Mn3O4/graphene nanocomposites: outstanding performances as highly efficient photocatalysts and microwave absorbers

Mn3O4 (M) incorporated graphenes (G) synthesized by a deposition–solvothermal process, formed at various nominal weight percentages (G1M1, G3M1 and G1M3), were efficiently used for the photodegradation of methylene blue dye (MB) under visible light illumination (l > 420 nm, 88 W, 20 ppm, 298 K) and under microwave irradiation (800 W, 2.45 GHz, 373 K). These materials were characterized using XRD, TEM-SAED, UV-Vis diffuse reflectance, N2 sorptiometry, FTIR and Raman techniques. Amongst the nanocomposites, G3M1 of polyhedral structure and an average domain equal to 10–12 nm has presented unique photo-degradation performance (100% degradation, 60 min, 0.0791 min 1 and TOC of 60%) exceeding the rest of the materials. This was mainly due to the extraordinary optical properties and to the strong interaction between Mn3O4 and graphene through which charge recombination is hampered. Based on the conduction and valence band edges together with the studied reactive species, it has been shown that cOH was the dominant species responsible for the MB degradation. Interestingly, the G3M1 nanocomposite has shown fascinating microwave absorption properties and is capable of degrading MB at a faster rate (0.287 min ) than the one conducted via photocatalysis. Scavenger studies have shown that cOH and electrons were responsible for the excellent performance of the MB microwave degradation. The microwave results were discussed in view of the marked increase in dielectric constant (3 ) and dielectric loss (300) in the studied frequency range of 1.0 Hz to 100 kHz, in addition to the electronic conductivity measurements. This work offers an exceptional approach for exploring high-performance microwave absorption as well as distinctive visible light photocatalytic reaction for organics degradation.


Introduction
The advances in industrial processes generate enormous amounts of highly toxic organics that pollute the aquatic environment. Photocatalysis as one of the promising green technologies, 1,2 has attracted the attention of many researchers due to its ability to decompose organic pollutants into CO 2 and water. So far, TiO 2 is a widely employed photocatalyst due to its superiority in oxidation power, nontoxicity, durability and stability. 3 However, TiO 2 limitation due to its restricted activation in near UV region as well as the high recombination of its electron-hole pair hinders its photocatalytic applications under visible irradiation. 4 Thus, great effort has been dedicated to develop unique photocatalysts, which could exhibit higher activity under visible light illumination. 5 Among them, Mnbased catalysts especially Mn 3 O 4 is known to be an effective catalyst for de-NOx reactions, raw materials for electronics and information devices. 6 Nanosized Mn 3 O 4 was also reported as a high-capacity anode material for rechargeable lithium batteries. 7 In spite of the high catalytic performance of Mn 3 O 4 nanoparticles (NPs), poor chemical and thermal stabilities of the material have stimulated the aggregation of NPs and thus, lessening their catalytic efficiencies. 8 In addition, increasing the band-gap energy of Mn 3 O 4 was generally uncontrollable and used to change as a function of the structure and/or the support. To get over these obstacles, carbonaceous materials especially graphene was exploited to support Mn 3 O 4 due to its high surface area and high electrical conductivity value. [9][10][11] This hybrid between graphene and Mn 3 O 4 has generated remarkable interest because of their synergistic effects in activating the electrocatalytic reduction of oxygen, 12 catalytic decomposition of organics, and as storage of charges in supercapacitors. 13,14 However, some limitations affected the performances of this hybrid most importantly, rapid and reversible transformation of the reduced phase (Mn 3 O 4 ) with the other oxidized forms (Mn 2 O 3 , MnO 2 ) 15 and the morphological deciencies based on preparing various structures of Mn 3 O 4 with high particles domain. 16 As a consequence, the hybrid Mn 3 O 4 /graphene never shows a photocatalytic oxidation to a dye under visible light irradiation as well as without oxidants such as H 2 O 2 and peroxymonosulfate (PMS). 9,14 Although these oxidants are environmentally friendly, they have some drawbacks such as costeffective, transport, storage, and pH adjusting requirements. 17 In view of the mentioned deciencies, we synthesized magnificent photocatalysts composed of Mn 3 O 4 /graphene nanocomposites via a facile non-template deposition-solvothermal route for the purpose of increasing the contacts between the 2D graphene nano-sheets and Mn 3 O 4 moieties. This indeed is expected to facilitate the electron transfer from the latter to that of the former; hindering e À -h + recombination, and thus can be employed in methylene blue oxidation in the absence of any oxidants under visible light illumination.
