Amine-terminated zirconium metal-organic framework as an efficient adsorbent for lanthanum removal

1 Introduction

Water is the most basic form of resource that sustains life on Earth, but the pollution of water by toxic metallic pollutants is one of the most persistent environmental problems of our century. Due to the upsurge of global industrialization, speedy urbanization practices, and intensification occurring on agricultural lands, hazardous metal ions are increasingly being released into water streams, namely, groundwater, surface water, and marine systems (Mitra et al., 2023; Obuotor et al., 2024). Heavy metals and rare earth elements are among the most lethal varieties of contaminants. Mining, processing, and industrial use of rare earth elements produce metal-containing effluents of concern (Mubarak et al., 2025; Guo et al., 2023). The mining and hydrometallurgical treatment of rare earth elements produce large volumes of acidic leachate streams and tailings that contain dissolved lanthanide ions. If these are not treated before discharging, they will contaminate nearby water bodies and ecosystems (Li et al., 2024). The recovery of rare earth elements from such secondary streams like industrial process water, acid mine drainage, and electronic waste leachates offers a double dividend; reducing pollution whilst recovering valuable strategic materials (Li et al., 2024; Bhadane et al., 2024).

Conventional methods used for the removal of rare earth elements from wastewater include chemical precipitation, coagulation-flocculation, ion exchange, solvent extraction, electrochemical methods, and membrane filtration (Wang et al., 2021; Adeleke et al., 2024). All of these approaches have their benefits but share significant limitations that restrict their effectiveness being used. Adsorption is considered as the most promising approach for the removal of heavy metals and rare earth elements because of its conceptual simplicity, low capital cost, operational flexibility, and efficiency in treating dilute solutions. Most conventional adsorbents, such as activated carbons, zeolites, clays, silica gels, etc., have relatively low adsorption capacities, poor tunability, and poor selectivity in multi-ion systems (Ali et al., 2012; Wen et al., 2024). With the development of metal-organic frameworks (MOFs), a new class of porous adsorbents has emerged and that competes with the zeolites. This class of materials is much more versatile, both structurally and chemically. MOFs have high surface areas (1,000–7,000 m²/g) and micro or mesoporous structures that make them a unique adsorbent (Furukawa et al., 2013; Magida et al., 2025). This means that various properties of a MOF materials can be altered in ways that would not be possible with a conventional porous materials. Some of the most widely studied MOFs for adsorption include UiO-66 family (Cavka et al., 2008), Cr- and Fe-based MIL-101 frameworks, and Zn-base ZIF-8 and ZIF-67 frameworks (Magida et al., 2025; Zhang et al., 2025) due to their high chemical and thermal stability.

In the past decade, the literature on MOFs for metal ion adsorption processes has grown. Strategies aimed at the functionalization of these MOFs, such as copolymer grafting, silica hybridization, phosphonic acid decoration, and core-shell nanostructure assembly, have been shown to significantly increase their adsorption capacity and selectivity for specific rare earth elements (Li et al., 2024; Bhadane et al., 2024). It is worth noting that composite systems like the 3D Agaric-like U6N@ZIF-8-20 MOF and phosphonic acid-functionalized ZIF-67@SiO₂ exhibited adsorption capacities close to or over 400 mg/g for lanthanide ions (Li et al., 2024). Zirconium-based MOFs (Zr-MOFs, UiO-66) are the most well-studied materials for aqueous-phase heavy metals and rare earth elements adsorption, upon which remarkable works have been done owing to their great chemical, thermal, and mechanical stability (Kim et al., 2024; Lei et al., 2023). For this reason, UiO-66 has become a benchmark family for environmental remediation studies (Chakraborty et al., 2024; Solís et al., 2022).

