NETWORK OF METAL FIBERS, METHOD FOR PRODUCING A NETWORK OF METAL FIBERS, ELECTRODE AND BATTERY

20210265636 · 2021-08-26

    Inventors

    Cpc classification

    International classification

    Abstract

    The present invention relates to a network of metal fibers, comprising a plurality of metal fibers fixed to one another; wherein at least some of the plurality of metal fibers have a length of 1.0 mm or more, a width of 100 μm or less and a thickness of 50 μm or less. The invention further relates to a method comprising step 1 of producing a plurality of metal fibers (2) by melt spinning; step 2 of providing a loose network of metal fibers (2) produced in step 1; and step 3 of fixating the plurality of metal fibers to one another by one of the following processes c1 to c4.

    Claims

    1. Network of metal fibers, comprising: a plurality of metal fibers fixed to one another; wherein the plurality of metal fibers have a length of 1.0 mm or more, a width of 100 or less and a thickness of 50 μm or less, wherein the metal fibers before and/or after fixing them to one another show an exothermic event when heated in a DSC measurement, and wherein the exothermic event releases energy in an amount of 0.1 kJ/g or more.

    2. The network according to claim 1, wherein the exothermic event releases energy in an amount of 0.5 kJ/g or more.

    3. The network according to claim 2, wherein the metal fibers are in electrical contact with one another.

    4. The network according to claim 1, wherein at least some of the metal fibers of the plurality of metal fibers are amorphous.

    5. The network according to claim 1, wherein at least some of the metal fibers of the plurality of metal fibers are sintered to one another.

    6. The network according to claim 1, wherein the metal fibers are obtainable by subjecting molten material of the metal fibers to a cooling rate of 102 K.Math.min-1 or higher by melt spinning.

    7. The network according to claim 1, wherein each of the metal fibers is in contact with one or more of the other metal fibers.

    8. The network according to claim 1, wherein the network is an unordered or ordered network.

    9. The network according to claim 1, wherein the network has open pores between the metal fibers of the plurality of metal fibers.

    10. The network according to claim 1, wherein at least some of the metal fibers of the plurality of metal fibers are at least partially coated.

    11. The network according to claim 1, wherein points of contact between metal fibers are distributed in an unordered or ordered manner throughout the three-dimensional structure of the network.

    12. Method for producing a network of metal fibers, the method comprising the steps of: producing a plurality of metal fibers having a length of 1.0 mm or more, a width of 100 μm or less and a thickness of 50 μm or less by melt spinning; providing a loose network of metal fibers; and fixating the plurality of metal fibers to one another by one of the following processes c1 to c4: c1: placing the plurality of metal fibers in a hot press and subjecting the plurality of metal fibers present in the hot press to a predetermined pressure and temperature for a predetermined period of time to produce the network by sintering the plurality of metal fibers one to another forming points of contact between the metal fibers, wherein in process c1 the pressure is between 0 and 20 GPa, and the temperature is between 10 to 80% of the melting temperature of the material of the metal fibers, wherein the melting temperature is determined by DSC measurement; c2: placing the loose network of metal fibers between two heating plates and adjusting the distance between the two heating plates to 0.1-1 mm and heating the heating plates to a temperature of 10 to 80% of the melting temperature of the material of the metal fibers wherein the melting temperature is determined by DSC measurement; c3: ultrasound welding; c4: hammering.

    13. The method according to claim 12, wherein process c3 or c4 is used for fixating the metal fibers to one another either over the complete surface area of the network or at a plurality of separated areas distributed over the surface area of the network.

    14. The method according to claim 12, wherein the metal fibers have a length of 1 to 20 cm.

    15. The method according to claim 12, wherein the method further comprises a step 4 of coating the metal fibers.

    16. Network of metal fibers, wherein the network of metal fibers comprises: a plurality of metal fibers fixed to one another; and wherein the network of metal fibers is obtained by a process comprising the following steps: step 1 of producing the plurality of metal fibers having a length of 1.0 mm or more, a width of 100 μm or less and a thickness of 50 μm or less by subjecting molten material from which the metal fibers are to be made to a cooling rate of 102 K.Math.min−1 or higher; step 2 of arranging the metal fibers obtained in step (1) to form a network of loose metal fibers (2); step 3 of sintering the metal fibers to one another by one of the following processes c1 to c4: c1: placing the plurality of metal fibers in a hot press and subjecting the plurality of metal fibers present in the hot press to a predetermined pressure and temperature for a predetermined period of time to pro-duce the network by sintering the plurality of metal fibers one to another forming points of contact between the metal fibers, wherein in process c1 the pressure is between 0 and 20 GPa, and the temperature is between 10 to 80% of the melting temperature of the material of the metal fibers, wherein the melting temperature is determined by DSC measurement; c2: placing the loose network of metal fibers between two heating plates and adjusting the distance between the two heating plates to 0.2-1 mm and heating the heating plates to a temperature of 10 to 80% of the melting temperature of the material of the metal fibers, wherein the melting temperature is determined by DSC measurement; c3: ultrasound welding; c4: hammering.

    17. The network of metal fibers according to claim 16, wherein in step 2 the metal fibers are arranged using carding, depositing or sedimentation from a liquid dispersion or from an air flow or by spraying.

    18. Electrode containing a network, the network comprising a plurality of metal fibers fixed to one another; wherein the plurality of metal fibers have a length of 1.0 mm or more, a width of 100 μm or less and a thickness of 50 μm or less, wherein the metal fibers before and/or after fixing them to one another show an exothermic event when heated in a DSC measurement, and wherein the exothermic event releases energy in an amount of 0.1 kJ/g or more.

    19. Battery comprising an electrode, the electrode containing a network, the network comprising a plurality of metal fibers fixed to one another; wherein the plurality of metal fibers have a length of 1.0 mm or more, a width of 100 μm or less and a thickness of 50 μm or less, wherein the metal fibers before and/or after fixing them to one another show an exothermic event when heated in a DSC measurement, and wherein the exothermic event releases energy in an amount of 0.1 kJ/g or more.

    20. The network according to claim 1, wherein at least some of the metal fibers of the plurality of metal fibers are nanocrystalline.

    21. The method according to claim 12, wherein step 4 is carried out after step 3.

    22. The electrode according to claim 18, wherein the network is present as a current collector.

    Description

    [0081] The invention will now be described in further detail and by way of example only with reference to the accompanying drawings and figures as well as by various examples of the network and method of the invention. In the drawings there are shown:

    [0082] FIG. 1a) Schematic drawing of a vertical melt spinning device.

    [0083] FIG. 1b) Schematic drawing of a horizontal melt spinning device.

    [0084] FIG. 1c) Photographic image taken from a movie of the deposition of a melt of an Al.sub.99Si.sub.1-alloy onto a rotating wheel.

    [0085] FIG. 2a) Photographic image of copper metal fibers.

    [0086] FIG. 2b) Photographic image of cobalt metal fibers.

    [0087] FIG. 2c) Photographic image of metal fibers of an Al.sub.99Si.sub.1-alloy.

    [0088] FIG. 2d) Photographic image of metal fibers of a Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy.

    [0089] FIG. 3a) i) Photographic image of Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers fabricated at a wheel-speed of 50 m/s with ii) showing a mean thickness and iii) showing a mean width of the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers.

    [0090] FIG. 3b) i) Photographic image of Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers fabricated at a wheel-speed of 25 m/s with ii) showing a mean thickness and iii) showing a mean width of the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers.

    [0091] FIG. 4a) X-ray diffraction pattern of copper metal fibers.

    [0092] FIG. 4b) X-ray diffraction pattern of Al.sub.99Si.sub.1-alloy metal fibers.

    [0093] FIG. 5a) Scanning electron micrograph (SEM) of copper metal fibers.

    [0094] FIG. 5b) Enlarged view of FIG. 5a).

    [0095] FIG. 5c) Scanning electron micrograph of Al.sub.99Si.sub.1-alloy metal fibers.

    [0096] FIG. 5d) Enlarged view of FIG. 5c).

    [0097] FIG. 5e) Further scanning electron micrograph of Al.sub.99Si.sub.1-alloy metal fibers.

    [0098] FIG. 5f) Scanning electron micrograph of gold metal fibers

    [0099] FIG. 5g) Scanning electron micrograph of Cu.sub.92Sn.sub.8-alloy metal fibers.

    [0100] FIG. 5h) Scanning electron micrograph of Cu.sub.96Si.sub.4-alloy metal fibers.

    [0101] FIG. 6a) Thickness and width of 527 Al.sub.99Si.sub.1-alloy metal fibers fabricated at a wheel-speed of 25 m/s.

    [0102] FIG. 6b) Distribution of the thickness of 527 Al.sub.99Si.sub.1-alloy metal fibers fabricated at a wheel-speed of 25 m/s.

    [0103] FIG. 6c) Distribution of the width of 527 Al.sub.99Si.sub.1-alloy metal fibers fabricated at a wheel-speed of 25 m/s.

    [0104] FIG. 6d) DSC measurements of CuSn.sub.8 microfibers after melt spinning (curve I) and after thermal equilibration of the same micro fibers (curve II) which demonstrates the exothermal process in case of heating melt spinned fibers in comparison to the equilibrated metal fibers.

    [0105] FIG. 7 SEM image of a network according to the invention of Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers after sintering (network of Example 2).

