Electrosorptive cavitation devices and methods for phase separation

Information

  • Patent Grant
  • 12017155
  • Patent Number
    12,017,155
  • Date Filed
    Tuesday, March 13, 2018
    6 years ago
  • Date Issued
    Tuesday, June 25, 2024
    7 months ago
Abstract
Provided herein are phase separation devices and methods for performing phase separation of an aqueous mixture. Said devices and methods may include a cavitation component for cavitating the aqueous mixture to produce a cavitated aqueous mixture; and an electrosorption component for performing an electrosorptive phase separation of the cavitated aqueous mixture. Such devices and methods may be used for phase separation of aqueous mixtures including algal aqueous mixtures and/or mining or waste water aqueous mixtures, for example.
Description
FIELD OF INVENTION

The present invention relates generally to devices and methods for phase separation of aqueous mixtures. More specifically, the present invention relates to phase separation devices and methods using electrosorptive cavitation.


BACKGROUND

Many industrial processes involve one or more phase separation steps in which aqueous mixtures are processed to separate hydrophilic and hydrophobic phases or components, and/or to separate out contaminants, desirable products, or other components present in an aqueous mixture. In the field of algae processing, for example, algal lipids or oils are extracted from algae cultivated in an aqueous environment. Phase separation is also performed as part of waste water management operations, and in industrial mining operations, for example.


Lipid extraction and/or phase separation of aqueous mixtures may be costly, time-consuming, and/or may involve high energy demands. By way of example, algal biofuel production typically involves cell disruption, harvesting, de-watering, and lipid extraction steps. Such harvesting and de-watering has conventionally involved filtration, reverse osmosis, electrocoagulation-electroflocculation (EF-EC), centrifugation, and/or drum drying for concentration, and lipid extraction involving disruption or conversion of algae using pressure, stress, heat, or solvent treatment.


Currently, mechanical and high energy cell disruption methods such as high-pressure homogenisers or sonication typically involve an energy input in the order of hundreds of MJ kg−1 of the dry mass. Heat based approaches following the concentration of the biomass to at least 10 wt % may also lead to a lower quality of fuel while consuming the biomass, complicating recovery of other potentially valuable products. In addition, many laboratory process methods are not readily scalable, making commercial or pilot scale biofuel production difficult.


Hydrodynamic cavitation (HC) has been previously used for cell disruption, however the cell disruption by HC was periplasmic and thus mainly confined to the cell wall and membrane, and involved solvents for extraction following the HC process and additional unit operations for separation of the phases.


Alternative, additional, and/or improved devices and/or methods for phase separation of aqueous mixtures are desirable.


SUMMARY OF INVENTION

In certain embodiments, there are provided herein devices and methods for phase separation of aqueous mixtures using electrosorptive cavitation. Cavitation of aqueous mixtures may be used to generate a cavitated aqueous mixture, which may be subjected to electrosorption, thereby providing phase separation of the aqueous mixture. Such combination of cavitation and electrosorption as described herein may be used to achieve phase separation in a variety of different aqueous mixtures, including algal aqueous mixtures and/or mining or waste water aqueous mixtures, for example.


In an embodiment, there is provided herein a phase separation device for performing phase separation of an aqueous mixture, said device comprising:

    • a cavitation assembly for cavitating the aqueous mixture to produce a cavitated aqueous mixture; and
    • an electrosorption assembly for receiving the cavitated aqueous mixture and performing an electrosorptive phase separation of the cavitated aqueous mixture.


In an embodiment of the above device, the device may be a lipid extraction and phase separation device for performing phase separation of an aqueous mixture, said device comprising:

    • a cavitation assembly for cavitating the aqueous mixture to produce a cavitated aqueous mixture with extracted lipids; and
    • an electrosorption assembly for receiving the cavitated aqueous mixture and performing an electrosorptive phase separation of the cavitated aqueous mixture.


In another embodiment of the above device or devices, the cavitated aqueous mixture may comprise cavitation voids comprising vapour, and the electrosorptive phase separation of the cavitated aqueous mixture may include electrosorptive vapour bubble generation.


In still another embodiment of the above device or devices, the electrosorption assembly may perform electrosorptive phase separation including microflotation.


In yet another embodiment of the above device or devices, the cavitation assembly may comprise a tangential flow aqueous mixture inlet, a vortex stabilizer in communication with the inlet, and a nozzle in communication with the vortex stabilizer, the nozzle comprising a divergent outlet in communication with the electrosorption assembly.


In another embodiment of the above device, the vortex stabilizer may comprise a vortex stabilizer insert comprising a conical portion having a longitudinal slit for allowing vortexing aqueous mixture to exit the vortex stabilizer and enter the nozzle.


In yet another embodiment of the above device or devices, the nozzle may comprise a restriction flow channel having a length L and a diameter d, wherein the restriction flow channel receives the aqueous mixture and directs the aqueous mixture to the divergent outlet.


In another embodiment of the above device, the restriction flow channel may have a d/L ratio of about 0.005-0.02.


In still another embodiment of the above device or devices, d may be about 0.03-0.0625 inches.


In another embodiment of the above device or devices, the divergent outlet may comprise a conical channel which progressively widens toward the electrosorption assembly with a divergent angle of about 30°-45°.


In still another embodiment of the above device or devices, the divergent outlet may have a diameter to length ratio of about 0.1875-0.3.


In another embodiment of the above device or devices, the electrosorption assembly may comprise a cathode and an anode, the cathode forming an outer shell defining an interior core, with the anode disposed within the core. In certain embodiments, the cathode may comprise a conductive spiral tube with a central core, and the anode may comprise a conductive member disposed within the central core. In certain embodiments, the cathode may comprise a longitudinal tube, coil, spiral, or other suitable structure, and may define an interior core having a longitudinal axis along which the anode is disposed. The cathode may be substantially continuous, or may comprise a series of individual members longitudinally oriented in series.


In yet another embodiment, the cathode may comprise a conductive spiral coil or tube comprising two or more turns, wherein each turn, in combination with the anode, may form a distinct electrochemical cell. In certain embodiments, the cathode may comprise an array of rings aligned in series, each forming a distinct electrochemical cell in combination with the electrode. The person of skill in the art having regard to the teachings of the instant application will be aware of suitable cathode/anode configurations, and will be able to select a suitable cathode/anode configuration suitable for a particular application.


In still another embodiment of the above device or devices, the cathode may comprise a carbon-based fiber conductive material, anodized aluminium, TiO2, brass, or stainless steel.


In yet another embodiment of the above device or devices, the device may be configured vertically or on an incline, such that the electrosorption assembly receives the cavitated aqueous mixture at a bottom end thereof and performs electrosorptive phase separation of the cavitated aqueous mixture as the cavitated aqueous mixture moves toward a top end of the electrosorption assembly, the top end being elevated relative to the bottom end. In certain embodiments, the cavitated aqueous mixture may exit the top end of the electrosorption assembly and be collected in a floatation cell or other phase separation tank.


In still another embodiment of the above device or devices, the electrosorption assembly may comprise a cathode and an anode, the cathode forming an outer shell defining an interior core, and the anode disposed within the core.


In yet another embodiment of the above device or devices, the cathode may comprise a continuous conductive member.


In another embodiment of the above device or devices, wherein the cathode may comprise a plurality of individual conductive members longitudinally aligned in series.


In still another embodiment, the cathode may comprise a tube, spiral, or coil. In yet another embodiment, the cathode may comprise two or more tubes, spirals, coils, or rings, or a combination thereof. In another embodiment, the tube, spiral, or coil may define a helix comprising two or more turns, wherein each turn, in combination with the anode, forms a distinct electrochemical cell. In still another embodiment, the cathode may comprise two or more spaced apart rings aligned in series, which form, in combination with the anode, a plurality of distinct electrochemical cells.


In certain embodiments, there is provided herein a method for continuous phase separation of an aqueous mixture, said method comprising:

    • cavitating the aqueous mixture to produce a cavitated aqueous mixture; and
    • performing electrosorptive phase separation of the cavitated aqueous mixture.


In another embodiment of the above method, the cavitating the aqueous mixture produces cavitation voids comprising vapour within the cavitated aqueous mixture, and wherein the electrosorptive phase separation of the cavitated aqueous mixture induces electrosorptive vapour bubble generation.


In yet another embodiment of the above method or methods, the electrosorptive phase separation may include microflotation.


In still another embodiment of the above method or methods, the step of cavitating may comprise passing the aqueous mixture through a cavitation assembly as defined hereinabove.


In another embodiment of the above method or methods, the step of performing electrosorptive phase separation may comprise passing the cavitated aqueous mixture through an electrosorption assembly as defined hereinabove.


In another embodiment of any of the above device or devices, the aqueous mixture may comprise a microalgal culture. In certain embodiments, the microalgal culture may comprise Chlorella vulgaris algal biomass, Scenedesmus algal biomass, or a combination thereof, for example.


In still another embodiment of any of the above method or methods, the aqueous mixture may comprise a microalgal culture. In certain embodiments, the microalgal culture may comprise Chlorella vulgaris algal biomass, Scenedesmus algal biomass, or a combination thereof, for example.


In another embodiment of the above device or devices, the electrosorptive phase separation may include algal lipid extraction.


In another embodiment of the above method or methods, the electrosorptive phase separation may include algal lipid extraction.





BRIEF DESCRIPTION OF DRAWINGS


FIG. 1 shows a perspective view of an embodiment of a phase separation device as described herein, with the outer surfaces of the cavitation and electrosorption assemblies made transparent to show interior components of the device;



FIG. 2 shows a front plan view of the device shown in FIG. 1, with the outer surfaces of the cavitation and electrosorption assemblies made transparent to show interior components of the device;



FIG. 3 shows a perspective view of the cavitation assembly of the device shown in FIG. 1, with the outer surfaces of the cavitation assembly made transparent to show interior components of the device;



FIG. 4 shows a front plan view of the cavitation assembly of the device shown in FIG. 1, with the outer surface of the cavitation assembly made transparent to show interior components of the device;



FIG. 5 shows (A) a front plan view of the conductive spiral coil cathode of the electrosorption assembly of the device shown in FIG. 1; (B) front plan views of the rod-shaped anode of the electrosorption assembly of the device shown in FIG. 1; (C) a front plan view of the outer enclosure of the electrosorption assembly of the device shown in FIG. 1; and (D) a perspective view of the assembled electrosorption assembly of the device shown in FIG. 1;



FIG. 6 shows (A) side plan, (B) perspective, (C) front plan, and (D) side plan views of an anode separator component of the electrosorption assembly of the device shown in FIG. 1;



FIG. 7 shows (A) perspective, (B) side plan, (C) side plan, and (D) front plan views of the vortex stabilizer of the cavitation assembly of the device shown in FIG. 1;



FIG. 8 shows (A) perspective, (B) side plan, (C) side plan, and (d) front plan views of the nozzle of the cavitation assembly of the device shown in FIG. 1;



FIG. 9 shows (A) perspective, (B) side plan, (C) side plan, and (D) front plan views of the flange of the cavitation assembly of the device shown in FIG. 1;



FIG. 10 shows (A) perspective, (B) side plan, (C) side plan, and (D) rear plan views of the vortex stabilizer insert of the cavitation assembly of the device shown in FIG. 1;



FIG. 11 shows examples of small and large-sized vortex stabilizer inserts;



FIG. 12 shows examples of small, medium, and large-sized divergent outlets of nozzle portions of cavitation assembly embodiments;



FIG. 13 shows an illustrative comparison of (A) electrosorption and (B) electrocoagulation processes;



FIG. 14 shows examples of aqueous mixtures for which phase separation may be desired, and industrial or commercial operations commonly associated therewith;



FIG. 15 depicts an experimental setup for an electrosorption process using 2.5 g/L of algal culture as shown in Example 2 with clear separation of phases;



FIG. 16 depicts an experimental setup for an electrosorption process using 2.5 g/L of algal culture with 30 mL/L of hydrochloric acid as shown in Example 2 with clear separation of phases;



FIG. 17 depicts an experimental setup for an electrosorption process using 2.5 g/L of algal culture with 8 mL/L of hydrochloric acid added as shown in Example 2 with clear separation of phases;



FIG. 18 depicts an experimental setup for an electrosorption process using 2.5 g/L of algal culture with 2 mL/L followed by 7 mL/L of hydrochloric acid as shown in Example 2 with clear separation of phases;



FIG. 19 depicts an experimental setup for an electrosorptive cavitation process using 2.5 g/L of algal culture processed through cavitation system with the addition of 10 mL of hydrochloric acid at 1200 psi (8375 kPa.) and electrosorption applied at 31 V and 1.45 A as shown in Example 2 with clear separation of phases;



FIG. 20 depicts an experimental setup for an electrosorptive cavitation process using 2.5 g/L of algal culture processed through 1 L batch system with electrosorption applied at 31 V and 1.45 A with the addition of 10 mL of hydrochloric acid/750 mL of culture as shown in Example 2 with clear separation of phases;



FIG. 21 depicts an experimental setup for an electrosorption process using 2.5 g/L of algal culture processed through 1 L batch system with electrosorption applied at 31 V and 1.45 A with the addition of 0.5 mL of hydrochloric acid/750 mL of culture as shown in Example 2 with clear separation of phases;



FIG. 22 depicts an experimental setup for an electrosorptive cavitation process using 2.5 g/L of algal culture processed through 1 L batch system with electrosorption applied at 31 V and 1.45 A with the addition of 1.5 mL of hydrochloric acid/750 mL of culture as shown in Example 2 with clear separation of phases;



FIGS. 23-29 depict experimental setups, and processed samples produced, in the testing performed in Example 2;



FIGS. 30-45 depict experimental setups, and processed samples produced, in the testing performed in Example 3;



FIG. 46 shows depictions of algal cultures, cavitation test setups, low pressure cavitation setups, and processed algal products obtained following treatment, in the testing performed in Example 5;



FIGS. 47-52 depict experimental setups, and processed samples produced, in the testing performed in Example 6;



FIG. 53 depicts a graph illustrating lipid yield as a function of biomass concentration following lipid extraction from Chlorella sorokiniana using electrosorptive cavitation;



FIG. 54 depicts micrographs taken of low lipid yielding algae in (A) untreated samples; (B) samples treated using electrosorptive cavitation—showing global agglomeration; (C) samples treated using cavitation with longer residence time; and (D) samples treated using electrosorption only—showing limited agglomeration;



FIG. 55 shows the UV-Vis spectra of a range of coal mine water samples treated to remove selenium using oxidation treatment methods and electrosorptive cavitation;



FIG. 56 shows micrographs of the biomass analysed using continuous electrosorptive cavitation compared with electrosorption and cavitation carried out in separate processes. (A) Raw feed 0.5 g/L Botryococcus braunii Race A; (B) Nozzle only; (C) Nozzle→electrode separately (Aqueous flow-through); (D) Nozzle→electrode separately (Foam); (E) Nozzle+electrode integrated (Aqueous flow-through); (F) Nozzle+electrode integrated (Foam); and (G) Nozzle+electrode integrated (Aqueous settled);



FIG. 57 shows photographs of the samples of biomass analysed using continuous electrosorptive cavitation compared with electrosorption and cavitation carried out in separate processes. (1) Raw Feed, (2) Cavitation Output (Nozzle only), (3) Cavitation with a residence time of 5 minutes followed by electrosorption (Nozzle+electrode separately) and (4) Electrosorptive Cavitation (Nozzle+electrode integrated);



FIG. 58 shows rapid selenium concentration increase from strong base anion exchange in metallurgical coal affected water outlet stream; and



FIG. 59 shows that selenium effluent concentration was not affected sufficiently by adsorption processing with Adsorbia As 600 adsorbent.





