The objective of this study was to develop a mechanistic model that predicts boron rejection performance under varying operating conditions such as pH and temperature with the goal of developing a tool for process design and operation optimization with higher boron removal. Bench scale cross-flow filtration experiments were performed to evaluate the rejection of boron as well as that of major ionic species (Na, Cl, Ca, Mg and SO₄) SR (Saehan Industries, Inc., Korea) seawater reverse osmosis (SWRO) membrane. The test unit was equipped with plat-and- frame test cells (73 mm length ? 38 mm width ? 5 mm height). Feed water stored in a 40 L polypropylene tank was circulated and pressurized through the cells by a positive-displacement high-pressure pump (Hydra-Cell D10S, Wanner Engineering, Minneapolis, MN). Feed pressure to the cell was controlled by a needle valve (Swagelok, Solon, OH) located downstream of the cells. To prevent the over-pressurization of the system, pressure regulating valve (C22AB, Wanner Engineering, Minneapolis, MN) was installed next to the pump outlet. The valve can regulate pressure by automatically bypassing a part of the feed flow if the system pressure exceeds the preset pressure. Bypass and feed flow rate was measured by hydraulic flow meter (King, Atlanta, GA). Pressure was measured with analogue gauges (Swagelok, Solon, OH) before and after the test cells. A temperature controller (Polystat, Cole parmer, Vernon Hills, IL) circulated the water through the heat-exchange coil immersed inside the feed tank to control the temperature. All experimental components were made of stainless steel 316 and/or Teflon® to avoid corrosion. The experiments were carried out with synthetic seawater (10,500 mg/L Sodium, 19,000 mg/L Chloride, 1,350 mg/L Magnesium, 450 mg/L Calcium, and 2,700 mg/L Sulfate) spiked with 5 mg/L of boron, following an orthogonal matrix of varying operating pressures (600 to 1,000 psi), pHs (6.2 to 9.5), and temperatures (15 to 30 ºC). Boron, sodium, calcium and magnesium concentrations of feeds and permeates were measured using Inductively Coupled Plasma Mass Spectrometer-Atomic Emission Spectrophotometer (Model ICAP 61E Trace Analyzer, Thermo Jarrell Ash, Franklin, MA). Chloride and sulfate concentrations were measured with a Dionex DX-600 ion chromatography system (Sunnyvale, CA) installed with IonPac AS16 analytical column (4 mm x 250 mm). Includes 10 references, figure.