Synthesis lafeo3 for photodegradation of pollutants in wastewater nghiên cứu tổng hợp vật liệu quang xúc tác lafeo3 cho phản ứng phân hủy hợp chất hữu cơ gây ô nhiễm trong nước thải

SEM images of LFO nanostructures with different morphologies: a nanocubes; b nanorodes; c nanospheres; d dendritic nanostructures e floral-like nanosheets; f nanowires; g nanofibers Yang

PHAN C

MINISTRY OF EDUCATION AND TRAINING

HA NOI UNIVERSITY OF SCIENCES AND TECHNOLOGY

MINISTRY OF EDUCATION AND TRAINING

HA NOI UNIVERSITY OF SCIENCES AND TECHNOLOGY

-

PHAN CHI NHAN

SYNTHESIS LAFEO3 FOR PHOTODEGRADATION OF POLLUTANTS

IN WASTEWATER

CHEMICAL ENGINEERING

ACKNOWLEDGEMENT

This work has been supported by the RoHan Project funded by the German Academic Exchange Service (DAAD, No 57315854) and the Federal Ministry for Economic Cooperation and Development (BMZ) inside the framework “SDG Bilateral Graduate school programme

I would like to express my deep gratitude to my principal supervisor, Associate Professor Pham Thanh Huyen I thank Assc Prof Pham Huyen wholeheartedly for her great academic guidance, valuable advice, and constant encouragement

I would like to give a big thank to Professor Malte Brasholz, who is my supervisor

at this time I worked in Germany Thank you for giving me an opportunity to study

in a good condition and learn more about interesting culture of Germany

Countless thanks are also dedicated to my Mum, my sister and younger brother, my other family members and my friends who have always supported and accompanied

me from my MsC Beginning to end

Finally, I would like to thank my father, Mr Phan Van Huan (1945 – 2017), who brought me to science, give me a passion about chemistry and he is the motivation for me to finish this project This thesis is a present, which I want to give to my father Thank him for all

TABLE OF CONTENTS

ACKNOWLEDGEMENT

TABLE OF CONTENT I CHAPTER 1: LITERATURE REVIEW 1

1 Introduction 1

2 Lanthanum Ferrite (LFO) Perovskite 3

2.1 Structure 3

2.2 Optical property 4

2.3 Magnetic property 5

3 Lanthanum Ferrite as Visible-light Photocatalyst 7

3.1 Application in textile wastewater treatment 7

3.2 Photocatalytic mechanism 8

4 Synthetic methods of LFO 10

4.1 Sol-gel method 11

4.2 Hydrothermal method 13

4.3 Sol-gel hydrothermal method 17

5 Factors affecting photocatalytic activity of LFO material 18

5.1 Surface area 18

5.2 Band gap energy 21

5.3 Morphology 22

5.4 Crystallinity 25

5.5 Dopants 27

CHAPTER 2: EXPERIMENTAL 30

1 Synthesis of LaFeO3 30

2 Characterization of LFO 30

3 Experiment of photocatalytic activity 31

CHAPTER 3 RESULTS AND DICUSSION 33

1 Thermal Analysis 33

2 XRD analysis 34

3 Morphological analysis 35

4 BET analysis 36

5 Band Gap energy result 37

6 Calibration curve of dyes and 17β-Estradiol 38

7 Photocatalytic activity of LFO nanoparticles 40

8 Efect of reaction condition on photodegradation efficiencies 42

9 Effect of catalytic concentration on photodegradation efficiencies 44

10 Effect of H2O2 concentration on photodegradation efficiencies 45

11 Comparison of catalytic activity for different dyes 46

12 Photodegradation of 17β-Estradiol 47

47

13 Effect of light intensity on photodegradation efficiencies 48

CONCLUSIONS 49

Recommendations for future research 49

REFERENCES 51

LIST OF TABLES

Table 1 Optical band gap values of LFO nanomaterials synthesized by various methods 5Table 2 Particle size and morphologies of LFO synthesized by sol-gel method using different templates 12Table 3 The influences of the hydrothermal temperature on the formation of LFO powders (Adopted from (Ji, Dai et al., 2013)) 17Table 4 Photocatalytic performance of LFO for the degradation of dyes 20Table 5 Degradation rate of dyes solution on different morphologies LFO nanoparticles 24Table 6 Effects of calcination temperature on the BET specific surface area and pore parameters of LFO nanoparticles 37

LIST OF FIGURES

Figure 1 Schematic crystalline structure of orthorhombic LFO (Misch, Birkel et al., 2014) 4Figure 2 Schematic diagram of LFO antiferromagnetic order (Lee, Yun et al., 2014) 6Figure 3 Schematic diagram of the reaction mechanism of LFO nanostructures for organic degradation (Adopted from (Thirumalairajan, Girija et al., 2013)) 9Figure 4 FESEM images of LFO powders from (Feng, Liu et al., 2011) (a) and (Liu and Xu, 2011) (b) 13Figure 5 XRD patterns of different LFO nanostructures (Thirumalairajan, Girija et al., 2013) 15Figure 6 Image of LFO nanospheres by HRSEM (Dhinesh Kumar and Jayavel, 2014) 16Figure 7 RhB degradation % by using LFO samples with different particle sizes (Thirumalairajan, Girija et al., 2012) 19Figure 8 SEM images of LFO nanostructures with different morphologies: (a) nanocubes; (b) nanorodes; (c) nanospheres; (d) dendritic nanostructures (e) floral-like nanosheets; (f) nanowires; (g) nanofibers (Yang, Huang et al., 2006; Leng, Li et al., 2010; Thirumalairajan, Girija et al., 2012; Thirumalairajan, Girija et al., 2013; Dhinesh Kumar and Jayavel, 2014; Thirumalairajan, Girija et al., 2014) 23Figure 9 (a) XRD pattern at different calcination temperatures; (b and c) degradation of RhB and MB in the presence of LFO calcined at 800°C (Thirumalairajan, Girija et al., 2014) 26Figure 10 (a) XRD pattern of LFO samples calcined at different temperatures; (b) Degradation of RhB with the use of different LFO samples and P-25TiO2 (Su, Jing

et al., 2010). 27Figure 11 TGA curve of synthesized LaFeO3 sample 33Figure 12 XRD patterns of LaFeO3 34Figure 13 SEM images of (a) LFO-C600, (b) LFO-C700, (c) LFO-C800, and (d) LFO-C900 35Figure.14 HRTEM image of the LFO-C800 36Figure.15 Nitrogen adsorption-desorption isotherms of the LFO nanoparticles calcined at different temperatures 36Figure 16 (a) LFO solid UV-vis absorption spectrum and (b) Schematic of determination of band gap energy 38Figure 17 Calibration curve of MB 38 39