The ability to add a variety of functions to UiO-66 is one of its main advantages as a metal adsorbent. The adsorption of Cr(VI) by amino groups on UiO-66-NH₂ is significantly enhanced (36.6 mg/g). The enhancement is attributed to electrostatic attraction and a coordination bond between the protonated amine and chromate anion (Chen and Lv, 2023). The adsorption capacities of chitosan-modified CS-UiO-66-NH₂ composites for Pb(II) can reach up to 555.56 mg/g (Hu et al., 2023). In the field of arsenic remediation, defect-engineered Zr-MOFs with glycine functionality, namely, Gly@UiO-66, exhibited high As(III) capacities up to 301.4 mg/g, which is amongst the highest reported values (Razavi et al., 2024). Methods that combine materials such as Fe₃O₄@UiO-66 enable magnetic recovery of the adsorbent for the removal of Se(IV) (Li et al., 2023). The Zr-MOFs functionalized with carboxyl and ethylenediamine groups, UiO-66-COOH-ED in particular, exhibited a 4.9-fold higher adsorption capacity for Gd³⁺ compared to the pure UiO-66 (Ahmed et al., 2019). Overall, the above studies suggested that the UiO-66 can be versatile, and water-stable adsorbent whose capacity can be tuned broadly for many metal ions through functionalization.

The main importance of this research lies in the incorporation of–NH₂ groups into the Zr-based UiO-66 framework, which generates chemically reactive sites capable of interacting with La³⁺, while the porous structure provides accessible adsorption areas. The solitary pair of electrons on nitrogen can engage with La³⁺, facilitating distinct chemical bonding alongside pore adsorption. The adsorption of La³⁺ may involve both the amino-functionalized organic linker and Zr–O/OH sites, resulting in various binding conditions. The protonation and deprotonation of–NH₂, along with La³⁺ hydrolysis, helps elucidate the mechanistic rationale for the ideal adsorption pH. The Zr–O clusters in UiO-66 confer significantly enhanced structural stability compared to numerous traditional MOFs, hence rendering aqueous La³⁺ separation more feasible. Even though, UiO-66-NH₂ is used for the removal of different metal ions, the removal of La³⁺ is scarce in the literature.

In this work, an amine-terminated zirconium MOF (UiO-66-NH₂) was synthesized and utilized as an effective adsorbent for the removal of lanthanum (La³⁺) from an aqueous solution. The adsorption process was optimized using a response surface methodology (RSM) model. Furthermore, we examined how several factors, such as pH, adsorbent dosage, and reaction time, impacted the removal of La³⁺ from the aqueous solution.

2 Experimental

2.1 Synthesis of UiO-66-NH₂

The synthesis of UiO-66-NH₂ was carried out based on a prior publication (Kavya et al., 2023). 0.04 g of ZrCl₄ and 0.03 g of 2-aminoterephthalic acid were dissolved in 20 mL DMF with continuous stirring. To this, 5 mL of acetic acid was added and stirred for 30 min. The resulting solution was moved to a solvothermal reactor and heated to 120° C for 16 h. After completion of the reaction, the product was centrifuged and washed repeatedly with alcohol and vacuum-dried at 80° C for 12 h.

2.2 Batch adsorption experiment

Experiments were conducted in 20 mL vials to examine the impact of different conditions on the removal of La³⁺. Each vial contained 10 mL of a 100 ppm lanthanum (III) nitrate solution. The study examined the effects of variables including pH levels from 3 to 11, reaction durations from 25 to 150 min, and adsorbent quantities of 3–10 mg per 10 mL of solution. Subsequent to the reactions, the UiO-66-NH₂ adsorbent was isolated from the mixture via 10 min centrifugation at 5000 rpm. Samples were collected at various time intervals, and the residual concentration of La³⁺ was measured. The quantity of La³⁺ adsorbed onto the material (mg/g) was subsequently determined using the following formula (Equation 1):

qt = C 0 − C t V m (1)

C _*0* - initial concentration of La³⁺, C _*t* - concentration of La³⁺ (mg/L) at time t. V - volume of adsorbate (L), and m - mass of the adsorbent (g). The percentage removal of La³⁺ from the aqueous solution was determined using the following Equation 2:

Removal % = C 0 – C t / C 0 × 100 (2)

3 Results and discussion

3.1 Characterization

In order to remove La³⁺, we have successfully synthesized UiO-66-NH₂ and the schematic illustration is presented in Figure 1. The X-ray diffraction pattern (XRD) of UiO-66-NH₂ is displayed in Figure 2a. The UiO-66-NH₂ showed a high-intensity peak between 5° and 30°, indicating the excellent crystalline nature. The (111) (200), and (006) planes of UiO-66-NH₂ (COD–4348,132) were identified as the main diffraction peaks at 7.4, 8.6, and 25.7°. The outcomes were in good agreement with the literature report (Kavya et al., 2023; Gokila et al., 2024). We also included the simulated XRD pattern in Figure 2a.

