Enhanced UV-Vis Photodegradation of Nanocomposite Reduced Graphene Oxide / Ferrite Nanoﬁber Films Prepared by Laser-Assisted Evaporation

: Nanocomposite ﬁlms of rGO / MFeO 3 (M = Bi, La) nanoﬁbers were grown by matrix-assisted pulsed laser evaporation of frozen target dispersions containing GO platelets and MFeO 3 nanoﬁbers. Electron microscopy investigations conﬁrmed the successful fabrication of MFeO 3 nanoﬁbers by electrospinning Part of nanoﬁbers were broken into shorter units, and spherical nanoparticles were formed during laser processing. Numerical simulations were performed in order to estimate the maximum temperature values reached by the nanoﬁbers during laser irradiation. X-ray di ﬀ raction analyses revealed the formation of perovskite MFeO 3 phase, whereas secondary phases of BiFeO 3 could not be completely avoided, due to the high volatility of bismuth. XPS measurements disclosed the presence of metallic bismuth and Fe 2 + for BiFeO 3 , whereas La 2 (CO 3 ) 3 and Fe 2 + were observed in case of LaFeO 3 nanoﬁbers. High photocatalytic e ﬃ ciencies for the degradation of methyl orange were achieved for nanocomposite ﬁlms, both under UV and visible light irradiation conditions. Degradation values of up to 70% after 400 min irradiation were obtained for rGO / LaFeO 3 nanocomposite thin layers, with weights below 10 µ g, rGO platelets acting as reservoirs for photoelectrons generated at the surface of MFeO 3 .


Introduction
Water purification from organic materials and other hazardous chemicals is of significant importance, given the high industrial demands of our society. One of the major pollutants found in wastewater is organic dyes, resulting from textile production and other industrial processes [1]. Different wastewater treatment methods are available, ranging from physical and chemical to biological procedures [2,3]. Photocatalytic decomposition of organic pollutants by using sunlight is an environmentally friendly approach developed for wastewater purification. Semiconductor transition metal oxides, most frequently TiO 2 or ZnO, are used in photocatalytic processes, due to their non-toxicity, photochemical stability, and low cost [4,5]. Even so, their absorption is limited to the UV range (~5% of total sunlight radiation) due to their large band gap (~3.2 eV). Their high electron-hole recombination rate is also inconvenient, as it must be overcome in order to reach high photocatalytic min −1 . A high voltage of 15-20 kV was applied with a power supply in order to establish an electric field (1.5-2.0 kV·cm −1 ) between the needle and the collector, kept 10 cm apart. The samples were then calcined at 600 °C for 5 h, at a heating rate of 1 °C min −1 [33,34].

Preparation of MAPLE Targets
Dispersions of 3 wt.% MFO nanofibers and 5 wt.% GO platelets (monolayers, NanoInnova Technologies, Madrid, Spain), as well as 3 wt.% MFO and 1 or 5 wt.% GO platelets in distilled water, were prepared for the growth of the reference single-component, as well as nanocomposite layers. The dispersions were then stirred and sonicated for 30 min, to ensure their homogeneity, and subsequently immersed in liquid N2.

Deposition of MFO Nanofiber/GO Nanocomposite Films by MAPLE Technique
A frequency-quadrupled Quantel Brilliant B Nd:YAG laser source (λ = 266 nm, τFWHM~5 ns, υ = 10 Hz, Quantel Lumibird, Lannion, France) was used for the irradiation of the frozen targets. The experiments were carried out under controlled ambient conditions, in a stainless-steel reaction chamber. Prior to irradiations, the chamber was evacuated down to a residual pressure of 0.1 Pa and then filled with N2 at 20 Pa. The angle between the incident laser beam and the targets' surface was 45°. The laser beam scanned the target over a surface area of approximately 1 × 1 cm 2 . The laser fluence on the target surface was set at 400 mJ·cm −2 . Then, 5000 laser pulses were applied for the deposition of each layer. F-doped SnO2 (FTO)/glass (Merck KGaA, Darmstadt, Germany) and Cu (Advent, Eynsham, UK) were used as substrate materials. Previously, the substrates were subsequently cleaned in a Hellmanex soap/water solution, deionized water, acetone and ethanol, and then dried in a hot air flow. The substrates were placed parallel to the target surface, at a separation distance of 4 cm, and were maintained at room temperature during the growth of the layers. Figure  1 illustrates the experimental approach employed during the preparation of the nanocomposite GO/MFO nanofiber films.

Morphological, Structural and Compositional Analyses
The morphological characterization of the films was performed by field emission scanning electron microscopy (FE-SEM) and high-resolution transmission electron microscopy (HRTEM), with the aid of a QUANTA FEI 200 FEG-ESEM (Thermo Fisher Scientific Inc., Waltham, MA, USA) and a JEOL JEM-2200FS 200 kV FEG TEM (Jeol Ltd., Tokio, Japan) with in-column filter and UHR pole piece electron microscopes. X-ray diffraction (XRD) patterns were recorded on a STOE STADI MP diffractometer (Mo source, λ = 0.7107 Å ).