It has been shown that microwave (MW) absorption for some materials can play signicant role in wastewater treatment and thus can save time and energy. During the previous decade, MW absorbing materials known as dielectrics have been the subject of great interest particularly for their marked roles in wastewater treatment. Materials, such as activated carbon, 18,19 CNTs 20,21 and polymers 22 are frequently employed in microwaveassisted degradation of organic pollutants. Among MW absorbents, MFe 2 O 4 are the most capable 23 because of the chemical properties of M 2+ and Fe 3+ and to the high magnetic permeability or electrical resistivity. 24 Accordingly, MnFe 2 O 4 -SiC has shown high MW absorption that assisted the degradation of RBR X-3B with an efficiency equal 92%. However, this catalyst suffers low stability. 25 Therefore, looking for other materials of higher MW absorbers and higher degradation performances is an essential task. Manganese oxides [MnO 6 ] have shown good degradation performances for methylene blue under microwave irradiation based on the signicant difference in the oxidative removal ability between akhtenskite and birnessite phases. 26 However, the hybrid between graphene and Mn 3 O 4 ; the most common dielectric and microwave attenuation material, has rarely been reported. Herein, we demonstrate the interfacial coupling between Mn 3 O 4 and graphene for the purpose of improving the dielectric properties of the nanocomposites and for microwave-enhanced oxidation of the MB dye with high reaction rates, short reaction times and great energy efficiency.
These ndings may open up an efficient improvement in the MW absorbers for rapid degradation of organic pollutants.

Reduction of GO
Graphene oxide that was synthesized based on the modied Hummer's method was rst dispersed in 30 mL distilled water and sonicated for 30 min. 27 This suspension was then heated to 100 C followed by the addition of 3 mL hydrazine hydrate. The suspension was then kept at the latter temperature for 24 h. Consequently, the reduced graphene was collected by ltration in the form of black powders. The obtained material was thoroughly washed using distilled water for several times followed by sonication to remove the excessive hydrazine amounts. The nal product was collected by vacuum ltration and dried at 80 C.  (25-28 wt%) was added into the previous solution to indicate changing of the colourless solution into to a light red without precipitation. An air purge at a rate equal 1.5 L min À1 was blown into the solution that heated up into 50 C for 5 min. During this process, the solution colour changes gradually into a deep black with a consequent decomposition to form a colloidal solution of high dispersion and stability. Centrifugation and drying at 50 C was then performed to produce the nano-sized Mn 3 O 4 crystals.

Synthesis of RGO/Mn 3 O 4
The RGO/Mn 3 O 4 composites were synthesized by solvothermal method via dispersing 20 mg RGO in 200 mL DMF followed by sonication for 30 min and heating to 80 C. Addition of 20 mL of 0.2 M Mn(Ac) 2 $4H 2 O was attained while stirring that extends about 1 h to prepare different composite loadings. The as-made composites were treated as above via using both the ammonia solution and purging air followed by the heating step carried out at 50 C for 5 min. These nanocomposites contained in 200 mL solution were autoclaved at 180 C for 10 h. Then, the products were collected by centrifugation using ethanol, followed by drying at 60 C for 48 h. The as-synthesized catalysts were denoted as G1M1, G3M1 and G1M3 where the numbers are denoted to the weight ratios.

Materials characterization
The X-ray powder diffraction patterns of various nanocomposites were carried out using a Philips 321/00 instrument and were run with Ni-ltered Cu Ka radiation (l ¼ 1.541Å) at 36 kV. The surface properties specically BET surface area, total pore volume (V p ) and mean pore radius (r) were determined from N 2 adsorption isotherms measured at À196 C using conventional volumetric apparatus. The pore size distribution was determined from desorption branch of the isotherm using the BJH analysis. Diffuse Reectance Ultraviolet-Visible Spectroscopy (UV-Vis DRS) of nanocomposites together with the edge energy (E g ) for allowed transitions were carried out at room temperature using a Perkin Elmer Lamda-900 spectrophotometer in the range of 200-800 nm. The Fourier transform infrared (FT-IR) spectra were recorded via a double beam Perkin Elmer Spectrometer with a resolution of 2 cm À1 using the KBr method. Raman spectra were measured with a U-1000 laser Raman spectrometer using the 514.5 nm line of an ArC laser as the excitation beam. Selected area electron diffraction (SAED) images and TEM micrographs were measured using a FEI; model Tecnai G20, super twin at an accelerating voltage of 200 kV.