Figure 18 Calibration curve of MB 39

Figure 19 Calibration curve of MO 39

Figure 20 Calibration curve of 17β-Estradiol 40

Fig 21 The photodegradation efficiencies of RhB under visible light irradation by different photocatalysts 40

Figure 22 The change in colour of RhB solution during the reaction 42

Fig.23 Photodegradation efficiencies of Metyl Orange under differences condition 43

Fig 24 Effect of catalytic concentration on photodegradation of MO 44

Fig.25 Effect of H2O2 concentration on photodegradation efficiencies 45

Fig.26 Photodegradation efficiencies for bleaching different dyes 46

Fig.27 The photodegradation efficiencies of 17β-Estradiol under visible light irradation 47

Fig.28 Photodegradation efficiencies of MB by different lamp 48

Fig.29 Photodegradation efficiencies of 7β-Estradiol by different lamp 48

CHAPTER 1: LITERATURE REVIEW

1 Introduction

In recent years, environmental pollution issue, especially wastewater pollution has been increasing alarmingly Due to the rapid development of textile industry and lack of modern technologies for textile wastewater treatment, a considerable amount

of harmful organic dyes has been discharged into environment (Yi, Chen et al., 2008) For several dyes with the concentration less than 1 ppm, their presence in water could easily be observed and undesirable (Robinson, McMullan et al., 2001) Among these, the most notable ones include Rhodamine B (RhB), methylene blue (MB) and methyl orange (MO) which have been used as coloured substances for printing or dyeing cotton, leather, silk, wool (Gupta, Suhas et al., 2004) MB causes not only permanent injury to eye but also difficulty in breathing to human and animals (Tan, Ahmad et al., 2008) Meanwhile, experimental research has proven the negative effects arising from RhB and MO on human well-being and ecological environment, including carcinogenicity, toxicity, and mutagenicity (Khataee and Kasiri, 2010) Consequently, wastewater treatment targeting at minimising the levels of these organic compounds has become essential

Conventional technologies for treatment of dye-containing water are not sufficiently effective to achieve the current stringent requirements for discharge To overcome the challenges, there has been much attention focusing on advanced oxidation processes (AOPs) which have been suggested as substitutions to previous treatment technologies Until now, there have been a variety of AOPs, including electrochemical oxidation (Gallios, Violintzis et al.,2010; Pang, Wang et al., 2013; Wang, Yue et al., 2014), supercritical water oxidation (Williams and Onwudili, 2006; Wang, Lv et al., 2013), ozonation (Prieto-Rodríguez, Oller et al., 2013), photocatalytic oxidation (Zangeneh, Zinatizadeh et al., 2010; Gümüş and Akbal, 2011; Rodríguez, Gallardo et al., 2012), wet air oxidation (Zhan, Li et al., 2010; Ovejero, Sotelo et al., 2011) and other combinations (Moreira, Vilar et al., 2012;

Couch, Mezyk et al., 2014) In particular, the AOP using photocatalysts - irradiated semiconductor systems has been suggested as a promising way in environment remediation, because it is efficient, cost-effective and environmental friendly by utilizing solar light or artificial irradiation, which is abundant wherever (Liu, Niu et al., 2013; Lai, Juan et al., 2014; Zhao, Tian et al., 2014) This shows significance in both of waste treatment and resource conservation

Photocatalysis is defined as the acceleration of a chemical transformation in the present

of a catalyst with light illumination (Augugliaro, Bellardita et al., 2012) Many semiconductors have been used as photocatalysts, including TiO2 (Nakata and Fujishima, 2012), ZnO (Yu, Shi et al., 2013), ZrO3 (Basahel, Ali et al., 2015), SnO2

(Cheng, Chen et al., 2011), CeO2 (Wu, Wang et al., 2015), and InO2 (Li, Zhang et al., 2013) for degradation of a wide range of environmental contaminants However, their applications are limited because these semiconductor catalysts have high activity only under UV illumination, which presents ~ 5% of solar energy spectrum For this reason, many efforts have been made in order to find different alternatives harvesting the solar light and afterwards utilizing it in large-scale application Recently, perovskite-based materials have been reported as excellent visible-light-driven photocatalysts (Kanhere and Chen, 2014; Wang, Tadé et al., 2015) Several types of perovskite materials have been studied widely, such as titanate perovskites (Yao, Xu et al., 2004; Zhang, Sun et al., 2013), tantalate perovskites (Li and Zang, 2009; Buršík, Vaněk et al., 2013), ferrite perovskites (Sun, Jiang et al., 2010; Soltani and Entezari, 2013) and complex perovskite materials (Zhu, Fu et al., 2008; Clark, Dyer et al., 2010) Among ferrite perovskites, lanthanum ferrites, LaFeO3, need to be further studied due to their interesting physical properties as well as potential applications in photocatalyst, fuel cell, sensors and permeation membranes (Gabal, Ata-Allah et al., 2006; Yoo, Kim et al., 2011)

Thus, the study of LaFeO3 materials with properties suitable for photocatalytic requirements is really necessary So I decided to choose a topic : “ Synthesize LaFeO3

materials for photodegradation of pollutants in wastewater”

