Ni, Co and Ni–Co codoping induced modification in shape, optical band gap and enhanced photocatalytic activity of CeO₂ nanostructures for photodegradation of methylene blue dye under visible light irradiation

Abstract

A simple method has been used to synthesize uneven sizes and shapes of CeO₂ nanostructures by Ni, Co and Ni–Co codoping without using any surfactant. All the samples were further characterized using X-ray diffraction (XRD), scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HR-TEM), energy dispersive X-ray spectroscopy (EDS), Raman spectroscopy, Fourier transform infrared (FT-IR) spectroscopy, ultraviolet-visible (UV-VIS) spectroscopy and photoluminescence (PL) spectroscopy measurements. The Ni–Co codoped CeO₂ nanostructures show broad absorption in the visible range (450–800 nm) as compared to undoped, Ni and Co doped CeO₂ nanostructures. The broad absorption feature (visible range) has made it a suitable material for obtaining enhanced photocatalytic activity under visible light irradiation. Further, Ni–Co codoping reduces the value of the optical band gap of CeO₂ nanostructures sharply from 3.46 to 2.5 eV. The recombination rate of photo-induced electrons and holes for Ni–Co codoped CeO₂ nanostructures is significantly reduced. A more realistic mechanism for superior photocatalytic activity of the Ni–Co codoped CeO₂ nanostructures is also proposed. In the CeO₂ matrix, the Ni and Co ion sites may act as electron and hole trap centers, which essentially improve the separation efficiency of the photo-induced electrons and holes in the Ni–Co codoped CeO₂ nanostructures.

---

1. Introduction

Doping is an important route for obtaining new and useful properties for many functional nanomaterials. Doping involves the introduction of atoms or ions into the host lattice and can be used for the controlled manipulation of the electronic, magnetic and optical properties of nanostructures.^1,2 Recently, it is found that doping has a significant influence on the nucleation/growth of some functional nanomaterials.³ In this context, a number of studies have been done to modify the properties of CeO₂ nanostructures. Chen et al.³ have reported a simple route for modifying the uneven size and shape of alkaline-earth fluoride nanophases to monodisperse ultrasmall nanospheres by lanthanide doping. Feng et al.⁴ have reported the conversion of CeO₂ nanopolyhedra into nanospheres by Ti⁴⁺ doping. Wang et al.⁵ have reported rational tunability of the size and phase of NaYF₄ nanocrystals by lanthanide doping. Gao et al.⁶ have reported the phase and shape controlled of VO₂ nanostructures by antimony doping.

Moreover, the doping of atoms or ions can also greatly affect the optical band gap of nanomaterials. Das et al.⁷ have presented a nice report on the band gap tuning in ZnO through Ni doping. They have reported a strong reduction in optical band gap of ZnO for the doping of Ni up to 4%, and on further higher doping, the band gap does not change much. The strong reduction in band gap at lower doping is mainly attributed to the interaction of the Ni 3d and O 2p states in both, valence and conduction bands. George et al.⁸ have reported the role of Fe doping in tuning the band gap of TiO₂ for the photo-oxidation-induced cytotoxicity paradigm. They showed that the band gap energy of TiO₂ was reciprocally tuned by the doping of Fe content in to TiO₂. Further, the tuning of band gap of semiconductors enhances the application of TiO₂ as an active photocatalyst to be used in photocatalytic devices. The improved photocatalytic activity of the catalysis may have a great impact on the degradation of organic pollutants (ethanol and dyes etc.) to save the environment and human health. Numerous efforts have also been made to enhance the photocatalytic activity of the catalysts using doping, mixing of two semiconductors, etc.^9–11 For this purpose, CeO₂ is widely studied as a catalyst by many research groups.^9–11 CeO₂ has stable chemical properties even at high temperature, closely lattice matched with silicon, fast transferring and large oxygen storage capacity, which make CeO₂ very interesting for the application as a photocatalyst.¹² Channei et al.¹⁰ have reported that the doping of Fe into CeO₂ matrix can enhance the photocatalytic performances of CeO₂ films, compared with undoped CeO₂ films. The enhanced photocatalytic activity was ascribed mainly in terms of the decrease in band gap energy and an increase in specific surface area of the material. Arul et al.¹¹ have reported the excellent photocatalytic activity in Co doped CeO₂ nanorods under the ultra-violet (UV) illumination. In the present work, Ni, Co and Ni–Co were chosen as doping elements into CeO₂ matrix and, we believe that the doped CeO₂ nanostructures may enhance the photocatalytic activity of CeO₂ nanostructures, significantly.

In the present study, we have synthesized the undoped, and Ni, Co and Ni–Co codoped CeO₂ nanostructures by hydrothermal method without using any template or surfactant. The effect of Ni, Co and Ni–Co codoping on structural morphology, optical band gap and photocatalytic activity of CeO₂ nanostructures on methylene blue (MB) dye was investigated in a systematic manner for the first time. Visible light of low power source (8 W) was chosen for illumination over the dye because UV light only composes, 3–5% of the photon flux reaching the earth's surface, while around 45% is in the visible light range. Thus, it would be quite beneficial to use visible light for illumination. The proper mechanism of photodegradation of MB dye has also been proposed in the present study.