Photocatalytic degradation experiments
A high pressure Hg lamp of 88 W with a special UV cut off lter (l > 420 nm) offering visible light source; with an average light intensity equal 40 mW cm À2 , was placed at a specied position using a special rod in the reactor. The photocatalyst (100 mg) was suspended in 100 mL aqueous solution of 20 mg L À1 methylene blue (MB). The solution was stirred in the dark for 60 min to ensure the establishment of an adsorption-desorption equilibrium. During irradiations, 2 mL aliquots were removed at denite time intervals and analyzed with a Shimadzu UV-2350 spectrophotometer to measure MB concentrations. The degradation percentages were calculated based on the following equation: removal (%) ¼ C 0 À C t /C 0 Â 100; where C 0 and C t correspond to the initial concentration and that at time t, respectively.
For exploring the reactive species might produced in the photocatalytic reaction, 28 we used different scavengers including isopropanol (a quencher of cOH), p-benzoquinone (a quencher of cO 2 À ), triethanol-amine TEOA (a quencher of h + ), and carbon tetrachloride (a quencher of e À ) at a concentration of 1.0 mM.

Microwave degradation of MB
A quartz vessel containing 100 mL of MB solution (20 ppm) and 0.1 g catalysts was inserted into a Milestone Start D Microwave Digestion System operating at 800 W and 2.45 GHz. The samples are heated from room temperature to 100 C in 5 min. This temperature was maintained while performing the reaction and then the samples were le to cool down to room temperature. The samples were then forwarded to the UV/Vis Spectrophotometer to analyze the residual MB in the wavelength region of 200-700 nm. The degradation percentages were calculated as above.

Electrical properties
The electrical properties of the prepared composites were demonstrated via compressing the powder of the sample under a pressure of 5 tons cm À2 to build up pellets. The two equivalent surfaces of the pellets (7 mm diameter and 1 mm thickness) were coated with silver paste to ensure good electrical contact. The electrical measurements were carried out at a constant voltage (1 volt) in a frequency range from 1.0 kHz to 300 kHz and at the temperature of 25 C; and if necessary raised to 100 C, using a programmable automatic LCR bridge (HIOKI: 3532-50). The dc electrical resistivity was measured with an electrical circuit consists of an electrometer (model 6517, Keithley), voltammeter (Keithley, 2182) and 5 kV dc power supply. The dcconductivity s dc of the material was calculated by the following equation s dc ¼ (l/A s )(1/R dc ) where R dc is the sample resistance, l is the length of the sample and A s is the crosssectional area. The complex dielectric permittivity was investigated through LCR meter (HIOKI: 3532-50) using the relation 3*(u) ¼ 3 0 (u) À j3 00 (u); where 3 0 is the real part of the permittivity and 3 00 is the imaginary part of the permittivity with j ¼ OÀ1. The dielectric constant of samples was measured using the equivalent capacitance (C). The values for 3 0 and 3 00 were estimated using the relations 3 0 ¼ (C/3 o )(d/A s ) and 3 00 (u) ¼ 3 0 (u) tan d; where C is the capacitance of the sample, 3 o is the permittivity of the vacuum, d is the thickness, A s is the cross-sectional area of the specimen, tan d is the dissipative factor where d-phase angle is determined via u ¼ 2pf, where f is the frequency of the applied electric eld. The surface morphology of the bare Mn 3 O 4 and the nanocomposites G1M1 and G3M1 was investigated using HR-TEM and SAED analyses (Fig. 2). The image formation of the bare Mn 3 O 4 shows spherical-like particles with an average size of 12 nm ( Fig. 2A). The selected-area electron-diffraction pattern (SAED) of the nanoparticles is in harmony with the Hausmannite Mn 3 O 4 phase and exposes strong ring patterns due to (101), (103), (211) and (220) planes. The spherical like particles were well distributed with a little tendency of forming clusters probably due to increasing the surface free energy of the nanoparticles. The TEM image shown in Fig. 2B for G1M1 reveals the existence of two types of morphological crystals for Mn 3 O 4 including square and spherical architectures decorating graphene nanosheets. The square shape indicates an average dimension of 44 nm where the spherical ones reveal a dimension equal 10 nm. The inset gure conrmed the formation of the Hausmannite Mn 3 O 4 phase via the selected area electron diffraction (SAED) pattern and thus exposed visible rings ascribed to 101, 103 and 211 lines. Another diffraction spot in the latter SAED pattern is also recognized and correlated to the 002 facet of graphene, giving a hint about the intimate contact between Mn 3 O 4 and graphene nanosheets. The TEM image of the G3M1 sample (Fig. 2C) shows polyhedral structures including spherical, regular tetrahedron and rectangular, as illustrated in the high resolution TEM image shown as inset in Fig. 2C. They have an average diameter in the 10-12 nm range. The selected area electron diffraction pattern indicates visible rings that expose lattice fringes 101, 103, 211, 220 and 002 extended from exterior to interior and indexed respectively to Hausmannite Mn 3 O 4 phase and graphene (002). This pattern indicates the well dispersion of Mn 3 O 4 nanoarchitectures and its overlaying on graphene and rather emphasizes the strong interaction between them. That high dispersion of Mn 3 O 4 within the great percentage of graphene sheets (G3M1) induces strong interaction between them and thus governs the particles size enlargement. A criterion about that interaction and consequences thereof concerning the agglomeration prevention of Mn 3 O 4 nanoparticles as a result of its distribution between graphene nanosheets, is indeed comes through varying the morphology of the parent Mn 3 O 4 and when it interacts with graphenes. That particle growth inhibition preserves the highsurface-area interface between Mn 3 O 4 and graphenes. The images of G3M1 ( Fig. 2D) indicated SAED pattern typical to all nanocomposites and rather shows spherical and square nanoarchitectures of dimension 12 nm and 50 nm, respectively.