    [0106] FIG. 8 Photographic image of a network according to the invention of sintered Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers before deformation.

    [0107] FIG. 9 Photographic image of the network of FIG. 8 in a deformed state.

    [0108] FIG. 10 Photographic image of a network according to the invention of copper metal fibers sintered for 5 minutes at a temperature of 300° C. under a pressure of 177 MPa (network of Example 9).

    [0109] FIG. 11 Photographic image of a network according to the invention of copper metal fibers sintered for 3 minutes at a temperature of 300° C. under a pressure of 177 MPa (network of Example 10).

    [0110] FIG. 12 SEM image of the network shown in FIG. 11.

    [0111] FIG. 13 Enlarged SEM image of the network shown in FIGS. 11 and 12.

    [0112] FIG. 14a) Photographic image of a network according to the invention of Cu.sub.92Sn.sub.8-alloy metal fibers sintered for 3 minutes at a temperature of 300° C. under a pressure of 78 MPa (network of Example 11).

    [0113] FIG. 14b) SEM image of the network of FIG. 14a).

    [0114] FIG. 14c) Enlarged SEM image of the network shown in FIG. 14b).

    [0115] FIG. 15 Schematic illustration of a hot press.

    [0116] FIG. 16a) Schematic drawing of a battery according to the invention.

    [0117] FIG. 16b) Schematic drawing of a reference battery according to prior art.

    [0118] FIG. 17 Comparison of the capacity of a battery according to the invention with a reference battery according to prior art.

    [0119] FIG. 18 Plot of charge discharge profile of a reference cell obtained by chronopotentiometry.

    [0120] FIG. 19 Plot of charge discharge profile of an inventive network cell obtained by chronopotentiometry.

    [0121] FIG. 20a) Plot of the second and last charge discharge profile of the reference cell obtained by chronopotentiometry as shown in FIG. 18.

    [0122] FIG. 20b) Plot of the second and last charge discharge profile of the of the inventive network cell obtained by chronopotentiometry as shown in FIG. 19.

    [0123] FIG. 21a) Plot of charge discharge profile of an inventive network cell and of a reference cell obtained by chronopotentiometry with current normalized time.

    [0124] FIG. 21b) Enlarged view of the plot of FIG. 21a showing the first cycles.

    [0125] FIG. 21c) Enlarged view of the plot of FIG. 21a showing the area around the last cycles of the reference cell.

    [0126] FIG. 22 Plot of the capacity over the number of cycles for an inventive network cell and a reference electrode.

    [0127] FIG. 23a) SEM image of an inventive electrode showing graphite flakes within the fiber network.

    [0128] FIG. 23b) EDX mapping of copper for the inventive electrode shown in FIG. 23a).

    [0129] FIG. 23c) EDX mapping of carbon for the inventive electrode shown in FIG. 23a).

    [0130] FIG. 24 Scheme of an exemplified carding machine for the formation of metallic fleece.

    [0131] FIG. 25 Photographic images of a manufacturing process according to the present invention in which metallic fibers are formed to fleece and combined with cotton fleece.

    [0132] FIG. 26 Copper based fleece, mechanically stabilized by ultrasonic treatment.

    [0133] FIG. 27 Schematic representation of a process according to the invention.

    [0134] For reasons of clarity, in some figures not all or no reference signs are shown.

    [0135] A schematic drawing of a melt spinning device 1 which can be used for producing metal fibers 2 suitable for forming a network according to the invention is shown in FIG. 1a). The melt spinning device 1 has a vertically rotating wheel 3 which can spin around an axis Z. Above the rotating wheel 3 a microstructured nozzle 4 is arranged through which drops 5 of a melt of the material from which the metal fibers 2 are produced can be deposited onto the rotating wheel 3. Alternatively, a horizontal melt spinner can be used (FIG. 1b; a horizontal melt spinner is disclosed in European patent application with application number EP19175749.1, the content of which is herewith incorporated by reference).

    [0136] FIG. 1b) shows a schematic drawing of a horizontal melt spinning device 1 which can also be used for forming metal fibers 2. In contrast to the melt spinning device 1 shown in FIG. 1a), the melt spinning device 1 shown in FIG. 1b) has a horizontally rotating wheel 3. Similar as in the melt spinning device 1 shown in FIG. 1a), in the melt spinning of FIG. 1b) the wheel 3 can spin around an axis Z. Above the rotating wheel 3 a microstructured nozzle 4 is arranged through which drops 5 of a melt of the material from which the metal fibers 2 are produced can be deposited onto the rotating wheel 3.

    [0137] FIG. 1c) shows a photographic image of the production of Al.sub.99Si.sub.1-alloy metal fibers. The photographic image shows the deposition of a melt of Al.sub.99Si.sub.1-alloy onto a rotating wheel 3 and are isolated from a movie taken at 40000 fps of depositing metal melts onto a rotating wheel 3 of a melt spinning device 1. The rotating wheel 3 is in the lower part and the microstructured nozzle 4 is in the upper part of the photographic image shown in FIG. 1c). The temperature of the melt was set in the range of 50 to 300° C. above the melting temperature (higher processing temperatures are possible). The melt forms a flow of metal which wets the rotating wheel 3 and is rapidly cooled to form metal fibers 2 from the metal flow 5.

    [0138] In FIG. 2a) to d) photographic images of copper metal fibers (FIG. 2a)), of cobalt metal fibers (FIG. 2b)), of Al.sub.99Si.sub.1-alloy metal fibers (FIG. 2c)) and of Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers (FIG. 2d)) are shown after the metal fibers 2 were produced using the melt spinning device 1 as shown in FIG. 1a) and b) and collected. The metal fibers 2 form an entangled network of metal fibers 2 in which the metal fibers 2 are not fixed to one another so that single metal fibers 2 can easily be separated from the entangled network of metal fibers 2.

    [0139] Further photographic images for Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers 2 are shown in FIG. 3a) and b). The photographic images were taken from Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers 2 produced with a wheel speed of 50 m/s and 25 m/s, respectively. From the corresponding thickness and width distribution also shown in FIG. 3a) and b) it can be seen that the higher wheel speed of 50 m/s (see FIG. 3a)) results in metal fibers 2 having a reduced thickness and width compared to metal fihers 2 produced with a wheel speed of 25 m/s (see FIG. 3b)).

    [0140] The x-ray diffraction spectra shown in FIG. 4a) to c) were recorded using a Bruker D8 advance XRD in Bragg-Bretano mode with a cobalt-source with 30 mA and 40 kV anode current and acceleration voltage. The data were collected with a Bruker VANTEC-1 detector and the measurements were done in air. The sharp peaks 20 observable for copper metal fibers 2 (see FIG. 4a)) and for Al.sub.99Si.sub.1-alloy metal fibers 2 (see FIG. 4b)) at specific angles indicate that these metal fibers 2 are polycrystalline, whereas the peak 22 extending over a comparatively large angle range, i.e. the absence of such sharp peaks 20, for the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers 2 (see FIG. 4c)) indicates that these metal fibers 2 are amorphous.

    [0141] Scanning electron micrographs of copper metal fibers 2 are shown in FIG. 5a) and b). The micrographs were recorded on a Zeiss Ultra 55 with an acceleration voltage of 3 kV. The scale bar in the lower left corner of FIG. 5a) indicates a length of 100 μm and the scale bar in the lower left corner of FIG. 5b) indicates a length of 2 μm. As can be seen from the micrograph of FIG. 5a), the metal fibers 2 are not fixed to one another and form an entangled network, in which the metal fibers 2 can be moved relative to one another so that single metal fibers 2 can easily be separated from the entangled network. It can also be recognized from the micrograph of FIG. 5a) that the metal fibers 2 have a substantially constant width over a length of millimeters. In fact, although not visible from FIG. 5a) and b), the width and thickness of the copper metal fibers is substantially constant also over a length of several centimeters. Further scanning electron micrographs shown in FIG. 5c) and d) were recorded in a similar manner as the scanning electron micrographs shown in FIG. 5a) and b) on Al.sub.99Si.sub.1-alloy metal fibers 2. The scale bar in the lower left corner of FIG. 5c) indicates a length of 100 μm and the scale bar in the lower left corner of FIG. 5d) indicates a length of 10 μm. It can be recognized that the Al.sub.99Si.sub.1-alloy metal fibers 2 form an entangled network with the metal fibers 2 not being fixed to one another. The thickness and width of these metal fibers 2 is also substantially constant over a length of millimeters. In fact, although not visible from FIG. 5c) and d), the width and thickness of the Al.sub.99Si.sub.1-alloy metal fibers 2 is substantially constant also over a length of several centimeters. FIG. 5e) shows a further scanning electron micrograph of an Al.sub.99Si.sub.1-alloy metal fiber 2. The scale bar in the lower left corner of FIG. 5e) indicates a length of 3 μm. In the micrograph shown in FIG. 5e) nanocrystalline domains can be recognized as grain structure. A scanning electron micrograph of a gold metal fiber 2 is shown in FIG. 5f). The scale bar in the lower left corner of FIG. 5f) indicates a length of 10 μm. In the micrograph shown in FIG. 5f) nanocrystalline domains can be recognized as grain structure. In FIG. 5g) a scanning electron micrograph of an Cu.sub.92Sn.sub.8-alloy metal fiber 2 is shown. The scale bar in the lower left corner of FIG. 5g) indicates a length of 1 μm. Also in the micrograph shown in FIG. 5g) nanocrystalline domains can be recognized. Nanocrystalline domains can also be found in Cu.sub.96Si.sub.4-alloy metal fibers 2 produced by melt spinning, as can be recognized from the grain structure visible in the scanning electron micrograph of such a fiber 2 which is shown FIG. 5h. The scale bar in the lower left corner of FIG. 5h) indicates a length of 1 μm.