2. Experimental

2.1 Synthesis

A simple hydrothermal technique was employed to synthesize undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures. For synthesis, CeCl₃·7H₂O (99.9%, Alfa Aesar), Co(NO₃)₂·6H₂O (98%, Sigma Aldrich) and Ni(NO₃)₂·6H₂O (98.5%, Sigma Aldrich) chemicals were used as precursors. All the precursors were purchased of analytical grade and used without further purification. In a typical synthesis of Ni–Co codoped CeO₂ nanostructures, the appropriate amount of CeCl₃·7H₂O was dissolved in 50 ml of distilled water to make 0.2 M solution. An appropriate amount of Co(NO₃)₂·6H₂O (5 at%) and Ni(NO₃)₂·6H₂O (5 at%) were added simultaneously into the mixed solution. Then, the solution was stirred vigorously for 2 h. After that, ammonia solution (Merck, 25 wt%) was added into the solution drop wise under vigorous magnetic stirring until the pH value of the solution is reached to 11 ± 0.05. Then the mixed solution was again stirred vigorously for 1 h. After being stirred, the mixed solution was transferred into a Teflon-lined autoclave of 50 ml capacity and heated for 12 h at 150 °C under autogenous pressure and static conditions. Thereafter, the autoclave was cooled down to the room temperature naturally. The light yellow colloidal solution was taken out from the autoclave. The precipitate was separated by centrifuging and then washed several times with distilled water. The Ni (5 at%) doped CeO₂ and Co (5 at%) doped CeO₂ powder were also prepared with the same procedure by adding the precursors of both transition metal into CeO₂ solution, separately. The undoped CeO₂ nanostructures were also obtained by the same method without adding any dopant into the initial solution. The synthesis procedure of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures are shown schematically in Fig. 1.

Fig. 1 Schematic presentation of synthesis procedure for undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures.

The formation of CeO₂ nanostructures involves several complicated chemical reactions. When ammonia solution was added into the precursor solution, Ce(OH)₃ precipitate was formed immediately due to the extremely low-solubility. In present case, the high alkaline environment due to high pH value favor the oxidation of Ce(OH)₃ to hydrated Ce(IV) with a subsequent hydrolysis to cerium hydroxide Ce(OH)₄ and finally CeO₂ nanostructures are formed as per following reaction:

Ce(OH)₄ → CeO₂·2H₂O → CeO₂ + 2H₂O

2.2 Instrumentation

The phase purity of the obtained undoped and Ni, Co and Ni–Co codoped CeO₂ samples was characterized by X-ray diffraction using powder diffractometer (Bruker AXS D8) with monochromatic Cu-Kα radiation (λ = 1.5406 Å). The sample was loaded onto an indented glass plate and diffracted signal was recorded for 2θ range from 20 to 70°. The structural morphology of the samples was obtained using SEM (ZEISS EVO15). The size and shape of the synthesized powder samples were determined using HR-TEM (Model 4000FX, JEOL, Boston, MA). The samples were prepared for TEM measurements by dispersing the powder sample in ethanol and putting drop wise on carbon coated copper grids, then drying in air at room temperature. EDS was used to investigate the chemical composition of the prepared samples. The room temperature Raman spectra of the powder samples were recorded in the spectral range, 300–800 cm⁻¹ using Thermo Scientific DXR-XI Raman Imaging Microscope. The 532 nm laser line of the Ar⁺ ions laser was used to illuminate the powder samples. FT-IR spectra of the synthesized samples have been recorded in the spectral range 400–4000 cm⁻¹ with the help of Perkin-Elmer 1600 Fourier transform instrument using the KBr pellet technique. The absorption spectra of the samples were recorded for the spectral range 250–800 nm using Perkin-Elmer Lambda 35 UV-visible spectrophotometer. The diffuse reflectance of the powder samples were also recorded for the spectral range 200–800 nm using Perkin-Elmer Lambda 35 UV-visible spectrophotometer equipped with an integrating sphere accessory. The PL emission spectra of the synthesized samples were recorded by using a spectro-fluorometer (Spex Flurolog3, FL3-22) with a 450 W xenon lamp as the excitation source with 340 nm excitation wavelength. For recording the PL spectra, emission slit width was kept to 3.0 nm.

2.3 Photocatalytic testing for the degradation of MB dye

The photocatalytic activity of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures has been tested by monitoring the photodegradation of MB dye dissolved in an aqueous solution under visible light. The solution containing catalyst was irradiated by the visible light at ambient temperature and pressure under continuous stirring. In the context for energy saving, a low power visible light source (Toshiba, FL8D 8W) was used for irradiation the solution. An amount of 50 ml of aqueous solution was prepared by dissolving 1 mg L⁻¹ of MB dye. Subsequently, undoped, Ni doped, Co doped and Ni–Co codoped CeO₂ nanostructured samples have been added separately in the prepared solution such that the concentration of catalyst in the final solution turns out to be 1 g L⁻¹. The mixed solution was equilibrated by stirring in dark for 1 h to stabilize the adsorption of the MB dye over the surface of catalyst. The dye solution was kept in the UV chamber, and the lamp was positioned exactly over the dye solution. The dye solution was irradiated by visible light using mercury lamp. A cut off filter was inserted between the lamp and the sample to filter out UV light portion (λ < 420 nm). The mixed solution was taken out from the UV chamber after exposing the solution for an interval of 20 min. The total exposure time of the mixed solution was kept to be 2 h. After exposing the mixed solution for each 20 min, the absorption spectrum of the solution was recorded. The absorption spectra of MB dye solution mixed with different CeO₂ samples were recorded within the range of 450–800 nm as a function of irradiated time. The absorption spectra of MB dye revealed a maximum absorption at ∼664 nm. The residual concentration of MB dye was monitored by measuring the intensity of absorption band centered at ∼664 nm using UV-VIS spectrometer (Lambda-35, Perkin-Elmer).

