Environmental green chemistry applications of nanoporous carbons

Abstract

Influence of surface properties of nanoporous carbons on activity and selectivity during the photooxidation of 4-chlorophenol on UV-irradiated TiO₂ was performed. Characterization by infrared spectroscopy, X-ray photoelectronic spectroscopy and X-ray absorption near edge structure spectroscopy confirm the presence of a contact interface between both solids and suggest the coordination of some functional organic groups of the carbon surface, mainly ethers and carboxylic acids, to metallic centre Ti⁺⁴ in TiO₂. Changes in surface pH of carbons from basic to neutral or acid remarkably increase the production of 4-chlorocathecol by a factor of 22 on TiO₂–Carbon in comparison of TiO₂ alone. A scheme of interaction between TiO₂ and carbon is proposed to the increased photoactivity of TiO₂ and a reaction mechanism for the different intermediate products detected is also proposed. Results showed that TiO₂–Carbon can be used as an alternative photocatalyst for environmental green chemistry and selective organic synthesis applications.

Introduction

The production of petroleum and gas is followed of significant production of polluted water commonly called production water. This waste presents the highest volume in all petroleum exploration and production process and it is composed of a complex mixture of organic and inorganic materials. For example, phenol or halophenols have been found at different amounts as a function of the characteristics of the reservoir and the age of the well and type of oil to be extracted. An interesting alternative in the treatment of these wasted effluents are the advanced oxidation processes, particularly heterogeneous photocatalysis. This technique using titanium dioxide (TiO₂) as photocatalyst has popularity for treatment and purification of water. However, TiO₂ have several operative limitations as low adsorption capabilities, only UV-absorption of the solar spectra, and low selectivity. Several works have showed beneficial cooperative effects between TiO₂ and different carbon forms [1–13]. We have reported synergistic effects between both solids in the photomineralization of phenol, halo-phenols and herbicides [4–6, 12, 13]. Also, an important challenge for industrial organic chemistry consists in developing chemical processes for a clean environment. This is the field of environmental green chemistry [14, 15]. An interesting alternative for green chemistry processes is heterogeneous photocatalysis, principally, because it directly concerns with several of the twelve principles of green chemistry [16]. TiO₂ is a non-toxic and biocompatible material and it is the most efficient photocatalyst that shows high efficiency for the oxidative photodegradation of hazardous aromatic molecules in polluted water and air. Also, it shows a high selectivity in functionalization of light n-alkanes, in selective mild oxidation of gas and liquid hydrocarbons and in selective oxidation of alcohols and hydroxyl-containing molecules [15]. In short, photocatalysis is a promising route for the 21st century organic chemistry [17] where selective photocatalytic conversions will play a major role, offering an alternative green route for the production of organics. However, in aqueous phase, TiO₂ has low selectivity because intermediate products of the photooxidation of aromatic molecules commonly are photomineralized. In previous works we have showed [12, 13] that TiO₂ photoactivity is strongly influenced by activated carbon surface properties. The objective of this work is verify the influence of a contact interface between TiO₂ and nanoporous carbons on the photocatalytic activity of TiO₂ in the 4-chlorophenol photodegradation a well-known halo-aromatic molecule commonly found in production waters. In addition, the influence of surface properties of some selected nanoporous carbons on the selectivity of intermediate products formed during the photooxidation of 4CP on UV-irradiated TiO₂ was studied as a possible strategy for high selective green organic synthesis.

Experimental

Materials

High purity 4-chlorophenol (4CP) was purchased from Aldrich. Photocatalyst was TiO₂ P25 from Degussa. Activated carbons (AC) were prepared from sawdust of Tabebuia Pentaphyla wood by two methods. Physical activation (AC_H-type or close AC) by gasification under CO₂ flow or by pyrolysis under N₂ flow at temperatures from 450 up to 1000 °C by 1 h [12]. These AC were denoted ${\text{AC}}_{{{{\text{CO}}_{2}{\text{-}}i}}}$ and ${\text{AC}}_{{{{\text{N}}_{ 2}{\text{-}}i}}}$, being i activation temperature. Chemical activation [13] was also performed (AC_L-type or open AC) after impregnation of precursor with different concentrations of ZnCl₂, H₃PO₄ and KOH followed by activation under N₂ flow at 450 °C by 1 h. These carbons were denoted as ${\text{AC}}_{{{{\text{ZnCl}}_{ 2}{\text{-}}i\% }}}$, ${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4}{\text{-}}i\% }}$ and AC_KOH-i% being i% concentration of ZnCl₂, H₃PO₄ and KOH, respectively.

