Metal fiber network and method of assembling a fiber network

Abstract

The invention relates to a method of assembling a network of fibers comprising a plurality of metal fibers, wherein the method comprises the steps of: providing a loose network consisting of a plurality of metal fibers at an assembly site; fixing the plurality of metal fibers to each other by heating the plurality of fibers to a fixing temperature selected from the range of 50-98% of the melting point temperature of the plurality of fibers at a heating rate higher than 50 K / min, in particular higher than 100 K / min, specifically higher than 200 K / min, preferably higher than 1000 K / min; and cooling the plurality of fibers at a cooling rate higher than 50 K / min, preferably higher than 100 K / min. The invention further relates to a network of metal fibers comprising a plurality of metal fibers fixed to each other at contact points, wherein the metal fibers comprise a non-circular cross-section, in particular a rectangular cross-section, a cross-section of a conic, a partly circular or an elliptical cross-section with a major axis and a minor axis, or wherein the metal fibers comprise a circular cross-section, and wherein the fibers comprise a width that is substantially constant along the length of the fiber, such that the variation of the width of the fiber along its length is less than 40%, preferably less than 30%, in particular less than 20%.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method of assembling a network of fibers comprising a plurality of metal fibers and to a network of metal fibers. BACKGROUND

[0002] Today, fiber networks are applied in a variety of applications, such as from filters to batteries.

[0003] Traditionally, the filtration of a gas, such as air, or a liquid is based on a network or foam of metal fibers. Today, such a network or foam has become part of a variety of devices, from oil filters in automotive applications to cleaning systems for liquids or gases, such as air.

[0004] Traditionally known filters are generally based on metal fibers comprising a circular cross-section, such as oil filters, or on carbon-based foams, such as HEPA filters. A characteristic of filters made of metal fibers having a circular cross-section is that such fibers have a high mechanical stability, while having a small surface to volume ratio. However, such filters generally have a considerable weight, since a large amount of fibers is required. On the other hand, filters made of carbon-based foams are mostly rather fragile, but light in weight and have a considerable internal surface area. Furthermore, it is noted that the filtration capacity of such traditionally known filters is not ideal.

[0005] In other fields of application, the performance of secondary batteries can also be improved when a network of metal fibers is used as a secondary electrode. Such a network of metal fibers can also contribute to improving the performance of catalytic materials in electrochemical applications, such as in fuel cells and water splitting, or as a component in electromagnetic shielding materials, as a filter, in polymer composites or as a tissue material and tissue hybrid material, which can also include as additives, for example, cotton, silk or wool.

[0006] Due to the diversity of the different fields of application, the need to be able to manufacture a network of fibers with different determined properties depending on the field of application is increasing.

[0007] In a traditionally known method of manufacturing a network of fibers, a plurality of fibers is provided in a hot press and subjected to a high pressure. The plurality of fibers is then placed in a furnace and slowly heated to a temperature close to the melting temperature of the fibers, while the plurality of fibers is still subjected to the pressure. The high temperature is maintained until the fibers are connected to each other. Then, the manufactured network is slowly cooled.

[0008] The above process is also referred to as “sintering”. Depending on the capacity of the used furnace, this process usually takes one hour or more. It is now recognized that conventional sintering allows the fibers to experience relaxations before reaching a temperature high enough to connect the fibers to each other. These relaxation processes release the stored energy of the fibers. For example, fibers obtained by fast cooling techniques such as melt spinning will have a large amount of stored energy.

[0009] The driving force for the above process is the reduction of the fiber surface and the corresponding reduction of its free energy AG. The free energy AG can be divided into a surface contribution AG S , a volume contribution AG V , and a grain boundary contribution AG B . This relationship is shown in equation (1). During fiber sintering, the volume fraction remains almost constant (AG V = 0), while the grain boundary fraction increases (AG B > 0) due to the transformation (i.e. reduction of the surface) and the volume fraction decreases (AG V < 0). The volume contribution AG V is significantly larger than the grain boundary contribution AG B , which leads to a negative change in the total free energy of the system (AG < 0) and the process happens automatically once a certain energy threshold (activation energy) is exceeded. During conventional sintering, the reduction of AG is also related to the rounding of the fibers, i.e. the transformation of the fiber diameter into a circle (also referred to as rounding herein).

[0010] AG T = AG V + AG S (1)

[0011] Here, the energy threshold to be exceeded is the activation energy for diffusion E A (equation (2)). Here, D0 is the temperature dependent diffusion constant, k is the Boltzmann constant, T is the absolute temperature, and D is the temperature dependent diffusion constant. The larger the temperature dependent diffusion constant D (in units of m 2 s -1 ), the faster the rounding of the fibers. Here, the temperature is not only responsible for fulfilling the activation energy E A , but also the speed.

[0012]

[0013] Therefore, this known sintering process happens by a reconstruction process at atomic level (diffusion) and not by a process of remelting the fibers. The thermodynamic goal is to obtain the largest volume with the smallest surface area possible. Its optimal ratio is achieved by the optimal sphere.

[0014] In order to produce, for example, a fiber network with a defined cross section of the fibers, the effect cannot really be controlled with the methods conventionally known for the manufacture of fiber networks. SUMMARY

[0015] It is therefore an object of the present application to provide a method of assembling a fiber network with enhanced control of the fiber cross section and a corresponding fiber network. This object is solved by the subject matter of the independent claims.

[0016] In particular, the present application provides a method of assembling a fiber network comprising a plurality of metal fibers, wherein the method comprises the following steps:

[0017] - providing a loose network consisting of a plurality of metal fibers at an assembly site;

[0018] - fixing the plurality of metal fibers to each other by forming contact points between the individual metal fibers by:

[0019] heating the plurality of fibers to a fixation temperature selected from the range of 50-98% of the melting point temperature of the plurality of fibers at a heating rate higher than 50 K / min, in particular higher than 100 K / min, in particular higher than 200 K / min, preferably higher than 1000 K / min; and

[0020] cooling the plurality of fibers, in particular to a temperature lower than 60% of its melting point, at a cooling rate higher than 20 K / min, preferably higher than 50 K / min, preferably higher than 100 K / min.

[0021] As mentioned above, fibers subjected to heat tend to rearrange on an atomic level in order to achieve the maximum possible volume with the minimum possible surface. From the fiber's point of view, the "perfect" state is a perfect sphere. Thus, in a conventional manufacturing process, due to the heating step, the fibers start to rearrange on an atomic level in order to reach a more preferred energy level by, for example, crystallization or by reducing defects in the fiber's crystal lattice. As a result, the fibers can even change their shape by converting their cross section from a flat or elliptical cross section to a circular cross section, i.e. a rounding of the fiber cross section shape occurs. During the transition to the thermodynamically most preferred spherical shape, not only a rounding effect of the fiber cross section shape can be observed, but also a change in the fiber diameter. Before reaching the spherical shape, the fibers show a cross section with a reduced diameter, herein referred to as a constriction. These constrictions further develop until the fiber breaks. Eventually, the fibers transform into a plurality of droplets, i.e. they reach the spherical shape.