    [0142] Thicknesses and widths for 527 Al.sub.99Si.sub.1-alloy metal fibers 2 fabricated with a wheel-speed of 25 m/s are shown in the graph of FIG. 6a). The corresponding thickness and width distributions are shown in the graphs of FIGS. 6b) and 6c), respectively. It can be recognized from FIG. 6a) and b) that the Al.sub.99Si.sub.1-alloy metal fibers have a thickness in the range between 3 and 17 μm with a mean thickness of 8.5 μm. The thickness distribution follows a narrow Gaussian function, such as indicated by the line in FIG. 6b). The width of the metal fibers is in a range of 5 to 80 μm, wherein the mean width is 39.5 μm and the median width is 35.0 μm. FIG. 6d shows two DSC measurements. Curve I in FIG. 6d) was obtained from CuSn.sub.8 microfibers directly after melt spinning and curve II in FIG. 6d) after thermal equilibration of the same microfibers. This demonstrates the exothermal process in case of heating melt spinned fibers in comparison to the equilibrated metal fibers.

    [0143] A scanning electron micrograph of a network 6 of metal fibers 2 according to the invention is shown in FIG. 7. The scale bar in the lower left corner of FIG. 7 indicates a length of 20 μm. The metal fibers 2 in the network 6 of FIG. 7 are amorphous metal fibers 2 of the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy. In contrast to the entangled networks of metal fibers 2 shown for example in the scanning electron micrographs of FIG. 5a) or c), in the network 6 of metal fibers 2 as shown in FIG. 7 the metal fibers 2 are fixed to one another at points of contact 7, where the metal fibers 2 are sintered with one another. As a result of the metal fibers being sintered to one another, the metal fibers 2 are fixed to one another, so that it is not possible to move these metal fibers 2 relative to one another and to separate one of the metal fibers 2 out of the network 6 without breaking the points of contact 7.

    [0144] It can be seen from FIGS. 7 to 9 that the networks 6 of metal fibers 2 according to the invention have voids 9 in the form of pores between the metal fibers 2. For a better overview, only some of the metal fibers 2, points of contact 7 and voids 9 are indicated by reference signs in FIG. 7. Reference signs for the voids 9 and also the points of contact 7 were omitted in FIGS. 8 and 9 and also only some of the metal fibers 2 are indicated by reference signs.

    [0145] In FIGS. 8 and 9 photographic images of another network 6 of amorphous metal fibers 2 of the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy are shown. It can be seen from these photographic images, that the network 6 can be held with tweezers 8 without separating metal fibers 2 from the network 6 of metal fibers 2 and that the network 6 is an unordered network 6, i.e. the metal fibers 2 do not have a preferred orientation but are oriented randomly.

    [0146] FIGS. 8 and 9 show photographic images of the same network 6 of metal fibers 2. In FIG. 8 the network 6 as obtained after production is held with tweezers 8. It can be recognized from FIG. 9 that the network 6 can be bent and that the metal fibers 2 are still fixed to one another and are not separated from the network 6 of metal fibers 2.

    [0147] Moreover, it can be seen from FIGS. 8 and 9 that the network 6 of metal fibers 2 has a porous structure, with interconnected pores extending throughout the network of metal fibers. The points of contact 7 between the metal fibers 2 are distributed randomly throughout the network 6 of metal fibers 2.

    [0148] Photographic images of networks 6 of metal fibers 2 of copper are shown in FIGS. 10 and 11. The production of the network 6 shown in FIG. 10 is described in the following as Example 9 and the production of the network 6 shown in FIG. 11 is described in the following as Example 10. Accordingly, the network 6 shown in FIG. 10 has been produced at a higher temperature compared to the network 6 shown in FIG. 11. While both networks 6 show voids 9 in the form of pores distributed throughout the complete network 6, in the network 6 shown in FIG. 10, the density of the pores, i.e. number of pores per surface area, is lower at the center of the network 6 and increases towards the edges of the network 6. For a better overview, reference signs for the metal fibers 2, points of contact 7 and voids 9 were omitted in FIGS. 10 and 11.

    [0149] In the network 6 shown in FIG. 11 the distribution of voids 9 in the form of pores is more evenly throughout the network 6 of metal fibers 2 compared to the distribution of the voids 9 in the network 6 shown in FIG. 10. Probably due to the higher processing temperatures used for producing the network 6 shown in FIG. 10, fractions of the metal fibers 6 were transformed into a metal foil. This was almost completely avoided by reducing the processing temperature, as found for the network 6 shown in FIG. 11.

    [0150] Scanning electron micrographs of the network shown in FIG. 11 are shown in FIGS. 12 and 13. Recording of the scanning electron micrographs was conducted similarly to the recording described above for the other scanning electron micrographs. The scale bar in the lower left corner of FIG. 12 indicates a length of 200 μm and the scale bar in the lower left corner of FIG. 13 indicates a length of 100 μm. It can be recognized from these scanning electron micrographs, that the structure of the copper metal fibers 2 was conserved, while the metal fibers 2 were sintered together at points of contact 7, so that they no longer form only an entangled network but are fixed to one another, so that it is no longer easily possible to separate single metal fibers 2 from the network 6 of metal fibers 2. It can also be recognized that voids 9 in the form of pores extend through the network 6 of metal fibers 2 and that the points of contact 7 are randomly distributed throughout the network 6. The stripes visible in the scanning electron micrographs of FIGS. 12 and 13 result from discs of thermax alloy used for producing the network 6 of metal fibers 2, as described below for the Examples. These discs of the thermax alloy were cut and as a result have very fine grooves (not shown) on their surfaces which were imprinted onto the network of metal fibers during production using a hot press 10 as schematically shown in FIG. 15.

    [0151] A further photographic image of a network 6 of metal fibers 2 according to the invention is shown in FIG. 14a). In this network 6 of metal fibers 2 the metal fibers are made of a Cu.sub.92Sn.sub.8-alloy. The production of the network 6 shown in FIG. 14a) is described in the following as Example 11. In the networks 6 shown in FIG. 14a) voids 9 in the form of pores are distributed throughout the complete network 6. The network 6 is an unordered network 6, i.e. the metal fibers 2 do not have a preferred orientation but are oriented randomly, with the metal fibers 2 being fixed to one another at points of contact 7, where the metal fibers 2 are sintered to one another. For a better overview, reference signs for points of contact 7 and voids 9 were omitted in FIG. 14a) and only some of the metal fibers 2 are indicated by reference signs in FIG. 14a). The metal fibers 2, points of contact 7 and voids 9 can be recognized from the SEM images of FIGS. 14b) and 14c) which were taken from the network 6 shown in FIG. 14a). The scale bar in the lower left corner of FIG. 14b) indicates a length of 100 μm and the scale bar in the lower left corner of FIG. 14c) indicates a length of 20 μm. The enlarged view provided in FIG. 14c) demonstrates that at a point of contact 7 more than two metal fibers 2 can be sintered together, so that at a single point of contact 7 a plurality of metal fibers 2 can be fixed to one another due to sintering. For a better overview, only some of the metal fibers 2, points of contact 7 and voids 9 are indicated by reference signs in FIGS. 14b) and 14c).

    [0152] A schematic illustration of a hot press 10 is shown in FIG. 15 which can be used for producing the network 6 of metal fibers 2 according to the invention. The hot press 10 is provided with upper and lower parts 11 which can exert a force onto discs 12 between which metal fibers 2 are placed. In the hot press the temperature at the site where the metal fibers 2 are located can be controlled. It is also possible to omit the discs 12 and to place the metal fibers 2 directly between the upper and lower parts 11. For producing the network 6 of metal fibers 2 from which the scanning electron micrographs are shown in FIGS. 12 and 13, the discs 12 were made of a thermax alloy having fine grooves (not shown) on their surfaces. These fine grooves have a width in the range of 30 to 60 μm.

    [0153] Schematic drawings of a half-battery 13a according to the invention and of a battery 13b according to the prior art are shown in FIGS. 16a) and 16b), respectively. In both half-batteries 13a, 13b, a current collector 14 is provided as a first electrode. The current collector 14 is coated with an active electrode material 15. Lithium 16 is provided as electrolyte. In the half-batteries 13a and 13b an electrolyte is provided which impregnates all components of the batteries 13a and 13b and transports lithium ions. In the half-battery 13a schematically shown in FIG. 16a) the current collector 14 is a network 6 of metal fibers 2 according to the invention, whereas in the battery 13b schematically shown in FIG. 16b) the current collector 14 is a copper foil. The structure and composition of the batteries 13a and 13b are described in more detail below in the context of describing the Examples.

    [0154] The results of capacity measurement of the half-batteries 13a and 13b are shown in FIG. 17. The battery according to the invention comprising the network of metal fibers has a capacity which is increased by around 50% compared to a reference battery comprising a copper foil instead of a network of metal fibers while keeping the composition and amount of active material constant for both half-batteries 13a and 13b.

