Introduction

The need for potable water is immensely expanding with the increasing human population. The inevitable textile, pigment and paint release substantial amounts of hazardous dyes to the aquatic system (Tan et al. 2015). Dyes represent a large portion of these pollutants, as about 15% of the 70 million tons of annual production of dyes run into watercourses (Yuan et al. 2020a, b). Consequently, the discharge of these unsafe wastes into the environment jeopardizes human health, causing respiratory problems, skin irritation, and cancer risk (de Lima Barizão et al. 2020; Lou et al. 2020). Moreover, dyes prompt a great chemical oxidation demand causing a nasty odor in wastewater (Midha and Dey 2008). The basic fuchsin (BF), also known as Rosaniline chloride or Basic Violet (Lee and Ong 2017), belongs to the triarylmethane class and possesses a great planar conjugated π-system (Soneta et al. 2006). It is a cationic dye used in various fields such as coloring agents in textile industries, biological stains, paper. (Hunger 2007; Ong et al. 2012). Nonetheless, the basic fuchsin (BF) dye possesses inflammable, anesthetic, and bacterial characteristics (Pathrose et al. 2016; Yamil et al. 2020). Also, due to its weak biodegradation, carcinogenicity, and toxicity (Huang et al. 2012), it has been considered a suspected carcinogen in many countries (Yang et al. 2015). Hence, great attention has been drawn to the elimination of BF from wastewater .

Conventional strategies such as filtration membrane (Aziz et al. 2020), electrochemical processing (Gökkuş 2016; de Almeida et al. 2019), coagulation-flocculation (Gökkuș and Yıldız 2014; Iloamaeke et al. 2021), and aerobic (Castro et al. 2020) are adopted to get rid of these dye contaminants. However, these standard procedures have quite a lot of disadvantages, such as ineffectiveness and a high cost. Instead. Adsorption is an inexpensive and remarkably effective method for eliminating pollutants (Çelebi 2019; Yuan et al. 2020a, b). Natural adsorbents, agricultural wastes, and palm oil waste are utilized for contaminant adsorption and removal (Ai et al. 2011; Ali 2018; Mehr et al. 2020). The efficacy of the adsorption method on the dye’s elimination count upon serval factors such as the dyes concentration, pH, coexisting ions, stirring speed, sorbent dose, contact time, and temperature. For example, Bessashia et al. investigated the influence of the eggshell membrane dose, BF initial concentration, temperature, pH, stirring speed, and contact time on the BF elimination from water (Bessashia et al. 2020). Recently, nanomaterials are employed to discard these hazardous materials from water and wastewater systems (Ashouri et al. 2019). Researchers are seriously charmed with the ZnO for its outstanding characteristics, including the raised binding energy (60 m eV) and sweeping band gap (≈ 3.4 eV) (Sangeetha et al. 2015). ZnO nanoparticles have exhibited an important adsorption affinity toward divers dyes (Chaudhary et al. 2016; Zhang et al. 2016; Palai et al. 2021). Several searches have been carried to explore doped nanomaterials as efficient sorbents for removing dyes from used water. However, doping modifies the proprieties of doped material compared to pure ones, such as the porosity and the surface area, driving to improve adsorption capacity (AC) (Klett et al. 2014; Guan et al. 2019). For instance, it has used several dopants such as chitosan ZnO NPs to remove Direct Blue 78, and Acid Black 26 (Salehi et al. 2010), Mg-doped ZnO (Rath et al. 2019) for the elimination of Congo red, and Pb-doped ZnO as a good adsorbent for Reactive Black 5 (Dassi et al. 2020). However, the utilization of yttrium-doped ZnO NPs, as a potential sorbent for the BF elimination from wastewater, has not yet been published. Therefore, this project aims for the first time to use synthesized Y2O3-doped ZnO (YZnO) for the elimination of BF from aqueous media and investigate the adsorption mechanism of BF.

In this work, YZnO nanoparticles were obtained using a mechanical ball milling approach and were characterized by different techniques (SEM, XRD, BET, and FTIR). The synthesized YZnO was used to eliminate BF dye from aqueous media. For this aim, the influences of the efficient parameters such as pH, initial BF concentration, and coexisting ions will be optimized. The adsorption kinetics will be modeled using four equations: pseudo-first-order, pseudo-second-order, intra-particle diffusion and Elovich. Meanwhile, Temkin, Dubinin–Radushkevich, Elovich, Freundlich, and Langmuir models were utilized to assess the experimental equilibrium. The mechanism of BF sorption on YZnO will also be examined.

