Patent Application
20250099917
Addressing the problem of water scarcity, which currently affects every continent and some 3 billion people around the world, is one of the greatest challenges in this century. Desalination, which separates fresh water from saline or contaminated water, proves to be one of the most promising methods to ease the pressure on global water shortages. As a hybrid thermal/membrane desalination technology, membrane distillation (MD) has recently gained much attention given its simple separation mechanism operating at low temperatures and pressure (see, e.g., Gonzalez and Suarez (2017), Renewable and Sustainable Energy Reviews 80, 238-259; Ali et al. (2018), Renewable and Sustainable Energy Reviews 81, 1-21; Desmukk et al. (2018), Energy & Environmental Science 11, 1177-1196).
Desalination by the MD process is based on the use of hydrophobic membranes, which contact the heated saline water (normally at 50 to 80° C.) at the feed side. Driven by the temperature difference across the membrane, water evaporates at the membrane-saline interface, diffuses through the pores of the membrane and condenses on the opposite side (normally at ˜20° C.). Due to its intrinsic water repellency, the hydrophobic membrane prevents the saline water from passing through while allowing for vapor transport. Thus, it separates the volatile (i.e., water) and nonvolatile species (i.e., salts) in the hot saline. As the MD process operates at much lower hydraulic pressures than reverse osmosis, and requires less complicated components than multiple-effect distillation, it is highly advantageous for small-scale, simple process-designed desalination facilities. MD also allows water recovery from highly concentrated brines, where reverse osmosis would require too high pressures to be practicably feasible. This includes applications apart from desalination, such wastewater treatment for water recovery, removal of key contaminants from aqueous solutions, or recovery of value-added products (Hussain et al. (2021), Emergent Materials; https://doi.org/10.1007/s42247-020-00152-8).
Despite intensive efforts, widespread adoption of MD is still hindered by the lack of durable hydrophobic membranes with a high distillation capacity. Theoretically, distillation flux J of a MD membrane can be expressed as a function of the membrane properties and partial vapor pressure difference (Δp) across the membrane:
where ε, τ, and σ are the membrane porosity, tortuosity, and thickness respectively; R is the universal gas constant; T is the mean membrane temperature; m is the molecular diffusion coefficient of vapor in air;
K is the Knudsen diffusion coefficient of vapor inside the membrane. At a fixed temperature, the ratio of the diffusion coefficient (
m/
K) is inversely proportional to the nominal pore diameter (dn) of the membrane. Therefore, a highly porous membrane with large pores will maximize the distillation flux.
Durable MD desalination with a high level of salt rejection necessitates excellent membrane wetting resistance, in order to prevent infiltration of salty water into the membrane. The wetting resistance of a membrane is usually assessed quantitatively by the liquid entry pressure (LEP), which is defined as the minimum required pressure for liquid solution entering the membrane pores. In a simple form based on the Young-Laplace equation, it can be calculated as LEP˜(−4γ cos θY)/dmax, where dmax is the maximum membrane pore diameter, γ is liquid surface tension, and θY is the contact angle of the membrane material. Accordingly, membranes with small pore size, narrow pore size distribution, and low surface energy typically show high LEP and excellent salt rejection.
These requirements of both high distillation flux and LEP for MD membranes pose a critical challenge when designing a membrane—large membrane pore size allows efficient distillation, and yet it inevitably increases the susceptibility to liquid penetration. To ensure stable desalination without risk of wetting, the nominal pore diameters of state-of-the-art membranes are typically less than 0.2 μm, which greatly reduce desalination efficiency. In order to balance the conflicting requirements of distillation flux and LEP, conceptual designs of composite membranes with different pore sizes have been proposed over the past years (Desmukk et al. (2018) ibid; Jiang et al. (2020), ACS Nano 14, 17376-17386), but fabrication of such membranes often involve multiple processing steps and use of fluorination agents, which pose environmental risks.
Therefore, to date, it still remains a challenge to produce a scalable polymeric membrane that can simultaneously enhance distillation flux and wetting resistance via a simple processing method.
In view of this situation, a main objective underlying the present invention is the provision of improved means for membrane distillation which overcome or considerably alleviate the drawbacks of the prior art, in particular in that they achieve a higher distillation flux and thermal efficiency of desalination than materials and methods of the prior art, are eco-friendly and can be a manufactured/used in a relatively simple and cost-efficient manner.
This main objective is achieved according to the present invention by providing the nanofilament coated membranes according to claim 1, the method for preparing the same according to claim 6 and the device according to claim 13. Additional aspects and preferred embodiments of the invention are the subject of further claims.