The thermal rise induced in the nanofiber/GO nanocomposite samples by laser irradiation during MAPLE deposition was simulated by solving the transient heat equation, using the finite

Morphological, Structural and Compositional Analyses
The morphological characterization of the films was performed by field emission scanning electron microscopy (FE-SEM) and high-resolution transmission electron microscopy (HRTEM), with the aid of a QUANTA FEI 200 FEG-ESEM (Thermo Fisher Scientific Inc., Waltham, MA, USA) and a JEOL JEM-2200FS 200 kV FEG TEM (Jeol Ltd., Tokio, Japan) with in-column filter and UHR pole piece electron microscopes. X-ray diffraction (XRD) patterns were recorded on a STOE STADI MP diffractometer (Mo source, λ = 0.7107 Å).
The thermal rise induced in the nanofiber/GO nanocomposite samples by laser irradiation during MAPLE deposition was simulated by solving the transient heat equation, using the finite element method, with COMSOL 5.3 Multiphysics software (COMSOL Inc., Stockholm, Sweden) [35][36][37][38][39][40][41][42]. The chemical composition of the samples was investigated by X-ray photoelectron spectroscopy (XPS). The measurements were performed at a residual pressure of~10 −9 mbar, using an ESCA M-Probe monochromatic Al Kα excitation source (λ = 8.33 Å) (Surface Science Instruments, Thermo Fisher Scientific Inc., Waltham, MA, USA). The acquisition of survey scans was performed with a detector pass energy of 158.9 eV, whilst a pass energy of 22.9 eV was used to record the high-resolution spectra. The C1s signal of adventitious carbon (284.8 eV) was used as reference for the correction of the binding energies for all spectra. Data analysis and peak deconvolution were performed with the CasaXPS software 2.3.15 (Casa Software Ltd., Teignmouth, UK).

Investigation of Photocatalytic Degradation Efficiency of MFO Nanofiber/GO Nanocomposite Films under UV or Visible Light Irradiation
The photocatalytic efficiency of the layers deposited by MAPLE technique was evaluated through the degradation of MO (C 14 H 14 N 3 NaO 3 S) organic dye molecules. The experiments were performed under UV or visible light irradiation, using an LZC-ICH2 photoreactor from Luzchem Research Inc., Ottawa, ON, Canada. 16 × 8 W UV lamps with an emission wavelength centered at 360 nm (300-400 nm emission range, LZC-UVA, Hitachi lamp) or 16 × 8 W visible lamps centered at 580 nm (420-700 nm emission range, LZC-Vis, OSRAM lamp, Sylvania, Wilmington, NC, USA) were used as light sources. 2 mL MO solutions with 1.25 × 10 −5 M (4 ppm) initial MO dye concentration (C 0 ) were irradiated in quartz vials. The changes of the peak absorbance of the MO solution at 464 nm wavelength was measured with the aid of a Shimadzu UV-2600 spectrophotometer (Shimadzu Scientific Instruments, Columbia, MD, USA). The MO dye concentration was calculated by using the Lambert-Beer law, from the measured peak absorbance value. The photocatalytic degradation efficiencies, η, of the samples were calculated with the formula η [%] = (C 0 − C) × 100/C 0 , where C is the MO concentration in the aqueous solution under UV or visible light irradiation, measured at regular time intervals.

Morphological, Structural and Compositional Characterization
The as-spun MFeO x (M = Bi, La) nanofibers possess a smooth surface morphology with rather uniform widths, with diameters of approximately 200-450 nm, while their lengths are in the tens of micrometers range (Figure 2a). The nanofibers' surface becomes porous, granular and rougher after calcination, with grains of around 50 to 100 nm in size. There is also a shrinkage in dimensions during annealing of MFO nanofibers, mainly due to PVP removal, diameters being around 100-250 nm ( Figure 2b) [43]. The presence of thicker regions is due to the merging/attachment of nanofibers during calcination and crystallization [44].
