Synthesis of novel g-C$_3$N$_4$/KBiFe$_2$O$_5$ composite with enhanced photocatalytic efficiency

(1)

Synthesis of novel g-C

₃

N

₄

/KBiFe

₂

O

₅

composite with enhanced photocatalytic efficiency

RANJAN RAI and MURALIKRISHNA MOLLI*

Department of Physics, Sri Sathya Sai Institute of Higher Learning, Prasanthinilayam, Anantapur 515134, India

*Author for correspondence (muralikrishnamolli@sssihl.edu.in) MS received 5 February 2020; accepted 27 March 2020

Abstract. Novel graphitic carbon nitride (g-C3N4)/potassium bismuth ferrite (KBiFe2O5) composite photocatalysts were synthesized using facile grinding and calcination method. X-ray diffraction, transmission electron microscopy, scanning electron microscopy and UV–visible diffuse reflectance spectroscopy analysis were carried out to investigate structural, morphological and optical properties of the prepared samples. The photocatalytic properties of the samples were studied by photocatalytic degradation of methylene blue under visible light irradiation. The composite showed enhanced photocatalytic efficiency compared to both pristine g-C₃N₄and KBiFe₂O₅. Out of four composites with different weight percentages of KBiFe2O5, one with 30 wt% showed the maximum photocatalytic efficiency. Samples with a higher content of KBiFe2O5showed decreased photocatalytic efficiency indicating 30 wt% as the optimum composition. The increase in the photocatalytic efficiency is mainly due to efficient charge separation of photo-generated electron–hole pairs in the composites. The possible mechanism for the photo-catalysis of g-C3N4/KBiFe2O5composites was also proposed.

Keywords. Photo-catalysts; composites; brownmillerite; graphitic carbon nitride; KBiFe₂O₅.

1. Introduction

Photocatalysis is one of the important advanced oxidation processes for pollutant degradation [1]. It also finds appli- cation in photocatalytic water splitting [2,3], hydrogen production, CO₂photoreduction, etc. [4,5]. TiO₂, ZnO and SnO₂ are a few semiconductors which are proven as efficient photocatalysts for the above applications owing to their efficient absorption capability in ultraviolet region of solar spectrum, high chemical stability, low cost and non- toxicity [6]. But due to their wide band gap they can absorb only in the ultraviolet region which forms only 5% of the solar spectrum. So, visible-light-driven photocatalysis is essential for practical applications [7].

Graphitic carbon nitride (g-C₃N₄) is one of the efficient metal-free photocatalysts [8]. It is highly stable even in more alkaline and acidic medium, has narrow band gap of 2.7 eV and an appealing electronic structure [9]. But pristine g-C₃N₄ suffers from a high recombination rate of photogenerated electron–hole pairs, low specific surface area etc., which leads to overall low photocatalytic efficiency [10]. Similarly, potassium bismuth ferrite (KBiFe₂O₅), a new brownmillerite compound with A2B2O5 crystal structure [11], has narrow band gap and efficient absorption in the visible region, making it suitable for photocatalytic applications. However, the photocatalytic efficiency of the material is still low due to high recombination rate of photogenerated electron–hole pairs and low specific surface area. Numerous strategies such

as non-metal doping [12,13], nanostructure engineering etc., are implemented in g-C₃N₄to achieve optimal photocatalytic efficiency [14–16]. To cope with the problem of high recombination rate of photogenerated electron–hole pairs in g-C₃N₄, its heterostructures are formed with semiconductors [17,18]. Reports on the enhanced photocatalytic properties of semiconductor g-C₃N₄ composite photocatalysts (such as TiO₂/g-C₃N₄ [19], CdS/g-C₃N₄ [20], Fe₂O₃/g-C₃N₄ [21]

etc.) are available in the literature. The difference in the energy levels of the constituents enables efficient charge transfer in the heterostructure [22].

In this study we report the synthesis of organic–inor- ganic, g-C₃N₄–KBiFe₂O₅composites using facile grinding and calcination method. The heterostructure between g-C₃N₄ and KBiFe₂O₅ was designed so as to reduce the recombination of light-induced electron–hole pairs in both g-C₃N₄and KBiFe₂O₅thereby enhancing the photocatalytic efficiency. We have investigated the photocatalytic efficiency of the composites by photocatalytic degradation of methylene blue under visible-light irradiation.