FTIR and UV-Vis spectroscopy
The FTIR spectra of the synthesized samples are presented in Fig. 3  graphene structure. The nanocomposites containing graphene that showed absorption enhancement in the whole region were in the order; G3M1 > G1M1 > G1M3. This indeed maximized the effect of graphene appropriate percentages relative to Mn 3 O 4 along with the color changes from black to grey. This was in concordance with the reported results of other GR-including nanocomposites. 41 The electronic absorption spectra of the nanocomposites also show broad bands from 390 to 530 nm in G1M3 where it extends into 600 nm in G1M1 and G3M1. This indeed is correlated to d-d crystal eld transitions on octahedral Mn 3+ species 42 and rather proposes the strong interaction between the moieties forming the composite. This is expected to increase the harvesting capability of the nanocomposites to have a great impact on their photocatalytic performances probably in the same sequence under visible illumination. The band gap energy of the samples is calculated via using the equation: where n, a, E g and A is respectively light frequency, absorption coefficient, band-gap energy and a constant. 43 The part n is a constant ts with 1/2 to comprehend that the optical transition is of a direct allowed type. 44 Accordingly, the estimated band-gap energies of G, Mn 3 O 4 , G1M1, G3M1 and G1M3 based on the latter equation are 1.39 eV, 3.1 eV, 1.44 eV, 1.41 eV and 2.05 eV, respectively (Fig. 4 inset). All the GM nanocomposite samples have band gap energies in the region between those of graphene and Mn 3 O 4 accomplishing the strong interaction devoted between them.

Raman spectroscopy
To gain more information about the structure of the graphene sheets and its interaction with Mn 3 O 4 , Raman spectra were recorded (Fig. 5). The Raman spectrum of the bare Mn 3 O 4 shows a strong peak at 641 cm À1 ascribed to the A 1g mode correlated to the oxygen vibrations in the MnO 6 unit. 45 The Raman spectrum of G1M1 displays the G peak at 1570 cm À1 correlated to the ordered sp 2 bonded carbon together with the D peak at 1347 cm À1 due to edges or disordered layers. 46 The small peaks in the margin 200-400 cm À1 ascribed to Mn 3 O 4 modes 47 were intensied in the G1M1 spectrum compared to the pristine Mn 3 O 4 to comprehend the proposed weak interaction between the moieties forming G1M1. Conversely, the G3M1 spectrum illustrates in addition to D and G peaks a very small peak due to This journal is © The Royal Society of Chemistry 2017 the free Mn 3 O 4 (641 cm À1 ) that constitutes 1/5 of its intensity in G1M1 and 1/4 to that in the bare Mn 3 O 4 . Taking into consideration the Mn to G ratios, one easily conrms that the interaction between Mn 3 O 4 with graphene was more intense in G3M1 compared to the other nanocomposites. Shiing the G band from 1570 cm À1 in G1M1 and G1M3 to 1580 cm À1 in G3M1 conrms the devoted strong interaction of Mn 3 O 4 with graphene and rather announces lowering of graphene layer thickness in the latter sample than those in the former. 47,48 Both of the nanocomposites show the 2D band; characteristics of two phonon lattice vibration process, 48 of the same shape and position extending from 2550 to 2850 cm À1 . Accordingly, the synthesized graphene is multilayered given that D in RG is very strong than the 2D. The intensity ratio of D and G peaks (I D /I G ) offers a precise measure of the disorder and crystallite size of the graphitic layers.