The present inventors developed a fluorine-free superhydrophobic membrane by coating a thin layer of nanofilament network onto the top of a micro-porous membrane matrix, which combines the advantages of multi-scale porous structures. This hierarchical topography greatly enhances the LEP of the membrane while retaining a high vapor transfer rate. Under standard desalinating conditions, the nanofilament-coated membrane demonstrated 60% higher distillation flux than that of commercial state-of-the-art membranes. Concurrently, the thermal efficiency of desalination was improved from 84% to 93%. Owing to its non-toxic hydrophobic nature and potential for scalable manufacturing, this advanced composite membrane offers an avenue to affordable clean water for the off-grid communities by using low-grade energy.
The nanofilament-coated membrane with hierarchical porous structures according to the present invention comprises
Principally, the microporous polymer support membrane in said nanofilament-coated membrane is not especially limited and may be any microporous polymer membrane known in the art and in particular any microporous polymer membrane which is used for membrane distillation and commercially available. The nominal pore diameter—which may be alternatively denoted as mean pore diameter—may be determined by any method known in the art and, for commercial products, corresponds to the product data provided by the respective supplier. Typically, the pore diameter is determined by means of capillary flow porometry, in particular according to ASTM F316-03 (2019): https://www.astm.org/f0316-03r19.html.
More specifically, the microporous support membrane comprises or consists of a polymer which is selected from the group consisting of polyethersulfone (PES), cellulose acetate (CA), polypropylene (PP), polyamide (nylon), polytetrafluoroethylene (PTFE), polyvinyl difluoride (PVDF) or polyethylene (PE).
The nominal pore diameter—which may be alternatively denoted as mean pore diameter—of the superhydrophobic fluorine-free nanoporous layer may be determined by any suitable method known in the art. The present inventors used scanning electron microscopy (SEM) and SEM image analysis for this purpose (as detailed in Example 2 and
Typically, the nanofilament-coated membrane of the invention has an apparent receding contact angle θrapp for water of more than 150°, measured at room temperature, and a roll off angle of less than 10° even after being immersed in hot water of 80° C. for 48 h. The nanofilament-coated membrane of the invention advantageously exhibits a liquid entry pressure LEP of at least 2 bar, preferably at least 5 bar.
A second aspect of the invention relates to methods for preparing the nanofilament-coated porous membrane as described above.
Such a method generally comprises at least the following steps:
In a specific embodiment thereof, the solvent is a mixture of n-heptane and toluene (for example a 1:1 mixture) and the formation and self-assembling of nanofilaments is completed within a predetermined period of time, typically in the range from 10 min to 12 h.
As already indicated above, the nanofilament-coated membrane according to the present invention is especially suited for uses/applications in the field of membrane distillation.
More specifically, the nanofilament-coated membrane may be used in, e.g., a process of desalination of saline or distillation of contaminated water or extraction of water from waste water or extraction of other volatile components from a feed solution.
A closely related aspect of the present invention, therefore, relates to a method for extraction of a volatile component from a feed solution, e.g. a saline aqueous medium, which comprises at least the following steps:
A still further aspect of the present invention relates to a device, in particular a membrane distillation device, comprising the nanofilament-coated porous membrane as defined above.
Principally, the construction of the membrane distillation device according to the present invention is not especially limited and may be based on any membrane distillation system known in the art.
In a specific embodiment, said membrane-distillation device of the invention further comprises at least the following components:
The following Examples are provided to illustrate the present invention in more detail, however, without limiting the same to the specific conditions and parameters thereof.
To fabricate a high-performance composite membrane as required by MD desalination, a superhydrophobic nano-porous layer was coated onto a micro-porous polyethersulfone (PES) membrane as follows. This commercially available 130-μm-thick PES membrane with nominal pore diameter (dn) of 8 μm (in the following denoted as PES-8) (
PES membranes, purchased from Sterlitech Corp., USA, were activated by using O2 plasma (2 min, 90 W, Diener Electronic Femto). The O2 flow rate was set to 7 ml/min. 0.6 ml of trichloromethylsilane (TCMS) was added to 300 ml of a 1:1 (volumetric) mixture of n-heptane and toluene. Before mixing, a trace amount of water was added to n-heptane and toluene. The water concentration of n-heptane and toluene was measured as 90 ppm and 275 ppm, respectively. Then the plasma activated PES membranes were immersed in the reaction solution for nanofilament growth. After 6 hours, the nanofilament-coated PES membranes were rinsed with n-hexane and dried using N2 flow.