Characterization

Characterization of TiO₂ and activated carbons was performed by B.E.T. surface areas (S_BET) and surface pH (pH_PZC) and reported elsewhere [12, 13]. The full isotherms of N₂ adsorption in the range of 0.03 up to 630 Torr were measured in a Micromeritics ASAP-2010 apparatus. pH_PZC were estimated by the drift method [18]. Characterization of some selected TiO₂–AC photocatalysts was also performed by infrared spectroscopy (FTIR), X-ray photoelectronic spectroscopy (XPS) and X-ray absorption near edge structure spectroscopy (XANES). TiO₂–AC photocatalysts were obtained by mixing TiO₂ with the selected AC in 5 mL of water continuously stirred for 80 min. After this, the mixture was filtered and dried at 120 °C, 12 h. Fourier transform Infra-red (FTIR) experiments were made on a spectrophotometer Magna-IR 560 from Nicolet. The powders were mixed with KBr in a 5% (w/w) mixture. The mixed powder was pressed to tablets of 1 cm diameter at 10 tons for 1 min. The transparent tablets were inserted in the apparatus and the spectra were recorded from 4000 to 400 cm⁻¹ with a resolution on 5 cm⁻¹. KBr reference spectrum and CO₂ from ambient have been subtracted to every spectrum. Ex-situ XPS was carried out in an ESCALAB 220i-XL spectrometer (VG scientific) equipped with a hemispherical electron analyzer and a double anode Mg–Al non-monochromatic X-ray source. The pressure in the analysis chamber was kept below 10⁻⁹Torr. Samples were protected from exposition to the atmosphere by immersion into an ultra-dry hydrocarbon solvent (purified heptanes) while transferring from the reactor to the preparation chamber of the spectrometer. The Ti K-edge XANES spectras at room temperature were recorded in the “fluorescence-yield mode” at the NSLS (Brookhaven National Laboratory) on beamline U7A using a Stern-Heald-Lytle detector with argon as detector gas. All data were normalized to make pre-edge = 0 and post-edge = 1.

Photocatalytic tests

Experimental conditions of photocatalytic tests have been previously reported [12, 13] but are summarized as follow: 50 mg TiO₂ and 10 mg AC were added under stirring in 25 mL at 20 °C of 100 ppm of 4CP and maintained in the dark by 80 min to reach adsorption at equilibrium. This adsorption time before irradiation was selected because in previous works [12, 13] we found that after this time 4CP achieve equilibrium of adsorption on TiO₂–AC photocatalysts. After this time, 4CP concentration remains constant at dark conditions. Batch photoreactor was a cylindrical flask (Pyrex, 60 mL) with a bottom optical window of 3 cm diameter and open to air [5]. Irradiation was provided by a high-pressure mercury lamp and IR was filtered by a circulating-water cell equipped with a 340 nm cut-off filter. Photons flux emitted by light source was determined by Actinometry using Uranyle Oxalate as actinometer and software Logicien Photon version 1.6. Photon flux estimated was 2.9 × 10¹⁵ photons cm⁻² s⁻¹. Disks (0.45 μm) were used to remove particulate matter from aliquots (0.3 mL) before analysis by HPLC with a UV absorbance detector (280 nm). Besides 4CP, hydroquinone (HQ), benzoquinone (BQ) and 4-chlorocatechol (4CT) were detected as the main intermediate products. From kinetic curves of 4CP disappearance as function of irradiation time, apparent first-order rate constants (k _app) were obtained as the best kinetic parameter to compare activity of photocatalysts.