DETAILED DESCRIPTION

Described herein are devices and methods for phase separation of aqueous mixtures using electrosorptive cavitation. Cavitation of aqueous mixtures may be used to generate a cavitated aqueous mixture, which may be subjected to electrosorption, thereby providing phase separation of the aqueous mixture. By way of example, in certain embodiments, the devices and methods described herein may be for lipid extraction from a lipid-containing aqueous mixture, such as an algal aqueous mixture. In such embodiments, cavitation of the aqueous mixture may generate a cavitated aqueous mixture comprising algal biomass, lipids, and water, which may be subjected to electrosorption. It will be appreciated that embodiments and examples are provided for illustrative purposes intended for those skilled in the art, and are not meant to be limiting in any way.


In an embodiment, there is provided herein a phase separation device for performing phase separation of an aqueous mixture, said device comprising:

    • a cavitation assembly for cavitating the aqueous mixture to produce a cavitated aqueous mixture; and
    • an electrosorption assembly for receiving the cavitated aqueous mixture and performing an electrosorptive phase separation of the cavitated aqueous mixture.


In another embodiment, there is provided herein a method for continuous phase separation of an aqueous mixture, said method comprising:

    • cavitating the aqueous mixture to produce a cavitated aqueous mixture; and
    • performing electrosorptive phase separation of the cavitated aqueous mixture.


As will be understood, the aqueous mixture may be any suitable aqueous mixture comprising water and one or more suitable components to be separated therefrom. By way of example, suitable aqueous mixtures may include algal aqueous mixtures and/or mining or waste water aqueous mixtures. In certain embodiments, the aqueous mixture may include industrial waste water, or industrial mining waste water. The aqueous mixture may comprise, for example, marine algae, freshwater algae, microalgae, colloidal waste water, mine effluent, gold mine effluent, cement mine effluent, and/or waste water from a gold ore processing and refining operation, metallurgical coal production, another suitable mine water application, and/or a municipal sewage water treatment, for example. The skilled person having regard to the teachings herein will be aware of suitable aqueous mixtures for which phase separation may be desired. FIG. 14 provides certain examples of aqueous mixtures for which phase separation may be desired, and industrial or commercial operations commonly associated therewith.


It is contemplated herein that, in certain embodiments, the aqueous mixture may include an aqueous mixture for which removal of oxyanions such as Arsenic and Selenium may be desired; recovery of base and/or precious metals from effluents may be desired; cyanide destruction may be desired; ammonia removal may be desired; fermentation process water treatment may be desired; removal of contaminants from oils may be desired; and/or lignocellulosic biomass pre-treatment may be desired.


As will be understood, phase separation may include any suitable separation of aqueous mixture components, and may involve partial, substantial, or full phase separation. In certain embodiments, hydrophobic components and/or components having low water solubility present in the aqueous mixture may be separated from a water fraction. By way of example, phase separation may involve formation of separate hydrophobic (i.e. oil, lipid, or oleaginous) and hydrophilic (i.e. water) layers, or isolation/extraction of hydrophobic (i.e. oil, lipid, hydrocarbon) components from the aqueous mixture, or isolation/extraction of mixture components having low or reduced solubility in water, or collection of mixture components such as metals or other cations/anions (which may be, optionally, oxidized) and/or lipids and/or other hydrophobic components on the respective cathode or anode portion of the electrosorption assembly, or any combination thereof.


In certain embodiments, phase separation devices as described herein may be in fluid communication with one or more conventional separation systems, wherein processed liquids exit the phase separation device and enter the conventional separation system for further phase separation and/or isolation. By way of example, phase separation devices as described herein may be in fluid communication with one or more settling chambers, flotation tanks, or skimming tanks (see, for example, U.S. Pat. No. 8,968,571, which is herein incorporated by reference in its entirety) which receive processed sample from the phase separation devices as described herein and allow for further phase separation thereof.


The skilled person will recognize that phase separation as described herein may be used, in certain embodiments, to separate and/or isolate particular components of a given aqueous mixture, and/or to extract/remove particular contaminants from a given aqueous mixture, for example. In certain embodiments, phase separation may involve extraction of desirable components such as, for example, algal lipids/oils, lignocellulosic material, or valuable metals from an aqueous mixture. In certain other embodiments, phase separation may involve regeneration of waste water through removal of one or more hydrophobic or metal contaminants. As will be recognized, the components to be extracted or separated from the aqueous mixture via phase separation will depend on the particular aqueous mixture used, and the particular downstream application(s) of the aqueous mixture and/or component(s) separated therefrom.


The cavitation assembly may be any suitable assembly which can receive the aqueous mixture to be processed, and cavitate the aqueous mixture to produce a cavitated aqueous mixture. As will be understood, the cavitation assembly may process (i.e. cavitate) the aqueous mixture to form cavitation voids comprising vapour in the aqueous mixture. In certain embodiments, and without wishing to be bound by theory, at least some of such cavitation voids may collapse, thereby generating nuclei which may assist in electrosorptive phase separation in the electrosorption assembly, as described in further detail below. As will be understood, in certain embodiments, the cavitation assembly may comprise a nozzle, in which the aqueous mixture is received and directed to a divergent outlet section, as described in further detail below. The skilled person having regard to the teachings herein will be aware of suitable cavitation assembly designs, and will be able to select a suitable cavitation assembly for the particular application.


Where, for example, an algal aqueous mixture is to be processed, cavitation by the cavitation assembly may result in at least some disruption or breakage of the algal cell, thereby exposing desirable algal components such as, for example, algal lipids, which may collect on the cavitation voids. Where, for example, a mining waste water aqueous mixture is to be processed, cavitation by the cavitation assembly and action of the forming/collapsing cavitation voids may result in formation of at least some hydroxyl radicals, which may oxidize metals in the mining waste (i.e. an advanced oxidation (AOP) process), thereby altering metal solubility and/or preparing the metals for cathode attachment in the electrosorption assembly. As will be understood, in certain embodiments, chemical oxygen demand (COD) of an aqueous mixture may be altered by treatment described herein.


In certain embodiments, the cavitation assembly may comprise a tangential flow aqueous mixture inlet, a vortex stabilizer in communication with the inlet, and a nozzle in communication with the vortex stabilizer, the nozzle comprising a divergent outlet in communication with the electrosorption assembly. The tangential flow aqueous mixture inlet and vortex stabilizer may generate a swirl or spiral flow of the aqueous mixture entering the nozzle. The nozzle section may receive the spiral-flowing aqueous mixture, and direct the mixture to a divergent outlet section, as described in further detail below. In certain embodiments, the vortex stabilizer may comprise a vortex stabilizer insert comprising a conical portion having a longitudinal slit for allowing vortexing aqueous mixture to exit the vortex stabilizer and enter the nozzle.


In certain embodiments, the cavitation assembly may comprise a nozzle which may comprise a restriction flow channel having a length L and a diameter d, wherein the restriction flow channel receives the aqueous mixture and directs the aqueous mixture to the divergent outlet. In still another embodiment, the restriction flow channel may have a d/L ratio of about 0.005-0.02, for example. In yet another embodiment, d may be about 0.03-0.0625 inches, for example. In another embodiment, the divergent outlet may comprise a conical channel which progressively widens toward the electrosorption assembly, with a divergent angle of about 30°-45°, for example. In still another embodiment, the divergent outlet may have a diameter to length ratio of about 0.1875-0.3, for example.


The electrosorption assembly may be any suitable assembly which can receive the cavitated aqueous mixture, and perform electrosorptive phase separation of the cavitated aqueous mixture. As will be understood, the electrosorption assembly may comprise a cathode and an anode, to which certain components of the aqueous mixture may attach or agglomerate, thereby facilitating their separation from the aqueous mixture, for example. In certain embodiments, and without wishing to be bound by theory, the electrosorption assembly may multiply cavitation voids in the cavitated aqueous mixture through an advanced oxidation process, thereby providing electrosorptive vapour bubble generation. In certain embodiments, such electrosorptive vapour bubble generation may facilitate phase separation via microflotation. The skilled person having regard to the teachings herein will be aware of suitable electrosorption assembly designs, and will be able to select a suitable electrosorption assembly for the particular application.


In certain embodiments, for example, the electrosorption assembly may comprise a cathode and an anode, the cathode forming an outer shell defining an interior core, with the anode disposed within the core. By way of example, the cathode may comprise a conductive spiral tube with a central core, and the anode may comprise a conductive member disposed within the central core. In certain embodiments, the cathode may comprise, for example, a longitudinal tube, coil, spiral, or other suitable structure, and may define an interior core having a longitudinal axis along which the anode is disposed. The cathode may be substantially continuous, or may comprise a series of individual members longitudinally oriented in series. In yet another embodiment, the cathode may comprise a conductive spiral coil or tube comprising two or more turns, wherein each turn, in combination with the anode, may form a distinct electrochemical cell. In certain embodiments, the cathode may comprise an array of rings aligned in series, each forming a distinct electrochemical cell in combination with the electrode. The person of skill in the art having regard to the teachings of the instant application will be aware of suitable cathode/anode configurations, and will be able to select a suitable cathode/anode configuration suitable for a particular application.


Where, for example, an algal aqueous mixture is to be processed, electrosorptive phase separation by the electrosorption assembly may result in at least some attachment and/or aggregation of algal lipids/oils on the surface of a cathode portion of the electrosorption assembly. Without wishing to be bound by theory, it is believed that algal lipids/oils may accumulate on cavitation voids, nuclei, and/or other voids or vapour bubbles created by action of the cavitation and/or electrosorption assembly, which may then become attached to the cathode portion of the electrosorption assembly. Collapse of cathode-attached cavitation voids and/or other voids or vapour bubbles having algal lipids or oils attached thereto may further promote algal lipid or oil clumping or agglomeration at the cathode. Algal lipids or oils may then be retrieved from the cathode, and/or may be retrieved through flotation or other density separation, for example. By way of example, liquid exiting the electrosorption assembly may be allowed to phase-separate, and the lipid-rich layer may be collected. In certain embodiments, electrosorptive vapour bubble generation in the electrosorption assembly may facilitate phase separation through phase separation, particularly where algal-size bubbles (i.e. about 2-5 μm) are generated through regulation of pressure, flow, and/or energy applied at the electrosorption assembly.


Where, for example, a mining waste water aqueous mixture is to be processed, electrosorptive phase separation by the electrosorption assembly may include accumulation of cationic or oxidized metal species at the cathode, and anionic species at the anode. Action of the hydroxyl radicals resulting from formation/collapse of cavitation voids may result in oxidation of metals in the mining waste, thereby altering metal solubility and/or facilitating metals attachment to the cathode of the electrosorption assembly and corollary anions formed are further reduced in the process.


In certain embodiments, the electrosorption assembly may comprise a cathode and an anode, the cathode comprising a conductive spiral coil or tube defining an inner central core or passage, and the anode comprising a conductive member axially disposed within the central core or passage. In certain embodiments, the conductive spiral coil or tube of the cathode may comprise two or more turns, wherein each turn, in combination with the anode, may form a distinct electrochemical cell. As such will be understood, in certain embodiments, the anode of electrosorption assemblies as described herein are not sacrificial anodes, and the skilled person may therefore select an appropriate anode material accordingly.


In certain embodiments, the spiral cathode and anode may be enclosed within an outer shell, such that the cavitated aqueous mixture exiting the cavitation assembly (and having a spiral flow pattern) may encounter a flow channel having a substantially circular cross-section defined by the outer shell, and may become exposed to the cathode and anode of the electrosorption assembly as it passes through the flow channel. In certain embodiments, the spiral flow of the cavitated aqueous mixture passing through the flow channel, and the spiral structure (i.e. handedness, etc.) of the cathode may be matched, thereby providing reduced fluid flow resistance through the electrosorption apparatus and providing increased contact and residence time between the cavitated aqueous mixture and the electrosorptive assembly.


In certain embodiments, the cathode of the electrosorption assembly may comprise a carbon-based conductive material, copper, aluminium (with and without anodization), TiO2, brass, or stainless steel. In certain embodiments, the anode of the electrosorption assembly may comprise carbon, stainless steel, iron, or aluminium (anodized or non-anodized). The person of skill in the art having regard to the teachings provided herein will be able to select suitable cathode and anode materials based on the particular application, particular contaminants involved, desired effluent discharge limits, and/or other such considerations.


In further embodiments the phase separation device may be configured vertically or on an incline, such that the electrosorption assembly receives the cavitated aqueous mixture at a bottom end thereof and performs electrosorptive phase separation of the cavitated aqueous mixture as the cavitated aqueous mixture moves toward a top end of the electrosorption assembly, the top end being elevated relative to the bottom end. In certain embodiments, the cavitated aqueous mixture may exit the top end of the electrosorption assembly and be collected in a flotation cell or other phase separation tank.


