(152n) Cyclodextrin-Modified Layered Double Hydroxide Thin-Film Nanocomposite Desalination Membrane for Boron Removal.

Boron is a naturally occurring element present in various water sources, including seawater. However, excessive levels of boron in drinking water can have negative impacts on human health. Hence, the elimination of boron from water sources is crucial. Recently, layered double hydroxide (LDH) membranes have been introduced for desalination purposes, and they have exhibited promising outcomes in eliminating boron from water sources [1-3]. This article examines the modification techniques of LDH for boron removal and the challenges linked with the development and application of LDH thin film nanocomposite (TFN) membranes. The study concludes that incorporating β-cyclodextrin (CD)-functionalized LDH into TFN membranes has significant potential for removing boron from brackish water.

To investigate the impact of CD-intercalated Mg₂Al-LDH on boron removal in desalination membranes, a 0.2 M CD solution was utilized to prepare the intercalated material. Different concentrations of CD (0.03, 0.06, 0.10 & 0.15 % w/v in trimesoyl chloride phase) were incorporated into the membrane, and the amount of CD intercalated was determined using thermogravimetric analysis (TGA). The same amount of CD was used for functionalization. The solvent mixture of hexane/xylene (95/5 % v/v) was optimized for the preparation of the desalination membrane, resulting in improved permeability. All membrane samples were post-treated with anhydrous ethanol for 5 min and stored in ultrapure water overnight before further characterization. The membrane samples were labeled iLDH_003, iLDH_006, iLDH_010, and iLDH_015, according to their respective loading of CD-intercalated Mg₂Al-LDH. Similarly, the membrane samples loaded with CD-functionalized Mg₂Al-LDH were labeled fLDH_003, fLDH_006, fLDH_010, and fLDH_015.

All of the analyzed membrane samples exhibited effective retention of salt exceeding 98 %, with membranes iLDH_003 and fLDH_010 both achieving an even higher level of 99.4 % (as demonstrated in Fig. 3). Conversely, the membrane that contained pristine CD at a 0.10 % loading did not demonstrate an enhanced salt rejection rate when compared to the control membrane. This finding has noteworthy implications, as it suggests that the combination of CD and Mg₂Al-LDH works in a synergistic manner to improve the desalination performance. As anticipated for iLDH and fLDH membranes, there is a clear negative correlation between the quantity of particles loaded and the flow rate of water through the membrane. Water permeance declined significantly from 2.20 LMH bar^-1 in iLDH_003 to 1.16 LMH bar^-1 in iLDH_015, showing that iLDH membrane permeabilities were substandard. On the other hand, water permeance for fLDH membranes only slightly decreased from 2.81 LMH bar^-1 in fLDH_003 to 2.53 LMH bar^-1 in fLDH_015, however, they sustained relatively high water permeance throughout.

The boron rejection efficiency of control membrane was found to be 71.3 % (Fig. 4). Conversely, membranes that were incorporated with LDH particles exhibited an improvement in their boron rejection efficiencies. The membrane variant fLDH_010 provided the highest boron rejection rate of 82.3 %, which then decreased to 80.2 % in the fLDH_015 membrane. This could be corroborated to the possible formation of agglomerated LDH particles within the interlayer of the polyamide membrane. On the other hand, the highest boron rejection efficiency offered by the iLDH incorporation was only 77.4 % in the iLDH_006 membrane. The enhancement in salt and boron retention relative to the control may be primarily attributed to the reduced pore size of the membranes, as evidenced by the diminished water permeance in iLDH membranes. Furthermore, the elevated water permeance rate in fLDH membranes may be ascribed to the expanded interlayer spacing, which generates additional water transport channels, as well as membrane surface wetting resulting from the incorporation of hydrophilic LDH particles. However, it is also essential to examine the adsorptive impact of LDH to determine its role in augmenting boron rejection efficiency. For the further investigations, 0.10 % was chosen as the optimal loading of LDH particles.

TFN membranes with 0.10 % LDH particle loading was chosen for long-term desalination stability assessment. It can be seen from Fig. 5 that the salt retention performance of both iLDH_010 and fLDH_010 were maintained (≥ 98.8 % in iLDH_010 and ≥ 99.1 % in fLDH_010) during the continuous desalination over 5 h. In contrast, there is a noticeable difference in the ability to reject boron between iLDH_010 and fLDH_010 membranes, with the latter demonstrating significantly greater selectivity. Both fLDH_010 and iLDH_010 membranes display an enhancement in their ability to reject boron over time, which may be due to the continuous compaction of the membranes at high pressure.