Materials and methods

Preparation of YZnO and ZnO

ZnO NPs were obtained via the sol–gel method. First, zinc acetate (39.9 mmoles) was dissolved in a methanol–water mixture (3:1) and agitated for 15 min. Hereafter, a saturated starch solution (30 mL) was appended dropwise, and the blending was stirred until the sol was made. Afterward, 60 ml of the aqueous tartaric acid (66 mmoles) was supplemented dropwise up to gel formation. The collected gel was dried at 105 °C for 10 h. Afterward, the obtained powder was crushed and annealed for 3 h at 500 °C.

YZnO NPs were prepared through a solid-state reaction with stoichiometric amounts of Y2O3 and ZnO by milling in a high-energy planetary mill. The milling process was realized in Fritsch Pulverisette P5 at room temperature for 24 h. The ball-to-powder weight ratio was appointed to 20:1, and the rotational speed was maintained at 450 rpm.

Adsorption experiments details

Experimental Kinetic and equilibrium studies were detailed in the supplementary information. The kinetic study is carried out using the different models, such as pseudo-first-order, pseudo-second-order, intra-particle diffusion, and Elovich. Detailed equations of the used models are listed in Table S1. Temkin, Dubinin–Radushkevich, Elovich, Freundlich, and Langmuir models were employed to assess the experimental equilibrium, and its detailed equations are given in Table S2.

Results and discussion

Structural description of nanomaterials

XRD characterization

The XRD patterns of YZnO and ZnO NPs divulge sharp diffraction peaks, indicating the obtaining of highly crystalline nanostructures. The peaks emerging at 2θ ≈ 69, 68, 66, 62, 56, 47, 36, 34 and 31 ͦ can be, respectively, designated to the (201), (112), (200), (102), (103), (110), (101), (002), and (100) characteristics planes of wurtzite hexagonal ZnO (Swarthmore 1972). Besides, the impurity peaks (indicated as * in Fig. 1 corresponding to the Y2O3 phase) developed due to doping process, connoting that Y was not fully incorporated into the host ZnO lattice, showing phase segregation (Kumar et al. 2015). The appearance of the phase peak (≈ 29 ͦ) indicates the formation of this separate phase. Moreover, the two XRD patterns comparing show diffraction peaks shift owing to the doping to lower values. Whenever the positive charge is higher, the repulsion increases, leading to diffraction angle reduction and inter-planar spacing widening. (Ye et al. 2013). A similar shift of (002) peaks was assigned to the modification of lattice parameters (Yang et al. 2008). The crystallite sizes obtained from Scherrer’s equation were 24.88 and 19.89 nm for YZnO and ZnO, respectively (Barrett 1943). The larger crystallite size of YZnO NPs than the undoped ZnO (Table 1) is consistent with the substitution of the smaller size Zn2+ (0.074 nm) with a larger ionic radius Y3+ (0.090 nm) through the doping process (Zheng et al. 2012). The lattice parameters a \(a (= \lambda /\sqrt{3}.sin{\theta }_{(100)}\)) and \(c (= \lambda /sin{\theta }_{(002)})\) (Modwi et al. 2018) are larger for ZnO. The lattice imperfection and/or distortions \(\varepsilon =\beta /4cos\theta\) increased with Y doping lead to an increase in crystallite size (Mote et al. 2012). The microstrain (εz) follows the same tendency of the microstrain (Karthika and Ravichandran 2015).

Fig. 1
figure 1

XRD patterns for YZnO and ZnO

Full size image
Table 1 The 2θ (101), β, IY, D, c, a and v values of the nanoparticles
Full size table

SEM, EDX and X-ray elemental mapping of YZnO and ZnO

The SEM images of YZnO and pure ZnO (Fig. 2a, b) exhibit an irregular spherical shape with little crystallized dimension nanoparticles. On the other hand, the agglomeration increased for the YZnO with the growth of non-uniform particle configuration. EDX analysis (Fig. 2c, d) was applied to detect the elements existing in ZnO and YZnO nanomaterials. The spectra portrayed sharp signals at 0.5 and 1 keV related to (O) and (Zn) for ZnO (Modwi et al. 2019), and 2, 1, and 0.5 keV for Y, Zn, and O, respectively. Besides, the small peak at 8.5 keV was due to the Zn surface plasmon resonance.