For comparison, the originally hydrophilic PES membranes were hydrophobized by using surface fluorination with 1H,1H,2H,2H-perfluorodecyltrichlorosilane (PFDTS). PES membranes were activated using O2 plasma treatment (2 min, 90 W, Diener Electronic Femto) at a O2 flow rate of 7 ml/min. Subsequently, PFDTS (180 μL, Alfa Aesar) was mixed with n-hexane (350 mL) and the activated membranes were immersed in the solution for 60 min, rinsed with n-hexane and dried under a nitrogen gas flow.
In order to characterize and evaluate the physical properties of the nanofilament-coated porous membranes prepared according to Example 1 above, the commercial polyethylene and polytetrafluroethylene micro-porous membranes were selected as the benchmarks for comparison and the properties of all tested membranes are indicated in Table 1 below.
SEM images of cross-sections (
To analyze the effective pore diameter of a nanofilament network coating, the geometrical features were evaluated based on the computer analysis of SEM images (
The pore size and pore size distribution of the nanofilament coating was measured by analyzing the top-down SEM images. The adopted methodology consists of the following steps. First, the image was denoised by means of a standard median filter to remove the salt-pepper image noise. Then, the intensity histogram was calculated and the threshold for identifying the membrane pores shape defined. Next, the image was converted to binary. From these binary images, membrane pores were identified. Finally, the sizes of detected membrane pores was calculated to obtain pore size histograms.
The computer image analysis revealed that the pore diameter of the nanofilament coating was distributed between 10 to 100 nm. Compared to the commercial membranes, the nano-porous coating resulted in not only a smaller pore size, but much higher surface roughness, which would substantially increase the liquid entry pressure of coated membranes according to the Young-Laplace equation.
Desalination performance using the MD process relies essentially on the liquid entry pressure (LEP) and gas permeability of the adopted membrane: the LEP highly affects the salt rejection and the gas permeability dominates the distillate flux. To quantitatively explore the improved wetting resistance of hierarchical membranes, the LEP of nanofilament-coated PES membranes was benchmarked against commercial PE membranes and PES membranes that were rendered hydrophobic by fluorination. The testing was carried out in a custom-made setup which can ramp up the transmembrane pressure difference (see Detailed Experimental Procedures). For convenient description here, NF-PES-0.1 to NF-PES-8 are used to denote the nanofilament-coated PES membranes with nominal pore diameter (dn) ranging from 0.1 to 8 μm; PE-0.2 to PE-2.5 are used to denote the PE membranes with dn ranging from 0.2 to 2.5 μm. For the membranes with single-scale porous geometry (e.g., PE and fluorinated PES membranes), the LEP values drop sharply with the increasing dn (
The present inventors found a greatly enhanced LEP for nanofilament-coated PES membranes, as the nano-porous outer layer withstands high capillary pressure. Note that NF-PES-0.1 and NF-PES-1.2 membranes exhibited an extremely high LEP, which even exceeded the limit of our testing setup (11.5 bar). However, when the pore size of PES membranes is considerably larger (e.g., dn>3 μm), the LEP of the nanofilament-coated membrane gradually goes down with the increasing pore size of matrix. The decline of LEP can be related to the imperfect growth of the nano-porous layer on the big membrane pores as it is difficult to cover the large openings completely with a nanofilament network. Nevertheless, the effective pore diameter of the NF-PES-3, NF-PES-5, and NF-PES-8 membranes were still smaller than 0.3 μm according to the theoretical estimation of the Young-Laplace model. Compared to the fluorinated PES membrane with the same pore diameter, the nanofilament-coated PES membranes raised the LEP by at least 16 fold. Considering the hydraulic pressure in the MD system, LEPs of all the nanofilament-coated PES membranes were well above the safety threshold, indicating their applicability in desalination.
To evaluate the membrane resistance to vapor transport, the gas permeation of PE and nanofilament-coated PES membranes under differing transmembrane pressures (see Detailed Experimental Procedures) was characterized. A linear dependence between gas permeation flux and transmembrane pressure difference was observed for both single-scale and multi-scale porous membranes (
During MD desalination, membranes need to remain in contact with hot water for hours or days, which may affect the physical properties and surface chemistry of membranes. To evaluate possible degradation of membrane surfaces, polytetrafluroethylene (PTFE), PE, and nanofilament-coated PES-8 membranes were immersed in Milli-Q water at 80° C. for 3 to 48 hours. After drying the tested membranes under a nitrogen stream, the apparent receding contact angle θrapp for water and the contact angle hysteresis θCAH on the membrane surfaces was measured.