2. Experimental

2.1 Synthesis of KBiFe₂O₅

KBiFe₂O₅ was synthesized using citrate-assisted sol–gel route. Aqueous solutions of bismuth nitrate (0.2 M),

Bull Mater Sci (2020) 43:257 ÓIndian Academy of Sciences

https://doi.org/10.1007/s12034-020-02232-1Sadhana(0123456789().,-volV)FT3](0123456789().,-volV)

(2)

ferric nitrate (0.4 M) and potassium nitrate (0.2 M) were prepared and mixed under vigorous stirring. Citric acid solution was added to the above mixture and stirred for 30 min. Ammonia solution was added to maintain a pH of 6.5–7 and stirred for 3 h. After a clear solution was formed the beaker was kept on a hot plate maintained at

130°C for evaporation of water. Gelation took place upon evaporation. Gel thus formed was heated at 220°C to cause the combustion of citrate. Brown coloured residue was left behind which was ground and calcined at 700°C for 2.5 h to get polycrystalline powder of KBiFe₂O₅.

Figure 1. (a) XRD pattern of different photocatalysts, (b) simulated XRD pattern of monoclinic crystal structure of KBiFe2O5.

Figure 2. TEM micrographs of (a) g-C3N4, (b) g-C3N4/KBiFe2O5(CN-KBF3) composite and (c) HRTEM image of KBiFe2O5showing a set of (110) lattice planes in KBiFe2O5.

257 Page 2 of 8 Bull Mater Sci (2020) 43:257

(3)

2.2 Synthesis of g-C₃N₄

g-C₃N₄was synthesized by heating melamine. About 10 g of melamine was placed in an alumina crucible with a cover to minimize the volatilization of melamine and heat treated at 500°C for 2 h. Further heat treatment was performed at 550°C to cause deammonation for 2 h. Yellow-coloured g-C₃N₄powder was thus obtained.

2.3 Synthesis of g-C₃N₄/KBiFe₂O₅composites

To prepare g-C₃N₄/KBiFe₂O₅(10 wt%) composite, 300 mg of g-C₃N₄ and 33.33 mg of KBiFe₂O₅ were mixed in 50 ml methanol under stirring for 12 h to obtain a homogeneous mixture. The mixture was kept for drying at 90°C for 12 h. The dried sample was collected and hand ground for 30 min using agate mortar and pestle. The sample thus obtained was heated at 400°C for 1 h to get the composite which was labelled as Figure 3. SEM micrographs of (a) KBiFe2O5, (b) g-C3N4, (c) CN-KBF1, (d) CN-KBF2, (e) CN-KBF3 and (f) CN-KBF4.

Figure 4. EDS chemical map of CN-KBF3 composite. Figure 5. EDS plot showing elemental analysis of CN-KBF3 composite.

Bull Mater Sci (2020) 43:257 Page 3 of 8 257

(4)

CN-KBF1. Similarly composites with 20, 30 and 40 wt%

KBiFe₂O₅were prepared following the above procedure and labelled as CN-KBF2, CN-KBF3 and CN-KBF4, respectively.

2.4 Characterization

The crystallinity and the phase purity of the sample was analysed using X-ray diffraction (PANalytical, X’Pert³ Powder, Cu-Ka radiation). Microstructural and the com- positional analysis of the sample was carried out using scanning electron microscope (JEOL JSM-IT300LV) and transmission electron microscope (JEM 2100F, JEOL).

Diffuse reflectance spectra were recorded using UV–Vis–NIR spectrophotometer (Perkin-Elmer Lambda 750 UV/Vis/NIR spectrophotometer) for the samples in the wavelength range 200–750 nm.

2.5 Photocatalytic measurements

The photocatalytic activity of the prepared samples was evaluated by degradation of methylene blue (MB) as a model pollutant. In a photocatalytic experiment, 0.05 g of photocatalyst was added to 100 ml (10 mg l^-1) aqueous solution of MB. Prior to irradiation the suspension was stirred in the dark for 0.5 h to achieve adsorption–desorption equilibria. The light source was kept axially about 7 cm from the reactor placed on top of a magnetic stirrer. The irradiation of the suspension was carried out under contin- uous stirring for 30, 60, 90, 120, 150 and 180 min. A vol- ume of 5 ml of suspension was withdrawn at regular intervals of 0.5 h upon which centrifugation was carried out for 0.5 h at 5000 rpm. Absorption spectra of the clear solution formed after centrifugation was recorded using UV–visible spectrophotometer (Shimadzu 2450). The dye concentration was determined from the absorbance as a function of the irradiation time.