Example 1—An Example Phase Separation Device


FIG. 1 shows a perspective view of an embodiment of an “electrosorptive cavitation” phase separation device as described herein. This illustrative device embodiment applies principles of fluid flow in a circular cross section flow channel (i.e. a pipe or tube) and the least resistant path of spiral flow, so as to encourage contact and residence time between the aqueous mixture to be treated and the electrosorption assembly. When applied to algal aqueous mixtures, the illustrated phase separation device may employ cavitating lipid extraction processes enhanced by electrosorption to extract and separate water, lipid and biomass phases in batch, or continuously.


The illustrated phase separation device (1) comprises a cavitation assembly (2) for cavitating an aqueous mixture to produce a cavitated aqueous mixture, and an electrosorption assembly (3) for receiving the cavitated aqueous mixture and performing an electrosorptive phase separation of the aqueous mixture. FIG. 2 shows a front plan view of the device shown in FIG. 1. Phase separation device (1) may be used for performing a method for continuous phase separation of an aqueous mixture, said method comprising: cavitating the aqueous mixture to produce a cavitated aqueous mixture; and performing electrosorptive phase separation of the cavitated aqueous mixture. In this example, use of device (1) for performing phase separation of an algal aqueous mixture to obtain algal lipids/oils for biofuel application is described, however it will be understood that such devices are not limited to processing algal aqueous mixtures, and may also be used to process a variety of other suitable aqueous mixtures.


Phase separation device (1) comprises a cavitation assembly (2), which is configured to receive the aqueous mixture to be processed, and cavitate the aqueous mixture to produce a cavitated aqueous mixture. The cavitation assembly (2) cavitates the aqueous mixture to form cavitation voids comprising vapour in the aqueous mixture, at least some of which may collapse, thereby generating nuclei which may assist in electrosorptive phase separation in the electrosorption assembly (3), as described in further detail below. Cavitation by cavitation assembly (2) results in at least some, or a majority of, the disruption or breakage of the algal cells of the aqueous mixture, thereby exposing desirable algal lipids, which collect on the surface of cavitation voids (bubbles) through lipid-bubble surface forces.


As shown in figure detail in FIGS. 3 and 4, cavitation assembly (2) comprises a tangential flow aqueous mixture inlet (6), which directs the aqueous mixture to vortex stabilizer (7) in communication with the inlet (6), and a nozzle (13) in communication with the vortex stabilizer (7), the nozzle (13) comprising a divergent outlet (9) in communication with the electrosorption assembly (3). FIG. 7 shows vortex stabilizer (7) of cavitation assembly (2) of device (1) in further detail. FIG. 8 shows nozzle (13) of cavitation assembly (2) of device (1) in further detail.


Low to high pressure (1724-11721 kPa, for example), which may depend on concentration of algal culture, algal cell size and shape, and/or desired throughput/flowrate, for example, may be applied to the tangential flow aqueous mixture inlet (6), and the tangential flow aqueous mixture inlet (6) and vortex stabilizer (7) may generate a swirl or spiral flow of the aqueous mixture entering the nozzle (13). In certain embodiments, for example, low pressure may be applied, such as a pressure of about 240 kPa to about 1825 kPa for example about 1724 kPa, and flow may be adjusted accordingly to maintain quality of lipid formed. The nozzle (13) may receive the spiral-flowing aqueous mixture, and direct the mixture to the divergent outlet (9). In the illustrated cavitation assembly (2), the vortex stabilizer (7) may comprise a vortex stabilizer insert (10) comprising a conical portion (11) having a longitudinal slit (12) for allowing vortexing aqueous mixture to exit the vortex stabilizer (7) and enter the nozzle (13). Vortex stabilizer insert (10) is shown in further detail in FIG. 10.


Nozzle (13) comprises a restriction flow channel (8) having a length L and a diameter d (see FIG. 4), wherein the restriction flow channel (8) receives the aqueous mixture and directs the aqueous mixture to the divergent outlet (9). In phase separation device (1), the restriction flow channel (8) employs a relatively long flow channel and low d/L ratio (i.e. a d/L ratio of about 0.005-0.02, for example, and a d of about 0.03-0.0625 inches, for example), thereby generating a small channel effect to control the cavitation zone length and location downstream. In certain embodiments, such a system may allow for a stable cavitation zone to be formed resulting in new bubbles entering from incipient cavities and collapsing bubbles at the other end resulting in a stable zone (length) or cavitation region. The divergent outlet (9) comprises a conical channel which progressively widens toward the electrosorption assembly (3), with a divergent angle of about 30°-45°, for example, and a small channel to divergent outlet ratio of about 0.1875-0.3, for example.



FIG. 11 shows examples of small and large-sized vortex stabilizer inserts, of which vortex stabilizer insert (10) is one example.



FIG. 12 shows examples of small, medium, and large-sized divergent outlets of nozzle portions of cavitation assembly embodiments, of which divergent outlet (9) is one example.


The cavitation assembly (2) cavitates the aqueous mixture to form cavitation voids comprising vapour in the aqueous mixture, at least some of which may collapse, thereby generating nuclei which may assist in electrosorptive phase separation in the electrosorption assembly (3). Aqueous mixture passing through divergent outlet (9) experiences partial pressure drop below the vapour pressure, resulting in the formation of such cavitation voids. When such cavitation voids collapse, sites of high-energy are created due to micro-implosions, resulting in generation of nuclei which enhance bubble production rate during electrosorption. In certain embodiments, and without wishing to be bound by theory, the bubble nuclei formed during cavitation may enhance the rate of formation of bubbles in the electrosorption and may lead to the propagation of bubble clouds. Lipid bubble attachment may occur in the bubble clouds, and lead to sorption of bubbles on the surface of the electrode.


Divergent outlet (9) of cavitation assembly (2) is in communication with electrosorption assembly (3), allowing cavitated aqueous mixture to exit cavitation assembly (2) and enter electrosorption assembly (3). In illustrated phase separation device (1), the cavitation assembly (2) and the electrosorption assembly (3) are coupled together via flange (14) as shown in FIGS. 1, 2, 4, and 9.


Electrosorption assembly (3) receives the cavitated aqueous mixture from the cavitation assembly (2), and performs electrosorptive phase separation of the cavitated aqueous mixture. As shown in FIG. 2, electrosorption assembly (3) comprises a cathode (4) and an anode (5), to which certain components of the aqueous mixture may attach or agglomerate, thereby facilitating their separation from the aqueous mixture.


In phase separation device (1), electrosorption assembly (3) comprises a cathode (4) and an anode (5), the cathode (4) comprising a conductive spiral coil defining an inner central core or passage, and the anode (5) comprises a conductive rod-shaped member co-axially disposed within the central core or passage defined by the spiral-shaped cathode (4). The conductive spiral coil of cathode (4) comprises multiple turns, wherein each turn, in combination with the associated portion of anode (5), forms a distinct electrochemical cell. In the device of this example, one or more anode separators (16) as shown in FIGS. 5(D) and 6 are used to position anode (5) within the central core defined by spiral cathode (4).


Although depicted as a spiral coil in this example, it will be understood that the cathode may adopt other suitable configurations. For example, in certain embodiments, the cathode may be configured as a longitudinal tube, coil, spiral, or other suitable structure, and may define an interior core having a longitudinal axis along which the anode is disposed. The cathode may be substantially continuous, or may comprise a series of individual members longitudinally oriented in series. The cathode may comprise a conductive spiral coil or tube comprising two or more turns, wherein each turn, in combination with the anode, may form a distinct electrochemical cell. In certain embodiments, the cathode may comprise an array of rings or other such units aligned in series, each forming a distinct electrochemical cell in combination with the anode. The person of skill in the art having regard to the teachings of the instant application will be aware of suitable cathode/anode configurations, and will be able to select a suitable cathode/anode configuration suitable for a particular application.


In this example, spiral cathode (4) and anode (5) are enclosed within an outer shell (15), which comprises a pipe or tube, such that the cavitated aqueous mixture exiting the cavitation assembly (2) (and having a spiral flow pattern) encounters a flow channel having a substantially circular cross-section defined by the outer shell (15) pipe, and becomes exposed to cathode (4) and anode (5) of the electrosorption assembly (3) as it passes through this flow channel. The spiral flow of the cavitated aqueous mixture passing through the flow channel, and the spiral structure (i.e. handedness, etc.) of cathode (5) are matched, thereby providing reduced fluid flow resistance through the electrosorption apparatus (3) and providing increased contact and residence time between the cavitated aqueous mixture and the electrosorptive assembly (3). FIG. 5 shows further details of (A) conductive spiral coil cathode (4), (B) rod-shaped anode (5), and (C) outer shell (15) of electrosorption assembly (3) of phase separation device (1) described in this Example. FIG. 5 (D) provides a perspective view of assembled electrosorption assembly (3), with the outer shell (15) comprising a transparent tube to allow for visual inspection during operation (drawing units are in inches, and are indicated for non-limiting and illustrative purposes intended for the person of skill in the art).


Cathode (4) of the electrosorption assembly (3) may comprise, for example, a spiral wound carbon-based fiber conductive material, activated or impregnated cloth conductive material, anodized aluminium, iron, TiO2, brass, or stainless steel, or other suitable conductive material. In phase separation device (1), a central core positive conductive source attached to an AC power input serves to enhance the advanced oxidation generated by cavitating nuclei, resulting in lipid-biomass-water phase separation through lipid-bubble interactions and reduced polarity of the algal cells rendering their effective separation and flotation of the lipid and algae phases with eventual settling of algae (by gravity) in the flotation cell in a single step.


In certain embodiments, the anode of the electrosorption assembly may comprise carbon, stainless steel, iron, or anodized aluminium. As will be understood, electrosorption assembly (3) performs electrosorptive function, and therefore does not comprise a sacrificial anode. As will be understood, electrosorption is based on sorption of material or contaminant on the electrodes, and does not involve release of electrode material. The electrode is thus non-sacrificial in the case of electrosorption.


Electrosorptive phase separation by electrosorption assembly (3) results in at least some attachment and/or aggregation of algal lipids/oils on the surface of cathode (4) of the electrosorption assembly (3). Without wishing to be bound by theory, it is believed that algal lipids/oils may accumulate on cavitation voids, nuclei, and/or other voids or vapour bubbles created by action of the cavitation and/or electrosorption assembly, which may then become attached to cathode (4) of electrosorption assembly (3). Collapse of cathode-attached cavitation voids and/or other voids or vapour bubbles having algal lipids or oils attached thereto may further promote algal lipid or oil clumping or agglomeration at cathode (4). Algal lipids or oils may then be retrieved from cathode (4), and/or may be retrieved through floatation or other density separation, for example. By way of example, liquid exiting electrosorption assembly (3) may be allowed to phase-separate, and the lipid-rich layer may be collected. In certain embodiments, electrosorptive vapour bubble generation in electrosorption assembly (3) may facilitate phase separation through phase separation, particularly where algal-size bubbles (i.e. about 2-5 μm) are generated through regulation of pressure, flow, and/or energy applied at the electrosorption assembly. Energy input by electrosorption assembly (3) may multiply cavitation voids in the cavitated aqueous mixture via an advanced oxidation process, thereby providing electrosorptive vapour bubble generation. Such electrosorptive vapour bubble generation may further facilitate phase separation via microflotation. Without wishing to be bound by theory, in certain embodiments, cavitation bubbles may be considered as nuclei which collapse with high intensity and tensile forces in algae culture, for example. This may allow for early nuclei to form in the electrosorption section and cause increased bubble generation. Separation may occur by lipid-bubble attachment with algae, the separated layers travelling up the tube into a microflotation cell, which may accelerate separation of phases, for example.


Phase separation device (1) described in this Example is designed to operate in single step-fashion, without requiring addition of added chemicals, solvents, flocculants, adsorbents, or pH buffer agents (it will be recognized that nutrients supplied to the culture during the algal growth in the photobioreactors may be present in examples where an algal aqueous mixture is to be processed). The combination of hydrodynamic cavitation and electrosorption design in a single step as described in this example may allow for scale-up and reduced complexity for one-step processes for both extraction and separation.


As will be recognized, phase separation device (1) employs electrosorption. Electrosorption is generally considered as potential-induced adsorption of molecules on the surface of charged electrodes, and is used herein for separation of phases with the use of high specific-area electrodes. Electrosorption involves charging and discharging the electrical double layer at the surface of electrodes by applying potential or current. In this respect, electrosorption is distinguished from conventional electrochemical or Faradaic processes such as electrocoagulation or electroflocculation, and offers in-situ regeneration of adsorbent with good energy consumption compared with the other separation processes (see, for example, Yang, K. L., Ying T-Y, Yiacoumi, S., Tsouris, C and Vittoratos, S. Electrosorption of Ions from Aqueous Solutions by Carbon Aerogel: An Electrical Double-Layer Model, Langmuir, 2001, 17 (6), pp 1961-1969; herein incorporated by reference in its entirety). FIG. 13 shows an illustrative comparison of (A) electrosorption and (B) electrocoagulation processes, indicating distinctions there between.


The phase separation device of this Example has been tested for continuous flow operation of up to about 1.5 L/min in the processing of algal aqueous mixtures having an algal concentration of about 0.1 g/L and 1 g/L.


Example 2—Direct Culture Extraction of Algae Using Electrosorptive Cavitation

A series of experiments were designed and performed in order to test phase separation of algal aqueous mixtures using phase separation devices as described herein. Testing included analysis of individual cavitation assemblies and electrosorption assemblies, and testing of combined electrosorptive cavitation phase separation devices as described herein.


The following tables provide experimental set-up and processing conditions of various experiments designed to assess the electrosorptive cavitation devices described herein. Experimental observations for each test are provided, along with accompanying figure(s) depicting experimental setup and processed samples produced.









TABLE 1





Bench Scale Test -1A
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste Unprocessed, non-cavitated


Additives to Culture
None


Capacitance Separation
Spiral (Segmented) cathode with high conduc-


Setup
tivity metal rod anode


Volume
1 L


Conductivity


pH


Pressure
N/A


Frequency
N/A


Voltage
31 V


Current
5.22 A (at start, decreased steadily)



5.10 A (after sometime- Not recorded)



2.00 A (after sometime - Not recoded)


Run Time
6:00 min run



3:00 min rest


Additional Details


Physical Observations
Clear separation of phases



White and deep green layers at top - algae and



lipid separation









Experimental setup, and processed sample produced, shown in FIG. 15.