The morphology of the as-synthesized UiO-66-NH₂ is depicted in Figure 2b, which was seen using a Field emission scanning electron microscope (FESEM). There were multimodal spherical particles of varying sizes that were seen in the UiO-66-NH₂ sample. The average size of the particles was less than 200 nm. Nevertheless, the particles were clumped together in a number of different locations. The N₂ adsorption/desorption isotherm was utilized in order to determine the Brunauer–Emmett–Teller (BET) surface area and Barrett-Joyner-Halenda pore size of UiO-66-NH₂. The graphs that resulted from these measurements are presented in Figures 2c, d. The UiO-66-NH₂ exhibited a type I isotherm and the pore size distribution provided evidence of hierarchical porosity. The computed values for the surface area, pore size, and pore volume were 682 m²/g, 1.79 nm, and 0.27 cm³/g, respectively. These values indicate that the material is microporous. However, pore size of 2.9 nm was also present in a very smaller fraction.

These values are in agreement with the range reported in the literature for solvothermally synthesized UiO-66-NH₂. Luu et al. reported a BET surface area of 873–876 m²/g with a pore volume of 0.379 cm³/g and a pore diameter of 1.9 nm for UiO-66-NH₂ (Luu et al., 2015). Cao et al. reported a BET surface area of 822 m²/g and a pore volume of 0.236 cm³/g for their UiO-66-NH₂, with pore sizes ranging between 1.6 and 2.2 nm (Cao et al., 2018). Yin et al. reported a surface area of 1,112 m²/g for pristine UiO-66-NH₂ (Yin et al., 2018). Kazemi et al. reported a surface area of 876 m²/g and pore volume of 0.38 cm³/g for UiO-66-NH₂ (Kazemi et al., 2023). Reported BET values of UiO-66-NH₂ are mostly within the range of ∼600–1,200 m²/g with pore volumes of 0.14–0.56 cm³/g and pore sizes in the micro-to small-mesoporous range (∼0.6–2.2 nm), depending on the synthesis route and activation conditions (Sultana et al., 2025). The properties of our as-synthesized UiO-66-NH₂ are in this reported range, confirming the successful generation of a porous, crystalline framework for La³⁺ adsorption.

3.2 Quadratic model

The Box-Berman model (BBM) was applied to improve the adsorption efficiency of La³⁺ with UiO-66-NH₂. Table 1 shows the experiment consisting of 13 trials and their corresponding adsorption results. A quadratic polynomial equation was developed to relate the removal efficiency to the adsorbent dosage (x₁) and reaction time (x₂). The experimental model describing La³⁺ removal is expressed by the following equation in terms of these process variables (Equation 3):

% removal of La 3 + = 7.62 x 1 + 6.29 x 2 + 1.97 x 1 x 2 + 8.35 x 1 2 + 2.24 x 2 2 (3)

TABLE 1. BBM design for La³⁺ adsorption using UiO-66-NH₂.

The equation illustrates how different variables affect the ability of UiO-66-NH₂ to adsorb La³⁺. The adsorption of La³⁺ reached nearly its highest possible level. To determine the most suitable model for this situation, experimental data was fitted to various models, as shown in Supplementary Table S1. Each model’s significance was assessed and the quadratic model was found to be significant. The cubic model was found to be aliased, meaning its parameters could not be independently determined.

An analysis of variance (ANOVA) confirmed the utility of the quadratic model, which was developed to define the connection between the outcome variables. As presented in Table 2, the ANOVA considered primary effects, interaction effects, and error components. The statistical significance of these components within the quadratic model was evaluated using p-values and F-values. Additionally, mean squares (MS), degrees of freedom (DF), and sums of squares (SS) were calculated. The F-value was derived by dividing MS by DF, and MS was obtained by dividing SS by DF, as per the methodology outlined by Sudamalla et al. (2012). The quadratic model demonstrated statistical significance, evidenced by an F-value of 6.44 and a p-value below 0.1, indicating its high significance.