Ferrite nanofibers were then deposited together with GO nanoplatelets by MAPLE, through the irradiation of flash-frozen targets (see Section 2.3). The film containing only BFO presents a surface morphology consisting of spherical particles with sizes between approximately 100 nm and 1 µm, and fiber-like structures are not present (Figure 2c). The LFO film is also formed by round particles with sizes similar to BFO sample. In addition, short wire-like structures with widths of~100-300 nm, and lengths of approximately between 500 nm and 2 µm, are also present. The formation of spherical nanoparticle aggregates has been previously attributed to partial melting of the material upon absorption of the laser radiation [45]. Thus, the fluence used in the current experiments (~400 mJ·cm −2 ) will likely melt part of the nanofibers, provided that the band gap of ferrite perovskites typically ranges from 2.0 to 3.0 eV and the laser photon energy is~4.7 eV, resulting in surface morphologies also observed for other oxide materials submitted to laser irradiation [25,46]. and this can be correlated with their lower relative concentration in the target dispersions and by the intrinsic physical properties of the GO platelets (lower absorption coefficient at the wavelength of the incident laser radiation) [47]. Additional areas could be identified with the presence of elongated structures which resemble broken nanofibers (Supplementary Figures S1 and S2). HRTEM images confirm the presence of GO platelets, round particles and nanofibers (Supplementary Figure S3). Numerical simulations were performed in order to estimate the temperature evolutions during the MAPLE deposition, both for MFO and GO platelets, and to explain the presence of nanoparticles on the surface of the MFO, as well as MFO/GO hybrid composite films. To this aim, the transient heat equation was solved, employing the optical and thermophysical magnitudes reported in References [35][36][37][38][39][40][41][42]. The laser pulse duration was assumed to have a Gaussian shape (FWHM~5 ns), and distilled water solvent was supposed to have a negligible absorption at the wavelength of the laser radiation. Figure 3 shows that temperatures reached due to GO platelets' heating through the absorption of laser radiation are larger when compared to the water ice boiling point and increase with the GO flake thickness. Moreover, the heating has a duration of tens to a few hundred ns. Upon interaction with the MFO nanofibers, the absorbed laser radiation induces a temperature rise, notably larger than the one achieved by GO platelets, overpassing the MFO melting point after the first 2 ns of the laser pulse (Tm (LFO/BFO) ~1250 K). Such rapid heating will also derive in the rising of the surrounding water ice temperature by heat transfer, leading to explosive boiling and vapor-driven ejection of MFO nanofibers and GO platelets toward the substrate surface. Numerical simulations also explain the presence of molten/spherical nanoparticles, since the vaporization temperature of LaFeO3 nanofibers situated at the surface of the MAPLE target, reported at ~2120 K [48], is surpassed during laser irradiation. On the other hand, BiFeO3 decomposes due to the volatility of Bi [49][50][51], and iron oxides evaporate at temperatures around 2000 K [52]. Furthermore, the fluence distribution throughout the target is inhomogeneous, since absorption of laser radiation occurs along the whole optical penetration depth. This shielding effect of the superficial material leads to temperatures below the The incorporation of the GO platelets into the film's structure can be clearly detected upon their addition to the starting dispersions (Figure 2c-e). The increasing amount of GO platelets in the nanocomposite films results in a reduction of the number of particles and fibers present at the surface, and this can be correlated with their lower relative concentration in the target dispersions and by the intrinsic physical properties of the GO platelets (lower absorption coefficient at the wavelength of the incident laser radiation) [47]. Additional areas could be identified with the presence of elongated structures which resemble broken nanofibers (Supplementary Figures S1 and S2). HRTEM images confirm the presence of GO platelets, round particles and nanofibers (Supplementary Figure S3).
Numerical simulations were performed in order to estimate the temperature evolutions during the MAPLE deposition, both for MFO and GO platelets, and to explain the presence of nanoparticles on the surface of the MFO, as well as MFO/GO hybrid composite films. To this aim, the transient heat equation was solved, employing the optical and thermophysical magnitudes reported in References [35][36][37][38][39][40][41][42]. The laser pulse duration was assumed to have a Gaussian shape (τ FWHM~5 ns), and distilled water solvent was supposed to have a negligible absorption at the wavelength of the laser radiation. Figure 3 shows that temperatures reached due to GO platelets' heating through the absorption of laser radiation are larger when compared to the water ice boiling point and increase with the GO flake thickness. Moreover, the heating has a duration of tens to a few hundred ns. Upon interaction with the MFO nanofibers, the absorbed laser radiation induces a temperature rise, notably larger than the one achieved by GO platelets, overpassing the MFO melting point after the first 2 ns of the laser pulse (T m (LFO/BFO)~1250 K). Such rapid heating will also derive in the rising of the surrounding water ice temperature by heat transfer, leading to explosive boiling and vapor-driven ejection of MFO nanofibers and GO platelets toward the substrate surface. Numerical simulations also explain the presence of molten/spherical nanoparticles, since the vaporization temperature of LaFeO 3 nanofibers situated at the surface of the MAPLE target, reported at~2120 K [48], is surpassed during laser irradiation. On the other hand, BiFeO 3 decomposes due to the volatility of Bi [49][50][51], and iron oxides evaporate at temperatures around 2000 K [52]. Furthermore, the fluence distribution throughout the target is inhomogeneous, since absorption of laser radiation occurs along the whole optical penetration depth. This shielding effect of the superficial material leads to temperatures below the MFO melting point in deeper zones, but above the water boiling point, enabling transportation of entire or partially molten nano-entities. The addition of GO also contributes to diminish the developed temperature in the irradiated targets, due to absorption of the laser radiation [42], which in turn derives in a lower vaporization rate compared to pure nanofiber samples.
