Removal of methylene blue (MB) by Bimetallic-Metal organic Framework

ABST CT: In this study, three improved versions of UiO-66 were synthesised successfully: Different ratios of Ca+2/Zr+4 were used to synthesize UiO-66, UiO-66-10%Ca and UiO-66-30%Ca. ese MOFs were used in batch adsorption experiments to remove different concentration of methylene blue (MB) from wastewater. When the concentration of MB was 50 ppm, UiO66-10%Ca had the highest affinity toward MB therefore it exhibited adsorption capacity of 50.25 mg.g–1. Furthermore, the MB adsorption capacity was 22.75 mg.g–1 and 14.84 mg.g–1 in UiO-66-30%Ca and UiO-66 respectively. For equilibrium study, Langmuir and Freundlich models were used to t the experimental data. Freundlich model was the best to describe the adsorption equilibrium of MO. A kinetics study was described by pseudo rst-order, pseudo second-order and intraparticle diffusion models. Pseudo second-order model demonstrated the best ing to the experimental data. e MOFs used in this study are suggested to be a ractive adsorbents to remove dyes from wastewater.


INTRODUCTION
Dyes exist where there is civilisation. ey are used to colour products, and employed in various industries, such as the food, paper, carpet, rubber, plastic, cosmetic, acrylic, wool, nylon, silk and textile industry [1-3].
Cationic methylene blue MB (tetramethylthionine chloride) is a basic thiazine dye as shown in Figure 1 [4]. As a basic dye, MB is not strongly hazardous, but may cause some harmful effects on humans and aquatic lives. It is also resistant to biological degradation [5]. Water is a precious resource for all living creatures on earth. A signi cant environmental challenge is the removal of dye pollutants from fabric and textile wastewater [6]. Use of dyes to colour products consumes signi cant volumes of water; consequently, a substantial amount of coloured wastewater can be generated [7]. Many approaches to dye removal have been proposed to treat the industrial wastewater [8,9]. e techniques are classi ed into three main types: physical, chemical and biological treatments. ese techniques include coagulation, membrane, separation process, adsorption process, ltration, so ening , reverse osmosis, electrochemical processes, chemical oxidation, and aerobic and anaerobic microbial degradation [10]. e adsorption process is the simplest technique for dye removal due to its low cost, easy availability, simplicity of design, high efficiency, ease of operation [11,12]. While activated carbon is presently believed to be the most operative adsorbent, its high cost means its production and regeneration remain uneconomical [13,14]. is limitations of using activated carbons have led researchers to seek low-priced dye sorbents, such as coal, y-ash, silica gel, wool waste, agricultural waste, wood waste, and clay materials [11,15]. In recent years, research and development in the eld of design and synthesis of MOFs has led to a rapid growth in practical and conceptual developments [16][17][18][19][20][21]. An extensive class of crystalline materials has become available because of metal organic framework (MOF) chemistry, which has superior characteristics such as high stability, tuneable metrics, organic functionality and porosity [16].
Its exceptional porousness means that MOFs have potentially numerous applications; their demonstrated applications in gas storage, separations, catalysis, energy technology fuel cells, supercapacitors and catalytic conversions has made them objects of intensive study, industrial-scale production and application [22][23][24][25].
e unique characteristics of MOF-type substances that make them the focus of much worldwide research are their pore geometry and high porosity [26,27], their central metals [28,29], open metal sites [30,31], functionalised linkers [32,33] and their loading of active species [34,35]. All these characteristics have been scienti cally employed to successfully improve interactions between the sorbates and MOFs. Speci cally, these characteristics distinguish MOFs from other porous material in the eld of adsorption processes for the effective removal of hazardous compounds [36]. Accordingly, MOFs are superior adsorbents because of their various host-guest interactions, acid-base [37,38], π-complexation [39], H-bonding [40,41] and coordination with open metal sites [31,36]. e pore size has a prime effect on the adsorption capacity of MOFs therefore exceptionally high dye uptake was demonstrated by mesoporous MOFs [42]. Very few studies was made based on bimetallic MOF [43]. UiO-66 has a ractive characteristics as an adsorbent to toxic chemicals from wastewater because it has a higher hydrothermal stability among other MOFs [44]. Bimetallic-UiO-66 was recently used to signi cantly adsorb anionic dyes [45].