2 Lanthanum Ferrite (LFO) Perovskite

Perovskites-type oxides present a class of inorganic crystalline solids, having a general formula of ABO3, where A and B are rare-earth metal and transition metal cations, respectively (Zhang and Li, 2013) Recently, the perovskite-based material LaFeO3 has attracted significant interest due to their useful application in biosensors (Liotta, Puleo et al., 2015), oxygen permeation membrane (Kida, Ninomiya et al., 2010), fuel cell (Li, Wang et al., 2014) and especially visible-light photocatalytic reaction (Li, Jing et al., 2007; Thirumalairajan, Girija et al., 2012; Dhinesh Kumar and Jayavel, 2014) Moreover, the desired properties of LaFeO3 for high photocatalytic activity under visible light, including narrow band gap energy, high surface area as well as high crystallinity, can be achieved by adjusting several parameters during the synthesis The doping with another rare-earth metal in A-site and transition metal in B-site of this material may lead to an enhancement in photocatalytic performance Therefore, in order to prepare a photocatalyst that meet above requirements, the in-depth study on the structure and properties of LaFeO3 is essential

2.1 Structure

Lanthanum ferrite, LaFeO3 (LFO), belongs to Pbnm space group with the lattice parameters, a = 5.557 Å, b = 5.565 Å, c = 7.854 Å (Vansutre, Das et al., 2000) It has an orthorhombic perovskite structure, which is derived from the distortion of the ideal cubic structure via the tilting of the FeO6 octahedral All of the Fe3+ ions are octahedral surrounded by oxygen ions and the La3+ ions are inserted in the interspaces between the FeO6 octahedra (Fossdal, Menon et al., 2004; Bellakki, Manivannan et al., 2009) Fig 1 illustrates the crystalline structure of LFO, in

which the La3+ cations are displayed as grey spheres and the FeO6 octahedra are in blue

Figure 1 Schematic crystalline structure of orthorhombic LFO (Misch, Birkel

Recently, the preparation and utilization of lanthanum ferrite with novel properties and morphologies have been reported by several researchers (Ahmed, Selim et al., 2011; Wang and Gong, 2011; Salah, Rashad et al., 2015) In particular, LFO nanomaterials are considered to be promising due to theirs large surface area and interfacial state, which lead to different optical, electrical and magnetic properties from bulk materials (Thirumalairajan, Girija et al., 2012) For the application as photocatalysts, the optical and magnetic properties of these materials are emphasized here

2.2 Optical property

Optical absorption performance of a semiconductor is related to its electronic structure

as well as band gap The band gap energy (Eg) is one of important features of the nanomaterial to evaluate its optical property Generally speaking, this energy illustrates the minimum amount of required energy so that an electron could jump from a valence band to a conduction band Based on corresponding wavelength of light that material absorbs, which allows to calculate its band gap value by following formula (Ziegler, Heinrich et al., 1981):

αhν = A(hν - Eg)1/2where hν is the photon energy, α is the optical absorption coefficient, A is a

constant and Eg is the band gap energy

It is believed that this optical band gap (Eg) value strongly depends on the preparation procedure of material as well as particle size (Roduner, 2006; Köferstein, Jäger et al., 2013) This is one of reasons to interpret for a range of band gap values of LFO nanoparticles, as summarized in Table 1 Generally, LFO nanoparticles possess narrow band gap energies, which enable them to be active under visible light This will be discussed in detail in the later section

Table 1 Optical band gap values of LFO nanomaterials synthesized by various

combustion 24-104 2.11-2.07 (Parida, Reddy et al., 2010) Emulsion 32 3.85 (Chandradass and Kim, 2010) Precipitation 60 2.33 (Tang, Fu et al., 2011)

are in pair, which indicates La3+ is non-magnetic and subsequently there is no magnetic interaction between La3+ and Fe3+ Therefore, the magnetization pattern of LFO nanoparticles is governed by the Fe sub-lattices (Wei, Wang et al., 2012) The

authors have reported that in the Fe sub-lattices, the magnetic moments of Fe3+ are slightly canted, as the source of the ferromagnetic character at room temperature of LFO nanoparticles, which were prepared by an auto-combustion process Moreover, LFO is well known as an antiferromagnetic material, which is caused by the collinear arrangement of the sub-lattices, as shown in Fig 2 (Lee, Yun et al., 2014) These antiferromagnetic particles often display increasing magnetization because of the presence of uncompensated surface spins (Kodama and Berkowitz, 1999; Lee, Pakhomov et al., 2010)

Figure 2 Schematic diagram of LFO antiferromagnetic order (Lee, Yun et al.,

2014)

Generally speaking, the magnetization of LFO nanoparticles is different from the bulk crystals For example, the spontaneous magnetization value of LFO bulk crystals is 0.1 emu/g at ambient temperature (Shen, Cheng et al., 2009) which is considerably smaller than those of LFO nanoparticles (~ 21.9 nm) prepared by sol-gel method (0.38 emu/g) (Saad, Khan et al., 2013), LFO nanofibers (~ 20 nm) fabricated by electrospinning (0.9 emu/g) (Lee, Yun et al., 2014), and LFO nanoparticles (~25-50 nm) synthesized via sol-gel combustion process (0.35 emu/g) (Hui, Jiayue et al., 2010) Obviously, the magnetization values of LFO nanomaterials in those studies are considerably different This also suggests that the preparation procedures, particle sizes and surfaces indeed play key roles in the properties of product (Kumar, Raja et al., 2014) It can be seen that a reduction in

LFO particle sizes leads to an increase of magnetization values (Phokha, Pinitsoontorn et al., 2014; Qiu, Luo et al., 2014) LFO exhibits a ferromagnetism, which should be retained or even enhanced, facilitating the recyclability of material after photocatalytic degradation of dyes with the use of magnetic field