    [0155] A further electrode was prepared using a network of metal fibers consisting of the copper alloy Cu.sub.96Si.sub.4. The network was infiltrated with a dispersion of 90% graphite and 10% binder, as further described below. As reference electrode, a copper foil was coated with a 50 μm layer of the same dispersion using a doctor blade. FIGS. 18 and 19 show plots of discharge profile obtained by chronopotentiometry for the inventive electrode having the network of metal fibers and of the reference cell.

    [0156] In FIG. 20a) the second and last charge discharge profile of the plot of FIG. 18 of the reference cell obtained by chronopotentiometry is shown and in FIG. 20b) the second and last charge discharge profile of the inventive network cell obtained by chronopotentiometry is shown. To better demonstrate the changes in the charge discharge profiled during cyclization FIG. 21a) shows a plot of charge discharge profile of an inventive network cell and of a reference cell obtained by chronopotentiometry with current normalized time. Enlarged views are provided in FIGS. 21b) and 21c). The development of the capacity over the number of cycles for the inventive electrode and for the reference electrode is shown in FIG. 22.

    [0157] In FIG. 23a) a SEM image of the inventive electrode having a network of metal fibers consisting of the copper alloy Cu.sub.96Si.sub.4 is provided. It can be seen from the SEM image that graphite flakes are within the metal fiber network, i.e. between the metal fibers. FIG. 23b) shows an EDX mapping of copper for the inventive electrode shown in FIG. 23a) and FIG. 23c) shows an EDX mapping of carbon for the inventive electrode shown in FIG. 23a). The EDX mapping demonstrates that the graphite is between the metal fibers.

    [0158] The following experiments were conducted:

    [0159] Production of metal fibers:

    [0160] A melt spinning device using a range of experimental parameters was used to form the metal fibers. On the one hand, the device was composed of a massive wheel 3 (copper-alloy) of diameter 200 mm placed in a chamber with Argon atmosphere at a pressure of 300 mbar (all typical experimental settings). On the other hand the wheel rotation-speed was increased up to 60 m/s. A pressure difference of up to 2000 mbar (or lower) between the crucible with nozzle and the surrounding chamber atmosphere triggered the ejection of the molten metal or metal alloy onto the rotating wheel-surface. As a result microdroplets of melt were formed and were shaped as metal fibers in the shape of microribbons by rapid quenching (see FIG. 1a) and b)) for micro droplets of different metal deposited on a spinning wheel). Details for the melt spinning device used are disclosed in EP19175749.1, WO2016/020493 A1 and in WO2017/042155 A1.

    [0161] Each single microdroplet 5 was transformed into a single metal fiber 2 or multiple metal fibers 2. The deposition rate of the molten alloy onto the rotating wheel 3 was reduced to be in the range of 1.0 to 10.0 mg.Math.s.sup.−1 or even lower. With such deposition rate, metal fibers 2 in the form of microribbons were formed in large quantities of up to 90-95% of the initial mass of molten alloy in the crucible. Optical images of the produced metal fibers 2 are shown in FIG. 2a) to d) after rapid quenching and collection, the metal fibers 2 form an entangled network, wherein the metal fibers slide easily against each other so that it is easy to isolate one single metal fiber from this network.

    [0162] A typical initial mass of melt was in the range of 5 to 12 g (but can be increase to 100 g). The distance between the nozzle 5 and the wheel surface was set in the range of 50 to 3000 μm, see FIG. 1b) which shows a photographic image isolated from a movie taken at 40000 fps of depositing a Al.sub.99Si.sub.1-alloy melt onto the rotating wheel. The temperature of the melt was set in the range of 50 to 300° C. above the melting temperature (higher processing temperatures are possible). The photographic images shown in FIG. 2a) to d) and in FIG. 3a) and b) were taken from correspondingly produced metal fibers.

    [0163] Structure of the metal fibers:

    [0164] Metal fibers 2 in the form of microribbons were produced of Co, Cu, Al as well as alloys of these elements with other elements such as Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16, Al.sub.99Si.sub.1 (none comprehensive list). The metal fibers 2 made with pure Cu or the alloy Al.sub.99Si.sub.1 had a polycrystalline structure of crystals with a maximum size up to 8 μm, as demonstrated by the X-ray diffraction spectra for the copper metal fibers 2 (FIG. 4a)) and for the Al.sub.99Si.sub.1-alloy metal fibers 2 (FIG. 4b)). The metal fibers 2 made with the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy had the typical structure of glassy metal, i.e. these metal fibers 2 were amorphous metal fibers 2, as can be seen from the X-ray diffraction spectrum of the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers 2 depicted in FIG. 4c).

    [0165] When using more complex Cu-alloys and Al-alloys, as well as adjusting the experimental parameters such as wheel speed and melt temperature, it is likely that metal fibers 2 mainly composed of Al or Cu can be fabricated with a nanocrystalline or glassy metal structure (as observed for the Co-alloy Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16).

    [0166] Dimensions of metal fibers:

    [0167] For the Co-alloy Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16, the metal fibers 2 had the following typical dimensions: a width of 2.0 to 25.0 μm, a thickness of 1.0 to 7.0 μm and a length of 2.0 to 100.0 mm (see FIGS. 3a) and 3b)). The ribbon-thickness distributions were narrow Gaussian distributions with a standard deviation as small as 0.4 μm, i.e. 68% (resp. 95%) of the metal fibers 2 had a thickness within an interval (centered on the mean thickness) as narrow as 0.8 μm (resp. 1.6 μm). With a wheel-speed of 25 m/s, the ribbon mean thickness is 5.80 μm (see FIG. 3b)). When the wheel speed is doubled, it decreases to 3.22 μm (see FIG. 3a)). The fiber-width distributions were either Gaussian or log-normal distributions. With a wheel-speed of 25 m/s, the fiber mean width is 14.2 μm and the median width is 13.2 μm, i.e. 50% of the fiber have a width below 13.2 μm. When the wheel speed is doubled, the mean and median widths decrease to 9.4 μm, i.e. 50% of the fibers have a width below 9.4 μm.

    [0168] For the Al-alloy Al.sub.99Si.sub.1, the mean and median thicknesses are 8.5±0.1 μm, i.e. 68% (resp. 95%) for the fibers and the fibers 2 had a thickness between 6.6 and 10.4 (resp. 4.8 and 12.2) μm. The mean width was 39.5±1.0 μm and the median width was 35.0±1.0 μm, i.e. 50% of the fibers had a width below 35.0 μm (FIG. 6a) to c)). SEM images for the Al.sub.99Si.sub.1-alloy metal fibers 2 are shown in FIGS. 5c) and 5d), respectively.

    [0169] The dimensions of the copper metal fibers 2 were similar to the ones of the Al-alloy metal fibers (or smaller). SEM images for the copper metal fibers 2 are shown in FIGS. 5a) and 5b). These SEM images demonstrate that the width of the metal fibers 2 remains constant over a length of at least 1 mm (this is even true for length over 1 cm). These ribbons have a crystalline structure: the observation of the top side, which was a liquid-gas interface before solidification upon rapid quenching, allows a direct visualization of the grains. The maximum size of the crystals is estimated to be in the range of 5-8 μm.

    [0170] The metal fibers 2 were used to produce networks 6 of metal fibers 2.

    [0171] Production of networks of metal fibers:

    [0172] Before producing networks 6 of metal fibers 2, it was determined by using differential scanning calorimetry (DSC) that the Co-alloy Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 has a crystallization temperature of 560° C. and a melting temperature of 1021° C.

    EXAMPLE 1

    [0173] Amorphous metal fibers 2 of Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 produced by the melt spinning process described above were placed between two discs 12 of aluminum oxide with diameters of 45 mm. The staple of the discs 12 of aluminum oxide having a diameter of 45 mm and the metal fibers 2 was then placed into the 400° C. preheated hot press 10 and the fibers 2 were pressed at a predetermined pressure of 377 MPa for a predetermined time of 30 minutes, to avoid thermal expansion effects due to heating.

    EXAMPLE 2

    [0174] The network 6 of Example 2 was prepared identical to the network 6 of Example 1, except that the pressure was reduced to 277 MPa. A SEM image of the network 6 of Example 2 is provided in FIG. 7, as can be seen, the metal fibers 2 are sintered together and form a stable network 6. The SEM image shows that apart from the contact faces with the aluminum oxide disc 12, the texture and appearance of the metal fibers 2 was conserved.

    EXAMPLE 3

    [0175] The network 6 of Example 3 was prepared identical to the network 6 of Example 1, except that instead of discs 12 of aluminum oxide discs 12 of a Thermax high-temperature alloy having diameters of 60 mm were used. The pressure applied was 283 MPa.

    EXAMPLE 4

    [0176] The network 6 of Example 4 was prepared identical to the network 6 of Example 3, except that the time was reduced to 20 minutes.

    EXAMPLE 5

    [0177] The network 6 of Example 5 was prepared identical to the network 6 of Example 3, except that the time was reduced to 10 minutes.

    EXAMPLE 6

    [0178] The network 6 of Example 6 was prepared identical to the network 6 of Example 3, except that the time was reduced to 5 minutes.

    EXAMPLE 7

    [0179] The network 6 of Example 7 was prepared identical to the network 6 of Example 3, except that the temperature was reduced 300° C.

    COMPARATIVE EXAMPLE 1

    [0180] For Comparative Example 1 amorphous metal fibers 2 of Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 were placed in an oven and heated for 30 minutes to 600° C. without applying external pressure. The fibers crystallized but did not sinter together.