[0022] The method of the present invention utilizes the kinetics of the rearrangement process. Rearrangement occurs only when the fibers have sufficient time to do so. By increasing the heating and cooling rates and by preferably maintaining a fixed temperature equal to or less than 30 minutes, it is ensured that contact points are formed where the metal fibers connect to each other. However, the rearrangement process is significantly reduced, and in particular, shape-changing effects, i.e., rounding and the formation of shrinkage and breakage, can be avoided. Therefore, in the method of the present invention, the heating and cooling rates are maintained above 20 K / min, preferably above 50 K / min, and preferably above 100 K / min. It should be noted at this point that the assembled network is preferably cooled to a temperature 60% below the fiber melting temperature at the aforementioned cooling rate. Once the fibers are cooled to this temperature rate, the cooling rate is no longer critical, and therefore the cooling rate can be reduced if necessary. Conventionally known sintering processes are carried out at heating / cooling rates of about 10 K / min to 20 K / min, which requires a much longer time (up to several hours) to heat / cool the fibers. When the heating rate is too low, a relaxation process may occur in the loose network of metal fibers before reaching a fixed temperature, thereby reducing the surface component of free energy ΔG. S and grain boundary component ΔG B Therefore, when a low heating rate is applied in the step of heating the metal fibers (also known as the first method step), an additional fixing step may be required, in which the fixing temperature is maintained for more than 30 minutes. By maintaining a low heating rate, the fixing time of 30 minutes or less in the additional fixing step may be insufficient to fix the fibers to each other, because the fixation cannot be achieved from the surface component ΔG. S and grain boundary component ΔG B Benefiting from the same expansion as when a higher heating rate is applied in the heating step. The need for more time to reach a fixed temperature and / or optionally maintaining the fiber at that temperature for a longer period causes the fiber to transform into a more thermodynamically favorable state, i.e., the fiber cross-section can change towards a circular cross-section. As mentioned above, when a fiber attempts to transform into a more thermodynamically stable state, not only may the cross-section change, but the fiber width may also become non-uniform and / or the fiber width may shrink. These shrinkages can even cause the fiber to break, resulting in a reduction in fiber length, as illustrated in the figures and discussed in more detail below.

[0023] By applying the method of the present invention, the loose fiber network can be assembled into a fixed fiber network with minimal (unwanted) impact on the atomic level. Thus, the cross-sectional shape and length of the fibers can be maintained. As mentioned above, this is not possible for the conventional known methods, because the stored energy is already released from the fibers during heating before the respective fixation temperature is reached. Thus, for the conventional sintering process, the fibers are in a thermodynamically more stable state when the fixation temperature is reached. In turn, higher fixation temperatures and higher fixation times are required to drive changes in the cross-section, diameter and / or length of the fibers.

[0024] In the method according to the present invention, the applied temperature, i.e. the fixation temperature, depends on the material of the metal fibers. In order to avoid crystallization of amorphous metal fibers during the welding process, the applied temperature is preferably below the crystallization temperature of these fibers. The crystallization temperature can be determined, for example, by performing a differential scanning calorimetry (DSC) measurement on the metal fibers in question. The DSC measurement can be performed using the following conditions: heating rate 10 K min -1 from the starting temperature of 30 °C to 1200 °C, continue at a cooling rate of 10 K min -1 to room temperature. The DSC measurement can be performed in an argon atmosphere, wherein the argon flow is 100 ml min -1 , a zirconium-oxygen-trap system is used for a completely oxygen-free atmosphere (STA 449 F3 Jupiter, Netzsch Bj. 2017).

[0025] In the context of the present description, “% of the melting temperature” refers to the melting temperature in °C as determined by, for example, a differential scanning calorimetry (DSC) measurement. Thus, if the melting temperature is 1000 °C, in the context of the present description, 20% of the melting temperature is 200 °C, 50% of the melting temperature is 500 °C, and 95% of the melting temperature is 950 °C.

[0026] In addition, as an additional effect, with the method according to the present application, the network can be assembled to be flexible and can repeatedly be deformed without causing a deterioration of the network, i.e. without separating individual metal fibers from the network of metal fibers due to the deformation. The metal fibers are fixed to each other such that the metal fibers are in contact with each other, i.e. the contact points cannot move relative to the metal fibers, for example in a non-woven mass of entangled metal fibers, such as a metal felt. As a result, the network of metal fibers according to the present application is mechanically stable but also flexible. In this context, mechanically stable means that the network of metal fibers is not a loose mass of metal fibers, i.e. even if small forces act on the network, the network does not disintegrate into isolated metal fibers. Thus, such a network of metal fibers can be deformed flexibly without breaking. The network of metal fibers can recover its shape after the deformation. However, if the network of metal fibers is folded, it can also be permanently reshaped.

[0027] The method according to the present application can also distribute the contact points throughout the assembled network, so that there are contact points in the three-dimensional structure of the network of metal fibers. Thus, the contact points are not only provided in specific regions of the network of metal fibers, for example in the center or the circumference of the network. Throughout the network, the contact points can be evenly distributed. It is also possible that throughout the network, the density of the contact points is gradient, i.e. the network has regions with a higher density of contact points and regions with a lower density of contact points. It is also possible to have an ordered or random spatial distribution of the contact points.

[0028] In this regard, it should also be noted that each of the metal fibers can have at least two contact points with other metal fibers, more preferably at least three contact points, even more preferably at least four contact points.

[0029] According to one embodiment of the present application, the method can further comprise the step of maintaining the fixed temperature for a fixed time, said fixed time being selected from the range of 0 seconds to 30 minutes, in particular from the range of 0 seconds to 15 minutes, preferably from the range of 0 seconds to 5 minutes, wherein the step of maintaining the fixed temperature is performed prior to cooling the plurality of fibers. Thus, as mentioned above, the method can comprise the additional step of maintaining the fixed temperature for a predetermined time. However, it can also be stated that by maintaining a high heating rate, the fixed time can be reduced to a minimum. That is, in some cases, the cooling process can even be started immediately upon reaching the fixed temperature, thereby essentially maintaining the temperature for 0 seconds. In this regard, it should be noted that in reality, when starting the cooling process immediately after reaching the fixed temperature, the fixed time is obviously not exactly 0 seconds, but slightly around 0.1 seconds or less. Thus, in the context of the present application, a fixed time of 0 seconds relates to the case that the cooling step is started directly after reaching the fixed temperature during the heating step. For some embodiments, the fixed time can be 1 second or more, 2 seconds or more, 3 seconds or more, 10 seconds or more, or even 30 seconds or more.

[0030] According to another embodiment, the method can comprise a cleaning step performed prior to the step of fixing the plurality of fibers to each other, said cleaning step comprising cleaning the plurality of fibers by heating the plurality of fibers to a cleaning temperature selected in the range of 20% to 60% of the fiber melting temperature. The lower limit can even be chosen to be at room temperature, in particular slightly above room temperature. It has been shown that metal fibers usually comprise different types of impurities and / or attachments on their surface, which are usually secondary products of their respective manufacturing process. By heating the fibers to said cleaning temperature, the impurities and / or attachments are decomposed (i.e. evaporated or combusted), such that the remaining fibers only comprise a clean surface. Such a cleaned surface is easier and better to sinter to each other.

[0031] In this regard, it should be noted that, ideally, the cleaning temperature should be well below the temperature at which the fibers start to lose their stored energy, i.e. below the temperature at which the above-mentioned rearrangement process tends to occur. In this way, the fibers can be cleaned to remove said attachments / impurities, without having already started to sinter to each other.

[0032] In the unfortunate case that the cleaning temperature is selected at a temperature at which the fibers have already started to reassemble on their atomic level, e.g. due to the nature of the attachments present on the fiber surface, special care has to be taken to maintain the fibers at said cleaning temperature for as short a time period as possible, i.e. only until the fibers are cleaned. In this case, it can be necessary to adapt the fixed temperature of the subsequent heating step to a higher temperature in order to be able to effectively sinter the cleaned fibers to each other. The minimum time period for cleaning the fibers can be determined by trial and error.