In this study, based on batch adsorption experiments, UiO-66 and UiO-66-Ca samples were used as sorbents to remove MB from an aqueous solution. Equilibrium and kinetic adsorption models were used to represent the experimental data. e equilibrium study was undertaken using Langmuir and Freundlich isotherms. e kinetics study was conducted using pseudo rst-order and pseudo second-order models as well as intraparticle diffusion.

Synthesis and Activation
All chemicals were supplied by Sigma-Aldrich (Australia) without further puri cations.
UiO-66 was synthesised successfully using a scaled-up procedure of a previously reported method [46]. Speci cally, 2.27 mmol of ZrCl 4 and 2.27 mmol 1,4-benzenedicarboxylic acid (BDC) were mixed with 405.38 mmol of N, N-dimethylformamide (DMF) in an autoclave and heated in a preheating oven at 393 K for 24 h. e produced UiO-66 was immersed in chloroform for 5 days to remove unreacted precursors. en, the crystalline product was ltered and dried under vacuum at 463 K for 48 h.
UiO-66-10%Ca was synthesised by mixing ZrCl 4 (1.5 g) with BDC (1.1 g) in 73 mL of DMF. A er mixing for 15 min, 0.15 g of Ca (NO3) 2 .4H 2 O was added and followed by the addition of 2 mL of H 2 O to the mixture. e solution is mixed for approximately 30 min; then transferred into a 125-mL Te on-lined autoclave, which is tightly sealed and placed in a preheated oven at 132 °C for 1 d.
UiO-66-30%Ca was synthesised by mixing ZrCl 4 (1.5 g, 6.44 mmol) with BDC (1.3 g, 7.82 mmol) in 70 mL of DMF. e solution was mixed for 30 min, then Ca (NO 3 ) 2 .4H 2 O (0.45 g, 2.86 mmol, 99%; Sigma-Aldrich) was added to the mixture. A er that, 5 mL of deionised water was added into the mixture. Eventually, the mixture was transferred to a Te on-lined autoclave which was tightly sealed and moved into a preheating oven at 430 K. e products were then ltered, dried and immersed in absolute methanol (100%, Sigma-Aldrich) for 5 d, a er that it was dried and heated under vacuum at 473 K overnight before use as adsorbents.
2.2 Characterisation e thermal stability of UiO-66, UiO66-10%Ca and UiO66-30%Ca were assessed by a thermogravimetric analysis (TGA) instrument (TGA/DSC1 STARe system; Me ler-Toledo). e samples were loaded into a pan and heated to 1173 K at a rate of 5 K/min. e air ow rate was maintained at 50 mL/min. FTIR spectra (Spectrum 100 FT-IR spectrometer, PerkinElmer, Waltham, USA) were obtained to assess the stability of the functional groups on the organic ligands. e spectra were scanned from 600 to 4000 cm -1 with a resolution of 4 cm -1 using an a enuated total re ectance technique. X-ray powder diffraction and pa erns were obtained with an X-ray diffractometer (D8 Advance, Bruker AXS) using Cu Kα radiation (λ = 1.5406 Å) with accelerating voltage and current of 40 kV and 40 mA respectively. Autosorb-1(Quantachrome, instruments) was used to determine N 2 adsorption/ desorption isotherms as well as the pore size and surface area of the MOFs. e samples were initially evacuated at 473 K for 24 h. en, the sample was analysed to determine surface area, pore size and pore volume.

Adsorption Study
An aqueous stock solution of MB (1000 ppm) was prepared by dissolving MB (C 16 H 18 ClN 3 S, ≥95%, Sigma-Aldrich) in deionised water. Aqueous solutions with different concentrations of MB (5-100 ppm) were prepared by successive dilution of the stock solution with water. A er obtaining the UV spectra of the solutions with a spectrophotometer (UV spectrophotometer), the MB concentrations were determined using absorbance at 668 nm wavelength of the solutions. A calibration curve was obtained from the spectra of the standard solutions (5-100 ppm).
