POROUS ETCHED ION-TRACK POLYMER MEMBRANE AS A SEPARATOR FOR A BATTERY

Abstract

The present invention relates to the use of a porous polymer etched ion-track membrane as separator for batteries comprising a positive electrode, a negative electrode and a liquid electrolyte comprising at least one salt of a cationic ion in solution in a solvent, and to batteries comprising such a membrane as porous separator.

Claims

1. A method to suppress or at least reduce the redox shuttle phenomenon in batteries comprising the use of a porous polymer etched ion-track membrane as a separator in a lithium-sulfur battery comprising at least one positive electrode, at least one negative electrode and at least one liquid electrolyte comprising at least one salt of a cationic ion selected in the group consisting of lithium, sodium, potassium and calcium in solution in at least one solvent, wherein: said polymer is selected from the group comprising polyethyleneterephtalate (PET), polycarbonate (PC), polyethylene (PE), polypropylene (PP), polyimide (PI) polytetrafluoroethylene (PTFE), and polyvinilidene fluorure (PVdF), said membrane has a frontside, a backside, and a thickness, the frontside of said membrane is designed to face the positive electrode, the backside of said membrane is designed to face the negative electrode, the thickness of said membrane between said frontside and said backside is ranging from about 5 to 60 m, said membrane comprises a plurality of nanochannels, the majority in number of said nanochannels is continuous and has an opening tip at the frontside and an opening tip at the backside of said membrane, the opening tip diameter of each of said nanochannels at the frontside of said membrane (D.sub.FS) ranges from about 10 to 200 nm, the opening tip diameter distribution of the nanochannels at the front side and at the back side is narrow.

2. The method according to claim 1, wherein: thickness of said membrane preferably ranges from 5 to 30 and/or the opening tip diameter of each of said nanochannels at the frontside of the membrane ranges from 50 nm to 150; and/or the opening tip diameter of each of said nanochannels at the backside of the membrane (D.sub.BS) ranges from 10 nm to 600 nm.

3. (canceled)

4. (canceled)

5. The method according to claim 1, wherein the opening tip diameter of each of said nanochannels at the backside of the membrane (D.sub.BS) is substantially identical to the opening tip diameter of each of said nanochannels at the frontside of the membrane (D.sub.FS) and ranges from 50 to 150 nm.

6. The method according to claim 5, wherein said nanochannels have a cylindric shape and their longitudinal axis are substantially parallel oriented to one another.

7. The method according to claim 5, wherein said nanochannels are interconnected.

8. The method according to claim 1, wherein the opening tip diameter of each of said nanochannels at the backside of the membrane (D.sub.BS) is greater or smaller than the opening tip diameter of each of said nanochannels at the frontside of the membrane (D.sub.FS) and ranges from 10 to 600 nm.

9. The method according to claim 8, wherein said nanochannels have a conical shape and their longitudinal axis are substantially parallel oriented to one another.

10. The method according to claim 1, wherein the said polymer is polyethyleneterephtalate.

11. The method according to claim 1, wherein the ratio of the total volume of the nanochannels/total volume of the membrane ranges from 2 to 50%.

12. The method according to claim 1, wherein the nanochannels density in said membrane ranges from to 10.sup.6 to 10.sup.12 nanochannels/cm.sup.2.

13. The method according to claim 1, wherein at least one of the surfaces of the porous membrane comprises a coating of at least one layer of oxide of a metal or of a metalloid.

14. The method according to claim 13, wherein said at least one surface is the surface designed to face the positive electrode.

15. The method according to claim 13, wherein said layer is a layer of TiO.sub.2, SiO.sub.2 or Al.sub.2O.sub.3.

16. A separator for a battery comprising at least one positive electrode, at least one negative electrode and at least one separator impregnated by a liquid electrolyte comprising at least one salt of a cationic ion selected in the group consisting of lithium, sodium, potassium and calcium in solution in at least one solvent, wherein said separator is a porous polymer etched ion-track membrane such as used in claim 1.