Fig. 2
figure 2

SEM images and EDX of ZnO (a), (c) and YZnO nanopowders (b), (d)

Full size image

The element’s weight proportions attained from the EDX results are presented in Table 2, and the findings were affirmed to be in suitable accordance with the composition.

Table 2 Percentage elements weight
Full size table

The X-ray elemental results of YZnO (Y, Zn, and O) were determined by EDX area scanning (Fig. 3a–c). The spectra were visibly confirmed by homogeneously distributed Y2O3 on the ZnO nanoparticles. Additionally, the X-ray elemental mapping of Zn, O, and Y was well defined with sharp contrast, indicating the successful mixing of Y2O3 with ZnO surface and the formation of YZnO composite construction.

Fig. 3
figure 3

X-ray elemental mapping of (a) Y–La, (b) Zn–Ka, and (c) O–Ka

Full size image

Surface area analysis of YZnO and ZnO

The surface and porosity properties of prepared nanomaterials were recorded employed BET along with that BJH graph. The isotherms correspond to Langmuir type II (Fig. 4a–b) with relative pressure (P/Po = 0.025–0.1), as displayed from obtained N2 sorption isotherms. Besides, the hysteresis loop is type H3 (Fig. 4a–b inset pore distribution) that confirms the mesoporous characteristic of material with slit-shape pores (Lippens and De Boer 1965). All graphs are positioned in the range of 25–160 nm, which is in accordance with the type II adsorption isotherm following IUPAC classification (Sotomayor et al. 2018). The effect of Y2O3 loaded into the ZnO nanomaterials has increased the specific surface area from 10.12 to 20.62 m2 g−1. Additionally, the pore distribution and the pore volume of the two samples equal to 23.46 and 29.75 nm, as well as 0.066 and 0.1641 cm3g−1, respectively. It can be shown that the YZnO structure has an evident influence on the pore size distribution and the surface area, leading to enhanced AC of the yttrium-doped ZnO as opposed to pure ones.

Fig. 4
figure 4

N2 adsorption–desorption cures (inset BJH plots showing pore distribution) of ZnO (a) and YZnO (b)

Full size image

Adsorption of BF

Influence of BF concentration

Experiments are preceded at a temperature of 25 °C and pH = 7 with different concentrations and a constant amount of adsorbent (15 mg of YZnO in 25 mL of dye solution). The BF concentration varied from 5 to 100 ppm. The obtained results (Fig. 5a) show that the adsorbed quantity of BF increases from 7.83 to 75.70 mg g−1 if the dye concentration increased. Herein, the effective driving force produced by increasing the initial BF concentration surmounts any resistance to BF molecules migration from the solution. The resulting fractional adsorption becomes reliant to the initial concentration (Elkady et al. 2011; Bessashia et al. 2020).

Fig. 5
figure 5

Influence of initial b BF dye concentration on the adsorption on YZnO (a), influence of pH on % elimination of BF (b), plot for the determination of pHZPC of YZnO (c), influence of coexisting cations on BF elimination efficiency by YZnO (pH = 11 and BF concentration = 25 ppm) (d),UV–Vis kinetic spectrum adsorption of BF on YZnO (e) and equilibrium time models comparison (f)

Full size image

Influence of pH on BF adsorption

The pH’s influence on BF adsorption onto YZnO is given in Fig. 5b. As shown, the upmost adsorption efficiency was obtained at higher pH values, and when the pH value decreased, the uptake decreased rapidly. The optimal pH for the BF elimination from an aqueous solution was 11. The pHZPC value of YZnO was ≈ 6.3 (pHZPC presents the pH when the surface charge on the YZnO is zero (Fig. 5c)). Accordingly, there develops an electrostatic repulsion between the sorbent and the BF dye at lower pH values leading to low adsorption efficiency (Futamata et al. 2011). The electrostatic attraction between YZnO and BF dye enhances the dye’s adsorption at higher pH (Moawed and Alqarni 2013). Therefore, the positive charge on the cationic dye makes them prone to the attraction by the negatively charged surface at pH above the pHZPC (Tahir and Rauf 2006). Analogous trends on the elimination of cationic dyes were previously reported for MB (Hameed and Ahmad 2009; Mittal et al. 2014) and methyl orange (Shen et al. 2015).