[0033] Therefore, the cleaning temperature can be selected based on the material of the fiber and the material from which the fiber appendages / impurities should be cleaned.

[0034] The step of cleaning multiple fibers can be performed at the same assembly site, i.e., in the same oven, as a step of heating multiple fibers to a fixed temperature. This arrangement is often referred to as batch processing. In other embodiments, different ovens or a single oven with different heating zones through which multiple fibers pass in each method step may be present. On the other hand, this process is called a continuous process and is particularly advantageous when assuming the processing of large quantities of fibers.

[0035] The step of cleaning multiple fibers may also include applying an airflow at the assembly site. This airflow can help remove evaporating / decomposing deposits around the multiple fibers, so that once the fibers are cooled again, the deposits cannot re-form onto the multiple fibers.

[0036] The gas flow can be provided, for example, by applying suction to the assembly site to draw away the combustion / evaporation adjuncts. Another possibility is to provide a gas flow, such as a reactive gas (e.g., oxygen or air) or an inert gas (e.g., nitrogen or argon), at the assembly site, configured to blow away the evaporated / combustion adjuncts. The method of providing the gas flow can be selected based on other conditions present at the assembly site, such as whether the assembly site is provided in a vacuum, in air, or in a protective gas. For example, in cases where the fiber is made of a material that tends to react with oxygen, the provided gas flow can be selected to clean the surrounding environment and thus be oxygen-free, for example, by providing an inert or protective gas.

[0037] It is also possible that the step of cleaning multiple fibers includes reducing the atmospheric pressure at the assembly site. This can be used, for example, for appendages containing high vapor pressure. The atmospheric pressure can be reduced to less than 80 kPa, less than 50 kPa, or less than 10 kPa. In some embodiments, the pressure can even be reduced to less than 1 kPa, less than 0.1 kPa, or even as low as 0.0001 kPa, i.e., a vacuum can be applied. The atmospheric pressure used herein corresponds to a pressure of 101 kPa.

[0038] The vapor pressure is defined as the pressure exerted by vapor in thermodynamic equilibrium with its condensed phase (solid or liquid) at a given temperature in a closed system. The equilibrium vapor pressure is an indicator of the liquid's evaporation rate. It is related to the tendency of particles to escape from the liquid (or solid). Substances with high vapor pressure at room temperature are generally called volatile substances. The pressure exhibited by vapor above the liquid surface is called vapor pressure. As the liquid temperature increases, the kinetic energy of its molecules also increases. As the kinetic energy of the molecules increases, the number of molecules that convert into vapor also increases, thus increasing the vapor pressure.

[0039] In another aspect of the application, cleaning the plurality of fibres further comprises determining the compound to be removed, i.e. the contaminant. Further, in this aspect of the application, the step of cleaning the plurality of fibres further comprises reducing the pressure and / or increasing the temperature based on the vapour pressure curve of the compound to be removed. In particular, the pressure is reduced and / or the temperature is increased in such a way that the fibres are eventually subjected to conditions in which the compound to be removed is in the gas phase according to its vapour pressure curve. The pressure and / or temperature change can be performed in a stepwise manner, i.e. by changing the pressure and / or temperature in incremental steps and subsequently keeping these parameters essentially constant for a certain time, or by changing the pressure and / or temperature in a continuous manner. The required pressure and / or temperature can vary significantly for different compounds to be removed. However, the available vapour pressure curves provide the skilled person with appropriate guidance for suitable cleaning conditions without over-heating the fibres to more than 60% of their melting point.

[0040] Thus, for some materials, a reduction of the atmospheric pressure at the assembly site can lead to evaporation of the material, thereby enhancing the cleaning effect. This can be combined with heating the fibres to a cleaning temperature.

[0041] According to one embodiment, before fixing the plurality of metal fibres to each other, the method further comprises the step of subjecting the plurality of metal fibres to a predetermined pressure, in particular less than 1 MPa, in particular less than 500 kPa. With the method according to the present application, a relatively low pressure can be applied to ensure that the fibres establish contact points at the locations where they are connected to each other. With previously known methods, a comparably high pressure has to be applied to establish the contact points.

[0042] It is further preferred to provide a protective gas, such as argon, nitrogen, Ar-W5 (5 vol.-% H2 in Ar), Ar-W2 (2 vol.-% H2 in Ar), forming gas (5 vol.-% H2 in N2) or other inert gases at the assembly site to avoid oxidation of the fibres during assembly. This step can be related to the step of heating the plurality of fibres to a fixing temperature and the step of keeping the fibres at the fixing temperature, if performed. In general, the method according to the present application can also be performed in vacuum. Thus, the precise conditions at the assembly site can be chosen, e.g. depending on the material used for the fibres. For example, some materials, such as iron and / or some steels, cannot be used with nitrogen, as they tend to be nitrided. Thus, for such materials, other protective gases can be used.

[0043] According to another embodiment of the present application, the step of heating the fibers is performed by a suitable heating device. Preferred examples for such a heating device are induction furnaces, infrared furnaces, high-temperature ceramic heating elements and / or zone furnaces, such as a conveyor furnace. Such a heating device can ensure a fast heating, i.e. a high heating rate, as well as a fast cooling, i.e. a high cooling rate, so that the plurality of fibers can be connected to each other without releasing too much energy due to a rearrangement and relaxation process or any similar process before reaching the fixation temperature. The heating device can be a continuous furnace or a batch furnace.

[0044] In some embodiments, the suitable heating device for performing the heating step can be a continuous furnace. Such a continuous furnace is typically the preferred choice in applications with high production rates, i.e. when a large amount of fibers is assumed to be processed (industrial applications).

[0045] Preferably, the fixation temperature is determined in-situ by electron microscopy. This can be done, for example, by placing the fibers in an in-situ SEM (scanning electron microscope) heating stage. Due to the almost non-existing heat transfer in high vacuum, the fibers need a good thermal connection to the heating stage. For this, a heat-stable graphite paper can be used. Thus, one sheet can be used as a support between the fibers and the heating stage and another sheet with a hole in the middle can be used to observe the fibers. These fiber sandwiches can then be transferred to the heating stage and pressed down. Then, the heating stage can be heated to a temperature close to but still below the fiber melting temperature. At the same time, the fiber cross-section can be observed with the SEM until the fibers start to connect to each other. In this respect, the fixation temperature can be determined. In a second experiment, the fiber sandwich can be heated to the determined fixation temperature with a heating rate as described above. Then, the fixation temperature can optionally be kept constant for a fixation time until the desired degree of connection and thus the desired connection strength is reached. Thus, in other words, in the second experiment, the method steps according to the present application can be performed in order to check whether the determined fixation temperature is correct.

[0046] The fixation temperature and time depend on the material of the fibers, the dimensions of the fibers, i.e. the width and the thickness, and the amount of energy stored in the fibers. For example, for thin fibers of a given material with a small thickness and a wide rounding process, the rounding tends to occur faster compared to thicker fibers of the same material. It is possible to determine the fixation temperature and time for a specific type of fibers using the in-situ electron microscope described above or using trial and error tests. The trial and error tests can be performed using the actual equipment for producing the network of metal fibers, i.e. under actual manufacturing conditions.

[0047] In this regard, it is noted that the fixation temperature is preferably selected from the range of 80% to 98% of the melting temperature of the metal fibers, in particular from the range of 90% to 98%. Fixation temperatures within said range have proven to be suitable for most materials. In this regard, it is noted that the exact fixation temperature can also depend on the fixation time. That is, the higher the fixation temperature, the shorter the fixation time, and vice versa.