Crystals 2020, 10, x FOR PEER REVIEW 6 of 18 MFO melting point in deeper zones, but above the water boiling point, enabling transportation of entire or partially molten nano-entities. The addition of GO also contributes to diminish the developed temperature in the irradiated targets, due to absorption of the laser radiation [42], which in turn derives in a lower vaporization rate compared to pure nanofiber samples. The crystalline phase of the MFO nanofibers, as well as nanocomposite films, was investigated by XRD ( Figure 4). The formation of a polycrystalline BFO phase is inferred by indexing the diffraction pattern ( Figure 4a). The space group corresponds to the R3 ̅ c with lattice parameters a~5.59 Å and c~13.87 Å (ICSD 15299). The most intense peaks of feasible impurity phases, such as Bi2Fe4O9 (ICSD 186440) and Bi25FeO40 (ICSD 41937), have been indexed in the diffraction pattern, to show that their presence is insignificant, although it cannot be completely discarded, particularly due to the rather strict temperature conditions for the fabrication of pure phase BFO, as well as the volatility of bismuth [49][50][51]. The LFO nanofibers pattern (Figure 4b) can be fitted to the pseudo-cubic perovskite Pm-3m structure with a lattice parameter aLFO = 3.926 Å (ICSD 29234). Some peaks that may be related to a Fe2O3 phase (ICSD 173655) were identified. The Debye-Scherrer formula was used to estimate an average crystallite size of 43 ± 16 nm (BFO) and 58 ± 15 nm (LFO) by fitting of the (110) peak. These values are smaller compared to the results reported by SEM, as expected, since XRD measures the small units that form larger grains [53]. The MFO nanofibers/rGO nanocomposite films deposited on FTO/glass present some differences between each other. On the one hand, it can be confirmed that both MFO phases remain after MAPLE deposition. The melting and recrystallization of oxides without phase segregation by laser irradiation is a phenomenon that has already been reported [25,54]. On the other hand, there is a substantial decrease in the BFO peak intensities of nanocomposite films, as compared to the corresponding single-component layer, which could be attributed to partial melting and decomposition of the nanofibers during laser-induced heating, as evidenced by our numerical simulations. Indeed, prolonged exposure to calcinations of BFO was reported to lead to bismuth evaporation and to the formation of secondary phases [55]. Nevertheless, the high signal-to-noise ratio does not allow the detection of other crystalline phases. The broad The crystalline phase of the MFO nanofibers, as well as nanocomposite films, was investigated by XRD ( Figure 4). The formation of a polycrystalline BFO phase is inferred by indexing the diffraction pattern ( Figure 4a). The space group corresponds to the R3c with lattice parameters a~5.59 Å and c~13.87 Å (ICSD 15299). The most intense peaks of feasible impurity phases, such as Bi 2 Fe 4 O 9 (ICSD 186440) and Bi 25 FeO 40 (ICSD 41937), have been indexed in the diffraction pattern, to show that their presence is insignificant, although it cannot be completely discarded, particularly due to the rather strict temperature conditions for the fabrication of pure phase BFO, as well as the volatility of bismuth [49][50][51]. The LFO nanofibers pattern (Figure 4b) can be fitted to the pseudo-cubic perovskite Pm-3m structure with a lattice parameter a LFO = 3.926 Å (ICSD 29234). Some peaks that may be related to a Fe 2 O 3 phase (ICSD 173655) were identified. The Debye-Scherrer formula was used to estimate an average crystallite size of 43 ± 16 nm (BFO) and 58 ± 15 nm (LFO) by fitting of the (110) peak. These values are smaller compared to the results reported by SEM, as expected, since XRD measures the small units that form larger grains [53]. The MFO nanofibers/rGO nanocomposite films deposited on FTO/glass present some differences between each other. On the one hand, it can be confirmed that both MFO phases remain after MAPLE deposition. The melting and recrystallization of oxides without phase segregation by laser irradiation is a phenomenon that has already been reported [25,54]. On the other hand, there is a substantial decrease in the BFO peak intensities of nanocomposite films, as compared to the corresponding single-component layer, which could be attributed to partial melting Crystals 2020, 10, 271 7 of 18 and decomposition of the nanofibers during laser-induced heating, as evidenced by our numerical simulations. Indeed, prolonged exposure to calcinations of BFO was reported to lead to bismuth evaporation and to the formation of secondary phases [55]. Nevertheless, the high signal-to-noise ratio does not allow the detection of other crystalline phases. The broad amorphous shoulder at angles below 17 • belongs to the glass substrate and probably some amorphized rGO.