2.4 Reactive species trapping experiments

In order to find the reactive species involved in photodegradation of MB dye, some sacrificial agents, such as 2-propanol (IPA), disodium ethylenediamine tetra-acetic acid (EDTA) and 1,4-benzoquinone (BQ) were used as the hydroxyl radical (OH˙) scavenger, hole (h⁺) scavenger and superoxide radical (O₂˙⁻) scavenger, respectively. The method was kept similar to the former photocatalytic activity test with the addition of 1 mMol of quencher in the presence of MB dye.

3. Results and discussion

3.1 X-ray diffraction analysis

Room temperature XRD patterns of the synthesized undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures are shown in Fig. 2. The XRD pattern of undoped CeO₂ powder sample shows characteristics reflections corresponding to (111), (200), (220), (311) and (222), planes which are located at 2θ = 28.41, 33.11, 47.50, 56.40 and 59.12 respectively. All the diffraction peaks are related to the cubic fluorite-type phase of CeO₂. The diffraction peaks appeared in XRD patterns are indexed using JCPDS card no. 75-0076. By looking at the XRD patterns of Ni, Co and Ni–Co codoped CeO₂ nanostructures, as shown in Fig. 2, it is clear that there is no any peak associated with spurious phases such as Ce₂O₃, CoO, Co₃O₄, NiO, Ni₂O₃ or metallic Co and Ni based phases in the samples. This suggests that after the substitution of Ce⁴⁺ by Co²⁺/Co³⁺ or Ni²⁺/Ni³⁺ into CeO₂ matrix, the cubic structure of CeO₂ is affected neither by the Co and Ni doping nor by Ni–Co codoping. In the XRD pattern of doped CeO₂, the position of XRD peaks slightly shifts to higher angle side which confirms that the doped elements (Ni and Co) are nicely incorporated in the CeO₂ matrix. The lattice parameters of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures are calculated from the XRD peak corresponding to (111) plane using the following relation:¹³
where d_hkl interplanar spacing, a is the lattice parameter, (h, k, and l) are the Miller indices of the plane.

Fig. 2 XRD patterns of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures.

The calculated values of lattice parameter for undoped and Ni, Co and Ni–Co codoped CeO₂ samples are presented in Table 1. The lattice parameter is found to be decreased slighltly for doped CeO₂ nanostructures as compared to the undoped CeO₂ sample.¹⁴

Table 1 The value of lattice parameter, a (Å) calculated for undoped and Ni, Co and Ni–Co codoped CeO₂ samples using XRD patterns

Sample	a (Å)
Undoped CeO2	5.44
Ni doped CeO2	5.41
Co doped CeO2	5.39
Ni–Co codoped CeO2	5.37

---

3.2 Microstructural analysis and growth mechanism

In order to know the shape and size of the synthesized samples, the samples were further characterized using SEM, TEM and HR-TEM. Fig. 3 shows the SEM image of undoped CeO₂ sample. It clearly shows the rod like structural morphology of CeO₂ nanostructures with the length in micrometer range. Further, Fig. 4(a) shows the TEM image of undoped CeO₂ sample. The undoped CeO₂ sample has a distribution of nanostructures that includes both nanorods and less well-defined nanoparticles. Actually, formation and growth of CeO₂ one dimensional (1D) nanostructures significantly depends on the solvent, cerium precursors, temperature and reaction time. If cerium chloride (as in present case) is used as precursor, than the presence of chloride anions play a key role to control the growth of CeO₂ 1D nanostructures. The chloride ions may affect the kinetics of hydrolysis which can influence the nucleation process and thus the overall structural morphology of the final products.¹⁵ Fig. 4(b) shows the TEM image of single nanorod of CeO₂. In this case, some of the nanoparticles can also be seen in the form bunches around the CeO₂ nanorods. The range of length of the nanorods starts from 0.2 μm and goes up to more than 2 μm. The diameter of rods ranges from 15 to 50 nm. As shown in the HR-TEM image (Fig. 4(c)), the ordered stripes of nanoparticles can be seen clearly and the interplanar spacing is calculated to be 0.314 and 0.271 nm that belong to the (111) and (200) reflection planes of CeO₂, respectively. Fig. 4(d) represents the HR-TEM image of CeO₂ nanorod. It confirms the presence of ordered strips belong to (111) reflection plane. The crystallinity and purity of the CeO₂ sample as determined by XRD were also confirmed by selected area electron diffraction (SAED) pattern as shown in Fig. 4(e). Similar conclusions can also be drawn by looking at the SAED pattern which has the smooth rings that can be indexed according to the CeO₂ structure. No additional diffraction spots or rings were detected within the experimental condition and sensibility.