TABLE 2. Analysis of variance for the quadratic model.

According to Table 2, x₁, x₁₂, and x₂₂ were recognized as significant factors due to their low p-values. Factors were excluded from this study because they had little influence on La³⁺ adsorption capacity but were statistically significant. This is because they had larger p-values. The coefficient of determination (R ²) might be used to assess the predictive capability of this model. With an R ² value of 0.82, the model explained 85.9% of the total variability and leaving 14.1% unexplained in this case. Due to the adjusted coefficient of determination, R² _adj = 0.69 < R ², we can say that the fit is less precise. Figure 3 highlights that the observed values of La³⁺ adsorption were in accordance with the predicted line as the data points cluster around the regression line. The observation suggested that the studied experimental parameters needed a quadratic model for effective prediction of La³⁺ removal.

3.3 Effect of process variables on La³⁺ adsorption

One further critical factor that influences the adsorption of La³⁺ is the solution’s pH. By maintaining a constant concentration of 100 ppm of La³⁺ and 10 mg/10 mL of adsorbent for 150 min, we were able to see how the pH of the solution affected the La³⁺ adsorption. We conducted the studies at a range of pH levels, from 3 to 11. Figure 4a shows that when the pH increased from 3 to 7, the percentage of La³⁺ removal fell steadily from 98.5% to 75.21%. One possible explanation is that in acidic conditions, the -NH₂ groups in the ligand can accept a proton to produce–NH₃ ⁺, which results in a positive surface charge for the UiO-66-NH₂. Additionally, the amount of La³⁺ removed is decreased because hydrogen ions in solution compete with La³⁺ for adsorption sites on the surface of UiO-66-NH₂. As the pH increased above 7, the electrostatic interaction between the positively charged La³⁺ and the negatively charged UiO-66-NH₂ may be the main adsorption process, which would explain why the effectiveness of La³⁺ removal increases as the pH rises. The optimal pH for La³⁺ elimination was thus determined to be 9.

An important variable in this study is the adsorption dose. To investigate the impact on La³⁺ adsorption, the adsorbent dosage was adjusted between 3 and 10 mg/10 mL. In order to study this effect, the following parameters were used: pH 9, 100 ppm of La³⁺, and 150 min of time. Figure 4b shows that the percentage of La³⁺ removed rose from 64% to 99.86% as the amount of UiO-66-NH₂ increased from 3 to 10 mg. More active sites become accessible for adsorbate as the dosage of the adsorbent increases, which directly impacts the adsorption capacity. The equilibration of La³⁺ adsorption onto the UiO-66-NH₂ has already occurred after 10 mg.

It is a common finding in adsorption study that the adsorption of adsorbate improves with increasing contact time. This is likely because there is more surface area available at the start for the adsorption of La³⁺ on the active sites. The rate of adsorbate transit from the surface to the inside of the adsorbent controls the uptake rate as the surface adsorption sites are used up. The pH remained steady at 9, the La³⁺ concentration was 100 ppm, and 10 mg/10 mL of adsorbent were kept constant. Time intervals ranging from 25 to 150 min were used to study the influence of contact time. The percentage of adsorption increased from 27.8% to 99.86% (Figure 4c) as the time increased from 25 to 150 min. Following that, no notable alteration was observed. Therefore, 150 min was chosen as the contact time for the subsequent investigation. Figure 4d represents the 3D surface mapping plot of reaction time and adsorbent dosage on removal of La³⁺ by considering pH and adsorbate concentration constant. The pH, amount of adsorbent, and duration of the La³⁺ adsorption process were the most important variables. Results showed that 10 mg of adsorbent, a pH of 9, and 150 min of contact time were the best conditions for La³⁺ removal. We also compared our results with the previously reported adsorbents for La³⁺ removal and tabulated in Table 3.

TABLE 3. Comparison of La³⁺ adsorption with previous reports.