Results and discussion

Characterization of AC

Tables 1 and 2 show a summary of S_BET and pH_PZC of AC. As expected, the higher activation or pyrolysis temperature the higher the pH_PZC which indicate the presence of basic functional groups on the surface of AC and a close topology (AC_H-type). A similar trend was observed for S_BET where an increase of the thermal treatment leads to an increase of the surface areas up to a maximum of 770 m² g⁻¹ around 800 °C (${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}800}}$) and of 590 m² g⁻¹ at 900 °C (${\text{AC}}_{{{\text{N}}_{ 2}{\text{-}}900}}$) under CO₂ and N₂ flows, respectively. It is important also to note that if wood is heat at higher temperatures than 800 and 900 °C under CO₂ and N₂ flow, respectively, a decrease in S_BET was observed. This could be due to excessive gasification by CO₂ or excessive pyrolysis promoted at high temperatures. Concerning to AC prepared by chemical activation, it can be seen from Table 2 that AC prepared with H₃PO₄ showed lower pH_PZC values than those AC prepared with ZnCl₂. It can be noted from Table 2 that pH_PZC of AC prepared with 1%w/w H₃PO₄ is clearly lower than that obtained for the same concentration of ZnCl₂ (4.7 against 6.4). This is the due to H₃PO₄ is a strong Brönsted acid while ZnCl₂ is a Lewis acid; therefore, the first one should introduce more positive zeta potential, and our case, more acidic groups on AC surface. Also, it must to be note that AC prepared with KOH showed an apparent unexpected behaviour. Thought KOH is a strong Lewis base, Table 2 show that the higher KOH concentration the lower pH_PZC of AC. However, this lowering is only from 7.7 down to 6.1 for the increase of KOH concentration from 1 to 65% w/w. This can be explained by the fact that AC prepared with KOH develops an open topology [19, 20] in comparison of AC_H-type. On the other hand, it can be seen from Table 2 that the higher the concentration of impregnation compounds the higher S_BET areas having a maxima of about 2485, 1987, and 476 m² g⁻¹ for ZnCl₂ (35% w/w), H₃PO₄ (35% w/w) and KOH (50% w/w), respectively. It is important to note that employing higher concentrations a clear decrease in S_BET is observed. This decrease in S_BET has been already reported and in the case of KOH [20] have been explained by a high concentration of steam produced during the thermal decomposition of KOH. In this case, activation would proceeds by a gasification reaction at low temperature: C + H₂O → H₂ + CO, and due to a higher concentration of steam a decrease in S_BET is expected because original microporous texture can be consumed by gasification.

Table 1 BET surface areas (S_BET) and surface pH (pH_PZC) of AC_H-type

Table 2 BET surface areas (S_BET) and surface pH (pH_PZC) of AC_L-type

Photocatalytic activity

Concerning to photocatalytic activity of TiO₂–AC, Fig. 1a shows an example of the kinetics of disappearance of 4CP under UV-irradiations obtained on some selected photocatalysts. It can be seen in Fig. 1a that direct photolysis is negligible. Also, Fig. 1a shows that AC alone, specifically ${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$, is not photoactive. This fact was similar for other AC, in concordance with previous results [2, 3]. Kinetics from Fig. 1a indicate that TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$ is clearly more efficient that TiO₂ alone requiring only 12 h of irradiation against about 28 h required by TiO₂ to total photodegradation of 4CP. Photodegradation of 4CP on TiO₂ and TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$ showed in Fig. 1a follow an apparent first order mechanism. Thus, to compare the influence of AC on photoactivity of TiO₂, the first-order apparent constant rate (k _app) were obtained from linear regressions ln(C _o/C) = k _app · (t), of kinetic data in Fig. 1a. Figure 1b shows several examples of linear regressions that confirm the apparent first order rate mechanism. Identical treatment of kinetic data was performed for the rest of photocatalyst and a summary of results is presented in Tables 3 and 4 showing the apparent first-order constant (k _app) and the interaction factor (I _F) developed between both solids. I _F is defined by comparison between the apparent constant obtained on the mixed system against value obtained on TiO₂ alone:

$$ I_{\text{F}} \, = \, \left[ {k_{\text{app}} \left( {{\text{TiO}}_{ 2}{-}{\text{AC}}_{\text{i}} } \right)/k_{\text{app}} \left( {{\text{TiO}}_{ 2} } \right)} \right]. $$

(1)

