Introduction

The development of economy and industrialization, caused water contamination especially those contained organic dye due and phenolic compounds to its toxic and harmful, has made people take it into a serious consideration [1,2,3,4,5]. 4-Chlorophenol (4-CP) is toxic and non-biodegradable, and is present in wastewater as by-products of pulp and paper, petrochemical industries, dyestuff, and pharmaceutical [6,7,8,9]. In recent years, titanium dioxide (TiO2) has been extensively investigated as an environmental friendly and clean photocatalyst [10,11,12]. TiO2 has wide applications in wastewater treatment, water splitting hydrogen generation, solar cells, gas sensors, and air purification [13,14,15]. TiO2 is a popular photocatalyst due to the good photocatalytic activity, low cost, non-toxicity, strong oxidizing potential, being inert, and stability for long time [10, 16,17,18,19]. It exists in four mineral forms including the anatase, rutile, brookite, and titanium dioxide (B) [20]. Phase transformation from amorphous to anatase occurred at 500 °C, and the transformation from anatase to rutile phase occurred at 700 °C. Complete rutile phase was formed at 900 °C with corresponding increase in the particle size [21]. Photocatalytic activity of TiO2 has a relationship with its size, shape, surface area, morphology, and dimensionality [16]. However, utilizing TiO2 has two drawbacks: first high energy bandgap (approximately 3.2 eV), which requires ultraviolet (UV) radiation for photo-induced activation. Another disadvantage is high electron (e−)–hole (h+) recombination rate [3, 22, 23]. Considerable research and development works have been carried out to overcome the aforementioned limitations. In recent years, doping metals such as Fe, Ag, Cr, Sn, and Pt are the most suitable solution [2, 21, 24,25,26]. Tong et al. [27] studied on photocatalytic methyl orange degradation. The preparation of Fe3+-doped TiO2 catalysts and their photocatalytic activity for methyl orange degradation were investigated in their report. Results showed that Fe3+ ions can be successfully incorporated into the crystal lattice of TiO2 [27]. Jia et al. [5] synthesized Fe3O4–TiO2 heterostructures through the co-precipitation method. In this work, the triple surface heterostructures, Ag@Fe3O4–TiO2, were prepared by a chemical solution-deposition process. The photocatalytic activity of Fe3O4–TiO2 was found to decrease due to increase in the rate of recombinations. Results investigated the composite coated with Ag metal ions, increasing its photocatalytic activity [5]. It is not easy to compare the results reported for doped semiconductor due to different experimental conditions and synthesis methods. The photocatalytic activity of the doped TiO2 relies on the nature and concentration of the dopant ion [22]. The optimum weight concentrations of Fe and Ag dopant technique are widely used in different studies. In this work, the amount of Fe chosen from optimum concentration of different researches and the photo-reduction were preferred for the fabricated of Ag on Fe3O4–TiO2.

In this study, the uniform nanocatalysts of Fe3O4–TiO2 were synthesized by a facile hydrothermal method. Ag-doped Fe3O4–TiO2 NCs were modified in silver acetate solution under UV light by a photochemical reduction technique. Then, the concentration effects of iron (Fe) and silver (Ag) doped on TiO2 from the viewpoint of morphology and degradation rate were investigated. The prepared catalyst was characterized through the XRD, SEM, FTIR, and BET techniques. Besides, the photocatalytic activity of Fe3O4–TiO2–Ag was studied by illumination of 4-CP aqueous solution under UV irradiation, and the reaction kinetics of removal were considered.

Materials and methods

Materials

All the chemical materials, including titanium tetrabutoxide, Ti(OC4H9-n)4 with purity of 98%, iron (III) nitrate [Fe(NO3)3·9H2O, 99%] as the source of Fe, ethanol (C2H5OH, 99.7%), glacial acetic acid (CH3COOH, 99.5%), sulfuric acid (H2SO4, 98%), and silver acetate (CH3COOAg) used in this study, were all analytical grade and purchased from the Merck company.