17. The separator of claim 16, wherein said membrane is inserted between two microporous polymer supports.

18. A lithium-sulfur battery comprising at least a negative electrode, a positive electrode, a liquid electrolyte comprising at least one salt of a cationic ion selected in the group consisting of lithium, sodium, potassium and calcium in solution in at least one solvent, and a porous separator, wherein said porous separator is a separator according to claim 1.

17. Method to suppress or at least reduce the redox shuttle phenomenon in batteries comprising the use of a porous polymer etched ion-track membrane as a separator in battery comprising at least one positive electrode, at least one negative electrode and at least one liquid electrolyte comprising at least one salt of a cationic ion selected in the group consisting of lithium, sodium, potassium and calcium in solution in at least one solvent, wherein: said polymer is selected from the group comprising polyethyleneterephtalate (PET), polycarbonate (PC), polyethylene (PE), polypropylene (PP), polyimide (PI) polytetrafluoroethylene (PTFE), and polyvinilidene fluorure (PVdF), said membrane has a frontside, a backside, and a thickness, the frontside of said membrane is designed to face the positive electrode, the backside of said membrane is designed to face the negative electrode, the thickness of said membrane between said frontside and said backside is ranging from about 5 to 60 m, said membrane comprises a plurality of nanochannels, the majority in number of said nanochannels is continuous and has an opening tip at the frontside and an opening tip at the backside of said membrane, the opening tip diameter of each of said nanochannels at the frontside of said membrane (D.sub.FS) ranges from about 10 to 200 nm, the opening tip diameter distribution of the nanochannels at the front side and at the back side is narrow, at least one of the surfaces of the porous membrane comprises a coating of at least one layer of oxide of a metal or of a metalloid.

20. The method according to claim 19, wherein the membrane, the nanochannel and the polymer are as described in claim 4.

21-32. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0112] FIG. 1 gives the results of specific capacity (mAh/g) as a function of cycle number for the batteries B1 and B4.

[0113] FIG. 2 gives the results of Coulombic Efficiency (%) as a function of cycle number for the batteries B1 and B4.

[0114] FIG. 3 gives the results of specific capacity (mAh/g) as a function of cycle number for the batteries B2 and B3.

[0115] FIG. 4 gives the results of Coulombic Efficiency (%) as a function of cycle number for the batteries B2 and B3.

[0116] FIG. 5a and FIG. 5b compares the results of specific capacity (mAh/g), and of coulombic efficiency (%) as a function of cycle number for coin cells characterized by different pore size at a fluence of 10.sup.7 cm.sup.2.

[0117] FIG. 6a and FIG. 6b compares the results of specific capacity (mAh/g), and of coulombic efficiency (%) as a function of cycle number for coin cells characterized by different pore size at a fluence of 10.sup.8 cm.sup.2.

[0118] FIG. 7a and FIG. 7b compares the results of specific capacity (mAh/g), and of coulombic efficiency (%) as a function of cycle number for coin cells characterized by different pore size at a fluence of 10.sup.9 cm.sup.2.

[0119] FIG. 8a and FIG. 8b compares the results of specific capacity (mAh/g), and of coulombic efficiency (%) as a function of cycle number for coin cells characterized by different fluences, and for an immersion time of 3.5 min.

[0120] FIG. 9a and FIG. 9b compares the results of specific capacity (mAh/g), and of coulombic efficiency (%) as a function of cycle number for coin cells characterized by different fluences, and for an immersion time of 4.5 min.

[0121] FIG. 10a and FIG. 10b gives the results of specific capacity (mAh/g), and of coulombic efficiency (%) as a function of cycle number for a coin cell characterized by membrane having conical pores.