TABLE 2





Bench-Scale Test-1B
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Unprocessed, non-cavitated


Additives to Culture
30 mL hydrochloric acid


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
1 L


Conductivity


pH


Pressure
N/A


Frequency
N/A


Voltage
31 V


Current
8.5 A (at start, decreased steadily)



4.5 A (after some time)


Run Time


Additional Details


Physical Observations
Clear separation of phases



White and deep green layers at top - algae and



lipid separation









Experimental setup, and processed sample produced, shown in FIG. 16.









TABLE 3





Bench Scale Test-1C
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Unprocessed, non-cavitated


Additives to Culture
8 mL hydrochloric acid


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
750 mL


Conductivity


pH
6.93 min (raw sample)



2.50 min (with HCl)



4:40 min (end of run)


Pressure
N/A


Frequency
N/A


Voltage
31 V


Current
1.31 A (at start)



0.62 A (after 5 mins)


Run Time
6:00 min


Additional Details


Physical Observations
Clear separation of phases









Experimental setup, and processed sample produced, shown in FIG. 17.









TABLE 4





Bench Scale Test-1D
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Processed through cavitation system for 1 run at



1200 psi and 30 Hz


Additives to Culture
2 mL hydrochloric acid, then additional 5 mL



added


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
750 mL


Conductivity
174.0 mV (end of run)


pH
6.90 (raw sample)



3.10 (with HCl)



2.60 (with additional HCl)



4.10 (end of run)


Pressure
1200 psi


Frequency
30 Hz


Voltage
31 V


Current
0.8 A (at start)



0.6 A (after 9 min)


Run Time
6:00 min - no change in solution, no separation



Paused test and added 5 mL acid



9:00 - solution separating slowly



Paused test and lowered both electrodes further



into solution



11:00 - solution cleared, test stopped


Additional Details


Physical Observations
Clear separation of phases









Experimental setup, and processed sample produced, shown in FIG. 18.









TABLE 5





Bench Scale Test-1E
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Processed through cavitation system for 2 runs at



1200 psi and 30 Hz


Additives to Culture
Hydrochloric acid


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
750 mL


Conductivity


pH
6.65 (raw sample)



2.50 (with HCl)



4.22 (end of run)


Pressure
1200 psi


Frequency
30 Hz


Voltage
31 V


Current
153 A (at start)



0.67 A (after 6 min)


Run Time
6:00 min


Additional Details


Physical Observations
Clear separation of phases









Experimental setup, and processed sample produced, shown in FIG. 19.









TABLE 6





Bench Scale Test-1F
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Processed through cavitation system for 2 runs at



1200 psi and 30 Hz


Additives to Culture
Hydrochloric acid


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
750 mL


Conductivity


pH
6.77 (raw sample)



2.50 (with HCl)



3.50 (end of run)


Pressure
1200 psi


Frequency
30 Hz


Voltage
31 V


Current
1.45 A (at start)


Run Time
3:00 min - solution close to full clarity



5:00 min - full clarity, test stopped


Additional Details


Physical Observations
Clear separation of phases









Experimental setup, and processed sample produced, shown in FIG. 20.









TABLE 7





Bench Scale Test-2A
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Sonicated


Additives to Culture
10 mL hydrochloric acid, added at 1:00 min


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
750 mL


Conductivity


pH
5.39 (raw sample)



3.95 (end of run)


Pressure


Frequency


Voltage
31 V


Current
1.95 A (after 2:00 min)



1.70 A (after 2:20 min)


Run Time
6:00 min - solution almost reached full clarity,



test stopped


Additional Details


Physical Observations
Clear separation of phases in



short amount of time with foaming of lipid.









Experimental setup, and processed sample produced, shown in FIG. 21.









TABLE 8





Bench Scale Test-2B
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Unprocessed, non-cavitated


Additives to Culture
0.5 mL hydrochloric acid


Capacitance
Spiral cathode with high conductivity metal rod


Separation Setup
anode


Volume
700 mL


Conductivity


pH
5.57 (raw sample)



4.00 (with HCl)



4.50 (end of run)


Pressure


Frequency


Voltage
31 V


Current
0.22 A


Run Time
6:25 min - solution cleared, test stopped


Additional Details


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 22.









TABLE 9





Bench Scale Test-2C
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Sonicated


Additives to Culture
0.5 mL hydrochloric acid



0.5 mL additional acid added at 3:30 min



0.5 mL additional acid added at 5:30 min


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
700 mL


Conductivity


pH
3.40 (raw sample)



5.40 (with HCl)


Pressure


Frequency


Voltage
31 V


Current
0.34 A


Run Time
7:00 min - solution cleared, test stopped


Additional Details


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 23.









TABLE 10





Bench Scale Test-3A
















Test Setup
Test in Broadband Sonication System


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Non-processed, non-cavitated


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
500 mL


Conductivity


pH
7.00 (raw sample)


Pressure


Frequency
574 Hz


Voltage
31 V


Current
0.30 A


Run Time
15:30 min - sonication time



15:00 min - electrosorption time


Additional Details


Physical Observations
Separated to almost full clarity









Experimental setup, and processed sample produced, shown in FIG. 24.









TABLE 11





Bench Scale Test-3B
















Test Setup
Test in Broadband Sonication System


Solution Type
Reconstituted algae from paste combined with



potable water


Concentration
2.5 g/L algae paste



Non-processed, non-cavitated


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
500 mL


Conductivity


pH


Pressure


Frequency
574 Hz


Voltage
15 V


Current
0.15 A


Run Time
3:00 min - power turned on (3:00 min of



sonication only)



7:25 min - changed power from CV to CC of



0.277 V


Additional Details


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 25.









TABLE 12





Bench Scale Test-4A.
















Test Setup
Test in 1 L beaker with spinning magnet



Technology)


Solution Type
Reconstituted algae from paste combined with



potable water



Added nutrient-rich water to increase conductivity



(Nutrient: BBM stock)


Concentration
2.5 g/L algae paste



Nutrient-rich water added to the existing potable



water + algae paste


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
800 mL


Conductivity
560 uS/cm


pH
6.70 (raw sample with nutrient-rich water)


Pressure
N/A


Frequency
N/A


Voltage
36 V


Current
0.9 A


Run Time
8:18 min - test stopped, solution clarity


Additional Details


Physical Observations
Notable lipid-water separation



Clarity reached by end of run









These experiments were performed to examine the effect of magnetism on the electrosorption process. Phase separation was obtained using a spinning magnetic stirrer. Under the conditions tested, the phase separation occurred after a long time, and no specific benefit was observed using a magnet. However, conductivity increase, as shown in Tables 13 and 14 (among others) showed shorter phase separation periods.


Experimental setup, and processed sample produced, shown in FIG. 26.









TABLE 13





Bench Scale Test-4B
















Test Setup
Test in Broadband Sonication System


Solution Type
Reconstituted algae from paste combined with



potable water



Added nutrient-rich water to increase conductivity



(Nutrient: BBM stock)


Concentration
2.5 g/L algae paste



Nutrient-rich water added to the existing potable



water + algae paste


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
700 mL


Conductivity
987 uS/cm


pH


Pressure
N/A


Frequency
574 kHz


Voltage
36 V


Current
3.8 A


Run Time
1:40 min - test stopped


Additional Details
Nutrient-rich water was added to the solution in



increments of 300 mL to increase the conductivity



Final conductivity of 987 uS/cm attained


Physical Observations
Clear separation of phases



Bubbling and raising of lipids to top of cylinder









Experimental setup, and processed sample produced, shown in FIG. 27.









TABLE 14





Bench Scale Test-5A
















Test Setup
Test in 1 L beaker with spinning magnet


Solution Type
Nutrient-rich water and algae paste


Concentration
Reconstituted algae: 0.08%



3.8 g algae paste in 3.5 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral cathode with high conductivity metal


Separation Setup
rod anode


Volume
800 mL


Conductivity
1730 uS/cm


pH


Pressure
N/A


Frequency
N/A


Voltage
36 V


Current
2.5 A at start, 2.8 A when power turned on for



second time


Run Time
3:00 min - power turned off



6:00 min - min rest period, then power turned on



8:00 min - test stopped


Additional Details


Physical Observations
At 40 seconds, can see separation of phases



beginning









Experimental setup, and processed sample produced, shown in FIG. 28.









TABLE 15





Bench Scale Test-5B
















Test Setup
Test in sonicator


Solution Type
Nutrient-rich water and algae paste


Concentration
Reconstituted algae: 0.08%



3.8 g algae paste in 3.5 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral cathode with high conductivity metal


Separation Setup
rod anode


Volume
700 mL


Conductivity
1730 uS/cm


pH


Pressure
N/A


Frequency
574 kHz


Voltage
36 V


Current
6.7 A


Run Time
2:24 min


Physical Observations


Additional Details









Experimental setup, and processed sample produced, shown in FIG. 29.


Example 3—Evaluation of Capacitance-Based Separation Using Reduced Cross Sectional Area

A series of experiments were designed and performed in order to test capacitance-based separation using reduced cross-sectional area. The following tables provide experimental set-up and processing conditions of a series of tests which were performed. Experimental observations for each test are provided, along with accompanying figure(s) depicting experimental setup and processed samples produced.









TABLE 16





Analysis Test 1
















Test Setup
Solution of algae paste in nutrient-rich water run



through low pressure nozzle system, then tested



using empirical evaluation using reduced cross



sectional area laboratory test


Solution Type
Nutrient-rich water and algae paste


Concentration
Reconstituted algae: 0.08%



3.8 g algae paste in 3.5 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
600 mL


Conductivity
1730 uS/cm


pH


Pressure
N/A


Frequency
N/A


Voltage
36 V


Current
10 A


Run Time
1:9 min


Additional Details


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 30.









TABLE 17





Analysis Test-2A
















Test Setup
Solution of algae paste in nutrient-rich water run



through low pressure nozzle system, then tested



using empirical evaluation using reduced cross



sectional area laboratory test


Solution Type
Nutrient-rich water and algae paste


Concentration
18% dry weight



116.67 g algae paste in 30 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
600 mL


Conductivity
1730 uS/cm


pH
6.49


Pressure
N/A


Frequency
N/A


Voltage
36 V


Current
6 A


Run Time
1:00 min


Additional Details



Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 31.









TABLE 18





Analysis Test-2B
















Test Setup
Solution of algae paste in nutrient-rich water run



through low pressure nozzle system, then tested



using empirical evaluation using reduced cross



sectional area laboratory test


Solution Type
Nutrient-rich water and algae paste


Concentration
18% dry weight



116.67 g algae page in 30 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
600 mL


Conductivity
1730 uS/cm


pH
6.48


Pressure
N/A


Frequency
N/A


Voltage
36 V


Current
7.4 A


Run Time
1:20 min


Additional Details



Physical Observations
Rest period allows for full clarity to be achieved









Experimental setup, and processed sample produced, shown in FIG. 32.









TABLE 19





Analysis Test-3
















Test Setup
Solution of algae paste in nutrient-rich water



run through low pressure nozzle system, then



tested using empirical evaluation using reduced



cross sectional area laboratory test (with and



without sonication)


Solution Type
Nutrient-rich water and algae paste


Concentration
18% dry weight



116.67 g algae in 30 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
600 mL


Conductivity



pH



Pressure
Nozzle system run at 1200 psi


Frequency
See Matrix Testing chart



Nozzle system run at 30 Hz


Voltage
See Motrix Testing chart


Current
See Motrix Testing chart


Run Time
See Motrix Testing chart


Additional Details
For nozzle system, approximate flow was 3 L/min


Physical Observations
Experiment end point: water phase separation



Lipid-bubble attachment as top layer (3-4 inch



thick layer)



Lipid-bubble cluster contains both lipid and



algal phases



Lipid-bubble flotation (around 60-120 seconds)










Matrix Testing















Current
Run Time



Test Type
Sample
Voltage (V)
(A)
(sec)
Freq (Hz)





Capacitance
Raw
36
8
86
N/A


Separation
1st Run
36
8
111
N/A


Test
2nd Run
36
8
80
N/A



3rd Run
36
9
72
N/A


Capacitance
2nd Run
36
7
120
575


Separation +
3rd Run
36
7
83
575


Sonication







Test









Experimental setup, and processed sample produced, shown in FIG. 33.









TABLE 20





Analysis Test-4
















Test Setup
Solution of algae paste in nutrient-rich water run



through low pressure nozzle system, then tested



using empirical evaluation using reduced cross



sectional area laboratory test (with and without



sonication)


Solution Type
Nutrient-rich water and algae paste


Concentration
16% dry weight (0.5 g/L dry)



93.75 g algae paste in 30 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode


Volume
600 mL


Conductivity



pH



Pressure
Nozzle system run at 1200 psi


Frequency
See Matrix Testing chart



Nozzle system run at 30 Hz


Voltage
See Matrix Testing chart


Current
See Matrix Testing chart


Run Time
See Matrix Testing chart


Additional Details
When solution was run through nozzle system,



approximate flow was 3 L/min


Physical Observations
Experiment end point water phase separation



Lipid-bubble attachment as top layer (3-4 inch



thick layer)



Lipid-bubble cluster contains both lipid and



algal phases



Lipid-bubble flotation (around 60-120 seconds)










Nutrient-Rich Water and Algae Poste Solution Matrix Testing















Current
Run Time



Test Type
Sample
Voltage (V)
(A)
(sec)
Freq (Hz)





Electrolysis
Raw
36
8
114
N/A


Test
1st Run
36
7
108
N/A



2nd Run
36
7
123
N/A



3rd Run
36
9
99
N/A


Electrolysis +
Raw
36
7
120
575


Sonication Test
1st Run
36
7
121
575









Experimental setup, and processed sample produced, shown in FIG. 34.