Catalyst preparation

Synthesis of Fe3O4–TiO2 NCs

The synthesize procedure under magnetic stirring was as follow: 20 mL acetic acid was added dropwise to a flask containing 10 mL of Ti(OC4H9-n)4 diluted in 30 mL C2H5OH, followed by the addition of 1 mL H2SO4 and a specific amount of Fe(NO3)3·9H2O (0.05, 0.01, and 0.3% wt. of Fe3O4–TiO2). Then, the obtained clear liquid was sonicated in an ultrasonic for 2 h, resulting in the formation of a milk-like solution which was further transferred into a stainless autoclave and kept at 453 K for 13 h. The resulting precipitates were separated from the mother liquor by centrifugation, washed thoroughly with distillated water and ethanol three times, respectively, and then dried at 373 K in oven for 12 h.

Synthesis of Fe3O4–TiO2–Ag NCs

For synthesizing of Ag loading (1.0 and 2.0% wt. Ag in Fe3O4–TiO2–Ag), the composite suspension was prepared by mixing Fe–TiO2 nanoparticles (0.5 g) with a specific amount of CH3COOAg aqueous solution and 50 mL of deionized water. Then, the prepared solution was stirred for 1 h at room temperature and 3 h under UV irradiation, continuously. The Fe3O4–TiO2–Ag nanocomposites were separated from the reaction media by centrifugation, washed completely with distillated water and ethanol, and then dried at 373 K in oven for 12 h. Finally, the prepared catalysts were calcined at 773 K for 4 h. Overall, at different weight percentages of Fe and Ag, six catalyst samples were prepared and the details are presented in Table 1.

Table 1 Synthesized catalysts at different conditions in this study
Full size table

Characterization

The X-ray diffraction (XRD) patterns were observed on a X-ray diffractometer using a Cu target Kα ray (λ = 0.15406 nm) to determine the crystalline structure and phase of nanomaterials by scanning in angular range (2Ɵ) from 20° to 80° (Philips, PW1730 diffractometer, Netherlands). The particle size and morphology of Fe3O4–TiO2–Ag NCs were studied with scanning electron microscopy (FESEM-TESCAN MIRA3, Kohoutovice, Czech Republic), which was equipped with an energy-dispersive spectroscopy (EDS). FTIR spectra were recorded between the wave number of 400 and 4000 cm−1 (Bruker, Model: VERTEX70, Germany). The specific surface area (BET method), specific pore volume, and average pore diameter (BJH method) were determined using nitrogen adsorption at 77 K (Model: BELSORP MINI 3, Czech Republic).

Experimental setup and procedure

The photocatalytic activity of Fe3O4–TiO2–Ag NCs with different operational conditions along with un-doped nano TiO2 for degradation of 4-CP in aqueous solution was carried out using an experimental setup. Setup included rectangular cubic glass photoreactor by a total volume of 480 mL (4 cm × 4 cm × 30 cm) without the upper face, which was equipped with two 8 W UVC lamps. For each test, a specific amount of catalyst was suspended into a glass reactor containing 100 mL of the 4-CP with specific concentration. The pH of solution was adjusted with 0.1 M HCl or 0.1 NaOH. Then, the above solution was stirred in the dark medium for 60 min to attain the adsorption–desorption equilibrium for 4-CP and dissolved oxygen on the surface of Fe3O4–TiO2–Ag NCs. After every given irradiation time, the mixture was sampled and centrifuged for 14 min at a rate of 11,000 rpm. The concentration of 4-CP in each test was determined through a UV–Vis spectrophotometer (Hach, DR 5000, USA) at λmax = 279 nm. Removal was analyzed through the initial and final 4-CP concentrations according to following equation:

$${\text{Removal}} \left( \% \right) = \frac{{\left( {C_{0} - C_{t} } \right)}}{{C_{0} }} \times 100,$$
(1)

where Co is the initial concentration of 4-CP (mg/L) and Ct is the concentration of 4-CP (mg/L) at specified reaction time t (min).

