CA2985244A1 - Membrane separation of emulsions produced from hydrocarbon recovery process - Google Patents

Abstract

A membrane for separating water and oil emulsion produced from a hydrocarbon recovery operation, the membrane including a permeable metal substrate and a nanotetrapodal oxide coating on the permeable metal substrate.

Description

MEMBRANE SEPARATION OF EMULSIONS PRODUCED FROM
HYDROCARBON RECOVERY PROCESS
FIELD
[0001] The present disclosure relates to the separation of emulsions produced from a hydrocarbon recovery process.
BACKGROUND

[0002] Extensive deposits of viscous hydrocarbons exist around the world, including large deposits in the northern Alberta oil sands that are not susceptible to standard oil well production technologies. The hydrocarbons in such deposits are too viscous to flow at commercially relevant rates at the temperatures and pressures present in the reservoir. For such reservoirs, thermal techniques may be utilized to heat the reservoir to mobilize the hydrocarbons and produce the heated, mobilized hydrocarbons from wells. One such technique for utilizing horizontal wells for injecting heated fluids and producing hydrocarbons is described in U.S. Patent No. 4,116,275, which also describes some of the problems associated with the production of mobilized viscous hydrocarbons from horizontal wells. Thermal in-situ techniques may include steam-assisted gravity drainage (SAGD), expanding solvent steam-assisted gravity drainage (ES-SAGD), cyclic steam stimulation (CSS), stearnflooding, solvent-assisted cyclic steam stimulation, toe-to-heel air injection (THAI), or a solvent aided process (SAP).

[0003] In the SAGD process, pressurized steam is delivered through an upper, horizontal, injection well, into a viscous hydrocarbon reservoir while hydrocarbons are produced from a lower, generally parallel, horizontal, production well that is near the injection well and is vertically spaced from the injection well. The injection and production wells are typically situated in the lower portion of the reservoir, with the production well located close to the base of the hydrocarbon deposit to collect the hydrocarbons that flow toward the base of the deposit.

[0004] The SAGD process is believed to work as follows. The injected steam initially mobilizes the hydrocarbons to create a steam chamber in the reservoir around and above the horizontal injection well. The term steam chamber is utilized to refer to the volume of the reservoir that is saturated with injected steam and from which mobilized oil has at least partially drained. As the steam chamber expands, viscous hydrocarbons in the reservoir and water originally present in the reservoir are heated and mobilized and move with aqueous condensate, under the effect of gravity, toward the bottom of the steam chamber. The hydrocarbons, the water originally present, and the aqueous condensate are referred to collectively as produced emulsion. The produced emulsion accumulates such that the liquid / vapor interface is located below the steam injection well and above the production well. The produced emulsion is collected and produced from a production well.

[0005] Separation of the water and oil in the produced emulsion is carried out to increase the efficiency of the hydrocarbon extraction and to meet regulatory requirements for the treatment of wastewater. The separated water may be reused to generate steam and the hydrocarbons treated for sale. The separation of the water from the hydrocarbons may include several processes that are capital intensive.

[0006] Improvements in the separation of produced emulsions are desirable.
SUMMARY

[0007] According to one aspect, a membrane is provided for separating a water and oil emulsion produced from a hydrocarbon recovery operation. The membrane includes a permeable metal substrate and a nanotetrapodal oxide coating on the permeable metal substrate.

[0008] According to another aspect, a process is provided for separating a water and oil emulsion produced from a hydrocarbon recovery operation. The process includes disposing a membrane that includes a permeable metal substrate having a nanotetrapodal oxide coating thereon, at a tilt angle. The process also includes applying the emulsion to the membrane to produce a roll-off fraction including at least a portion of the produced water in the emulsion and a permeate fraction that passes through the membrane. The permeate fraction includes at least a portion of the oil in the emulsion. The permeate fraction is oil-rich and the roll-off fraction is water-rich. The membrane may be disposed at a tilt angle before or after the emulsion is applied to the membrane.

[0009] According to yet another aspect, there is provided a method of producing a membrane for separating a produced water and oil emulsion from a hydrocarbon recovery operation. The method includes heating a metal foil in the presence of oxygen to a temperature sufficient to cause oxidation of the metal foil and the production of nanotetrapodal oxide structures, dispersing the nanotetrapodal oxide structures in a solvent to provide a dispersion solution, and coating a permeable metal substrate with the dispersion solution.

[0010] Other aspects and features of the present disclosure will become apparent to those of ordinary skill in the art upon review of the following description of specific embodiments in conjunction with the accompanying figures.
BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Embodiments of the present application will now be described, by way of example only, with reference to the attached Figures, wherein:

[0012] FIG. 1A is a flowchart illustrating a method of producing a membrane for separating a water and oil emulsion produced from a hydrocarbon recovery operation;

[0013] FIG. 1B shows a method of producing zinc oxide nanotetrapods;

[0014] FIG. 1C shows a method of producing membranes for separating a water and oil emulsion, the membranes including a nanotetrapodal oxide coating on a permeable metal substrate;

[0015] FIG. 2A is a flowchart illustrating a process for separating a water and oil emulsion utilizing the membrane including a nanotetrapodal oxide coating on a permeable metal substrate;

[0016] FIG. 28 is a schematic illustration of a membrane configuration;

[0017] FIG. 3, FIG. 4, FIG. 5, and FIG. 6 are field emission scanning electron microscope (FE-SEM) images of a stainless steel mesh with a ZnO
nanotetrapodal coating;

[0018] FIG. 7 is a Raman spectrum acquired for a ZnO nanotetrapodal coating on a stainless steel mesh;

[0019] FIG. 8 is an x-ray diffraction (XRD) pattern for the ZnO
nanotetrapodal coating on the stainless steel mesh;

[0020] FIG. 9 is a stereomicroscopy image showing contact of water on a stainless steel mesh;

[0021] FIG. 10 is a stereomicroscopy image showing contact of water on the stainless steel mesh with a ZnO nanotetrapodal coating;

[0022] FIG. 11, FIG. 12, and FIG. 13 are a sequence of stereomicroscopy images showing the wettability of hexadecane on a stainless steel mesh with a ZnO nanotetrapodal coating;

[0023] FIG. 14 is a schematic illustration of an experimental configuration utilized to test the separation of water and hexadecane emulsions;

[0024] FIG. 15 is a representative graph illustrating water purity of a roll-off fraction, and thus the separation efficacy, as a function of effective length of a membrane;

[0025] FIG. 16 is a bar graph illustrating the purity of the roll-off fraction, and thus the separation efficacy, for various samples along an effective length of membrane of about 21 cm;

[0026] FIG. 17 is a photograph showing the interaction of sales oil with stainless steel mesh;

[0027] FIG. 18 is a photograph showing the interaction of sales oil with stainless steel mesh having a first ZnO nanotetrapod loading;

[0028] FIG. 19 is a photograph showing the interaction of sales oil with stainless steel mesh having a second ZnO nanotetrapod loading;

[0029] FIG. 20 is a photograph showing the interaction of sales oil with stainless steel mesh having a first ZnO nanotetrapod loading and having been functionalized by a first concentration of a fluorinated silane;

[0030] FIG. 21 is a photograph showing the interaction of sales oil with stainless steel mesh having a first ZnO nanotetrapod loading and having been functionalized by a second concentration of a fluorinated silane;

[0031] FIG. 22A is a photograph showing the interaction of reconstituted emulsion with diluent with stainless steel mesh having a first ZnO
nanotetrapod loading; and

[0032] FIG. 22B is a photograph showing the resulting roll-off and permeate from the interaction of reconstituted emulsion with diluent with stainless steel mesh having a first ZnO nanotetrapod loading as shown in FIG.
22A.