DETAILED DESCRIPTION

Examples

Example 1: Preparation of a PET Porous Etched Ion-Track Membrane According to the Invention, Optionally Comprising a Surface Layer of TiO.SUB.2

[0122] 1.1. Swift Heavy-Ion Irradiation

[0123] A PET foil has been irradiated by Au-ion at the Universal Linear Accelerator (or UNILAC) of the GSI Helmholtz Centre for Heavy Ion Research in Darmstadt, Germany. The main branch of this apparatus consists of two ion source terminals followed by a Radio Frequency Quadrupole and by an Interdigital linac IH linac accelerator resonating at 36 MHz up to the energy of 11.4 MeV per nucleon (MeV/u) corresponding to approximately 15% of the velocity of light. The main part then is operated by a classical linac of the Alvarez type which resonates at 108 MHz. Final energy adjustment can be performed in the last section consisting of a series of single-gap resonators. Ion beams of such high energy have a penetration range in polymers of about 120 m. Given this large range, foil stacks (e.g., ten foils 12 m thick, or four foils 30 m thick) can be irradiated. Each ionic projectile induces electronic excitation and ionisation processes in a cylindrical zone along its trajectory. In polymers, chemical bonds are destroyed and small volatile fragments (e.g., H.sub.2, CO, CO.sub.2, hydrocarbons) easily outgas. This damaged region is called the ion track and has a typical diameter of few nanometres.

[0124] By suitable adjustment of the ion beam and monitoring the flux (beam current), the applied ion fluence can be adjusted over a wide range, from exposure to a single ion (single track) up to more than 10.sup.12 ions/cm.sup.2. At the UNILAC beamline of the GSI facilities, irradiation with a broad homogenous beam is obtained by magnetic defocusing. Samples of up to several square centimetres in size can be exposed. The resulting ion tracks are stochastically distributed and oriented in parallel across the sample. The ions are registered either by a three-foil detector or by a solid-state particle detector placed behind the sample. As soon as the given detector has registered the desired number of ion impacts, the entire ion beam is deflected by an electrostatic chopper system. This can go down to the detection of single ion impacts (Toimil-Molares M. E., Beilstein Journal of Technology, 2012, 3, 860-883, see in particular FIG. 1).

[0125] In practice, circular PET foils having the following dimensions: thickness: 19 m; diameter: 30 mm, were exposed to 11.4 MeV/u Au ions at the X0 beamline of the UNILAC accelerator for about 10 seconds.

[0126] Other ions such as Bismuth ions, Uranium ions or also lighter ions like Xenon ions can equally be used to produce a polymer foil having ion tracks.

[0127] 1.2. Chemical Etching

[0128] After the irradiation process, the irradiated PET foils were chemically etched by immersion in an etching solution to selectively dissolve the ion tracks and subsequently enlarge them into nanochannels.

[0129] In practice, the irradiated PET foils obtained at the end of step 1.1. were exposed to UV light and subsequently immersed in a 6M NaOH solution at 50 C. for various times to obtain membranes with nanochannels with diameter varying between about 115 to about 198 nm.

[0130] The diameter of the nanochannels increases as a function of the immersion time is given in the following Table 1:

TABLE-US-00001 TABLE 1 Immersion Diameter of the time nanochannels Number of Membrane (in min) (in nm) nanochannel/cm.sup.2 M1 3.5 115 10.sup.9 M2 4.5 161 10.sup.8 M3 5.5 198 10.sup.8 M5 3.5 115 10.sup.7 M6 3.5 115 10.sup.8 M7 4.5 161 10.sup.7 M8 4.5 161 10.sup.9 M9 5.5 198 10.sup.9

[0131] At the end of the chemical etching process PET porous etched ion-track membranes having the following characteristics were obtained: [0132] thickness: 19 m [0133] shape of the nanochannels: cylindrical [0134] opening tip diameter of each of said nanochannels at the frontside of about 115 nm for membranes M1, M5 and M6, 161 nm for membranes M2, M7 and M8; and 198 nm for membranes M3 and M9 respectively, [0135] opening tip diameter of each of said nanochannels at the backside: of about 115 nm for membranes M1, M5 and M6, 161 nm for membranes M2, M7 and M8; and 198 nm for membranes M3 and M9.

