquinone and ammonia. A similar pathway n was proposed by Kumar and Mathur for the photocatalytic degradation of amines [34] . The NH3 concentration in the post-reaction solutions was 13.8 mg L 1 for oxidation and 48.7 mg L 1 for photooxidation, respectively, confirming the proposed diagram. The presence of phenol and chlorobenzene in the solutions indicates a high degree of decomposition because these compounds are the primary and secondary products of oxidation and photooxidation with H2O2; chlorobenzene is generated using the chlorine ions added to correct the pH. The proposed degradation pathway for PPA is presented in Fig. 8.

OH

OH

+

As

OH

OH OH Fig. 7. Suggested decomposition of PPA in the primary reactions.

generating hydroquinone. Under the experimental conditions, two degradation pathways for nitrophenol are possible: hydroxyl radical oxidation and photolysis. For oxidation, the reaction most likely proceeds according to the pathway proposed by Di Paola et al. [31]. During photo-oxidation, the mechanism proposed by Alif et al. [33] is possible. These mechanisms explain why benzoquinone and hydroquinone may be formed, regardless of the process conditions. The pathway proposed by Di Paola et al. [31] and Alif et al. [33] invoke the NO2 ion or radical as a process products that will create HNO3 during reactions with water. The denitration of nitrophenol generates inorganic nitrogen species.

3.5. Characteristics of the solid phase The generation of insoluble residues in acidic environment was observed only during the photodegradation, oxidation with H2O2 and photo-oxidation with UV/H2O2 of PPA. The solid phase was generated throughout the process; however, the residue continued to form up to 72 h after the process. Therefore, the solid phase was separated from the solutions 72 h after the experiments were finished. The efficiency of the residue generation varies between processes. The most efficient solid phase formation was observed during photo-oxidation, while the lowest was observed during photodegradation. For the latter, 0.04 g L 1 of solid phase was formed, while during the photo-oxidation with hydrogen peroxide, the mass of the precipitated solid phase was 1.7 g L 1 (Table 2). The arsenic accounted for 33.4–36.7% of solid phase mass, depending on the process, and the highest value was observed during the photodegradation process (Table 2). After photodegradation, 1.5% arsenic, which had been introduced into the reaction system as PPA, was removed as a residue. The oxidation of PPA using hydrogen peroxide removed 50% of the

369

M. Czaplicka et al. / Chemical Engineering Journal 243 (2014) 364–371

Fig. 8. Pathway proposed for photooxidation of PPA.

Table 2 Arsenic content in solid phase generated in photodegradation, oxidation and photooxidation of p-arsanilic acid, at pH = 2. Agent process

Mass of residue (g L

UV H2O2 UV/H2O2

0.042 1.491 1.677

1

)

arsenic as a solid phase, while during the analogous photo-oxidation, 57% was removed. The high efficiency of the H2O2 mediated PPA oxidation reached 100% after 160 min, and the precipitation of 50% of the arsenic as a residue suggests that some of the arsenic does not change its oxidation state. The oxidation of As(III) to As(V) occurs between pH 7 and 11.5. The oxidation of As(III) to As(V)

As content in residue (%)

Mass of precipitated As (g L

36.7 33.4 34.2

0.015 0.500 0.573

1

)

under the process conditions for the oxidation and photo-oxidation with hydrogen peroxide results is limited by the reaction environment, specifically the pH < 2 and the presence of As(OH)3 in its non-dissociated form. As proven by Pettine et al. [8], when pH < 2, As (III) only reacts with hydroxyl radicals to a small degree. The oxidation of the As (III) generated by the photolytic reductive

370

M. Czaplicka et al. / Chemical Engineering Journal 243 (2014) 364–371

cleavage of the As–C bond may occur under acidic conditions; the dominant As species (As(OH) and H2AsO ) give values of Gibbs energy DG = 51.2 kcal mol 1 [8]. Due to the diverse selectivities of oxidisers, the formation of the solid phase is stimulated by hydroxyl radicals that did not react with the acid selectively. During ozonation, the acidic environment significantly limits the reactions that generate this type of radical while significantly increasing the oxidation potential of the solu-

tion; therefore, the generated arsenic(V) compound dissolves easily in the acidic environment. The X-ray diffraction spectra of the selected sediments that contain As-compounds were analysed from a Bragg angle of 15–110°. The normalised patterns (intensity vs. 2a angle) of the examined samples are presented in Figs. 9 and 10 and are restricted to the most informative angular range (15–36°). The dominant phases are listed in Table 3.

Fig. 9. X-ray diffraction patterns of the samples (sediments deposited on a paper substrate).

Fig. 10. X-ray diffraction patterns from Fig. 9 with marked lines of the selected phase identified in the examined samples.