[0033] FIG. 23 is a graph illustrating % water content of the permeate and membrane permeation temperature as a function of pore size with varying ZnO
nanotetrapod loading.

[0034] FIG. 24 is a series of graphs, (A) showing a 3D plot of flux rate as a function of pore size and ZnO loading; (B) showing a 2D plot of % water content as a function of pore size at different ZnO loadings; (C) showing a 3D plot of %
water content as a function of pore size and ZnO loading; and (D) showing a 2D

plot of flux rate as a function of pore size at different ZnO loadings.
DETAILED DESCRIPTION

[0035] It will be appreciated that for simplicity and clarity of illustration, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements. In addition, numerous specific details are set forth in order to provide a thorough understanding of the embodiments described herein. However, it will be understood by those of ordinary skill in the art that the embodiments described herein may be practiced without these specific details. In other instances, well-known methods, procedures and components have not been described in detail so as not to obscure the embodiments described herein. Also, the description is not to be considered as limiting the scope of the embodiments described herein.

[0036] As described above, a steam-assisted gravity drainage (SAGD) process may be utilized for mobilizing viscous hydrocarbons. In the SAGD
process, a well pair, including a hydrocarbon production well and a steam injection well are utilized.

[0037] During SAGD, steam is injected into the injection well to mobilize the hydrocarbons and create a steam chamber in the reservoir. In addition to steam injection into the injection well, light hydrocarbons, such as the C3 through C10 alkanes, either individually or in combination, may optionally be injected with the steam such that the light hydrocarbons function as solvents in aiding the mobilization of the hydrocarbons. The volume of light hydrocarbons that are injected is relatively small compared to the volume of steam injected.
The addition of light hydrocarbons is referred to as a solvent aided process (SAP). Alternatively, or in addition to the light hydrocarbons, various non-condensing gases, such as methane or carbon dioxide, may be injected. Viscous hydrocarbons in the reservoir are heated and mobilized and the mobilized hydrocarbons drain under the effect of gravity. The produced emulsion, which includes the mobilized hydrocarbons along with produced water, is collected in a generally horizontal segment of the production well.

[0038] Separation of the water and oil in the produced emulsion is carried out to increase the efficiency of the hydrocarbon extraction and for the treatment of wastewater.

[0039] Membranes and demulsification systems for water and oil separation, based on differential affinity, density, flow characteristics, and wettability, utilize polymers and have limited viability at the high temperatures and pressures utilized in hydrocarbon extraction processes as such membranes are susceptible to degradation and fouling.

[0040] Known filtration systems are also not suitable for separating emulsions that contain a variety of different droplet sizes of one phase dispersed within the second phase. Such systems are typically utilized to separate oil and water droplets based on droplet size differentials. The pore size of the filter in such systems must be smaller than the smallest droplet size contained within the emulsion, which is difficult for submicron size droplets of emulsified oil. At high pressures, the droplets may deform and pass through the filter, substantially degrading the separation efficiency. Use of filters with submicron-sized dimensions, however, requires unrealistically high pressure gradients and yield extremely low liquid fluxes.

[0041] According to the present disclosure, a membrane for separating water and oil emulsion produced from a hydrocarbon recovery process includes a permeable metal substrate and a nanotetrapodal oxide coating on the permeable metal substrate.

[0042] The inorganic membrane including a permeable metal substrate and a nanotetrapodal oxide coating is utilized for separating the water and oil components of emulsions based on the differential wettability of the two liquids on the textured surface. The permeable metal substrate may be a metal or metal oxide substrate, for example a ceramic, a sintered metal, or a metal mesh.
The permeable metal substrate may be a metal mesh substrate of stainless steel, aluminum, brass, bronze, copper, polytetrafluoroethylene coated stainless steel, galvanized low alloy steel, nickel-coated low alloy steel, and an acid-resistant nickel. The metal mesh substrate may be of from about 60 gauge to about 500 gauge. For example, the metal mesh substrate may be 316 stainless steel mesh of from about 150 gauge (104 microns) to about 500 gauge (30 microns), 304 stainless steel mesh of from about 150 gauge (104 microns) to about 500 gauge (30 microns), aluminum mesh of up to about 200 gauge (74 microns), brass wire mesh of up to about 100 gauge (152 microns), bronze wire mesh of up to about 325 gauge (43 microns), copper mesh of up to about 200 gauge (76 microns), polytetrafluoroethylene coated 304 stainless steel mesh of up to about 325 gauge (43 microns), or acid-resistant nickel mesh of up to about 200 gauge (74 microns). The nanostructured oxide coating may be a coating of one or more of ZnO, A1203, MgO, Fe203, Fe304, Si02, Ti02, V205, Zr02, Hf02, Mo03, or W03.

[0043] Optionally, a further functional layer, for example, a silane layer, such as n-octadecyltrichlorosilane may be disposed on the nanotetrapodal oxide coating. Alternatively or in addition to silane, one or more of a phosphonic acid, a carboxylic acid, a sulfonate, an alcohol, a thiol, and an amine may be disposed on the nanotetrapodal oxide coating.

[0044] Referring now to FIG. 1A, a method of producing a membrane for separating water and oil emulsion produced from a hydrocarbon recovery operation is illustrated. The method may contain additional or fewer processes than shown or described.

[0045] Nanotetrapodal oxide structures are formed at 102. The nanotetrapodal oxide structures may be created by heating a metal foil in the presence of oxygen at a heating rate of about 43 C/min, or at a heating rate of about 10 C/min to about 1,000 C/min, and to a temperature of about 900 C to about 950 C to oxidize the metal foil and form crystalline oxide nanotetrapodal structures (as shown in FIG. 18). The metal foil may be, for example, zinc, aluminum, magnesium, iron, silicon, titanium, vanadium, zirconium, hafnium, molybdenum or tungsten. The nanotetrapodal oxide structures may be ZnO, A1203, MgO, Fe203, Fe304, Si02, Ti02, V205, ZrO2, Hf02, Mo03, or W03. An acceptable temperature range may be from about two thirds of the melting point of the metal foil to about three times the melting point of the metal foil.

[0046] The nanotetrapodal oxide structures are dispersed in a solvent such as 2-propanol, to provide a dispersion solution at 104.

[0047] A permeable metal substrate is coated by applying the dispersion solution to the permeable metal substrate at 106. As indicated above, the permeable metal substrate may be a metal mesh substrate may be stainless steel, aluminum, brass, bronze, copper, polytetrafluoroethylene coated stainless steel, galvanized low alloy steel, nickel-coated low alloy steel, or an acid-resistant nickel. For example, the metal mesh substrate may be 316 stainless steel mesh of from about 150 gauge (104 microns) to about 500 gauge (30 microns), 304 stainless steel mesh of from about 150 gauge (104 microns) to about 500 gauge( 30 microns), aluminum mesh of up to about 200 gauge (74 microns), brass wire mesh of up to about 100 gauge (152 microns), bronze wire mesh of up to about 325 gauge (43 microns), copper mesh of up to about 200 gauge (76 microns), polytetrafluoroethylene coated 304 stainless steel mesh of up to about 325 gauge (43 microns), or acid-resistant nickel mesh of up to about 200 gauge (74 microns). The nanotetrapodal oxide structures may blow through metal mesh of greater size during application of the nanotetrapodal oxide structures to the substrate and thus, a size of 500 gauge (30 microns) or less is advantageous for the application of the nanostructure oxide coating.