    COMPARATIVE EXAMPLE 2

    [0181] The network of Comparative Example 2 was prepared identical to the network 6 of Example 1, except that the pressure was reduced to 157 MPa.

    [0182] The network disintegrated when being removed from the hot press 10 which indicated that the metal fibers 2 were not sufficiently sintered together.

    COMPARATIVE EXAMPLE 3

    [0183] The network of Comparative Example 3 was prepared identical to the network 6 of Example 3, except that the pressure was reduced to 177 MPa.

    COMPARATIVE EXAMPLE 4

    [0184] The network of Comparative Example 4 was prepared identical to the network 6 of Example 4, except that the temperature was reduced to 100° C.

    COMPARATIVE EXAMPLE 5

    [0185] The network of Comparative Example 5 was prepared identical to the network 6 of Example 4, except that the temperature was reduced to 200° C.

    [0186] Table 1 summarizes the time, pressure and temperature used for the preparation of the networks of Examples 1 to 7 and Comparative Examples 1 to 5.

    TABLE-US-00001 TABLE 1 Exam- Exam- Exam- Exam- Exam- Exam- Exam- Comp. Comp. Comp. Comp. Comp. Parameter ple 1 ple 2 ple 3 ple 4 ple 5 ple 6 ple 7 Example 1 Example 2 Example 3 Example 4 Example 5 time (min) 30 30 30 20 10 5 20 30 30 30 20 20 pressure (MP) 377 220 283 283 283 283 283 ambient 157 177 283 283 temperature 400 400 400 400 400 400 300 600 400 400 100 200 (° C.) network properties: network ++ ++ ++ ++ ++ + ++ −− −− −− −− − stability amorphous ++ ++ ++ ++ ++ ++ ++ −− ++ ++ ++ ++ fibers

    [0187] In Comparative Example 2 time and temperature were identical to Examples 1 and 2. However, the network of Comparative Example 2 disintegrated when being removed from the press 10. This indicates that the pressure was not sufficient for sintering the amorphous Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy metal fibers 2 to provide a fixation of the metal fibers 2 to one another. It can be concluded that the pressure is the driving force for the sintering of the metal fibers 2. Since amorphous materials have a lower density than crystalline materials, the atoms at the interphase are set into motion when pressure is applied. This causes these atoms to make a transition to an energetically preferred state in view of the pressure applied. As a result of the movement of the atoms, the metal fibers 2 are sintered together permanently.

    [0188] The networks 6 of metal fibers of Examples 3 to 5 did not show substantial differences. When shortening the time to only 5 minutes in Example 6, the fibers 2 were not fixed to each other as strong as in Examples 3 to 5. This demonstrates that the process of fixing the metal fibers 2 of the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy to one another has a certain time dependency but is finished within several minutes.

    [0189] It can be seen that the process of sintering the metal fibers 2 together takes place although the temperature the metal fibers 2 are subjected to is only around 40% or 30%, respectively, of the melting temperature of the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16-alloy, such as demonstrated by Examples 4 and 7 in which the temperature was 400° C. and 300° C., respectively. Comparative Examples 4 and 5 show that if the temperature is reduced to 100° C. or 200° C., respectively, the movement of the atoms is too low to provide for a fixation of the metal fibers 2 to one another by sintering.

    EXAMPLE 8

    [0190] The network 6 of Example 8 was prepared identical to the network 6 of Example 3, however, the temperature was set to 500° C. and the time was set to 20 minutes. In addition, in the network 6 of Example 8 more fibers 2 were used to obtain a network 6 having a thickness of 0.7 mm. After 20 minutes at a temperature of 500° C. under a pressure of 283 MPa, the network 6 of Example 8 was completely sintered, i.e. the stability of network 6 was comparable to that of Example 3. In FIG. 8 an image of the network 6 of Example 8 is shown before deformation and in FIG. 9 an image of the network 6 of Example 8 in a deformed state is shown. Accordingly, it can be recognized from FIG. 9 that even when the network 6 is highly deformed, no metal fibers 2 are separated from the network 6. This indicates the formation of a highly stable network 6 that is not easily damaged by deformation.

    EXAMPLE 9

    [0191] The network 6 of Example 9 was prepared identical to the network 6 of Example 3, however, instead of amorphous metal fibers 2 of Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16, polycrystalline fibers 2 of copper (Cu) as described above were used, the time was set to 5 minutes, the pressure was set to 177 MPa and the temperature was set to 300° C.

    EXAMPLE 10

    [0192] The network 6 of Example 10 was prepared identical to the network 6 of Example 9, however, the time was set to 5 minutes.

    COMPARATIVE EXAMPLE 6

    [0193] The network 6 of Comparative Example 6 was prepared identical to the network of Example 9, however, the time was set to 30 minutes and the temperature was set to 500° C.

    [0194] Table 2 summarizes the time, pressure and temperature used for the preparation of the networks of Examples 9 and 10 and for Comparative Example 6.

    TABLE-US-00002 TABLE 2 Parameter Example 9 Example 10 Comp. Example 6 time (min) 5 3 30 pressure (MPa) 177 177 177 temperature (° C.) 300 300 500 network properties: network stability ++ ++ ++ porosity + ++ −−

    [0195] In Comparative Example 6 the metal fibers could not be recognized in the sintered product and the obtained product was a copper foil. If a light source is placed behind the copper foil, a certain inhomogeneity could be recognized. The networks 6 of Examples 9 and 10 have a thickness from 0.15 mm up to 0.25 mm.

    [0196] For producing the network 6 of Example 9, time and temperature were reduced to 5 minutes and 300° C. while the pressure was identical to the one applied in Comparative Example 6. It was possible to recognize in the resulting product the metal fibers 2 and a porous structure, i.e. voids 9, could be observed at least in some areas. Consequently, for producing the network 6 of Example 10 the time was further reduced to 3 minutes, while time and pressure were kept at the same values as for Example 9. It was found that the porous structure, i.e. the voids 9, was present substantially homogeneously throughout the whole sample.

    [0197] Images of the networks 6 of Examples 9 and 10 are shown in FIGS. 10 and 11, respectively. It can be seen from FIG. 10 that the network 6 of Example 9 has some areas in which the porosity is reduced. Particularly close to the edges of the network 6 of Example 9 many pores can be observed. By reducing the time from 5 to 3 minutes, the porosity was increased, as can be seen by the image of the network 6 of Example 10 provided in FIG. 11.

    [0198] SEM images of the network 6 of Example 10 are shown in FIGS. 12 and 13. The SEM images show the porous structure of the network 6 and that the metal fibers 2 are sintered together. The stripes which can be seen in FIGS. 12 and 13 are produced by the surface of the discs 12 of Thermax high-temperature alloy used for pressing the metal fibers 2. The stripes in the discs of the Thermax high-temperature alloy are the result of cutting the Thermax high-temperature alloy.

    [0199] Without being bound to a theory, it is assumed that the copper metal fibers 2 show an improved capability to be sintered as a result of the high energy which is stored as a result of the fast cooling rates resulting from the melt-spinning process used for preparing the fibers. The melt spinning process provides cooling rates of up to 10.sup.6 K.Math.min.sup.−1 which freezes the movement of atoms of a system before they can arrange into energetically favored states. Of course, also copper specific effects concerning atomic diffusion could play a role.

    [0200] It is possible to bend the networks 6 of the Examples described above without permanently deforming them. If these networks 6 are folded, it is possible to provide them with a stable new shape.

    [0201] Further experiments were conducted using metal fibers of the Cu-alloy Cu.sub.92Sn.sub.8. The metal fibers of the Cu-alloy Cu.sub.92Sn.sub.8 were prepared analogously to the other metal fibers described above. The metal fibers of Cu-alloy Cu.sub.92Sn.sub.8 were dispersed in 200 mL of demineralized water containing 50 mg of SDS (sodium dodecyl sulfate) and dried after vacuum filtration. In the entangled networks of metal fibers obtained this way, the metal fibers are distributed homogenously but are non-oriented.

    EXAMPLE 11

    [0202] Polycrystalline metal fibers 2 of Cu.sub.92Sn.sub.8 produced by the melt spinning process described above and treated with an aqueous solution of SDS, as also described above, were placed between two discs 12 of thermax alloy with diameters of 60 mm. The staple of the discs 12 of thermax alloy having a diameter of 60 mm and the metal fibers 2 was then placed into the 300° C. preheated hot press 10 and the fibers 2 were pressed at a predetermined pressure of 78 MPa for a predetermined time of 3 minutes, to avoid thermal expansion effects due to heating. This way a stable network 6 of Cu.sub.92Sn.sub.8 metal fibers was obtained with similar stability to deformation as the networks 6 of Examples 1 to 10 described above. The network 6 of Example 11 has a thickness in the range from 0.15 mm up to 0.25 mm. A photographic image of the network 6 of Cu.sub.92Sn.sub.8 metal fibers is shown in FIG. 14a).SEM images of the network of Cu.sub.92Sn.sub.8 metal fibers are shown in FIG. 14b) and FIG. 14c).

    COMPARATIVE EXAMPLE 7

    [0203] The network 6 of Comparative Example 7 was prepared identical to the network 6 of Example 11, however, the temperature was kept at room temperature, i.e. around 20° C. The metal fibers were not sintered together and only a weak mechanical stability was observed. The mechanical stability is a result of the metal fibers being deformed due to the applied pressure but not as a result of a sintering of the metal fibers.