Results and discussion

XRD analysis

X-ray diffraction was used to confirm the phase purity and crystallinity of the catalysts. The XRD patterns of samples containing different Fe and Ag contents are shown in Fig. 1. All samples consist of anatase as the unique phase. The peaks at scattering angles of 25.4°, 37.9°, 48.2°, 54°, 55.2°, and 62.9° are well supported with the JCPDS 21–1272, and correspond to the reflections from the (101), (004), (200), (105), (211), and (204) crystal planes, respectively. XRD patterns indicate that the crystal structure of TiO2 remains unchanged after sensitization process, as reported in similar work [28]. The identified peaks located at 44.6° (400), 54° (422), and 62.8° (440) were attributed to Fe3O4 diffractions (JCPDS 65–3107) [29]. When Ag+ ions are incorporated into crystal lattice of Fe3O4–TiO2 NCs, the peaks are found at 38.7° (111), 44.5° (200), and 62.9° (220) that confirmed the JCPDS 04–0783 [5, 22, 27, 30, 31]. After the addition of Fe species, the crystallinity of the Fe3O4–TiO2 nanocomposites changed and was confirmed to be more addition of Fe amount that has great effect on the crystallinity intensity of the TiO2. The characteristic diffraction peaks corresponding to the anatase phase of TiO2 remained unchanged in all samples without any peak shifting. Furthermore, the existence of new diffraction peaks of Fe3O4 was not identified, suggesting that the low loading of Fe3O4 might be dispersed well on the surface of the TiO2. The average crystallite domain sizes were calculated using the Scherrer equation from the fitting of the (101) peak corresponding to the anatase (2θ = 25.4°) [32,33,34]. The calculated average crystallite sizes for samples (a), (b), and (c) were 23.2, 19.2, and 16.8 nm, respectively. All samples have the same peaks, nearly; however, the average crystallite size of sample (c) is lower than that of other samples. Figure 1 does not show all peaks that satisfy Fe and Ag, due to their lower concentration (i.e., below the detection limit of the XRD).

Fig. 1
figure 1

XRD pattern of a sample-1, b sample-3, and c sample-6

Full size image

FESEM and EDX analysis

Figure 2 shows the FESEM images of some samples which were prepared by a novel and facile ultrasonic-assisted hydrothermal method with mainly sphere-like shape. It is obvious that the Fe and Ag are successfully doped on the surface of TiO2. The size of particles varies from 12 to 50 nm, which represents that a particle is formed from many crystals. According to the literature, doping of Fe and Ag caused grain growth, which is restrained during the hydrothermal, and decreases the crystal size of TiO2 (Degussa P25) [27, 35]. These phenomena may prevent particle agglomeration, forming well-defined nanocrystalline powders with high surface area [26]. As shown in Fig. 2, the sample morphologies have no evident differences and all samples consist of spherical particle sizes below 50 nm. However, as the Fe and Ag contents increased to 0.3 and 2, respectively, it can be seen that the aggregation phenomenon occurred and many nanoparticles transformed into obviously bigger nanoparticles. Moreover, EDX results (Fig. 3) reveal that Fe3O4–TiO2–Ag is composed of Ti, O, Ag, and Fe elements, which was confirmed to the XRD results. According to Table 2, the experiment evaluated the weight percentage of Fe and Ag in sample-2 similar to the theoretical ones, which confirms that doping is perfect and complete through this condition.

Fig. 2
figure 2

Typical FESEM images of catalyst a sample-2, b sample-3, c sample-4, d sample-6, and e particle size of sample-6

Full size image
Fig. 3
figure 3

EDX analysis of the sample-6

Full size image
Table 2 Quantitative element results from EDX technique
Full size table

FTIR analysis

FTIR analyses using KBr as reference show the characteristic peaks. FTIR spectra of Fe3O4–TiO2–Ag NCs (Fig. 4) showed the peaks corresponding to stretching vibrations of O–H groups on the catalyst surface and bending vibrations of water adsorbed on the surface around 3350–3450 and 1620–1635 cm−1, respectively. Moreover, the ~ 600 cm−1 band was associated with the Ti–O–Ti stretching vibration, which has been shifted to the lower wavelength as the addition of Fe3+ and Ag+ ions. The peak at 1113 cm−1 was described by the C–H vibration of the acetate ion [3, 6, 36, 37]. The Fe3O4–TiO2–Ag NCs by observing a peak at ~ 1080 cm−1 of the Fe3O4 along with the peaks of Ti–O–Ti (~ 600 cm−1) and the O–H vibrations (~ 3400 cm−1) confirm the formation of products, which corresponded to the open literature [36]. Almost all samples have the same results of FTIR.