[0048] The permeable metal substrate may be any suitable shaped membrane. For example, the permeable metal substrate may be cylindrical, semi-cylindrical, or may be any other suitable shape. In a particular example, the substrate is a stainless steel pleated filter cartridge.

[0049] The permeable metal substrate may be coated by spray coating the dispersion solution while heating the permeable metal substrate to provide a nanotetrapodal oxide coated substrate. The heating temperature may be from about half to about two times the boiling point of the solvent used for coating the substrate.

[0050] Alternatively, the permeable metal substrate may be coated utilizing a Plasma-Enhanced Atomic Layer Deposition (PE-ALD) or other Chemical Vapour Deposition (CVD) process to grow the nanotetrapodal oxide structures on the substrate, which may be a mesh substrate.

[0051] Optionally, the nanotetrapodal oxide is further adhered to the substrate by a modified StOber method wherein tetraethylorthosilicate (TEOS) is used as the precursor for a conformal amorphous SiO2 coating (TEOS-derived) at 107. The use of a ceramic ZnO/SiO2coating may allow for compatibility with high-temperature operations and may further lend mechanical resilience to the coating. Ceramic refers to the higher thermal stability and mechanical resilience that may be provided by the coating compared to other polymeric or metallic materials. Different configurations of SiO2 coatings may be utilized, for example:
(1) a configuration wherein the TEOS precursor solution for generating Si02 is applied onto a stainless steel mesh prior to spray-coating of the ZnO

nanotetrapods; (2) a configuration wherein the ZnO nanotetrapods are dispersed in the TEOS precursor solution for generating Si02 and the mixture is directly coated onto the stainless steel mesh; and (3) a configuration wherein the TEOS-derived Si02 layer is sprayed after already having deposited the ZnO
nanotetrapods onto the stainless steel mesh (topcoat). S102 loading of about 3.9 pL/cm2 was tested. A range of Si02 loadings may be suitable, for example, of from about 2 pL/cm2 to about 400 pL/cm2.

[0052] Optionally, the nanotetrapodal oxide coated metal substrate is coated at 108 with a further functional coating to further increase differential wettability of the membrane by hydrocarbons and water. The further functional coating may be, for example, silane, such as n-octadecyltrichlorosilane or heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane, a phosphonic acid, such as 1H,1H,2H,2H-perfluorooctanephosphonic acid, a carboxylic acid, an amine or any other suitable coating or combination of coatings.

[0053] Nanotetrapods (broadly called nanostructures) provide a nanoscale texture (nanotexture) on the surface of the permeable metal substrate that is not dependent on orientation of the nanostructures. As a result, precise lithographic patterning is not required. In addition, close packing of the nanotetrapods is inhibited by their geometries, yielding a porous network that facilitates permeation of liquid that wets the surface of the nanotetrapodal structures.

[0054] Referring to FIG. 2A, a process for separating a water and oil emulsion produced from a hydrocarbon recovery operation, utilizing the membrane including nanotetrapodal oxide on a permeable metal substrate, is illustrated. The process may contain additional or fewer operations than shown or described, and may be performed in a different order.

[0055] The membrane is disposed at a downward angle relative to a horizontal axis to facilitate the flow of liquid downwardly along the membrane at 202. The angle of the membrane may be about 10 to about 450 as determined by the rate of roll-off. For example, the membrane may be disposed at an angle of about 5 relative to the horizontal axis. The emulsion from the hydrocarbon recovery process is then applied to the membrane at 204 to produce a roll-off fraction that includes a major portion of the produced water in the emulsion and a permeate fraction that passes through the membrane. The permeate fraction includes a major portion of the oil in the emulsion. For example, the roll-off fraction may include up to about 100% of the water and about 1% of the oil in the emulsion and the permeate fraction may include up to about 100% of the oil in the emulsion with no visible water present. The emulsion is applied to the membrane by spray-coating, electrophoretic deposition, or dip-coating.

[0056] At least a portion of the roll-off fraction is treated and recycled to generate steam for the hydrocarbon recovery operation at 206. Treatment of the roll-off fraction may include several processes, including, for example, skim tank oil removal, induced static flotation/induced gas flotation for removal of suspended matter such as oil and solids, oil removal filtration, warm lime softening, and ion exchange for further removal of calcium and magnesium ions.

A portion of the roll-off fraction may be disposed of if it does not meet the quality required for steam generation.

[0057] The permeate fraction is treated prior to transporting at 208, for example, by mixing the permeate fraction with a diluent, for example, natural gas condensate, refined naphtha, or synthetic crude oil to meet pipeline transportation specifications. Treating the permeate fraction may include testing for oil quality in terms of basic sediment and water (BS&W). The permeate fraction may contain about <0.5% BS&W. The roll-off fraction may be tested for water quality in terms of residual oil content and further separation of oil and water may be required if the residual oil content is about 1 /0. If the residual oil content of the roll-off fraction is about <1%, the roll-off fraction may treated in a series of processes including, but not limited to skimming, flotation, oil filtering, lime softening (e.g., warm lime softening), lime softener filtration, and ion exchange processes (e.g., primary strong acid ion exchange, secondary weak acid ion exchange). Such processes may be utilized to remove oil, silica, calcium, magnesium, and iron from the roll-off fraction prior to utilizing the water for steam generation in SAGD.

[0058] The high surface area of nanotextured surfaces results in differential wettability that is utilized to separate the dissimilar liquids. Without wishing to be bound by theory, the effect of nanotexturation on the wettability of different liquids may be quite dissimilar depending on the intrinsic wettability of the surface by the liquid, which is dependent on the solid¨liquid interfacial energy and thus the specific intermolecular interaction. Intrinsic differences in wettability are greatly increased by hierarchical texturation. The vastly different surface tensions of the two liquids, i.e., water, which has a surface tension of 72.80 x 10-3 N/m at a temperature of 293K, and hydrocarbon, such as hexadecane, which has a surface tension of 27.47 x 10-3 N/m at a temperature of 293K, and the contrasting interaction with the oxide nanotetrapods on a permeable metal substrate facilitate effective separation.

[0059] The separation is a result of the wettability of the two liquids, water and hydrocarbons, as modified by texturation utilizing ZnO and molecular modification by silane treatment. When a liquid droplet comes into contact with a solid surface, the extent of dispersion of the liquid on the surface and the eventual shape of the droplet is determined by the balance between interfacial energies at the solid¨vapor, liquid¨vapor, and vapor¨liquid interfaces. As a liquid spreads onto a surface, the existing solid¨vapor interface is replaced by new liquid¨vapor and vapor¨liquid interfaces. The equilibrium contact angle (6,) reflects the balance between the three types of interfacial energies and may be simplified as:
case, = Y" ___________________________ Ysi.... ( 1) YLV
where ee is the equilibrium contact angle, and the y terms are the interfacial energies for the solid¨vapor (SV), solid¨liquid (SL), and liquid¨vapor (LV) surfaces.