    COMPARATIVE EXAMPLE 8

    [0204] The network 6 of Comparative Example 8 was prepared identical to the network 6 of Example 11, however, only a weak pressure of around 2 kPa was applied. There was no fixation between the metal fibers to one another obtained.

    [0205] It is demonstrated by Example 11 and Comparative Examples 7 and 8 that the combination of pressure and temperature is required to sinter the metal fibers to one another so that the metal fibers 2 are fixed to one another. Without being bound by a theory, it is likely that due to the pressure the metal fibers 2 are brought into close contact with each other and matching contact faces are formed between the metal fibers 2 due to mechanical deformation. The increased temperature promotes the atom movement in the direction of the pressure and therefore results in a sintering of the metal fibers 2 so that the metal fibers 2 are fixed to one another.

    EXAMPLE 12

    [0206] Preparation of a Fleece by Carding

    [0207] Carding is a mechanical process that disentangles, cleans and intermixes fibers 2 to produce a continuous fleece 26. A scheme of a carding machine 24 is shown in FIG. 24. The continuous fleece 26 is obtained by passing the fibers 3 between differentially moving surfaces, e.g. rotating drums 28, covered with card clothing 30. The carding cloth 30 breaks up locks and unorganized clumps of fiber 2 and then aligns the individual fibers 2 to be parallel with each other. While carding is well known for wool it has not been used for organizing metallic fibers 2 to form a fleece 26.

    [0208] FIG. 25 shows examples of fleece after carding. Here, ca. 10 cm long, 30 μm wide and 2 μm thick fibers of brass as prepared by melt spinning are disentangled by a carding device. Two layers are formed which are stacked on top of each other. It may also stacked in between layers of cotton in order to form cotton/metallic hybrid fleece or tissue. Also combinations of cotton and metallic fiber nets can be disentangled by one carding step and form a cotton/metallic fiber hybrid fleece.

    [0209] FIG. 25 shows photographic images of a manufacturing process according to the present invention. At first metallic fibers were disentangled by a carding device, as shown in the upper left image of FIG. 25. Reference signs are not shown in FIG. 25 for a better overview. Next layer 1 and layer 2 of differently oriented fibers, as shown by the two middle photographs in the upper line of FIG. 25 were stacked on top of each other to provide a bilayered fleece, as sown in the upper right image of FIG. 25. As a mediating material a cotton fleece such as the one shown in the lower right image of FIG. 25 can be integrated in metal fibers to provide a layered fleece, the upper side and underside of which are shown in the lower left and lower middle image in FIG. 25.

    EXAMPLE 13

    [0210] Ultrasound welding was applied to mechanical fix all fibers 2 in a metal network 6 or just fibers 2 at different sites. In principle, longitudinal and vertical ultrasound welding is possible. Respective machines are commercially available. Vertical ultrasound is the preferred technique. Here, a hammer is moving up and down at high frequencies. In principle, this is also possible by hammering a suitable object onto the fleece made of metal fibers 2. An example of a copper based fleece is shown in FIG. 26. In FIG. 26 no reference signs are shown for a better overview. The left portion of FIG. 26 shows a photographic image of a network 6 of metal fibers 2 according to the present invention. The clearly visible square shaped portion in the middle of the photographic image on the left side is an area in which the metal fibers 2 are fixed to one another via ultrasound welding, as described above. Images numbered with 1 to 3 in FIG. 26 are enlarged views of the sections indicated in the photographic image on the left side of FIG. 26 by corresponding boxes and numbers. The image beneath enlarged view number 1 is a further enlarged view corresponding to the box indicated in enlarged view number 1 and shows fixation of the metal fibers 2 to one another due to the applied ultrasound welding.

    [0211] Preparing Electrodes and Lithium Ion Batteries:

    [0212] Inventive Electrode:

    [0213] Circular discs having diameters of 6 mm were cut from the network 6 of Example 10 and infiltrated with a dispersion of 80% SnO, 10% carbon black and 10% binder. In this inventive electrode, the network 6 of sintered metal fibers 2 of copper acts as current collector 14.

    [0214] Non-Inventive Electrode:

    [0215] A copper foil was coated with a dispersion of 80% SnO, 10% carbon black and 10% binder to obtain a copper foil having a coated layer of active material on its surface. The thickness of the coating layer was adjusted to 50 μm by using tape casting. The copper foil acts as current collector 14.

    [0216] The Following Materials were Used to Prepare the Dispersion:

    [0217] SnO: Tin(II) oxide, 99.9 wt %, trace metal basis (AlfaAesar (Art. Nr. 11569))

    [0218] Carbon black: (Carbon Nanopowder <100 nm, Sigma Aldrich, P-Code: 633100-25)

    [0219] Binder: Polyvinylidenfluorid (PVDF) (AlfaAesar (Art. Nr. 44080))

    [0220] Batteries:

    [0221] A half-cell 13a was assembled, as schematically shown in FIG. 16a) including electrodes 14 and 15, separator 17 and Li-foil 16. In this half-cell 13a the electrode described above as inventive electrode was used, i.e. the network 6 of Example 10 which is coated with an active electrode material 15.

    [0222] As a reference, a further half-cell 13b was assembled, as schematically shown in FIG. 16b). In this half-cell 13b, the electrode described above as non-inventive electrode was used, i.e. a copper foil coated with 50 μm active electrode material as described above.

    [0223] If the amount of active electrode material 15 is increased in the reference half-cell 13b, i.e. a layer thicker than 50 μm is tape casted, the capacity drops, since the electrons have to pass through a thicker layer of active electrode material. It was found that by using the assembly shown in 13a, the amount of active electrode material could be increased by incorporating it into a conductive network by a factor of 60 compared to the standard assembly shown in 13b. Furthermore, it was possible to obtain a homogenous charge distribution throughout the electrode by using a metal fiber network and it was further possible to significantly increase the capacity per mass unit.

    [0224] A comparison of the capacity of these two half-cells 13a, 13b is provided in FIG. 17. As can be seen, the capacity for the half-cell 13a (according to the invention) is increased by almost 50% compared to the reference half-cell 13b.

    [0225] The capacity measurements were conducted using a Metrohm M204 electrochemical measurement system run with the software NOVA Battery 1.0. The cell was assembled in a Swagelok™ type cell, using lithium foil (Sigma Aldrich (99.8 wt %) as counter electrode, separion polymer vlieβ (Sigma Aldrich, Whatman® glass microfiber filters, Grade 934-AHO) as separator and 1 M LiClO.sub.4 (Sigma Aldrich) dissolved in 1:1 EC/DMC (Ethylcarbonate/Dimethylcarbonate (ALFA AESAR)) as electrolyte.

    [0226] In order to measure the capacity of the half-cell, a constant current of 100 mAh/g was applied, normalized to the amount of active material used for the respective electrode. The potential was simultaneously measured and as vortex points for the potential 0.0125 V (lower vortex point) and 2.2 V (higher vortex point) were measured. The resulting set of data points includes values for the potential at any given point of time. Since a constant current is supplied, the capacity can then be calculated by multiplying the time between lower (completely discharged) and higher vortex point (completely charged) with the applied current.

    [0227] In addition to the above mentioned half-cells 13a further electrodes were prepared and assembled together with a counter electrode, a separator and an electrolyte and analyzed. The details are the following:

    [0228] Fibers from the copper alloy Cu.sub.96Si.sub.4 were dispersed, vacuum filtrated and afterwards pressed between two 60 mm diameter Thermax alloy with a pressure of 300 kN at 300° C. for 3 min. From the resulting sintered network (mechanical stable) a 10 mm diameter circular disc was punched out and infiltrated with a dispersion of 90% graphite and 10% binder. Here, the copper-alloy network acts as current collector. As reference, a copper foil was coated with a 50 μm layer of the same dispersion using a doctor blade.

    [0229] The graphite and binder used were the following:

    [0230] Graphite: powder, <20 rim, synthetic (SigmaAldrich (Art. Nr. 282863))

    [0231] Binder: Polyvinylidenfluorid (PVDF) (AlfaAesar (Art. Nr. 44080))

    [0232] The capacity measurements were conducted using a Metrohm M204 electrochemical measurement system run with the software NOVA Battery 1.0. The cell was assembled in a Swagelok™ type cell, using lithium foil (Sigma Aldrich (99.8 wt %) as counter electrode, glas fiber (Sigma Aldrich, Whatman® glass microfiber filters, Grade 934-AH®) as separator and EC:DMC (1 M LiPF.sub.6) as electrolyte (EC: ethylene carbonate; DMC: dimethyl carbonate).

    [0233] In order to measure the capacity of the half-cell, a constant current of 382 mA/g was applied, normalized to the amount of active material used for the respective electrode. The potential was simultaneously measured and as vortex points for the potential 0.0125 V (lower vortex point) and 2.2 V (higher vortex point) were measured. The resulting set of data points includes values for the potential at any given point of time. Since a constant current is supplied, the capacity can then be calculated by multiplying the time between lower (completely discharged) and higher vortex point (completely charged) with the applied current.