I _F permits to verify the influence of AC on TiO₂ photoactivity. For example, this parameter indicate that most of TiO₂–AC photocatalysts showed a synergistic effect (I _F > 1) between both solids for 4CP degradation and the best mixed system developed a maximum of about three times higher photoactivity than that of TiO₂ alone (Table 3, for TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$). We talk about synergy effect if I _F > 1 because AC is not photoactive. On the other hand, if I _F < 1, there is an inhibiting effect on TiO₂ photoactivity. We attributed these effects to different kind of interactions between TiO₂ and AC. In an early work [2] supported in changes in Langmuir’s parameters of phenol adsorption, we reported that this interaction is consequence of a contact interface [2, 4] spontaneously created between both solids. The physicochemical properties of AC, principally surface area (S_BET) and surface pH (pH_PZC), clearly influences the magnitude of this interface and concomitantly plays an important role on TiO₂ properties, including photoactivity. In general, the higher surface area the higher photoactivity of TiO₂ but this is not the only parameter to be considered because data of Tables 3 and 4 indicate that surface pH plays an important role. For example, in spite that some AC_L-type (Table 2) have remarkably high S_BET, as ${\text{AC}}_{{{\text{ZnCl}}_{ 2}{\text{-}}65\% }}$, ${\text{AC}}_{{{\text{ZnCl}}_{ 2}{\text{-}}35\% }}$, ${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4}{\text{-}}65\% }}$, ${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4}{\text{-}}35\% }}$, contrary to the expected, these AC poorly influence or indeed inhibit Titania photoactivity (I _F < 1, Table 4). This inhibition have been ascribed to the fact that pH_PZC of those AC are very acidic [13]. Therefore, there is a combination between both physicochemical properties of AC that influence the interaction with TiO₂. The interaction between both solids was confirmed by FTIR, XPS and XANES.

Fig. 1

a Kinetics of 4CP disappearance. filled diamond TiO₂, open diamond Photolysis, filled triangle${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$, Open trianlge TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$. b Linear regression from kinetic data of some cases in a