Prior to adsorption, the adsorbents were dried overnight under vacuum at 373 K. Several glass containers were cleaned, dried and lled to 20 mL with MB of different concentrations ranging from 5 to 50 ppm. An exact amount of the MOF adsorbent (20 mg) was then put in each container. e dye solutions containing the adsorbents were mixed well with a magnetic stirrer and maintained for a period from 5 min to 24 h at 298 K. Samples for analysis were collected by a syringe lter at different sampling intervals. UV spectrometer was used to investigate the dye content in the supernatant.
Adsorbed amounts of MB by the Zr-MOFs at each time interval of time, the equilibrium and percentage removal of MB were computed according to the following equations:

Applied Materials and Technology
(1) Where: qt : the amount of MB adsorbed per unit weight of MOF at any time t (mg/g) qe : the amount of MB adsorbed per unit weight of MOF at equilibrium (mg/g) C 0 : the initial concentration of the MB solution at time zero (mg/L) C t : the concentration of the MB solution at time t (mg/L) C e : the concentration of the MB solution at equilibrium (mg/L) V : volume of the MB solution in batch adsorption process (L) R% : percentage removal of MB [5] m : Zr-MOF mass used in adsorption batch process (g) [1, 5,47].
Kinetics study: Adsorption mechanism and rate of diffusion were estimated using three kinetic models: pseudo rst-order, pseudo second-order [48][49][50] and intraparticle diffusion model [51,52]. e nonlinear form of the Lagergren pseudo first-order kinetic equation can be wri en as follows [54,55]: e linear form of the pseudo first-order kinetic equation can be expressed as follows: e nonlinear form of the pseudo second-order kinetic equation can be wri en as follows [56]: e linear form of the pseudo second-order kinetic equation can be wri en as follows: Where: qe : the amount of MB adsorbed per unit weight of MOF at equilibrium (mg/g) qt : the amount of MB adsorbed per unit weight of MOF at any time t (mg/g) k1 : pseudo first-order rate constant (min -1 ) t : time (min) k 2 : pseudo second-order rate constant (g/mg min).
A linear plot of the pseudo rst-order model (ln [q e -q t ]) against time provides the values for the kinetics sorption parameters, such as rate constant (k 1 ), equilibrium adsorption capacity (q e ) and the linear regression coefficient (R 2 ). Likewise, a linear plot of the pseudo second-order model (t/q t ) against time also provides the rate constant (k 2 ), equilibrium adsorption capacity (q e ) and the linear regression coefficient (R 2 ).
As a result of the limitations of the pseudo first-order and pseudo second-order kinetic equations, the lack of an identi ed adsorption mechanism and the rate-limiting steps in the adsorption process, Weber and Morris established intraparticle diffusion model [117]. In general, the migration of sorbate molecules in bulk to the surface of a solid sorbent by intraparticle diffusion process is what controls the rate of most liquid/solid sorption systems. e analysis using Weber and Morris's intraparticle diffusion model is as follows [51,52]: Where: qt : the amount of MB adsorbed per unit weight of MOF at any time t (mg/g) kp : intraparticle diffusion rate constant (mg/g min 0.5 ) t : time (min) C : constant represents the surface adsorption [57][58][59].
Equilibrium study: e adsorbents' adsorption behaviours were simulated using the Freundlich and Langmuir adsorption isotherms [48][49][50]53]. e nonlinear form of the Langmuir isotherm can be expressed as: while the linear form can be wri en as [60]: Where: qm : Langmuir maximum loading capacity (mg/g) k L : Langmuir constant related to the energy of adsorp tion and affinity of binding sites (L/mg)[61] C e : Equilibrium concentration of dye in solution (mg/L) qe : Amount of dye adsorbed at equilibrium per unit mass of sorbent (mg/g).