In conclusion, LFO has attracted great research interest for various applications as mentioned before; because of its unique structure, optical and magnetic properties Among these applications, LFO has shown great potential for use in the field of photocatalysis by virtue of narrow band gap energy, which makes it active in visible region Moreover, if its inherent ferromagnetic behaviour is enhanced, LFO would

be a photocatalyst which is capable to be recycled and reused

3 Lanthanum Ferrite as Visible-light Photocatalyst

Photocatalysis employing irradiated semiconductors has been proven to be very effective in decomposition of a wide range of organic pollutants (Li, Chen et al., 2010; Sayama, Hayashi et al., 2010) Numerous efforts have been made in synthesizing the semiconducting materials, such as TiO2, ZnO, CdS; however, they are only active under ultraviolet irradiation because of their large band gap energy Therefore, the selection of alternative, which has an appropriate band gap, with enhanced photocatalytic activity in visible region is necessary (Reddy, Martha et al., 2012) Taken account of narrow band gap energy as aforementioned, LFO nanostructures exhibit promising photocatalytic activities under visible light illumination (Li, He et al., 2014; Thirumalairajan, Girija et al., 2014) and in turn has been investigated as potential photocatalyst candidates to degrade organic dyes in wastewater This section reviewed the literature on the applications of LFO in wastewater treatment as well as discussed photocatalytic mechanism

3.1 Application in textile wastewater treatment

Many researchers have studied photocatalytic degradation of dye solutions by using LFO nanoparticles (Tang, Tong et al., 2013; Dhinesh Kumar and Jayavel, 2014; Li,

Wang et al., 2014; Gaikwad, Sheikh et al., 2015) Most of studies have focused on degradation of Rhodamine B (RhB), methylene blue (MB) and methyl orange (MO) (Su, Jing et al., 2010; Gao and Li, 2012; Sacco, Stoller et al., 2012; Wei, Wang et al., 2012; Tang, Tong et al., 2013; Xiao, Hong et al., 2013) because they are one of the common dyes which is widely used in a variety of industrial applications This will be discussed in detail in later section Apart from the decomposition of dyes such as RhB, MB, MO, LFO nanomaterials were also used as photocatalysts for

degradation of several other dyes For example, Abazari et al., carried out the

photodegradation of toluidine blue O (TBO) under solar light condition using LFO nanoparticles which were fabricated in emulsion nanoreactors in the presence of cetyltrimethyl ammonium bromide (CTAB) (Abazari, Sanati et al., 2014) They concluded that the TBO dye was decomposed completely (99.98%) after 90 min

exposure to the solar light Yang et al., conducted investigations on degradation of

Acid Red 18 in the presence of visible light using LFO as nanocatalysts They proposed that the nanoparticles displayed a relatively high activity; that 60% of Acid Red 18 was decomposed after 60 min (Yang, Zhong et al., 2009) Meanwhile, 68.2% of dye Acid Red 3B was decomposed after 2 hour illumination by LFO prepared via citric-based sol-gel method (Wang, Shen et al., 2012)

In my study, some of the most notable dyes will be selected, including RhB, MB,

MO and 17β-Estradiol, which have been widely used as coloured substances in industries In order to enhance degradation of these dyes by LFO, photoreaction mechanism should be well understood

3.2 Photocatalytic mechanism

Photodegradation of organic contaminants is based on the use of light irradiation to initiate photoreaction Fig 3 illustrates the process of a LFO nanoparticle absorbing light energy to generate electron-hole pairs

Figure 3 Schematic diagram of the reaction mechanism of LFO

nanostructures for organic degradation (Adopted from (Thirumalairajan,

When photocatalytic processes carry out in aqueous solution, the photo-induced holes (h+) react with hydroxide ions and water to yield hydroxyl radicals (OH•) which are responsible for the photocatalytic oxidation of the dye into non-toxic products (Cheng, Huang et al., 2010; Zhang, Zhou et al., 2011) It is reported that repeated attack of organic dyes by OH• result in complete oxidation (William IV, Kostedt et al., 2005) The process of producing OH• can take place by two pathways The first route is that the electrons react with dissolved O2 presenting in

water to form superoxide anion radicals O2 •− (the reduction of O

2), which subsequently combine with H+ to form the hydroperoxy radicals HO2• and rapidly decompose to OH• (Hou, Jiao et al., 2011) The second one is the oxidation of OH- The reaction mechanism of LFO by light could take place in the following steps, as

suggested by Chong et al (Chong, Jin et al., 2010)

VB -> hν + (or heat) recombination

Because of the complexity of photocatalytic mechanism, some researchers assumed that photocatalytic performance has been mainly determined by the light absorption capability of photocatalyst and the charge separation as well as light utilization yield to carry out reduction and oxidation reactions (Wang, Huang et al., 2010)

The research on the application of LFO in photocatalytic degradation of organic dyes has found its properties and then activity is strongly affected by the synthetic method selected, each of which usually involves specific preparation conditions Therefore, different synthesis methods have been employed to prepare LFO

4 Synthetic methods of LFO

To date, several methods have been studied for the preparation of perovskite LFO, such as co-precipitation (Nakayama, 2001), solid state reaction (Chu, Zhou et al., 2009), sol-gel process (Qi, Zhou et al., 2003), sonochemical method (Sivakumar, Gedanken et al., 2004), hydrothermal synthesis (Zheng, Liu et al., 2000) and microemulsion method (Giannakas, Ladavos et al., 2004) However, the particles

produced by the conventional solid state reaction exhibit slow kinetics, nonuniform particle size and low surface area (Popa and Moreno, 2011) Meanwhile, the co-precipitation method needs to use more chemicals as well as longer time to obtain LFO powders (Nakayama, 2001) By contrast, the hydrothermal method and sol-gel process have attracted increasing attention in the synthesis of nanostructured photocatalysts, because the produced LFO exhibits high crystallinity, good purity, controllable morphology and narrow particle size distribution at a relatively low synthetic temperature