TABLE 21





System Test-1
















Test Setup
Nozzle system test


Solution Type
Nutrient-rich water and algae paste


Concentration
0.5 g/L dry weight



25 g algae paste in 48 L nutrient rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal rod anode


Separation Setup
36 in. length


Volume
48 L


Conductivity
1860 uS/cm (initial)


pH



Pressure
Low pressure


Frequency
N/A


Voltage
19 V


Current
10 A


Run Time
25 hours


Resonance Time
240 seconds


Additional Details
Spiral coil cathode isolated from acrylic transparent tube by insulated



spacers



High conductivity metal rod anode fixed inside spiral coil cathode using



insulated and perforated rings



Inlet: tangential


Physical Observations
Water phase separation observed as solution is released into exit tank



Lipid-bubble attachment as top layer



Lipid-bubble cluster contains both lipid and algal phases










Electrode Setup














Spring
Material
Zinc and carbon based conductive materials




Outer Diameter
  1.5 in




Inner Diameter
1.176 in




Wire Diameter
0.162 in




Length
  36 in



Rod
Material
High conductivity activated electrodes - activated carbon,





titanium dioxide, iron and aluminum




Diameter
 0.5 in




Length
  36 in



Tube
Material
Acrylic




Outer Diameter
   2 in




Inner Diameter
 1.75 in




Thickness
0.125 in









Experimental setup, and processed sample produced, shown in FIG. 35.









TABLE 22





System Test-2
















Test Setup
Electrosorptive Continuous Cavitation


Solution Type
Nutrient-rich water and algae paste


Concentration
0.5 g/L dry weight



16.7 g algae paste in 32 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal rod anode


Separation Setup
36 in. length


Volume
32 L


Conductivity



pH



Pressure
Low pressure operation (250 psi)


Frequency
N/A


Voltage
12 V


Current
10 A


Run Time
1.5 hours


Resonance Time
240 seconds


Additional Details
Perforations in insulated rings around anode increased to allow for better



flow



Angle increased to approximately 32″



Inlet: tangential


Physical Observations
Water phase separation observed as solution is released into exit tank



Lipid-bubble attachment as top layer (thick layer in tank)



Lipid-bubble cluster contains both lipid and algal phases










Electrode Setup












Spring
Material
Zinc and carbon based conductive materials



Outer Diameter
 1.5 in



Inner Diameter
1.176 in



Wire Diameter
0.162 in



Length
  36 in


Rod
Material
High conductivity activated electrodes - activated carbon,




titanium dioxide/Zinc



Diameter
 0.5 in



Length
  36 in


Tube
Material
Acrylic



Outer Diameter
   2 in



Inner Diameter
 1.75 in



Thickness
0.125 in


Distance
Between Spring and Rod
0.338



Between Spring and Tube
0.125









Experimental setup, and processed sample produced, shown in FIG. 36.









TABLE 23





System Test-3
















Test Setup
Electrosorptive Continuous Cavitation


Solution Type
Nutrient-rich water and algae paste


Concentration
0.5 g/L dry weight



16 g algae paste in 32 L nutrient-rich water



(Concentration = 0.5 g/L)


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
32 L


Conductivity
Start: 1635 uS/cm



End: 1520 uS/cm


pH
Start: 6.53



End: 9.48


Turbidity
Start (6″ beaker): 25.6



End (6″ beaker): 4.0



End (1″ diameter graduated cylinder #1): 2.98



End (1″ diameter graduated cylinder #2): 1.53


Pressure
Low pressure


Frequency
N/A


Voltage
Start: 20 V



Middle: 28 V



End: 35 V


Current
10 A


Run Time



Resonance Time



Additional Details
Inlet: axial


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 37.









TABLE 24





System Test-4
















Test Setup
Nozzle system test


Solution Type
Nutrient-rich water and algae paste


Concentration
0.5 g/L dry weight



16 g algae paste in 32 L nutrient-rich water


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
32 L


Conductivity
Start: 1642 uS/cm



End: 1560 uS/cm


pH
Start: 6.55



End: 8.88


Turbidity
Start (6″ beaker): 29.80



End (6″ beaker): 9.30



End (6″ beaker, after 24 hours): 4.75



End (1″ diameter graduated cylinder #1): 8.72



End (1″ diameter graduated cylinder #2): 7.85


Pressure
250 and 500 psi


Frequency
11 Hz at 250 psi



16 Hz at 500 psi


Voltage
17 V


Current
10 A


Run Time



Resonance Time



Flow
14 L/min at 250 psi



2.0 L/min at 500 psi


Additional Details
Inlet: axial


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 38.









TABLE 25





System Test-5
















Test Setup
Electrosorptive Continuous Cavitation


Solution Type
Nutrient-rich water and algae paste


Concentration
0.1% dry


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
18 L


Conductivity
Start: 1544 uS/cm



End: 1403 uS/cm


pH
Start: 6.59



End: 8.30


Turbidity
Start (6″ beaker): 41.0



End (6″ beaker): 16.1



End (6″ beaker, after 7 hours): 5.09



End (1″ diameter graduated cylinder): 9.3



End (1″ diameter graduated cylinder, after



7 hours): 4.95


Pressure
250 psi


Frequency
11 Hz


Voltage
17 V


Current
10 A


Run Time
13 minutes 24 second


Resonance Time
72 seconds


Flow
1.4 L/min


Mass of lipid
79.97 g


Additional Details
Inlet axial


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 39.









TABLE 26





System Test-6
















Test Setup
Electrosorptive Continuous Cavitation


Solution Type
Nutrient-rich water and algae paste


Concentration
0.1% dry


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
18 L


Conductivity
Start: 1535



End: 1467


pH
Start: 6.53



End: 9.17


Turbidity
Start (6″ beaker): 46.8



End (6″ beaker): 28.7



End (6″ beaker, after 7 hours): 14.9



End (1″ diameter graduated cylinder): 15.5



End (1″ diameter graduated cylinder, after



7 hours): 9.55


Pressure
500 psi


Frequency
17 Hz


Voltage
16 V


Current
10 A


Run Time
8 minutes 38 seconds


Resonance Time
0:22 min


Flow
2.1 L/min


Mass of lipid
78.46


Additional Details
Inlet: axial


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 40.









TABLE 27





System Test-7
















Test Setup
Nozzle system test


Solution Type
Nutrient rich water and algae paste


Concentration
0.1% dry


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
60 L


Conductivity
Start: 1560



End: 1459


pH
Start: 6.53



End: 9.94


Turbidity
End (after 30 min): 42.0 NTU


Pressure
Low pressure


Frequency



Voltage
26 V


Current
10 A


Run Time



Resonance Time
2:25 min


Flow
0.375 L/min


Mass of lipid



Additional Details



Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 41.









TABLE 28





System Test-8
















Test Setup
Electrosorptive Continuous Cavitatior


Solution Type
Nutrient rich water and algae paste


Concentration
0.1% dry


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
32 L


Conductivity



pH



Turbidity



Pressure
Low pressure


Frequency



Voltage
19 V


Current
10 A


Run Time



Resonance Time



Flow



Mass of lipid



Additional Details
Sample collected into beakers of increasing radius



for physical observation of algae separation



over time. Three setups were used



I Cylinder separatior with low pressure system



II Cylinder separatior with low pressure system



and additional inlet tube



III Cylinder separatior with 250 psi system and



additional inlet tube


Physical Observations
See photos









Experimental setup, and processed sample produced, shown in FIG. 42.









TABLE 29





System Test-9
















Test Setup
Electrosorptive Continuous Cavitatior


Solution Type
Nutrient rich water and algae paste


Concentration
0.5% dry


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
40 L


Conductivity
1525 uS/cm


pH



Turbidity



Pressure
250 psi


Frequency
 8 Hz


Voltage
15 V


Current
10 A


Run Time
1 hour


Resonance Time
1:20 min


Flow
2.1 L/min


Mass of lipid
78.46


Additional Details
Inlet: tangential


Physical Observations









Experimental setup, and processed sample produced, shown in FIG. 43.









TABLE 30





System Test-10
















Test Setup
Electrosorptive Continuous Cavitatior


Solution Type
Nutrient rich water and algae paste


Concentration
0.05% dry


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
40 L


Conductivity
Start: 1477 uS/cm



End: 714 uS/cm


pH



Turbidity:
End: 4.07 NTU



End (after 3.4 hours): 3.14 NTU


Pressure
Low pressure


Frequency
8 Hz


Voltage
15 V


Current
10 A


Run Time
1 hour


Resonance Time
1:20 min


Flow
2.1 L/min


Mass of lipid



Additional Details
Inlet: tangential



Collectior in 3 L capacity tube


Physical Observations
Algae mass settling at bottom of collection tube



Lipid remaining at top of collection tube









Experimental setup, and processed sample produced, shown in FIG. 44.









TABLE 31





System Test-11
















Test Setup
Electrosorptive Continuous Cavitatior Test



complete a twice first at 250 psi and then



repeated with extremely low pressures


Solution Type
Nutrient rich water and algae paste


Concentration
0.05% dry


Additives to Culture
None


Capacitance
Spiral coil cathode with high conductivity metal


Separation Setup
rod anode 36 in. length


Volume
40 L


Conductivity
1515 uS/cm


pH



Turbidity



Pressure
250 psi test:



Low pressure test:


Frequency



Voltage
250 psi test: 16 V



Low pressure test: 10 V


Current
250 psi test: 10 A



Low pressure test: 10 A


Run Time
250 psi test



Collection time in vertical 3.5 L collectior tube =



3:00 min



Settling time = 2 hours (not completely settled)



Complete run time = 11:00 min



Low pressure test:



Collection time in vertical 2.5 L collectior tube =



10:00 min


Resonance Time
250 psi test:



Low pressure test: 3:00 min


Flow



Mass of lipid
Low pressure test: Lipid mass = 1000.86 g


Additional Details



Physical Observations
250 psi test: After 2 hours of settling time,



collection tube not cleared completely, cannot



observe full separatior of phases



Low pressure test: Much clearer separation of



phases in less time









Experimental setup, and processed sample produced, shown in FIG. 45.


Example 4—FAME Analysis for Lipid Extract and Dried Algal Biomass

FAME (Fatty Acid Methyl Ester) analysis experiments were performed to assess lipid extracts and dried algal biomass products produced according to methods described herein (see Example 5).


The following table (Table 32) provides results of FAME analysis of 4 samples: CV2 Biomass, CN2 Biomass, CV2 Lipid, and CN1 Lipid. Experimental protocol is provided in Example 5. Two biomass samples (CN2 and CV2) from the 1500 psi and 18 psi experimental runs in Example 5 were submitted for spectroscopic analysis and 1 lipid sample from the 18 psi run (CN1) was also analyzed. Lipids were extracted from CV2 biomass with solvents using modified Bligh and Dwyer method and then analyzed as shown in Table 32.


Materials and Methods

    • FAME standards: Supelco® 37 component FAME mix and PUFA-3 (Supelco, USA)
    • Methyl tricosanoate (23:0 Methyl ester): Sigma-Aldrich, USA
    • BF3 Solution in MeOH: Sigma-Aldrich, USA
    • NaOH: Sigma-Aldrich, USA
    • Hexane: Caledon, Canada


Experimental Protocol


Direct trans-esterification-saponification by NaOH followed by BF3-catalyzed methylation was performed. Protocol was as follows:

    • 1. Add 10 mg (or exact weight in mg) of sample in a dry screw-capped vial.
    • 2. Add 1.0 mL MeOH containing 0.1 mg internal standard (methyl tricosanoate; 23:0 methyl ester).
    • 3. Sonicate for 10 min.
    • 4. Add 0.5 mL 1.5 N NaOH solution in MeOH, blanket with nitrogen, cap and mix.
    • 5. Heat 5 min at 100° C. and cool.
    • 6. Add 1.0 mL 14% BF3 solution in MeOH blanket with nitrogen, mix and heat at 100° C. for 30 min.
    • 7. Add 0.5 mL distilled water and mix.
    • 8. Add 2.0 mL hexane, blanket with nitrogen and vortex for 30 sec.
    • 9. Centrifuge using benchtop centrifuge (Fisher Scientific centrific model 228) for 1 min.
    • 10. Transfer hexane layer (300-600 μL) to a GC vial, blanket with nitrogen and store the sample at −20° C. before analysis.
    • 11. Mass of EPA and DHA in the sample will be calculated as mg/g sample, as follows:

      EPA or DHA, mg/g=(Ax·WIS·CFx/AIS·WS·1.04)·1000


Where AX=area counts of EPA or DHA; AIS=area counts of internal standard (tricosylic acid methyl ester), CFX=theoretical detector correlation factor for EPA or DHA (0.99 for EPA, 0.97 for DHA); WIS=weight of IS added to sample in mg; WS=sample weight in mg; and 1.04 is factor necessary to express result as mg fatty acid/g sample.


The weight of the rest of the fatty acids were also calculated using the same formula as described above, assuming that all fatty acids regardless of the structure have the same response as the internal standard (i.e. theoretical detector correlation factor is equal to 1).


Reference: AOAC official method 991.39. Fatty acids in encapsulated fish oils and fish oil methyl and ethyl esters. Gas chromatography methods.









TABLE 32







Fatty Acid (FA) content, expressed in mg/g sample











Fatty Acid (FA)
CV2 biomass
CN2 biomass
CV2 lipid
CN1 lipid





C8:0






C10:0






C11:0






C12:0 (Lauric acid)






C13:0 (Tridecanoic acid)






C14:0 (Myristic acid)



3.29


C14:1 (Myristoleic acid)

0.18
1.29
1.05


C15:0 (Pentadecanoic acid)


0.45
3.88


C15:1 (cis-10-Pentadecenoic acid)


0.30



C16:0 (Palmitic add)
2.60
0.75
24.39
107.60


C16:1 n-7 (Palmitoleic acid)


1.21



C17:0 (Heptadecanoic acid)


0.97
1.22


C17:1 (cis-10-Heptadecenoic acid)


3.97
0.98


C18:0 (Stearic acid)
0.21
0.08
2.67
10.13


C18:1 n-9c/t (Oleic acid/Elaidic acid)
0.75
0.51
13.33
47.06


C18:1 n-7


1.65



C18:2 n-6c/t (Linoleic acid/Linolelaidic acid)
0.49
0.11
22.05
5.09


C18:3 n-6 (γ-linolenic acid)


0.16



C18:3 n-3 (α-linolenic acid)
0.15
0.08
0.12
3.71


C20:0 (Arachidic acid)


0.33



C20:1 n-9 (cis-11-Eicosenoic acid)


0.32
1.75


C20:2 (cis-11,14-Eicosadienoic acid)


0.08



C20:3 n-6 (cis-8,11,14-Eicosatrienoic acid)


0.48



C21:0 (Henicosanoic acid)


0.07



C20:3 n-3 (cis-8,11,14-Eicosatrienoic acid)


0.37
3.37


C20:4 n-6 (Arachidonic acid)


0.04
1.11


C20:5 n-3 (Eicosapentaenoic acid)


0.07



C22:0 (Behenic acid)


0.22
0.42


C22:1 n-9 (Erucic acid)


0.45
0.43


C22:2 n-3 (Docosadienoic acid)


0.68
1.67


C24:0 (Lignoceric acid)


0.07



C22:6 n-3 (Docosahexaenoic acid)






unknown RT 6.419
0.15

1.70
6.54


unknown RT 6.674
0.20

7.14
1.54


unknown RT 11.743


10.45






(—) not detected






Example 5—Additional Testing of Algal Aqueous Mixtures

Two high pressure tests (1500 psi) were performed, and in both tests no changes occurred after 30 min and 60 min settling periods (i.e. no separation was seen in collection tanks located at the top of the vertical system). During the low pressure tests, a previous batch of C. sorokiniana showed very clear separation in the collection vessel. See Example 6 for experimental protocol details. The high and low pressure experiments performed at 1500 psi and at 18 psi described are the “1” and “2” samples reported in Table 32.