Crystals 2020, 10, x FOR PEER REVIEW 7 of 18 amorphous shoulder at angles below 17° belongs to the glass substrate and probably some amorphized rGO. XPS investigations were performed in order to determine the surface composition and metal oxidation states of the samples. The XPS spectra in Supplementary Figure S4 and Reference [13] show the elemental composition of MFO nanofibers. From the survey spectra (Supplementary Figure S4a), the elemental ratios for BFO nanofibers are estimated to be 17.0 (Bi), 6.8 (Fe), 41.2 (O) and 34.8 at.% (C), whereas the composition for LFO nanofibers is 7.0 (La), 6.6 (Fe), 56.4(O) and 29.7 at.% (C). The presence of carbon was mainly attributed to adventitious carbon species [56], although some carbon residues stemming from the decomposition of organic material during the nanofiber synthesis process might also be present (Supplementary Figure S4b) [13,34]. The presence of lanthanum carbonate La2(CO3)3 on the LFO sample surface was also detected, due to its high reactivity with ambient CO2 [13,57]. The La-Fe-O elemental ratio in LFO fibers is estimated to be close to the expected 1:1:3, whereas the nonstoichiometric ratio of Bi-Fe-O for BFO has been attributed to several origins. XPS investigations were performed in order to determine the surface composition and metal oxidation states of the samples. The XPS spectra in Supplementary Figure S4 and Reference [13] show the elemental composition of MFO nanofibers. From the survey spectra (Supplementary Figure S4a), the elemental ratios for BFO nanofibers are estimated to be 17.0 (Bi), 6.8 (Fe), 41.2 (O) and 34.8 at.% (C), whereas the composition for LFO nanofibers is 7.0 (La), 6.6 (Fe), 56.4(O) and 29.7 at.% (C). The presence of carbon was mainly attributed to adventitious carbon species [56], although some carbon residues stemming from the decomposition of organic material during the nanofiber synthesis process might also be present (Supplementary Figure S4b) [13,34]. The presence of lanthanum carbonate La 2 (CO 3 ) 3 on the LFO sample surface was also detected, due to its high reactivity with ambient CO 2 [13,57]. The La-Fe-O elemental ratio in LFO fibers is estimated to be close to the expected 1:1:3, whereas the nonstoichiometric ratio of Bi-Fe-O for BFO has been attributed to several origins. The XPS detection limit of light elements, such as iron and oxygen, is known to increase in presence of heavy elements, like bismuth [58]. Furthermore, it has been reported that the accuracy in the quantification of some transition metals is influenced by the orbitals selected for quantification; in the case of iron, it can even vary by a factor four [59]. The Bi 4f region was deconvoluted by fitting two sets of doublets (Supplementary Figure S4c). The peaks with higher binding energies, 159.1 and 164.4 eV, respectively, were correlated with Bi-O bonds in BiFeO 3 . The low-intensity lines located at 157.4 and 162.7 eV were attributed to the presence of metallic bismuth (21.08 at.%) [60][61][62][63], which are often observed after calcination due to the high volatility of bismuth [49][50][51]. Elucidation of oxygen vacancies concentration has been investigated through deconvolution of the O 1s spectrum [64][65][66]. Three peaks were identified at binding energies of 529.95 eV (81.20%), 531.5 eV (9.44%) and 532.1 eV (9.35%) ( Figure S4d), which were respectively assigned to oxygen lattice bonds in BiFeO 3 [67], oxygen deficient regions and absorbed hydroxides, as well as organic species that may remain after calcination of the BFO nanofibers [67,68]. By taking into account the ratio between the former two peaks [66], we estimate a concentration of oxygen vacancies of~12 %. In the Fe 2p high-resolution spectrum (Supplementary Figure S4e), the Fe 2p 3/2 signal could be fitted with a multiplet of five peaks situated at 709.7, 710.7, 711.5, 712.6 and 713.6 eV, binding according to typical values of Fe 3+ in Fe 2 O 3 or BiFeO 3 [69]. The absence of the typical Fe 2+ satellite peak at 715.4 eV indicates that no significant amount Fe 2+ species is present. For LFO nanofibers [13], the La 3d spectrum has two contributions, corresponding to lanthanum carbonate and LFO. In the Fe 2p spectrum, we could identify the presence of Fe 2+ with an amount of 29. In all samples of the BFO nanocomposite film series, the M:Fe:O stoichiometry ratios are slightly different from the nanofiber counterparts (Figure 5a). The excess oxygen observed in the LFO survey is most probably related to the formation of lanthanum carbonate (Figure 5b). The metallic bismuth peak is also found in the high-resolution Bi 4f spectra with a maximum percentage of~30% (Figure 6a and Supplementary Figure S5a,b). Since it is well-known that oxidized bismuth species can be thermally reduced in the presence of carbon, we assume that the presence of rGO, as well as the laser-driven heating demonstrated by the temperature simulations, may be promoting the formation of Bi 0 during the laser deposition process [70]. On the other hand, the La 3d spectra for rGO/LFO films also show the presence of larger amounts of La 2 (CO 3 ) 3 , i.e., between 10% and 15% more, as compared to LFO nanofibers (Figure 6b and Supplementary Figure S5c,d). Besides the possible reaction with atmospheric humidity and CO 2 during storage [71], the laser-driven reaction of LFO with the carbon present in the rGO platelets could explain this increase.
