Energy and water shortages are serious hand-in-hand challenges we are facing in 21st century. As the water demands increase exponentially with the expansion of human territory in all aspects, the current energy source that we heavily rely on, the fossil fuel, is going to be depleted in the next several decades. As a result, the economic efficiency of the seawater and brackish water desalination, the only two sectors that could increase water supply for us, has spearheaded in the water purification related research and technologies. Forward osmosis (FO), a natural process in which water naturally permeate from a low salt concentration feed solution to a high salt concentration draw solution due to osmotic pressure difference, has provided a smart approach to lower the energy cost of reverse osmosis processes. Through this process, both fresh water and renewable energy are able to be produced through salinity gradient, provided that appropriate membrane and system configuration can be developed. However, the efficiency of FO membranes is severely inhibited by an intrinsic bottleneck—internal concentration polarization (ICP); that is, when water diffuses through the FO membranes, the salt concentration inside the tortuous support layers of FO membranes is dramatically diluted. This causes the detrimental loss of osmotic driving force, resulting in substantial decline of water production rate. Many endeavors have been attempted, but none have delivered a sufficient approach to address this ICP problem. In this work, we designed a novel nanofiber composite FO (NC-FO) membrane, which has a highly porous nanofiber support layer with inter-connected and low-tortuosity pore structure. This design provides direct paths for salt and water diffusion in the support layers. This NC-FO membrane provides an efficient answer for elimination of the ICP bottleneck, showing high water production rate of 37.8 L m-2h-1 using 0.5 M NaCl as draw solution. In another application of the FO membrane, namely the pressure retarded osmosis (PRO) process, a hydraulic pressure (less than osmotic pressure of draw solution) is applied against the draw solution to harvest the salinity energy in seawater reverse osmosis (SWRO) brine. The FO membrane used in the PRO process, which generally requires much higher mechanical strength but lower solute rejection, is addressed as the PRO membrane. The crucial component to determine the energy recovery rate of the PRO process is the semi-permeable PRO membrane. In this part, we report the fabrication and optimization of thin-film nanofiber composite PRO (TNC-PRO) membranes with unique support membrane structure (inter-connected, low tortuousness and highly porous properties), aiming to overcome the ICP problem. With low structure parameter (S) value (150 μm) nanofiber support membranes (NSMs), the optimum water permeability (A) and solute permeability (B) of TNC-PRO membrane for power generation was determined to be 4.1 L m-2h-1Bar-1 and 1.74 L m-2h-1 respectively. This highly efficient TNC-PRO membrane can achieve a power density of 15.2 W m-2 and maximum energy recovery of 0.86 kWh m-3, using synthetic brackish water (80 mM NaCl, π=3.92 bar) and seawater brine (1.06 M NaCl, π=51.8 bar) as feed and draw solution, respectively. For a more diluted synthetic river water (0.9 mM NaCl, π=0.045 bar) feed solution, the same membrane can achieve a higher power density of 21.3 W m-2. The main performance limiting factors in PRO application such as ICP, External Concentration Polarization (ECP) and Reverse Solute Permeation (RSP) are quantified and their values are related with A, B and S values of TNC-PRO membranes. Furthermore, the influence of hydrophilicity, additives, post treatment, and non-woven substrate on the water permeability coefficient (A) and solute permeability coefficient (B) of the TNC membrane was systematically investigated. It was found that the nanofiber FO membranes fulfilled an empirical relationship of B=k*A3 with fitting coefficient k=0.025 bar3 m4 h2 L-2. The permeability coefficients were related to ICP index (I) and ECP index (E), which are proposed in this study to quantify the performance limiting factors, ICP, ECP and RSP. It was found that in a FO process (i.e., active layer facing draw solution), loss of effective osmotic pressure difference (eff) to coupled effect of convective dilution and low mass transfer coefficient in support membrane (i.e. D/S value) is the most severe, resulting in lower JWFO. In PRO process (i.e. active layer facing feed solution), loss of eff to coupled effect of convective dilution and higher mass transfer coefficient in bulk solution (i.e. k value) is less severe, resulting in higher JWPRO. Loss of eff to coupled effect of solute leakage and low mass transfer coefficient in support membrane (i.e., D/S value) is minor. To conclude, the NC-FO membranes have been systematically investigated focusing on the theory, fabrication, optimization, characteristics and its application. The results of NC-FO membranes presented in this thesis re-invents the wheel of osmotic pressure-driven membranes and shed a light on overcoming the hurdle of ICP towards higher efficiency production of clean water and renewable energy using salinity gradient. Doctor of Philosophy (CEE)