Influence of coexisting ions

Wastewater always included, along with dye, a broad category of coexisting ions that can influence the adsorption process of dyes. Therefore, to investigate the impact of coexisting cations on BF adsorption onto YZnO, NaCl and MgCO3 were chosen as commonly coexisting salts, and the experiments were carried out using two concentrations of cations (25 and 50 ppm). From Fig. 5d, it was established that the addition of Na+ and Mg2+ cations leads to a decrease in the BF elimination efficiency. Moreover, the elimination efficiency of BF decreased when concentration changed from 25 to 50 ppm from 89 to 80% with Na+ and from 70 to 50% with Mg2+. The results moreover affirmed the electrostatic interaction of BF with YZnO, in accordance with the result obtained in the pH study.

Kinetic study

The contact time impact on the adsorption process onto YZnO was explored at BF concentration of 25 ppm and for shaking time from 0 to 1440 min. Figure 5e exhibits a continuous decrease in the population of BF monomers (λmax = 545 nm) and the BF dimers (λmax = 486 nm) as the adsorption progressed (Singha et al. 2017). As given from Fig. 5f, the AC increases quickly within the initial 150 min and achieves equilibrium at about 180 min.

The low correlation coefficient (R2 = 0.8851 (Table 3)) and the deviation of the AC from the experimental data conclude the inapplicability of pseudo-first-order kinetics to the adsorption of BF onto YZnO. The great value of R2 (= 0.9991) and the nearness of the calculated qe (= 26.02 mg g−1) compared to the experimental qe value (25.20 mg g−1) indicate the agreement of the pseudo-second-order kinetic with adsorption (Zamouche et al. 2020). The shorter half-sorption time (t1/2 = 1/( k2.qe) = 19.20 min) and the high initial rate of sorption (h0 = k2.qe2 = 0.0523 mg g−1 min−1) are evidence of favored high rate adsorption of the dye (Li et al. 2012).

Table 3 Used kinetics models for BF adsorption by YZnO
Full size table

Testing experimental data for fitting with the Elovich equation is done by plotting qt versus ln t. The great value of R2 (= 0.9904) for the linear plot indicates the correspondence of the experimental data with the Elovich model (Ho and McKay 2004). The R2 values are more than 0.96, signifying the pertinence of the data to the model. However, the linear plot does not pass per the origin, suggesting another sorption modes involvement (Okello et al. 2017).

The intra-particle diffusion graph (Fig. 6) is multi-linear reflected by the two sorption stages (Yazdani et al. 2016). The sharp first stage may have developed from the dye transport from the BF solution to the external YZnO surface via limit layer, whereas the secondary stage can designate the ultimate equilibrium as the intra-particle diffusion begins to decrease owing to the reduced BF concentration gradient. Table 3 exhibits a lessening in diffusion rate at a long contact time owing to the small pores available for diffusion as the BF molecules formerly pass into the inside voids of YZnO, that proved through the lesser kdif2 value as opposed to kdif1. The higher C value in secondary stage conjectures a bigger limit layer influence. (Milosavljević et al. 2011). Thus, the surface and limit layer sorption or intra-particle diffusion contribute to dye elimination by adsorbent (Ali et al. 2018a, b).

Fig. 6
figure 6

Kinetics data of BF adsorption onto YZnO

Full size image

Adsorption equilibrium

Adsorption isotherm parameters are presented in Table 4. The highest regression coefficient (R2 = 0.9957) and the good fitting of the line (Fig. 7) designate the accordance of the data with Freundlich isotherm (Vijayaraghavan et al. 2006).

Table 4 Used isotherm model for BF adsorption onto YZnO
Full size table
Fig. 7
figure 7

Adsorption of BF equilibrium models

Full size image

The large magnitude of the exponent n > 2 provides a clue on the favorability of adsorption as commonly asserted that n values in the gamut 2–10 signify excellent adsorption characteristics (Treybal 1980). The higher R2 (= 0.9957) for the Freundlich model than Langmuir model (R2 = 0.9611) advocates adsorption on the YZnO heterogeneous surface rather than monolayer adsorption (Ali et al. 2018a, b). The adsorption process is attributed, whether it is primarily physical or chemical, according to the value of E in the Dubinin–Radushkevich isothermal model (Demiral et al. 2008). The physical adsorption happens when the E value is fewer than eight kJ mol−1 (Mondal and Basu 2019), while the process is chemisorption when the E value is among 8 and 16 kJ mol−1 (Zhang et al. 2019). The calculated mean energy value (E) was 7.23 kJ mol−1, confirming that the process of BF adsorption on YZnO was defined as physical.