[0048] The cooling rate in the cooling step is preferably maintained for a time sufficient to cool the metal fibers below 60% of the melting temperature of the metal fibers.

[0049] The method steps according to the present application, i.e. the heating and cooling steps, optionally also the step of keeping the fibers at a fixation temperature, are performed within a combined time period of preferably less than 30 minutes, preferably less than 15 minutes, in particular less than 5 minutes, specifically less than 1 minute. It has been shown that the faster the method is performed, the less negative side effects occur, such as energy release, formation of shrinkage and / or breakage of the fibers and changes in the fiber cross-sectional shape. In this context, it is noted that the cooling step does not necessarily need to cool to room temperature. The step can be terminated after cooling to 60% of the melting temperature of the metal fibers.

[0050] In this regard, it is noted that the predetermined time period can be equally divided between the steps. In other embodiments, the step of keeping the fibers at a fixation temperature, if applied, can take longer than the steps of heating and cooling the fibers. In an ideal experiment, for example, the steps of heating and cooling the fibers can take 1 minute, while the step of keeping the fibers fixed can take 30 seconds or even less. In other experiments, however, the heating step can be performed within 1 to 5 minutes, the step of keeping the fibers at a fixation temperature within 0.5 to 1 minute and the cooling step can take 10 minutes, until the assembled network is cooled to a temperature of about 60% of the melting temperature of the fibers used. Further cooling of the assembled network to room temperature can take for example an additional 1 to 2 hours.

[0051] Preferably, the metal fibers comprise a length of 1.0 mm or more, and / or a width of 100 pm or less, and / or a thickness of 50 pm or less. With metal fibers having such dimensions, it is possible to produce a network of metal fibers that are fixed to each other without the need to heat the metal fibers for more than 30 minutes to temperatures close to their melting point. Conventional sintering techniques require temperatures close to or even slightly above the metal melting temperature to be maintained for a relatively long time. This would result in the material of the metal fibers to melt or at least to soften to such an extent that the metal fibers form a metal foil rather than a network, especially when a relatively high pressure is applied during sintering. Since the network of metal fibers is not a metal foil, the structure of the metal fibers used to manufacture the network of metal fibers can still be identified in the network of metal fibers. Thus, in a cross-sectional view of the network of metal fibers, there are voids between the metal fibers of the network fibers that are not part of the metal fibers.

[0052] According to the present application, preferably, the metal fibers show an exothermic event when heated in a DSC measurement before the metal fibers are fixed to each other, wherein the amount of energy released by the exothermic event is 0.1 kJ / g or more, more preferably 0.5 kJ / g or more, even more preferably 0.1 kJ / g or more, most preferably 1.5 kJ / g or more. The absolute amount depends to a large extent on the metal or metal alloy used. The extent of the exothermic event can be determined by comparing the DSC measurement of the metal fibers before and after thermal equilibration. In other words, the metal fibers showing such an exothermic event are not in their thermodynamic equilibrium at ambient temperature. During heating in the DSC measurement, the metal fibers can transform from a metastable state to a thermodynamically more stable condition, for example by crystallization, recrystallization or other relaxation processes reducing defects in the metal atomic lattice. For example, observed exothermic events for the metal fibers when heated during the DSC measurement indicate that the metal fibers are not in their thermodynamic equilibrium, for example, the metal fibers can be in an amorphous or nanocrystalline state containing defect energy and / or crystallization energy which is released during heating of the metal fibers due to the occurrence of crystallization or recrystallization. Such events can be identified, for example, using DSC measurements. It has been found that a network of metal fibers showing such an exothermic event has an improved strength after the metal fibers are fixed to each other.

[0053] According to another embodiment, the metal fibers comprise a non-circular cross-section, in particular a rectangular cross-section, a conic cross-section, a partially circular cross-section, or an elliptical cross-section with a major axis and a minor axis. Such cross-sections generally put the fibers out of their thermal equilibrium, i.e. in a metastable state, which will be beneficial for certain applications.

[0054] In this regard, it is noted that obviously the value of the minor axis has to be smaller than the value of the major axis. In case the minor axis comprises a larger value (i.e. a larger length) than the major axis, the definitions of "small" and "large" have to be simply interchanged.

[0055] Preferably, the ratio of the minor axis to the major axis is in the range of 1 to 0.05, preferably in the range of 0.7 to 0.1, in particular in the range of 0.5 to 0.1. Generally, the higher the ratio of the length of the minor axis and the major axis of the ellipse, the more the ellipse looks like a circle for which the ratio would be 1. The smaller the ratio, the more flattened, i.e. elliptical, it is. Thus, the ratio of the minor axis to the major axis is in particular smaller than 1.

[0056] Alternatively, the metal fiber can comprise a circular cross-section. For such a cross-section, the ratio of the “large” axis to the “small” axis is obviously exactly 1. The circular cross-section comprises an energetically more preferred state, said cross-section comprising an aspect ratio of less than 1. Thus, the fiber having a circular cross-section is energetically closer to its equilibrium state than a fiber having a cross-section of another shape.

[0057] According to another embodiment of the present application, the molten material of the metal fiber is subjected to a cooling rate of 10 2 K min -1 The metal fiber can be obtained in particular by a vertical or horizontal melt spinning process. Due to the fast cooling applied during the melt spinning process, such metal fiber produced by the melt spinning process can comprise spatially confined regions in a high energy state, i.e. in a metastable state. The fast cooling in this respect refers to a cooling rate of 10 2 K min -1 or more, preferably 10 4 K min -1 or more, more preferably 10 5 K min -1 or more.

[0058] Moreover, the fiber obtained by the melt spinning process typically comprises a rectangular or semi-elliptical cross-section, which is preferred for certain fields of application as they are far from their equilibrium state. Examples of melt spinning machines with which such fibers are produced can be known, for example, from the not yet published international application PCT / EP2020 / 063026 and the published applications WO2016 / 020493A1 and WO2017 / 042155A1, which are incorporated herein by reference.

[0059] According to another example, at least some of the plurality of metal fibers are amorphous, or at least some of the plurality of metal fibers are nanocrystalline. Nanocrystalline metal fibers contain crystalline domains. At temperatures heated to about 20-60% of the melting temperature of the nanocrystalline metal fibers, these domains undergo recrystallization, resulting in an increase in the average size of the crystalline domains compared to the average size of the initial crystalline domains in the nanocrystalline metal fibers before heating. It is also possible to mix non-equilibrium (e.g. nanocrystalline or amorphous fibers) with equilibrium (e.g. annealed) fibers.

[0060] In some applications, it is preferred that the metal fibers are in electrical contact with each other. This will be preferred if the assembled network is intended for electrochemical applications, such as batteries, fuel cells or any similar applications.

[0061] According to one embodiment, the metal fibers are in direct electrical contact with each other, such that the electrical conductivity can be increased to a maximum. In this respect, it is particularly preferred that all metal fibers are sintered to other metal fibers, most preferably directly sintered to other metal fibers, without the need for additional binders, such as polymeric binders. It is thus further preferred that the metal fibers are fixed to each other without polymeric binders, as such polymeric binders often have poor electrical conductivity and high temperature performance.