Another interesting feature is the change in the peak structure of the Fe 2p region by increasing the rGO content (Supplementary Figure S6). As we mentioned before, the Fe 2p 3/2 line of all films presents the characteristic structure of Fe 3+ related to BFO and LFO. In the BFO series, the appearance of Fe 2+ peaks can be identified, indicating the reduction of Fe 3+ to Fe 2+ due to the presence of rGO. The LFO films present an equivalent satellite structure to that of BFO films. Three peaks could be identified in the high-resolution O1s spectra ( Figure S7) that can be associated to lattice oxygen (529.9 eV), oxygen vacancies and hydroxides (531.5 eV) and organic oxygen (532.1 eV) from carbon-oxygen groups (C-O-C, C-OH, C=O and O-C=O).  Another interesting feature is the change in the peak structure of the Fe 2p region by increasing the rGO content (Supplementary Figure S6). As we mentioned before, the Fe 2p3/2 line of all films presents the characteristic structure of Fe 3+ related to BFO and LFO. In the BFO series, the appearance of Fe 2+ peaks can be identified, indicating the reduction of Fe 3+ to Fe 2+ due to the presence of rGO. The LFO films present an equivalent satellite structure to that of BFO films. Three peaks could be identified in the high-resolution O1s spectra ( Figure S7) that can be associated to lattice oxygen (529.9 eV), oxygen vacancies and hydroxides (531.5 eV) and organic oxygen (532.1 eV) from carbon-oxygen groups (C-O-C, C-OH, C=O and O-C=O).
The high-resolution C1s peak of the GO precursor, reported elsewhere [32,72,73], shows a large content of oxygen functional groups. At binding energies of 284.5 and 284.8 eV, we can respectively find the peaks associated to sp 2 graphitic carbon bonds (C=C), adventitious carbon and sp 3 bonded carbon (C-C, C-H). As mentioned before, it is possible that these peaks have some contribution   Another interesting feature is the change in the peak structure of the Fe 2p region by increasing the rGO content (Supplementary Figure S6). As we mentioned before, the Fe 2p3/2 line of all films presents the characteristic structure of Fe 3+ related to BFO and LFO. In the BFO series, the appearance of Fe 2+ peaks can be identified, indicating the reduction of Fe 3+ to Fe 2+ due to the presence of rGO. The LFO films present an equivalent satellite structure to that of BFO films. Three peaks could be identified in the high-resolution O1s spectra ( Figure S7) that can be associated to lattice oxygen (529.9 eV), oxygen vacancies and hydroxides (531.5 eV) and organic oxygen (532.1 eV) from carbon-oxygen groups (C-O-C, C-OH, C=O and O-C=O).
The high-resolution C1s peak of the GO precursor, reported elsewhere [32,72,73], shows a large content of oxygen functional groups. At binding energies of 284.5 and 284.8 eV, we can respectively find the peaks associated to sp 2 graphitic carbon bonds (C=C), adventitious carbon and sp 3 bonded carbon (C-C, C-H). As mentioned before, it is possible that these peaks have some contribution The high-resolution C1s peak of the GO precursor, reported elsewhere [32,72,73], shows a large content of oxygen functional groups. At binding energies of 284.5 and 284.8 eV, we can respectively find the peaks associated to sp 2 graphitic carbon bonds (C=C), adventitious carbon and sp 3 bonded carbon (C-C, C-H). As mentioned before, it is possible that these peaks have some contribution coming from residual carbon due to PVP decomposition during nanofiber calcination, despite that the resulting nanofibers are completely crystalline [13,34]. The peaks corresponding to single-bonded carbon-oxygen groups, such as epoxy (C-O-C) and hydroxyl (C-OH), can be found at approximately 286.3 eV, whereas the peaks at binding energies of 287.8 and 288.8 eV are typically assigned to carbonyl (C=O) and carboxyl (O-C=O) groups [74,75]. The C1s spectra of the GO/BFO composite (Figure 6c and Supplementary Figure S8b), as well as GO/LFO composite layers (Figure 6d and Supplementary Figure S8d) confirm the significant reduction of the number of oxygen-containing functional C-OH, C-O-C, C=O and COOH groups of GO platelets during laser processing. In addition, the low-intensity peaks centered at 291 eV can be assigned to the π-π* shake-up satellite of the sp 2 graphitic C peak at 284.5 eV, indicating the restoration of π conjugated domains, and is characteristic for aromatic structures [76].

