1 Introduction

The increase in pollution due to civilization and development in human activities cause global warming and health disorders. Particularly, water pollution because of dum** of industrial wastes into the rivers and other water bodies instigate an imbalance in the water resulting in severe contamination and death of aquatic species. Thus, there is a necessity to curb water pollution through cost-effective scientific methods. Semiconductor mediated photocatalysis has attracted the attention of scientific industry due to its broad applications in the decomposition of environmental pollutants and production of clean hydrogen energy [1,2,3,4,5]. The researchers successfully synthesized a large number of semiconducting materials for the degradation of organic contaminants present in water and air. However, many of these materials respond to UV light only due to their wide bandgap energy (Eg) higher than 3.2 eV. Since only about 4–5% of the solar spectrum falls in the UV range, extensive research has been carried out to investigate the visible light sensitive oxide semiconductors in view of the better utilization of solar energy or indoor artificial illuminations [6,7,8,9,10]. Modification of band edge (valence band or conduction band) position of wide bandgap energy (Eg) materials by do** of cation/anion into their lattice is one of the strategies to make them visible-light sensitive photocatalysts [11, 12]. Noble metals such as platinum (Pt), silver (Ag) and gold (Au) substituted semiconductors have paid more attention mainly in heterogeneous photocatalysis due to their remarkable photocatalytic properties [13, 14]. The d-orbitals of these metals can hybridize with the O 2p orbitals in the valence band of material, increasing the energy of valence band maximum and results in a decrease in Eg.

Silver-based oxides have been used as efficient photocatalysts for the degradation of organic dyes [6, 15, 16]. Thus, the design and development of visible light active photocatalyst, AgAlO2, has been chosen. However, the direct preparation of silver-based oxides by the solid-state reaction encountered practical problems due to the decomposition of Ag2O to elemental Ag at higher temperature [17]. A process employing a cation exchange reaction has been chosen for the preparation of Ag-based oxides. Hence, in the present investigation, AgAlO2 is prepared from LiAlO2 by cation exchange solid reaction process. To the best of our knowledge, synthesis of a flower-like hierarchical β-AgAlO2 and its photocatalytic activity against methylene blue (MB) and rhodamine B (RhB) degradation studies under visible light irradiation have not been reported. The structure and photocatalytic properties of LiAlO2 and AgAlO2 are studied and compared.

2 Experimental

2.1 Synthesis of LiAlO2

Lithium aluminum oxide, LiAlO2, is prepared using the sol–gel method.

Solution “A”: Stoichiometric amount of LiNO3 was dissolved in 15 ml of double distilled water.

Solution “B”: Calculated amount of Al(NO3)3.9H2O was dissolved in 15 ml of double distilled water.

The solution A is mixed with solution B. Aqueous citric acid solution was added to the mixture such that the mole ratio of citric acid to metal ion was 2:1. The pH of the resultant metal citrate solution was adjusted to 6–7 by dropwise addition of dilute ammonia solution. The solution was then slowly evaporated on a magnetic stirrer till a viscous liquid was obtained. At this stage ethylene glycol (gelating reagent) was added (mole ratio of citric acid to ethylene glycol was 1:1.2). This mixture was heated on a hot plate at 100 °C for 2–3 h. The temperature was increased to 160–180 °C to obtain a fluffy solid material. This material was ground in an agate mortar using spectral grade acetone and heated to about 300-400 °C in small amounts in an electric burner to obliterate the organic matter. The resultant ash colored solid (precursor) was heated in a muffle furnace at 700 °C temperature for 14 h to obtain the desired LiAlO2 (heat after named as LAO).

2.2 Synthesis of AgAlO2

LAO powder was mixed with AgNO3 and KNO3 in the molar ratio of 1:2:3 in an agate mortar. Here, the KNO3 is used as a flux. This mixture was crushed thoroughly for 60 min and heated in air at 300 °C for 6 h. The resultant mixture was washed with distilled water to remove unreacted reactants. This material was dried in air at 60 °C and labeled as AAO.