Fig. 4
figure 4

FTIR spectra of the a sample-2, b sample-3, and c sample-6

Full size image

BET analysis

N2 adsorption–desorption measurements reveal that nanoporous structure of catalyst was present for all samples. It was observed from Fig. 5 that the prepared samples possess mesoporous surface as the isotherms are of type IV [21, 38]. Using the Barrett–Joyner–Halenda model, the pore distribution of the nanocomposites is centered at 20 nm.

Fig. 5
figure 5

a and b N2 adsorption–desorption isotherms. c and d Macro pore size distribution plots of sample-2 and sample-6

Full size image

The average pore diameter from adsorption branch is 14.02 nm for sample-2 and sample-6. SBET of nanospherical structure was estimated to be 44.60 and 43.62 m2/g from adsorption curve. Table 3 shows that Fe and Ag loading led to an increase in the BET surface area and decreased its mean pore size diameter of the catalyst. This fact observed due to the obstacle of TiO2 pores by metal oxidic species can take place, as their interactions with the support, which may be physical or chemical adsorption [38].

Table 3 Properties of as-prepared sample-2 and sample-6 nanocomposites
Full size table

In our previous work, the optimum values of Fe and Ag were chosen as 0.3 and 2% wt. in which the Fe3O4–TiO2–Ag photocatalyst showed considerable enhancement in the degradation of 4-CP compared to bare nano TiO2 [39]. The introduction of the Fe and Ag loading led to an increase in the BET surface area and reduced its mean pore size diameter of the catalyst compared to the bare TiO2 [39].

Photocatalytic activity

The photocatalytic activities of the Fe- and Ag-loaded samples for 4-CP decomposition were compared with un-doped TiO2 (P25) under UV irradiation. The photocatalytic activity tests were performed in Fig. 6 at conditions including the pH = 3, irradiation time = 120 min, 4-CP concentration = 30 mg/L, and catalyst dosage of 1 g/L. According to Fig. 6, sample-6 was chosen as desired catalyst. According to Fig. 2, for sample-6, the Fe and Ag were distributed very well on TiO2 in comparison to other samples. Besides, the crystal size of this sample was lower than others. Then, the operation conditions were optimized and removal efficiency increased to 97%. Table 4 also presents the reusability of the prepared nanocomposite after recycling. The photocatalytic degradation of Fe3O4–TiO2–Ag NCs still maintained 94% efficiency, even after five cycles. In our previous work [40] under UV and solar irradiations, some nanocomposites exhibited higher photocatalytic efficiency compared to the bare TiO2 (around 20 to 40%), which demonstrated the benefits of the nanocomposites in TiO2-based photocatalytic systems.

Fig. 6
figure 6

Degradation results of different synthesized catalysts

Full size image
Table 4 Optimal conditions and reusability of catalyst with RSM and experimental ones
Full size table

Mechanism of 4-CP photocatalytic degradation

When UV light falls on the surface of Fe3O4–TiO2–Ag, photogenerated electrons and holes are generated. Fe3+ ions can act as electron and hole traps, hence forming Fe 2+ and Fe 4+ ions, which are less stable compared to Fe3+ ions. So, tends to return to Fe3+. This leads to the generation of OH· radical and O2 anion [27, 34, 41].

Although the Ag Fermi energy level is lower than the conduction band of TiO2, the oscillation hot electron around Ag is excited to higher energy level. So it could transfer to conduction band of TiO2 by hot electron injection, furthermore these electrons were used to generate free radicals which enhances the photocatalytic activity [14, 42].

The photocatalytic reaction generally includes photo-excitation, charge separation and migration, and surface oxidation–reduction reactions [43]. The reactive species generated during illumination of photocatalysts are h+, OH−, and O2−. To understand the mechanism of Fe and Ag for degradation 4-CP, it is necessary to detect which reactive species plays a major role in the photocatalytic degradation process. According to the results of the intermediate analysis and considering the degradation pathways suggested by existing literature studies [44,45,46], among the different active species, the OH· radical has been reported as the most important active species in the photocatalytic oxidation process. Since the OH· radicals are electrophilic and OH groups have electron-donating character, the OH· radicals attack the para and ortho positions of the benzene ring [40].

