Completed Research
Mathematical Modeling of Flowback Water Treatment Using Reverse
Osmosis
and Nanofiltration Technologies
Pedro Henrique Casa Grande Rosa, MS, 2016
(Thesis: University of Pittsburgh ETD)
The main objective of this study is to assess, through mathematical modeling, the potential use and feasibility of deploying nanofiltration and reverse osmosis technologies in the treatment of flowback water. Field data of flowback water flow rates and chemical composition were used in the models in order to provide an accurate assessment of each technology. Operating conditions based on the current commercial reverse osmosis and nanofiltration membranes for water treatment were also considered. Mathematical models for the reverse osmosis and nanofiltration processes were developed to assess the performance of these processes in the treatment of flowback water produced during the hydraulic fracturing for natural gas production from shale plays. The models, based on the mass balance and thermodynamics, were verified and implemented in Matlab version R2015.
The models were used to perform a sensitivity analysis for the two processes in order to determine the effect of the operating variables on the membrane performance in terms of solute concentration and filtration time. For the reverse osmosis, it was found that pressure drop, inlet flow rate and membrane area were the major parameters governing the process. For nanofiltration, on the other hand, pressure drop, reflection coefficient and membrane area were the most important parameters affecting the process performance.
The models were also used to assess and compare the performance of four different commercial reverse osmosis and three nanofiltration membranes using actual field data, such as inlet flowrate and flowback water composition. The predictions of the two models showed that the reverse osmosis was significantly superior to the nanofiltration membranes in the removal of Na+ and Ca2+. Nanofiltration membranes, however, exhibited higher removal efficiencies for Cl- than that of the reverse osmosis membranes. This behavior was attributed primarily to the nature of both processes; since the reverse osmosis is mainly driven by the chemical potential of chlorine, whereas, the nanofiltration is controlled by the molecule size.