Fig. 3 SEM image of undoped CeO₂ nanostructures.

Fig. 4 (a) TEM micrograph of undoped CeO₂ (b) TEM image of single CeO₂ nanorod (c) HR-TEM image of bunches of particles showing well aligned (111) and (200) planes (d) HR-TEM image of CeO₂ nanorod (e) SAED pattern of undoped CeO₂ sample.

Fig. 5 shows the TEM and HR-TEM images of Ni doped and Co doped CeO₂ samples. From the images it is clear that both the samples have same morphology i.e. distribution of both nanorods and nanoparticles, with a slight difference in population. SAED pattern, as shown in Fig. 5(c) and (f), of Ni doped and Co doped CeO₂ samples, respectively also confirm the absence of any impurity phase in the samples within the experimental condition and sensibility. Fig. 6(a) and (b) show the TEM images of Ni–Co codoped sample at different resolution. From these figures, it is clear that hexagon nanosheets are formed after codoping of Co and Ni into the CeO₂ matrix and 1-D structures are completely disappeared. It means that the shape of the nanostructures is completely modified to hexagon nanosheet by the codoping of Ni (5 at%) and Co (5 at%) in to the CeO₂ matrix. Further, HR-TEM image of hexagon of Ni–Co codoped sample is shown in Fig. 6(c). The well defined 2D lattice planes can clearly be seen in Fig. 6(c). The lattice planes with d-spacing of 0.312 and 0.311 nm are also observed that correspond to (111) planes of CeO₂. The angle between these two (111) planes is calculated to be ∼70.6°. According to crystal geometry, the common perpendicular plane of two (111) planes is (110) plane. Therefore, the surface of the Ni–Co codoped CeO₂ sample is dominated by (110) crystal planes. In previous theoretical study, (110) crystal plane is proven to have highest surface energy and exhibit more catalytic activity for CO oxidation.^16,17 SAED pattern (Fig. 6(d)) of Ni–Co codoped CeO₂ sample also confirms the pure cubic phase of CeO₂ and no any extra ring is observed corresponding to any impurity within the experimental condition and sensibility. Moreover, the Ni–Co codoped CeO₂ sample is also characterized through EDS measurement. The EDS spectrum is presented in Fig. 6(e). The EDS measurements show the peaks corresponding to Ni and Co along with Ce and O in the Ni–Co codoped CeO₂ sample. In addition to these peaks, some more peaks related to Cu and C can also be seen that correspond to the carbon coated copper grid substrate.

Fig. 5 (a) TEM micrograph of Ni doped CeO₂ (b) HR-TEM image of Ni doped CeO₂ nanorod composed of (111) plane (c) SAED pattern of Ni doped CeO₂ sample (d) TEM micrograph of Co doped CeO₂ (e) HR-TEM image of Co doped CeO₂ nanorod composed of (111) plane (f) SAED pattern of Co doped CeO₂ sample.

Fig. 6 (a) TEM micrograph of Ni–Co codoped CeO₂ showing hexagon type of structures (b) TEM micrograph of Ni–Co codoped CeO₂ (c) HR-TEM image of Ni–Co codoped CeO₂ hexagon nanosheet composed of (111) plane (d) SAED pattern of Ni–Co codoped CeO₂ sample showing the absence of spurious phases in the sample (e) EDS spectrum of Ni–Co codoped CeO₂ sample.

In this section, we have mainly focused to understand the evolution of uneven shapes and sizes of CeO₂ samples by the substitution of Co²⁺/Co³⁺ or Ni²⁺/Ni³⁺ in place of Ce⁴⁺. Without doping, distribution of nanorods and nanoparticles are observed with preferential growth direction along (111) reflection plane. With substitution of one ion of Co²⁺ or two ions of Co³⁺ for Ce⁴⁺ in CeO₂ matrix requires an extra oxygen vacancy for charge compensation. The higher surface energy of nanoparticles favors oxygen vacancies to aggregate on their surface which induces transient electric dipoles with outward positive pole.⁶ These transient dipoles restrict the nucleation and growth of CeO₂ nanostructures along a preferential direction. This may be a reason for obtaining less number of 1-D nanostructures in Ni or Co doped CeO₂ as compared to undoped CeO₂ sample. When we further increase the doping concentration (as in the case of Ni–Co codoped CeO₂) in CeO₂, this will further cause the formation of more number of transient dipoles, which essentially minimizes the particle size. This may be the reason to obtain nano hexagon sheets in Ni–Co codoped CeO₂ sample. Thus, it is suggested that an extra repulsion force, as induced by the defects due to dopants in the host matrix, may play a significant role in modifying the structural morphologies, shape and size of the nanoparticles.