[0062] Preferably, the metal fibers comprise at least one of copper, silver, gold, nickel, palladium, platinum, cobalt, iron, chromium, vanadium, titanium, aluminum, silicon, lithium, manganese, boron, combinations of the foregoing, and alloys comprising one or more of the foregoing, such as CuSn8, CuSi4, AlSi1, Ni, stainless steel, Cu, Al or Vitrovac alloys. Vitrovac alloys are iron- and cobalt-based amorphous alloys. It is particularly preferred that the metal fibers are made of copper or aluminum or stainless steel alloys. Different types of metal fibers can be combined with each other, such that the filter can comprise metal fibers made of, for example, copper, one or more stainless steel alloys and / or aluminum. The network made of metal fibers is particularly preferred, wherein the metal fibers are made of copper, aluminum, cobalt, stainless steel alloys containing copper aluminum silicon and / or cobalt.

[0063] According to another aspect of the present application, there is provided a network of metal fibers, wherein the network comprises a plurality of metal fibers fixed to each other at contact points, and wherein the metal fibers comprise a non-circular cross-section, such as a rectangular cross-section, a conic cross-section, a partially circular cross-section or an elliptical cross-section with a large and a small axis, or wherein the metal fibers comprise a circular cross-section. The fibers further comprise a width that is substantially constant along the length of the fiber, such that the variation in the width of the fiber along its length is preferably less than 40%, more preferably less than 30% or even more preferably less than 20%.

[0064] Preferably, the width of the fibers varies less than 20%, more preferably less than 10%, even more preferably less than 5%, or most preferably less than 1% along their length. The variation in fiber width here refers to a comparison of the fiber width before and after the fibers have sintered to each other.

[0065] In a conventionally known network, the fibers typically comprise a random shape, such that it cannot be ensured that the variation in width of the individual fibers along their length is not large. For example, if a fiber would comprise a large variation in its width, it is possible that the fiber would break at the part comprising the smaller width, i.e. the fiber has a constriction and then becomes a break. On the other hand, fibers having a (nearly) constant width have the advantage that the individual fibers can be connected to each other at any given point along their length without the risk of a break occurring in the process.

[0066] In this regard, it is noted that preferably, the width of the metal fibers is substantially constant, i.e. the variation in width of the fibers along their length is preferably less than 40%, more preferably less than 30%, or even more preferably less than 20%. As mentioned above, when the metal fibers are heated at a low heating rate, a rearrangement process occurs on an atomic level in order to reach an energy level closer to its equilibrium state. This sometimes even leads to a change in fiber shape, as a perfect sphere would be the most preferred state. When this shape change starts to occur, it is possible that the fibers start to disintegrate by building up a constriction that can cause the fiber to break. Eventually, when heated too long, the fibers transform into metal droplets. The method according to the present invention makes use of a fast heating and cooling rate, and a reduced fixation time, if applied. In the resulting network of fixed metal fibers, these fibers are substantially free of such breaks, thereby maintaining the length of the fibers. Furthermore, due to the high heating and cooling rate, the shape change of the fiber cross-section can be avoided, i.e. there is a kinetic control on the fiber shape. Thus, the method of the present invention provides a high precision control on the fiber shape.

[0067] Preferably, the fibers in the plurality of fibers can sinter to each other, more preferably directly sinter to each other. This ensures that no additional frame or anything similar is needed to hold the fibers together. Furthermore, by sintering the metal fibers directly to each other, the connection points are electrically conductive. This provides a relatively low internal resistance for the network of metal fibers.

[0068] Preferably, the ratio of the small axis to the large axis is in the range of 1 to 0.05, preferably in the range of 0.7 to 0.1, in particular in the range of 0.5 to 0.1. As already mentioned above, fibers having a more flattened cross-section are more far away from their equilibrium state in terms of energy, such that they can store more energy compared to fibers comprising e.g. a circular cross-section.

[0069] In fact, by using fibers with a high or low ratio of the length of the small axis and the length of the large axis, as described above, it is possible to select a characteristic network depending on the application of the network. Thus, by using fibers with a low ratio, the mechanical stability of the network and the weight are reduced, while by using fibers with a high ratio, the mechanical stability of the network and the weight are increased. This can be selected depending on the application where the characteristics are more important. Due to the dynamics control provided by the present invention, the shape of the fibers is substantially maintained, i.e. the aspect ratio of the fibers is substantially maintained. As a result, by starting with fibers having the desired final shape, the network characteristics can be easily adjusted.

[0070] According to one embodiment, the network is an ordered network or a disordered network. Such a disordered network has, for example, good electrical conductivity and anisotropic fluid properties in all directions. Furthermore, it is easier to produce a disordered network of metal fibers than an ordered network of fibers. However, in some applications, it is preferred that the fibers in the network are combed in different directions to provide directionality of the individual fibers. Thus, it can be preferred that in the network some or all of the fibers have an orientation, i.e. the length of the fibers is not randomly oriented but has a predominant orientation in one or more spatial directions. By having a predominant orientation of the metal fibers, the filter can have isotropic fluid properties.

[0071] According to another embodiment of the present invention, the network has open pores between the metal fibers of the plurality of metal fibers. The porosity of the network is preferably up to 95 vol.%. More preferably, the porosity of the network is greater than 80 vol.%. Even more preferably, the porosity is in the range of 80 vol.% to 95 vol.%. An active material can be incorporated into the open pores, for example an active electrode material or an active catalyst material. It is further preferred that in the network according to the present invention at least some of the plurality of metal fibers are at least partially coated. The coating can be, for example, an active material, for example an electrode active material that interacts with lithium ions in a battery, or a catalytically active material that covers CO to CO2 or is active in hydrolysis. The coating can also be applied on the metal fibers, which improves the fixation of the metal fibers to each other, thereby increasing the mechanical strength of the network. The porosity can be determined using a micro computed tomography machine to reproduce the network structure and then using the bubble point method described below to assess the porosity.

[0072] For example, such active electrode materials for batteries are: for the anode: graphite, silicon, silicon carbide (SiC) and tin oxide (SnO), tin dioxide (SnO2), and lithium-titanium dioxide (LTO); for the cathode: lithium-nickel-manganese-cobalt-oxide (NMC), lithium-nickel-cobalt-aluminum-oxide (NCA), lithium-cobalt-oxide (LiCoO2), and lithium-iron-phosphate (LFP).

[0073] The network can comprise a mean pore size selected in the range of 0.1 pm to 100 pm, preferably in the range of 0.5 pm to 50 pm, in particular in the range of 1 pm to 10 pm. The mean pore size can be determined using micro computed tomography to reproduce the fiber structure and then using the bubble point method to assess the mean pore size. This bubble point method determines the largest sphere diameter that would fit between two fibers to be considered as a pore size. In more detail, a point is set in the center between two fibers and the radius of a bubble is increased, centered on this point, until it comes into contact with the surface of both fibers. The diameter of the bubble corresponds to the pore size. If at any given parameter the bubble diameter only contacts one fiber, the center point is shifted towards the fiber that the bubble does not contact.

[0074] It is particularly preferred that the network of metal fibers according to the present application are fixed to each other, in particular directly, at contact points that are randomly distributed throughout the network of metal fibers. According to another aspect of the present application, it is preferred that the contact points are not randomly distributed, but are arranged for example in a peripheral region of the network of metal fibers, or the metal fibers are ordered such that the contact points are also ordered. It is further preferred that the contact points that fix the metal fibers to each other are positioned in specific regions and are not evenly arranged over the entire network of metal fibers. As the contact points that fix the metal fibers to each other only exist in separate regions, the fibers between these regions can have a high flexibility, while at the same time ensuring mechanical stability and good electrical conductivity.