[0136] 1.3. Surface Modification by ALD

[0137] Membrane M3 obtained at step 1.2. has then been coated by a layer of TiO.sub.2 by ALD, using a Picosun R-200 Basic flow-type reactor. The precursors used were titanium isopropoxide (Ti(OCH(CH.sub.3).sub.2).sub.4, TTIP) and water, forming a TiO.sub.2 layer at the end of each cycle. The temperature was held at 110 C. to prevent the condensation of H.sub.2O. Nitrogen gas (N.sub.2) was used for the purging gas, at a flow rate of 120 sccm. Each cycle consists of the following:

[0138] 1. A 2 s pulse of TTIP

[0139] 2. 50 s of N.sub.2 purging

[0140] 3. A pulse H.sub.2O pulse with N.sub.2 carrier gas at 200 sccm for 0.4 s

[0141] 4. 50 s of N.sub.2 purging

[0142] This process led to a layer of TiO.sub.2 having a thickness of about 20 nm onto the surface of the polymer membrane M3 which results in a final nanochannel diameter of about 160 nm. This membrane was denominated M3-TiO.sub.2.

[0143] All the porous membranes prepared in this example can then been used as a separator in batteries operating by circulation of cationic ions, in particular in a lithium-sulfur battery as described in the following example.

Example 2: Preparation of a Lithium-Sulfur Battery Using the PET Porous Ion-Track Membrane of Example 1 as Separator

[0144] 2.1. Preparation of the Positive Electrode Composite

[0145] Mesoporous carbon: MC and sulfur have first been heat-treated under 300 C. and 60 C. respectively for 24 hours under vacuum to remove any moisture present. The two treated materials were then mixed in a MC to sulphur 50:50 weight percentage mixture, ground in a mortar, and heated in a sealed flask to impregnate the MC with the sulphur, at 155 C. for 6 hours, with a slow 0.5 C. per minute ramp. A slurry composed of the carbon-sulphide composite (80 w %), carbon Super-P (10 w %), and poly (vinylidene fluoride-hexafluoropropylene) (PVdF-HFP) (10 w %), in 40% of dry acetone was made and casted over a thin aluminium foil (15 m). Carbon Super-P is used to increase the surface area of the composite, while PVdF-HFP is a binder for the slurry. The foil has then been cut into 8 mm diameter circular discs for later integration into coin cells for testing. Each disc contained 0.5 mg of sulfur.

[0146] 2.2. Assembly of the Lithium-Sulfur Battery

[0147] A drying pre-treatment has then first been applied to some of the different parts of the battery in order to remove all moisture from the following materials: [0148] 2 Celgard separators in the form of discs having a diameter of 12 mm and a thickness of 10 m, [0149] PET porous etched ion-track membranes M1, M2 and M3-TiO.sub.2 as prepared in Example 1, cut into 13 mm discs, and [0150] the positive electrode composite, cut into a 11 mm disc.

[0151] This pre-treatment has been carried out for 24 hours at 70 for the Celgard separator and PET porous etched ion-track membranes and for 12 hours at 60 C. in vacuum for the carbon-sulphur positive electrode composite.

[0152] A battery B1 in accordance with the present invention was then assembled in a coin cell as follows.

[0153] Membranes were transferred from a sealed vacuum sealed chamber to the glove box without exposure to air. The layers are placed one atop another in the stainless steel coin cell top lead in the quick succession: stainless steel spring, stainless steel current collector with lithium foil (thickness: 100 m, diameter: 11 mm) pressed onto it, first Celgard separator, 5 l drop of an electrolyte composition E1 comprising 1M LiTFSI in a mixture of TEGDME/DOL (1:1 by volume), PET porous etched ion-track membrane M1 of example 1, second Celgard separator, 5 L drop of the electrolyte composition E1, carbon-sulphur composite positive electrode, stainless steel current collector, and finally, the stainless steel bottom lead. The cell is then pressed together to seal it completely. The cell has immediately been transferred for battery cycling tests.