[0060] For a surface that is non-wettable towards a liquid (ee > 120 ), ysL
is substantially greater than ysy. In other words, the intrinsic surface energy of the surface, a solid¨vapor surface energy, is very low and the solid¨liquid interfacial energy must be very high. Conversely, for the liquid to completely wet the surface (t9e = 00), the intrinsic surface energy corresponding to the solid¨vapor surface energy is greater than the solid¨liquid interfacial energy and the latter term is very small. When considering the wettability of a single surface by two different liquids, the ysv term is the same in both cases and thus both the sign and magnitude of cosee and ultimately the wettability of the two liquids is dictated by the relative value of ysL with respect to ysv. Two parameters strongly affect this balance: (1) the surface tension of the liquid, or the cohesive forces and the nature of the liquid itself, and (2) the chemical compatibility of the surface with the liquid, which is a function of the molecular interactions at the interface. For a specific range of ysv, two liquids with very different values of ysi_ may yield opposite signs of cosee for the same surface, thereby facilitating selective retention of one liquid and flow of the second liquid through without wetting the surface. Water and hydrocarbons, such as hexadecane, have very different surface tension values. In particular, water has a surface tension of 72.80 x 10-3 N/m and hexadecane has a surface tension of 27.47 x 10-3 N/m at 293 K. The lower surface tension of hexadecane implies that the ysi_ term is likely to be smaller, facilitating more readily wetting of a surface by hexadecane. This difference in surface tension values and interfacial interactions with the membranes facilitates the effective separation of the two liquids.

[0061] The surface roughness of a substrate may greatly enhance the interfacial surface area or alter the proportion of the surface across which the solid and liquid are actually in contact. A change in surface roughness changes the intrinsic wettability of a surface without changing the sign of cosee. The texturation by integrating ZnO nanotetrapods onto a permeable metal substrate such as a micrometer-sized mesh, referred to herein as multiscale texturation, greatly increases the differential wettability of hexadecane and water by rendering the surface more wettable to hexadecane and more repellant to water.

Thus, for two liquids with opposite signs of cose, multiscale texturation increases the differences in wettability. Modifying the surfaces with silane monolayers further allows for modulation of the interfacial interaction term ysL.

[0062] The membranes described above are utilized to separate the hydrocarbons from the water in emulsions produced from hydrocarbon recovery operations, such as steam-assisted gravity drainage (SAGD), expanding solvent steam-assisted gravity drainage (ES-SAGD), cyclic steam stimulation (CSS), steamflooding, solvent-assisted cyclic steam stimulation, toe-to-heel air injection (THAI), or a solvent aided process (SAP). For such treatment, the membrane may be installed within or in-line with a free-water knockout (FWKO) or a treater utilized after degassing to treat produced emulsion to remove water.
Alternatively, the membrane may be installed in place of one or both of the FWKO or the treater. Instead of separation of oil and water based on a difference in densities, the membranes described above are utilized for separation of oil and water based on a difference in wettabilities. The membranes may function as illustrated in FIG. 2B, in which oil permeates radially through a membrane configured as a cylinder (tube), while water, which may include a portion of residual oil, flows through the tube. The membrane may be disposed inside a housing and various operating conditions (e.g., pressure, temperature, emulsion flow rate, membrane capacity) may be applied to the membrane, the housing, or both the membrane and the housing. Processes for delivering emulsion to the membrane may be automated. The permeate fraction, the roll-off fraction, or both fractions may be re-delivered to the membrane one or more times before the permeate fraction, the roll-off fraction, or both are further processed, stored or transported.

[0063] The membrane illustrated in FIG. 2B may be utilized in a process for separating hydrocarbons from the water in emulsions produced from hydrocarbon recovery operations, such as SAGD, ES-SAGD, CSS, steamflooding, solvent-assisted cyclic steam stimulation, THAI, or a SAP. As indicated above, the membrane may be installed within or in-line with a free-water knockout (FWKO) or a treater utilized after degassing to treat produced emulsion to remove water. Alternatively, the membrane may be installed in place of one or both of the FWKO or the treater.

[0064] The feed emulsion from the hydrocarbon recovery operation enters the membrane. The hydrocarbons permeate radially through the membrane while the water does not permeate radially through the membrane, thereby separating the hydrocarbons and the water. Heat exchangers may optionally be employed to recover heat from the hydrocarbons and the water. The water may be reused after separation. Optionally, the water may be further purified after separation and prior to reuse. The hydrocarbons are further treated to remove additional water or contaminants or both. Optionally, a diluent may be added to the hydrocarbons to facilitate transportation, for example, through a pipeline or to a railcar.

[0065] A blower may be utilized to drive gas such as air, methane, natural gas, or nitrogen (N2) through the membrane and through the heat exchanger utilized to recover heat from the hydrocarbons. The blower, also referred to as a back pulse system, may blow pulses of the gas to clean out the membrane between uses.

[0066] Coated tubular membranes as described herein may be utilized in the separation of hydrocarbons and water from a feed, such as a produced emulsion from a hydrocarbon recovery operation, including high temperature and pressure applications, such as in SAGD facilities. The temperature of the emulsion that is applied to the membrane may be in the range of from ambient temperature up to about 250 C. Preferably, the temperature is in the range of from about 120 C to about 225 C, for example, in the range of from about 175 C

to about 220 C.

[0067] Data may be acquired to monitor the separation. For example, the feed, as well as the permeate and roll-off fractions may be monitored and analyzed as the emulsion is fed to the membrane. Such analysis may include chemical and analytical testing. Other data may also be monitored and analyzed, including the monitoring of flux through the membrane, transmembrane pressure (TMP), temperature, phase separation, and other parameters. The feed temperature may be from about 120 C to about 225 C, the pressure may be in the range of about 1000 kPa to about 2500 kPa, and the TMP may be about 200 kPa.

[0068] Testing of suitability of such membranes may include sample acquisition at regular intervals for analysis of oil and grease by infrared, total organic compounds, water-in-oil content, and any other suitable information.

[0069] Surprisingly, emulsions are effectively separated based on wettability differences alone utilizing the nanotetrapods and microscale features of the underlying substrate, such as meshes, without any need for lithographically defining specific morphologies. Utilizing metal mesh coated with a ceramic, the resultant membrane may be utilized at high temperatures and pressures and in corrosive environments, for example, the resultant membrane may be utilized within a range of process conditions for thermal in-situ techniques for hydrocarbon recovery. Such process conditions would be readily understood by a person of ordinary skill in the art given the present description.
The treatment of the surfaces of such membranes utilizing silanes further improves selective permeability of oil in water.
Examples Example 1: Hexadecane

[0070] Particular examples of membranes were fabricated and evaluated for the ability to separate water and oil emulsion produced from a hydrocarbon recovery operation.

[0071] Meshes may be from 60 gauge to 325 gauge. Membranes of 80 gauge stainless steel mesh and 180 gauge stainless steel mesh were tested.

[0072] The membranes in the present examples included a 316 stainless steel mesh with a pore size of about 84 pm, a 316 stainless steel mesh with a pore size of about 84 pm coated with ZnO nanotetrapods, a 316 stainless steel mesh with a pore size of about 84 pm coated with ZnO nanotetrapods and further treated with n-octadecyltrichlorosilane, and a 316 stainless steel mesh with a pore size of about 84 pm coated with ZnO nanotetrapods and further treated with heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane.

[0073] To fabricate the stainless steel mesh with a ZnO nanotetrapodal coating, Zn metal sheets (99% purity on metals basis) were cut into samples of about 3 mm x 3 mm in size. The Zn samples were placed onto a stainless steel mesh and inserted into a quartz tube of about 1" (2.54 cm) diameter, which in turn was placed in a Lindburg/BlueMTm tube furnace. The samples were heated at a rate of about 43 C/min with the ends of the tube furnace open to achieve a temperature of about 950 C. The samples were recovered after heating for about 1 min at 950 C.