[0075] The thickness of the network of the present application is not particularly limited. However, it is preferred that the network has a thickness of 0.01 mm or more. More preferably, the thickness of the network is 0.03 mm or more, even more preferably 0.05 mm or more, even more preferably 0.07 mm or more, most preferably 0.1 mm or more. If the thickness of the network is less than 0.01 mm, there is a risk of insufficient mechanical stability of the network. The upper limit for the thickness of the network is not particularly limited. However, depending on the application, the upper limit can be 3.0 mm or less, or 2.5 mm or less. For battery applications, a preferred thickness of the network is in the range of 0.1 mm to 0.5 mm. A network having a thickness in this range is advantageous for the stacking and rolling of the network coated with active material for the production of a battery. Other preferred thickness ranges are in the range of more than 0.5 mm to 5 mm, more preferably in the range of 1 mm to 3 mm.

[0076] According to another aspect of the present application, a network of metal fibers is provided, wherein the network can be for example a network according to the present application, which can be obtained by a method according to the present application. BRIEF DESCRIPTION OF DRAWINGS

[0077] The present invention and the various examples of the network and method of the present invention will now be described in further detail, by way of example only, with reference to the accompanying drawings and pictures. Shown in the drawings are:

[0078] Figure 1 : exemplary scheme illustrating a typical process that can occur during sintering;

[0079] Figure 2 : video frames showing sintering and relaxation of CuSi4;

[0080] Figure 3 : video frames showing sintering and relaxation of AlSi1 ;

[0081] Figure 4 : micrograph of a fiber network of conventional sintering;

[0082] Figure 5 : micrograph of sintered CuSi4 fibers with a flattened shape;

[0083] Figure 6 : micrograph of sintered AlSi1 fibers with a flattened shape; and

[0084] Figure 7 : micrograph of cleaned and sintered AlSi1 fibers with a flattened shape. DETAILED DESCRIPTION

[0085] Figure 1 A typical diffusion process that leads to relaxation during sintering is shown. Different arrows show surface diffusion 1, lattice diffusion (from the surface) 2, evaporation and condensation 3, grain boundary diffusion 4, lattice diffusion (from the boundary region) 5, and volume diffusion 6. While processes 1 to 3 do not lead to shrinkage and only connect the fibers to each other, processes 4 to 6 remove material from the boundary region and deposit it on the sintering necks. The reason for this is, as already explained above, the reduction of the fiber surface and the corresponding reduction of its free energy ΔG.

[0086] By the method of the present invention, only the boundary regions of the metal fibers are thermally activated, so that the fibers 10 sinter together, but the entire fiber 10 is not rounded in order to preserve the fiber shape and size, which will be associated with an enlarged surface that provides beneficial properties for many applications, such as electrochemical applications or filtration applications.

[0087] According to the method of the present invention, a loose network of fibers 10 is provided at assembly sites 12. Then, the fibers 10 are fixed to each other by forming contact points 14 between the individual fibers 10. In order to create the contact points 14, the method according to the present invention provides three steps:

[0088] A. The plurality of fibres 10 is heated to a fixed temperature selected from the range of 50% to 98% of the melting point temperature of the fibres 10 at a heating rate higher than 50 K / min, in particular higher than 100 K / min, in particular higher than 200 K / min, preferably higher than 1000 K / min.

[0089] B. Then, optionally, the fixed temperature can be maintained for a fixed time selected from the range of 0 seconds to 30 minutes, in particular from the range of 0 seconds to 15 minutes, preferably from the range of 0 seconds to 5 minutes.

[0090] C. Finally, the plurality of fibres 10 is cooled at a cooling rate higher than 50 K / min, preferably higher than 100 K / min, in particular to a temperature lower than 60% of its melting point.

[0091] Additionally, before performing the three steps A, B and C, a pressure can be applied to the fibres 10 to ensure that the individual fibres 10 are in contact with each other. The pressure can be relatively low (i.e. in the range of 0.05 GPa to 1 GPa) and serves to form a contact between the unconnected metal fibres. It is not necessary to maintain the pressure while performing steps A, B and C, i.e. it is sufficient to briefly compress the fibres by applying the pressure only before performing steps A, B and C, but not during the performance of steps A, B and C. Preferably, no external pressure is applied during steps A, B and C. By avoiding external pressure, the risk of the metal fibres transforming into a metal foil can be avoided, in particular when operating at a fixed temperature close to the melting temperature.

[0092] Further, before performing the above steps A, B and C, an additional cleaning step can be performed which comprises heating the plurality of fibres 10 to a cleaning temperature such that the attachments and / or impurities that can be present on the surface of the fibres are decomposed (i.e. evaporated or combusted), thereby resulting in a clean surface of the fibres 10. The cleaning step is explained in more detail below in connection with Figure 7 The step of cleaning the fibres 10 is further explained.

[0093] It can be seen that steps A, B and C are performed much faster compared to the conventionally known methods. The longest time to perform all three steps can be defined to be less than 45 minutes, even in the range of about 15 minutes. It has been shown that with the method according to the present application, times below 5 minutes, even below 1 minute are possible. This is far below the times for sintering which usually take up to several hours.

[0094] To be able to realize such a short time frame, step A, optionally also steps B and C are performed with a furnace or other heating device configured to provide a high heating and cooling rate (e.g. induction furnace, infrared furnace, high temperature ceramic heating elements and / or zone furnaces, such as a conveyor furnace (not shown in the figures)).

[0095] Of particular interest is to heat the fibers 10, which are to be considered connected to each other, to a precise fixed temperature in the range of 50% to 98%, in particular in the range of 80% to 98%, more particularly in the range of 90% to 98% of the melting temperature of said fibers 10. The precise fixed temperature depends on the material used for the fibers 10 (see also Tables 1 and 2 below). Selecting the right fixed temperature allows to connect the fibers 10 to each other without starting to change their shape (i.e. to round) due to the above mentioned relaxation process or without starting to melt them.

[0096] To be able to determine said fixed temperature and time, trial and error experiments and / or electron microscopy can be performed on a sample of the actual metal fibers. For electron microscopy, the fibers are placed in a SEM (scanning electron microscope) in situ heating stage. For this, the fibers need a good thermal connection to the heating stage, since there is almost no heat transfer in a high vacuum. Therefore, heat stable graphite paper can be used: for example, one sheet as a support between the fibers and the heating stage, and another sheet with a hole in the middle to observe the fibers. This fiber sandwich is then transferred to the heating stage and pressed down. The heating stage is then heated to a temperature close to the melting temperature. The fiber cross section is then observed with the SEM until the fibers start to connect to each other. In this way the fixed temperature is determined. In a second experiment, at least steps A and C described above are performed until the desired degree of connection is reached, so that the desired connection strength is reached. For this trial and error experiment, an amount of fibers is placed in a rapid heating furnace. To achieve contact points between the fibers, the network can be pressed together or placed on a plate with space holders and a cover plate. After removing the air / oxygen in the furnace and setting the test atmosphere, the furnace is heated to the possible (i.e. determined) fixed temperature and kept for a certain time (which can be a fixed time). Depending on the result of the fibers, for example depending on whether the fibers connect to each other and / or whether the fibers change their shape, the parameters have to be adjusted. In this regard, it is noted that three possible results can be expected: 1) the fibers are not sintered, 2) the fibers are sintered but rounded, or 3) the fibers are not sintered but rounded. For the first result, the fixed temperature and / or the fixed time should be increased. For the second result, the fixed temperature and / or the fixed time should be decreased, for the third result, the heating rate should be increased, and the fixed temperature and / or the fixed time should be decreased.