The schematic of pollutant oxidation removal through the TiO2 are illustrated in Fig. 7. The beneficial effect of Fe3+ may be explained by considering the formation of Fe2+ species by means of a transfer of photo-generated electrons from TiO2 to Fe3+ [Eq. (3)]. According to the crystal field theory, Fe2+ is relatively unstable due to the loss of d5 (half-filled high spin) electronic configuration and tends to return to Fe3+ (d5). Subsequently Fe2+ could be oxidized to Fe3+ by transferring electrons to absorbed O2 on the surface of TiO2 [Eq. (5)]. The Fe2+/Fe3+ energy level lies close to Ti3+/Ti4+ level. As a consequence of this proximity, the trapped electron in Fe2+ can also be easily transferred to a neighboring surface Ti4+ [Eq. (6)], which then leads to interfacial electron transfer. That is to say, Fe3+ can be an effective electron trap in anatase phase. Meanwhile, Fe3+ can also serve as hole trap [Eq. (12)] due to the energy level for Fe3+/Fe4+ above the valence band edge (Evb) of anatase TiO2. The trapped hole embodied in Fe4+ has longer lifetime because of the immobilized electron in Fe2+. Therefore, Fe3+ can act as both hole and electron traps, according to the following reactions [3, 5, 17, 47,48,49]:

Fig. 7
figure 7

Schematic illustration on removal of pollutants by the formation of photo-induced charge carriers (e/h+) in a semiconductor TiO2 particle surfaces [48]

Full size image
$${\text{TiO}}_{2} + {\text{h}}\upnu \to e^{ - } + h^{ + }$$
(2)
$${\text{Fe}}^{3 + } + h^{ + } \to {\text{Fe}}^{4 + }$$
(3)
$${\text{Fe}}^{3 + } + e^{ - } \to {\text{Fe}}^{2 + }$$
(4)
$${\text{Fe}}^{2 + } + {\text{O}}_{2} \left( {\text{ads}} \right) \to {\text{Fe}}^{3 + } + {\text{O}}_{2}^{ - }$$
(5)
$${\text{Fe}}^{2 + } + {\text{Ti}}^{4 + } \to {\text{Fe}}^{3 + } + {\text{Ti}}^{3 + }$$
(6)
$${\text{Ti}}^{3 + } + {\text{O}}_{2} \left( {\text{ads}} \right) \to {\text{Ti}}^{4 + } + {\text{O}}_{2}^{ - }$$
(7)
$${\text{Fe}}^{4 + } + {\text{OH}}^{ - } \left( {\text{ads}} \right) \to {\text{Fe}}^{3 + } + {\text{OH}}{ \cdot} \left( {\text{ads}} \right)$$
(8)
$${\text{Fe}}^{4 + } + e^{ - } \to {\text{Fe}}^{3 + }$$
(9)
$${\text{Fe}}^{2 + } + h^{ + } \to {\text{Fe}}^{3 + }$$
(10)
$${\text{Fe}}^{2 + } + {\text{OH}}{ \cdot } \to {\text{Fe}}^{3 + } + {\text{OH}}^{ - }$$
(11)
$${\text{Fe}}^{3 + } + {\text{h}}\upnu \to {\text{Fe}}^{4 + } + e^{ - }$$
(12)
$$e^{ - } + {\text{O}}_{2} \left( {\text{ads}} \right) \to {\text{O}}_{2}^{ - }$$
(13)
$$4{\text{-CP}} + {\text{OH}}{ \cdot } \to {\text{Intermediates}} + {\text{OH}}{ \cdot} /{\text{OH}}^{ - } /{\text{O}}_{2}^{ - } \to {\text{CO}}_{2} + {\text{H}}_{2} {\text{O}} + {\text{simpler}}\,{\text{molecules}} .$$
(14)

The conductive band electron further reacts with adsorbed O2 to form O2−, while Fe4+ reacts with surface hydroxyl group to produce hydroxyl radical. Thus, 4-CP was photo-degraded even under the UV irradiation [Eq. (14)]. Besides, a decrease in the semicircle diameter of the titanium nanocomposites compared to the bare TiO2 demonstrated a reduction in the resistance of interfacial charge transfer, leading to fast electron transfer and effective separation of photo-induced charge carriers. This phenomenon was related to very good conductivity properties of doped materials and the \(\pi - d\) junction between the doped material and titanium which impressively facilitated the mobility of the electron and through charge separation, preventing the recombination of electrons and holes [40].