3.3 Raman and FT-IR spectroscopy

Since the presence of defects and small grains of impurity phases is difficult to detect by XRD and TEM techniques, the undoped and Ni–Co codoped CeO₂ samples were characterized by Raman spectroscopy at room temperature. A small change in structure can easily be seen by looking at the Raman features of the materials. Fig. 7 shows the Raman spectra of undoped and Ni–Co codoped CeO₂ samples in the range of 300–800 cm⁻¹ measured at room temperature. In the Raman spectrum of undoped CeO₂ sample, the characteristic F_2g mode can be observed at 457 cm⁻¹ confirming the fluorite cubic symmetry of the system.¹² The F_2g Raman band is found to be positioned at 445 cm⁻¹ for the Ni–Co codoped CeO₂ sample. This behavior shows the electron molecular vibrational coupling due to substitution of Co and Ni ions into the host lattice. The shifting in Raman band may be attributed to inhomogeneous strain effect produced by Ni–Co codoping into the host matrix. Further, the asymmetric line broadening of F_2g mode is also found to be increased in Ni–Co codoped CeO₂ sample as compared to undoped CeO₂ sample. This asymmetric broadening may be attributed to the change in confinement due to the change in shape and size caused by the Ni–Co codoping in CeO₂ matrix. It is also important to note that there is no any existence of additional peak in the Raman spectra of undoped and Ni–Co codoped CeO₂ samples which further confirm the absence of any spurious phases of dopant in the samples.

Fig. 7 Raman spectra of undoped and Ni–Co codoped CeO₂ nanostructures.

FT-IR spectroscopy is usually employed as an additional probe to detect the OH functional groups as well as other inorganic and organic species present in the samples. All the undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructured samples are characterized by the FT-IR spectroscopic techniques. The FT-IR spectra are recorded in the spectral 400–4000 cm⁻¹ and shown in Fig. 8.

Fig. 8 FT-IR spectra of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures.

By looking at Fig. 8, it is clear that the CeO₂ nanostructures still contain water molecule since H₂O and CO₂ molecules have property to be chemisorbed easily on the CeO₂ surface when they exposed to the atmosphere. All undoped and Ni, Co and Ni–Co codoped CeO₂ samples show a broad band at ∼3428 cm⁻¹, which is attributed to the O–H stretching vibration.^18,19 A band at ∼1626 cm⁻¹ corresponds to the bending vibration of associated water molecule.^18,19 These two vibrational modes show that the residual water and hydroxyl groups are associated to the surface of CeO₂ samples. The band at ∼554 cm⁻¹ is the characteristic peak of the Ce–O stretching vibration that indicates the formation of CeO₂.¹² The FT-IR bands observed at ∼854, ∼1059 and ∼1518 cm⁻¹ are similar to those bands as reported for CeO₂ nanostructures in earlier studies.^19,20 The bands at ∼854 and ∼1059 cm⁻¹ are attributed to CO₃²⁻ bending vibration and C–O stretching vibration, respectively.²¹ The bands at ∼1518 cm⁻¹ is attributed to carbonate species vibrations.²¹ In addition to these bands, two additional bands are also visible at ∼2341 and ∼1368 cm⁻¹. These vibrations are attributed to the gaseous CO₂ present in the spectrometer because these signals are also observed during the background scan of the FTIR instrument.¹⁹

3.4 UV-VIS spectroscopy

In the CeO₂, the valence band (VB) and conduction band (CB) are composed of O 2p and Ce 5d states, respectively.²² According to density functional theory (DFT) calculations, the gap between the VB (O 2p) and CB (Ce 5d) is lying in range of 6.0–7.5 eV.^22–24 However, the experimental results revealed the value of band gap in the range of 5.5–6.5 eV.^25,26 The Ce 4f state is located inside the gap between VB and CB and forms a narrow vacant band just above the Fermi level.²⁷ Thus, the band gap from the VB to Ce 4f state is generally referred as the optical gap of CeO₂ and its value is reported to be ∼3.6 eV.²⁸

In the present study, UV-VIS absorption and diffuse reflectance measurements are done to study the effect of shape, size and doping of Ni, Co and Ni–Co on the value of optical band gap of CeO₂ nanostructures. The optical absorption spectra of undoped and Ni, Co, Ni–Co codoped CeO₂ nanostructures dispersed in distilled water are shown in Fig. 9(a). The absorbance spectra of all the samples were recorded in the spectral range 250–800 nm. The absorption peaks of undoped, Ni, Co and Ni–Co codoped CeO₂ nanostructures are observed at ∼300, ∼318, ∼309 and ∼345 nm, respectively. The strong absorption band at ∼300 nm in undoped CeO₂ sample is due to the charge transfer from O 2p to Ce 4f states in CeO₂. The doping of Ni and Co ions to the host matrix change the optical properties of CeO₂ nanostructures, which can easily be seen as shifting in absorption maximum with doping compared to undoped CeO₂ sample. The absorption band is found to be shifted towards higher wavelength side compared to its position in undoped sample which could be an indication for well incorporation of Ni and Co into the CeO₂ matrix. The doped samples have large tendency to absorb the light lying in the visible region. Further, in case of Ni–Co codoped CeO₂ samples, a higher absorption feature extending from 450 to 800 nm, completely lying in the visible range has also been observed. Fig. 9(b) shows the reflection spectra of undoped and Ni, Co and Ni–Co codoped CeO₂ powder samples. The reflectance rate of doped CeO₂ samples is lower than that of the undoped CeO₂ sample. In the most significant region i.e. visible region (400–800 nm) Ni–Co codoped CeO₂ samples is found to have the lowest reflectance rate which means high absorption power of the sample. Since, for Ni–Co codoped CeO₂ sample the absorption range is mainly lying in the visible range, it could be a better candidate for achieving enhanced photocatalytic activity under the visible light and thus it could be used as a good photocatalyst for the photodegradation of long organic molecules.