e equilibrium experimental data were ed using the linear form of the Langmuir isotherm equation (Equation  9). Speci cally, the Langmuir parameters q m , K L , and R 2 were obtained from the plot of (C e /q e ) against C e . e dimensionless constant separation factor, R L , is vital to the Langmuir isotherm, and can be found in the following equation [62][63][64][65]:

Applied Materials and Technology
(2) Where C 0 is the initial concentration of adsorbate (mg/L) and K L (L/mg) is the Langmuir constant. e shape of the isotherm depends on R L , because this factor indicates the adsorption process as: Unfavourable (R L > 1) Linear (R L = 1) Favourable (0 < R L < 1) Irreversible (R L = 0). e nonlinear form of the Freundlich isotherm is wri en as: Whereas the linear form of the Freundlich isotherm equation can be wri en as [60,66]: Where K F is the calculated Freundlich equilibrium constant ([mg/g] [L/mg] 1/n ) and is an indicator of adsorption capacity, and n is a measure of the deviation from linearity of adsorption (g/L). Figure 2 shows the N 2 adsorption/desorption isotherms for UiO-66-Ca samples and UiO-66. Hysteresis in the desorption isotherm was distinguishably demonstrated by UiO-66-10%Ca which had a sharp increase in adsorption at relative pressures close to 0.999. is observation is strong evidence that the mesopore and macropore sizes were enhanced [67]. Table 1 Textural properties of the adsorbents based on N 2 adsorption/isotherm.

Characterisation
In addition, Table 1 presents the textural properties of all adsorbents, according to the calculations of the N 2 adsorption isotherm. e speci c surface area (S BET ) decreased with increasing content of a second metal. BET surface area in UiO-66 was 1585.50 m 2 .g -1 and then decreased to 918.115 and 557.68 m 2 .g -1 in UiO-66-10%Ca and UiO-66(Zr)-30%Ca respectively due to increasing the content of Ca in the synthesis process. e current BET values are acceptable when they compared with that in previous studies [68] .
However, the pore volume and average pore size were enhanced in the MOFs with the lowest content of the second metal. e highest pore volume and pore size were seen in UiO-66-10%Ca, which were 1.10 cc.g -1 and 2.39 nm, respectively. e results indicate that the addition of low concentrations of the second metal in the single-pot synthesis, followed by the activation process using the solvent exchange method, enhanced the pore volume and pore size replacing the second metal by methanol molecules which were discarded by the heating in the second stage of the activation process [69].  e results demonstrate that the integrity of the structure was maintained in activated samples, which indicates that the synthesis and activation procedures succeeded reliably without suspected impurities of a metal oxide inside the pores. e XRD pa erns of activated samples are similar to the XRD pa ern of UiO-66 in previous studies [46,70,71]. e pa erns of the samples a er using in the adsorption experiments in Figure 3a shows that UiO-66-30% Ca demonstrated higher stability than other samples because the pa ern of this sample displayed all peaks as same as those of activated samples. However, other samples were distinguished by the main characteristic peak in 2 theta of 7° while other peaks were signi cantly reduced. Figure 3(b) shows that the spectra of all samples, including that of UiO-66, exhibit the same vibration bands with slight deviations in the position of some peaks with increases in the content of a second metal. In addition, the peaks in the mixed-metal samples were broader than the peaks in the single-metal (Zr) sample, which indicates a difference in the dipole between ground state and excited state in the mixed-metal UiO-66 as a result of incorporating a second metal in the metal centre [72,73]. e vibration band of 1615-1580 cm -1 was a ributed to C=C-C stretching in the aromatic ring of terephthalate salts; however, this band extended from 1590 to 1525 cm -1 in the mixed-metal UiO-66 [74]. Further, the bands at 1500 and 1390 cm -1 were a ributed to the stretching vibrations of symmetric COOand asymmetric COOin coordinated organic linkers, as shown in the spectrum of UiO-66.
Moreover, the weak bands at 881, 812 and 785 cm -1 were assigned to Zr-O whereas the peak at 730 cm -1 in the UiO-66 spectrum was assigned to the stretching vibration of C-H and out-of-plane bending of aromatic ring in the main skeleton of UiO-66; this peak was shi ed to 744 cm -1 in the  [73,75]. In addition, the band at 1017 cm -1 belonged to C-H stretching in the MOF. Figure 3(c) presents the results of thermogravimetric analysis for all adsorbents in this study. All samples appear to have the same thermal stability, with structural stability at increasing temperatures up to 725 K. uptake at the commencement of the adsorption process is very rapid; a er an initial period of time, it proceeds at a slower rate until the saturation is a ained [76][77][78].