    [0234] For the reference cell with reference electrode, a decrease of 76% of the capacity (from 1183 mAh/m.sup.2 to 289 mAh/m.sup.2) was noticed after 50 cycles. For the cell with inventive network electrode, a decrease of only 9% of the capacity (from 1492 mAh/m.sup.2 to 1381 mAh/m.sup.2) was noticed. This demonstrates that the inventive network electrode is more electrochemically stable than the common reference electrode with a nearly constant capacity over 50 cycles whilst using the same active material. This is related to the 3D network of the inventive electrode, which supports the effective distribution of the stress formed within the active material due to swelling during ion intercalation. Upon intercalation of Lithium ions the active material undergoes swelling (up to 8 vol % for graphite), which leads to a decrease of capacity over cycling in the reference cell.

    [0235] Furthermore, these results indicate that the electronic conductivity is improved by the inventive metal fiber network used as current collector. The network improves the electrode conductivity by shortening the electron conduction path compared to the reference electrode where the conductivity gradient is present through the electrode. Thus for the case of the inventive electrode, it was possible to charge and discharge the half-cell 50 times within 6 hours compared to the reference electrode, which was charged 50 times in 16 hours. In addition, as mentioned above the capacity for the inventive network electrode could be preserved (1492 mAh/m.sup.2.fwdarw.1381 mAh/m.sup.2), which was not the case for the reference electrode (1183 mAh/m.sup.2.fwdarw.289 mAh/m.sup.2).

    [0236] FIGS. 18 and 19 present plots of charge discharge profiles of the reference cell and the inventive network cell respectively. FIGS. 20a) and 20b) presents the first and last cycle of the same measurement. FIGS. 21a) to 21c) present the charge discharge profiles with a current normalized time for a better view of the changes while the cyclization. FIG. 22 presents the capacity over the cycles.

    [0237] For all calculations, the first cycle of each measurement was excluded.

    [0238] FIG. 23a) presents a cross section SEM image of the inventive network electrode, which visualizes graphite flakes between the copper fibers. The same spot investigated by EDX (EDAX model ZEISS Ultra 55, 132-10) mapping is shown in FIGS. 23b) and 23c).

    [0239] Microstructure of Metal Fibers:

    [0240] To investigate the influence of the microstructure of the metal fibers in the network of metal fibers, fibers of CuSi4, A199Si1, Cu92Sn8, Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 and FeNiB were produced using the above described method. The alloys listed in Table 3 were thermally pretreated with the parameters also listed in Table 3 in order to reduce the stored defect energy, but without causing a change in the grain structure caused by recrystallization. The amorphous/nanocrystalline fibers were additionally thermally treated above the crystallization temperature to investigate the difference in the microstructural state.

    TABLE-US-00003 TABLE 3 defect removal (thermal pretreatment) crystallization Temperature Duration Temperature Duration material [° C.] [h] [° C.] [h] CuSi.sub.4 300 1 — — Al.sub.99Si.sub.1 200 1 — — Cu.sub.92Sn.sub.8 300 1 — — Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 300 1 600 1 Fe.sub.40Ni.sub.40B.sub.20 300 1 450 1

    [0241] Parts of the fibres in each state (before thermal pretreatment, after thermal pretreatment and after crystallization) were melted twice with a Netzsch STA 449 F3 under argon atmosphere. Thermal pretreatment was performed using the parameters described in Table 3 using also the Netzsch STA 449 F3. All samples were heated at a constant heating and cooling rate of 10 K/min from 30° C. to 1200° C. or 900° C., respectively, in the case of Al.sub.99Si.sub.1, and then cooled down again to 30° C. The samples were then heated to 1200° C. or 900° C., respectively. The temperature was kept at 30° C. for 1 h between the individual melting steps or where applicable between thermal treatment and melting. By subtracting the second from the first hating cycle, the measurement could be adjusted so that only the pure energy quantity of the fibers was measured. The difference of the area integrals between the measurements with and without thermal pretreatment corresponds to the stored defect energy or crystallization energy. The amount of stored energy is shown in Table 4 as defective energy and crystallization energy. No energy could be determined for CuSi.sub.4 because during thermal pretreatment the fibers were transferred into a thermodynamically stable phase with a two-phase structure. Due to the resulting two-phase structure, a correct measurement of the stored defect energy is not possible.

    TABLE-US-00004 TABLE 4 Defective Crystallization material energy [kJ/g] energy [kJ/g] Cu.sub.96Si.sub.4 — — Al.sub.99Si.sub.1 1.7 — Cu.sub.92Sn.sub.8 1.8 — Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 0.1 2.6 Fe.sub.40Ni.sub.40B.sub.20 2.8 3.8

    [0242] It can be seen from the results of table 4 that metal fibers obtained by melt spinning, as described above, have a significant amount of stored energy in form of defective energy and/or crystallization energy, i.e. these fibers are not in their thermodynamic equilibrium. Even though no values are indicated for the alloy Cu.sub.96Si.sub.4, it is noted that also metal fibers of this alloy have a significant amount of defective energy; however, it is not possible to make a meaningful quantification thereof, since the material is transferred into a state of thermodynamic equilibrium when subjected to the thermal pretreatment conditions specified in table 3. The combined amount of defective energy and crystallization energy for Cu.sub.96Si.sub.4 is estimated to be around 2.3 kJ/g.

    [0243] The fibers were then weighed against the values given in Table 5 and wet laid to form uniform non-woven structures thereof. These were then sintered using pressure-induced low-temperature sintering as described above for Example 1 with the parameters also listed in Table 5. Before sintering, the electrical conductivity of the non-woven structures of the copper and aluminium alloy fibres was determined by means of a 4-point measurement and an impedance measurement. These measurements were repeated after sintering. The measured values before and after sintering are listed in Table 6.

    TABLE-US-00005 TABLE 5 Initial weights and pressure sintering parameters. Weights Temperature Pressure Duration material [g] [° C.] [MPa] [min] Cu.sub.96Si.sub.4 1 300 35 5 Al.sub.99Si.sub.1 0.7 200 35 5 Cu.sub.92Sn.sub.8 1 300 35 5 Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 1 400 140 5 Fe.sub.40Ni.sub.40B.sub.20 0.7 400 140 5

    TABLE-US-00006 TABLE 6 Electrical conductivities. Loose fibers Sintered network (before sintering) (after sintering) 4-point Impedance 4-point Impedance material [Ohm/m] [Ohm/m] [Ohm/m] [Ohm/m] Cu.sub.96Si.sub.4 57.14 45.71 2.48 2.39 Al.sub.99Si.sub.1 230.77 538.46 2.71 2.7 Cu.sub.92Sn.sub.8 300 1000 5.45 3.61

    [0244] It can be clearly seen that the conductivity of the sintered samples is many times higher than that of the unsintered samples. It is worth mentioning that the distance between the contacts which was used to measure the material's conductivity was only a 5 mm in the case of the loose fibers (before sintering). Increasing the distance between these contacts increased the resistance to more than 100 fold. This results from the non-sintered fibers which do not form a stable conductivity in between fibers. Contrary, while increasing the distance between contacts in case of sintered networks the conductivity was barely depending on the distance. This is because of the high conductivity in between fibers due to sintering.

    [0245] In order to investigate the mechanical stability of the samples, 10 mm wide strips were cut from each of the samples and examined by means of a tensile test at a tensile rate of 0.01 mm/s. The results of the tensile measurements are presented in Table 7. The same number of fibres per cross section was chosen for the standardization of the samples, as all samples were made of homogeneous fibres from the same production with the same basis weight. Accordingly, a sample from a network of metal fibers with the fibers being subjected to thermal pretreatment as described in table 3 had the same density of fibers compared to a sample from a network of metal fibers with the fibers not being subjected to such a thermal pretreatment.

    TABLE-US-00007 TABLE 7 Mechanical stability of sintered fiber networks Untreated fibers, as Fibers subjected to obtained from the melt conditions in table 3 spinning process before sintering Tensile Young's Tensile Young's strength modulus strength modulus material [N] [N] [N] [N] CuSi.sub.4 12.5 1741 8.1 1151 Al.sub.99Si.sub.1 7.1 1118 0.4 2.3 Cu.sub.92Sn.sub.8 17.6 5288 9.9 1754 Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 7.1 858 — — Fe.sub.40Ni.sub.40B.sub.20 10.7 1894 6.3 913

    [0246] It can be clearly seen that by thermal treatment of the samples and the associated degradation of stored defect the mechanical properties of the sintered networks are negatively influenced. In other words, using metal fibers with structures that are not in thermodynamic equilibrium the strength of the resulting sintered network is improved. This becomes particularly clear with the Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 samples subjected to thermal pretreatment. The thermally pretreated Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 samples could not be sintered at all, in contrast to the untreated ones. The crystallized samples of the amorphous/nanocrystalline alloys Co.sub.66Fe.sub.4Mo.sub.2B.sub.12Si.sub.16 and Fe.sub.40Ni.sub.40B.sub.20 could not be sintered at all; samples where annealed before sintering process. They disintegrated during pressing into fine fiber particles without any mechanical cohesion.

    [0247] It can be summarized that only fibers produced by melt spinning can be mechanically firmly bonded to each other via pressure-induced low-temperature sintering in case the samples are not annealed before sintering. The results presented here show how large the influence of the defect energy introduced by the manufacturing process is on the degree of sintering and thus on the mechanical and electrical properties of the 3D meshes. For the use of 3D nets as current collectors, it is essential that the fibers are firmly connected to each other to ensure a constant electrical conductivity across the entire battery electrode. Nevertheless, it is required that the structure of the fibres is maintained during the sintering, i.e. the fibers are not pressed to a metal foil without open pores.

