**1. Introduction**

Semiconductor-based catalysis is a green technology, which gained considerable attention owing to its potential environmental applications, such as wastewater treatment, air purification and degradation of different organic contaminants [1–3]. In the past few decades, titanium dioxide (TiO2) as an n-type semiconductor has an attractive extensive interest as photocatalysts because of its easy availability, inertness, low costs, nontoxicity and chemical stability [4]. However, the large band gap energy for TiO2 (3.0–3.2 eV), requiring UV light that occupies around 5% of solar energy for excitation, limits its applications to a great extent [5]. Another difficulty is the high recombination rate of the photoexcited electron-hole pairs in TiO2 [6]. Many attempts have been developed to retard this electron-hole recombination and to increase the photocatalytic efficiency of TiO2, such as surface modification using a suitable metal ion and nonmetal dopant to increase the visible light absorbance and coupling with another semiconductor to enhance the charge separation efficiency [7,8]. Although in some cases improved photocatalytic activities were reported, very often the doping increased the number of structural defects acting as unwanted recombination centers. Therefore, the development of cost-effective, efficient and alternative photocatalysts with intrinsic narrow band gaps to increase the

visible light response has become a research focus [9,10]. Mixed metal oxides and oxynitrides attracted interest since many of them are visible-light active, cheap, non-toxic and stable [11].

Iron is highly abundant in the earth crust and thus cheap. Many mixed metal oxides containing iron, i.e., ferrites, offer suitable band gap energy for visible light absorption. Furthermore, the position of their valence band edges is more positive than the oxidation potential of O2/H2O (1.23 V vs. NHE) rendering them suitable for the photooxidation of water [12]. The high activity of ferrites for degradation of pollutants has been proven in many studies [13–15]. Ferrites with a perovskite structure, with a general formula of ABO3 with for example, A = rare-earth metal ion and B = Fe3+ ion, exhibit a wide range of ferro-, piezo-, and pyro-electrical properties rendering them suitable as magneto-optical material, electrode materials, structural materials, sensors and refractory materials [16]. The perovskite LaFeO3 is employed as a catalyst, e.g., in solid oxide fuel cells, but also in devices using its good dielectric properties and high piezoelectricity.

However, LaFeO3 has also been used as a photocatalyst; several studies focused on the synthesis and the activity for photodegradation of several organic dyes under visible light irradiation [17–19]. Thirumalairajan et al. synthesized floral-like LaFeO3 by a surfactant-assisted hydrothermal technique and found that the porous floral nanostructure led to higher photoactivity compared to bulk LaFeO3 for the degradation of different dyes, such as rhodamine B (RhB) and methylene blue (MB) [20]. Su et al. prepared large surface area nanosized LaFeO3 particles by employing SBA-16 as a hard template and compared its visible light activity for RhB degradation with that of LaFeO3 prepared by the citric acid assisted sol-gel route [21]. Yang et al. prepared LaFeO3 by conventional co-precipitation and enhanced its activity by post-treatment in molten salt [22]. Tijare et al. [23] formed nano-crystalline LaFeO3 perovskite by the sol-gel route and claimed activity for photocatalytic hydrogen generation under visible light irradiation.

In the present work, we applied the same synthetic route for LaFeO3 as Tijare et al. but altered (i) the duration of the thermal treatment and (ii) used a pyrolysis step at 400 ◦C instead of using ultra-sonication or drying at 90 ◦C in an oven. Citric acid assisted sol-gel was chosen as a synthesis route because, in general, it is a suitable method for the synthesis of nanopowders with a well-developed high specific surface area obtained at low calcination temperature and short times without employing expensive sacrificial structure-directing agents or template structures. The visible light activity for degradation of RhB and 4-chlorophenol (4-CP) as model organic pollutants was investigated. As for Tijare et al., we also attempted hydrogen generation, however, failed with that and suspected it was based on the Mott-Schottky plots calculating band positions that the conduction band edge of LaFeO3 was too positive than the reduction potential of H2/H2O (0 V vs. NHE) to create electrons which were reductive enough to react with protons to hydrogen.