[0097] For some materials, it is also beneficial if a protective gas (e.g. argon, nitrogen Ar-W5 (5 vol.-% H2 in Ar), Ar-W2 (2 vol.-% H2 in Ar), forming gas (5 vol.-% H2 in N2) or other inert gas) is provided at the assembly site 12 to prevent the metal fibers from being oxidized. Whether such a protective gas needs to be provided can be selected depending on the material(s) of the fibers 10.

[0098] The contact points 14 of the assembled network can be distributed in an ordered or disordered manner throughout the network, depending on the application of the assembled network, and fix the fibers to each other. Furthermore, the amount of contact points 14 can be selected depending on the application of the network by subjecting the fibers 10 to a higher or lower pressure before at least performing step A and step C, thereby creating more or less contact points 14. The fiber density (i.e. the amount of fibers per volume) and / or the fineness of the fibers can also be used to adjust the number of contact points 14.

[0099] The contact points 14 also enable electrical conductivity throughout the assembled network. Therefore, a large number of contact points 14 is advantageous for applications requiring a high electrical conductivity network. On the other hand, for filters, it will not be important how many contact points 14 are provided throughout the network as long as the contact points 14 still hold all fibers 10 together.

[0100] The fibers 10 used to assemble the network according to the present application comprise a length of 1.0 mm or more and / or a width of 100 pm or less and / or a thickness of 50 pm or less (see Figures 2 to 6 ). Such fibers can for example be produced by the so-called vertical or horizontal melt spinning process, which is described in the documents PCT / 10 / 063026 (not yet published), WO2016 / 020493A1 and WO2017 / 042155A1. These fibers 10 typically have an elliptical, rectangular or flat cross-section. Furthermore, the fibers 10 produced by the melt spinning process typically store a large amount of energy.

[0101] In order to better understand the method according to the present application, several experiments have been carried out, which are described below in connection with Figures 2 to 7 .

[0102] Copper alloy fibers (CuSi4 (4 wt.-% Si and 96 wt.-% Cu) and AlSi1 (1 % (weight) Si and 99 % (weight) Al)) were sintered together while maintaining the flat ribbon structure of the fibers. In order to systematically examine these processes, the fibers were heated in an electron microscope at a heating rate of 10 K / min and a video was recorded. Figure 2A single frame from the CuSi4 video is shown at a special point such as the beginning of sintering (to see the sharp transition 14 between the fibers 10 blurred, left) and the point at which the fibers 10 start to round, thus forming a constriction 15 and a break 16 to disintegrate the fibers 10 (right). The corresponding temperatures are very close to each other, which is why the highest temperature accuracy and control are needed to obtain good results.

[0103] Figure 3 A video frame taken using AlSi1 fibers under the same conditions as the video of Figure 2 is shown. Figure 3 The same characteristic points are shown as in Figure 2 , namely the beginning of sintering and the beginning of the rounding process. It can be observed that the fibers sinter together at approximately 602°C, while at 624°C they sinter together. Furthermore, between 602°C and 624°C, the fibers transform from flat ribbon fibers to fibers with a circular cross section. This can be identified by the fact that the fibers at 624°C are thinner than the fibers at 602°C. It is important to note that the frames of the videos as shown in Figure 2 and Figure 3 were recorded under low heating conditions (10 K / min) in high vacuum conditions. These conditions are different from the conditions of the present invention. This is the reason for the values shown in Figure 2 and Figure 3 to be different from the values described in the table below. However, Figure 2 and Figure 3 show that it is difficult to sinter fibers under conventional methods that employ lower heating rates.

[0104] For conventional thermal sintering using a resistance heating furnace, the fibres 10 are heated at a rate of 10 K / min - 20 K / min, i.e. relatively slowly. During this time, the fibres 10 undergo a so-called relaxation process and the energy stored in these fibres during their production, for example by means of a melt spinning process, is slowly released and is no longer available for forming the contact points between the metal fibres. The release of the energy stored during the slow heating not only influences the mechanical properties of the fibres 10, but also increases the energy requirement during the actual sintering, since the fibres 10 are no longer in their thermodynamic disequilibrium after production. For this reason, untreated fibres 10 obtained from a melt spinning process and fibres 10 for comparison which have been tempered at 300°C for one hour are brought to the sintering temperature in a rapid heating furnace, here an infrared furnace, within 1 minute. This temperature is maintained for 1 minute and then the fibres are cooled as quickly as possible (from the sintering temperature to below 600°C in less than 30 seconds). Possible heating devices other than infrared heaters are for example ceramic heaters or induction heaters. The very short treatment time of only 1 minute or less is sufficient for the fibres 10 to sinter to one another at the contact points 14, but the energy and time are insufficient for the fibres 10 to transform into the thermodynamically favourable rounded shape. This is not possible when applying conventional heating and cooling rates, which take a very long time to reach the target temperature (from the sintering temperature to below 600°C in several hours). Applying conventional heating and cooling rates still makes it possible for the fibres 10 to sinter to one another. However, the sintered fibres subsequently adapt to the ideal rounded shape and are damaged due to the constriction 15 or even possibly a break 16, for example at the twist point. Due to the very long diffusion paths when the fibres round, high temperatures and / or long times are necessary for the transformation into the thermodynamically favourable rounded shape. This can be avoided by using fibres which contain stored energy, for example produced by melt spinning. The stored energy can be measured for example by means of a DSC measurement, in which the stored energy can be observed in the form of an exothermic event.

[0105] It was also tested how long the fibres made of AlSi1 and CuSi4 respectively have to be heated at a certain temperature until they reach the ideal rounded shape. The fibres of AlSi1 have a ribbon-like structure with an average length of 30 mm, an average width of 75 pm and an average thickness of 15 pm. The fibres of CuSi4 have a ribbon-like structure with an average length of 20 mm, an average width of 35 pm and an average thickness of 7 pm. For these tests, the fibres were heated to a certain fixed temperature shown in the table below within 1 minute. The fixed temperature was maintained for a certain period of time before rapid natural cooling, for CuSi4 from 30 seconds to about 500°C and about 20 minutes to room temperature and for AlSi1 from 30 seconds to about 330°C and 15 minutes to room temperature. After cooling, it was checked whether the fibres had a rounded cross-section. The experiment was repeated at the respective fixed temperature with increasing fixed time. The results of these tests are shown in the table below:

[0106]

[0107] It can be clearly seen that the higher the chosen fixing temperature, the shorter the time spent to sinter the fibres 10 to each other without the fibres 10 changing their external shape. Moreover, it can be seen that the temperature depends significantly on the material the fibres 10 are made of. Furthermore, the fibre size, in particular the thickness and width, has a certain influence on the speed at which the cross-sectional shape of the fibres is transformed from flat to round. The above experiments demonstrate how a person skilled in the art can determine suitable conditions by simple trial and error for each fibre material.

[0108] Although different conditions can be required to fix the metal fibres to each other depending on their material and / or size, the above experimental study demonstrates that the fibres 10 can be connected to a network without the fibres 10 changing their length, shape and / or diameter if the time to sinter the fibres 10 to each other is reduced to a minimum.

[0109] Figure 4 A regular sintered network of previously flat fibres 10 is shown. The formation of the constriction 15 and the break 16 of the fibres 10 can be clearly seen. Moreover, the cross-section of the fibres is transformed from flat to round. The formation of the sintering neck, i.e. the constriction, corresponds to the Figure 1 the current theory.