    [0248] In the following, the present invention is further described in regard to a process of producing a battery. The Battery production consists of 7 process steps:

    [0249] 1. Production of metal fibers (step 1 in FIG. 27)

    [0250] 2. Carding of fiber nets for laying fiber fleece (step 2, b1, in FIG. 27); alternatively depositing fibers by liquid dispersion or air flow (step 2, b2, in FIG. 27).

    [0251] 3. Sintering of metal fibers for the formation of a metal fiber net electrode (step 3 in FIG. 27); in a hot press (c1, c2), by ultrasound welding (c3) or hammering (c4).

    [0252] 4. Formation of anode and cathode by loading the metal fiber net with electrode active material (step 4 in FIG. 27)

    [0253] 5. Calendering of electrodes (step 5 in FIG. 27)

    [0254] 6. Ultrasound welding of conductive wire on electrode as connector (step 6 in FIG. 27)

    [0255] 7. Assembly of battery (step 7 in FIG. 27)

    [0256] A schematic representation for these process steps is provided in FIG. 27, wherein step 1 shows production of metal fibers, step 2 shows arranging of fiber nets for layer fiber fleece, step 3 shows different sintering methods of metal fibers for the formation of a mechanically stable metal fiber network. Sintering can be performed by pressure induced low temperature sintering (part c1) or by thermal sintering (part c2), ultrasound welding (part c3) or hammering (part c4). In part d the anode and cathode are formed by loading the metal fiber network with active material (step 4). The electrodes are densified in part e using calendering (step 5). As a next step, a conductive foil is attached onto the network of metal fibers in step 6. Battery assembly is shown in step 7 of FIG. 27. These process steps are also described in the following.

    [0257] Step 1. Production of metal fibers

    [0258] Metal fibers are produced by melt spinning. Two principal different melt spinning technologies can be used to produce metal fibers: a) vertical melt spinner, b) horizontal melt spinner; the vertical melt spinner has technical limitations which makes the product more expensive and less efficient compared to production using a horizontal melt spinner. Therefore, the horizontal melt spinner is preferably used in the present invention.

    [0259] Step 2. Laying fiber fleece

    [0260] The metallic fibers are treated much like cotton for textiles the reason why the resulting net can be referred to as a metallic textile. First, the metallic “wool” is disentangled and the fibers aligned by carding, such as schematically shown in FIG. 24. This step requires microfibers of substantial length. By using melt spinning as described herein, it is possible to produce fibers having a length of several centimeters; no liquid dispersion step is necessary or possible due to the substantial length of the fibers. This process produces a 3-dimensional fiber net for subsequent processing. In step 2, b1, the fibers are carded by a carding comb, as known in the art of cotton treatment, to obtain a homogenous fiber network structure and such ordered layers are stacked on top of each one. To do that, the fibers preferably have a length of 5-18 cm and have no connections between them before carding. In this step no liquid dispersion step is necessary. Avoiding liquid dispersions is possible due to the long extension of the fibers.

    [0261] Alternatively, non-ordered fiber nets are produced by deposition from liquid dispersions or from air flow (step 2, b2).

    [0262] Step 3. Sintering of metal fibers for the formation of a 3D metal fiber nets The uniform fiber network structure can be sintered between two heating plates (distance adjustable between 0.2-1 mm, here 0.5 mm; Al.sub.99Si.sub.1 at 650° C. for 1.5 h; Cu.sub.96Si.sub.4 at 950° C. for 2 h), such as schematically shown in step 3, c2 of FIG. 27.

    [0263] This leads to strong mechanical connections between fibers at their crossing points. The electrical conductivity increases substantially once the metal fibers are mechanically connected and sintered.

    [0264] Alternatively, the 3D metal fiber net is pressure-sintered between 2 thermo plates. For this, the fibers are placed on two polished thermo plates and pressed 1 min with an pressure of 10 GPa at 150° C. for Al.sub.99Si.sub.1 and 300° C. for Cu.sub.96Si.sub.4 in a preheated uniaxial press, such as schematically shown in step 3, c1 of FIG. 27. After this, the mechanical stable 3D fiber-network can be easily zipped off from the substrate.

    [0265] Alternatively, metal fibers are locally fixed by ultrasound welding (FIG. 27, step 3, c3) or hammering (FIG. 27, step 3, c4).

    [0266] Step 4. Formation of anode and cathode by loading the 3D metal fiber net with electrode active material

    [0267] The active material used in the examples is commercially available from Custom Cells as a slurry. Graphite was used for the Anode side and NMC_111 for the cathode side.

    [0268] Next, utilizing a standard doctor blading process (from doctor blade; Rakelprozess) the 3D metal fiber net was loaded with a slurry of active material, such as illustrated in FIG. 27, step 4, d. A siliconized PMMA foil was placed onto the plate, wetted before with ethanol/acetone. After flattening the foil, the sintered fiber network was placed onto the foil. Then, the rather liquid slurry of active material was drop-casted onto the 3D metal fiber net. The 3D metal fiber net structure offers capillary forces which pulled the slurry into the network and coated it homogenously. The, more viscous slurry was poured onto the network and the gap doctor blade was used to remove residual slurry (height 0.650 mm). Subsequently, the samples were dried (Anode at RT and Cathode at 30° C.).

    [0269] The 3D metal fiber network was formed without being in contact with active material, i.e. the active materials were applied onto the networks of metal fibers after sintering of these networks was completed.

    [0270] Lamination/Post-Processing of the Electrodes

    [0271] Step 5. Calendering of electrodes

    [0272] After drying the electrodes, they were laminated using a gap of 0.2 mm and a weight limited calendaring process of 40 kg per roller. Additional experiments with the following parameters were also conducted: [0273] 0.4 mm gap, 40 kg [0274] no gap, 40 kg [0275] no gap, 120 kg [0276] no gap, 112 kg, 160° C.

    [0277] A schematic illustration of such a calendering process is illustrated in FIG. 27, step 5, e.

    [0278] Step 6. Ultrasound welding of conductive foil on electrode

    [0279] Finally, a Ni foil was attached to the side of an electrode by ultrasound welding. These Ni foils were the contact electrodes of the battery. Ultrasound welding of contact electrodes to the network 6 is schematically illustrated in FIG. 27, step 6, f.

    [0280] Step 7. Assembly of battery

    [0281] The packaging of the 3D metal fiber networks started with punching out the active material loaded electrodes in the desired dimension/geometry. The samples were placed into a uniaxial press and punched out in the desired form. Subsequently, the samples were glued using PVDF Binder in acetone onto the respective sides of a separator (PP/PE stretched), which was punched out beforehand with an overlap of 1-2 mm in order to avoid internal shorting of the electrodes. Then, they were placed into the laminated pouch and dried at 110° C. for 48 h in a furnace. After 48 h the samples were transferred into a glovebox, filled with electrolyte (EC/DMC, 1 M LIPF.sub.6) and sealed in order to ensure an airtight packaging. After wetting the sample for 3 h, the surplus of electrolyte was removed using a syringe and a vacuum pump with a liquid filter and sealed again tightly directly below the electrode.

    [0282] Results:

    [0283] The battery according to the invention which was obtained by the process described above was compared to a comparative battery not containing the network of metal fibers as current collector but a planar foil. The results are provided below in tables 8 and 9.

    TABLE-US-00008 TABLE 8 values for gravimetric capacity and gravimetric energy density Inventive battery comparative battery gravimetric 140 Ah/kg (0.1 C)  63 Ah/kg (0.1 C) capacity 135 Ah/kg (0.5 C)  52 Ah/kg (0.5 C)  67 Ah/kg (1 C)  29 Ah/kg (1 C) gravimetric 519 Wh/kg (0.1 C) 233 Wh/kg (0.1 C) energy 499 Wh/kg (0.5 C) 196 Wh/kg (0.5 C) density 249 Wh/kg (1 C) 107 Wh/kg (1 C)

    TABLE-US-00009 TABLE 9 values for volumetric capacity and volumetric energy density Inventive battery comparative battery volumetric  87 Ah/l (0.1 C)  28 Ah/l (0.1 C) capacity  81 Ah/l (0.5 C)  24 Ah/l (0.5 C)  41 Ah/l (1 C)  18 Ah/l (1 C) volumetric 320 Wh/l (0.1 C) 106 Wh/l (0.1 C) energy 301 Wh/l (0.5 C)  89 Wh/l (0.5 C) density 150 Wh/l (1 C)  67 Wh/l (1 C)

    [0284] In above tables 8 and 9 the c-rate used for determining values for gravimetric and volumetric capacities and energy densities are indicated as 0.1C, 0.5C and 10, respectively.

    REFERENCE SIGNS

    [0285] 1 melt spinning device [0286] 2 metal fibers [0287] 3 rotating wheel [0288] 4 microstructured nozzle [0289] 5 drop [0290] 6 network [0291] 7 point of contact [0292] 8 tweezers [0293] 9 voids [0294] 10 hot press [0295] 11 upper and lower parts [0296] 12 disc [0297] 13a battery [0298] 13b prior art battery [0299] 14 current collector [0300] 15 active electrode material [0301] 16 lithium [0302] 17 separator [0303] 20 sharp peak [0304] 22 peak [0305] 24 carding machine [0306] 26 fleece [0307] 28 rotating drums [0308] 30 carding cloth