is phenomenon can be explained thus: the first available MB molecules are favourably adsorbed onto the most active sites of the single-metal and bimetal Zr-MOF, and the high initial MB uptake is possible because of the accessibility of many active sites. A longer contact time between the MOFs and MB results allows to increase the removal of MB until equilibrium adsorption capacity is reached [78]. Another explanation is that a higher initial concentration of MB provides more MB molecules and greater driving force of the aqueous phase (MB) against the solid phase (MOFs) to overcome mass-transfer resistance.
is fact gives rise to increase collisions between MB molecules and active sites on the adsorbent [79,80]. For instance, the adsorption capacity for MB onto UiO-66 at equilibrium increased from 2.151 to 14.837 mg.g -1 with increase in MB concentration from 5 to 50 mg. L -1 . Also, the adsorption capacity in UiO-66-Ca 10% and UiO-66-30%Ca was higher than that of UiO-66. It was reported that vacant metal sites in bimetallic MOFs are enhanced a er removing the second metal by the solvent exchange activation [81]. erefore, removing of Ca from UiO-66-Ca increased the active sites and consequently enhanced the adsorption capacity of MB [81]. e pseudo rst-order and pseudo second-order model were employed for the adsorption of MB onto UiO-66, UiO-66-10%Ca and UiO-66-30%Ca. e linear regression correlation, R 2 , was calculated to identify the model of best t; higher R 2 values mean a be er t for the experimental data. e results of the correlational analysis of the amount of adsorbed dye (mg/g) against contact time, for the various initial concentrations of MB (5, 15, 30 and 50 ppm) are shown in Figure 4. e results indicate that the amount of dye loading (q t [mg/g]) increases with contact time at each level of MB concentration. In addition, the amount of MB adsorbed increased with increasing in the initial MB concentration [47]. e kinetics of the adsorption process in the laboratory -based on batch experiments enables the prediction of the rate at which a pollutant is removed from bulk solutions, which informs the design of adsorption treatment plant columns [82]. However, the physical and chemical properties of the adsorbent signi cantly affect its adsorption kinetics, which in turn, affects the sorption mechanism. Statistics from kinetics studies of pseudo first-order and pseudo second-order kinetics model equations have been investigated for t with contact time data [76]. Tables 2 and  3 below present their main characteristics as calculated kinetic constants (k 1 , k 2 ) and correlation coefficients (R 2 ) for C i = 5, 15, 30 and 50 ppm.
According to the R 2 values obtained, they have been consistent and closer to unity for the pseudo second-order kinetic equation than for the pseudo first-order kinetic equation.
erefore, based on R 2 values, the sorption kinetics of MB removal using single-metal and bimetal Zr-MOF were well described by the pseudo second-order kinetic equation. Further, the calculated equilibrium adsorption capacity agreed with the experimental Applied Materials and Technology equilibrium adsorption capacity, further indicating that the sorption of aqueous MB onto single-metal and bimetal Zr-MOF perfectly obeyed pseudo second-order kinetics that indicates strong interactions happened between MB and active sites in UiO-66 [83]. Speci cally, the sorption of MB by single and bimetal Zr-MOFs occurred through chemisorption (the exchange or sharing of electrons between the sorbate and sorbent via covalent forces and ion exchange) [56,83].
Based on the mechanism underlying pseudo second-order kinetics, the effects of the initial concentration on the adsorption kinetics of MB onto the three MOFs (i.e., all the sorbent systems) were similar over time. UiO-66-10% Ca was taken to be a representative adsorbent and was used to explain the effects of the initial concentration on the rate of adsorption. Precisely, Table 2 shows that the adsorption rate constants (k 2 ) of Pseudo second-order model on UiO-66-10% Ca were higher than those on other adsorbents. Speci cally, k 2 on UiO-66-10% Ca was 0.86348, 0.07616, 0.04628 and 0.02259 g mg -1 .min -1 at initial MB concentrations of 5, 15, 30 and 50 mg.L -1 , respectively, signifying a decrease in adsorption rate at higher initial concentrations of MB. Reductions in the amount adsorbed at higher initial concentrations may be due to MB molecules having to enter the pores through a longer diffusion path. On the other hand, with less amounts of MB adsorbed, MB molecules tend to be rapidly adsorbed into the open pores of MOFs, which eventually increases the adsorption rate (K 2 ). Table 3 shows the kinetic constant of Pseudo rst-order model (K 2 ) and R 2 . K 1 of UiO-66-10% Ca was also higher than that of UiO-66 and UiO-66-30% Ca. For instance, it was 0.0101, 0.2669 and 0.0105 respectively at initial concentration of 5 ppm. e lower rate constant for MB adsorption onto the UiO-66-30% was tentatively ascribed to MB diffusion into the micropores of the MOF [84].