[0074] The heat treatment resulted in highly crystalline nanostructures, which were collected and dispersed in a sufficient quantity of 2-propanol to provide dispersions with a concentration of about 20 mg/mL. The dispersions were spray coated onto 316 stainless steel meshes, each with a pore size of about 84 pm, utilizing an airbrush with a nozzle diameter of about 0.5 mm and an air compressor at an output pressure of about 45 psi (310 kPa). The coated meshes had a ZnO loading of about 3.5 mg/cm2. ZnO loadings of about 3.0 mg/cm2 to about 6.0 mg/cm2 were tested. A range of ZnO loadings may be suitable, for example, from about 2 mg/cm2 to about 25 mg/cm2. All spray coatings were carried out while heating the 316 stainless steel mesh on a heating plate with a surface temperature of about 120 C.

[0075] The membranes that were further treated with n-octadecyltrichlorosilane or heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane were fabricated as described above and further treated by tethering silanes to the ZnO nanotetrapodal coating to provide monolayers with pendant fluorinated or hydrocarbon chains. Treatment (functionalization) compound concentrations of about 2.7 mM, about 8.1 mM, and about 27 mM
were tested. A range of functionalization compound concentrations may be suitable, for example, from about 1.5 mM to about 250 mM. Stock solutions of about 2 vol.% of silane were prepared by combining about 400 pL of deionized water (p = 18.2 MQ/cm), about 400 pL of 28-30% ammonium hydroxide, and about 400 pL of the silane, followed by dilution to 20 mL utilizing n-butanol.
The silanes utilized included heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane and n-octadecyltrichlorosilane.

[0076] Stainless steel meshes with a ZnO nanotetrapodal coating were immersed in the butanol solutions including heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane for about one hour and then air dried to provide the membranes further treated with heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane.

[0077] Similarly, stainless steel meshes with a ZnO nanotetrapodal coating were immersed in the butanol solutions including n-octadecyltrichlorosilane for about one hour and then air dried to provide the membranes further treated with n-octadecyltrichlorosilane.

[0078] Referring now to FIG. 3 through FIG. 6, field emission scanning electron microscope (FE-SEM) images of the stainless steel mesh with ZnO
nanotetrapodal coating are shown. FIG. 3 through FIG. 5 show the porosity of the membrane provided by the network of ZnO nanotetrapods that extend across the pores of the stainless steel meshes. The FE-SEM images in FIG. 5 and FIG.

show that the nanotetrapods provided a multiscale textured morphology. The sharp thorn-like structure of the ZnO tetrapods brings about the nanoscale texturization of surfaces, thereby rendering metallic meshes hydrophobic and facilitating the use of orthogonal wettability to separate liquids with disparate surface tensions. (See also O'Loughlin, T. E.; Martens, S.; Ren, S. R.; McKay, P.;
Banerjee, S. Adv. Eng. Mater. 2017, 19 (5), 1600808.)

[0079] FIG. 7 is a Raman spectrum acquired for the ZnO nanotetrapodal coating and FIG. 8 is an x-ray diffraction (XRD) pattern for the ZnO
nanotetrapodal coating. The Raman spectrum shown in FIG. 7 is consistent with stabilization of the hexagonal form of ZnO. The XRD pattern of the ZnO
nanotetrapods was indexed to Joint Committee on Powder Diffraction Standards (JCPDS) # 36-1451, indicating the formation of phase-pure ZnO in the hexagonal zincite phase.

[0080] The wettability of the surface of each of the membranes towards water and hexadecane was tested and characterized by the application of a drop size of 10 pL of doubly distilled and deionized water and 10 pL of hexadecane.
A
mechanical pipette was utilized to apply the liquids and contact angles were determined utilizing at least three averaged values.

[0081] The hydrophobicity of the stainless steel mesh surfaces is increased by the deposition of ZnO nanotetrapods on the stainless steel mesh, as illustrated by the contact angles shown in FIG. 9 and FIG. 10, which show contact of water on a stainless steel mesh in FIG. 9 and contact of water on the stainless steel mesh coated with ZnO nanotetrapods in FIG. 10. FIG. 9 and FIG.

show the increase in contact angles measured for water from 119 2 in FIG.
9 to up to 154 1 in FIG. 10. The hydrophobicity of the stainless steel mesh coated with ZnO nanotetrapods was evident by the rolling off of water droplets, akin to the "lotus leaf" effect when the substrates were tilted.

[0082] FIG. 11 through FIG. 13 show the wettability of hexadecane on the stainless steel mesh coated with ZnO nanotetrapods by a sequence of images taken at 0 seconds in FIG. 11, 0.24 seconds in FIG. 12, and 0.48 seconds in FIG.
13, after contact of the drop with the membrane. FIG. 11 through FIG. 13 illustrate complete, flash spreading to a contact angle of 0 and permeation of hexadecane within 0.5 seconds.

[0083] For the purpose of testing the membranes for separation of water from hydrocarbons, emulsions were made by combining 15 mL of hexadecane and 15 mL of deionized water and shaking vigorously until the two phases were completely mixed. Because water and hexadecane are both colourless, blue food dye (propylene glycol, FD&C blue1 and red 40, propylparaben) was added to the water to visually distinguish the two components (not shown in the figures).

[0084] The membranes were mounted at a downward angle of about 5 relative to a horizontal axis to facilitate the flow of liquid downwardly along the membrane. A first vessel was utilized to collect the roll-off fraction and a second vessel was utilized to collect the permeate fraction. Hexadecane and water volumes were measured utilizing a graduated cylinder. Multiple samples were utilized and tests replicated for various effective lengths of membranes.

[0085] FIG. 14 is a schematic illustration of the experimental configuration utilized to test the separation of the water and hexadecane emulsions. Because of the wettability of the surface towards hexadecane, the permeate fraction was entirely hexadecane, as was apparent from visual observation and verified colourimetrically by the absence of a discernible spectroscopic signature of the blue dye. In contrast, the roll-off fraction was enriched in water with the specific water to oil ratio dependent on the effective length of the membrane.

[0086] To quantify the efficacy of the separation, the water purity was quantified as:
Water purity= x 100%...(2) VR
where Vw is the volume of water and VR is the total volume (oil and water) of the roll-off fraction. Because the starting emulsion was a 1:1 mixture of hexadecane and water, the initial water purity of all the samples was 50%.

[0087] FIG. 15 shows the water purity of the roll-off fraction, and thus the separation efficacy, as a function of the effective length of the membrane.
The plot has a clearly sigmoidal shape. A sample membrane length of about 200 cm yielded a roll-off fraction that was greater than 99% water. The efficacy of separation as a function of path length was approximated well by a sigmoidal Boltzmann function. Without being limited to theory, the probability of permeation through the membrane may increase with increasing path length given the energetic preference for wettability of hexadecane. In other words, permeation may correspond to a low-energy state. The continued removal of hexadecane may give rise to an open two-phase system, which may drive the system towards increasing water purity. In FIG. 15, the white circles represent the lower energy state or the permeated hexadecane and the black circles represent the higher energy state where the hexadecane has not permeated the mesh.

[0088] Referring to FIG. 16, the purity of the roll-off fraction, and thus the separation efficacy, for various samples along an effective length of membrane of about 21 cm is shown. Treating, also referred to as functionalizing, the surfaces with n-octadecyltrichlorosilane resulted in pendant octadecyl groups, thereby increasing the hydrophobicity and oleophilicity of the substrates. The separation efficiency was increased by >6%, in terms of the purity of water recovered for a given effective length of the membrane, for the stainless steel meshes with a ZnO nanotetrapodal coating further treated with n-octadecyltrichlorosilane by comparison to stainless steel meshes with a ZnO nanotetrapodal coating without further silane treatment.