4.1 Sol-gel method

Sol-gel method has been suggested as a potential route to synthesize LFO particles with a uniform composition and good crystallinity at a relatively low synthetic temperature (Tien, Mittova et al., 2014) Generally, sol-gel process includes four main steps from a precursor to final product: precursor solution to form a gel, aging

of a gel, drying and calcination to obtain the end product In the typical synthesizing

of LFO by sol-gel method, one starts with the certain amounts of lanthanum nitrate and iron nitrate, which is dissolved in deionized water The citric acid is added in the nitrates solutions and then gently evaporated at 60-70oC under magnetic stirring, resulting in a gel state with high viscosity Then, the prepared sample is dried at

70oC overnight Finally, LFO is obtained by calcining the dried sample at appropriate temperature (600-700oC) for several hours

In this process, several parameters could affect the properties of final product, including composition and concentration of precursor (starting material), type of solvent, concentration, temperature, heating rate, etc

For example, the composition of starting materials has found to affect particle sizes

and shapes of LFO nanoparticles (Ita, Murugavel et al., 2003) Ita et al revealed

that nano-sized particles of LFO could be slightly tailored by changing the ferric source The particles were spherical in shape with an average size of 41.4 nm and

42.5 nm when synthesized with the use of iron nitrate and iron acetylacetonate, respectively

The selection of an appropriate “chelating agent” or “structure-directing agent” or

“template” could considerably improve the quality of the product The morphology

or pore size of the catalyst strongly depends on the type of templates selected This could be explained by each of structure-directing agent will hold different volumes

in the gel network after complexing with cations in the synthetic solution (Wang, Xiang et al., 2013) Table 2 shows the different morphologies of LFO produced after sol-gel method by using different templates

Table 2 Particle size and morphologies of LFO synthesized by sol-gel method

using different templates

Templates

Formation temperature (oC)

Particle size (nm) Morphology References

Citric acid 680 Not

mentioned Nanowires

(Yang, Huang

et al., 2006) Polyvinyl

alcohol 600 70-90

Irregular shape

(Feng, Liu et al., 2011)

Ethylene

glycol 600 40

Not mentioned

(Aono, Tomida et al., 2009)

Glucose 500 30 Agglomerated

particles

(Liu and Xu, 2011)

Citric acid 800 Not

mentioned Highly porous

(Gosavi and Biniwale, 2010)

For example, by a sol-gel method using La2O3, Fe(NO3)3.9H2O and polyvinyl alcohol (PVA) as raw materials, single-phase and well-crystallized LFO particles

were obtained at 700oC with an average particle size of approximately 50 nm (Feng, Liu et al., 2011), as shown in Fig 4a In comparison, by adding glucose as a

structure-directing agent, Liu et al., successfully synthesized nanosized LFO with a

diameter of about 30 nm at lower synthetic temperature (500oC) with the mole ratios of glucose to metal ions 3:5 and 3:10 via sol-gel route (Fig 4b) (Liu and Xu, 2011)

Figure 4 FESEM images of LFO powders from (Feng, Liu et al., 2011) (a) and

(Liu and Xu, 2011) (b)

To conclude, sol-gel method has been widely used to synthesize LFO particles because it offers a facile way to control the size and morphology of the particles However, depending on target applications, the synthetic parameters should be carefully controlled to obtain the desired morphology as well as particle size

4.2 Hydrothermal method

Hydrothermal method is a dominant method to prepare nanosized materials The advantages of this method include easy control of size and morphology, facile fabrication process and cost-effectiveness In particular, the hydrothermal method can directly synthesize LFO at low temperature, as compared to the sol-gel method

So far, many researchers have used this method to synthesize LFO nanoparticles (Ji, Dai et al., 2013; Thirumalairajan, Girija et al., 2013; Dhinesh Kumar and Jayavel, 2014)

Typically, hydrothermal process for synthesizing LFO nanoparticles is based on a reaction of aqueous solution of lanthanum salts and ferric salts with or without structure-directing agent The reaction takes place at a temperature of >100oC and a pressure of > 1atm by using a sealed system, called autoclave, to produce LFO Afterwards, the obtained product is centrifuged, washed with water and ethanol for several time and dried Finally, the LFO powder is obtained after calcination Different parameters in the hydrothermal synthesis can be varied to control the sizes and morphologies of product, including precursor material, structure-directing agent, and temperature

In general, LFO nanoparticles were produced from hydrothermal method by complexing lanthanum nitrate, ferric salt and structure-directing agent Interestingly, the size of final product could be controlled by changing starting materials According to the experimental results of Kumar and Jayavel, 45 nm LFO nanospheres were formed by hydrothermal process with La(NO3)3.6H2O, Fe(NO3)3.9H2O and citric acid as the starting materials (Dhinesh Kumar and Jayavel, 2014) On the other side, by using another ferric source - K3[Fe(CN)6] to replace Fe(NO3)3.9H2O acting with La(NO3)3.6H2O and citric acid, Thirumalairajan and co-workers successfully fabricated LFO nanospheres with the average crystallite size of 52 nm (Thirumalairajan, Girija et al., 2013)

So far, different structure-directing agents have been studied, including citric acid (Dhinesh Kumar and Jayavel, 2014), cetyltrimethyl ammonium bromide (CTAB) (Yao, Wang et al., 2013), sodium carbonate (Zheng, Liu et al., 2000), etc It is reported that synthesis of LFO nanoparticles with or without structure-directing agent produced three different morphologies via the hydrothermal method (Thirumalairajan, Girija et al., 2013) In this work, the hydrothermal synthesis was carried out by selecting La(NO3)3.6H2O and K3[Fe(CN)6] as starting materials in the presence or absence of structure-directing agent, (NH2)2CO or C6H8O7.H2O, at

180oC for 12 h Figure 7 shows the change of product morphology synthesized under different conditions The LFO nanocubes which are produced in the absence

of structure-directing agents were indexed according to the perovskite phase with a cubic structure (Fig 5a); whereas the XRD pattern of LFO nanorods and nanospheres, formed with the use of urea or citric acid (Fig 5b-c), suggested the samples as a perovskite phase of orthorhombic structure By using Scherrer’s fomular, the average crystallite size was found to be 52, 64, 85 nm for nanospheres, nanorods and nanocubes, respectively Therefore, the selection of an appropriate structure-directing agent plays an important role in producing the desire sizes and morphologies of products

