| JP2004227902 | FUEL CELL EQUIPPED APPARATUS |
| JP2002216822 | FUEL CELL SYSTEM |
| JP2005255505 | HYDROGEN SUPPLY METHOD |
HONG, Liang (3 Research Link, Singapore 2, 11760, SG)
LIU, Zhaolin (3 Research Link, Singapore 2, 11760, SG)
TAY, Siok Wei (3 Research Link, Singapore 2, 11760, SG)
ZHANG, Xinhui (3 Research Link, Singapore 2, 11760, SG)
HONG, Liang (3 Research Link, Singapore 2, 11760, SG)
LIU, Zhaolin (3 Research Link, Singapore 2, 11760, SG)
TAY, Siok Wei (3 Research Link, Singapore 2, 11760, SG)
| Claims A method of forming a polymer electrolyte membrane comprising the steps of: providing a porous polymer membrane matrix in solvated state, the porous polymer membrane matrix having opposite-facing surfaces and, between said surfaces, having an internal structure defining a three-dimensional network of interconnected passage and pathways; and polymerizing ionic monomers disposed in the three-dimensional network of interconnected passage and pathways while said matrix is in said solvated state to thereby form an ion-conductive polymer disposed substantially throughout the bulk of the matrix. A method as claimed in claim 1 , wherein said ionic monomers are anionic .monomers that form a proton- conductive phase■ throughout the bulk of the matrix. A method as claimed in claim 1 or claim 2, wherein during said polymerizing step, the matrix is expose to a solvent and anti-solvent mixture. A method as claimed in claim 3, wherein there is a volume excess of anti-solvent to solvent in said mixture . A method as claimed in claim 4., wherein the volume ratio of solvent to anti-solvent is in the range of 1: 10 to 3: 10. 6. A method as claimed in any one of the preceding claims, wherein during the polymerization step, the matrix is also exposed to a solution containing a cross-linking agent . 7.. A method as claimed in any one of the preceding claims, wherein during the polymerization step, the matrix is also exposed to a solution containing a ultraviolet initiator. 8. A method as claimed in any one of the preceding claims, wherein said polymer membrane matrix is a thermoplastic polymer. 9. A method as claimed in any one of the preceding claims, wherein said thermoplastic polymer is selected from the group consisting of polysulfones, polyamides, polyphenyleneoxides and polybenzimidazole. 10. A method as claimed in any one of claims 2 to 9, wherein said anionic monomers are selected from at least one of the group comprising carboxylic acids (salt), carboxylates , sulfonic acid (salt), sulfonates, acrylates, and acrylonitriles . 11. A method as claimed in claim 10, wherein the anionic monomers may be selected from the group comprising polyacrylonitrile , hydroxyethyl methacrylate, sodium 4-styrenesulfonate, 3-sulfopropyl acrylate potassium salt, and 3-sulfopropyl methacrylate potassium salt. A method as claimed in any one of the preceding claims wherein said polymerization ' step is undertaken for at. least 1 hour. A method as claimed in any one of the preceding claims, wherein said providing step comprises the step of forming the polymer membrane matrix by phase inversion. A method as claimed in claim 13, wherein said phase inversion is undertaken in a coagulant bath containing a mixture of anti-solvent and solvent. A method as claimed in claim 14, wherein the volume ratio of anti-solvent to solvent is 0.5:1 to 1:1.5. A method as claimed in claims 13-15, wherein said phase inversion is undertaken for at least 12. hours. A method as claimed 'in claims 13-16, wherein said phase inversion is. undertaken for 12 hours to 24 hours . 18. A method as claimed in any one of the preceding claims, comprising the step of exposing said polymer membrane matrix to the ionic monomers before said polymerizing step. A method as claimed in claim 18,. wherein said exposing of said polymer membrane matrix to the ionic monomers before said polymerizing step is undertaken for at least 1 hour. A method, as claimed in claim 19, wherein said exposing of said polymer membrane matrix to the ionic monomers, is undertaken for a period of time to form a hydrogel between the anionic monomers and a solvent to the polymer membrane matrix. A method as claimed in any one of claims 3 to 20, wherein the solvent is an organic ~ phase solvent. A method as claimed in any one of claims 3 to 21, wherein the anti-solvent is an aqueous phase solvent. |
Technical Field
The present invention generally relates to a method for forming a polymer electrolyte membrane.
Background
Polymer electrolyte membrane (PEM) fuel cells, also known as proton exchange membrane fuel cells, deliver high- power density and offer the advantages of low weight and volume compared with other fuel cells. PEM fuel cells are thus an attractive option for use in various applications including automobiles and stationary power generation. However, sulfonated perlfluoropolymer polymer electrolyte membranes (PEMs) such as Nafion® Aciplex®, Flemion® and Dow membranes, which are commonly used in such fuel cells, are generally costly. Thus, the commercialization of. PEM fuel cells has proven to be challenging, due to the high cost of the PEMs, as well as the added costs associated with using platinum (Pt) as electrode catalysts.
One possible way to reduce the cost of PEM fuel cells is to develop non-fluorinated PEMs with cost-effective polymer backbones to which sulfonated side chains, are grafted. Currently, aromatic polymers such as poly(ether ketone) , polyimide, polybenzimidazole, polysulfone, and polyphenyleneoxide, are of particular interest for use as polymer backbones because of their excellent mechanical strength. However, there are a number of major hurdles impeding the development of PEMs based on aromatic polymer backbones. For example, random sulfonation on the backbone of a commercially available aromatic polymer causes either low proton conductivity or an excessively high water uptake and poor mechanical properties. Although well-defined sulfonated polymer chain structures can be synthesized in order to avoid the aforementioned problems, the synthesis of a well-defined chain structure requires the use of special monomers and is very expensive.