[0089] The stainless steel meshes with a ZnO nanotetrapodal coating further treated with heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane, were more hydrophobic and more oleophobic by comparison to stainless steel meshes with a ZnO nanotetrapodal coating without further silane treatment and by comparison even to stainless steel meshes without a ZnO nanotetrapodal coating. Thus, the separation efficacy was reduced by treatment with heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane. The resulting surface was omniphobic and substantially no separation occurred at room temperature and pressure.

[0090] The octadecyl-terminated silane rendered the surface more oleophilic and more hydrophobic and thus further increased the magnitude of the wettability difference between water and hydrocarbons, and enhanced the separation efficacy, whereas the fluorinated hydrocarbon rendered the surface more hydrophobic but also more oleophobic and thus diminished the wettability difference, thereby degrading the separation efficiency at room temperature and pressure. Alternatively, at high temperatures at which water has a lower surface tension, the increased hydrophobicity provided by functionalizing with n-octadecyltrichlorosilane, heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane, or another functionalization compound may be useful for maintaining high orthogonal wettability.
Example 2: Wellhead Emulsion & Sales Oil

[0091] Further examples of membranes were utilized to test the permeability of those membranes to hydrocarbons other than hexadecane, including wellhead emulsion and sales oil. The membranes included a 316 stainless steel mesh with a pore size of about 84 pm; a 316 stainless steel mesh with a pore size of about 84 pm coated with ZnO nanotetrapods at a loading of about 3 mg/cm2; a 316 stainless steel mesh coated with ZnO nanotetrapods at a loading of about 6 mg/cm2; a 316 stainless steel mesh coated with ZnO
nanotetrapods at a loading of 3 mg/cm2 and treated with a 0.02 vol.% solution of heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane in butanol, and a 316 stainless steel mesh coated with ZnO nanotetrapods at a loading of 3 mg/cm2 and treated with a 2.0 vol.% solution of heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane in butanol.

[0092] Experiments were performed utilizing wellhead emulsion and utilizing sales oil. The wellhead emulsion was initially decanted to separate the water phase from the oil (bitumen). The recovered water phase was recombined with the oil of the emulsion in a 4:1 (v/v) ratio of water/oil (reconstituted emulsion) and heated in a glass container submerged within a water bath, to a temperature of 60 C. The samples were vigorously agitated for 30 minutes in an Astro 4550A Air Operated Paint Shaker. A diluent, Klean StripTM odorless mineral spirits, which is hydrotreated light distillate, was added to the oil phase of the wellhead emulsion samples in order to reduce the viscosity at room temperature to obtain fluid flow when a droplet is placed on a solid substrate. Diluents other than mineral spirits, for example, natural gas condensate, refined naphtha, synthetic crude oil, hydrocarbon solvents, for example, butane, pentane, or hexane, or a combination thereof, may be used to reduce the viscosity of the emulsion prior to membrane separation.

[0093] Sales oil is substantially more viscous than hexadecane, but has rheological properties such as low surface tension that allow membrane permeation and flow of sales oil at room temperature. Sales oil is also much less viscous than emulsions created at the wellhead due to the addition of diluent during wellhead emulsion processing.

[0094] The experiments described in Examples 1 and 2 were carried out to examine the wettability of the membranes by hydrocarbons. These experiments included hexadecane (Example 1) and sales oil and a 7:1 (v/v) oil:water reconstituted wellhead emulsion with diluent (that is, 4:3:1 (v/v/v) recovered water phase from wellhead emulsion:Klean StripTM diluent:oil of the emulsion) (Example 2), which have similar properties to the properties of wellhead emulsions at room temperature. The addition of diluent to some of the emulsion samples was utilized to provide less viscous samples to approximate high temperature conditions.

[0095] As demonstrated in FIG. 17 through FIG. 21, the sales oil only permeated the membranes (which are shown positioned at the top of a test tube in each of FIG. 17 through FIG. 21) including 316 stainless steel coated with ZnO
nanotetrapods (see FIG. 18 and FIG. 19, which reflect ZnO nanotetrapod loadings of 3 mg/cm2 and 6 mg/cm2, respectively), whereas the uncoated stainless steel mesh (FIG. 17) and fluorinated ZnO nanotetrapodal coated meshes (see FIG. 20 and FIG. 21, which reflect functionalization with 0.02 vol. /0 and 2.0 vol.% solutions, respectively, of heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane in butanol) remained impermeable to the sales oil. The permeation of sales oil through the 316 stainless steel coated with ZnO
nanotetrapods began immediately and lasted about 1 minute. The sales oil did not permeate the uncoated stainless steel meshes and the fluorinated ZnO
nanotetrapodal coated meshes, even after 10 hours of exposure. The ZnO
nanotexturation clearly facilitates permeation of the sales oil through the membrane and fluorination negatively impacted the oil permeation at the conditions tested.

[0096] The uncoated 316 stainless steel meshes were wetted by the hexadecane in the emulsions created (see Example 1) because of the low surface tension of hexadecane, but were not wetted by the sales oil (Example 2).
However, the texturation by deposition of ZnO nanotetrapods onto micrometer-sized 316 stainless steel mesh, referred to herein as multiscale texturation, greatly increased the intrinsic oleophilicity of the 316 stainless steel mesh and expanded the range of liquid surface tension values that wet the surface, thereby enabling permeation of the more viscous sales oil. In contrast, treatment of the surfaces of the ZnO nanotetrapods by heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane rendered the surfaces more oleophobic and even a low grafting density achieved with a 0.02 vol.% of the heptadecafluoro-1,1,2,2-tetrahydrodecyl trimethoxysilane rendered the 316 stainless steel meshes coated with ZnO nanotetrapods impermeable to sales oil.

[0097] The reconstituted wellhead emulsions with a 4:1 (v/v) ratio of water/oil to which diluent was added were diluted to form a 7:1 (v/v) mixture of oil/mineral spirits to reduce the viscosity and facilitate flow at room temperature.
Permeate and roll-off fractions were collected for the reconstituted emulsions with added diluent in a similar manner to that detailed with reference to FIG.
14.
In this example, the multiscale structured membrane was a stainless steel mesh with a ZnO nanotetrapod loading of 3 mg/cm2. The results showed that the reduced viscosity of the oil phase facilitated permeation of the multiscale structured membranes by the hydrocarbons in the emulsion as shown in FIG.
22A and FIG. 22B.
Example 3: Higher Temperature Emulsion Separation

[0098] Further examples of membranes were utilized to test the permeability of those membranes to bitumen and water emulsions at higher temperature. The bitumen and water emulsions utilized in this Example 3 were obtained from a Northern Alberta oil sands SAGD production facility. Emulsion viscosity was about 140,000 mPa=s at 25 C and water content was about 30 vol.%. The emulsions were utilized to study the efficacy of mesostructured meshes towards oil-water separation. The ability of such meshes to separate the water and oil fractions of these emulsions and the flux rates attainable for such a separation process were evaluated as a function of mesh pore size and ZnO
loading. These studies resulted in up to almost 98% reduction of water content in bitumen and water emulsions from about 30 vol.% water to about 0.69 vol.%
water by the selective permeation of bitumen and rejection of water through the inorganic membrane.