[0154] A second coin cell, identical to B1, but using the membrane M2 prepared hereabove at example 1), step 1.2., has also been assembled according to exactly the same process. This cell was denominated B2.

[0155] A third coin cell, identical to B1, but using the membrane M3-TiO.sub.2 prepared hereabove at example 1), step 1.3., has also been assembled according to exactly the same process. This cell was denominated B3.

[0156] On a comparative purpose, a battery B4 not forming part of the present invention was also assembled according to the same process except that the battery B4 did contain only one Celgard separator and no PET porous etched ion-track membrane of example 1.

[0157] 2.3. Battery Cycling Tests

[0158] The batteries B1, B2, B3 and B4 were tested by galvanostatic cycling with potential limitation (GCPL), being the limiting reduction and oxidation potentials 1.5 V and 3.0 V, respectively. A given current value was imposed on the batteries, but limiting the potential for both charge and discharge.

[0159] The rate used was C10, meaning that the discharge of the batteries occurs over a period of 10 hours. A constant temperature of 20 C. was utilized. Due to the standardized 0.5 mg of sulfur content per cell, 0.0419 mA was imposed on the cell for charge and discharge at C/20, in accordance with the calculated theoretical capacity of the cell. For clarity, only the first few curves will be presented in the annexed FIGS. 1 and 2.

[0160] FIG. 1 gives the results of specific capacity (mAh/g) as a function of cycle number. On FIG. 1, the curve with filled triangles corresponds to the charge capacity of battery B4 not forming part of the present invention, the curve with empty triangles correspond to the discharge capacity of the battery B4, the curve with filled circles corresponds to charge capacity of the battery B1 in accordance with the present invention and the curve with empty circles corresponds to the discharge capacity of the battery B1.

[0161] FIG. 2 gives the results of Coulombic Efficiency (%) as a function of cycle number for the battery B4 not forming part of the present invention (empty squares) and for the battery B1 in accordance with the present invention (filled squares).

[0162] These results show that the coulombic efficiency of the battery B1 according to the present invention, i.e. comprising the PET porous etched ion-track membrane M1 of example 1, is 97% compared to 83% for the battery B4 not forming part of the present invention because not comprising such a PET porous etched ion-track membrane. This indicates a sharp reduction of the shuttle redox reactions within the cell with the use of the membrane.

[0163] FIG. 3 gives the results of specific capacity (mAh/g) as a function of cycle number. On FIG. 3, the curve with filled circles corresponds to the charge capacity of battery B2 according to the present invention but having no TiO.sub.2 coating layer, the curve with empty circles corresponds to the discharge capacity of said battery B2, the curve with filled squares corresponds to charge capacity of the battery B3 in accordance with the present invention having a TiO.sub.2 coating layer and the curve with empty squares corresponds to the discharge capacity of said battery B3.

[0164] FIG. 4 gives the results of Coulombic Efficiency (%) as a function of cycle number for the battery B2 of the present invention but having no TiO.sub.2 coating layer (filled circles) and for the battery B3 in accordance with the present invention having a TiO.sub.2 coating layer (filled squares).

[0165] Data presented on FIG. 4 show a clearly higher coulombic efficiency for the cell B3 with TiO.sub.2 coated channels at approximately 95%, while the cell B1 with the non-coated sample gave 90-92%.

[0166] Without willing to be bound by any theory, the Inventors think that the combined use of TiO.sub.2 and nanochannel membranes might have created a tandem effect of some surface adsorption of the polysulphides and filtering by the nanochannels.