371

M. Czaplicka et al. / Chemical Engineering Journal 243 (2014) 364–371 Table 3 The dominating phases identified in the examined samples with estimated volume fraction (estimation error: ±5 vol.%). Identified phase [ICDD code]

Arsenic oxide hydroxide As3O5(OH) [01–075-2110] Hydrogen arsenate hydrate H6(AsO4)2(H2O) [04–013-7680] Carbon amide nitrite C2(NH2)2(NO2)2 [04–013-1416]

Based on the diffraction spectra, the volume fractions of the dominant phases were estimated and are listed in Table 3. The X-ray diffraction analyses of the solid phases indicated that the As3O5(OH), H6(AsO4)2(H2O) and C2(NH2)2(NO2)2 phases dominated the samples after processing. The volume fraction of the As3O5(OH) phase decreased insignificantly, in contrast to the H6(AsO4)2(H2O) phase; this phase decreased dramatically (a few times) after oxidation and photo-oxidation with H2O2 compared to the photodegradation process. In addition, the volume fraction of C2(NH2)2(NO2)2 changed significantly (over two times). The numerical values presented in Table 3 are the results of a rough estimation but still illustrate the significant trends after treatment. The carbon amide nitrite phase was most likely the decomposition product of the secondary products of the processes.

4. Conclusions The efficiency of the PPA degradation depends on the type of process. The highest efficiency is observed during the photo-oxidation with H2O2, while the lowest occurs during photodegradation. The irradiation at 254 nm influences the kinetic oxidation processes and efficiency. The identified products suggested that the processes follow different mechanisms. During ozonation and photo-ozonation, the products suggested that PPA reacts with molecular ozone and chlorine ions to form primarily chloronitrobenzene. During the reactions with H2O2, both p-nitrophenol and aminophenol are generated before undergoing secondary reactions that generate benzoquinone and other products. During the photodegradation, oxidation and photo-oxidation with H2O2, a solid phase was generated that remained insoluble at pH < 2; this material was primarily composed of As3O5(OH), H6(AsO4)2(H2O) and C2(NH2)2(NO2)2.

References [1] D.R. Killelea, J.H. Aldstadt, Identification of dimethylchloroarsine near a former herbicide factory by headspace solid-phase microextraction gas chromatography–mass spectrometry, Chemosphere 48 (2002) 1003–1008. [2] L.X. Yao, G.L. Li, Z. Dang, Z.H. He, C.M. Zhou, B.M. Yang, Phytoavailability of roxarsone and its metabolites for turnip as affected by soil pH, Geoderma 154 (2009) 48–51. [3] X. Dou, Y. Zhang, M. Yang, Y. Pei, X. Huang, T. Takayama, S. Kato, Occurrence of arsenic in groundwater in the suburbs of Beijing and its removal using an iron cerium bimetal oxide adsorbent, Water Qual. Res. J. Can. 41 (2006) 140–146. [4] M. Czaplicka, R. Kurowski, L. Bratek, K. Jaworek, Application of advanced oxidation process for cleaning of industrial water generated in wet dedusting of shaft furnace gases, Environ. Technol. 34 (2013) 1455–1462. [5] A.R. Roerdink, J.H. Aldstadt, Sequential injection absorption spectrophotometry using a liquid-core waveguide: determination of parsanilic acid in natural waters, Anal. Chim. Acta 539 (2005) 181–187. [6] A.R. Roerdink, J.H. Aldstadt, Sensitive method for the determination of roxarsone using solid-phase microextraction with multi-detector gas chromatography, J. Chromatogr. A 1057 (2004) 177–183. [7] D. Chen, H. Zhang, Y. Tao, Y. Wang, L. Huang, Z. Liu, Y. Pan, D. Peng, X. Wang, M. Dai, Z. Yuan, Development of a high-performance liquid chromatography method for the simultaneous quantification of four organoarsenic compounds in the feeds of swine and chicken, J. Chromatogr. B 879 (2011) 716–720. [8] M. Pettine, L. Campanella, F.J. Millero, Arsenite oxidation by H2O2 in aqueous solutions, Geochim. Cosmochim. Acta 63 (1999) 2727–2735.