[0099] The zinc metal sheets were cut into small substrates that were about 3 mm x 3 mm in size. The zinc substrates were then placed onto a boat like stainless-steel mesh (not to be confused with the membrane itself, but for preventing the ZnO nanotetrapods from becoming affixed to the quartz tube itself during preparation) and then placed within a 1" diameter quartz tube, which was then placed within a tube furnace (Lindburg/BlueMm). The substrates were heated at a rate of about 43 C/min until a temperature of about 950 C was attained. The furnace was then held at 950 C for 1 min and then allowed to cool.

[00100] After the ZnO nanotetrapods cooled, the crystalline nanostructures were then dispersed in 2-propanol to obtain dispersions with a concentration of about 20 mg/mL. The dispersion was then spray coated onto stainless steel mesh substrates with a variety of pore sizes using an airbrush with a nozzle diameter of 0.5 mm, and an air compressor with output pressure of 45 psi. To facilitate the removal of solvent during the coating process, the stainless steel meshes were held at a temperature of about 120 C. Using a modified Stober method, a layer of Si02 was additionally deposited to optimize the adhesion and the mechanical resilience of the ZnO nanostructures (see StOber, W.; Fink, A.;
Bohn, E. J. Colloid Interface Sci. 1968, 26 (1), 62-69). The deposition of an amorphous Si02 shell helps prevent the absorbed nanostructures from being readily sloughed off the mesh under harsh conditions. To deposit a Si02 shell, tetraethylorthosilicate (TEOS) was used as the precursor. The solution spray coated on the stainless-steel mesh comprised a mixture of 80 vol.% ethanol, deionized 18.5 vol.% water (p=18.2MC2cm-1) , an aqueous solution of 1 vol.% of 28-30% Nh140H, and 0.5 vol.% TEOS. The substrates were held at a temperature of about 120 C during the spray coating of the TEOS solutions to facilitate solvent removal.

[00101] The ZnO nanotetrapod morphology was imaged utilizing a JEOL
JSM-7500F field-emission scanning electron microscope (FE-SEM). The instrument was equipped with a high brightness conical FE gun with a low aberration conical objective lens. The source was a cold cathode UHV field emission conical anode gun. An accelerating voltage of 10 kV was used to image the structures.

[00102] A thermal autoclave testing apparatus (comprised of glass for in situ permeation temperature measurements) that can operate at temperatures up to 200 C and pressures of up to 900-1000 kPa was used for this purpose. The system was filled with 250 mL of water and heated to temperatures of 110-200 C. A custom glass insert was placed inside the thermal autoclave to observe the permeation of bitumen through the membranes. The autogenous pressure from the heating of water did not form a pressure gradient across the membrane as the insert was configured with a small hole to prevent such a pressure build up. During subsequent separation of emulsions under conditions of high temperature and pressure, the permeation temperature was recorded as soon as visible bitumen was seen permeating through the membrane in the autoclave.

[00103] Regarding separation efficiency, the Dean-Stark's method along with optical microscopy were utilized to evaluate the water content within the permeated heavy oil fractions. For samples with water content too low for quantitation via Dean Stark, Karl Fischer titration was performed on a representative set of samples. In the Dean-Stark's method for determination of water content, given the viscous nature of the permeated bitumen, a solvent (toluene) was mixed with the bitumen and then refluxed for 2 h to collect and measure water content in the permeate. To obtain statistically meaningful results, a minimum of 3 membranes with the same pore size and ZnO loading were tested. Water content was measured by the Dean-Stark method for permeate samples of about 10 mL recovered upon filtration; quantitation was possible using Dean-Stark when at least 0.1 mL of water was recovered as a distillate, establishing a detection limit of about 1 vol.%. Samples that did not yield a measurable amount of water were deemed to have a water content below 1%. A representative set of such samples were analyzed using Karl-Fisher titration. Karl-Fisher titration was performed using a Mettler-Toledo C20 Coulometric Titrator with a diaphragm. The electrolyte used for both the anolyte and the catholyte was Hydranal Coulomat E from Sigma Aldrich. The water content was determined to be about 6945 ppm.

[00104] Mesoscale porosity was defined by the interconnected network of ZnO nanotetrapods that spanned across the pores of the metal meshes. To test the separation efficiency, different stainless steel meshes with varying pore sizes were used, and the loading of ZnO on the membranes was systematically varied.
The meshes included 180-gauge, 250-gauge, 325-gauge, 400-gauge mesh, and 500-gauge mesh corresponding to pore sizes of about 84 pm, 61 pm, 43 pm, 38 pm, and 30 pm, respectively. A ZnO loading of about 7.0 rng/cm2 was utilized in each case. SEM images (not shown) indicated that the nanotetrapods defined an interconnected network that precludes close packing. The TEOS loading and associated parameters were previously varied to obtain good adhesion as verified by the American Society for testing of Materials (ASTM) tests D3359 and D2197 (see O'Loughlin, T. E.; Waetzig, G. R.; Davidson, R. E.; Dennis, R. V.;
Banerjee, S. 2017 Encycl. Inorg. Bioinorg. Chem. 1-21; O'Loughlin, T. E.; Dennis, R. V.;

Fleer, N. A.; Alivio, T. E. G.; Ruus, S.; Wood, J.; Gupta, S.; Banerjee, S.
Energy & Fuels 2017, 31 (9), pp 9337-9344).

[00105] For meshes with larger pore sizes, 180 and 250-gauge corresponding to pore sizes of 84 pm and 61 pm respectively, it was observed that most of the oil started penetrating through the membrane at low temperatures. As a result of the complex nature of the emulsions, water droplets entrained within oil droplets may permeate through if a separation is engineered at low temperatures based on surface tension alone. Thus, a temperature of greater than about 130 C may crack (break) the emulsions and thus permeation should occur only above temperatures where the concentric nature of the emulsions has been disrupted. In other words, permeation at low temperatures may yield samples with high degrees of water contamination since only free water may be separated under these conditions.

[00106] The use of high temperatures is relevant to the operating conditions under which emulsions from hydrocarbon recovery operations such as SAGD are extracted and handled. FIG. 23 depicts that smaller dimensions (pore size) necessitate the use of higher temperature for permeation of the bituminous phase. Likewise, with increased ZnO loading, higher permeation temperatures were required to permeate the bitumen providing opportunities for reducing water content.

[00107] The permeate fractions were further examined by optical microscopy. In order to perform the analysis, the permeated fraction was deposited onto a thin microscope slide with no additional dilution. Optical microscopy results indicated that the SAGD emulsions had a complex hierarchical structure with oil droplets dispersed within a continuous water phase containing water droplets that further contained asphaltene residues. The nature of these reconstituted emulsions may make oil-water separation challenging, as higher temperatures are requred to crack the emulsions.

[00108] Starting from bitumen emulsion that had not been treated by permeation through membranes, and which contained about 30 vol.% water, optical microscopy images of the emulsion were compared with permeated fractions from various stainless steel meshes all with a ZnO nanotetrapod loading of 14 mg/cm2. Water and oil were distinguishable in the images (not shown) with water being the lighter and more transparent fraction. With decreasing pore size and identical loading, it was apparent that the water droplet size and frequency decreased. Karl-Fisher titration results suggested a reduced water content of 0.69 vol. % for the permeate recovered using a 500-gauge membrane with a pore size of 30 pm and ZnO loading of 14 mg/cm2.