Intraparticle diffusion modelling study
A multistep adsorption process consists of the mass transfer of MB from the solution to the surface of single-metal and bimetal UiO-66; this transfer determines the extent of reaction throughout the whole adsorption process [85]. Adsorption process mechanism of MB onto MOFs can be arranged into the following three stages: 1. Film diffusion: the initial stage of rapid adsorption 2. Successive intraparticle diffusion: the second stage of the process during which the adsorption rate slows 3. e nal stage: the adsorption a ains equilibrium and lasting constant.
Film diffusion is very fast because of the rapid sorption of MB to the surface of the MOF. is stage is featured by quick surface mass transfer caused by a large differential which acts as a driving force. is stage is when the most is adsorbed by adsorbents, according to Weng et al. [86]. Such a nding establishes MOF-MB systems as entailing a fast adsorption process. Consequently, these adsorbent systems are favourable alternatives for removing cationic dyes from wastewater effluent. e second stage, intraparticle diffusion, is slower because the occupation of MB molecules on many of the available external sites in the rst step slows the diffusion of MB molecules into the pore spaces of the MOF [85].
e mechanism of MB sorption on the surface of MOF was investigated using contact time data. Speci cally, experimental data were ed to the intraparticle diffusion model (Equation 13) and the outcomes interpreted by plo ing q t versus t 1/2 in Figure 5. e most important aspects of the intraparticle diffusion plot are rst, the linear portion and the intercept of the plot (c), which indicates the effects of the boundary layer on the adsorption process. e second linear portion of the plot can be used to interpret intraparticle diffusion. e plot can be used to derive values for parameters, such as k p (the diffusion rate), C and R 2 , as presented in Table 4. e k p can be determined from the slope of the plot. e slope can be used to estimate the driving force of diffusion, which plays a critical role in the adsorption reaction. Experimental data analysis demonstrated that the k p values increased from 0.0991 to 0.638 mg.g -1 min -(1/2) with increases in the initial MB concentration from 5 to 50 mg.L -1 . erefore, higher initial concentrations of MB increase the driving force and subsequently increase the MB diffusion rate. Further, increasing initial MB concentrations over a similar range led to increases in the intercept value (C) from 0.5017 to 3.8144 mg.g -1 , suggesting that an initial high concentration of basic dye is associated with a stronger boundary layer effect in the sorption process. In addition, an increase in the intercept value (C) can indicate the availability of MB on the boundary layer of UiO-66.