Volume fraction (vol.%) Photodegradation

Oxidation by H2O2

UV/H2O2

64 16

58 2

56 4

19

40

40

[9] H. Yang, W.-Y. Lin, K. Rajeshwar, Homogeneous and heterogeneous photocatalytic reactions involving As(III) and As(V) species in aqueous media, J. Photochem. Photobiol. A: Chem. 123 (1999) 137–143. [10] S.-H. Yoona, J.H. Leea, S.E. Oh, J.E. Yang, Photochemical oxidation of As(III) by vacuum-UV lamp irradiation, Water Res. 42 (2008) 3455–3463. [11] G.-S. Zhang, J.-H. Qu, H.-J. Liu, R.-P. Liu, G.-T. Li, Removal mechanism of As (III) by a novel Fe Mn binary oxide adsorbent: oxidation and sorption, Environ. Sci. Technol. 41 (2007) 4613–4619. [12] T.V. Nguyen, S. Vigneswaran, H.H. Ngo, J. Kandasamy, H.C. Choi, Arsenic removal by photo-catalysis hybrid system, Sep. Purif. Technol. 61 (2008) 44–50. [13] S. Maeda, A. Ohki, T. Kawabata, M. Kishita, Degradation of organoarsenic compounds by a hydrothermal or photo-oxidation process, Appl. Organometal. Chem. 13 (1999) 121–125. [14] T. Nakajima, T. Kawabata, H. Kawabata, H. Takanashi, A. Ohki, S. Maeda, Degradation of phenylarsonic acid and its derivatives into arsenate by hydrothermal treatment and photocatalytic reaction, Appl. Organometal. Chem. 19 (2005) 254–259. [15] T.L. Xu, P.V. Kamat, K.E. O’Shea, Mechanistic evaluation of arsenite oxidation in TiO2 assisted photocatalysis, J. Phys. Chem. A 109 (2005) 9070–9075. [16] M.N. Chong, B. Jin, C.W.K. Chow, C. Saint, Recent developments in photocatalytic water treatment technology: a review, Water Res. 44 (2010) 2997–3027. [17] A.H. Fostier, M.D.S. Silva Pereira, S. Rath, J.R. Guimaraes, Arsenic removal from water employing heterogeneous photocatalysis with TiO2 immobilized in PET bottles, Chemosphere 72 (2008) 319–324. [18] T. Xu, P.V. Kamat, S. Joshi, A.M. Mebel, Y. Cai, K.E. O’Shea, Hydroxyl radical mediated degradation of phenylarsonic acid, J. Phys. Chem. A 111 (2007) 7819–7824. [19] S.-H. Yoona, S. Leea, T.-H. Kimb, M. Leeb, S. Yub, Oxidation of methylated arsenic species by UV/S2 O28 , Chem. Eng. J. 173 (2011) 290–295. [20] T.L. Xu, Y. Cai, K.E. O’Shea, Adsorption and photocatalyzed oxidation of methylated arsenic species in TiO2 suspensions, Environ. Sci. Technol. 41 (2007) 5471–5477. [21] Z. Xu, C. Jing, F. Li, X. Meng, Mechanisms of photocatalytic degradation of monomethylarsonic and dimethylarsinic acids using nanocrystalline titanium dioxide, Environ. Sci. Technol. 42 (2008) 2349–2354. [22] S. Zheng, Y. Cai, K.E. O’Shea, TiO2 photocatalytic degradation of phenylarsonic acid, J. Photochem. Photobiol. A: Chem. 210 (2010) 61–68. [23] A.J. Bednara, J.R. Garbarinob, I. Ferrera b, D.W. Rutherford, R.L. Wershaw, J.F. Ranvillea, T.R. Wildemana, Photodegradation of roxarsone in poultry litter leachates, Sci. Total Environ. 302 (2003) 237–245. [24] M. Czaplicka, Determination of phenols and chlorophenols in bottom sediments, Chromatographia 53 (2001) S470–S473. [25] M. Czaplicka, T. Man´ko, J. Czarnul, J. Wypych, Application of solid phase microextraction for determination of dihydroxybenzenes, trihydroxybenzenes and chlorohydroxybenzenes in water and wastewater, Fresenius Environ. Bull. 18 (2009) 491–498. [26] J. Hoigné, H. Bader, The role of hydroxyl radical reactions in ozonation processes in aqueous solutions, Water Res. 10 (1977) 377–386. [27] M. Lescano, C. Zalazar, A. Cassano, R. Brandi, Kinetic modelling of arsenic (III) oxidation in water employing the UV/H2O2 process, Chem. Eng. J. 211–212 (2012) 360–368. [28] Y. Li, L. Ding, Q. Zhang, W. Wang, MD and QM/MM study on catalytic mechanism of a FAD-dependent enzyme ORF36: for nitro sugar biosynthesis, J. Mol. Graph. Model. 44 (2013) 9–16. [29] H. Tang, J. Li, Y. Bie, L. Zhu, J. Zou, Photochemical removal of aniline in aqueous solutions: switching from photocatalytic degradation to photo-enhanced polymerization recovery, J. Hazard. Mater. 175 (2010) 977–984. [30] J. Sarasa, S. Cort!es, P. Ormad, R. Gracia, J.L. Ovelleiro, Study of the aromatic byproducts formed from ozonation of anilines in aqueous solution, Water Res. 36 (2002) 3035–3044. [31] V. Di Paola, V. Augugliaro, L. Palmisano, G. Pantaleo, E. Savinov, Heterogeneous photocatalytic degradation of nitrophenols, J. Photochem Photobiol. A Chem. 155 (2003) 207–214. [32] Y. Hori, A. Nakatsu, S. Suzuki, Heterogeneous photocatalytic oxidation of nitrogen oxide ion (NO2 ) in aqueous suspension of various semiconductor powders, Chem. Lett. 9 (1985) 1429–1432. [33] A. Alif, J.F. Pilichowski, P. Bode, Photochemistry and environment XIV: phototransformation of nitrophenol induced by excitation of nitrite and nitrate ions, J. Photochem. Photobiol. A Chem. 59 (1991) 357–367. [34] A. Kumar, N. Mathur, Photocatalytic degradation of aniline at the interface of TiO2 suspensions containing carbonate ions, J. Colloid Interface Sci. 300 (2006) 244–252.