Photocatalytic Degradation of Methyl Orange
The photocatalytic degradation efficiency of the MFO (M = Bi, La) nanofiber/rGO nanocomposite films was investigated both under UV and visible light irradiation (Supplementary Figures S9-S13). Photodegradation experiments were carried out in quartz reactors containing the MO solutions. Control experiments in the absence of photocatalysts were performed under UV and visible light irradiation, which evidenced the high photostability of MO molecules (Supplementary Figure S9). The photodegradation efficiency of single-component GO, BFO and LFO films was found to be very low, reaching only around 5% after 450 min of UV light exposure (Figure 7a,b). The inclusion of rGO in the composition of the layers led to the gradual increase of the degradation efficiency, both in case of BFO and LFO. The photodegradation efficiency of the BFO/rGO and LFO/rGO composites with 1 and 5 wt.% rGO was found to be approximately 43% and 70%, respectively, after 450 min of UV light irradiation (Figure 7). Our results are in good agreement with previous results reported in the literature concerning photocatalytic degradation efficiency of MFO/rGO composite layers. The inclusion of rGO improved the photocatalytic activity, demonstrating considerably better degradation of methylene blue [77] or Rhodamine B [20] dye molecules and higher hydrogen production by water-splitting, as compared to pure BiFeO 3 [77].    The photocatalytic properties of the LFO/rGO composite photocatalyst containing 5 wt.% rGO, with the highest degradation efficiency under UV light irradiation, were also tested during visible light irradiation (Figure 8). Similar to the previous experiments, around 70% of the initial dye molecules were decomposed after 450 min of light exposure. However, the degradation efficiency of the layer decreased until around 27% during the second consecutive degradation cycle.
This process can also take place under visible light irradiation. The maximum emission wavelength of the visible lamps used for the photocatalytic tests is 580 nm, corresponding to a photon energy of 2.1 eV, while their total emission range is 420-700 nm, (1.8-2.9) eV total photon energy. Thus, part of the emitted radiation can be absorbed by the BFO and LFO nanofibers, i.e., photon energies equal or higher than their band gap. The positively charged h + oxidize the water molecules in contact with the nanofibers' surface, generating hydroxyl radicals, which later oxidize the MO dye, forming inorganic products [79,80], CO2 and H2O (Equations (2) and (3)): *OH + pollutant → H2O + CO2 + inorganic products (SO4 2− , NO3 − , NH4 + ) The photogenerated e − and h + can be trapped in the MFeO3 lattice and rapidly recombine [81,82]. However, the addition of rGO and its high electrical conductivity supplied by the π-π conjugated structure and the low amount of oxygen-containing functional groups further enhance the electrical conductivity and promote the separation of e − and h + , improving the degradation capabilities as compared to MFO or rGO alone [83,84]. In addition, the photogenerated e − reacts with the adsorbed O2, forming a superoxide radical anion (O2* − ), also contributing to pollutant degradation (Equations (4) and (5)) [29,85]: O2* − + pollutant → H2O + CO2 + salts (5) Moreover, rGO also contributes to the generation of e − and h + under UV and visible light irradiation (Equation (6)) [29]: The apparent photodegradation rate constant, k app , of the BFO/rGO and LFO/rGO composite photocatalyst was calculated according to first-order kinetics, −dC/dt = k app C, both for UV and visible light irradiation. After the integration of the equation between the limits C 0 at t = 0 and C at irradiation time, t, the apparent rate constant is given by k app = ln (C 0 /C)/t. Moreover, k app was calculated for each sample, from the linear part of the ln (C 0 /C) plots versus t [78], before the saturation limit, t sat , was reached for different time intervals. The results are presented in Table 1. As can be observed, the k app values are similar under UV light irradiation, being slightly lower in the case of the samples with a low rGO concentration. Moreover, saturation has been reached earlier in case of BFO/rGO, as compared to the LFO/rGO layers, and at lower rGO concentration. The saturation behavior can be attributed to adsorption of dye molecules by the BFO nanomaterial during the photocatalytic test. The relative concentration of BFO is higher in the case of the BFO/rGO composite containing 1 wt.% rGO, accompanied with a low degradation rate, led to a faster saturation. Similarly, adsorption and photodegradation processes simultaneously take place during visible light irradiation of MO dye solutions in the presence of LFO/rGO composite layers. The process was slowed down after 120 min of light exposure, due to the effective degradation of MO molecules, as compared to UV light irradiation. During the second degradation cycle, part of the dye molecules adsorbed during the previous degradation run remained on the surface after the washing step, causing a faster saturation (Figure 8). The UV light, with maximum emission wavelength of 360 and 300-400 nm emission range corresponds to a photon energy of 3.4 eV for the maximum emission wavelength and 3.1-4.1 eV total photon energy range. These photon energy values are larger than the band gaps of BFO and LFO (E g = 2.0-2.7 eV), ejecting electrons (e − ) to the conduction bands (CB) of BFO and LFO, leading to the formation of holes (h + ) in their valence bands (VB) (Equation (1)) [18,29]: This process can also take place under visible light irradiation. The maximum emission wavelength of the visible lamps used for the photocatalytic tests is 580 nm, corresponding to a photon energy of 2.1 eV, while their total emission range is 420-700 nm, (1.8-2.9) eV total photon energy. Thus, part of the emitted radiation can be absorbed by the BFO and LFO nanofibers, i.e., photon energies equal or higher than their band gap. The positively charged h + oxidize the water molecules in contact with the nanofibers' surface, generating hydroxyl radicals, which later oxidize the MO dye, forming inorganic products [79,80], CO 2 and H 2 O (Equations (2) and (3)): The photogenerated e − and h + can be trapped in the MFeO 3 lattice and rapidly recombine [81,82]. However, the addition of rGO and its high electrical conductivity supplied by the π-π conjugated structure and the low amount of oxygen-containing functional groups further enhance the electrical conductivity and promote the separation of e − and h + , improving the degradation capabilities as compared to MFO or rGO alone [83,84]. In addition, the photogenerated e − reacts with the adsorbed O 2 , forming a superoxide radical anion (O 2 * − ), also contributing to pollutant degradation (Equations (4) and (5)) [29,85]: Moreover, rGO also contributes to the generation of e − and h + under UV and visible light irradiation (Equation (6)) [29]: The degradation process is schematically presented in Figure 9.