One of the current methods for manufacturing PEMs, which can overcome such problems, is to fill a microporous polymer membrane matrix (grid) with a proton conducting phase (plug) . Commercial microporous polymer membranes are made from different thermoplastics, such as polyethylene, poly (propylene ) , poly ( tetrafluroroethylene) , polyimide, polyacrylonitrile and poly (ethyleneterephthalate ) , which have a fiber-piling matrix or foam-like matrix. To enhance the filling effect of the proton conducting phase, surface modification of the porous membranes may be carried out, with subsequent crosslinking of the proton conducting phase in the pores in order to secure the proton conducting phase to the membranes. However, a major problem with manufacturing a 'plug-grid' membrane structure by using the method described above is the incomplete filling of the proton conducting phase within the porous membrane. This is because the polymer electrolytes typically used have a very slow mass diffusion rate due to their large molecular weight. In addition, the porous membranes formed using the above method also tend to have small and tortuous pores. Thus, a slow vaporization of solvent is required in order to avoid precipitation of the proton conducting phase. ' However, regardless of how slowly the evaporation is conducted, the polymer can only partially fill the pores.
This results in a double-layer structure, comprising the base material layer (porous polymer membrane matrix) with a thin electrolyte layer on top of it. This double- layer structure, with incompletely filled pores, is highly undesirable for. fuel cell applications because the added thickness of the membrane increases electrical resistance and causes deformation of the membrane. Furthermore, the incompletely filled pores require that the membranes be kept in an environment with a high degree of moisture throughout an entire fuel cell operation process. Otherwise, the unfilled pore channels would collapse. Such structural features are unsuitable for a fuel cell operated at elevated temperatures.
There is therefore a need for a method for producing a polymer electrolyte membrane that can overcome, or at least ameliorate one or more of the disadvantages described above.
Summary
According to a first aspect, there is provided a method of forming a polymer electrolyte membrane comprising the steps of :
providing a porous polymer membrane matrix in a solvated state, the porous polymer membrane matrix having opposite-facing surfaces and, between said surfaces, having an internal structure defining a three-dimensional network of interconnected passages and pathways; and
polymerizing ionic monomers disposed in the three- dimensional network of interconnected passages and pathways while said matrix is in said solvated state to thereby form an ion-conductive polymer disposed substantially throughout the bulk of the matrix.
Advantageously, the use of a porous polymer membrane matrix in a solvated state allows the passages and pathways of the porous polymer membrane to be accessible for. the monomers through diffusion. Due to the small molecular sizes of the ionic monomers, the ionic monomers can fill the pores of the porous polymer membrane matrix with higher diffusion kinetics. Hence, the electrolyte polymer formed may be physically anchored to the porous polymer membrane matrix in-situ.
During said polymerization step, the matrix may be exposed to an anti-solvent and solvent solution. The use of a higher volume of solvent relative to the volume of anti-solvent during the polymerization step may advantageously lead to a reduction in the surface pore apertures due to excessive shrinking of the polymer chains at the interfacial regions. Advantageously, this may aid in the polymerization of the ionic monomers.
Prior to said polymerization step, the matrix may be exposed to ionic monomers. The polymer membrane matrix may be exposed to the ionic monomers for a period of time to form a hydrogel between the ionic monomers and a solvent to the polymer membrane matrix. Advantageously, the ionic monomers may combine with a solvent to the polymer membrane matrix to form a hydrogel, which leads to polymerisation of the ionic monomers.. More advantageously, this may result in a higher flux of the ionic monomers into the passages and pathways of the matrix.
Definitions
The term "monomer" refers to a molecule that, can undergo polymerization, thereby contributing constitutional, units, (i.e., an atom, a group of atoms, and/or groups of. atoms), to the essential structure .of a polymer.
The term "ionic monomer" is used to refer collectively to monomers having the ionic functional groups in. the molecular structure. The ionic functional group is a functional group capable of ionizing mainly in an aqueous solution to form a cation or an anion and includes, particularly, carboxyl group, sulfonyl group, sulfonate group, acrylate group, acrylonitrile group, phosphonic group,, quaternary ammonium group and quaternary phosphonium group, as well as those groups described from such functional groups such as alkali metal salts and halogen salts thereof, with no particular restriction only thereto.
The term "proton conductive phase" in the context of this specification refers to a polymer phase within which a proton can be transported. For example, in some embodiments, a proton is transported from an anode through a proton conductive polymer to a cathode.
The term "polymerizing" means the condensation' of monomers to form a molecule of higher molecular weight than the monomers and also includes curing/hardening of a polymerizable material, such as due to cross-linking.
The term "solvated state" is to be interpreted broadly to refer to the interaction of a solute with a solvent, and whereby at least one ion in a solution is complexed by solvent molecules.
The term "anti-solvent" refers, to a fluid which promotes precipitation from the solvent of the product (or of a precursor for the product).. The anti-solvent may comprise a fluid which promotes the precipitation via a chemical reaction, .or which decreases the solubility of the product in the solvent; it may be the same liquid .as the solvent but at a different temperature, or it may be a different liquid from the solvent. .
The term "hydrogel" as used herein refers to a polymeric material which exhibits the ability to swell in an aqueous phase and to retain a significant portion of water within its structure without dissolution.
The word "substantially" does not exclude "completely" e.g. a composition which is "substantially free" from, Y may be completely free from Y. Where necessary, the word "sμbs antially" may be omitted from the definition of the invention.
Unless specified otherwise, the terms "comprising" and "comprise", and grammatical variants thereof, are intended to represent "open" or "inclusive" language such that they include recited elements but also permit inclusion of additional, unrecited elements.
As used herein, the term "about", in the context of concentrations of components of the formulations, typically means +/- 5% of the stated value, more typically +/- 4% of. the stated value, more typically +/- 3% of the stated value, more typically, +/- 2% of the stated . value, even more typically +/- 1% of the stated value, and even more, typically +/- 0.5% of the stated value.
Throughout this disclosure, certain embodiments may be disclosed in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the disclosed ranges. Accordingly, the description of a range should be considered to . have . specifically disclosed all the possible sub-ranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed sub-ranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.
Certain embodiments may also be described broadly and generically herein. Each of the narrower . species and subgeneric groupings falling within the generic disclosure also form part of the disclosure. This includes the generic description of the embodiments with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein.