FIG. 46 shows depictions of algal cultures, cavitation test setups, low pressure cavitation setups, and processed algal products obtained following treatment.


Example 6—Symbiotic Culture Evaluation

Experimental Protocol:


Algal culture (Chlorella vulgaris) was received from Symbiotic Algae Inc. The received culture was obtained nutrient-depleted and with a low conductivity of 680 uS/m and was not stressed during operation to produce lipids. To find out the dry content of wet algae, 2.0 g of wet algae was dried in an oven overnight at 105° C. and the dry content was calculated. Multiple tests were performed using 400 ml (lab scale) to 20 L batches (pilot scale) in the electrosorptive cavitation system under low to medium to high pressure cavitation (18 psi, 250 psi, 800 psi and 1100 psi). The system is connected to an AC power input (36 V, 10 amps) that serves to enhance the advanced oxidation generated by cavitating nuclei and results in lipid-biomass-water phase separation through lipid-bubble interactions and reduced polarity of the algal cells rendering their effective settling in a single step. There are several experiments carried out to validate the developed system from laboratory scale to pilot scales.


In certain embodiments, electrosorptive cavitation may be a continuous process and may be further designed to operate in a single step without the addition of added chemicals, flocculants, adsorbents, or pH buffer agents beyond the nutrients supplied to the culture during the algal growth in photobioreactors. Electrosorptive cavitation process may include a batch/continuous electrosorptive-cavitating flow system to perform both extraction and separation processes by applying a complex cavitating flow which applies a high pressure to a tangential flow inlet.


A summary showing differences between low and high pressure electrosorptive cavitation is benchmarked against a system with electrosorption only:

















System test using






electrosorptive
System test using
System test using
System test using


Test Setup
cavitation
electrosorptive cavitation
electrosorptive cavitation
electrosorption only







Solution Type

Chlorella vulgaris with


Chlorella vulgaris with


Chlorella vulgaris with


Chlorella vulgaris with




F2 media
F2 media
F2 media
F2 media


Additives to Culture
None
None
None
None


Volume
20 L
20 L
20 L
20 L


Conductivity
Raw sample: 680 μs/cm
Raw sample: 680 μs/cm
Raw sample: 680 μs/cm
Raw sample: 680 μs/cm


pH
7.10
7.10
7.10
7.1


Pressure
250 psi
800 psi
1100 psi
18 psi


Frequency
11 Hz
20 Hz
20 Hz



Run Time
7:57 min
5:05 min
5:15 min
25:40 min


Resonance Time
1:17 min
0:57 min
00:50 min
6:15 min


Flow
0.9 L/min
1.8 L/min
2 L/min
0.2 L/min


Mass of Algae and
78.6 (g), centrifuged
54.21 (g), centrifuged
91.72 (g), centrifuged
24.46 (g), centrifuged


water
with 1000 rpm for 10 min
with 1000 rpm for 10 min
with 1000 rpm for 10 min
with 1000 rpm for 10 min


Physical
After 3:00 min settling
After 3:30 min settling
Brown color in raw sample.
After 3:00 min settling


Observations
time, water in float cell
time, water in float cell
Water in float cell
time, water in float cell



appeared clear.
appeared clear.
appeared clear
was not very clear.









Additional experiments testing treatment of algal culture were performed. Algal culture (Chlorella vulgaris) was tested. The culture was nutrient depleted, and was reported with a low conductivity of 680 μS/m and was not stressed during operation to produce lipids. Two pails of samples were brownish in color, with a distinct odour of degradation. It can be inferred that the samples may have partially degraded due to storage time and/or temperature.


Multiple tests were performed using 20 L batches and electrosorptive cavitation treatment. Results are shown below for both low and high pressure cavitation. Algae cells were analyzed using a microscope and were observed to be between 2 and 7 microns in size. The culture was separated and water samples and both culture and water samples are prepared for spectroscopic analysis. The conditions of operation and collected results are summarized in the following


Tables and Figures.









TABLE 33







Symbiotic Culture Evaluation - System Test 1










Test Setup
System test using nozzle







Solution Type

Chlorella vulgaris algae





with nutrient depleted media



Concentration




Additives to Culture
None



Capacitance Separation
Spiral coil cathode (steel) with high



Setup
conductivity metal rod anode




(aluminum) 36 in. length



Volume
20 L



Conductivity
Raw sample: 680 μs/cm



pH
Raw sample: 7.10



Turbidity




Pressure
250 psi



Frequency
11 Hz



Voltage
36.1 V (constant voltage)




34 V (after 5:00 min,




switched to constant current)



Current
7.7 A (constant voltage)




10 A (after 5:00 min, switched to




constant current setting)



Run Time
7:57 min



Resonance Time
1:17 min



Flow
0.9 L/min



Mass of lipid




Mass of Algae and H2O
78.6 (g), centrifuged with 1000



after centrifuged
rpm for 10 min



Additional Details
Pail 1 of 3



Physical Observations
After 3:00 min settling time,




water in float cell appeared clear.










Experimental setup, and processed sample produced, shown in FIG. 47.









TABLE 34







Symbiotic Culture Evaluation - System Test 2








Test Setup
System test using nozzle





Solution Type

Chlorella vulgaris algae with nutrient




depleted media


Concentration



Additives to Culture
None


Capacitance Separation
Spiral coil cathode (steel) with high


Setup
conductivity metal rod anode



(aluminum) 36 in. length


Volume
20 L


Conductivity
Raw sample: 680 μs/cm


pH
Raw sample: 7.10


Turbidity



Pressure
800 psi


Frequency
20 Hz


Voltage
33.40 V (constant voltage)



32 V (after 4:00 min, switched to



constant current setting)


Current
10.10 A (constant voltage)



10 A (after 4:00 min, switched to



constant current setting)


Run Time
5:05 min


Resonance Time
0:57 min


Flow
1.8 L/min


Mass of lipid



Mass of Algae and H2O
54.21 (g), centrifuged with 1000


after centrifuged
rpm for 10 min


Additional Details
Pail 1 of 3


Physical Observations
After 3:30 min settling time,



water in float cell appeared clear.









Experimental setup, and processed sample produced, shown in FIG. 48.









TABLE 35







Symbiotic Culture Evaluation - System Test 3








Test Setup
System test using nozzle





Solution Type

Chlorella vulgaris algae with nutrient




depleted media


Concentration



Additives to Culture
None


Capacitance Separation
Spiral coil cathode (steel) with high


Setup
conductivity metal rod anode



(aluminum) 36 in. length


Volume
20 L


Conductivity
Raw sample: 680 μs/cm


pH
Raw sample: 7.10


Turbidity



Pressure
18 psi


Frequency



Voltage
36 V


Current
7.8 A (at start)



9 A (after 17:00 min)



8 A (after 24:00 min)


Run Time
25:40 min


Resonance Time
6:15 min


Flow
0.2 L/min


Mass of lipid



Mass of Algae and H2O
24.46 (g), centrifuged with 1000


after centrifuged
rpm for 10 min


Additional Details
Pail 2 of 3


Physical Observations
After 3:00 min settling time, water



in float cell was not very clear.



Brown color in raw sample.









Experimental setup, and processed sample produced, shown in FIG. 49.









TABLE 36







Symbiotic Culture Evaluation - System Test 4








Test Setup
System test using nozzle





Solution Type

Chlorella vulgaris algae with




nutrient depleted media


Concentration



Additives to Culture
None


Capacitance Separation
Spiral coil cathode (steel) with high


Setup
conductivity metal rod anode



(aluminum) 36 in. length


Volume
20 L


Conductivity
Raw sample: 680 μs/cm


pH
Raw sample: 7.10


Turbidity



Pressure
1100 psi


Frequency
20 Hz


Voltage
36 V


Current
7.8 A (after 2:00 min)



8.5 A (after 3:00 min)



9.8 A (after 4:30 min)


Run Time
5:15 min


Resonance Time
00:50 min


Flow
2 L/min


Mass of lipid



Mass of Algae and H2O
91.72 (g), centrifuged with 1000


after centrifuged
rpm for 10 min


Additional Details
Pail 2 of 3


Physical Observations
Brown color in raw sample.









Experimental setup, and processed sample produced, shown in FIG. 50.









TABLE 37







Symbiotic Culture Evaluation - System Test 5










Test Setup
System test using nozzle







Solution Type

Chlorella vulgaris algae with





nutrient depleted media



Concentration




Additives to Culture
None



Capacitance Separation
Spiral coil cathode (steel) with high



Setup
conductivity metal rod anode




(aluminum) 36 in. length



Volume
20 L



Conductivity
Raw sample: 680 μs/cm



pH
Raw sample: 7.10



Turbidity




Pressure
1100 psi



Frequency
20 Hz



Voltage
36 V



Current
8 A (at start)




9.5 A (after 3:00 min)



Run Time
4:00 min



Resonance Time
00:40 min



Flow
2 L/min



Mass of lipid




Mass of Algae and H2O
82.24 (g), centrifuged with 1000



after centrifuged
rpm for 10 min



Additional Details
Pail 3 of 3



Physical Observations
3:00 min settling time










Experimental setup, and processed sample produced, shown in FIG. 51.









TABLE 38







Symbiotic Culture Evaluation - System Test 6










Test Setup
System test using nozzle







Solution Type

Chlorella vulgaris algae with





nutrient depleted media



Concentration




Additives to Culture
None



Capacitance Separation
Spiral coil cathode (steel) with high



Setup
conductivity metal rod anode




(aluminum) 36 in. length



Volume
20 L



Conductivity
Raw sample: 680 μs/cm



pH
Raw sample: 7.10



Turbidity




Pressure
250 psi



Frequency
11 Hz



Voltage
36 V



Current
6.6 A (after 1:30 min)




9.3 A (after 7:00 min)



Run Time
7:55 min



Resonance Time
1:18 min



Flow
1 L/min



Mass of lipid




Mass of Algae and H2O
154.56 (g), centrifuged with 1000



after centrifuged
rpm for 10 min



Additional Details
Pail 3 of 3



Physical Observations
3:00 min settling time










Experimental setup, and processed sample produced, shown in FIG. 52.


Example 7—Metals Separation and Processing of Mining Affected Water

Experiments were performed with mine water, and metals separation was assessed following treatment. The experimental results demonstrated a significant reduction in metals content in the water.


Mine affected water was subjected to electrosorptive cavitation in a laboratory scale system. Mine affected water, rich in aluminium, iron and sulphate was batch tested with electrosorptive cavitation in the system described earlier for algae-water-lipid extraction and phase separation.


During the test, 20 L of mine water was subjected to electrosorptive cavitation and samples were collected for ICP analysis. ICP results are shown below:









TABLE 39







Metals Separation from Mining Water










Feed
Electrosorptive


Concentration (ppm)
water
Cavitation












Ag




Al
90.64
31.77


As




B




Ba
0.0043
0.0074


Be
0.0032



Bi




Ca
188.4
179


Cd

2.84


Co
0.321
0.277


Cr




Cu
1.74
8.48


Fe
155.2
10.77


K
1.44
1.66


Li




Mg
97.55
92.94


Mn
6.94
9.63


Mo




Na
21.48
21.27


Ni
0.277
0.142


P




Pb




S
656.8
345.2


Sb




Se




Si
14.9
7.48


SO4
1968
1034


Sr
0.402
0.389


Te




Ti




Tl




V




Zn
0.914
1.06


Zr
0.006









Example 8—Algal Culture Processing Using Electrosorptive Cavitation

1. Extraction of Lipids from Chlorella sorokiniana



Chlorella sorokiniana samples treated by cavitation demonstrated an increasing trend of crude lipid recovery relative to biomass loading, approaching 50% recovery in the lipid fraction at 0.1 wt % dry solids biomass loading. Results are shown below in Table 40, and in FIG. 53.









TABLE 40







Lipid Recovery Using Electrosorptive Cavitation


on Chlorella sorokiniana samples









Conductivity (μS/cm)











Algal Biomass
Lipid recovery
lipid/biomass
Before
Post-


(wt %)
(mg)
(%)
treatment
treatment














0.021
0.5
11.90
6.9
17.3


0.052
3
28.85
14.9
16.9


0.078
5.5
35.26
17.5
20.6


0.105
9.5
45.24
20.4
38.2









2. Conductivity Analysis of Different Algal Cultures


Results are shown below in Table 41. With the exception of C. vulgaris, the culture conductivity generally increased following ultrasound treatment compared to untreated samples, indicating the release of polar intracellular components.









TABLE 41







Conductivity of Different Algal Cultures Before and After Treatment












Conductivity




Loading
(μS/cm)













Strain
(% w/v)
Pre
Post
















Botryococcus #66-02

0.008
9.7
13.5




0.020
19
15.3




0.030
17
19




0.040
19
21




Chlorella vulgaris #208-01

0.023
9.1
8.5




0.057
19.2
13.5




0.086
13.5
5




0.114
20.5
25




Chlorella sorokiniana #282-02

0.021
6.9
17.3




0.052
14.9
16.9




0.078
17.5
20.6




0.105
20.4
38.2




Scenedesmus sp. #173-13

0.024
7.9
9.5




0.060
13.8
15




0.090
17.8
21.3




0.120
38.7
39.8









Example 9—Viscosity Analysis Using Tri-County Protein Soybean Oil

Soybean oil, a high viscosity liquid, was subjected to cavitation and did not demonstrate a significant change using either low or high frequency cavitation. 200 mL samples were treated with varying times and duty cycles at varying frequency at 4/6 amplitude. Viscosity was recorded pre and post-treatment. Results are shown below in Table 42.