Photocatalytic degradation of the MB dye
The photocatalytic degradation of as-synthesized nanocomposites RG-Mn 3 O 4 (G3M1, G1M1 and G1M3) was studied via using the methylene blue (MB) dye at a concentration of 20 ppm and at room temperature. These samples were initially le for 60 min in MB solutions before irradiation for ascertaining equilibrium adsorption. Upon visible light irradiation (88 W, l > 420 nm), the G1M3 sample did not exhibit any activity apart from that related to the dye absorption and comprised of 30% (Fig. 6). The photocatalyst G1M1 exhibits MB degradation equal 92%, showing fast oxidative decomposition upon increasing the RG ratio. It seems also that the RG ratio affected the adsorption process to reach 52% in this latter sample. A further increase in the RG ratio as in the G3M1 sample has indicated an efficient MB degradation via achieving 100% in    Fig. 6) with slopes indicative to the reaction rate constants. These lines indicate that the oxidation reaction follows pseudo-rst order rate kinetics. The rate constant values that performed at 298 K were in the order: G3M1 (0.0791 min À1 ) > G1M1 (0.0428 min À1 ) > Mn 3 O 4 (0.0114 min À1 ) > G1M3 (0.0005 min À1 ). Since, the nanocomposite G3M1 presented the most promising photoactivity result, it has been used for performing further reactions. At certain reaction intervals, the UV-Vis absorption spectra of the MB dye were undertaken while tracing its photocatalytic degradation using the nanocomposite G3M1. As shown in Fig. S2, † two absorption peaks at 615 and 664 nm are observed characterizing the MB typical peaks. 50 These absorption peaks diminish with time and the solution turns colourless gradually within 60 min irradiation time. The TOC% of the same sample taken during the 60 min reaction time; inset in Fig. S2, † indicates 60% elemental carbon representing the photodegradation ratio of the dye organic carbons. However, the difference between C/C o and TOC% values; typical of 40%, are mostly correlated to the presence of non-degradable intermediates produced during the photo-degradation process. However, extending the time into 120 min has accomplished the 100% TOC degradation verifying the complete transformation of organic carbons into elemental carbons.
To shed an idea about the reactive species could be involved in the oxidation process, some recognized scavengers were added to investigate their role on the reaction rate. Accordingly, the effects of the addition of benzoquinone; BQ, isopropanol; IPA, triethanolamine; TEOA, and carbon tetrachloride; CCl 4 , on the catalytic oxidation of the MB dye over the photocatalyst G3M1 have been investigated under identical experimental conditions (Fig. 7). It has been shown that the reaction rate was relatively retained upon using BQ reecting the small effect of the cO 2 À species. The experiment with IPA has indicated a higher decrease than that evoked via BQ, proposing the inuential effect of cOH.
Based on the presented results, the photocatalytic mechanism can be proposed as follows (Scheme 1). In the Mn 3 O 4 /RG composites, Mn 3 O 4 attached on the surface of RG produces electron-hole pairs as a result of photons absorption when exposed to visible light irradiation. Effective separation of the charges is attained given that RG acts as an effective scavenger for the produced electrons. Herein, electrons absorbed by RG can react with the adsorbed O 2 to yield cO 2 À that by its turn transferred into cOH together with the OH À ions dedicated as well to produce cOH. Also, cOH can be obtained via direct water oxidation or reduction of H 2 O 2 . Investigation that the latter moieties is in situ formed is traced by measuring the light absorption of the titanic-hydrogen peroxide compound. 52 Accordingly, cOH moieties compile their denite effects in effective MB dye oxidization. Nevertheless, one must bear in mind that our RG is functionalized by residual oxygen moieties on its surface; in conformity with E g data ascertaining its semiconducting property, enabling it to take part in the photoreaction products. Accordingly, the photo-generated holes could be capture by not only the lattice oxygen in Mn 3 O 4 but also those on RG sheets to form cOH active species. On the other hand, the photogenerated h + in the VB is also amenable for generating cOH via the reaction with H 2 O. In addition, it is very important to conrm that the photodegradation obtained for the nanocomposites is consistent with the values estimated from incident photon to current efficiency (IPCE) spectra (Fig. 8). This is because photodegradation is entirely dependent on the incident light and the consequences thereof concerning the impinged photons. As shown in Fig. 8, the G3M1 electrode possessed the highest IPCE values over the wide range from 350 to 750 nm compared to rest of the samples. This result was in harmony with the devoted UV-Vis absorption spectra (Fig. 4) in which G3M1 has indicated the highest absorption throughout all the studied range (200-800 nm) and rather it indicates the lowest E g value (1.41 eV). Increasing IPCE of the G3M1 electrode in the wavelength range of 600-700 nm could also be due to increasing the absorption capacity of the MB dye (Fig. 8). Increasing the IPCE of same sample in the narrow wavelength range of 470-490 nm; in which the MB dye has a low absorption for incident light, reects that this promotion is directly correlated to the phenomenon of light scattering. This result further conrmed the superior light absorption and scattering properties of the G3M1 electrode in the long wavelength region. This explains the higher efficiency of G3M1 in photodegrading the MB dye compared to rest of the catalysts. Exposing the tetrahedral structure of Mn 3 O 4 in G3M1 than the octahedral one; as deduced from FTIR results, may substantiate the facile mobility of Mn 3 O 4 in enhancing the photocatalysis of this specic sample together with its decreased particles size.