During the 48-hour hot immersion test, NF-PES-8 membranes maintained their super liquid-repellency with θrapp always greater than 155° and θCAH less than 5° (
In contrast to this, the commercial PE and PTFE membranes showed significant deterioration.
After being immersed in hot water for 48 hours, θrapp on PE membranes declined from ˜88° to ˜24° and θCAH increased from ˜35° to ˜75°, indicating a considerable loss of liquid-repellency. Snapshots in
The hydrophobicity also decreased on commercial PTFE membranes, although PTFE is widely considered to be thermally stable. After 48 hours in hot water, the θrapp on PTFE membranes decreased from ˜130° to ˜96° and θCAH increased from ˜20° to ˜50°. Based on previous studies of PTFE polymer and membranes, the inventors assume that the increasing hydrophilicity is mainly caused by the change in surface structure and polymer crystallinity at elevated temperatures. In the case of commercial micro-porous membranes, the loss of surface hydrophobicity could cause a gradual infiltration of liquid into membranes, decrease desalination efficiency, and even cause the contamination of the distillate by salty water.
The durable superhydrophobicity of NF-PES-8 membranes indicates that by reducing the wetted area on membrane surface, there is an effective way of retarding the polymer degradation in hot water. The fraction of the water contact area on membrane surfaces was estimated by analyzing the receding contact angle. For superhydrophobic arrays of cylindrical micropillars, the apparent receding contact angle (θrapp) can be expressed as
Here, Or is the receding contact angle for water on a flat surface of same material as the membrane. The wetted area φls=Als/Aw is the ratio of the projected area of liquid-solid interface (Als) to the projected area of total wetted region (Aw), and φlv=1−φls is the fractional area of the liquid-vapor interface.
By adopting such a specific geometrical model in eq. 2, the wetted area fractions on PE and NF-PES-8 membranes were estimated to be ˜20% and ˜1.5% at the beginning of the immersion test, respectively. An alternative approach is to take the liquid droplet in global thermodynamic equilibrium and apply the Cassie-Baxter equation to estimate the φls surface fraction. In this case, the predicted φls for PE and NF-PES-8 membranes were ˜40% and ˜5%, respectively. Whatever the approach, the extremely small wetted area φls on the NF-PES-8 membrane explains its long-term thermal stability in the immersion test, and also implies a higher energy efficiency in desalination. This is because the heat loss by conduction between hot water and the membrane is greatly suppressed. The nanofilament coating also increases the water evaporation area on the membrane surface due to its superhydrophobic nature. As indicated by the measured θrapp, the NF-PES-8 membrane kept a large fractional area of liquid-vapor interface with φlv=˜98.5% throughout the 48-hour test. In the theoretical analysis of the MD process, φlv is practically equivalent to the surface porosity (c). Therefore, from Eq. 1 distillation flux (J) is proportional to fly, i.e., the large φlv on the NF-PES-8 membrane would lead to a substantial increase in the overall production of distillate.
The water contact angles of all membranes (PE, PTFE, fluorinated PES and nanofilament-coated PES membranes) were measured to characterize the surface wettability. Contact angle and contact angle hysteresis of a water droplet were measured using a DataPhysics OCA35 goniometer. During the measurement, a 5 μl droplet was deposited on the membrane surface, and afterward 20 μl of water was added to and then removed from the droplet. The measurement was consecutively repeated three times at the same position, and at three different positions per substrate. The error of the advancing and receding contact angle measurements was estimated to be ±2°.
To measure the LEP of membranes, a custom-designed apparatus was built. The tested membrane was mounted inside a filter holder, which connected to a syringe pump. By slowly pumping the salty water into the filter holder (0.1 mL/min), the hydrostatic pressure applied on the tested membrane gradually increased. The hydrostatic pressure was monitored using a pressure sensor (IPSLU-M12, RS-Pro) and the data was recorded using a data acquisition system (PCI 6251, National Instruments). Once the applied pressure exceeded the capillary pressure of the membrane pores, liquid penetrated the membranes, leading to a pressure drop. The obtained peak value of the pressure measurement gives the LEP of the tested membrane.
A gas permeability test was employed to analyze the mass transfer resistance of different membranes. The permeation flux of nitrogen through the dry membranes was measured under transmembrane pressures ranging from 10 to 1000 mbar. With increasing transmembrane pressure, the gas flow rate was obtained by using flow sensors with respective ranges (SMC Corp., PFMV5 series). The effective area of the tested membrane was 63 mm2.