TABLE 42







Soybean oil viscosity results following cavitation.













Cavitation Period



Sample
Viscosity (cP)
(min.)














soybean oil untreated
63.4
0



soybean oil − near
54.4
2.5



sonic frequency





soybean oil + higher
54.4
2.5



frequency





soybean oil +
66.4
5



hydrodynamic





cavitation





soybean oil higher
63.7
5



frequency/cavitation









As a slight change was observed in the case of lower frequency operations, electrosorptive cavitation at higher flow rates may allow for a higher removal efficiency.


Example 10—Electrosorptive Cavitation with B. braunii

Dispersions were prepared in 20 L BBM/tap water at 720 μS and used for two runs at low/high pressure. Lipids accumulated in the collections float cell were skimmed along with the biomass with clear high conductivity water without biomass in the continuous operation. Due to lipid bubble attachment, the cells were found to be trapped in the interstitial spaces between the lipid-bubble generated froth. To overcome this separation issue, in the last trials, collected foam was broken in a sieve, centrifuged, and the top foam layer collected by aspiration for CHCl3 extraction to achieve better separation with minimal water present. Solids separated by centrifugation were freeze-dried to determine the final dry biomass to establish stoichiometry. Results are shown below in Table 43.









TABLE 43







Large-scale lipid recovery results with B. Braunii









Phases










Lipid
Biomass













Feed Culture
Biomass
Lipid
wt
wt















Pressure
Solids
Solids
recovered
recovered
recovered
starting


Sample
(psig)
(wt %)
(g)
(g)
(mg)
biomass
biomass


















Botryococcus

300
0.094117
5.65
1.5901
8
0.14
0.50


#66-02










Botryococcus

1100
0.094117
5.65
1.1043
8.5
0.15
0.77


#66-02









During the operation, the electrosorptive system was operated at 30V DC and 10 A to obtain clear water flow without any lipids or biomass.


Example 11—Nanochloropsis Strain Grown with Marine Media with Low Lipid Yield
















Concentration

Conductivity
Lipid



(wt %)
Treatment
μS
(mg)








0.019434
DTE
79.3
4.5



0.019434
eCav
85.1
6.5









During this study, the focus was on determining the technology potential for low lipid yielding algae and to compare the potential of the electrosorptive cavitation method of the present application (eCav) to electrosorption or cavitation independently. Lipid extraction recovery was greater following treatment, and conductivity also increased indicating lipid release.


Untreated samples were compared to cavitation, 5 min 36 V electrosorption, and electrosorptive cavitation treatment. Micrographs of treatments (FIG. 54) were obtained to determine the benefits of the individual processes and compare them to electrosorptive cavitation treatment. Results are shown below in Table 44.









TABLE 44







Comparison of Cavitation, Electrosorption, and Electrosorptive


cavitation on Conductivity and Biomass Recovery











Concentration



Recovered


(wt %)
Treatment
Conductivity
pH
Biomass (g DM)














0.097171
Control
408
8.2
0.135


0.097171
Cavitation
429
7.78
0.123


0.097171
Electrosorption
334
8.5
0.236


0.097171
Electrosorptive
338
8.67
0.206



cavitation









Results:


Biomass yield was nearly 2× for electrosorption treatment as compared to untreated samples or treatment by cavitation alone, and was similar to electrosorptive cavitation, where the precipitates of the ionic species were not observed. Supernatants following centrifugation after electrosorption were much clearer than untreated samples or those treated by cavitation alone. Micrographs (FIG. 54) demonstrated planktonic cells in untreated and cavitation treatment samples, while samples treated by electrosorption and electrosorptive cavitation showed agglomeration.


Due to the brackish nature of the growth media, the system is subjected to other contaminants and the precipitation of the ionic species is critical and can be represented by reduced conductivity.


Electrosorptive cavitation conductivity decrease is substantially higher than cavitation as the advanced oxidative process initiated during cavitation is further enhanced by the immediate initiation of electrosorption, an advanced oxidation process that does not release lipids but the initiated cavitation voids propagate further energy dissipation and resulting added lipid due other than promoting agglomeration of algal cells. Thus, an indication of the benefits of electrosorptive cavitation is provided through the higher conductivity and pH within the solution. To confirm the findings, the micrographs shown in FIG. 54 show agglomeration of the cells throughout the solution using electrosorptive cavitation. This was not observed in the case of electrosorption treatment, where sporadic areas of agglomeration were observed.


Example 12—Mine Affected Water Treatment Using Electrosorptive Cavitation

1. Development of an Optimum Condition for Selenium Removal from Coal Mine Affected Waters


A range of advanced oxidation treatment methods were applied in addition to electrosorptive cavitation on coal mine waters to determine the removal efficiency. Ultrasound based methods are limited for large scale mine waters. The objective of this work was to develop an approach to address high volume (throughput) treatment to remove selenium from coal mining affected waters.


Modified coal mining effluent water with initial pH 7.68. Different experimental conditions as below:

    • S1: Cavitation treatment with frequency of 574 kHz, power input 50%, sonication time 30 min.
    • S2: Cavitation treatment with frequency of 40 kHz, amplitude 50%, sonication time 1 hour.
    • S3: Cavitation treatment with frequency of 20 kHz, amplitude 80%, sonication time 1 hour.
    • S4: Oxidation treatment by adding 1 ml H2O2, 0.2 g FeSO4.7H2O, pH 7.68, shaking for 1 h at 200 rpm.
    • S5: Oxidation-Cavitation treatment with addition of 0.2 g FeSO4.7H2O, pH 7.68 with frequency of 40 kHz, power input 80%, sonication time 1 hour, sample volume 350 mL
    • S6: Ozonation treatment with ozone dose 60 L/h, ozonation time 30 min, sample volume 350 mL, pH 7.68
    • S7: Electrosorptive cavitation treatment
    • S8: Oxidation treatment by adding 1 ml H2O2, 0.2 g FeSO4.7H2O, pH (2.5-3.5), shaking for 1 h at 200 rpm, sample volume 350 mL


The selenium concentration was measured by using UV-Vis spectrophotometry after each treatment. The tests were repeated three times. The results were compared with the blank solutions with known concentration of selenium. All the results obtained by UV-Vis were compared with ICP-MS results for validation of the method. The UV-Vis spectra of sample analysis are presented in FIG. 55. Different results obtained during the treatment process are presented below in Table 45.









TABLE 45







Selenium concentration obtained by UV-Vis and ICP-MS


after different treatments of coal mining affected waters.










Total selenium
















ICP-MS
COD
Ammonia-N



Sample name
UV-VIS
(mg/L)
mg/L
mg/L
















Blank

0.398





Effluent
0.5961
0.587
99.6
3.58



S1
0.4843
0.420

2.6



S2
0.4902
0.416
72
2.36



S3
0.3676
0.390
67.2
2.89



S4
0.3524
0.392
80
2.79



S5
0.3473
0.364
81.6
2.08



S6
0.5127
0.447
73.6
3.48



S7-
0.4931
0.430
72
3.47



Electrosorption









Tuning of Cavitation System: Using the above conditions, cavitation in the electrosorptive cavitation system was operated at a reduced pressure to only form incipient cavities.















Cavitation Pressure
Selenium concentration in effluent



Psi
mg/L


















1500
430



1100
170



600
78



300
19



250
8 (incipient cavitation)









2. Gold Mine Process Water Treatment


i. Selenium Removal


A sample of gold mine processing water, sample CAV-S2, was obtained from a gold refining operation. Treatment was focused on selenium removal only. As seen below in Table 46, substantial reduction of selenium from the original mine water (at a concentration of Se=1 ppm) to 0.369 ppm was achieved through this treatment. Results for samples CAV-S10, CAV-S11 AND CAV-S12 show the effects of change in cavitation pressure on selenium removal in the electrosorptive cavitation system.









TABLE 46





Results for the removal of selenium from gold processing waters using


electrosorptive cavitation.





















Aluminum
Antimony
Silver
Arsenic
Barium













Sample
Flow
Sample
Date
Unit


















No
(l/min)
Description
Sampled
Parameter
μg/L
μg/L
μg/L
μg/L
μg/L










G/S
5-100

0.26
6.0



Raw Water


CAV-S-2
Jul. 28, 2017
8600542
9.00
1.00
<0.2
10.3
10.00







RDL
5
1
0.2
0.3
2


Condition












300 psi 9 Hz-
10.300
1.104
CAV-S-10
Jul. 28, 2017
8600550
206.00
1.00
<0.2
8.2
10.00


9.1 V, 9.7 A



















50sek-



920 ml

















600 psi 12 Hz
11.600
1.48
CAV-S-11
Jul. 28, 2017
8600551
393.00
1.00
<0.2
20.6
10.00


8.9 V, 9.7 A



















30sek-



740 ml

















900 psi, 16 Hz
12.900
1.89
CAV-S-12
Jul. 28, 2017
8600552
12.00
1.00
<0.2
6.8
30.00


11.0 V 9.9 A



















30sek-



945 ml

















Boron
Cadmium
Chromium
Cobalt
Copper
Manganese












Sample
Flow
Sample
Unit

















No
(l/min)
Description
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L










0.09

0.002-0.004
2-4



Raw Water


CAV-S-2
<40
0.2
<1
<0.5
6.00
3.00






40
0.1
1
0.5
1
1


Condition











300 psi 9 Hz-
10.300
1.104
CAV-S-10
77.00
0.3
2.00
<0.5
10.00
22.00


9.1 V, 9.7 A


















50sek-



920 ml
















600 psi 12 Hz
11.600
1.48
CAV-S-11
34.00
0.3
2.00
<0.5
10.00
18.00


8.9 V, 9.7 A


















30sek-



740 ml
















900 psi, 16 Hz
12.900
1.89
CAV-S-12
93.00
0.4
<1
<0.5
14.00
1.00


11.0 V 9.9 A


















30sek-



945 ml

















Molybdenum
Nickel
Lead
Selenium
Sodium
Zinc












Sample
Flow
Sample
Unit



















No
(l/min)
Description
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L










73
25-150
1-7
1

30



Raw Water


CAV-S-2
2.00
46.00
<0.5
1,010.00
17.90000
9.00







1
1
0.5
1
 2
3



Condition












300 psi 9 Hz-
10.300
1.104
CAV-S-10
2.00
22.00
<0.5
383.00
17.10000
6.00



9.1 V, 9.7 A


















50sek-



920 ml


















600 psi 12 Hz
11.600
1.48
CAV-S-11
2.00
24.00
<0.5
658.00
17.10000
8.00



8.9 V, 9.7 A


















30sek-



740 ml


















900 psi, 16 Hz
12.900
1.89
CAV-S-12
2.00
70.00
<0.5
369.00
17.40000
10.00



11.0 V 9.9 A


















30sek-



945 ml

































Parameter



























Anti-


Bar-


Chrom-








Aluminum
mony
Silver
Arsenic
ium
Boron
Cadmium
ium



















Unit























Sample
Date

μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
















Descrip-
Sam-

G/S























tion
pled

5-100

0 25
5 0


0 09






Raw Water


CAV.S.2
Jul. 28,
8600542
9 00
1 00
<0.2
10.3
10.00
<40
0.2
<1






2017

















Sample
Flow



RDL




















Condition
No
(l/min)



5
1
0.2
0 3
2
40
0.1
1





300 psi.
10-300
1 104
CAV.S.10
Jul. 28,
8600550
206 00
1 00
<0 2
 8.2
10 00
77 00
0.3
  2.00


9 Hz



2017




























9 1 V. 97A


50 sek-














920 ml





























600 psi.
11-600
1 48
CAV.S.11
Jul. 28,
8600551
393 00
1 00
<0 2
20.6
10 00
84 00
0.3
  2 00


12 Hz



2017




























8 9 V 97A


30 sek-














740 ml





























900 psi.
12-900
1.89
CAV.S.12
Jul. 28,
8600552
 12 00
1 00
<0 2
 6.8
30 00
93 00
0.4
<1


16 Hz



2017




























11 0 V 9.9A


30 sek-














945 ml




















Parameter



























Man-
Molyb-


Selen-









Cobalt
Copper
ganese
denum
Nickel
Lead
ium
Sodium
Zinc

















Unit

























μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L















Sample
Date
G/S























Descrip-
Sam-
0 002-













tion
pled
0 004
2-4

73
25-150
1-7
1

30





Raw Water


CAV.S.2
Jul. 28,
<0 5
6 00
3 00
2 00
46 00
<0.5
1,010.00
17.900 00
9 00






2017
















Sample
Flow


RDL




















Condition
No
(l/min)


0.5
1
1
1
1
0.5

2
3





300 psi.
10-300
1 104
CAV.S.10
Jul. 28,
<0 5
10.00
22.00
2 00
22.00
<0 5
383.00
17.100 00
 6 00


9 Hz



2017




























9 1 V. 97 A


50 sek-














920 ml





























600 psi.
11-600
1 48
CAV.S.11
Jul. 28,
<0 5
10 00
18 00
2 00
24 00
<0 5
658.00
17.100 00
 8 00


12 Hz



2017




























8 9 V 97 A


30 sek-














740 ml





























900 psi.
12-900
1.89
CAV.S.12
Jul. 28,
<0 5
14 00
 1 00
2 00
70 00
<0 5
369.00
17.400 00
10 00


16 Hz



2017




























11 0 V 9.9 A


30 sek-














945 ml









i. Salt Removal


Treatment of gold mining and extraction waters that are not selenium rich necessarily contain contaminants, such as salt and copper. Table 47 shows the salt and copper removal efficiency at incipient cavitation conditions of 250 psi and electrosorption operation a 30 V and 10 A.