[00109] FIG. 24 shows a series of graphs, (A) to (D). FIGs. 24 (A) and (B) together illustrate that the flux rate through the membrane was observed to be inversely correlated to water purity of the permeate. That is, higher flux rates and higher % water content in the permeate were observed for membranes having larger pore sizes and lower ZnO loadings. Flux rate was measured at the permeation temperature and it was observed that higher temperatures enhance =
the flux rate without compromising water purity. For larger pore dimensions and lower ZnO loadings, the flux rate was high (reaching 20 mL/h, see FIG. 24 (A)) despite the relatively low permeation temperatures. In contrast, the flux rate was diminished for smaller pore dimensions and higher ZnO loadings. The decrease in flux rates engendered by increased ZnO loading may be rationalized by the smaller effective pore diameter. For a 500-gauge mesh with a ZnO
loading of 28 mg/cm2, no permeation was observed up to a temperature of 190 C.

[00110] By using the Dean-Stark method, the amount of water that permeated through the different membranes was quantified. The water content of the hydrated bitumen emulsion utilized as the precursor in all experiments described in this Example 3 was 30 vol. %. The water content recovered in the permeate was found to be a function of the permeation temperature and thus also of pore size and ZnO loading. As shown in FIG. 24(B), membranes with larger pore dimensions and relatively low ZnO loadings, such as 180 and 250-gauge meshes with ZnO loadings of 7 mg/cm2 permeated 18.5% and 15.7% of water, respectively. While the water content was substantially reduced from the bitumen and water emulsion, most of the water content eliminated was free water and emulsified water droplets were largely permeated given the relatively low permeation temperatures of 117 C and 127 C, respectively. In contrast, the 500-gauge mesh with a ZnO loading of 22.5 mg/cm2 yielded a permeate with a water content of 0.69 vol. %. FIGs. 24 (C) and (D) depict related plots of %
water content in the permeate and flux rate (mL/h), respectively, as a function of the mesh pore size and ZnO loading. These figures further illustrate that higher oh water content in the permeate and higher flux rates of emulsion through the membrane were observed for membranes having larger pore sizes and lower ZnO loadings.

Conclusion

[00111] While undiluted bituminous emulsions were too viscous to permeate the multiscale structured membranes, permeability was observed for the samples diluted by the addition of mineral spirits, and by the sales oil samples.
Uncoated 316 stainless steel meshes were impermeable to both liquids. Thus, the ZnO
nanotetrapodal coating on the 316 stainless steel meshes increases the wettability by oil. Separation of emulsions obtained from Northern Alberta oil sands was observed based on orthogonal wettability of hydrocarbons and water towards nanotextured surfaces. The separation efficiency and flux rate were tuned by adjusting the permeation temperatures as a function of pore size and ZnO loading. The membranes significantly reduced the quantity of water present in the emulsions and achieved a permeate water content as low as 0.69 vol.%.

[00112] The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the claims should not be limited by the preferred embodiments set forth in the examples, but should be given the broadest interpretation consistent with the description as a whole. All changes that come with meaning and range of equivalency of the claims are to be embraced within their scope.

Claims (27)

What is claimed is:

1. A membrane for separating a produced water and oil emulsion from a hydrocarbon recovery operation, the membrane comprising:
a permeable metal substrate and a nanotetrapodal oxide coating on the permeable metal substrate.

2. The membrane according to claim 1, wherein the permeable metal substrate comprises a metal mesh.

3. The membrane according to claim 1 or claim 2, comprising a silane disposed on the nanotetrapodal oxide coating.

4. The membrane according to claim 3, wherein the silane comprises n-octadecyltrichlorosilane.

5. The membrane according to any one of claims 1 to 3, comprising at least one of a phosphonic acid, a carboxylic acid, and an amine disposed on the nanotetrapodal oxide coating.

6. The membrane according to claim 2, wherein the metal mesh comprises at least one of stainless steel, aluminum, brass, bronze, copper, polytetrafluoroethylene coated stainless steel, galvanized low alloy steel, nickel-coated low alloy steel, and an acid-resistant nickel.

7. The membrane according to claim 2, wherein the metal mesh comprises at least one of 316 stainless steel mesh of from about 150 gauge (104 microns) to about 500 gauge (30 microns), 304 stainless steel mesh of from about 150 gauge (104 microns) to about 500 gauge (30 microns), aluminum mesh of up to about 200 gauge (74 microns), brass wire mesh of up to about 100 gauge (152 microns), bronze wire mesh of up to about 325 gauge (43 microns), copper mesh of up to about 200 gauge (76 microns), polytetrafluoroethylene coated 304 stainless steel mesh of up to about 325 gauge (43 microns), and acid-resistant nickel mesh of up to about 200 gauge (74 microns).

8. The membrane according to any one of claims 1 to 7, wherein the substrate comprises a metal mesh having a pore size of up to about 43 microns and a ZnO
loading of greater than or equal to 7 mg/cm2.

9. The membrane according to any one of claims 1 to 8, wherein the substrate comprises a metal mesh having a pore size up to about 30 microns and a ZnO
loading of about 22.5 mg/cm2.

10. The membrane according to any one of claims 1 to 9, wherein the nanotetrapodal oxide coating comprises at least one of ZnO, Al2O3, MgO, Fe2O3, Fe3O4, SiO2, TiO2, V2O5, ZrO2, HfO2, MoO3, and WO3.

11. The membrane according to claim 1 or claim 2, a SiO2 coating disposed on the nanotetrapodal oxide coating on the metal substrate.

12. A process for separating a produced water and oil emulsion from a hydrocarbon recovery operation, the process comprising:
applying the emulsion to a membrane comprising a permeable metal substrate having nanotetrapodal oxide coating thereon, to produce a roll-off fraction comprising at least a portion of the produced water in the emulsion and a permeate fraction that passes through the membrane, the permeate fraction comprising at least a portion of the oil in the emulsion.

13. The process according to claim 12, wherein the membrane is disposed at a tilt angle during application of the emulsion.

14. The process according to claim 12 or claim 13, wherein the applying the emulsion to the membrane comprises applying the emulsion to a tubular membrane.

15. The process according to claim 12, wherein applying the emulsion comprises applying an emulsion produced from a SAGD hydrocarbon recovery operation.

16. The process according to claim 12, wherein applying the emulsion comprises applying an emulsion at a temperature of from about 120°C to about 225°C.

17. The process according to claim 16, wherein applying the emulsion comprises applying an emulsion at a temperature of from about 175°C to about 220°C.

18. The process according to claim 12, comprising treating the roll-off fraction prior to utilizing the roll-off fraction to generate steam for the hydrocarbon recovery operation.

19. The process according to any one of claims 12 to 14, comprising treating the permeate fraction prior to transporting the permeate fraction.

20. The process according to claim 19, wherein treating the permeate fraction comprises adding a diluent to the permeate fraction to facilitate transportation.

21. The process according to any one of claims 12 to 20, comprising adding a diluent to the emulsion prior to applying the emulsion to the membrane.

22. A method of producing a membrane for separating a produced water and oil emulsion from a hydrocarbon recovery operation, the method comprising:
heating a metal foil in the presence of oxygen to a temperature sufficient to cause oxidation of the metal foil and the production of nanotetrapodal oxide structures;
dispersing the nanotetrapodal oxide structures in a solvent to provide a dispersion solution;
coating a permeable metal substrate with the dispersion solution.

23. The method according to claim 22, wherein coating the permeable metal substrate comprises coating a metal mesh.

24. The method according to claim 22 or claim 23, wherein coating comprises spray coating the dispersion solution onto the metal substrate.

25. The method according to claim 24, wherein coating comprises heating while spray coating.

26. The method according to claim 22 or claim 23, comprising applying a SiO2 coating to the nanotetrapodal oxide structures on the metal substrate.

27. The membrane according to claim 22 or claim 23, comprising disposing a silane on the nanotetrapodal oxide structures on the metal substrate.