Detailed Description of Embodiments
Exemplary, non-limiting embodiments of a method of forming a polymer electrolyte membrane will now be disclosed.
The ionic monomers may be disposed in the passages and pathways of the porous polymer membrane matrix by contacting the matrix in a solution comprising an ionic monomer. In one embodiment, the ionic monomers may be anionic monomers that form a proton-conductive phase throughout the bulk of the matrix. Exemplary examples of monomers include carboxylic acids or salts, carboxylates, sulfonic acid or salts, sulfonates, acrylates, and acrylonitriles , (meth) acrylic acid or salts thereof; maleic acid or salts thereof, fumaric acid or salts thereof, itaconic acid or salts thereof; crotonic acid or salts thereof; vinyl sulfonic acid or salts thereof; vinyl benzene sulfonic acid or salts thereof; acrylamine alkyl sulfonic acids such as 2-acrylamide-2-methylpropane sulfonic acid or salts thereof; and (meth) acryloylalkylsulfonic acid such as 2-acryloylethane sulfonic acid, 2-acryloylpropane sulfonic acid, 2- inethachloylethane sulfonic acid or salts thereof..
The ionic monomers may be anionic monomers that form proton-conducting polymers upon polymerization. Exemplary proton-conducting polymers that may be used as a material for an electrolyte for a fuel cell include, perfluoro-based polymers, benzimidazole-based polymers, polyimide-based polymers, polyetherimide-based polymers, polyphenylene sulfide-based polymers, polysulfone-based polymers, polyethersulfone-based polymers, polyetherketone-based polymers, polyether-etherketone-based polymers, polyphenylquinoxaline-based polymers, and combinations thereof. Preferably, the proton-conducting polymers may include but are not limited to poly (perfluorosulfonic acid), poly (perfluorocarboxylic acid), co-polymers of tetrafluoroethylene and fluorovinylether containing sulfonic acid groups, defluorinated polyetherketone sulfides, aryl ketones, poly ( 2 , 2 ' - (m-phenylene ) -5, 5 ' - bibenzimidazole) , poly (2, 5-benzimidazble) , and combinations thereof. Accordingly, the anionic monomers utilized during the polymerization step may be derived from anionic monomeric groups of the above-mentioned polymers. In one embodiment, the anionic monomers may be selected from the group. comprising polyacrylonitrile, hydroxyethyl methacrylate, sodium 4-styrenesulfonate, 3-sulfopropyl acrylate potassium salt and 3-sulfopropyl methacrylate potassium salt. Copolymers of the monomers described above may be used.
The method may comprise the ' step . of exposing said polymer membrane matrix to the ionic monomers before said polymerizing step. The exposing of said polymer membrane matrix to the ionic monomers before said polymerizing step; may be undertaken for at least 1 hour. The step of exposing said polymer membrane matrix to the ionic monomers before said polymerizing step may be undertaken in the time range . of 1 to 6 hours,. 2 to 6 hours, 3 to 6 hours, 4 to 6 hours, 5 to 6 hours, 1 to 5 hours, 1 to 4 hours, 1 to 3 hours, or 1 to 2 hours. In one preferred embodiment, said exposing step may be undertaken in the range of 2 to 4 hours. The exposing of said polymer membrane matrix to the ionic monomers may be undertaken for a period of time to form a hydrogel between the anionic monomers and a solvent to the polymer membrane matrix. The formation of a hydrogel may enhance polymerisation of the ionic monomers. The formation of said hydrogel may occur within ' the passages and pathways of the polymer membrane matrix. Hence, formation of said hydrogel may allow the electrolyte polymer formed to be physically anchored to the porous polymer membrane matrix in-situ. Advantageously, the presence of a hydrogel may result ' in a higher flux of the ionic monomers into the passages ' . and pathways of the matrix .
During the exposing step, the ionic monomers in solution and may be at a concentration of 1.5 mmol/ml to 4.5 mmol/ml. The solution used during said exposing step may be a mixture of solvent, such as N-methylpyrrolidone (NMP) and anti-solvent such as water to the polymer membrane matrix, preferably with an excess of anti-solvent to solvent. With an increase in the ionic monomer concentration, this may result in a higher flux of the ionic monomers into the passages and pathways of the matrix. More preferably, the ionic monomer concentration during said exposing step is 4.0 mmol/ml. In one embodiment, the matrix may be exposed to a cross-linking agent during said polymerization step. In another embodiment, the matrix may be exposed to an ultraviolet (UV) initiator. The porous polymer membrane may be contacted with . the ionic monomer (s), cross-linking agent ( s) and an UV initiator as a mixture in a solvent which is compatible with these reactants and the porous polymer membrane so that the desired polymerization and cross-linking may be achieved.
Suitable cross-linking agents and UV initiators for the monomers set forth above are well known in the art. Cross-linking agents having difunctionality or higher functionality may be utilized. Exemplary examples of cross-linking agents include ethylene glycol, dimethacrylate and divinylbenzene . Exemplary examples of UV initiators include ammonium persulfate, potassium persulfate and sodium persulfate.
The solution may further comprise a polymerization initiator. Suitable polymerization initiators for the monomers set forth above are well known in the art. -; When utilizing vinyl compounds as the monomer, suitable polymerization initiators include ammonium persulfate, potassium persulfate, 4 , 4 ' -azobis ( -cyanovaleric acid) 2,2' -azobis (2-amidinopropane) hydrochloride, potassium hydrogen persulfate (Oxone) or the like.
During the polymerization step, the matrix may be exposed to an anti-solvent, such as an organic solvent like NMP and solvent solution such as an aqueous solvent, for example water. In one embodiment, there may be a volume excess of the anti-solvent to the solvent. Typically, the volume ratio of solvent to anti-solvent may be selected from the group consisting of 1:10 to 5:10, 2:10 to 5:10, 3:10. to 5:10, 4:10 to 5:10, 1:10 to 4:10, 1:10 to 3:10, or 1:10 to 2:10. Preferably, the volume ratio of the solvent to anti-solvent may be in the range of 1:10 to 3:10. More preferably, the volume ratio of the solvent to . anti-solvent is 2 : 8. The use of a higher volume of solvent relative to the volume of anti-solvent during the polymerization step may aid in the polymerization of the ionic monomer.