TABLE 47







Results for the removal of salt and copper from gold


processing waters using electrosorptive cavitation












Initial
Controlled


Assays
Gold Mining
Cavitation
Cavitation












Metals
Unit
RDL
Affected Water
Processing
after Tuning















Aluminum
μg/L
50
18.00
18,800.00
541.00


Antimony
μg/L
1
434.00
433.00
2.00


Silver
μg/L
0.2
0.4
0.5
<0.2


Arsenic
μg/L
3
586.00
547.00
<3


Barium
μg/L
2
49.00
55.00
60.00


Boron
μg/L
40
586.00
560.00
<40


Cadmium
μg/L
0.2
<0.1
<0.1
<0.2


Chromium
μg/L
1
2.00
15.00
<1


Cobalt
μg/L
0.5
240.00
234.00
<0.5


Copper
μg/L
1
23.00
184.00
3.00


Iron
μg/L
60
<60
142.00
<60


Manganese
μg/L
1
22.00
40.00
18.00


Molybdenum
μg/L
1
20.00
54.00
<1


Nickel
μg/L
1
0.5
2.3
11.00


Selenium
μg/L
1
18.00
19.00
<1


Sodium
μg/L
650
340,000.00
351,000.00
14,400.00


Zinc
μg/L
3
4.00
31.00
7.00









Example 13—Comparison of Continuous Electrosorptive Cavitation with Electrosorption and Cavitation Carried Out in Separate Processes

To demonstrate that the two processes involved in the present apparatus and method cannot be de-coupled with equal results, the electrosorptive cavitation equipment was operated (i) as a continuous electrosorptive cavitation process, and (ii) as 2 different process stages with an intermediate tank after cavitation that was pumped at the same flow rate using a separate pump into the electrosorptive process stage.


An aqueous feed consisting of Botryococcus braunii culture dispersed at 0.5 g/L in basal mineral media with conductivity of 720 μS was subjected to 2 different configurations 1. Electrosorptive cavitation and subjected to low pressure cavitation at 300 psi fed directly into the electrosorptive system at 36 VDC/10 A without additional residence time and 2. The same culture was subjected to low pressure cavitation and the cavitation outlet flow was collected in a reservoir and then pumped into the electrosorption system An aqueous feed consisting of Botryococcus culture dispersed at 0.5 g/L in basal mineral media with conductivity of 720 μS was subjected to cavitation through a nozzle at 300 psi, and directly fed into a column containing steel coil and aluminum rod electrodes at 36 VDC/10 A for electrosorption.


To compare the effect of separate cavitation and electrosorption treatments in sequence, the feed was also run through the cavitation nozzle followed by a resting period of 5 minutes, to allow dissipation of generated cavitation bubbles prior to electrosorption. It was then passed through the electrode column at the same flow rate used in the integrated cavitation-electrosorption experiment, producing similar residence times in each treatment but removing any synergistic effects caused by the presence of active cavitation bubbles during electrosorption with a 5 minute residence time in the process.


Photographs of the samples, micrographs of the biomass, and optical density measurements were taken and are shown in Table 48 below and in FIGS. 56 and 57. A large increase in the amount of lipid layer (foam) is shown for the integrated electrosorptive cavitation process relative to the same sample subjected to separate cavitation followed by a delay prior to electrosorption.


Optical density measurements were reduced by approximately 50% relative to the feed, indicating removal of biomass from the dispersion.









TABLE 47







Optical density measurements









OD 600



(Represents optical density of algae


Culture Sample
to determine biomass agglomeration)











Raw feed 0.5 g/L Botryococcus
0.538


Nozzle only
0.595


Nozzle -> electrode separately
0.314


(Aqueous flow-through)



Nozzle -> electrode separately
8.67


(Foam)



Nozzle + electrode integrated
0.321


(Aqueous flow-through)



Nozzle + electrode integrated
7.68


(Foam)



Nozzle + electrode integrated
0.12


(Aqueous settled)









Example 14—Gold Mining Effluent Water Treatment

Gold mine effluent water with a high turbidity of 5.4 was subjected to electrosorptive cavitation (e-Cav) and the water quality is reported below in Table 48. Assays performed to determine metals content before and after e-Cav show that there is a significant reduction in sodium (reduced by ˜45%). This indicates that the process can be applied for de-salting applications from mine water effluents.









TABLE 48







Analysis of gold mine effluent water subjected to electrosorptive cavitation



















Sample
Sample
Aluminum
Silver
Arsenic
Cobalt
Copper
Iron
Manganese
Molybdenum
Nickel
Selenium
Sodium


Name
Tag
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(mg/L)
(mg/L)






















LDR
LDR
0.04
0.020
0.05
0.01
0.1
0.1
0.01
0.01
0.01
0.15
0.200


Effluent
S1-R-C
0.35
0.209
0.14
0.67
33.1
1.1
0.71
0.05
7.64
0.16
919.00


(PARC-1)














PARC-1
S1-P-
13.1
0.026
0.09

16.8
1.5
0.42
0.03
4.31
0.15
540.00


Processed
600-C













(e-Cav














Process














at 600 psi)









To examine the impact of lower feed turbidity on the same mine affected water, pre-treated samples with reduced turbidity were subjected to the electrosorptive cavitation process. The initial and processed effluent characteristics are shown below in Table 49.


The e-Cav process shows a reduction in metals with a significant reduction in selenium—a hard to remove contaminant.









TABLE 49







Analysis of pretreated mine affected water (from Table 48 feed) with a reduced


turbidity of 2.49.





















Silver
Barium
Boron
Cadmium
Chrome
Cobalt
Copper
Iron
Manganese
Nickel
Lead
Selenium
Zinc










No.
Units





















Tag
C/N
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
























LDR
LDR
0.2
2
40
0.1
1
0.5
1
60
1
1
0.5
1
3


S3-
8780230
4.10
37.00
121.00
<0.1
<1
242.00
390.00
<60
35.00
42.00
<0.5
117.00
33.00


R-C-N
















S3-
8780220
2.10
51.00
556.00
0.20
13.00
235.00
520.00
168.00
48.00
97.00
8.10
21.00
36.00


AX-
















600









Example 15—Benchmarking of Selenium Removal Between Best Available Physico-Chemical Technologies (Ion Exchange and Adsorption)

The selenium extraction results described herein were further benchmarked with a best available technology—ion exchange process.

    • Ion Exchange with a Strong Base Anion Resin (Amberlite IRA 410) was evaluated to benchmark the removal of selenium from metallurgical coal waters (The same metallurgical coal affected water samples above were subjected to ion exchange and adsorption). (See FIG. 58)
    • Much lower selenium at less than 1 ppb was obtained with a fully regenerated bed. However, in less than 9 bed volumes selenium breakout above 60 ppb (beyond all known environmental discharge limits) was achieved and full exhaustion occurred at 15 bed volumes showing that the process was not suitable for large volume mine affected water.


To compare the performance of ion exchange vs. electrosorptive cavitation, the same water sample was run through the e-Cav system and the effluent selenium concentration was determined to be 8 ppb consistently for over 100 bed volumes (equivalent).


Benchmarking of the process with the same metallurgical coal affected mine water was also performed using a speciality oxy-anion adsorbent. (See FIG. 59). ADSORBSIA™ As600 media is an easy to use granular titanium oxide with strong affinity for arsenic (exists as arsenate or arsenite-oxy-anion form of arsenic in water, lead and other heavy metals. This distinct media is designed for non-regenerative applications. The inherently high adsorptive capacity of Dow's titanium oxide based technology has been formulated into a mechanically stable granulation suitable for use in a broad range of potable water applications. Because it is based on titanium, ADSORBSIA As600 does not support bacterial growth as will iron based media. However, as shown in FIG. 59, the selenium concentration of coal affected mine effluent was not significantly affected by the use of As-600 over 15 bed volumes and the effluent concentration was 400 ppb or higher throughout the evaluation. The exhaustive experiment showed a reduction to 175 ppb over 20 bed volumes but was far from the effluent discharge limits at the end of the process cycle.


One or more illustrative embodiments have been described by way of example. It will be understood to persons skilled in the art that a number of variations and modifications can be made without departing from the scope of the invention as defined in the claims.

Claims
  • 1. A phase separation device for performing continuous phase separation of an aqueous mixture, said device comprising: a cavitation assembly for cavitating the aqueous mixture to produce a cavitated aqueous mixture; wherein the cavitation assembly comprises a tangential flow aqueous mixture inlet, a vortex stabilizer in communication with the tangential flow aqueous mixture inlet, and a nozzle in communication with the vortex stabilizer, wherein the nozzle comprises a restriction flow channel and a divergent outlet in communication with the electrosorption assembly; andan electrosorption assembly for receiving the cavitated aqueous mixture and performing an electrosorptive phase separation of the cavitated aqueous mixture; the electrosorption assembly comprising a non-sacrificial cathode and a non-sacrificial anode.
  • 2. The device of claim 1, wherein the nozzle comprises the restriction flow channel having a length L and a diameter d, wherein the restriction flow channel receives the aqueous mixture and directs the aqueous mixture to the divergent outlet; wherein the cavitated aqueous mixture comprises cavitation voids comprising vapour, and wherein the electrosorptive phase separation of the cavitated aqueous mixture includes electrosorptive vapour bubble generation.
  • 3. The device of claim 2, wherein the restriction flow channel has a d/L ratio of 0.005-0.02.
  • 4. The device of claim 2, wherein d is 0.03-0.0625 inches.
  • 5. The device of claim 1, wherein the vortex stabilizer comprises a vortex stabilizer insert comprising a conical portion having a longitudinal slit for allowing vortexing aqueous mixture to exit the vortex stabilizer and enter the nozzle.
  • 6. The device of claim 1, wherein the divergent outlet comprises a conical channel which progressively widens toward the electrosorption assembly with a divergent angle of 30°-45°.
  • 7. The device of claim 1, wherein the divergent outlet has a diameter to length ratio of 0.1875-0.3.
  • 8. The device of claim 1, wherein the cathode comprising a conductive spiral tube with a central core and the anode comprising a conductive member disposed within the central core.
  • 9. The device of claim 8, wherein the cathode comprises a carbon-based conductive material, TiO2, copper, brass, aluminum, or stainless steel.
  • 10. The device of claim 8, wherein the conductive spiral tube of the cathode comprises two or more turns, wherein each turn, in combination with the anode, forms a distinct electrochemical cell.
  • 11. The device of claim 1, wherein the device is configured vertically or on an incline, such that the electrosorption assembly receives the cavitated aqueous mixture at a bottom end thereof and performs electrosorptive phase separation of the cavitated aqueous mixture as the cavitated aqueous mixture moves toward a top end of the electrosorption assembly, the top end being elevated relative to the bottom end.
  • 12. The device of claim 1, wherein the cathode forms an outer shell defining an interior core, and the anode is disposed within the core.
  • 13. The device of claim 12, wherein the cathode comprises a continuous conductive member, or a plurality of individual conductive members longitudinally aligned in series.
  • 14. The device of claim 13, wherein the cathode comprises a tube, spiral, or coil.
PCT Information
Filing Document Filing Date Country Kind
PCT/CA2018/050299 3/13/2018 WO
Publishing Document Publishing Date Country Kind
WO2018/165750 9/20/2018 WO A
US Referenced Citations (18)
Number Name Date Kind
7762715 Gordon et al. Jul 2010 B2
8002992 Foret Aug 2011 B2
8673129 Gordon et al. Mar 2014 B2
8673154 Kniep et al. Mar 2014 B2
8771499 McCutchen et al. Jul 2014 B2
8808529 McAlister Aug 2014 B2
8845906 Henley Sep 2014 B2
8968571 Lee et al. Mar 2015 B2
20070056611 Martin Mar 2007 A1
20100290307 Gordon et al. Nov 2010 A1
20110147231 Gordon Jun 2011 A1
20120129244 Green et al. May 2012 A1
20120205301 McGuire et al. Aug 2012 A1
20130042756 Oda et al. Feb 2013 A1
20140017754 Kale et al. Jan 2014 A1
20140116942 Gordon et al. May 2014 A1
20140251904 Ranade Sep 2014 A1
20140263093 Sprague Sep 2014 A1
Foreign Referenced Citations (16)
Number Date Country
WO2010123903 Oct 2010 CA
2672646 Jan 2005 CN
201467824 May 2010 CN
101948154 Jan 2011 CN
101948154 Jan 2011 CN
101948154 Jan 2011 CN
102442746 May 2012 CN
2756068 Jul 2014 EP
52-132546 Nov 1977 JP
2004-122045 Apr 2004 JP
2004-174325 Jun 2004 JP
2012-523849 Oct 2012 JP
2015-508016 Mar 2015 JP
625741 Sep 1978 SU
2010123903 Oct 2010 WO
2013116357 Aug 2013 WO
Non-Patent Literature Citations (13)
Entry
Supplementary European Search Report dated Dec. 11, 2020 from EP Application No. 18767266.2.
Yujie, Feng, Applications of Electrochemical Technology in Environmental Engineering, Chemical Industry Press, pp. 103-104, May 2002 (Chinese publication).
Yujie, Feng, Applications of Electrochemical Technology in Environmental Engineering, Chemical Industry Press, pp. 103-104, May 2002 (English translation).
Moshen, Cui, High-Pressure Water Jet Technologies, Coal Industry Press, pp. 108-110, Oct. 1993 (Chinese publication).
Moshen, Cui, High-Pressure Water Jet Technologies, Coal Industry Press, pp. 108-110, Oct. 1993 (English translation).
Office Action dated Dec. 3, 2021 from corresponding application EP 18767266.2.
International Search Report and Written Opinion for Application No. PCT/CA2018/050299.
International Preliminary Report on Patentability for Application No. PCT/CA2018/050299 dated Sep. 26, 2019.
Japanese Office Action dated Jan. 5, 2022.
English translation of Japanese Office Action dated Jan. 5, 2022.
Lee, Andrew K.; Lewis, David M.; Ashman, Peter J., Microalgal cell disruption by hydrodynamic cavitation for the production of biofuels, J Appl Phycol 27, 1881-1889 (2015), Dec. 16, 2014.
Yang, Kun-Lin; Ying, Tung-yU; Yiacoumi, Sotira; Tsouris, Costas; Vittoratos, E. Steven, Electrosprption of ions from aqueous solutions by carbon aerogel: an electrical double-layer model, Langmuir 2001, 1961-1969, Feb. 16, 2001.
Office Action dated Jun. 2, 2022 from corresponding Chinese Application No. 201880031308.9.
Related Publications (1)
Number Date Country
20200086237 A1 Mar 2020 US
Provisional Applications (1)
Number Date Country
62471060 Mar 2017 US