3. Results and discussion

Figure1a shows the XRD pattern of different photo-catalysts. The XRD patterns show the formation of well crystallized samples. The diffraction pattern of KBiFe₂O₅ indicates that the sample crystallized into monoclinic crystal system and matched with simulated XRD pattern as given in figure1b. The diffraction pattern of g-C₃N₄shows two distinct peaks at 13.1 and 27.4°which can be indexed as (100) and (200) planes, respectively. The diffraction patterns of the composites are a combination of XRD patterns of KBiFe₂O₅ and g-C₃N₄ as seen in figure1a. As evident from figure1a the peak intensity of g-C₃N₄ grad- ually decreases as the wt% of KBiFe₂O₅increases. So, the Figure 6. (a) Absorbancevs.wavelength curves for KBF, CN and CN-KBF composites and (b) Tauc plot for estimating the energy band gap.

Figure 7. Emission spectrum of 50 W LED bulb used for irradiation.

(5)

XRD patterns of the composites demonstrate two-phase composition: g-C₃N₄and KBiFe₂O₅.

Figure2 gives the TEM micrograph of g-C3N4 which shows the formation of nanosheets. For composites the KBiFe₂O₅ are observed to be well anchored on to the

sheet-like structure of g-C₃N₄. Figure3 shows the SEM micrographs of different photocatalysts. SEM micrographs reveal the formation of irregularly shaped KBiFe2O5grains with some agglomeration. Lamellar structures with large aggregates are observed for g-C₃N₄. For composites the Figure 8. Temporal evolution of absorbance of MB in aqueous solution under visible-light irradiation for different photocatalysts.

(6)

KBiFe₂O₅ particles are observed to be anchored onto the surface of g-C₃N₄layered structure and KBiFe₂O₅particles are well distributed in the composites. Energy-dispersive spectroscopic analysis indicates the presence of all ele- ments, namely, N, C, Bi, K, Fe and O as can be seen in figures4 and 5. EDS chemical map (figure4) shows the homogeneous distribution of KBiFe₂O₅ particles on the surface of g-C₃N₄.

3.1 UV–visible analysis

Optical properties of the samples were studied using UV–

visible diffuse reflectance spectroscopy. Figure 6a depicts the absorbancevs. wavelength plots for the photocatalysts.

As observed the absorbance of the composites extended in the whole range of visible spectrum compared to g-C₃N₄ and also the absorbance increased with increase in the wt%

of KBiFe₂O₅ in the composites compared to pristine g-C₃N₄. The band gap calculation from the Tauc plot which is the plot of [F_KM*ht]²vs. ht, whereF_KMis the Kubelka- Munk function and ht is the photon energy, is shown in figure6b. The calculated values of E_g are 1.75 eV for KBiFe₂O₅(KBF), 2.1 eV for CN-KBF3 and 2.78 eV for CN (g-C3N4). So it is observed that the band gap value of the composite CN-KBF3 is narrower compared to pristine g-C₃N₄. This result is consistent with the observation that for composite, the absorption edge undergoes a redshift compared to pure g-C₃N₄. This redshift in the absorption edge and the narrowing of band gap in case of the composite can be attributed to the formation of the heterostructure between g-C₃N₄ and KBF. As shown in figure10, the conduction band minimum (CBM) of KBF lies below the CBM of g-C₃N₄so the forbidden energy gap

in g-C₃N₄is bridged due to the presence of energy orbitals of KBF thereby narrowing the band gap in the composite material.

3.2 Photocatalytic properties of g-C₃N₄/KBiFe₂O₅ The visible-light-induced photocatalytic activity of the composites was evaluated by degradation of MB under visible light irradiation. A 50 W LED bulb was used as the visible-light source for irradiation which has the emission spectrum as shown in figure7. The photocatalytic activity of g-C₃N₄and KBiFe₂O₅was also carried out under similar conditions for comparison.

Figure8represents the temporal evolution of absorbance of MB for various photo-catalysts. We observed an increase in the photocatalytic degradation of MB with g-C₃N₄/ KBiFe₂O₅ composites compared to both pristine g-C₃N₄ and KBiFe₂O₅.

The photocatalytic kinetics for various photo-catalysts was also studied. The photocatalytic reaction rate depends on the concentration of the organic pollutants and follows the kinetic model called Langmuir-Hinshelwood model [28]. The model is described by equation (1):

ln C₀

C ¼kt; ð1Þ

where C₀ is the concentration of MB after adsorption–

desorption equilibria, C is the concentration of MB after timet, andkis the rate constant obtained from the slope of the plot ln(C₀/C)vs. t.