It's of important notifying that at the 25 wt% of RGO loading (as in G1M3), no photoactivity is exhibited where aer increasing the weight ratio of RGO compared to Mn a superior photoactivity is attained. This is due to at the former ratio, an oxidized GO is obtained due to the excessive amounts of Mn 3 O 4 (75 wt%) and this prohibits the facile electron transfer due to its scavenger throughout oxygen functional groups leading to high loss of charges. Conversely, increasing the ratio into 75% graphene (as in G3M1) reduces tremendously the residual oxygen, as reveled from IR results, and thus acts as an electron acceptor/donor capable of performing magnicent role in the dye degradation. Convincingly, the strong adhering of Mn 3 O 4 into RG; which retained the particles size of the former at $12 nm, has shown exceptional optical and texturing properties those in turn caused high oxidative decomposition for the MB dye in the absence any oxidants unlike many comparable nanocomposites. 9,14,25 The stability and recyclability of the G3M1 nanocomposite was evaluated by successive tests for the MB decolorization ( Fig. 9). It is established that the G3M1 photocatalyst can work at least four cycles without noticeable loss in the catalytic activity, demonstrating the long-term durability of this nanocomposite. Whereas, the h run indicates retaining of the 85% of the catalyst activity proposing an excellent stability while reusing for 300 min. Fig. 10 shows the normalized degradation curves of MB treated in the microwave reactor; for 30 min at 373 K and under 1 atm pressure, with the nanocomposite catalysts as well as with the free Mn 3 O 4 catalyst. During the microwave treatment, the nanocomposite G3M1 catalyst indicates the best MB degradation performance by signifying 100% color removal. This only takes 10 min compared to 60 min in the photocatalytic reaction performed for the same sample. Accordingly, the consequence of the studied nanocomposites was in the order; G3M1 > G > G1M1 > Mn > G1M3. These results maximized the role of the synergistic effects between Mn 3 O 4 and graphene at high percentages of the latter (G3M1) since, it shows higher degradation percentages than pristine Mn 3 O 4 and graphene nanoarchitectures. It was found that the G1M1 catalyst exhibits a microwave catalytic activity comprised of 70% degradation whereas that of G1M3 indicates 40%, both in 30 min irradiation time. The microwave degradation of MB by a direct microwave irradiation in the absence of a catalyst was almost zero. Additionally, to exclude that the degradation has nothing to do with the sample heating, the nanocomposites were heated to 373 K with the MB solution for 1 h away from microwave irradiations to comprehend zero degradation. Whereas, the microwave degradation of MB in presence of nanocomposites was found to follow pseudo-rst-order decay kinetics. 53 Based on the Langmuir-Hinshelwood model, the estimated reaction rate (Fig. 10 inset) was in the order G3M1 (0.287 min À1 ) > G (0.085 min À1 ) > G1M1 (0.045 min À1 ) > Mn (0.026 min À1 ) > G1M3 (0.016 min À1 ) (Fig. 11). It's interesting notifying that the activity of microwave irradiation of G3M1 while degrading MB exceeded that obtained via photocatalysis by 3.5 times.

Microwave absorption of Mn 3 O 4 /graphene for MB degradation
The scavenger studies for detecting the reactive species could be responsible for MB degradation while employing the microwave absorption property of nanocomposites were performed (Fig. 11). Notably, when TEOA and BQ used, the degradation efficiency is overall similar to that of the blank without scavengers and thus nullies the inuence of holes and cO 2 À species. Whereas, upon using IPA and CCl 4 a signicant decrease in activity is perceived relative to that of the blank implying that cOH and electrons are the major reactive species for G3M1. This advocates that the coupling effect between microwave and the active cOH species on Mn 3 O 4 incorporated Fig. 9 The recycle test during degradation of MB using G3M1 under visible light illumination.  graphenes is participated in the MB degradation. It has been notied that cOH is produced considerably by h + in the VB of Mn 3 O 4 /graphene (G3M1) owing to its more negative potential than the standard redox potential of OH/cOH (E Q ¼ 2.4 eV). 53 The violent motion of polar materials; exerted due to collisions between reactant molecules due to microwave irradiation, can compete with the catalytic reactive species to accelerate degradation of the MB dye.