The time period for the polymerization step may be determined by the formation of said hydrogel (gellation) . During the formation of said hydrogel, the viscosity of the monomer solution may increase. Hence, the time period for said polymerization step may be determined by an increase in the viscosity of the solution containing the monomer. In one embodiment, the polymerization step may be undertaken for at least 0.5 hours. The polymerization step may be undertaken in the time range of 0.5 to 3 hours, 0.5 to 2.5 hours, 0.5 to 2 hours, 0.5 to 1.5 hours, 0.5 to 1 hour, 1 to 3 hours, 1.5 hours to 3 hours, 2 hours to 3 hours, or 2.5 hours to 3 hours. Preferably, the polymerization step may be undertaken for at least 1 hour.;
The pore density of the porous polymer membrane matrix may be selected from the group consisting of about 10 8 to about 10 12 pores cm "2 , about 10 9 to about 10 12 pores cm "2 , about 10 10 to about 10 12 pores cm "2 , about 10 11 to about 10 12 pores cm "2 , about 10 8 to about 10 9 pores cm "2 , about. 1Ό 8 to about 10 10 pores cm "2 and about 10 8 to about 10 11 pores cm "2 .
The porosity of the porous polymer membrane matrix may be in the range of about 30% to about 70%, about 40% to about. 70%, about 50% to about 70%, about 60% to about 70%, about 30% to60%, about 30% to 50%, or about 30% to 40%. Preferably, the porosity of the porous polymer membrane matrix may be about 50%. The polymer membrane matrix of the present invention may comprise of a thermoplastic polymer. Thermoplastic polymers employed may be characterized by the specific mechanical and chemical requirements arising in a hydrogen/oxygen fuel cell. The thermoplastic, polymers provided may be suitable as a constructional material for supporting parts of the fuel cell structure, i.e. possess dimensional stability under pressure and at increased temperatures. In addition, the thermoplastic polymers must not be attacked chemically by dry and moist hydrogen or oxygen, and in addition must be hydrolytically stable. Further, the thermoplastic . material must retain all the above-mentioned properties stably up to a minimum temperature of 80 °C.
Thermoplastic polymers employed may be characterised by high tensile strength, impact resistance, resistance to heat deformation and to chemicals, flame resistant and self-extinguishing. The thermoplastics polymers may have an "aromatic spine" and may be soluble in suitable solvents. Examples of such polymers are:
Polysulphones (I)
[-S0 2 -R 1 -0-R2-C(CH 3 )2- 3-0-R 4 -]x (I) with
Ri, R2, R3, R =-C6H ~, -ΟιοΗβ-;
Polyether sulphones (II)
[ [-S0 2 -Ri-] n [-0-R 2 -] m ] x (II)
with
Rif R2 =_ C6H4~ , -ΟιοΗβ-;
n,m=l or 2 Polyether ketones (III)
[ [-CO-R 1 -] n [-0-R 2 -] m ] x (III)
with
Ri, R2 =- C 6 H 4 -, -ΟιοΗβ- n, m=l or 2
Polyphenylene sulphides (IV) .
with - ' ■
Ri =- C 6 H 4 -,
x being capable of varying within wide limits, yet it is required that the polymer be soluble in a suitable solvent. Therefore, x preferably lies between 5 and 10,000, depending on the polymer and the solvent.
In one embodiment, the thermoplastic polymer may be selected from the group consisting of polysulfones , polyimides, polyamide, polyphenyleneoxide and polybenzimidazole . One preferred embodiment is the use of polyether sulphone (PES) as a thermoplastic polymer. PES has no aliphatic CH-bonds, but only aromatic CH-bonds.. Due to the higher bonding energy of aromatic CH-bonds compared to that of aliphatic CH-bonds, PES possesses the stability for use in an oxygen-containing environment in a fuel cell. In addition, PES possesses no easily-hydrolysed functional groups, for example ester functional groups, but ' only sulphone or ether groups which do not participate in hydrolytic reactions. Advantageously, PES may undergo phase inversion to produce a highly porous polymer membrane matrix. A further advantage of PES is that PES may be processed by injection moulding or by extrusion ' . In this regard, the aromatic rings in the PES polymer spine offer the possibility of introducing ionic groupings chemically to produce ion-conductive polymers..
The polymer membrane matrix .. may be prepared using phase inversion process known in the art. In the phase inversion process, for example, a liquid film may.; be developed by casting a homogenous polymer solution ontio ' a glass panel, and immersing it into a coagulation bath consisting of a mixture of anti-solvent and the solvint. During the immersion, the solvent and anti-solvent on the surface of membrane may undergo inverse diffusion, driven by osmotic pressure. Such solvent exchange process causes the polymer chains in the polymer solution to shrink and coalesce, resulting in the formation of a porous polymer membrane matrix. In one preferred embodiment, said phase inversion is undertaken in a coagulant bath containing a mixture of solvent and anti-solvent.
The volume ratio of the anti-solvent to solvent used in the coagulation bath may be in the range of 0.5:1 to 1:1.5. Preferably, the volume ratio of the anti-solvent to solvent used in the coagulation bath is 1:1. An increase in the volume ratio of anti-solvent to solvent used in ' , the coagulation liquid may result in a reduction of surface pore apertures due to excessive shrinking of the polymer chains at the interfacial region. On the contrary, a decrease in the volume ratio of water to solvent used in the coagulation liquid may result in smaller pore volumes in the polymer membrane due to incomplete phase separation.
The phase inversion step may be undertaken for at least 12 hours. The phase inversion step may be undertaken in the time range of 12 hours to 48 hours, 18 hours to 48 hours, 24 hours to 48 hours, 30 hours to 48 hours, 36 hours to 48 hours, 42 hours to 48 hours, 12 hours to 42 hours, 12 hours to 36 hours, 12 hours to 30 hours, 12 hours to 24 hours. Preferably, the phase inversion step may be undertaken . in the time range of 12 hours to 24 hours.