2.3 Characterization

Powder X-ray diffraction (PXRD) data for a sample was collected using a Philips X pert pro X-ray diffractometer with Cu Kα radiation with a scan speed of 0.02 degrees/sec. FT-IR spectral analysis was carried out by Shimadzu spectrometer using KBr pellets. The SEM images were taken on a HITACHI SU-1500 variable pressure scanning electron microscope. The UV–Vis diffuse reflectance spectra (UV–Vis DRS) of the samples were recorded using JASCO V-650 UV–Vis spectrophotometer and BaSO4 was used as reference material. Reflectance spectra were collected in the wavelength range 900–200 nm. The reflectance was transformed into the Kubelka–Munk (KM) function. Photon energy at the absorption edge was estimated by extrapolating the linearly ascending portion of the spectrum to the x-axis, and the value was defined as an optical bandgap. The photoluminescence spectra are recorded with a JASCO FP-8500 spectrofluorometer.

2.4 Photocatalytic activity measurements

The photocatalytic activity of as-synthesized catalysts was tested using HEBER visible annular type photoreactor equipped with a 300 W Tungsten lamp (wavelength range 380–840 nm). In the experiment, 50 ml of 10−5 M aqueous MB (or RhB) solution comprising of 50 mg of AAO (or LAO) catalyst is aerated in darkness for 60 min to achieve the adsorption–desorption equilibrium. Then, the resultant solution was illuminated by a 300 W Tungsten lamp and collected aliquots of the aqueous dye solution at regular 30 min intervals. The absorbance of illuminated dye samples was recorded using a spectrophotometer. The MB (or RhB) degradation efficacy has been determined as follow

$$Degradation \;(\% ) = \frac{{C_{0} - C_{t} }}{{C_{0} }} \times 100$$

where C0 and Ct are the absorbance of MB (or RhB) solution before and after photoirradiation with a UV–Vis spectrophotometer.

2.4.1 Mechanistic photodegradation pathway studies

Controlled photocatalytic experiments using different radical scavengers were carried out under similar experimental conditions to realize the mechanism of photocatalytic degradation of RhB in the presence of AAO. The reaction mixture consists of 3 mL of 2 mM scavenger, 50 mL of RhB and 0.05 g of catalyst. Ammonium oxalate (AO), benzoquinone (BQ) and isopropanol (IPA) were used as scavengers for photogenerated holes, superoxide radical and hydroxyl radical species, respectively.

3 Results and discussions

High-temperature (> 900 °C) preparation of LiAlO2 and its crystal structure has been studied by many authors [18,19,20]. In the present investigation, LAO is prepared at 700 °C using the sol–gel method. It is subjected to PXRD to identify its phase purity. The powder patterns of this sample are well matched with that of γ-LiAlO2 [JCPDF: 73-1338] (Fig. 1). Thus, LAO was crystallized in tetragonal structure with space group P412121. The structure of this phase is shown in Fig. 2a. In this structure, Li and Al are tetrahedrally coordinated to four oxygen atoms such that O-edge shared LiO4 and AlO4 tetrahedra and O-corner shared LiO4–AlO4 pairs. Besides this tetragonal form, LiAlO2 has two other forms, α and β, which are crystallized in trigonal and orthorhombic structures respectively. The phase of the LAO depends on its sintering temperature. At lower temperatures, α and β LiAlO2 phases were obtained beside γ-LiAlO2 [21].

Fig. 1
figure 1

The powder XRD patterns of LAO and AAO

Fig. 2
figure 2

The schematic structures of a γ-LAO and b β-AAO

An X-ray diffraction pattern of ion exchanged product is consistent with β-AgAlO2 (JCPDF No. 21-1332 & 21-1070) (Fig. 1). It is reported that AgAlO2 crystallizes in delafossite structure (rhombohedral system with the space group \(R{\kern 1pt} \overline{3} {\kern 1pt} m\)) or cristobalite related structure (an orthorhombic system with the space group Pna21) [22, 23]. The formation of these forms depends on the starting materials and method of preparation [22,23,24]. The crystal structure of AAO is shown in Fig. 2b. The lattice parameters of both samples are determined using Powd software by least square fitting of their d-lines. The obtained values are close to that of reported values and given in Table 1. In this structure each atom is tetrahedrally coordinated i.e., the structure can be described as corner shared AgO4, AlO4, and OAg2Al2 tetrahedra.