Equilibrium modelling study
Recent research has revealed that initial concentration of MB has a detrimental effect on adsorption process. Initial concentration of MB plays a role in determining removal efficiency of MB (R %) and equilibrium adsorption capacity (q e ); indeed, the initial concentration of MB has profound consequences for R% and q e . e initial concentration of MB positively affected q e and negatively affected R% [60,78]. e observed decrease in MB removal (R% values of 43.03% to 29.67%) by UiO-66 samples was representative of the adsorption process in all systems and con rmed the occupation of all accessible active sites on the UiO-66 above a certain concentration of MB. However, the increase in equilibrium adsorption capacity (q e ) from 2.15 to 14.83 mg/ g can be a ributed to the higher adsorption rate and the use of all available active sites on UiO-66 samples for sorption at higher concentrations of MB. Equilibrium isotherms were examined using the Langmuir and Freundlich isotherms. e assumption of the Langmuir isotherm is monolayer coverage of sorbate over a sorbent with homogenous surface [5,63]. It assumes that the adsorption process occurs at speci c homogenous sites over the adsorbent; that is, when an MB molecule occupies a speci c site, additional sorption cannot happen again at the same site. Successful implantation of the Langmuir adsorption isotherm has been undertaken to explain the adsorption of basic dyes such as MB from aqueous solutions [66]. According to Freundlich model, the favourability of adsorption can be estimated by the magnitude of the exponent (1/n), which predicts the feasibility of the adsorption process. e values of n must be greater than one for conditions to be favourable for an adsorption process [5,87]. e constant n values of UiO-66, UiO-66-10%Ca and UiO-66-30%Ca have been calculated to be 1.29, 5.01 and 3.09, respectively. ese values con rm the favourability of adsorption of MB onto single-metal and bimetal Zr-MOF. e results of the correlational analysis for K F , n and the linear regression coefficient (R 2 ) for the plot of the linear form of the Freundlich model are presented in Table 5. Figure 6 illustrates the experimental equilibrium data and the predicted theoretical isotherms for the adsorption of MB onto single-metal and bimetal Zr-MOFs. It is apparent, from Figure 6 and the R 2 values in Table 5, that there is closer t between the experimental data and Freundlich isotherm compared to that with the Langmuir isotherm, at higher values of R 2 .
Analyses and calculations of the Langmuir and Further, Freundlich constants (K F ) related to the bonding energy of MB molecules with single-metal and bimetal Zr-MOFs were greater than Langmuir constants which were related to the affinity of MB molecules to single-metal and bimetal Zr-MOF in all cases. As a result, the adsorption of MB onto single-metal and bimetal Zr-MOF occurred as multilayer adsorption on a heterogeneous surface. e calculated maximum monolayer adsorption capacity (q m ) of Zr-MOF for MB is 50.25 mg/g for UiO-66-10%Ca, a relatively satisfactory adsorption capacity (see Table 4). According to Langmuir isotherm, the calculated results for the separation factor (R L ) are (0.89-0.44), (0.005 -0.0005) and (0.28-0.03) for UiO-66, UiO-66-10%Ca and UiO-66-30%Ca, respectively. R L values for the sorption of MB onto single-metal and bimetal UiO-66 are in the range of 0 < R L < 1, indicating that the adsorption was favourable. Further, higher initial MB concentrations in the adsorption process can make it irreversible [88]. Table 6 lists the maximum adsorption capacity (q m ) of the UiO-66 adsorbents in this study for MB, relative to those reported in the literature for different adsorbents of MB. e performance of UiO-66 in MB removal is relatively effective by comparison.

CONCLUSION
e main goal of the current study was to assess the adsorption capacity of MB in UiO-66, UiO-66-10%Ca and UiO-66-30%Ca. e Zr-MOFs were prepared according to a single pot solvothermal methods with a modi cation using trace additives of Ca. Compared with the UiO-66 without the modi cation; the textural properties of the modi ed UiO-66 were enriched while their performances were enhanced to remove MB from wastewater. e kinetics of MB sorption onto UiO-66, UiO-66-10% Ca and UiO-66-30% Ca were ed by the pseudo rst and second-order models. e second model offered the best t for the experimental data for all systems studied. e mechanism of MB sorption onto the surface of MOFs was investigated using contact time data. Speci cally, the ing of experimental data to the intraparticle diffusion model identi ed three stages in the sorption process.
Langmuir and Freundlich plot analyses and calculations revealed that the values of the linear regression correlation coefficient (R 2 ) for the Freundlich model were greater than those for the Langmuir model, for UiO-66, UiO -66-10%Ca and UiO-66-30%Ca. As a result, the adsorption of MB onto single-metal and bimetal Zr-MOFs was considered to occur as multilayer adsorption on a heterogeneous surface.
Langmuir maximum loading capacity (q m ) was compared with other reported adsorbents in previous studies. e values of the separation factor (R L ) indicated that the adsorption was a favourable process. Using the Freundlich linear model, constant n values for UiO-66, UiO-66-10%Ca, and UiO-66-30%Ca were found to be more than one (i.e., n > 1). ese values con rm the favourability of MB adsorption onto single-metal and bimetal Zr-MOF. is study can suggest the bimetallic UiO-66 as an a ractive adsorbent to remove dyes from wastewater.