Fig. 6 illustrates the outcomes of an investigation on the boron adsorption isotherm of LDH particles and LDH-incorporated thin-film nanocomposite membranes. Both iLDH and fLDH particles effectively adsorbed boron with isotherms that were well-suited to both Langmuir and Freundlich. The Langmuir adsorption model provided high R² values of 0.995 (iLDH) and 0.985 (fLDH), which estimated maximum adsorption capacities of 27.0 mg g^-1 and 89.5 mg g^-1, respectively. The Langmuir constants were utilized to calculate ∆G^o (iLDH) = -14.8 kJ mol^-1 and ∆G^o (fLDH) = -13.8 kJ mol^-1, indicating that both LDH particles spontaneously adsorbed boron in salt solution. The thermodynamically favorable adsorptions were also supported by the computed 1/n (from Freundlich model), with values (0.69 for iLDH, and 0.78 for fLDH) that were significantly less than unity. The study examined the adsorptive effect of iLDH_010 and fLDH_010 membranes, but there was minimal static adsorption observed due to the insignificant amount of particles in the membranes. The membranes were exposed to 15 mg L^-1 of boron and subjected to "boron desorption" through sonication in 1 M HCl and 1 M NaOH solutions. The boron contents in the combined supernatant were analyzed, and only 0.18 mg L^-1 and 0.41 mg L^-1 of boron were detected from the desorption of iLDH_010 and fLDH_010 membranes, respectively. Therefore, it can be concluded that adsorption has a negligible contribution to boron rejection during desalination.

The study also examined the kinetics of boron adsorption for both iLDH and fLDH particles. The results displayed in Fig. 7 indicate that iLDH particles reached adsorption saturation much faster (within 30 minutes) compared to fLDH particles (around 120 minutes). However, fLDH particles had a higher adsorption capacity than iLDH particles, even at shorter durations. To determine the rate-determining step, the boron adsorption kinetics were analyzed using both pseudo-first-order and pseudo-second-order models. The results showed that boron adsorption by iLDH particles fit better with the pseudo-first-order kinetic model, with a high R² value of 0.981, indicating that physisorption is the rate-determining step. On the other hand, boron adsorption by fLDH particles fit best with the pseudo-second-order kinetic model (R² = 0.995), although both kinetic models had high R² values (R² = 0.985 for the pseudo-first-order model). This suggests that the rate-determining step of boron adsorption by fLDH particles is mainly due to chemisorption.

Several studies have been conducted to investigate the factors affecting boron rejection with RO membranes. It has been found that the boron rejection rate is influenced by several factors, including the feed water pH, the feed water boron concentration, the membrane type and properties, and the operating conditions [4,5]. In this study, it has been reported that a high boron rejection rate can be achieved without pH adjustment (i.e. pH 8), even at high feed water boron concentration (i.e. 15 mg L^-1). In conclusion, while high boron rejection rates with RO membranes can be challenging, several strategies have been investigated to overcome this issue by introducing selective boron adsorbent nanofiller within the polyamide TFN interlayer.

(1) Sajid, M., Sajid Jillani, S. M., Baig, N., & Alhooshani, K. (2022). Layered double hydroxide modified membranes for water treatment: Recent advances and prospects. Chemosphere, 287, 132140. https://doi.org/10.1016/j.chemosphere.2021.132140

(2) Zhao, Q., Zhao, D. L., Feng, F., Chung, T. S., & Chen, S. B. (2022). Thin-film nanocomposite reverse osmosis membranes incorporated with citrate-modified layered double hydroxides (LDHs) for brackish water desalination and boron removal. Desalination, 527, 115583. https://doi.org/10.1016/j.desal.2022.115583

(3) Zhao, Q., Zhao, D. L., Chung, T. S., & Chen, S. B. (2023). In-situ growth of layered double hydroxides (LDHs) onto thin-film composite membranes for enhanced reverse osmosis performance. Desalination, 547, 116235. https://doi.org/10.1016/j.desal.2022.116235

(4) Han, D., Hwang, M., & Kim, I. S. (2017). Effect of boron rejection and recovery rate on a single-pass design of SWRO using hybrid membrane inter-stage design (HID) concept. Desalination, 404, 215–223. https://doi.org/10.1016/j.desal.2016.11.007

(5) Ruiz-García, A., & Nuez, I. (2021). Performance evaluation and boron rejection in a SWRO system under variable operating conditions. Computers & Chemical Engineering, 153, 107441. https://doi.org/10.1016/j.compchemeng.2021.107441