Forward Osmosis as an Approach to Manage Oil Sands Produced Water: Membrane Fouling and Organic Removal

• Author / Creator

  Zhu,Shu

• Currently, large amounts of oil sands process-affected water (OSPW) are stored in tailing ponds, leaving its environmental impact a significant concern. To better manage OSPW, numerous treatment approaches have been investigated including adsorption, advanced oxidation and biological treatment, among others. Forward osmosis (FO) as an emerging membrane desalination technology is gaining increasing research interests. FO utilizes only the osmotic difference between two solutions to draw the water molecules from the less concentrated side to the concentrated side, which potentially reduces the energy consumption and eliminates the membrane fouling. To date, the application of FO has already been evaluated on water and wastewater treatment, especially for the oil and gas wastewater. The overall objective of the current research is to better understand the feasibility of using FO in the treatment of OSPW in terms of membrane fouling and organic removal. In the first set of investigations, FO was proposed to manage OSPW, using on-site waste basal depressurization water (BDW) as the draw solution. To investigate its feasibility, both short and long-term OSPW desalination experiments were carried out. By applying FO process, the volume of OSPW was decreased >40% and high rejections were achieved, especially, the major organic toxicity source — naphthenic acids (NAs). Although comparable low water flux (≥3 L/m2 hr) was obtained, water flux caused by membrane fouling can be completely recovered using clean water backwash. Moreover, calcium carbonate precipitation was observed on OSPW-oriented membrane side and with respect to flux decline, FO (active layer facing feed solution) and PRO (support layer facing draw solution) mode did not demonstrate a significant difference on anti-fouling performance. The advantages provided by this approach include zero draw solution cost, less and reversible membrane fouling and beneficial reuse/recycle of diluted BDW. In the second set of investigations, the effects of pH and draw solutions on the rejection of NA model compounds including cyclohexane carboxylic acid (CHA), 1-adamantaneacetic acid (AAA) and the refined Merichem mixture of NAs in forward osmosis were studied. The rejection behavior of CHA and AAA were pH-depend (from pH = 3 to 9), which further suggested that electrostatic repulsion was the dominant rejection mechanism. The rejection efficiency of Merichem NAs was maintained above 95%, which was not affected by the pH range from 6 to 9. A decline trend was observed on water flux using AAA and Merichem NAs as feed solution and surface fouling on Merichem NAs rejected membrane was confirmed by scanning electron microscopy (SEM) analysis. Four inorganic salts — sodium chloride (NaCl), ammonia chloride (NH4Cl), sodium sulphate (Na2SO4), and calcium chloride (CaCl2) — were introduced as the draw salts and no significant difference was found between the these draw solutes regarding the CHA rejection. Furthermore, the reverse salt diffusions for the draw solutes remained stable, except CaCl2. The reverse salt flux along with the water decline indicated that using CaCl2 as the draw solution caused membrane surface precipitation. Our experimental results also suggested that the exposure of NA model compounds might alter the membrane characteristics, which needs to be further investigated. In the third set of investigations, aquaporin (AQP)-based and cellulose triacetate (CTA)-based FO membranes, used to treat oil sands produced water, were compared in terms of membrane characteristics, NA model compounds adsorption and rejection, membrane fouling, and NA and inorganic salt rejections. Results of the nano-filtration (NF) test indicated the AQP membrane had higher water permeability than the CTA membrane. At pH = 9, each of the membranes showed low adsorption and high rejection of NA model compounds, due to the electrostatic forces between the negatively charged membrane surface and the model compounds. Our study demonstrated that CTA membrane was more anti-fouling in treating OSPW. OSPW fouling associated with the presence of calcium was observed on both membranes, evidencing the possibility of cake enhanced concentration polarization. In addition, it was seen that the rejections of NAs and salt were more related to pretreatment methods and produced water type than membrane selection. In the last set of investigations, to understand better the fouling mechanism in FO in terms of membrane materials and functional groups in NAs, which are the main source of organic concentration in OSPW, the direct force measurement was conducted using surface force apparatus (SFA) between three membranes including cellulose triacetate (CTA)-FO, aquaporin (AQP)-FO and polyamide (PA)-RO membrane and three functional groups including carboxyl, hydroxyl and hydrophobic functional group. Moreover, the adsorption phenomenon of the tested membranes using two NA model compounds (cyclohexanecarboxylic acid and cholic acid) was investigated and the FO fouling test using 0.45 μm filtered OSPW as the feed solution was also performed. In the force measurement, only repulsive force was observed on -OH and -COOH functional group regardless of membrane types. It was also found that the adhesive forces caused by hydrophobic-hydrophobic interaction and compared to the other two membranes, AQP-FO exhibited the strongest hydrophobic interaction. The results from OSPW fouling experiment showed that AQP-FO suffered a more severe flux decline, which supported the observations in the interactive force analysis and subsequently suggested that the AQP-FO membrane might be easier to cause membrane fouling, comparing two other membranes.

• Subjects / Keywords

  • Forward osmosis
  • Oil sands process-affected water
  • Membranes

• Graduation date

  Fall 2017

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/R3WW77D38

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.