Table 3 Summary of kinetics results of 4CP photodegradation of TiO₂–AC_H-type

Table 4 Summary of kinetics results of 4CP photodegradation of TiO₂–AC_L-type

FTIR, XPS, XANES characterization. Interface TiO₂–AC

Figure 2 shows the FTIR spectra of ${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$, pure TiO₂ and TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$. It can be seen that compared with the spectrum of pure TiO₂, a remarkably increase in the stretching absorption band in 500–1000 cm⁻¹ of the skeletal O–Ti occur in presence of ${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$. At the same time, a clear decrease in the cyclic ether (C–O–C) stretching signal of ${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$) suggest that interaction between both solids would occur by the coordination of oxygenated functional organic groups on AC’s surface to TiO₂. Under the effect of oxygen atoms from surface of AC, the symmetry of Ti–O tetrahedron in TiO₂ is broken and therefore, and increase in the absorption band is observed. The influence of functional groups AC on TiO₂ FTIR spectra have been confirmed with other kind of AC. For example, a band corresponding to the stretching of Ti–O–C surface was detected [6] for the case of interaction of TiO₂ with AC. In that work we found that some acidic AC, particularly those with acetate or carboxylate functional groups (for example, ${\text{AC}}_{{{\text{ZnCl}}_{ 2}{\text{-}}5\% }}$ and ${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4}{\text{-}}5\% }}$) increased TiO₂ photoactivity as can be seen in the present work in Table 4. Concerning to XPS spectra, it can be seen in Fig. 3 that in presence of AC, a decrease in the binding energy corresponding to Ti 2p signal. This is indicative of reduced Ti states [21] and corresponds to a partial reduction of the oxidation state of Ti from Ti⁺⁴ to Ti⁺³ of about 1.3 eV when TiO₂ interact with the more basic activated carbon, ${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$ (9.1 pH_PZC) in stead of only 0.5 eV in presence of ${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}450}}$ (6.3 pH_PZC). Therefore, it seems clear that pH_PZC of AC influence the valence state of Ti in TiO₂–AC. This is in agreeing with previous results [12, 13] where surface aggregation of TiO₂ nanoparticles on AC was detected by scanning electron microscopy for the case of acid AC [13] while dispersion of TiO₂ was improved in presence of basic AC [12]. This interfacial interaction between both solids was confirmed by XANES studies as follows. Figure 4 shows Ti K-edge XANES spectra for TiO₂ and TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}450}}$, TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}600}}$, TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$. In Fig. 4 can be seen that changes in the shifts of the Ti K-edge position are minor suggesting that oxidation state of the Ti ions change only a little. However, the three pre-edge features give more information because changes in the pre-edge intensity are associated with reduced Ti states [22]. These changes are highly correlated to the energy position of the Ti K-edge. As observed in the Fig. 4, the larger is the Ti K-edge position, the smaller is the pre-edge intensity. The changes in the pre-edge intensity may be due to the oxygen vacancies produced in the crystal lattice of TiO₂. This is because the 4-(TiO₄), 5-[(Ti = O)O₄] and 6-(TiO₆) coordinated Ti are attributed to the pre-edge peaks between 4969 and 4972 eV [22], commonly assigned 1s-to-3d transitions. The oxidation state of the Ti ions is expected to be higher with fewer amounts of oxygen vacancies inside the crystal lattice. Thus, the intensity changes of the pre-edge feature may be explained by the different amount of oxygen vacancies inside the crystal lattice. Larger amount of oxygen vacancies will produce a more significant pre-edge feature in the XANES spectra. In other words, oxidation state of Ti ions is expected to be lower with more oxygen inside the crystal lattice of TiO₂ as suggested the increase of intensity in the pre-edge of Ti in XANES spectra in Fig. 4. In short, present results suggest that an increase in pH_PZC of AC clearly introduces an enhancement in Titania’s photoactivity during the 4CP degradation. Figure 5 shows a schematic representation of the interaction between TiO₂ and both types of AC. In the case of TiO₂–AC_H-type (Fig. 5a), activated carbons play the role of electron carrier that could inhibit the recombination of photoelectrons. Figure 5a shows that excited photoelectrons once excited and in conduction band are driven out from TiO₂ surface. This is perfectly logical to think because AC_H-type behaves as an electron semiconductor in totally concordance with a recent study or Peralta et al. [23] whose found that the level of the conduction band of TiO₂ is −0.5 eV versus NHE (normal hydrogen electrode), and the reduction potential of oxygen for the one-electron reduction is −0.2 eV versus NHE. Since this value is less negative than that of the conduction band of TiO₂, the electron transfer from the semiconductor to the carbon particle is thermodynamically favored. On the other side, Fig. 5b shows the contrary behaviour. Photoelectrons can be added to TiO₂ surface because AC_L-type have specific functional organic groups that in solution are dissociated to produce surface anions able to drive out electron density from aromatic system to TiO₂. This is in agreeing with works from other groups [24, 25] that show that the addition of low concentrations of organic acids as ascorbic [24] or acetic [25] to aqueous suspensions with TiO₂ introduce important modifications to the surface of the semiconductor. Therefore, if the surface concentration of such groups on AC is enough low, which is perfectly valid to consider in the present case because AC prepared at low impregnation concentrations of acids showed the best photocatalytic results, for example TiO₂–${\text{AC}}_{{{\text{ZnCl}}_{ 2}{\text{-}}5\% }}$ and TiO₂–${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4}{\text{-}}5\% }}$ (Table 4), then, it can be suggested that we are in presence of a similar behaviour than those studies [24, 25]. In other words, for the case of the mixed system TiO₂–AC_L-type, the synergy effect detected could be attributed not only to a proper surface area but also to the presence of low surface concentration of acidic oxygenated groups, particularly carboxylic acids. The analogy is perfectly valid because once dissociated in aqueous phase, the carboxylic acid produce carboxyl anions (–COO⁻) clearly stabilized by transfer electrons to the aromatic ring of graphene layers in AC. Having this in mind, Fig. 6 shows a general mechanism for the degradation of 4CP involving the interaction between both solids. In a first step, it can be seen in Fig. 6 that after dissociation in aqueous phase, carboxylic or acetic acid as functional groups on AC’s surface coordinated to Ti⁺⁴ metallic centre of TiO₂. As described above, this interaction could inhibit the recombination of photogenerated species in the semiconductor (e−, h+). In consequence, a higher concentration of hydroxyl radical (^•OH) and superoxide anion radical (O₂ ^•−) can be formed as indicate the second step of mechanism in Fig. 6. Then the third step shows once adsorbed 4CP, it suffer degradation with ^•OH and/or O₂ ^•−. The influence of O₂ ^•− anion radical is very important because the higher concentration of O₂ ^•− the higher the production of HO₂ ^• radicals that enhance the photoactivity of TiO₂ in the photomineralization of 4CP as indicated third step in mechanism of Fig. 5.