After polymerization, the resultant polymer matrix may display a grid-plug morphology, in which both the ionic- conductive polymer and the matrix may be discerned in the scale of less than 1 micrometer.
The resultant polymer matrix may be subjected to UV irradiation to induce further polymerization of the monomers absorbed by the pores of the porous polymer membrane. Typically, the UV irradiation may be carried out in an UV oven that utilizes an UV-A irradiation source.
Hence, the resultant polymer electrolyte membrane may exhibit properties such as higher proton conductivity, enhanced cell voltages, higher ' glass transition temperature, and reduced thermal degradation.
Brief Description of Drawings
The accompanying drawings illustrate a disclosed embodiment and serve to explain the principles of the disclosed embodiment. It is to be understood, however, that the drawings are designed, for purposes, of illustration only and not as a definition of the limits of the invention.
Fig. 1 is a graph showing solvent uptake of a porous PSU membrane and a dense PSU membrane with different water- NMP volume ratios produced by Example 1.
Fig.- 2 is a graph showing the variation of the hydrophilic PAMPS-MBA phase loading in the porous PSU matrix against varying concentrations of an AMPS monomeric solution in NMP/H20 and in pure H 2 0.
Fig. 3 shows a series of field emission scanning electron microscopy (FESEM) images of the cryofractured surfaces of (a) the pristine porous PSU membrane, (b) the PAMPS- BA-filled PSU membrane of Sample 1 in Table 1, (c) the PAMPS-MBA-filled PSU membrane of Sample 2 in Table 1, (d) the PAMPS-MBA-filled PSU membrane of Sample 3 in. Table; 1, and (e) the PAMPS-MBA-filled PSU membrane of Sample 4 in Table .1.
Fig. 4 is a graph showing thermogravimetric analyzer (TGA) data for the unfilled PSU membrane and the PAMPS-MBA- filled PSU membrane (Sample 3).
Fig. 5 is a graph showing differential scanning calorimetry (DSC) profiles of- the PAMPS-MBA-filled PSU membranes .
Fig. 6 is a schematic illustration of entrapment of PAMPS-MBA micro-gels on the inner tube wall of porous PSU membrane .
Fig. 7 is a graph showing experimental and theoretical ion exchange capabilities (IEC) values of the PAMPS-MBA filled PSU membranes.
Fig. 8 is a graph showing the relationship between proton conductivity, and ion exchange capacities of different membranes, at room temperature.
Fig. 9 is a graph showing the influence of temperature on the conductivity of the membranes, of Samples 1 to 4.
Fig. 10 is a graph showing the electrochemical performance of the membranes of Samples 3 to 4 and a. Nafion membrane in a hydrogen-driven single fuel cell at room temperature.
Examples
Non-limiting examples of the invention will be further described in greater detail by reference to specific Examples, which should not be construed as in any way limiting the scope of the invention.
Example 1
Preparation of porous membrane
In this example, a porous polysulfone (PSU) membrane is fabricated using a phase inversion method. 0.8 g of PSU is dispersed in a mixed solvent of 2.5 mL of N- Methylpyrrolidone (NMP) and 0.8 mL of ethanol, and stirred at room temperature for 24 hours to form a homogenous viscous solution. It is understood that the choice of solvent is dependent on the polymer used to form the porous membrane .
The solution is then cast on a clean glass panel using a film applicator with a gate thickness, of 150 μπι. The glass panel was swiftly immersed in a coagulation bath consisting of NMP and water in the ratio of V NM p:V water = 1:1 for 24 hours. During the immersion, the solvent phase (NMP) and the non-solvent phase (water) on the membrane surface undergo inverse diffusion driven by osmotic pressure. This solvent exchange process causes the PSU chains to shrink and coalesce. The deposition of aggregated PSU chains then forms the porous matrix. A white film of the porous polymer matrix is obtained on the glass panel after the inverse solvent extraction. This porous PSU membrane is used in the following examples.
An appropriate initial concentration gradient of NMP across the boundary where the solvent exchange takes place affects the porous structure of the PSU membrane generated eventually. Specifically, it is important not to use water as a bored medium to precipitate the PSU matrix, as a dense film surface would result otherwise. This is illustrated in Fig. 1 which shows that an increase in the water content in the coagulation bath brings about a reduction of surface pore apertures due to excessive shrinking of the polymer at the interfacial region..
Referring now to Fig. 1, it is shown that with an increase in the volume of water i the coagulation bath from V„ater : NMP = 1:1 to V wat er : NM p = 9:1, the coagulation bath solvent uptake of the PSU membrane decreases because of the smaller pore volume due to excessive polymer, shrinking. On the other hand, a decrease in the water content also results in smaller pore volumes throughout the membrane due to incomplete phase separation. Specifically, an increase in the NMP volume from V wat er : NM p = 1:1 to V wat er : NM p = 1:9 gives rise to lower solvent uptake because of the incomplete phase separation. Therefore, the diffusion equilibrium of V NMP : V wat er = 1:1 gives the highest coagulation bath solvent uptake and is thus used in this Example.
The porous PSU membrane is made up .of pore tubes and the membrane matrix. Accordingly, the coagulation bath solvent must, occupy the pore tubes as well as the PSU membrane matrix. Referring again to Fig. 1, the solvent uptake for the PSU membrane matrix arid the total solvent uptake for the whole PSU membrane is shown. The solvent uptake for the pore tubes can thus be gauged by deducting the solvent uptake of the PSU membrane matrix ' from the total liquid uptake of the porous PSU membrane. From Fig. 1, it can be seen that the V NMP :V water of 1:1 leads to the highest solvent uptake for the pore tubes, indicating that pore volume is the greatest when the membrane is immersed in this solvent composition. On the basis of this assessment, it can be inferred that the solvent occupying the pore tubes must have similar composition to the coagulation bath solution used. However, the solvent occupying and swelling the PSU membrane matrix must have a higher percentage of NMP than the coagulation bath solution because NMP is a good solvent of PSU. It is imperative to maintain such a liquid-borne state because it allows the pore tubes throughout the PSU membrane to be accessible for the monomeric solution, i.e. AMPS-MBA- Initiator solution, to enter through diffusion in Example 2 below.