To congure the role of electrons during the microwave MB degradation performance; as committed from the trapping experiments, the electrical conductivity of as-synthesized nanocomposites as well as the bare Mn 3 O 4 was measured at temperature range from room temperature to 373 K (Fig. 12). The electrical conductivity values were found to decrease from 0.033 U À1 cm À1 , 0.028 U À1 cm À1 , 0.002 U À1 cm À1 and 6.0 Â 10 À9 U À1 cm À1 for G3M1, G1M1, G1M3, and Mn 3 O 4 , respectively. This indicates that the combination of graphene with Mn 3 O 4 nanocomposites can strongly increase the electronic conductivity compared to the bare Mn 3 O 4 . It also indicates the potentiality of graphene specically in the nanocomposite G3M1, reecting the synergism between the components forming this composite. Based on the high absorptivity of carbon containing materials including graphenes to microwave, hot-spots are formed to act as oxidation centres for the pollutants oxidation. 53 That selective heating stimulates molecular rotation causing a decrease in the activation energy (see Fig. 10).
Since microwave assists the rise of temperature of the samples thus, an increase in the number of carriers and lattice vibrations is expected to be obtained. 52 Indeed, increasing the temperature increases the number of carriers and consequently it leads to conductivity elevation (inset in Fig. 12). Scheme 1 also illustrates the effect of cOH and electrons evoked while microwave irradiation on the MB degradation. Of particular interest, although the nanocomposites G1M1 and G1M3; of alike thickness, show the presence of small amounts of MnO 2 and Mn 2 O 3 species; as devoted from XRD and IR results, they still indicate lower conductivity values than G3M1. This declares that the geometric defects obtained in G1M1 and G1M3 samples such as dislocations and grain boundaries 54 are great; as exactly devoted from surface properties and TEM results. Consequently, such grain boundaries in graphene can disrupt the electron transfer and thus being trapped causing the depletion in charge carriers. The strong interaction devoted between Mn 3 O 4 and graphene in G3M1; conrmed via IR, Raman, TEM-SAED and surface texturing data, minimizes the presence of discrete boundaries, structural disorders and surface imperfections between the moieties forming this sample. On the other hand, the lattice vibrations create phonons and thus an expected interaction between electron-phonon can be perceived. 54 Undoubtedly, G1M1 and G1M3 must have shown more charge carriers than G3M1 due to evoking Mn 4+ and Mn 3+ species on their surfaces; as depicted via XRD and IR results. However, the devoted decrease in their conductivities explains the possibility of hampering the electrons by phonons and as a result affecting the oscillation mode occurs in their lattice structures inuencing their conductivities (Fig. 12).
Increasing the microwave catalytic activity of G3M1 was due to enhancement of the microwave absorption as a result of the change occurring in dielectric constant and dielectric loss caused by the addition of Mn 3 O 4 to graphene. Under the alternating electromagnetic eld, interfacial and space-charge polarization can easily be formed at the interface of Mn 3 O 4 / graphene and also due to the conducting nature of graphene and the dielectric nature of Mn 3 O 4 . Fig. 13a shows that the nanocomposites dielectric constant (3 À ); inferred to the storage ability of electric current, decrease continuously with increasing frequencies to be in the order; G3M1 > G1M1 > G1M3 > Mn. The monotonous decline in dielectric constants on increasing frequencies may be attributed to the combined contribution caused by electronic, ionic, and interfacial polarization. 58 The observed dielectric distribution at low frequencies can be explained on the basis of the Maxwell-Wagner theory of interfacial polarization 59 by which the dielectric structure of the composite consists of two layers. The number one layer stands for a large number of grains that act as conducting layer at lower frequencies and the other layer consists of grain boundaries that act as highly resistive medium at higher frequencies. At low frequencies, the polarization process in composites of high percentages of Mn 3 O 4 (G1M3) can be described as a local displacement of electrons via hopping mechanism between Mn 3+ and Mn 4+ and an orientation of electric dipole in the direction of the applied eld. Nevertheless, the monitored decrease in the latter sample dielectric constant is in part due to the presence of residual oxygenated functional groups; depicted via IR results. That presumably lessens the direct contact between graphene nanosheets and thus facilitates the electron transfer through oxygen functional groups leading to high dielectric loss (Fig. 13b). Increasing the microwave absorption property of G3M1 could also be caused by dielectric relaxation and interfacial scattering. 55,56 The dielectric relaxation and polarization was mainly induced by interfacial multipoles, which took place along the boundaries between graphene sheets and Mn 3 O 4 moieties. 