Kinetics of 4-CP photocatalytic degradation

Experimental studies were carried out at optimum conditions, pH = 4.86, catalyst dosage of 1.5 g/L, and 4-CP concentrations of 40.4 mg/L, to determine the degradation kinetic mechanism. The reactor was stirred by magnetic stirrer to complete photo-degradation process. Samples were taken from one reactor at 30, 60, 90, 120, 150, and 180 min after initiation of reaction. The models widely used are zero-, pseudo-first-, and second-order kinetic models. The degradation of 4-CP can be shown as follows:

$$4{\text{-CP}} + {\text{OH}}{ \cdot } \to {\text{Oxidation}}\,{\text{products}} .$$
(15)

Because OH· is a very reactive free radical, it does not accumulate in the reactor and its concentration is chosen as a steady-state value [50,51,52]. Pseudo-first-order kinetic respect to the concentration of the 4-CP is shown below:

$$\frac{{{\text{d}}\left[ {4{\text{-CP}}} \right]}}{{{\text{d}}t}} = K_{\text{obs}} \left[ {4{\text{-CP}}} \right]\left[ {{\text{OH}}{ \cdot} } \right].$$
(16)

Since [OH·] is constant at steady state, Kobs [OH·] is equal to Kapp, where Kobs and Kapp are absolute and apparent rate constants, respectively [53]. Equation (16) can be written as follows:

$$\frac{{{\text{d}}\left[ {4{\text{-CP}}} \right]}}{{{\text{d}}t}} = K_{\text{app}} \left[ {4{\text{-CP}}} \right].$$
(17)

With integration of above equation, we have first-order kinetic equation:

$${\text{Ln}}\frac{{\left[ {4{\text{-CP}}} \right]_{0} }}{{\left[ {4{\text{-CP}}} \right]_{t} }} = K_{\text{app}} t,$$
(18)

where [4-CP]0 and [4-CP] are the concentrations of 4-CP at the beginning and time t, respectively. The second-order kinetic equation is given as:

$$\frac{1}{{\left[ {4{\text{-CP}}} \right]_{t} }} - \frac{1}{{\left[ {4{\text{-CP}}} \right]_{0} }} = K_{\text{app}} t.$$
(19)

The zero-order kinetic equation can be written as follows:

$$\left[ {4{\text{-CP}}} \right]_{0} - \left[ {4{\text{-CP}}} \right]_{t} = K_{\text{app}} t.$$
(20)

Modeling of 4-CP removal by zero-, pseudo-first-, and pseudo-second-order equation is shown in Fig. 8, respectively.

Fig. 8
figure 8

a Zero-, b pseudo-first-, and c pseudo-second-order kinetic modeling of 4-CP removal

Full size image

The correlation coefficient for the 4-CP removal kinetics of zero, pseudo-first, and second order was obtained 0.9956, 0.9499, and 0.7795, respectively. So, 4-CP degradation follows the zero-order kinetic models.

Conclusions

Nanocomposites of Fe3O4–TiO2–Ag with different weight concentrations of iron (Fe) and silver (Ag) doped TiO2 were synthesized in this investigation. The synthesized catalysts were characterized which confirmed that the average nanosphere size was from 12 to 50 nm and Ag is well dispersed on to Fe3O4–TiO2. Furthermore, the Fe3O4 magnetic core facilitates the convenient recovery of the catalyst by applying an external magnetic field. Fe3+ can be easily integrated into the crystal lattice of TiO2 and reduced recombination of photo-electrons. Silver nanoparticles combine with semiconductor, which promotes the separation of charges, produces more photo-generated charges in Fe3O4–TiO2–Ag, and enhances bandgap absorption of TiO2.

According to characterization and primary photocatalytic activity tests, the catalyst with 88% degradation was chosen as the desired catalyst. Then, the operation conditions were optimized and removal efficiency increased to 97%. The most interesting feature from an application point of view is the combination of its photocatalytic response and the possibility to easily extract the photocatalyst by the deposition method. Furthermore, maximum degradation of 4-CP for the optimal conditions is 97% and photocatalytic degradation of Fe3O4–TiO2–Ag NCs still maintains 94% efficiency, after five cycle recovery. Besides, 4-CP degradation kinetic rate was considered in which the degradation follows the zero-order kinetic.