Fig. 2

FTIR of TiO₂, ${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$ and TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$

Fig. 3

XPS of TiO₂ and selected TiO₂–AC in the region 2p Ti

Fig. 4

Ex situ Ti K-edge XANES. (2) TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}600}}$; (3) TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}900}}$; (5) TiO₂–${\text{AC}}_{{{\text{CO}}_{2}{\text{-}}450}}$; (6) TiO₂

Fig. 5

Interaction mechanism between AC and TiO₂. a TiO₂–AC_H-type. b TiO₂–AC_L-type

Fig. 6

Mechanism of 4CP degradation on TiO₂–AC_L-type

Photoselectivity. Kinetics of appearance and disappearance of intermediate products

According to the type of carbon materials employed, the proportion of the intermediate 4-chlorocatechol detected is clearly increased. This fact can be clearly seen from Fig. 7 that shows the kinetic trends corresponding to the sum of HQ + BQ (Fig. 7a) and 4CT (Fig. 7b). In general, it can be seen from Fig. 7 that for TiO₂ alone and most of TiO₂–${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4} }}$ photocatalysts, the main intermediate products are HQ + BQ (Fig. 7a). This figure shows that the more acid is the AC employed the higher is the production of intermediate products. Figure 7b clearly shows that the CT formation is clearly higher in the most acid TiO₂–${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4} }}$ (65, 35, 5% w/w) photocatalysts and practically negligible for the TiO₂–${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4}{\text{-}}1\% }}$ and TiO₂ alone. This fact indicates that different surface sites governed by pH_PZC more than by texture characteristics leads to different types of 4CP adsorption mechanism and therefore influence the intermediate product distributions. These results are discussed as follow. Intermediate products detected agree with chlorine substitution by p-hydroxylation yielding HQ and BQ that are commonly in equilibrium and to o-hydroxylation yielding 4CT. This general trend was also observed for TiO₂–${\text{AC}}_{{{\text{N}}_{ 2} }}$ and therefore, Table 5 shows a summary of changes detected in orto/para products in the 4CP photooxidation. R(o/p) compares the maximum quantity detected of 4CT against maxima of HQ and BQ. Values in Table 5 indicate a clear increase of R(o/p) in comparison of that obtained on TiO₂ alone (0.08). R(o/p) yield a maxima value of about 7.453 for TiO₂–${\text{AC}}_{{{\text{N}}_{ 2}{\text{-}}450}}$ (pH_PZC 6.05). A similar trend was detected for relation F(o/o) that compares the maxima quantity of 4CT detected on TiO₂–AC against maxima 4CT on TiO₂ (0.037 μmol). It can be seen from Table 5 that for the case of ${\text{AC}}_{{{\text{N}}_{ 2}{\text{-}}700}}$ against ${\text{AC}}_{{{\text{N}}_{ 2}{\text{-}}1000}}$ a change in only one unity of surface pH (from 8.9 to 7.9) leads to a remarkably increase in F(o/o) from 1.60 up to a maxima of about 22.1 for TiO₂–${\text{AC}}_{{{\text{N}}_{ 2}{\text{-}}700}}$. A similar behaviour is detected for the other type of AC which is clearly more acid than those prepared by pyrolysis. In addition, having in mind that 4CP photooxidation occurs on TiO₂ surface, the present results clearly showed that light changes in pH_PZC of AC induce important changes in the surface of semiconductor and this clearly affect the reaction mechanism. Figure 8 shows the influence of pH_PZC of the different carbons, ${\text{AC}}_{{{\text{N}}_{ 2} }}$ (Fig. 8a) and ${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4} }}$ (Fig. 8b). It can be seen in Fig. 8 that the lower the pH_PZC of AC the higher the values of R and F indicated in Table 5. In the photooxidation of 4CP there are two possible reaction mechanisms. First, the oxidation of 4CP can occur by nucleophilic substitution of chlorine by the attack of aromatic ring with ^•OH radicals that are promoted at neutral or basic pH. This route yield preferentially HQ and its successive oxidation yield BQ. On the other hand, the oxidation of 4CP would occur by oxidation of aromatic ring by electrophilic addition of anion radical O₂ ^•− yielding selectivity 4CT. This mechanism is favored at acid pH. These results are in totally concordance with previous studies concerning to photooxidation of phenol [2] and 2,4-dichlorophenoxiacetic acid [5]. These remarkably differences in products selectivity as function of AC’s pH_PZC can be associated with interfacial interaction detected between both solids discussed above. For example, for the case of the transfer of electrons from Titania to AC (Fig. 5a), this kind of interaction permits to increase ^•OH radicals in agreeing with the strong oxidation of 4CP by means of a mechanism to mainly produce para-products (Table 5) as hydroquinone (HQ) and benzoquinone (BQ). This p-hydroxylation mechanism is showed in Fig. 9. This mechanism is induced by basic surface pH of AC and it involves the abstraction of a hydride and chloride from 4CP to produce bencine (structure 2A) which is readily oxidized to produce HQ, commonly in equilibrium with BQ. On the other hand, the interaction between TiO₂ and AC by the way showed in Fig. 5b, promotes higher concentrations of O₂ ^•− and HO₂ ^• (obtained from O₂ ^•− + H⁺) radicals whose induce remarkable changes in selectivity of products (Table 5) to mainly produce orto-product (Table 5) as 4-chlorocathecol (4CT). This o-hydroxylation mechanism is showed in Fig. 10. It is induced by acid surface pH of AC and involves the abstraction of an electron from 4CP to produce an aromatic cation radical (structure 6A) which is stabilized by resonance to produce the structure 6C (Fig. 10) which is readily oxidized to produce selectively 4CT and H₂O₂, a common sub-product obtained in the present experimental conditions. According to this, the interpretation of pH effects on the selective photooxidation of 4-chlorophenol is a very difficult task because the pH of the solution influences adsorption and dissociation of substrate, catalyst surface charge, oxidation potential of the valence band and other physicochemical properties of the system. In reality, the interpretation of pH effects on the selective photooxidation of 4-chlorophenol is a very difficult task because the pH of a solution influences adsorption and dissociation of substrate, catalyst surface charge, oxidation potential of the valence band and other physicochemical properties of the system. Therefore, we are working on characterization of TiO₂–AC composites by surface techniques as X-ray photoelectronic spectroscopy, and electron and photoelectronic conductivity to better elucidate the effect of TiO₂–AC for the selective photooxidation of 4-chlorophenol a little more deeply.