Example 2
Conducting pore-filling via impregnation and polymerization
The. porous PSU membrane produced in Example 1 is transferred directly from the coagulation bath to a specially formulated feed solution. The solution contains an ionic monomer (2-acrylamido-2-methylpropane sulfonic acid (AMPS) ) , a crosslinker (methylene bisacrylamide (MBA) ) , and an UV initiator (ammonium persulfate (APS) ) , in a mixture of NMP and water ( V NMP : V wa ter = 2:8). The experiment was repeated with different monomer formula compositions as detailed in Table 1 below.
Table 1: Different compositions of monomer formula (AMP.S- MBA-APS) in the mixed solvent of NMP and water for filing PSU porous membrane : ■ Sample Monomer formula composition
1 2 3 4
AMPS (mmol/ml) 1-5 3.0 4.0 4.5
NHiS 2 be (APS) 1.65X10- 2 3.30x10 "2 4.40x10- 2 5.00x10 "2 (mmol/ml)
MBA immol/ml) 1.0x10 "2 2.0Χ1Ό 2 2.7 10 2 3.0x10 "2
Mixed solvent (ml) of
19.2 19.2 19.2 19.2 N P and H 2 0 (v/v=2/8)
The PSU membrane was soaked in the feed solution for 2 hours to allow the monomers to fill the pore tubes. Diffusion of the monomers into the pore tubes of the membrane takes place due to the concentration gradients. The volume ratio of NMP to H0 used in the preparation of the monomer solution is an important factor affecting the filling of the pore tubes. When a higher volume of NMP is used, the monomer formula composition is prone to polymerization at room temperature because AMPS molecules show a tendency to form a pseudo-gel when NMP volume increases. This gel is the precursor leading to polymerization at ambient temperature. In contrast, the use of pure water results in relatively lower solvent loading.
Referring now to Fig. 2, it can be seen, that the loading percentage when pure H 2 0 was used is lower than; the corresponding loading percentage when an NMP/H 2 0 solution in the ratio of 2:8 was used. Specifically, it was found ' that the ratio of V NMP :V H2 o = 2:8 provides the ' best solvent condition to achieve the highest filling result. As seen in Fig. 2, the maximum loading was obtained from the 4.0M AMPS solution (Sample 3) but a further increase in the AMPS concentration to 4..5M in Sample 4 led to a slight reduction in the PAMPS loading and without being bound by theory, this is thought that this is due to the converse relationship between viscosity and diffusivity. Additionally, a higher concentration of monomeric solution results in a gelling effect that reduces polymerization conversion.
In order to achieve the maximum loading of monomers in the pore tubes, the AMPS concentration was increased to the highest possible level of 4.5 mmol/mL. After soaking for a length of time, the end point was determined when gellation occurred. The monomer-loaded membrane was sandwiched between two glass plates and subjected to UV-irradiation in a UV reactor. (Hg lamps, 8x8 ) for about 2 hours to polymerize the entrapped monomers. Finally, the resulting membranes were soaked in DI water at room temperature for 24 hours to remove any unbound water-soluble species.
With an increase in monomer concentration, the resulting membranes show an increase in filling. This was verified by the change of their matrix morphologies. Referring to Figs. 3c-3e, the FESEM images of the membrane matrices display a grid-plug morphology, in which both the PSU membrane, and the PAMPS-MBA phases can be discerned in the scale of less than 1 μπι. The two phases form an interpenetrating network since the previously-formed pores of the PSU membrane are mutually inter-connected.
Example 3
Characterization experiments on the pore-filling polysulfone-AMPS membranes
Thermal analysis of the membranes
The cross-section of the membranes by cryogenic fracture was examined by field emission scanning electron
2] microscopy (FESE ) on a JEOL JSM-6700F instrument obtained from JEOL, Japan, and operated at 5kV.
The thermal degradation of the AMPS filler and the PSU. grid matrix of the membranes from Example 2 was recorded on a high resolution . thermogravimetric analyzer (TA Instruments Q500 obtained from TA Instruments, New Castle, Delaware, United States of America). The samples were heated to 800°C under a dry nitrogen purge at a rate of 100 ml/min with a heating rate of 10°C/min. The TGA data for the membranes is shown in Fig. 4. Referring to Fig. 4, the pristine unfilled PSU membrane showed negligible thermal degradation before 500°C, while the filled membrane reveals a steep mass-loss slope in the temperature range from 200°C to about 320°C due to thermal degradation of the PAMPS-MBA phase. After that, the slower mass-loss gradient of the. filled membrane in the range from 320°C to 500°C was due to further degradation of the residues from the preceding process. It is thus rational to take the gap between these two curves at the medium temperature range of 320°C-500°C as the approximate loading of the ion conducting PAMPS-MBA phase.
Next, the glass transition behavior of the composite membranes was measured on a differential scanning calorimeter (DSC 2920, TA Instruments, New Castle.,. Delaware, United States of America) . About 5 to 10 mg of each sample were loaded into a pressure DSC cell and the test was conducted through a two-scan mode. In the first scan, the samples were heated from room temperature to 200°C at a rate of 10°C/min and cooled down to -20°C at the same rate of 10°C/min. The same heating rate was applied during the second scan to heat the samples from -10°C to 250°C. The energy-temperature profile produced by the second scan is shown in Fig. 5.