Figure 5 XRD patterns of different LFO nanostructures (Thirumalairajan,

Girija et al., 2013)

Moreover, the changes in the ratio of metal ions to structure-directing agent should

be carefully control to improve the purity of the end products Dhinesh Kumar and Jayavel suggested the optimum condition to obtain the pure orthorhombic LFO be 1:2 molar ratios between metal ions and citric acid The HRSEM image (Fig 6) shows the produced LFO nanoparticles with a sphere-like morphology and uniform

size distribution of approximately 45 nm However, the use of metal and citric acid with a molar ratio of less than 2 formed impurities in the product; whilst most of the final particles were La(OH)3 when the ratio between metal and citric acid reached 1

- 1.5 (Dhinesh Kumar and Jayavel, 2014)

Figure 6 Image of LFO nanospheres by HRSEM (Dhinesh Kumar and

Jayavel, 2014)

Ji et al., studied the effect of hydrothermal temperature on the crystallinity and

particle size of final products (Ji, Dai et al., 2013) The hydrothermal process took place at 110, 140, 170 and 200 °C for 14 h, respectively These samples were denoted as LFO-110, LFO-140, LFO-170 and LFO-200 The authors realized that the hydrothermal temperature had a significant influence on the obtained products

In particularly, the average crystallite size of the LFO-110 was much smaller than those of LFO-200 sample, but much bigger than that of LFO-140 and LFO-170 samples The detailed results of processing parameters on the formation of LFO are listed in Table 3

Table 3 The influences of the hydrothermal temperature on the formation of

LFO powders (Adopted from (Ji, Dai et al., 2013))

Catalyst XRD results BET surface

area (m2/g)

Average crystallite size (nm) LFO-110 Orthorhombic LaFeO3 18.6 50.5

LFO-140 Orthorhombic LaFeO3 23.6 32.8

LFO-170 Orthorhombic LaFeO3 25.8 36.5

LFO-200 Orthorhombic LaFeO3

and trace hexagonal La(OH)3

15.5 71.2

Consequently, there is no doubt that hydrothermal method enables the fabrication of highly crystallized and well-defined powders at relatively low temperature (Lu, Yuan et al., 2013) Moreover, this method have allowed to produce LFO with various morphologies, including spheres (Dhinesh Kumar and Jayavel, 2014), floral-like nanostructure (Thirumalairajan, Girija et al., 2014), nanowires (Thirumalairajan, Girija et al., 2012), cubes and nanorods (Thirumalairajan, Girija

et al., 2013) It appears that such differences in size, crystallinity and morphology of photocatalyst could be better controlled during hydrothermal process

4.3 Sol-gel hydrothermal method

Taking the double merits of both hydrothermal and sol-gel process, hydrothermal method offers high purity, high crystallinity, well-controlled morphology and narrow particle size distribution of the product while carrying out the reaction under moderate temperature (Li, Hou et al., 2005; Wang, Cao et al., 2013) In perovskite family, various materials have been synthesized by sol-gel-hydrothermal method for photocatalyst purpose, including NaTaO3 (Hu, Tsai et al., 2009), SrTiO3 (Yu, Ouyang et al., 2011), Bi2WO6 (Liu, Tang et al., 2014), BiFeO3

sol-gel-(Liu and Zuo, 2012); however, to the best of our knowledge, synthesizing LFO nanoparticles by using sol-gel-hydrothermal method has not been reported in open literature

From above discussion, it can be concluded that LFO nanoparticles with different morphologies as well as particle sizes can be fabricated by either employing different synthetic methods or altering processing parameters The effect of particle characteristics on the photocatalytic performance of LFO will be discussed in detail

in a following section

5 Factors affecting photocatalytic activity of LFO material

Perovskite-based photocatalyst prove to be promising materials with good photostability and high photocatalytic activity due to unique structure and optical property (Wu and Zhang, 2004; Xiong, Tian et al., 2015) Moreover, the vacant metal or oxygen sites in the structure of perovskite are able to promote oxygen adsorption onto the surface of photocatalyst, which boosts the reaction (Pena and Fierro, 2001; Worayingyong, Kangvansura et al., 2008) Generally, the photocatalytic performance could be affected by particle size, band gap energy, morphology and crystalline structure From a technological point of view, the properties of photocatalyst, including morphology, surface area, and crystallinity are strongly dependent on the preparation procedures

5.1 Surface area

From an aspect of photocatalytic application, the specific surface area of semiconductors considerably affects the activity of photocatalyst It is believed that surface area is closely related to particle size of material In particular, during the photoreaction, an undesirable step is that the recombination of electron–hole pair before they migrate to the photocatalyst surface (Parida, Reddy et al., 2010) The small size of photocatalyst reduces the distance from the place of electron–hole pair generation to photocatalyst surface; thereby photogenerated charge carriers have more chance to migrate to surface before recombination taking place These captured carriers are able to initiate photoreaction, causing the enhancement in photocatalytic activity This has been demonstrated by the degradation efficiency of RhB when using LFO nanoparticles with different particle sizes as photocatalysts in

Figure 7 (Thirumalairajan, Girija et al., 2012) Nano-sized LFO with an average particle size of approximately 90 nm was found to decompose more than 76% of RhB after 180 min (Thirumalairajan, Girija et al., 2013) The efficiency could be further improved to 93% and 96% if the particles possessed an average size of 85

nm and 70 nm, respectively (Thirumalairajan, Girija et al., 2012) On the other side, the activity of photocatalyst decreased with increasing particle size

Figure 7 RhB degradation % by using LFO samples with different particle

sizes (Thirumalairajan, Girija et al., 2012)