Fig. 7

Kinetics of appearance and disappearance of main intermediates products in 4CP photooxidation in presence of ${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4} }}$ a HQ + BQ. b 4CT

Table 5 Intermediate products in the photooxidation of 4CP on TiO₂–AC

Fig. 8

Influence of pH_PZC of activated carbon on selectivity. a ${\text{AC}}_{{{\text{N}}_{ 2} }}$, b ${\text{AC}}_{{{\text{H}}_{ 3} {\text{PO}}_{ 4} }}$

Fig. 9

para-Hydroxylation mechanism induced by basic surface pH and close topology of AC

Fig. 10

ortho-Hydroxylation mechanism induced by acid surface pH and open topology of AC

Finally, it must to be remark that nowadays a challenge for the heterogeneous photocatalysis consists in improving the photocatalytic efficiency of TiO₂ by shifting its optical response to the visible range [10]. One of the most interesting ways to modify this semiconductor to absorb visible light consists in doping with transition metals and other elements. For example, several attends have been performed by doping with iron [9], carbon [10] and more recently, the effect of rare-earths [26] and nitrogen [27, 28] as TiO₂–doping elements have been studied for the degradation of dyes and azo-dyes and results obtained have been promising ones. Regarding our own experience, we are studying TiO₂–doped photocatalysts with Fe, N, P and C for tune the process for visible range and up to now the best results (preliminary ones) have been obtained with C and P as doping elements.

Conclusions

FTIR, XPS and XANES characterization of TiO₂–AC suggest a strong interaction between TiO₂ and AC and from these evidences, a mechanism for the interaction between both solids was proposed. This mechanism explains both the improved photoactivity of TiO₂ as a function of the surface pH of different types of activated carbon and changes in products distributions. The design of pH_PZC of AC permits the control in the selectivity of products during 4CP photooxidation on TiO₂–AC. Results suggest the possibility to use TiO₂–AC as an alternative environmental green photocatalyst both in treatments of wasted waters and in selective organic synthesis.