Conclusions
Nanocomposite films consisting of MFO (M = Bi, La) nanofibers and rGO platelets were synthesized by using the MAPLE technique. The layers were used as photocatalysts for organic dye degradation. Formation of shorter nanofiber units was caused by breakage of nanofibers. Temperature simulations confirmed that nanoparticles' formation takes place due to the partial melting of nanofibers during laser processing and subsequent recrystallization. The MFO/rGO nanocomposite films presented enhanced photocatalytic activities toward degradation of methyl orange dye molecules, under UV or visible light irradiation, with LFO/rGO showing values around two times higher than BFO/rGO layers. The injection of photogenerated charge carriers from the MFeO3 into the π-π conjugated structure of rGO is the main mechanism responsible for the improved photocatalytic degradation efficiency as compared to pure MFO or GO layers. In addition, the formation of intra-gap states due to the presence of oxygen vacancies allows for the extension of the absorption range to the visible range of the solar spectrum.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1. Figure S1: Additional SEM micrographs of BFO/GO nanocomposites. Figure S2: Additional SEM micrographs of LFO/GO nanocomposites. Figure S3: HRTEM of LFO/GO nanocomposites. Figure S4: XPS spectra of BFO nanofibers. Figure S5: High-resolution XPS spectra of Bi 4f and La 3d lines. Figure S6: High-resolution XPS spectra of Fe 2p lines. Figure S7: High-resolution XPS spectra of O 1s lines. Figure S8: High-resolution XPS spectra of C 1s lines. Figure S9: Absorption spectra of methyl orange references. Figure S10: Absorption spectra of methyl orange with a GO sample. Figure S11: Absorption spectra of methyl orange under UV light for BFO/GO nanocomposites. Figure S12: Absorption spectra of methyl orange under UV light for LFO/GO nanocomposites. Figure S13

Conclusions
Nanocomposite films consisting of MFO (M = Bi, La) nanofibers and rGO platelets were synthesized by using the MAPLE technique. The layers were used as photocatalysts for organic dye degradation. Formation of shorter nanofiber units was caused by breakage of nanofibers. Temperature simulations confirmed that nanoparticles' formation takes place due to the partial melting of nanofibers during laser processing and subsequent recrystallization. The MFO/rGO nanocomposite films presented enhanced photocatalytic activities toward degradation of methyl orange dye molecules, under UV or visible light irradiation, with LFO/rGO showing values around two times higher than BFO/rGO layers. The injection of photogenerated charge carriers from the MFeO 3 into the π-π conjugated structure of rGO is the main mechanism responsible for the improved photocatalytic degradation efficiency as compared to pure MFO or GO layers. In addition, the formation of intra-gap states due to the presence of oxygen vacancies allows for the extension of the absorption range to the visible range of the solar spectrum.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4352/10/4/271/s1. Figure S1: Additional SEM micrographs of BFO/GO nanocomposites. Figure S2: Additional SEM micrographs of LFO/GO nanocomposites. Figure S3: HRTEM of LFO/GO nanocomposites. Figure S4: XPS spectra of BFO nanofibers. Figure S5: High-resolution XPS spectra of Bi 4f and La 3d lines. Figure S6: High-resolution XPS spectra of Fe 2p lines. Figure S7: High-resolution XPS spectra of O 1s lines. Figure S8: High-resolution XPS spectra of C 1s lines. Figure S9: Absorption spectra of methyl orange references. Figure S10: Absorption spectra of methyl orange with a GO sample. Figure S11: Absorption spectra of methyl orange under UV light for BFO/GO nanocomposites. Figure S12: Absorption spectra of methyl orange under UV light for LFO/GO nanocomposites.