3.4 Adsorption isotherm

Two isotherm models were used to study the distribution of La³⁺ in the equilibrium liquid/solid interphase. While the Langmuir isotherm explains monolayer adsorption on a solid surface and the Freundlich isotherm depicts multilayer adsorption with a random distribution (Haldorai et al., 2015; Nandi et al., 2009; Haldorai and Shim, 2014). At a pH of 9, the adsorption capabilities of La³⁺ were tested. The experiment involved equilibrating a UiO-66-NH₂ adsorbent with an adsorbate concentration of 100 ppm for 150 min using 10 mg of the adsorbent. The Langmuir (Equation 4) and Freundlich (Equation 5) models were used to study the equilibrium isotherms.

C e q e = 1 K L q m + C e q m (4)

ln ⁡ q e = ln ⁡ K F + 1 n ln ⁡ C e (5)

In this context, C _*e* represents the concentration of La³⁺ (mg/L), q _*e* denotes the quantity of La³⁺ adsorption (mg/g), q _*m* indicates the maximum adsorption capacity, n signifies the heterogeneity coefficient, while K _*L* and K _*F* are the Langmuir and Freundlich constants associated with adsorption capacity and the adsorbent’s affinity for La³⁺, respectively.

Figure 5a illustrates that the Langmuir isotherm plot of C _*e* versus C _*e* /q _*e* produced a linear relationship with a R ² value of 0.9916. The q _*m* , and K _*L* values were determined to be 312.5 mg/g and 0.057, respectively. The Freundlich isotherm (Figure 5b) exhibited a R ² value of 0.9986 for the log C _*e* versus log q _*e* plot. The n values of 1.98 corroborated the efficient recovery of La³⁺ and suggested advantageous adsorption. The linear plot indicated that the adsorption of La³⁺ onto the UiO-66-NH₂ adhered to the Freundlich isotherm. The findings suggest that the as-synthesized UiO-66-NH₂ may serve as an exceptional adsorbent for the elimination of La³⁺. The adsorption efficacy of UiO-66-NH₂ is attributable to two primary factors: (i) its microporous characteristics, which encompass a substantial surface area and significant pore volume, and (ii) the electrostatic interaction between nitrogen atoms and La³⁺, wherein the nitrogen atoms within the MOF serve as donor groups, effectively attracting La³⁺ due to their high electronegativity.

3.5 Adsorption kinetics

The Weber–Morris model was employed to better examine the intraparticle diffusion mechanism (Equation 6).

q t = K diff t 1 / 2 + C (6)

Where, K _*diff* is the intraparticle diffusion rate constant (mg/g/min^1/2), q _*t* is the amount of adsorbate adsorbed at time t (mg/g), and C is the intercept related to the boundary-layer thickness (mg/g).

The Weber–Morris plot illustrated in Figure 5c displayed a single linear line, suggesting that a diffusion-dominated mechanism significantly influences the adsorption kinetics throughout the examined timeframe. The derived linear correlation produced a K _*diff* value of 7.14 mg/g/min^1/2 and a positive C value of 12.8 mg/g. The non-zero intercept signifies that the regression line did not intersect the origin, illustrating that intraparticle diffusion is not the exclusive rate-limiting factor. The favourable C value may be linked to boundary-layer resistance, indicating that both external mass transfer and intraparticle diffusion play a role in the comprehensive adsorption mechanism. Nonetheless, the single linear line suggested that no clear transition between multiple diffusion regimes was seen within the study timeframe.

The kinetics of the adsorption process was investigated by analyzing the experimental data with both the pseudo first-order model (Fakhri, 2017) and the pseudo second-order model (Rengaraj et al., 2017). The pseudo first-order kinetic data can offer insights into the adsorption mechanism and the elements that influence the process’s pace, including mass transfer and chemical reactions. In contrast, the pseudo second-order model indicates that chemisorption, characterized by the formation of chemical bonds through electron exchange between the adsorbate and adsorbent, may be the rate-limiting step. The mathematical representations for the first-order (Equation 7) and second-order (Equation 8) rate laws are provided subsequently.

log q e − q t = log q e − k 1 2.303 t (7)

t q t = 1 k 2 q e 2 + t q e (8)

Where, k _*1* and k _*2* signify the rate constants for the Lagergren pseudo first-order (min^-1) and pseudo second-order adsorption (g/mg/min) processes, respectively. The variable t denotes the time elapsed during adsorption (min), and q _*t* and q _*e* represent the amount of substance adsorbed at time t and at the point of equilibrium, respectively.