The photocatalytic degradation by various photo-catalysts are compared in figure9a. Figure9b gives the plot of -ln(C/C₀) vs. t for various photo-catalysts. The obtained Figure 9. (a) Comparison of photocatalytic degradation of MB under visible-light irradiation for different photocatalysts and (b) kinetic study of photocatalytic degradation for various photocatalysts.

(7)

value ofkfrom the slopes are 0.00163 min^-1for KBiFe₂O₅, 0.00404 min^-1 for CN, 0.0083 min^-1 for CN-KBF1, 0.00869 min^-1for CN-KBF2, 0.0157 min^-1for CN-KBF3 and 0.0062 min^-1for CN-KBF4. So we observed that the composites have higher degradation efficiency and CN- KBF3, i.e., CN:KBF with weight ratio 7:3 shows the highest efficiency. So 30 wt% loading is the optimal amount of KBiFe₂O₅in the composites. The enhancement factor for the photocatalytic performance of CN-KBF3 is about 3.9 times the pristine g-C₃N₄ and 9.6 times pure KBF. This value of enhancement factor is comparable to and higher than many other g-C₃N₄-based type-II heterostructures as listed in table 1.

3.3 Mechanism

The reaction rate for CN-KBF3 was found to be 0.0157 min^-1 which is about 3.9-fold enhancement compared to

pristine g-C₃N₄ and 9.6-fold enhancement compared to KBiFe₂O₅. This photocatalytic enhancement in CN-KBF3 and other composites can be ascribed to enhancement in the light absorption capacity and also due to inhibition of recombination of electron–hole pairs because of synergetic effect between g-C₃N₄and KBiFe₂O₅[29].

Using Mullikan electronegativity theory, the valence band (VB) and conduction band (CB) edge potentials of g- C₃N₄ and KBiFe₂O₅ were calculated using the following equations [30,31]:

E_CB¼XE^Cþ 0:5E_g; ð2Þ

E_VB ¼E_CBE_g; ð3Þ

whereXandE_grepresent the absolute electronegativity and energy band gap, respectively. E^C is the free energy of electron on the hydrogen scale (4.5 eV). The calculated value of CB and VB edge potentials of g-C₃N₄are -1.14 and 1.54 eV, respectively [32], while those for KBiFe₂O₅ were calculated to be?0.012 and 1.76 eV, respectively (see Table 1. Comparison of photocatalytic performance of g-C3N4 based type-II heterojunction photocatalysts used for photocatalytic degradation of organic pollutants.

Photocatalysts Enhancement factor of photocatalytic performance References

TiO2/g-C3N4 3 times higher than pristine g-C3N4 [23]

CdS/g-C3N4 3 times higher than pristine g-C3N4 [24]

MoO3/g-C3N4 4.2 and 1.9 times higher than pure MoO3and g-C3N4 [25]

AgI/g-C3N4 2.3 times higher than g-C3N4and 1.3 times higher than AgI [26]

WO3/g-C3N4 4.2 times higher than WO3and 2.9 times higher than g-C3N4 [27]

Fe₂O₃/g-C₃N₄ 1.8 times higher than g-C₃N₄ [21]

KBiFe2O5/g-C3N4 9.6 times higher than pure KBiFe2O5and 3.9 times higher than g-C3N4 Present work

Figure 10. Schematic showing the photocatalytic mechanism in the g-C₃N₄/KBiFe₂O₅composite.

(8)

figure10). Due to the narrow band gap in both g-C₃N₄and KBiFe₂O₅, electron–hole pairs are generated under visible- light irradiation. Since the CB position of g-C₃N₄is more negative than that of KBiFe₂O₅, electrons formed in the CB of g-C₃N₄migrate to CB of KBiFe₂O₅. Similarly, holes from the VB of KBiFe₂O₅migrate to the VB of g-C₃N₄. As a result, no recombination of photo-generated electron–hole pairs takes place and the electrons and holes get accumulated in the CB of KBiFe₂O₅and VB of g-C₃N₄, respectively [33].

4. Conclusions

We have synthesized g-C₃N₄/KBiFe₂O₅ composites with different weight percentage of KBiFe₂O₅ content. Overall photocatalytic efficiency of the composites was found to be higher than the pristine g-C₃N₄and KBiFe₂O₅. CN-KBF3, i.e., composite with 70 wt% g-C₃N₄and 30 wt% KBiFe₂O₅ showed the maximum efficiency with reaction rate of 0.0157 min^-1 which is about 3.9-fold enhancement compared to that of g-C₃N₄and 9.6-fold enhancement compared to KBiFe₂O₅. This enhancement in the photocatalytic efficiency of composites can be ascribed to efficient charge transfer, decrease in the recombination rate of photo-generated electron–hole pairs and increase in their light-ab- sorbing capacity.