Utilization of the partially solvated PSU matrix formed by phase inversion is distinguished from the utilization of dry porous membranes because of the presence of. a dynamic matrix along the pore walls. Hence, the dynamic matrix allows the monomer molecules to diffuse from the pore walls into the matrix when the monomers fill the pores. As a result, the polymer electrolyte formed would be physically anchored to the host matrix in-situ. The anchoring mechanism was demonstrated by the DSC analysis of the four membranes in Fig. 5. Referring to Fig. 5, a peak at a temperature of about 190°C for all the four samples of filled membranes shows that the glass transition temperature (T g ) of pristine PSU was about 190°C. The filled membrane of Sample 1 (PSU-PA PS-MBA-1) showed both the T g of PSU and another glass transition temperature at 152°C. The T g of 152°C was indicative of the presence of the ion conducting PAMPS-MBA gel phase in the pore tubes of the PSU This assignment was in line with the surface morphology of the same Sample 1 as seen in Fig. 3(b) . Sample 2 (PSU- PAMPS-MBA-2) had an increased PAMPS-MBA loading and displayed a. much stronger T g peak of the PAMPS-MBA gel phase as compared to PSU-PAMPS-MBA-1. . However, there is a tangible shift of the T g of PAMPS-MBA down to 135°C, " which indicated that the majority of the PAMPS-MBA gel was ;freed from being entrapped in the matrix of pore wall in contrast to Sample 1, as illustrated in Fig. 6. The same trend is reflected in Samples 3 and 4 ( PSU-PAMPS-MBA-3 and PSU- PAMPS-MBA-4) with a downward shift of T g to about 100°C. However, the T g peak is broadened which is likely due to the variation of crosslinking densities in the PAMPS-MBA gels formed.
Still referring to Fig. 5, it is further observed that for PSU-PAMPS-MBA-3, a shoulder■ peak of the T g of pristine PSU emerged, and it became stronger in the DSC diagram of PSU-PAMPS-MBA-4. The shoulder peak could be interpreted to. indicate the presence of loosened PSU matrices close to the tube wall due to the inter-penetration with the PAMPS-MBA gel phase. It is apparent in the DSC diagram of PSU-PAMPS- MBA-4 in Fig. 5 that the infiltration of the PAMPS-MBA into the pore walls is enhanced by the increase in monomer concentration . .
In short, the DSC analysis results prove the presence of the glass transition behaviors of three phases, i.e. the PSU phase, the PAMPS-MBA gel phase and the interfacial boundary.
Assessment of proton conductivity
The ion exchange capacity (IEC). values of the membranes were determined using a titration technique. Each sample was soaked in 15 ml of a 0.05M sodium chloride solution for 24 hours to reach an ion exchange equilibrium between the proton and sodium ions . The liquid phase was then titrated to a pH of 7.0 using 0.05M. of odium hydroxide aqueous solution, in which the pH is checked by a pH meter obtained from Schott AG, Germany. The IEC of each membrane is determined by three repeated titrations and ■ the results are shown in Fig. 7.
Referring to Fig. 7, as compared to the theoretical IEC, the experimental IEC measures the accessibility of the pendant sulfonic acid groups in the PSU polymer matrix and therefore reflects the situation of hydrophobic blockage. The theoretical IEC is determined by the Helfferich' s- method, which is the rati'o of the mole number of ' AMPS in the feedstock to the mass of PSU resin used with the units of meq/g. ..
As seen in Fig. 7, the experimental IEC was lower than, its theoretical value and it is postulated that there may be two reasons for this: (i) the conversion of ionic AMPS monomers to poly-AMPS cannot be quantified; and (ii) there is always the presence of a small number of acidic ionic groups which are blocked by the hydrophobic polymer segments and therefore cannot be detected by the titration to produce an equivalent value as the theoretical IEC.
It is shown in Fig. 7 that the discrepancy between the theoretical and experimental IEC values are closer for Samples 3 and 4 than that for Samples 1 and .2. The smaller discrepancy indicates a high rate of polymerization and a highly interconnected hydrophilic PAMPS-MBA phase. Evidently, the higher the loading of the PAMPS-MBA phase, the better the inter-connection of the ion conducting phase throughout the membrane.
Fig. 7 . further confirms that the experimental IEC values and the AMPS loading of the. different membranes detected by TGA in Fig. 2 are proportional. Specifically, the loading percentage and the IEC. values increase as the AMPS concentration increases and peaks at Sample 3. " Both the loading percentage and the IEC value then decrease in Sample 4. The proportional variations prove that these two analytical methods confirm the ■ concentration of PAMPS-MBA in each membrane sample.
Next, there are two common methods to measure proton conductivity. The first method is the two-probe transverse measurement method which measures proton conductivity through a membrane. The second method is the four-probe longitudinal measurement . method which measures planar conductivity of a membrane.
The four-electrode- technique is used because by this approach, the interfacial impedance between the measurement probe and membrane will not affect the main signal when the scanning frequency is low. The measurement kit consists of two rectangular Teflon plates. One ' of the Teflon plates comprises two parallel stainless strips 2 cm apart as the outer current-carrying electrodes. The other plate comprises two gold (Au) wires 1 cm apart and the Au wires are fixed as the inner potential-sensing electrodes.
A measurement membrane sample 1 cm wide and 4 cm long was sandwiched between the two plates. The impedance values were recorded on an Autolab instrument at galvanostatic mode with an AC current amplitude of 0.1 mA over . a frequency range from 50 Hz to 1 MHz. A spectrum of frequency-dependent impedance, i.e. the Nyquist plot, was obtained from the frequency scanning. The resistance (R) due to proton transfer in the membrane could be read from the real part of the spectrum. The proton conductivity (C) was calculated using the formula
C = L/RWd
where
. L is the distance between the. two potential-sensing ' electrodes ;
W is the width of the membrane; and
d is the thickness of the membrane.
The conductivity of the membranes was evaluated at temperatures of up to 90°C at 100% relative humidity. For each measurement at a setting temperature, a 30 minute span was set to allow the measurement setup to reach a steady state before the impedance spectrum was recorded.
Fig. 8 shows the relationship between IEC and proton conductivity. Referring to Fig. 8, the conducting threshold takes place at the experimental IEC value of Sample 3 at 2.43 meq/g. This ' indicates the efficiency of the proton conducting channels developed in the PSU membrane matrix at room temperature because the theoretical IEC value of about 2.5 meq/g is very close to the experimental IEC value. Accordingly, the ionic groups in the proton conducting channels were not significantly blocked by the hydrophobic polymer segments and were available to contribute to ionic exchange or conduction.