57,58 It has been documented that charges would transfer throughout the interface in a metal graphene heteroarchitecture due to their different work function and thus an electron transfer from Mn 3 O 4 to graphene can be anticipated. Accordingly, a charge transfer process can rationally be assumed at the interface of G3M1, leading to the introduction of the free carriers into graphene sheets. The latter carriers would vibrate with the microwave motivation giving rise to electric polarization in graphene and thus increase the value of 3 0 in G3M1 compared to rest of the composites (Fig. 13a). Furthermore, 3 0 of all nanocomposites decreases with increasing frequency (Fig. 13a 1 ) owing to the relaxation of polarization and thus shows a dielectric dispersion. The smallsize particles (12 nm) observed in G3M1 involves large number of particles per unit volume resulting in an increase of the dipole moment per unit volume and thus induces the highest dielectric constant. In the meantime, the motion of the introduced free carriers would attenuate the microwave energy, resulting in the enhancement of the dielectric loss 3 00 (Fig. 13b). Attenuation of dielectric loss with increasing frequencies was also attained; as seen in the corresponding Fig. 13b 1 . Diminishing the dielectric constant with increasing frequency is dedicated to the polarization decrease of the dipoles when the electric eld propagates with high frequencies. Interestingly, the G3M1 sample has indicated the lowest decrease in dielectric loss with increasing frequency between all samples. This latter behaviour is explained on the basis that in dielectric nanostructured materials, interfaces are produced with large volume fractions involving great number of defects such as vacancies, dangling bonds and microporosities. 58 These can cause a change in negative and positive space charge distribution at the interfaces. Such space charges when subjected to an electric eld, they trapped by defects to form lots of dipole moments. On the other hand, these dipole moments at low frequencies follow the change of the electric eld 59 and thus both the dielectric loss and the dielectric constant display high values.
The dielectric properties of the composites are usually correlated to the surface morphology of reduced graphene nanosheets, since it increases as the graphene ratio increase ( Fig. 13a and b). Accordingly, the random distribution of the conserved sp 2 and distorted sp 3 carbons decorated by various oxygen containing functional groups can be anticipated. 60 Such distribution in RGO works as a nanocapacitor electrode separated by the Mn 3 O 4 dielectric material. Thus, the revealed considerable improvement in dielectric constant can be attributed to the formation of huge number of nanocapacitors inside the composite. 61 This is believed to originate due to the presence of residual oxygenated functional groups that apparently lessen the direct contact between graphene nanosheets. 62 Consequently, the large capacitance caused by each nanocapacitor results in a noticeably high local electric eld. An environment is aerwards created which is conducive for migration and accumulation of charge carriers at the RGO/ Mn 3 O 4 interface. That interfacial polarization named Maxwell-Wagnar effect is considered to be responsible for the large dielectric constant at low frequency region. 63 Conversely, the escape current is principally responsible for the high dielectric loss in nanocomposites, as validated in G1M1 and G1M3. This explains that oxygenated functional groups on RGO are not as an insulating layer or intrinsic barriers that limits the current leakage 64 but facilitates the electron transfer through oxygen functional group leading to high dielectric loss.

Conclusions
The polyhedral nanoarchitecture graphene/Mn 3 O 4 formed at the ratio of G3M1 has shown fascinating photocatalytic performances; under visible light illumination without any oxidants, and microwave irradiation for the purpose of the MB degradation. The experimental results of this type of smart material evokes some important factors by which the enhancement was revealed such as (i) high optical properties explained via exceeding the visible light absorption of the sample in the 400-800 nm region as well as increasing the IPCE% value (ii) the devoted strong interaction between the moieties forming G3M1 and the consequence thereof concerning the decrease in E g as well as the delay in the charges recombination (iii) increasing the conductivity of this sample at room temperature (i.e. indicative to photocatalysis) and rather its enhancement at 373 K (i.e. indicative to microwave irradiation) (iv) the superior increase in the dielectric constant at the low frequency margin, at which microwave absorption takes place, explained the extraordinary photocatalytic MB degradation. The excellent microwave absorption seen in G3M1 was attributed to the charge transfer at the latter nanocomposite interface as well as the interfacial polarization that well explained in view of the Maxwell-Wagnar effect.