Acknowledgements

We express our gratitude to Bhagawan Sri Sathya Sai Baba, the founder chancellor of SSSIHL, for his constant guidance and inspiration. We are grateful to our university, SSSIHL, for providing constant support and Central Research Instruments Facility. We also acknowledge the financial support from DST-FIST (sanction no. SR/FST/PSI-172/

2012).

References

[1] Chong M N, Jin B, Chow C W K and Saint C 2010Water Res.442997

[2] Hou Y, Wen Z, Cui S, Guo X and Chen J. 2013Adv. Mater.

256291

[3] Li Y, Takata T, Cha D, Takanabe K, Minegishi T, Kubota J et al2013Adv. Mater.25125

[4] Zhang J, Guo F and Wang X 2013 Adv. Funct. Mater.23 3008

[5] Yang S, Gong Y, Zhang J, Zhan L, Ma L, Fang Zet al2013 Adv. Mater.252452

[6] Jing L, Zhou W, Tian G and Fu H 2013Chem. Soc. Rev.42 9509

[7] Zhang N, Chen D, Niu F, Wang S, Qin L and Huang Y 2016 Sci Rep.626467

[8] Wang X, Maeda K, Thomas A, Takanabe K, Xin G, Carlsson J Met al2009Nat. Mater.876

[9] Zheng Y, Liu J, Liang J, Jaroniecc M and Qiao S Z 2012 Energy Environ. Sci.56717

[10] Zhang Y, Liu J, Wu G and Chen W 2012Nanoscale45300 [11] Zhang G, Wu H, Li G, Huang Q, Yang C, Huang Fet al2013

Sci. Rep.31265

[12] Zhang J, Sun J, Maeda K, Domen K, Liu P, Antonietti M et al2011Energy Environ. Sci.4675

[13] Hong J, Xia X, Wang Y and Xu R 2012J. Mater. Chem.22 15006

[14] Jun Y S, Park J, Lee S U, Thomas A, Hong W H and Stucky G D 2013Angew. Chem. Int. Ed.5211083

[15] Lin Z and Wang X 2013Angew. Chem. Int. Ed.521735 [16] Liu J and Antonietti M 2013Energy Environ. Sci.61486 [17] Qu Y and Duan X 2013Chem. Soc. Rev.422568

[18] Wang Y, Wang Q, Zhan X, Wang F, Safdara M and He J 2013Nanoscale58326

[19] Gu L, Wang J, Zou Z and Han X 2014J. Hazard. Mater.268 216

[20] Zhang J, Wang Y, Jin J, Zhang J, Lin Z, Huang Fet al2013 ACS Appl. Mater. Interfaces510317

[21] Ye S, Qiu L-G, Yuan Y-P, Zhu Y-J, Xia J and Zhu J-F 2013 J. Mater. Chem. A13008

[22] Li Q, Zhang N, Yang Y, Wang G and Ng D H L 2014 Langmuir308965

[23] Ma L, Wang G, Jiang C, Bao H and Xu Q 2018Appl. Surf.

Sci.430263

[24] Jiang F, Yan T, Chen H, Sun A, Xu C and Wang X 2014 Appl. Surf. Sci.295164

[25] Huang L, Xu H, Zhang R, Cheng X, Xia J, Xu Yet al2013 Appl. Surf. Sci.28325

[26] Liu L, Qi Y, Yang J, Cui W, Li X and Zhang Z 2015Appl.

Surf. Sci.358319

[27] Huang L, Xu H, Li Y, Li H, Cheng X, Xia J et al 2013 Dalton Trans.428606

[28] Shang M, Wang W and Xu H 2009Cryst. Growth Des.9991 [29] Li W, Feng C, Dai S, Yue J, Hua F and Hou H 2015Appl.

Catal. B168465

[30] Liu W, Wang M, Xu C and Chen S 2012Chem. Eng.209386 [31] Xian T, Yang H, Di L J and Dai J F 2015J. Alloys Compd.

6221098

[32] Sun J-X, Yuan Y-P, Qiu L-G, Jiang X, Xie A-J, Shen Y-H et al2012Dalton Trans.416756

[33] Vadivel S, Maruthamani D, Habibi-Yangjeh A, Paul B, Dhard S S and Selvam K 2016J. Colloid Interface Sci.480 126