Table 1 Unit cell parameters of LAO and AAO

The scanning electron microscopy (SEM) images of LAO and AAO powders measured in different magnifications (2 μm, and 1 μm) are shown in Fig. 3. It is found to be that both the samples have flower-like hierarchical shapes. The optical properties of these samples are studied by UV–visible diffused reflectance spectra (Fig. 4). The absorption edge of LAO and AAO was observed in UV (340–360 nm) and visible (550–600 nm) regions respectively. The bandgap energy (Eg) of the semiconductor is related to its band edge. The ionic radius of the dopant and the host cations, as well as the chemical nature of the dopants, will influence the relative shift of the absorption edge of the semiconductor [25]. The bandgap energy of the oxides, LAO and AAO, can be obtained from the Kubelka–Munk (KM) plot of (Khν)1/2 versus hν, where K is reflectance transformed according to Kubelka–Munk [\(K = \frac{{(1 - R)^{2} }}{2R}\), where R is reflectancy (%)] and hν is photon energy (Fig. 4). Extrapolation of the linear portion of the plot to (Khν)1/2 = 0 (i.e., onto the x-axis) gives an estimation of bandgap energy. The bandgap energy of LAO and AAO samples is found to be 3.38 and 2.03 eV, respectively.

Fig. 3
figure 3

SEM images of LAO and AAO

Fig. 4
figure 4

Absorbance and KM plot of LAO and AAO

The photocatalytic activity of LAO and AAO was tested by studying the degradation of methylene Blue (MB) and rhodamine B (RhB) aqueous solutions under visible light irradiation. The degradation of MB and RhB in the presence of LAO was found to be negligible. Figure 5 shows the temporal changes in the concentration of the aforementioned organic dyes in the presence of AAO. The degradation of MB/RhB was measured by following the variations in maximal absorption in the UV–visible spectra at 664/554 nm. Before the irradiation experiments, the samples were kept in the dark chamber for attaining the adsorption–desorption equilibrium. The adsorbed amount of MB and RhB on AAO during the dark reaction was found to be 20 and 7%, respectively. The concentration of MB/RhB decreased with visible light irradiation time. The degraded amount of MB and RhB in 3 h of irradiation was about 92 and 63%, respectively (Fig. 5). For semiconductor photocatalytic reactions, light absorption by the material, generation, and migration of light-induced electron–hole pairs are important factors to determine the efficiency of the photocatalyst. The light absorption capacity of the material depends on its bandgap energy. Since the bandgap energy of AAO (2.03 eV) in the visible light region, it absorbs the visible light and generates electron–hole pairs. These electron–hole pairs can migrate to the surface of the semiconductor and combine with surface \(OH^{ - }\) species and dissolved oxygen to form the \(^{ \cdot } OH\) and \(O_{2}^{ \cdot - }\), respectively. Subsequently, these \(OH^{ - }\) and \(O_{2}^{ \cdot - }\) react with dye molecules and degrade them into simple inorganic minerals.

Fig. 5
figure 5

Temporal changes in the concentration of a MB and b RhB in the presence of LAO and AAO under the visible light irradiation

