Boron removal from geothermal waters by ceramic membrane process

In recent years, the growing interest in clean and renewable energy sources has made the sustainable use of geothermal resources important. Additionally, the increasing demand for clean and sustainable water sources has highlighted the need for effective treatment technologies for geothermal waters with high boron (B) content. Geothermal waters are emerging as an alternative source in areas such as electricity generation, district heating, and agricultural applications. However, the high B content of these waters poses environmental and agricultural hazards when discharged without proper treatment. Excess boron negatively affects agricultural productivity by causing soil structure degradation, groundwater contamination, and plant toxicity. This thesis aims to develop an innovative and cost-effective method for B removal from geothermal waters using a tubular ceramic ultrafiltration (fine-UF) membrane with a pore size of 1 kDa. The main objective of the study is to evaluate B removal from model and real geothermal waters using polyol-supported (mannitol, sorbitol, and xylitol) membranes and to determine the optimal operating conditions. Ceramic membranes offer significant advantages over polymeric membranes due to their superior chemical, thermal, and mechanical durability in high-temperature geothermal water applications. However, the limited boron retention capacity of existing ceramic membranes restricts the widespread use of these technologies. To overcome this limitation, a hybrid approach using polyol compounds that can form stable complexes with boron is proposed. Polyol compounds increase their molecular size by forming complexes with boric acid (H₃BO₃), enabling them to be more effectively retained by ceramic membranes. The thesis study was conducted in four main phases. In Phases I and II, the performance of ceramic membranes was evaluated using model boron and model geothermal water samples; in Phases III and IV, polyol-supported boron removal was investigated using model and real geothermal water samples. Filtration experiments conducted in Phases I and II were performed at different operating pressures (1, 2, and 4 bar) and pH levels (natural and pH 9.5). The results obtained showed that the boron removal efficiency of ceramic membranes alone ranged between 0-11% (model geothermal water) and that these levels did not meet irrigation water standards. In Phase III and Phase IV experiments, laboratory-scale cross-flow ceramic membrane tests were conducted; membrane pressure (2 and 4 bar), feed water pH (natural and 9.5), and different boron concentrations (10 and 20 ppm B-model geothermal water) using model and real geothermal water samples (18 ppm B). The data obtained showed that boron removal efficiency varied significantly depending on the pH of the solution, the applied pressure, and the type of polyol used. In general, sorbitol provided higher boron removal under natural pH conditions, while mannitol was more effective under pH 9.5 conditions; xylitol showed moderate performance under both conditions. In the sorbitol-added model geothermal water, the highest boron removal of 41% was achieved at a boron concentration of 20 ppm, pH 9.5, and 4 bar pressure. Similarly, in model water containing mannitol (10 ppm B, pH 9.5), 38% removal was achieved, reducing the B concentration in the filtrate to ~6 ppm. This value is compatible with the recommended irrigation water limits for tolerant plants (4.0-6.0 ppm B: alfalfa, purple flowered plant, parsley, red beet, sugar beet, tomato) and highly tolerant plants (6.0-10.0 ppm B: sorghum, cotton, celery). The highest boron removal efficiency for actual geothermal water was recorded at 34% with sorbitol at 4 bar and pH 9.5. However, boron concentrations in the treated water remained between ~12-14 ppm in actual water samples, failing to meet agricultural irrigation standards (≤10.5 mg/L). In filtration tests, the filtrate flow rate was affected by operating pressure and pH. The highest permeate flux was measured at 38 LMH in model geothermal water containing mannitol at 4 bar pressure and pH 7.5, and at 29 LMH in real geothermal water containing sorbitol at 4 bar pressure and pH 7.4. This study quantitatively demonstrates for the first time that the boron retention efficiency of ceramic UF membranes, which have very low boron removal capacity (0-11%) under specified conditions, can be increased up to 41% by adding polyol. Thus, while preserving the structural advantages of the membranes, functional selectivity has been imparted. However, the boron concentration in the treated real geothermal water (~12 ppm) remains above the limits prescribed for agricultural irrigation. This indicates that the proposed approach alone is not sufficient; however, it shows that multi-stage treatment systems can be an effective pre-treatment option before advanced treatment technologies such as reverse osmosis. Polyols increased boron removal by forming stable complexes with borate ions, but did not cause membrane fouling. SEM and SEM-EDX analyses confirmed the structural integrity and pore homogeneity of the ceramic membranes. In the economic evaluation, the use of xylitol, in particular, increased operating costs. However, the long-lasting and clog-resistant structure of ceramic membranes suggests that they may offer advantages in geothermal applications in the long term. These findings highlight the potential of ceramic membranes and polyol use for boron removal; however, they also indicate the necessity of additional treatment steps to meet irrigation standards. Future studies should aim to improve process efficiency while targeting environmental and economic sustainability.