An examination of the Lagergren pseudo first-order model produced a straight-line graph when log (q _*e* −q _*t* ) was plotted against t, achieving an R ² value of 0.9827. For the pseudo second-order kinetics, the model was evaluated by plotting t/q _*t* versus t, from which the q _*e* and k₂ were derived from the slope and intercept, respectively. The experimental conditions were set at an initial adsorbate concentration of 100 ppm, 10 mg of adsorbent, a pH of 9, and a contact time of 150 min. The quality of each kinetic model was assessed based on the R ² value. It was found that the pseudo second-order model was a better fit for La³⁺ adsorption due to a higher R ² value of 0.9971 (Figure 5d). The values of q _*e* and k _*2* were calculated as 212.8 mg/g and 8.6 × 10⁻⁴ g/mg min. The findings imply that surface reaction was the critical factor affecting the reaction speed, while mass transfer effects were negligible.

3.6 Selectivity

Selectivity studies were conducted utilizing 50 mL of 100 ppm La³⁺ spiked seawater containing 10 mg of UiO-66-NH₂ adsorbent. After 150 min of agitation, the adsorbent was recovered, and the solution was analyzed using an inductive coupled plasma mass spectrometer (ICP-MS) to determine the initial and residual La³⁺ concentrations. After each cycle, 5 mL of 0.1 M nitric acid was added to extract the La³⁺ from the adsorbent. The recovered adsorbent was then washed with excess water. The efficiency of the adsorbent in removing La³⁺ in seawater containing competing cations, such as Na⁺ (∼9,000 ppm) and Mg²⁺ (∼1,000 ppm) was used to assess its selectivity for La³⁺ (Figure 6). Since the concentration of La³⁺ is always much lower than that of coexisting cations like Na⁺ and Mg²⁺, the adsorbent UiO-66-NH₂ exhibited remarkable selectivity toward La³⁺. The adsorption procedure was repeated for five times, with the first cycle recovering 87.5% of La³⁺, 1.1% of Na⁺, and 0.5% of Mg²⁺ from seawater. There were no significant differences in La³⁺ recovery between the second and third cycles. The concentrations of Na⁺ and Mg²⁺ were 1.8% and 1.2% in the second cycle, and 2.4% and 1.4% in the third cycle, respectively. The fourth and fifth cycles saw a very small decrease in La³⁺ recovery to 85%. The Na⁺ and Mg²⁺ concentrations were 2.9% and 1.9% in the fourth cycle, and 3.2% and 2.2% in the fifth cycle, respectively. The results showed that the presence of coexisting cations had no substantial effect on the adsorption of La³⁺.

4 Conclusion

In conclusion, we successfully synthesized UiO-66-NH₂ adsorbent that showed a high La³⁺ adsorption efficiency. FESEM revealed a spherical morphology, while XRD verified the adsorbent’s crystalline nature. The BBM was utilized to enhance the adsorption efficacy of La³⁺ using UiO-66-NH₂, and the quadratic model was determined to be relevant. The adsorption process was mainly dependent on the primary variables, such as the dosage of the adsorbent and the duration of contact. In the first 50 min, the adsorption capacity rose at a rate of 63% according to the experimental data. The subsequent 150 min saw a steady rise in removal efficiency, reaching 99.86%. The 2D porous adsorbent’s high surface area and the electrostatic attraction between La³⁺ and nitrogen atoms in the ligand are likely responsible for its outstanding adsorption ability. Adsorption and kinetic studies followed Freundlich isotherm and pseudo second-order reaction, respectively. The results revealed that while intraparticle diffusion significantly influences the adsorption process, it is not the sole rate-determining factor. Based on the results of the ICP-MS study, it was determined that the UiO-66-NH₂ had the ability to selectively adsorb La³⁺ from seawater in comparison to competing ions like Na⁺ and Mg²⁺.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmats.2026.1944659/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Yun Kong, Yangtze University, China

Reviewed by: Rabelani Mudzielwana, University of Venda, South Africa

Selçuk Demir, Recep Tayyip Erdoğan University, Türkiye