An attempt is made to verify the generation and involvement of these photogenerated holes (h+), hydroxyl (\(^{ \cdot } {\text{OH}}\)), and superoxide (\({\text{O}}_{2}^{ \cdot - }\)) in the RhB degradation process. For this purpose, ammonium oxalate (AO), isopropyl alcohol (i-PrOH), and benzoquinone (BQ) were used as radical quenchers to scavenge the holes (h+), hydroxyl (OH), and superoxide (\({\text{O}}_{2}^{ \cdot - }\)) radicals, respectively, during the photodegradation reaction [26,27,28]. These experiments were carried out as described in Sect. 2.4, but with 2 mM of AO, 2 mM of i-PrOH, and 2 mM of BQ separately under identical experimental conditions (Fig. 6). It is observed from Fig. 6 that the RhB degradation was reduced noticeably by the addition of 2 mM of AO, 2 mM of i-PrOH, and 2 mM of BQ to the RhB solution in the presence of the AAO catalyst. The percentage of degradation of RhB after the addition of AO, i-PrOH and BQ quenchers was found to be 20, 29 and 35% (it was 62% without AO, i-PrOH and BQ separately), respectively. Thus, a decrease in the percentage of degradation of RhB indicates the generation of holes (h+), hydroxyl (\(^{ \cdot } {\text{OH}}\)), and superoxide (\({\text{O}}_{2}^{ \cdot - }\)) species.

Fig. 6
figure 6

Photodegradation Rhodamine B with Scavengers in the presence of AAO catalyst

The probable mechanism is illustrated below:

$$\begin{array}{*{20}l} {AAO + hv \to e_{CB}^{ - } + h_{VB}^{ + } } \hfill \\ {OH^{ - } + h_{VB}^{ + } \to {}^{ \cdot }OH + H^{ + } } \hfill \\ {O_{2} + e_{CB}^{ - } \to O_{2}^{ \cdot - } } \hfill \\ {O_{2}^{ \cdot - } + {}^{ \cdot }OH + RhB \to Byproducts} \hfill \\ \end{array}$$

As stated earlier, LAO has not shown any degradation of MB/RhB under identical conditions (Fig. 4). The bandgap energy of the material and photogenerated electron–hole pairs and their recombination rate play significant role in the photodegradation process. In the present investigation the poor activity of LAO compared to that of AAO can be attributed to (i) wide bandgap energy (3.38 eV), (ii) lower number of electron–hole pairs generated and (iii) higher electron–hole pair recombination rate. The photoluminescence (PL) emission measurements have been widely used to investigate the efficiency of charge carrier migration, trap**, and transfer to understand the fate of electron–hole pairs in semiconductors [29]. It is believed that the higher the intensity of PL signal higher will be electron–hole pair recombination rate. As shown in Fig. 7, the PL intensity in the presence of LAO is maximum indicating higher electron–hole pair recombination rate. Under identical conditions, the PL intensity in the presence of AAO is negligible indicating negligible electron–hole pair recombination. However, in Fig. 5a, 29% of MB degradation is observed in the presence of LAO. It is due to the self-photolysis of MB in the presence of visible light irradiation [30]. The dissimilarities in the rate of degradation of MB and RhB over AAO may be due to their molecular structure, the extent of ionization in aqueous solution and/or adsorbing capacity of the catalyst [30].

Fig. 7
figure 7

Photoluminescence spectra of LAO and AAO (excitation wavelength 390 nm)

4 Conclusions

LiAlO2 was synthesized by a sol–gel method at 700 °C. AgAlO2 is successfully prepared from the mixture of LiAlO2, AgNO3, and KNO3. LAO and AAO are crystallized in tetragonal and orthorhombic structures respectively. The bandgap energy of LAO and AAO was found to be 3.38 eV and 2.03 eV respectively. Both the samples have flower-like hierarchical shape particles. The degraded amount of MB and RhB in the presence of AAO was about 92 and 63%, respectively in 3 h of visible light irradiation. The present study indicates that it was feasible to prepare visible light-sensitive semiconductor material by the ion-exchange method and AAO can be used as photocatalyst for the degradation of organic pollutants in an aqueous medium. Based on competitive photocatalytic reactions using scavengers, it was understood that holes (h+), hydroxyl (\(^{ \cdot } {\text{OH}}\)) radicals and super oxide (\({\text{O}}_{2}^{ \cdot - }\)) radicals played a pivotal role in the photocatalytic degradation of RhB. AAO shows higher photocatalytic activity, which can be ascribed to its lower bandgap energy and more charge separation of electron–hole pairs during the photocatalytic process.