Fuel cell test
To ' make the membrane electrode assembly (MEA) , the anode and cathode, obtained from SGL, Germany are carbon sheets coated with a layer of 20 wt% carbon-supported platinum (Pt) catalyst obtained from E-TEK, Natick, Massachusetts, United States of America. The catalyst was loaded at 2 mg/cm 2 on either electrode, or the Pt was loaded at 0.4 mg/cm 2 . The effective area . was 5 cm 2 on both electrodes.
When evaluating the performance of the membranes in the .single fuel cell, the flow rates of the two respective feed streams purging the two electrodes were varied . /in order to change the current density. The anodic ; feed stream consisted of pure H 2 and the cathodic feed stream consisted of pure 0 2 , and the molar ratio of hydrogen gas to oxygen gas (H 2 /0 2 ) was fixed at 1.15:2. The MEA was operated at 1 bar and the anodic feed stream was not humidified. The electrochemical performance of the membranes ' in the fuel cell was measured and the results are shown in Fig. 9.
Accordingly, referring to Fig. 9, it is known that proton conductivity can reflect the flux of protons across the membrane with an increase in the voltage of a fuel cell, Accordingly, the four-electrode method described above was used to measure the proton conductivities of the different membrane samples and Fig. 9 shows the influence of temperature on the conductivity of the membranes.
Referring to. Fig. 9, Sample 1 has a relatively gentler slope as compared to the other three samples. This indicates that Sample 1 contains slim proton conducting channels and is therefore almost kinetically stable. On the contrary, Samples 2 to 4 display similar slopes, indicating that they have similar proton conducting kinetic constants. This can be related to the fact that the filled PAMPS-MBA phase in these membranes becomes a hydrogel when it absorbs water. Accordingly, the proton conducting mechanism is evidently similar in Samples 2 to 4 but differs only by the concentration of the pendant sulfonic acid groups. Furthermore, although Sample 3 has been detected to have a greater IEC value than Sample 4, Sample 4 still exhibited higher proton conductivity over the low temperature range than Sample 3. This discrepancy can be explained by the extent of the crosslinking of the PAMPS-MBA hydrogel phase of Samples 3 and 4 in accordance with the feed composition; given in Table 1. Specifically, the data of Fig. 9 shows' that MBA possesses a higher diffusion coefficient than AMPS to enter the pores of the PSU membrane. As a result, the extent of crosslinking of the PAMPS-MBA gel phase in Sample 3 is lower than that in Sample 4. The slightly different glass transition temperatures of Samples 3 and 4 in Fig. 5 confirm this. The higher crosslinking extent of PAMPS-MBA gel in Sample 4 results in an underestimation of its actual IEC.
In addition, the proton conduction activation energy of. the pore-filled membranes was significantly lower than that of Nafion-117 membrane as evidenced in . Table 2 below. This outcome is attributed to the trait of hydrogel. A proton hydrogel matrix cannot be used alone in a PE due to its poor mechanical integrity, even though ^ it possesses high conductivity. However, the disclosed, pore-filling tactic allows the pores of the mechanically strong PSU matrix to be filled with PAMPS-MBA hydrogel. In contrast, the proton transport in the Nafion matrix takes place through hydrophilic channels in nano-scale and hence, the narrower pore channels have a higher activation energy barrier.
Table 2: Activation energies (£a) of the filled PES membranes and Nafion membrane
Finally, the pore-filled membranes of Samples 3 and 4 are examined in a single hydrogen fuel cell. In line with the conductivity measurements shown in Fig. 9, Sample 4 also, exhibited better performance as compared to Sample 3 at room temperature. According to the polarization curves of Fig. 10, Samples 3 and 4 hold higher cell voltages with an increase in current density than the Nafion-117 membrane This outcome also confirms the previous conclusion that the embedded PAMPS-MBA hydrogel phase allows a higher flux of protons than the hydrophilic channels in the Nafion membrane. Applications
The disclosed method may be. used for producing polymer electrolyte membranes suitable for use in hydrogen, fuel cells . (including proton exchange membrane fuel cells, direct methanol fuel cells, alkaline fuel cells), electrodialysis cells for desalination, pervaporation membrane reactors and food preservation membranes.
Advantageously, the disclosed method involves the use of a porous polymer membrane matrix in a solvated state which allows the passages and pathways of the porous polymer membrane to be accessible for the monomers through diffusion. Due to the small molecular sizes of the ionic monomers, the ionic monomers can fill the pores of the porous polymer membrane matrix with higher diffusion kinetics. Hence, the electrolyte polymer formed may be physically anchored to the porous polymer membrane matrix in-situ. Also, the disclosed method may not suffer from the problem of forming polymer electrolyte membranes with pores that are incompletely filled with the proton conducting phase .
During the polymerization step of the disclosed method, the porous polymer membrane matrix may be exposed to an anti-solvent and solvent solution. Advantageously, the use of a higher volume of solvent relative to the volume of anti-solvent during the polymerization step may aid in the polymerization of the ionic monomer because a higher portion of solvent resists or at least ameliorates over- swelling of the passages and pathways of the polymer matrix, thereby keeping the passages and pathways open for polymerixation and cross-linking of the ionic monomeric groups to the passages and pathways of the polymer matrix. This results in a ionic phase being present throughout the bulk of the membrane matrix.
Polymer electrolyte membranes formed . using. the disclosed method may advantageously exhibit properties such as . higher proton conductivity, enhanced cell voltages, higher glass ' transition temperature, and reduced thermal degradation relative to those polymer electrolyte membranes which have been formed by a plug-grid method.
It will be apparent that various other modifications and adaptations of the invention will be apparent to the person skilled in the art after reading the foregoing disclosure without departing from the spirit and scope of the invention and it is intended that all such modifications and adaptations come within the scope of the appended claims.