Example 3: Comparative Study of Coin-Cell Type Battery Cycling Tests Comprising Different Membranes with Cylindrical Pores

[0167] 3.1. Preparation of the Membranes and of the Coin Cells

[0168] Membranes M5-M9 were prepared as described in example 1. The irradiated PET foils (thickness 19 m or 12 m when the pores are cylindrical) were exposed to UV light (312 nm wavelength) for three hours on each side, and subsequently immersed in a 6M NaOH solution at 50 C. for various times to obtain membranes with nanochannels with diameter varying between about 115 to about 198 nm.

[0169] Coin cells were prepared as described in example 2. The cell comprises 20 ul of electrolyte per mg of active material, and thus 10 ul of electrolyte per cell.

[0170] 3.2. Battery Cycling Tests

[0171] The batteries were tested by galvanostatic cycling with potential limitation (GCPL), being the limiting reduction and oxidation potentials 1.5 V and 3.0 V, respectively. Upon reaching 1.5V, the battery is considered completely discharged, and upon reaching 3.0V, the battery is considered completely charged, regardless of time elapsed. A given current value was imposed on the batteries, but limiting the potential for both charge and discharge.

[0172] The rate used was C10, meaning that the discharge of the batteries occurs over a period of 10 hours. A constant temperature of 20 C. was utilized. Due to the standardized 0.5 mg of sulfur content per cell, 0.0419 mA was imposed on the cell for charge and discharge at C/20, in accordance with the calculated theoretical capacity of the cell.

[0173] The results are presented on FIGS. 5-10.

[0174] FIG. 5a compares the results of specific capacity (mAh/g) (charge and discharge) as a function of cycle number, for a coin cell comprising a membrane M5 or M7 characterized by an immersion time of 3.5 min and 4.5 min (and thus nanochannels with an average pore diameter of about 115 nm and 161 nm), for a fluence (number of nanochannels/cm.sup.2) of 10.sup.7 cm.sup.2.

[0175] FIG. 5b compares the results of coulombic efficiency as a function of cycle number, for a coin cell comprising a membrane M5 or M7 characterized by an immersion time of 3.5 min and 4.5 min (and thus nanochannels with an average pore diameter of about 115 nm and 161 nm), for a fluence of 10.sup.7 cm.sup.2. The coulombic efficiencies of both batteries are very good.

[0176] FIG. 6a compares the results of specific capacity (mAh/g) (charge and discharge) as a function of cycle number, for a coin cell comprising a membrane M6 or M2 characterized by an immersion time of 3.5 min and 4.5 min (and thus nanochannels with an average pore diameter of about 115 nm and 161 nm), for a fluence of 10.sup.8 cm.sup.2.

[0177] FIG. 6b compares the results of coulombic efficiency as a function of cycle number, for a coin cell comprising a membrane M6 or M2 characterized by an immersion time of 3.5 min and 4.5 min (and thus nanochannels with an average pore diameter of about 115 nm and 161 nm), for a fluence of 10.sup.8 cm.sup.2. Coulombic efficiencies obtained for both batteries are improved upon cycling compared that of the battery of reference B4 shown in FIG. 1.

[0178] FIG. 7a compares the results of specific capacity (mAh/g) (charge and discharge) as a function of cycle number, for a coin cell comprising a membrane M1, M8 or M9 characterized by an immersion time of 3.5 min, 4.5 min and 5.5 min (and thus nanochannels with an average pore diameter of about 115 nm, 161 nm and 198 nm), for a fluence of 10.sup.9 cm.sup.2.

[0179] FIG. 7b compares the results of coulombic efficiency as a function of cycle number, for a coin cell comprising a membrane M1, M8 or M9 characterized by an immersion time of 3.5 min, 4.5 min and 5.5 min (and thus nanochannels with an average pore diameter of about 115 nm, 161 nm and 198 nm), for a fluence of 10.sup.9 cm.sup.2.