Moreover, the smaller average particle size leads to an increase in specific surface area of photocatalyst and in turn enhances dye decomposition The large specific surface area creates favourable conditions for photocatalytic reactions In terms of nanometer regime, many studies had reported the inverse relation between specific surface area and average particle size of photocatalyst (Abazari, Sanati et al., 2014); that is the larger particles of LFOexhibits smaller surface area Therefore, specific surface area of LFO nanoparticles is believed as a key factor in evaluating their photocatalytic activities Table 4 summarizes some of recent studies on the fabrication of nano-sized LFO by using different methods and the resulting photocatalytic performances

76

0 20 40 60 80 100 120

As seen in Table 4, a catalyst with high surface area is more desirable because it can achieve much more active sites where pollutant molecules can be absorbed to undergo its degradation by reacting with charge carriers coming from excitation of semiconductor For example, the nanosized LFO having a specific surface area of

278 m2/g displayed excellent photocatalytic performance for RhB decomposition as the yield was found to be 80% (Xiao, Hong et al., 2013) These results were found

to be relatively higher compared to the reported value of synthesized LFO/SAB-16, with 85 m2/g of surface area and decomposing efficiency of RhB was 35% (Su, Jing et al., 2010) By contrast, LFO with a specific area of 6.4m2/g evidently has showed low photocatalytic activity The degradation rate of phenol just achieved 30% (Wu, Hu et al., 2015) Therefore, the higher the surface area is, the higher the photocatalytic activity of LFO nanoparticle is Consequently, the high surface area of LFO nanomaterial could benefit the degradation of dyes in aqueous solution

Table 4 Photocatalytic performance of LFO for the degradation of dyes

Method Particle

size (nm)

Surfac

e area (m2/g)

Dye types

Degradati

on efficienc

y (%)

Illumination time (h)

al., 2010) Sol-gel

(citric)

Not mentioned

21.90 RhB 24 3 (Li, Jing et

al., 2007) Sol-gel

(glucose)

mentioned

Not-6.4 Phenol 30 3 (Wu, Hu et

al., 2015) Sol-gel

(citric)

mentioned

Not-5.2 Phenol 95 24 (Rusevova,

Köferstein et al., 2014)

5.2 Band gap energy

The narrow band gap of the semiconductor can absorb more photons to generate electrons in the CB and holes in the VB, which then carry out redox reactions (Dong, Xu et al., 2009) For this reason, semiconductor photocatalyst with low band gap energy would be more desirable for efficient light harvesting, thus improving photocatalytic activity

It is reported that LFO is a high visible-light-responsive photocatalyst as a result of its narrow band gap (Yang, Zhong et al., 2009; Ding, Lü et al., 2010) The Eg value

of LFO nanoparticles have been reported in the range of 2.1 eV - 3.85 eV (Bellakki, Kelly et al., 2010; Popa and Moreno, 2011; Sorescu, Xu et al., 2011; Tang, Fu et al., 2011), which is suitable for visible region, accounting for ~ 43% of the total energy from the sun, with a wavelength ranging from 400 nm to 800 nm This absorption range is interesting because LFO nanoparticles could be developed to be a new visible-light-driven photocatalyst, which has potentially useful for photocatalytic as well as other photoactive applications

In particular, Parida et al., successfully synthesized LFO nanoparticles by a sol-gel

auto-combustion method having a band gap energy of 2.1 eV (Parida, Reddy et al.,

2010) Phokha et al., have reported the direct band gaps in the range of 2.15 eV –

2.23 eV for all samples of LFO nanoparticles prepared by polymerized complex method, which was synthesized on polyester network of ethylene glycol and citric

acid solution to better distribution of ions (Phokha, Pinitsoontorn et al., 2014) Meanwhile, LFO which were synthesized by calcining an emulsion method had the energy band gap level of 3.85 eV (Chandradass and Kim, 2010) UV-Vis NIR spectrum of this sample indicated that the LFO nanocrystals synthesized by this method could be a type of photocatalyst

5.3 Morphology

Particle morphologies play an important role in determining photocatalytic performance Figure 8 shows LFO with diverse morphologies prepared by various methods; and Table 5 summarizes the photocatalytic performances of LFO with different morphologies on dyes Seen from Figure 8 and Table 5, the photocatalytic activities of LFO nanoparticles under visible light vary considerably according to the morphologies of samples The LFO nanospheres displayed the relatively high photocatalytic performance, which could be ascribed to the small crystallite size, large surface area and uniform pore size distribution

Figure 8 SEM images of LFO nanostructures with different morphologies: (a) nanocubes; (b) nanorodes; (c) nanospheres; (d) dendritic nanostructures (e) floral-like nanosheets; (f) nanowires; (g) nanofibers (Yang, Huang et al., 2006; Leng, Li et al., 2010; Thirumalairajan, Girija et al., 2012; Thirumalairajan, Girija et al., 2013; Dhinesh Kumar and Jayavel, 2014; Thirumalairajan, Girija

et al., 2014)

Three different morphologies of LFO nanostructure, including nanorods, nanocubes and nanospheres were reported by Thirumalairajan and co-workers (Thirumalairajan, Girija et al., 2013) The efficiency of RhB decolorization under visible light was found highest, 90.80%, as compared with those of the samples with other morphologies, 88.36% for nanorods and 76.81% for nanocubes, respectively In the same photocatalytic activity test condition such as visible light source, irradiation time (180 min), the MO degradation rate with the use of LFO nanospheres was about 83% (Dhinesh Kumar and Jayavel, 2014) Other recent studies have also explored and studied the influence of different morphologies on the photo-decomposition of organic contaminants For example, Thirumalairajan et

el found the floral-like LFO nanostructure comprised of nanosheets exhibited better photocatalytic efficiency (92.85%) than LFO nanofibers (80%) in photodegradation

of MB (Li, He et al., 2014; Thirumalairajan, Girija et al., 2014)

Table 5 Degradation rate of dyes solution on different morphologies LFO

nanoparticles

Catalyst Morphology Dye type and

concentration

Catalyst loading (mg)

Degradation rate (%)