Fig. 9 (a) UV-VIS absorption spectra (b) reflectance spectra, of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures.

The value of direct band gap (E_g) of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures are calculated by extrapolating the linear portion of (ahν)² vs. hν plots to intercept the photon energy axis, as shown in Fig. 10. The values of optical band gap are calculated to be 3.46, 3.24, 3.27 and 2.5 eV for undoped, Ni doped, Co doped and Ni–Co codoped CeO₂ nanostructures. It is clear that the optical band gap decreases significantly with doping. A huge decrease in the value of optical band gap from 3.46 to 2.5 eV is found for Ni–Co codoped CeO₂ as compared to undoped CeO₂ sample. The reduction in the value of optical band gap may be due to the impurity/defect level introduced by the influence of Ni and Co doping.² When Co²⁺/Co³⁺ or/and Ni²⁺/Ni³⁺ are substituted in CeO₂ matrix, the oxygen vacancy is naturally formed to adjust the charge neutrality. Therefore, the individual doping of Ni and Co would also modify the electronic structure of CeO₂ by introducing the defect energy level between VB and CB. This will essentially cause band gap narrowing and resulting a red shift in the absorption band.

Fig. 10 (ahν)² versus photon energy (hν) plots to calculate the optical band gap of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures.

3.5 Photoluminescence spectroscopy

In order to determine the electron–hole recombination characteristic of the undoped, Ni, Co and Ni–Co codoped samples, all the undoped and doped samples are further characterized by PL spectroscopy. It is well known that PL emission results from the recombination of excited electrons and holes. In the emission spectra, low emission intensity signifies the low recombination rate of electron and hole pairs. The room-temperature photoluminescence spectra of undoped and Ni, Co and Ni–Co codoped CeO₂ samples are recorded in the spectral range of 370–650 nm with the excitation wavelength of 340 nm and the recorded spectra are presented in Fig. 11. The emission spectrum of undoped CeO₂ sample exhibit a broad band centered at ∼402 nm. Our experimental results are well consistent with the value of emission peak reported in earlier study.^2,29 The emission bands of CeO₂ lying in the spectral range 400–500 nm are associated with the transition from the defect level to the O 2p level of CeO₂.¹⁵ Here, the strong blue emission line centered at 402 nm would therefore correspond to the transition from defect state to (including oxygen vacancies) electronic energy levels below the 4f band.²⁹ It further indicates that the CeO₂ nanostructures possess higher oxygen vacancy levels, which is quite helpful for fast oxygen transport.

Fig. 11 PL spectra of undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures.

Colis et al.²⁹ have reported that the emission intensity of ∼405 nm band of CeO₂ nanostructure is increased upon Co doping. The increased intensity of the band at ∼405 nm is attributed to the existence of a large amount of oxygen vacancies in doped CeO₂ matrix. However, in the present study, the intensity of emission band is decreased with the doping of Co or Ni into CeO₂ matrix. A similar kind of result was also obtained by Ranjith et al.² for Co doped CeO₂ nanoparticles. This essentially means that the recombination of photo-generated electrons and holes is effectively suppressed by codoping. It is one of the reasons to expect the enhanced photocatalytic activity for Ni–Co codoped CeO₂ sample. Moreover, from Fig. 11, it can be seen that all Ni, Co and Ni–Co codoped CeO₂ samples show a red shift from 402 to 410 nm, which may be attributed to the shifting of localized energy levels (impurity based) lying in between Ce 4f and O 2p levels.

3.6 Photocatalytic activity

The UV-VIS spectra of neat MB dye and its aqueous solution mixed with the catalyst samples (undoped, Ni, Co and Ni–Co codoped) are recorded for the spectral range 450–800 nm as function of irradiation time. The UV-VIS spectra are recorded after irradiating the solutions by the visible light (not shown here). From the absorption spectra, we found that the intensity of main absorption band centered at ∼664 nm is gradually decreased with the increase of irradiation time at regular interval of 20 min. The decrease in the intensity of absorption band represents the degradation of the MB dye with the increase of irradiation time. The normalized residual concentration of MB dye is calculated using the following relationship:³⁰
where C_o is initial concentration of MB dye, C is residual concentration of MB dye, A_t is intensity of absorption band after any irradiated time t and A_o is intensity of absorption band at time t = 0. The calculated value of photodegradation as a function of irradiation time is presented in Fig. 12. When the solution of neat MB dye is irradiated for 120 min, the degradation of MB dye is found to be very small (∼3%). This indicates that MB was stable and difficult to decompose in absence of photocatalyst. From the catalytic experiments, Ni–Co codoped sample was found to be more photoactive for photodegradation of MB dye as compared to undoped, Ni doped and Co doped CeO₂ samples. The highest degradation of MB dye within the limit of irradiation time is found to be nearly 60% for Ni–Co codoped CeO₂ sample, which is the excellent photocatalytic activity of the Ni–Co codoped sample as compared to other samples. The enhancement in photocatalytic activity for Ni–Co codoped CeO₂ sample is basically attributed to the red shift of optical band gap in Ni–Co codoped sample. The cycling runs for the photodegradation of MB dye with Ni–Co codoped CeO₂ as photocatalyst were performed to evaluate its photocatalytic stability and recyclability and it is shown in Fig. 13. After reusing three cycles, the photodecomposition rate of MB dye is still remained approximately same.