For the utilization in the field, it is beneficial to compare the AC of YZnO with various sorbents available for BF. Table 5 displays the AC of several sorbents for BF elimination confronted with the YZnO. It is noticeable that the AC of the YZnO is higher than reported sorbents, such as industrial sludges and Fe-MgO-coated kaolinite. This result revealed that YZnO is an effective adsorbent for BF, giving a high potential for dye removal in wastewater.

Table 5 BF adsorption capacities of several sorbents
Full size table
Adsorption mechanism

The study of the pH’s influence implies that the electrostatic attraction could dominate (control) the mechanism adsorption of BF onto YZnO. The BF molecules and YZnO surface were identically charged at low pH values, impeding BF adsorption owing to the electrostatic repulsions. The adsorption efficiency noted at pH < pHZPC proves the existence of other interactions between BF and YZnO. Furthermore, the two primary amine in the molecular structure of BF can establish hydrogen bonds with OH groups of YZnO (Al-Ghouti et al. 2003). Therefore, two reverse mechanisms (hydrogen bonding and electrostatic repulsions) can occur simultaneously at the YZnO surface. However, it can be shown that the BF removal was increased at a pH higher than 6.3; this can be due to hydrogen bonds and the electrostatic attractions between YZnO and BF molecules. FTIR spectra of YZnO, BF, and YZnO@BF (Fig. 8a) were registered in the range of 200–4000 cm−1 to additionally explain the adsorption mechanism. BF spectrum (Fig. 8a) shows many bands that can be distinguished as follow: aryl CH wagging (904–837 cm−1), C = N stretching (1630; 1328 cm−1), and NH2 bending (3296; 1567 cm−1) (Mohammed and Yahia 2018). The YZnO spectrum (displayed in Fig. 8a) presents an important band at 3485 cm−1 attributed to the O–H stretching vibration of water in ZnO (Khezami et al. 2017). After BF adsorption, the stretching O–H band slightly shifted to 3456 cm−1 owing to the presence of hydrogen bond interactions between the lone-pair electrons of amine groups of BF and OH groups of YZnO (Singha et al. 2017). In addition, several characteristic bands of the BF have appeared and changed positions compared to those of the free molecules indicating that interactions may exist between BF and YZnO. For example, the apparition of a novel band at 1641 cm−1 in YZnO@BF spectra implied the establishment of ionic interactions between the negative YZnO charged surface and amino groups of BF dyes (Singha et al. 2017). Liu et al. have proved that the cationic dyes are adsorbed via electrostatic attraction (Liu et al. 2020). The recommended sorption mechanism of the BF onto the YZnO involves hydrogen bonds and the electrostatic attractions between OH groups on the YZnO surface and BF molecules. (Fig. 8b).

Fig. 8
figure 8

FTIR spectra of YZnO, BF, and YZnO @BF (a) and proposed adsorption mechanism of BF onto YZnO NPs (b)

Full size image

Regeneration/reusability experiments

Since cyclic availability is an important property, it is necessary to testify the regeneration and the reusability of an adsorbent. The used YZnO composite was regenerated by calcination for one hour at 500 °C. After the adsorption experiment, the used YZnO was recuperated by filtration and then calcinated at 500 °C before it was reused. Figure 9 shows the reusability results. It is found that YZnO has effectively employed for the BF elimination, at least in four continual cycles.

Fig. 9
figure 9

Reusability Efficiency of YZnO NPs

Full size image

Conclusion

In this work, the YZnO nanoparticle, as an effective sorbent for the BF elimination, was obtained using a mechanical ball milling and was characterized. The sorption experiments were achieved to examine the influence of initial BF concentration, the coexisting ions, and pH. The effect of initial BF concentration had an important effect on BF sorption onto YZnO, and the % BF elimination was established to increase remarkably with pH. The equilibrium time of YZnO was 180 min, and the adsorption amount of YZnO was 75.53 mg g−1 at the optimum pH equal to 12. The great values of R2 indicate the compliance of adsorption to the pseudo-second-order kinetic and Freundlich isotherm models. The adsorption mechanism of BF is associated with the electrostatic interaction and hydrogen bond, as indicated by the pH, the coexisting ions, and the FTIR studies. BF adsorption capacities of YZnO compared to those detailed in the literature predicted that YZnO is an effective adsorbent for BF and has great potential application in BF dye elimination. Hence, YZnO could be an effective sorbent to eliminate other cationic dyes such as Crystal violet, Auramine O, and Malachite green.