[0180] FIG. 8a compares the results of specific capacity (mAh/g) (charge and discharge) as a function of cycle number, for a coin cell comprising a membrane M1, M5 or M6 characterized by an immersion time of 3.5 min (and thus nanochannels with an average pore diameter of about 115 nm), for a fluence of 10.sup.7, 10.sup.8, and 10.sup.9 cm.sup.2.

[0181] FIG. 8b compares the results of coulombic efficiency as a function of cycle number, for a coin cell comprising a membrane M1, M5 or M6 characterized by an immersion time of 3.5 min (and thus nanochannels with an average pore diameter of about 115 nm), for a fluence of 10.sup.7, 10.sup.8, and 10.sup.9 cm.sup.2.

[0182] FIG. 9a compares the results of specific capacity (mAh/g) (charge and discharge) as a function of cycle number, for a coin cell comprising a membrane

[0183] M2, M7 or M8 characterized by an immersion time of 4.5 min (and thus nanochannels with an average pore diameter of about 161 nm), for a fluence of 10.sup.7, 10.sup.8, and 10.sup.9 cm.sup.2.

[0184] FIG. 9b compares the results of coulombic efficiency as a function of cycle number, for a coin cell comprising a membrane M2, M7 or M8 characterized by an immersion time of 4.5 min (and thus nanochannels with an average pore diameter of about 161 nm), for a fluence of 10.sup.7, 10.sup.8, and 10.sup.9 cm.sup.2.

[0185] From FIGS. 7a and 7b, 8a and 8b and 9a and 9b, the highest coulombic efficiency is that of B1 which comprises a membrane having an average pore diameter of about 115 nm with a fluence of 10.sup.9 cm.sup.2.

Example 4: Cycling Tests of Batteries Using Polymer Etched Ion-Track Membranes of the Present Invention with Conical Pores

[0186] 4.1. Preparation of the Membranes and of the Coin Cells

[0187] Membrane M10 was prepared based on the method described by Apel et al. (Nanotechnology 18 (2007) 305302). Briefly, a PET foil (thickness of 12 m) was irradiated as described in Example 1 and exposed to UV light of 312 nm wavelength on both sides for 3 hours and subsequently clamped into a two chamber system maintained at 60 C. The two chambers are then filled simultaneously with 60 C. 6M NaOH solution containing a surfactant on one side, and 60 C. 6M NaOH without surfactant on the other. The chambers are then emptied after 3 minutes and washed with water. In the present example 0.05% (w/w) of a concentrated (about 45%) aqueous solution of sodium dodecyl diphenyloxide disulfonate was used as surfactant.

[0188] Membranes produced in this way comprises nanochannels with opening diameter of about 20 nm to 50 nm on one side and about 150 nm to 300 nm on the other side of the membrane. Membrane M10 had an opening tip diameter of about 20 nm on one side and an opening tip diameter of about 200 nm on the other side. Since especially very small opening tip diameters are difficult to determine precisely, these opening tip diameters might vary by +/10-15%.

[0189] Other methods to prepare etched ion-track membranes with conical pores or pores that have an opening tip diameter that is bigger on one side than on the other are known in the art and can equally be used.

[0190] Coin cells were prepared as described in example 2.

[0191] 4.2. Battery Cycling Tests

[0192] The cell comprises 20 l of electrolyte per mg of active material, and thus 10 l of electrolyte per cell. The batteries were tested

[0193] FIG. 10a gives the results of specific capacity (mAh/g) (charge and discharge) and coulombic efficiency as a function of cycle number, for a coin cell comprising conical pores.

[0194] FIG. 10b gives the results of coulombic efficiency as a function of cycle number, for a coin cell comprising conical pores in different configurations.

[0195] From FIG. 10a and FIG. 10b capacity and capacity retention of the cell with the membrane is comparable with the reference electrode whereas a very good coulombic efficiency close to 100% is measured with our membrane showing suppression of polysulfides redox shuttle.