Fig. 12 Photodegradation efficiencies of aqueous solution of neat MB dye, MB dye mixed with undoped, Ni doped, Co doped and Ni–Co codoped CeO₂ nanostructures as a function of irradiation time under visible light irradiation.

Fig. 13 Cycling runs for the photocatalytic degradation of MB dye for Ni–Co codoped CeO₂ nanostructures under visible light irradiation.

In order to detect the main reactive species responsible for photodegradation of MB dye, IPA, EDTA and BQ were employed as the hydroxyl radical (OH˙) scavenger, hole (h⁺) scavenger and superoxide radical (O₂⁻˙) scavenger, respectively to the MB dye solution having Ni–Co codoped CeO₂ catalyst. By following the same procedure, residual concentration of MB dye measured at different time is presented in Fig. 14. It can be seen that the photodegradation of MB dye is decreased greatly when IPA is added into the solution of MB dye with catalyst. This confirms that hydroxyl radicals (OH˙) are the main reactive species involve in degradation of MB dye.

Fig. 14 Photocatalytic degradation of MB dye with Ni–Co codoped CeO₂ nanostructures alone and with different scavengers such as; IPA, EDTA, and BQ.

3.6.1 Mechanism of photodegradation. In order to calculate the band positions of VB and CB using optical band gap energy values, following empirical formulas are used:^30,31

E_VB = X − E^e + 0.5E_g

E_CB = E_VB − E_g

where E_VB is the valence band edge potential, E_CB is the conduction band edge potential, X is the electronegativity of the semiconductor, which is the geometric mean of the electronegativity of the constituent atoms, E^e is the energy of free electrons on the hydrogen scale (about 4.5 eV), E_g is the band gap energy of the semiconductor. The CB and VB edge potentials of undoped CeO₂ sample are determined to be −0.67 eV and 2.79 eV, respectively. On the basis of above result, the probable photocatalytic mechanism in Ni–Co codoped CeO₂ nanostructures is presented by a schematic diagram as shown in Fig. 15. The doping of Ni and Co ions into the CeO₂ matrix sharply reduced the band gap by adding the energy levels between VB and CB of CeO₂. When visible light is incident on MB dye suspension, the electron–hole pairs are generated due to the ejection of electron from VB that creates a hole in the VB. The decreased band gap of Ni–Co codoped CeO₂ is more favourable to absorb the exciting light and generate more number of electron–hole pairs compared to undoped CeO₂ sample. The doped Ni and Co ions energy level can act as a trapping site for photoinduced electrons and holes, respectively.^32,33 The doped Ni ions sites preferably act as an electron trapping centre while Co ions sites act as hole trapping centre to facilitate charge carrier localisation, which mainly slow down the recombination rate of electron–hole pairs. Therefore, Ni–Co codoping may be more effective for the production of those materials that can delay electron–hole recombination rate. The increase in the lifetime of the electron–hole separation supports the charge carrier transfer to the catalyst surface. The adsorption of MB dye takes place through coulombic interaction with OH⁻ present on the surface of catalyst since MB dye has cationic configuration (Fig. 15). The generated holes are allowed to react with OH⁻ ions and electrons may react with dissolved O₂ and creates the OH˙ radical into the aqueous solution. These OH˙ radicals can attack to C–S⁺=C functional group of MB dye, attached through coulombic interaction with the surface of catalysts³⁴ and degrade the MB dye into the final products such as; CO₂, SO₄²⁻, NO₃⁻, H₂O and H⁺. Thus, the photocatalytic activity and durability of Ni–Co codoped CeO₂ sample will greatly improve the removal of organic dyes and pollutants present in the aqueous solution.

Fig. 15 Schematic presentation of photodegradation mechanism of MB dye with Ni–Co codoped CeO₂ nanostructures under visible light irradiation.

4. Conclusions

In summary, undoped and Ni, Co and Ni–Co codoped CeO₂ nanostructures were successfully synthesized by a simple, one step hydrothermal method at low temperature (150 °C) without using any surfactant. This study has offered a simple route for modifying the size and shape of CeO₂ nanostructures. XRD, Raman and FT-IR analysis showed that all the samples have face-centered cubic structure. EDS analysis has confirmed the presence of Ni and Co into the CeO₂ matrix. Ni–Co codoped CeO₂ was found to have excellent photocatalytic activity for the degradation of MB dye under visible light irradiation. Hydroxyl radicals (OH˙) were found to be main reactive species involved for the degradation of MB dye. The Ni–Co codoping was mainly involved to reduce the value of optical band gap from 3.46 to 2.5 eV. The Ni–Co codoping has made the CeO₂ nanostructures suitable for absorption of visible light especially in the range of 450–800 nm. The reduction in intensity of emission band for Ni–Co codoped CeO₂ sample, as compared to undoped CeO₂, revealed suppression in the recombination rate of photo-generated charge carriers, which mainly enhances the photocatalytic activity and stability of Ni–Co codoped CeO₂ nanostructures.