Reference

s

LFO Nanorods RhB, 1g/l 100 88.36 (Thirumal

airajan, Girija et al., 2013) LFO Nanocubes RhB, 1g/l 100 76.81 (Thirumal

airajan, Girija et al., 2013) LFO Nanosphere

s

MB, 10mg/l 20 ~ 100 (Tang,

Tong et al., 2013) LFO Nanosphere

s

RhB, 1g/l 100 90.80 (Thirumal

airajan, Girija et al., 2013) LFO Nanosphere

nanosheet

RhB, 1g/l 100 90.66 (Thirumal

airajan, Girija et al., 2014) LFO Floral-like MB, 1g/l 100 92.85 (Thirumal

nanosheet airajan,

Girija et al., 2014) LFO Nanofibers MB, 10mg/l 100 75 (Li, He et

al., 2014)

5.4 Crystallinity

Crystallinity, as another factor that can effect on the photocatalytic performance of LFO photocatalyst, strongly depends on the nature of the starting material as well as the synthesis method (Bessekhouad, Robert et al., 2003) It is commonly known that the crystal structure is one of essential properties of the particles, thus the investigation on it is needed (Guo, Quan et al., 2007)

A photocatalyst with a good degree of crystallinity, which is presented by narrow and strong diffraction peaks in XRD spectrum, is necessary because it helps suppress recombination between photogenerated charge carriers, which improves photocatalytic efficiency (Li, Liu et al., 2013; Martin, 2014) In contrast, the amorphous characteristics of compounds induce a large amount of defects and each one acts as a recombination centre when electrons and holes are generated under

illuminations (Bessekhouad, Robert et al., 2003) For example, Thirumalairajan et

al reported that when the sample of LFO with a floral structure is calcined at

600°C, the amorphous phase disappeared and characteristic peaks of LFO appeared with low intensity at 650, 700 and 750°C Well crystalline powders were obtained after increasing the calcination temperature to 800°C, which possessed a high degradation efficiency to RhB (90.66%) and MB (92.85%) under illumination of visible light (Thirumalairajan, Girija et al., 2014), as shown in Figure 9

Figure 9 (a) XRD pattern at different calcination temperatures; (b and c) degradation of RhB and MB in the presence of LFO calcined at 800°C

(Thirumalairajan, Girija et al., 2014)

Su and co-workers have also proven that the increase of calcination temperature led

to an enhanced photocatalytic activity; the LFO nanoparticles calcined at 900 °C exhibited the best photocatalytic performance, with more than 50% of RhB degraded (Su, Jing et al., 2010) It is worth noted that all the LFO samples exhibited higher photocatalytic activity in visible light region than P25 TiO2 Figure 10 displayed the X-ray diffraction patterns of LFO calcined at different temperatures and photocatalytic performances of as-prepared LFO compared to P25 TiO2 in visible light for degradation of RhB solution

Figure 10 (a) XRD pattern of LFO samples calcined at different temperatures;

To conclude, one of factors affecting photocatalytic performance is attributed to the high crystallinity of sample This implied that the high crystallinity of LFO nanoparticles which caused by high calcination temperature could promote the photoinduced charge separation, thus enhancing the photocatalytic activity (Jin, Wu

et al., 2006)

5.5 Dopants

Previous studies have reported that the enhanced photocatalytic activity and superparamagnetic property of perovskite-based photocatalysts could be achieved

by several methods, including doping or adding a desirable component to the A or

B site One of the purposes of doping strategy is to supress the recombination of electron – hole pairs by reducing the particle size (Wang, Wang et al., 2012; Wei, Wang et al., 2012) For example, among the various doping strategies, Ca and Zn were used to dope LFO nanoparticles at A and B site, respectively Typically, after the substitution of A site with Ca, 10 mol% of Ca-doped LFO (La1-xCaxFeO3) photocatalyst exhibited a larger light absorption capability at λ > 420 nm than that

of pure LFO, thus improving the photodecomposition efficiency of MB from 48.9%

to 77.5% after 1 h photoirradiation (Li, Liu et al., 2010) The results revealed that the substitution of Ca2+ for La3+ led to a decrease of crystallite size and improve visible light adsorption capability of the photocatalyst as well as photocatalytic

performance The other reason is related to enhancement of photocatalytic performance could be discussed as follow Photocatalytic performance is associated with the formation of O vacancies and CaLa’donor centres due to displacement of

La3+ by Ca2+ ion When loading of dopant was low, the number of CaLa’donor centres increased, resulting in an increase of photogenerated electrons-holes and thus favouring the photocatalytic performance However, when the doping concentration was too high, the dopant acted as an photoinduced electron-hole recombination centre, causing the deactivation of photocatalyst (Asiltürk, Sayılkan

et al., 2009)

In comparison with B-site doping process, Dong et al., evaluated photodegradation

rate of MB by using nano-sized LaFe1-xZnxO3 synthesized by a sol-gel combustion method Based on their results, we can see that the photocatalytic activity of LaFe1-xZnxO3 nanoparticles with Zn doping at B site was higher than that

auto-of undoped LFO, which reached to 75% degradation rate as compared with 45% degradation rate of undoped LFO This was because the partial substitution of Fe3+

in LFO with Zn2+ led to the increase of lattice constant of Zn-doped LFO, whilst its crystal size decreased and in turn specific surface area increased (Dong, Xu et al., 2009)

So far, different metals, e.g Li, Al, Ag, have been utilized as dopants to vary the inherent properties of perovskite-based photocatalysts (Hou, Sun et al., 2006; Acharya and Chakrabarti, 2010; Desai and Athawale, 2013); whilst non-metal doping is considered as a new route to improve the photocatalytic performance of LFO for dye decomposition The photocatalytic performance of 3.5% B-doped LFO and undoped LFO were studied by decomposing of phenol under the condition of simulated sunlight irradiation After 300 min exposed to light source, the degradation rate of 3.5% B-doped LFO photocatalyst reached 87.9%, which was higher than that of undoped LFO (30.9%) (Wu, Hu et al., 2015)