Exploration and Production Technologies
Novel Fouling - Reducing Coatings for Ultrafiltration, Nanofiltration and Reverse Osmosis Membranes

DE-FC26-04NT15547

Project Goal 
The project goal is to provide new alternatives to purify oilfield produced water by removing emulsified oil, particulates, and dissolved solids to enable inexpensive beneficial use of produced water for agricultural and landscape irrigation and drinking water. Such beneficial use would significantly reduce injection costs associated with the current practice of disposing of produced water and provide a new source of purified water in what are often water-starved regions. This objective would be achieved by exploiting recent discoveries to dramatically improve the fouling resistance of polymer membranes. New, fouling-reducing membrane coatings could be developed and then applied to commercially available reverse osmosis (RO), ultrafiltration (UF), and nanofiltration (NF) membranes to reduce membrane fouling and markedly improve membrane lifetime for produced-water purification.

Performer
University of Texas, Austin, TX

Results
Initial work on this project has shown that commercial RO membrane performance for oil-in-water emulsion filtration falls far below that of pure water filtration, demonstrating the need for fouling-resistant materials. Thorough characterization of polyethylene glycol (PEG)-based hydrogels confirm the materials’ hydrophilic nature and support their use as a means to prevent adhesion by oil. Surface chemical analysis of RO membranes grafted with poly(ethylene glycol) diglycidyl ether (PEG diepoxide) indicates that PEG is present on the surface. Initial fouling studies of coated and grafted RO membranes suggest that these modified materials are more resistant to fouling than untreated commercial membranes. Based on crossflow results of anionic and cationic surfactants, focus has expanded to include surface charge as a significant variable in membrane fouling.

Benefits
The main benefit of the project would be the development of new coatings for membranes that will be more resistant—or completely resistant—to fouling and thus would result in better overall separation performance than that of uncoated membranes.

Background
In 2002, >14 billion barrels of produced water was handled in U.S. oil and gas operations. This water contains salts, heavy metals, emulsified oil, and other organics that render it unsuitable for human or animal consumption, agricultural use, or possibly direct discharge offshore. Often, the most viable disposal option today is subsurface injection, with costs of $0.50-1.75 per barrel, representing an annual operating cost of >$7 billion for U.S. oil and gas producers.

However, water is an extremely valuable commodity in the arid western and midwestern states. Produced water could provide a valuable water source of irrigation, industrial, and even potable water if the organic content and salinity could be reduced to acceptable limits.

When commercially available UF, RO, and/or NF membranes are exposed to mixtures of salt, emulsified oil droplets, and other particulate matter, their lifetime decreases catastrophically due to dramatic and largely irreversible permeate flux reduction, which causes fouling of the membranes by organic components (primarily from emulsified oil droplets). Fouling is the most significant roadblock to wider adoption of membrane technology for desalination specifically and water purification in general. This research proposes a novel approach to dramatically improve the fouling resistance of commercial UF, RO, and NF membranes by applying a very thin coating of fouling-resistant polymer to the surface of these membranes. The proposed research optimizes fouling-resistant polymer chemistry and its attachment to UF, RO, and NF membranes.

Summary
The project proposes a systematic experimental exploration of coating materials chemistry to optimize water throughput and resistance to fouling by organic components in produced water, such as emulsified oil droplets.

Flat-sheet AG RO membranes manufactured by GE Infrastructure Water Process and Technologies were obtained for characterization and modification studies. The AG membrane is sold commercially for brackish water desalination and is a polyamide thin film composite RO membrane. Flat-sheet membranes from Dow FilmTec were also obtained for use in characterization and modification studies. Focus has been centered on the LE (low energy) and XLE (extra-low energy) RO membranes.

This project focused on surface modification of commercial RO membranes by coating and grafting hydrophilic materials to the membrane surface. First, the commercial RO membranes, the LE and XLE by Dow FilmTec and the AG from GE, were thoroughly characterized. Crossflow testing conditions must be carefully controlled in order to measure accurate, reliable values of water flux and NaCl rejection in accordance with the manufacturer’s specifications. Feed pH and the use of prefiltration of the feed water were found to be critical variables in membrane testing. The LE/XLE and AG membranes must be tested under different conditions (pH 8 with unprefiltered feed and pH 7 with prefiltered feed, respectively) to obtain their best performance. Concentration polarization also must be accounted for to find the true salt rejection capabilities of the membranes.

Before applying coatings to membranes, thorough characterization of the coating materials was performed. Three series of PEG-based copolymers were systematically studied to relate chemical composition and structure to polymer properties such as water and NaCl permeability. Acrylic acid, 2-hydroxyethyl acrylate, or poly(ethylene glycol) acrylate were each copolymerized with poly(ethylene glycol) diacrylate to form a highly hydrophilic, crosslinked hydrogel. All of the copolymers exhibited large water uptake, with the PEGA copolymers having the largest uptake amounts. Water permeability was directly proportional to the water uptake; higher uptake materials also had higher water permeability, regardless of chemical composition. Ethylene oxide content and crosslink density were major contributors to water sorption and transport behavior. Increasing ethylene oxide content and decreasing crosslink density both promoted increased water transport. NaCl diffusion and partition coefficients were also measured. The salt transport properties were similar to the water transport properties; higher water uptake materials also had a larger salt uptake, and high water permeability lead to large diffusion coefficients and high salt permeability. Contact angle measurements confirmed the hydrophilic nature of the copolymer surfaces. All copolymer contact angles were less than the contact angle of a commercial RO membrane, indicating that the copolymer surfaces would be less conducive to oil adhesion.

Preparation of composite membranes using the synthesized hydrogels proved difficult. Progress was made in elucidating optimal coating conditions and techniques. A basic model was applied to use composite water flux to gauge coating thickness. SEM images confirmed the viability of this model. Initial crossflow fouling tests showed the coated membrane to foul less than an uncoated membrane.

Grafting of hydrophilic PEG molecules was performed by reaction of membrane surface amine groups with the epoxide endgroups of poly(ethylene glycol) diglycidyl ether (PEG diepoxide). XLE membranes were dip coated in heated solutions of PEG diepoxide for 10 minutes to ensure reaction with the membrane surface. Dead end water flux testing revealed the effect of PEG diepoxide molecular weight and concentration on surface coverage and pure water flux (more surface coverage, lower pure water flux). Crossflow testing of high water flux candidates demonstrated that PEG diepoxide-grafted XLEs have better fouling resistance than their unmodified counterparts (LE and XLE membranes).

Current Status (December 2008)
This project is completed. The final report is listed below under "Additional Information".

Funding
The project was selected under DE-PS26-04NT15460-02, Produced-Water Management.

Project Start: September 1, 2004
Project End: August 31, 2008

Anticipated DOE Contribution: $576,000
Performer Contribution: $115,000

Contact Information 
NETL – Jesse Garcia (jesse.garcia@netl.doe.gov or 918-699-2036)
University of Texas - Benny Dean Freeman (freeman@che.utexas.edu or 512-232-2803)

Additional Information

Final Project Report [PDF-992KB]
StayConnected Facebook Twitter LinkedIn RssFeed YouTube