Acknowledgements

AKO is thankful to DST for providing the financial assistance in the form of a research project grant no. SERB/F/4663/2013-14.

References

1. F. Wang and X. Liu, Chem. Soc. Rev., 2009, 38, 976.
2. K. S. Ranjith, P. Saravanan, S. Chen, C. Dong, C. L. Chen, S. Chen, K. Asokan and R. T. R. Kumar, J. Phys. Chem. C, 2014, 118, 27039.
3. D. Chen, Y. Yu, F. Huang, P. Huang, A. Yang and Y. Wang, J. Am. Chem. Soc., 2010, 132, 9976.
4. X. Feng, D. Sayle, Z. Wang, M. Paras, B. Santor, A. Sutorik, T. Sayle, Y. Yang, Y. Ding, X. Wang and Y. Her, Science, 2006, 312, 1504.
5. F. Wang, Y. Han, C. Lim, Y. Lu, J. Wang, J. Xu, H. Chen, C. Zhang, M. Hong and X. Liu, Nature, 2010, 463, 1061.
6. Y. Gao, C. Cao, L. Dai, H. Luo, M. Kanehira, Y. Ding and Z. Wang, Energy Environ. Sci., 2012, 5, 8708.
7. S. C. Das, R. J. Green, J. Podder, T. Regier, G. Chang and A. Moewes, J. Phys. Chem. C, 2013, 117, 12745.
8. S. George, S. Pokhrel, Z. Ji, B. L. Henderson, T. Xia, L. Li, J. Zink, A. E. Nel and L. Madler, J. Am. Chem. Soc., 2011, 133, 11270.
9. H. Yang, K. Zhang, R. Shi and A. Tang, J. Am. Ceram. Soc., 2007, 90, 1370.
10. D. Channei, B. Inceesungvorn, N. Wetchakun, S. Ukritnukun, A. Nattestad, J. Chen and S. Phanichphant, Sci. Rep., 2014, 4, 5757.
11. N. Arul, D. Mangalaraj, P. Chen, N. Ponpandian, P. Meena and Y. Masuda, J. Sol-Gel Sci. Technol., 2012, 64, 515.
12. S. Khan, M. Faisal, M. Rahman and A. Jamal, Sci. Total Environ., 2011, 409, 2987.
13. M. A. Wahab, Solid State Physics, Narosa Publishing House, New Delhi, 2nd edn, 2010.
14. R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751.
15. C. Sun, H. Li, H. Zhang, Z. Wang and L. Chen, Nanotechnology, 2005, 16, 1454.
16. T. X. T. Sayle, S. C. Parker and C. R. A. Catlow, Surf. Sci., 1994, 316, 329.
17. J. Conesa, Surf. Sci., 1995, 339, 337.
18. S. Kumar, S. Bhunia and A. K. Ojha, Phys. E, 2015, 66, 74.
19. S. Phoka, P. Laokul, E. Swatsitang, V. Promarak, S. Seraphin and S. Maensiri, Mater. Chem. Phys., 2009, 115, 423.
20. H. Li, G. Wang, F. Zhang, Y. Cai, Y. Wang and I. Djerdj, RSC Adv., 2012, 2, 12413.
21. L. Truffault, M. Ta, T. Devers, K. Konstantivov, V. Harel, C. Simmonard, C. Andreazza, I. P. Nevirkovets, A. Pineau, O. Veron and J. Blondeau, Mater. Res. Bull., 2010, 45, 527.
22. C. W. M. Castleton, J. Kullgren and K. Hermansson, J. Chem. Phys., 2007, 127, 244704.
23. R. Gillen, S. J. Clark and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2013, 87, 125116.
24. H. Jiang, R. I. Abal, P. Rinke and M. Scheffler, Phys. Rev. Lett., 2009, 102, 126403.
25. F. Marabelli and P. Wachter, Phys. Rev. B: Condens. Matter Mater. Phys., 1987, 36, 1238.
26. A. Pfau and K. D. Schierbaum, Surf. Sci., 1994, 321, 71.
27. P. J. Hay, R. L. Martin, J. Uddin and G. E. Scuseri, J. Chem. Phys., 2006, 125, 034712.
28. S. Guo, H. Arwi, S. N. Jacobsen, K. Jarrendahl and U. Helmersson, J. Appl. Phys., 1995, 77, 5369.
29. S. Colis, A. Bouaine, G. Schmerber, C. U. Bouillet, A. Dinia, S. Choua and P. Turek, Phys. Chem. Chem. Phys., 2012, 14, 7256.
30. S. Kumar and A. K. Ojha, J. Alloys Compd., 2015, 644, 654.
31. F. Guo, W. Shi, X. Lin, X. Yan, Y. Guo and G. Che, Sep. Purif. Technol., 2015, 141, 246.
32. M. Nishikawa, S. Hiura, Y. Mitani and Y. Nosaka, J. Photochem. Photobiol., A, 2013, 262, 52.
33. M. Luo, Y. Liu, J. Hu, H. Liu and J. Li, ACS Appl. Mater. Interfaces, 2012, 4, 1813.
34. A. Houas, H. Lachheb, M. Ksibi, E. Elaloui, C. Guillard and J. M. Herrmann, Appl. Catal., B, 2001, 31, 145.

---