In order to demonstrate the technical potential of exemplary nanofilament-coated PES membranes in water desalination, membrane distillation (MD) experiments were conducted in a custom-made air gap membrane distillation (AGMD) system (
In said AGMD experiments, nanofilament-coated PES membranes were tested at different feed water temperatures for more than 12 hours. The conductivity of saline feed water σf in the MD test was stabilized at the level of seawater (σf=54.0±0.5 mS/cm at 25° C.). The weight and conductivity of distillate σd were continuously monitored over time to characterize the distillation flux and salt rejection rate in the desalination process. For comparison, MD performance of commercial PE and PTFE membranes were assessed as being the benchmark. In the following, PTFE-0.1 and PTFE-0.2 are used to denote the PTFE membranes with nominal pore diameters of dn=0.1 and 0.2 μm, respectively.
The inventors calculated the salt rejection rate by (1−σd/σf)×100%, where σd and σf are conductivity of distillate and feed, respectively. As commonly used membranes in the MD process, hydrophobic PE membranes were able to remove more than 99.9% salt from the feed saline (i.e., σd<54 μS/cm) when the pore diameter was below dn≤0.9 μm. With the increasing pore size, the distillation flux of PE membranes goes up from 11.5 to 17.7 Lm−2h−1, but the water conductivity rises from 1.1 to 2721 μS/cm. The decline in salt rejection performance arises from the low wetting resistance of PE membranes with a large pore size (e.g., dn>1.5 μm). Considering the standard conductivity of distilled water (0.5-3 μS/cm), only a PE-0.2 membrane conforms to the acceptance criteria.
In contrast, for all the nanofilament-coated PES membranes, the conductivity of purified water remained low and independent on the membrane pore size and distillation flux. Although the surface tension of feed water decreased from 0.07 N/m at 50° C. to 0.063 N/m at 80° C., the superhydrophobic nanofilament coating maintained high wetting resistance. Even when the nominal membrane pore size was 8 μm, the NF-PES-8 membrane performed with an excellent salt rejection (>99.995%). Given the substantial increase of distillation flux (18.2 Lm−2h−1), the hierarchical PES membrane shows significant advantages over the commercial membranes.
To further test the performance of various membranes under different environments, their desalination ability was determined at differing feed water temperatures.
Maximizing the thermal efficiency of water production is imperative to the future development and industrialization of MD in terms of the water-energy nexus. The thermal efficiency (n) of tested MD membranes, defined as the ratio of heat utilized for distillation to the total heat consumption at the feed side, is determined by=qd/qf. Here, qf is the total heat transfer rate through the membrane and qd is vaporization heat transfer rate associated with the distillation flux.
As shown in
In order to clearly summarize membrane performance for water desalination, a selection of different membranes was rated in a diagram with the two most important parameters: LEP and distillation flux (
Membrane distillation tests were performed using a custom-made AGMD setup, which consisted of AGMD module, feed water and coolant circulating loops, digital balance, conductivity meter, and data acquisition system. The tested membrane was mounted in the AGMD module, between a feed flow channel and a condensing surface. A support mesh (˜0.5 mm thick) was used to hold the membrane in a planar shape and reduce the membrane deformation due to the pressure difference between feed flow and air gap. An acrylic spacer was used in the MD module to create the required air gap. The total air gap width between membrane and condensing surface was ˜4.5 mm. Feed saline water was heated to the desired temperature and pumped to the AGMD module using a magnetic coupling water pump. The condensing surface temperature was controlled by the coolant flow loop using a refrigerated water bath circulator. When distilled water slid off the condensing surface by gravity, it was collected in a glass flask. A digital balance (SPX 2202, Ohaus) continuously recorded the weight of collected distilled water for determining the distillation flux of tested membranes. The conductivities of feed and distilled water were measured by the conductivity meter for calculating the salt rejection during membrane distillation. Four Pt100 temperature probes (PM-1/10-1/8-6-0-P-3, Omega) were adopted to measure the liquid temperature at inlet and outlet of feed flow channel and coolant flow channel, respectively. Two flow meters (FT110, Gems) and two pressure transducers (IPSLU-M12, RS-Pro) were installed in the pipelines to continuously monitor the flow rate and pressure in the feed and coolant loops. All the sensors in the AGMD testing setup were electrically connected to a data acquisition system, which consisted of two National Instruments (NI) analog input modules (PCI 6251 and NI-9216). The measured data during MD experiments were transferred to the computer, which could be monitored in real-time and stored using a self-written LabView code.
| Number | Date | Country | Kind |
|---|---|---|---|
| 22150541.5 | Jan 2022 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2023/050117 | 1/4/2023 | WO |