[0110] Table 1 shows the sintering temperature (in each case for a holding time of 1 minute) for CuSi4 fibres 10 which were not heat treated (obtained from melt spinning) and tempered (1 hour at 300°C under an argon atmosphere), wherein the ribbon-like structures have an average length of 20 mm, an average width of 35 pm and an average thickness of 7 pm. Table 2 similarly shows the same cases for AlSi1 fibres 10 with an average length of 30 mm, an average width of 75 pm and an average thickness of 15 pm. The comparison of the untreated fibres and the tempered fibres was carried out using a tube furnace under a protective gas atmosphere (argon) which provides a heating rate of 10 K / min. It was found that for the tempered CuSi4 fibres 10 a temperature of at least 950°C and a holding time of at least one hour are necessary in order to sinter the fibres 10 together. After sintering, the previously tempered fibres 10 are almost completely rounded and in some cases the length is strictly limited by the constrictions 15 and breaks 16. In contrast, the sintering of the not heat treated CuSi4 fibres starts at a significantly lower temperature (between 890°C and 910°C) compared to the tempered fibres (sintering starts above 950°C) and is completed within 0.5 to 5 minutes for the CuSi4 fibres and within 0.5 to 5 minutes for the AlSi1 fibres, depending on the fixing temperature and the lower fixing temperature which requires a longer fixing time.

[0111] Table 1 : CuSi4 sintering parameters

[0112] Fixed temperature °C Untreated fibre Tempered fibre 890 Unsintered Unsintered 910 Sintered Unsintered 930 Sintered Unsintered 950 Sintered Unsintered; rounding begins 970 Sintered; rounding begins Sintered; rounding

[0113] Table 2: AlSi1 sintering parameters.

[0114] Temperature °C Untreated fibre Tempered fibre 620 Unsintered Unsintered 630 Sintered Unsintered 640 Sintered Sintered; rounding begins 660 Sintered Sintered; rounding

[0115] Contrast experiments with relaxed fibers 10 (heat treated at 300°C under protective gas for 1 hour, shape did not change, only defects were degraded and stored energy was released) show that the sintering described here is not possible or only at higher temperatures compared to untreated fibers 10. For relaxed fibers the temperature window between the onset of sintering and the change of fiber shape is very narrow. However, when using fibers with stored energy (e.g. fibers showing an exothermic signal during DSC measurement) the temperature window for sintering the fibers to each other without rounding is much wider. With the slow heating and cooling rates known from sintering processes the fibers 10 undergo a relaxation process before the sintering temperature can be reached, the stored energy is released too early and therefore cannot be used to drive the sintering process. When applying low heating rates the fibers are tempered before the sintering temperature is reached. Therefore their behavior will be similar to the tempered fibers reported in table 1 and table 2. The higher the energy stored in the fibers 10 during the manufacturing process the lower the required sintering temperature and time.

[0116] The maximum possible energy can be introduced by high quenching rates (e.g. by known melt spinning processes). Due to the underlying mechanisms of this process the method according to the present invention can be transferred to almost all metals, metal-inorganic and comparable alloys and materials as long as enough energy can be stored in them.

[0117] Figure 5 CuSi4 fibers are shown which have been sintered with the method according to the present invention so that they still keep their original flat shape. Figure 6 Similar AlSi1 fibers 10 are shown. It can be clearly seen that Figure 5 and Figure 6 the fibers 10 in table 2 include a nearly constant width along their length, i.e. no constrictions 15 or breaks 16 are formed which could lead to a complete destruction of the fibers 10. That is, the width of the fibers 10 in the network according to the present invention varies by no more than 40%, preferably even by no more than 30%.

[0118] Finally, Figure 7 a network of sintered AlSi1 fibers with flat cross section is shown and the surface of this network has been cleaned with the cleaning step described above before the fibers 10 have been heated to a fixed temperature. From Figure 7 it can be seen that the fibers 10 in Figure 6The surface of the fiber 10 looks smoother compared to the AlSi1 fiber shown. It can thus be concluded that the surface of the fiber 10 has been cleaned from previously existing impurities. Such impurities usually arise with secondary products of the manufacturing process of the fiber 10 and cannot be avoided. Therefore, in order to obtain a good sintering result, an additional step of cleaning the fiber 10 by heating the fiber 10 to a temperature in the range of 40% to 60% of its melting point can be preferred.

[0119] In order to show the effectiveness of the additional step of cleaning the fiber 10, three experiments have been conducted to prove the advantages of this cleaning step.

[0120] Experiment 1 : Sintering (melt spinning) of metal fibers in a ceramic heating element oven

[0121] For experiment 1, a self-constructed oven comprising ceramic heating elements was used. A plurality of fibers 10 made of an aluminum-silicon alloy (1 wt.-% Si in Al) and produced by a conventional melt spinning process (as described above) were placed on the heated surface of the oven. The heated surface then heated the fibers to a temperature of 640°C within four minutes, which correlates to an average heating rate of about 155 K / min. It is also to be noted at this point that conventional ovens start heating at a higher heating rate and tend to reduce the heating rate once a higher temperature is reached.

[0122] After heating the fibers 10 to the fixed temperature of 640°C as described above, the temperature was kept for 10 seconds, 20 seconds, 30 seconds and 60 seconds, respectively. Thus, all fibers 10 were connected to each other, i.e. sintered, regardless of the duration of time the fibers 10 were kept at the fixed temperature.

[0123] The following cooling step was naturally performed, i.e. the fibers 10 were cooled without external interference. After about 1 minute, the temperature of the fibers 10 was already below 500°C and after about 5 minutes, the temperature of the fibers 10 was already below 300°C, which correlates to an average cooling rate of about 68 K / min to cool the fibers from 640°C to 300°C.

[0124] It can be seen from the results that the fibers 10 sintered to each other without changing the cross section during the process.

[0125] Experiment 1 as described above was further performed under a protective gas, i.e. Ar, to prevent oxidation of the fibers 10.

[0126] Experiment 2: Sintering of metal (extractor wheel) fibers in a ceramic heating element oven

[0127] This experiment 2 was performed in the same way as experiment 1. However, even though the fibers 10 used were made of the same alloy (i.e. AlSi1), they were manufactured using the so-called extraction wheel method. Said extraction wheel method is a conventional method for preparing (metal) fibers (see for example Cramer, A., et al. Tailored magnetic fields in the melt extraction of metallic filaments. Metallurgical and Materials Transactions B, 2009. 40(3): p. 337-344 or Park, M.H., Y.S. Song, and J.H. A Study on the Fabrication of Metal Fiber by Fine Melt Extraction Process. in Advanced Materials Research. 2007. Trans Tech Publ.). Due to the use of the extraction wheel method, the fibers 10 produced therewith are thicker, i.e. not as thin as the fibers produced by the melt spinning process. Therefore, an additional cover plate has been placed on top of the fibers 10 so that the plurality of fibers 10 can be better connected to each other.

[0128] After the heating and cooling steps under the same conditions as described above, it can be seen that these sintered fibers 10 also do not change their cross-section after the sintering process, so that they still comprise their half-moon shaped cross-section (which is caused by the production method) after being connected to each other.

[0129] Experiment 3: Sintering of metal (melt spun) fibers in a ceramic heating element oven and performing an additional cleaning process

[0130] This experiment 3 was performed in the same way as experiment 1. However, before being sintered to each other, the fibers 10 were additionally heated to a temperature of about 400°C to prove that impurities and / or attachments can remain on the surface of the fibers 10 during the production of the fibers 10. For example, thin paraffin oil as well as PVA (polyvinyl alcohol) were used. For both cases, the fibers 10 were held at the cleaning temperature described above for about 5 minutes so that the attachments present on the surface decompose and can be removed with the air flow applied at the assembly sites 12 of the fibers 10.

[0131] After removal of the decomposed (i.e. evaporated / decomposed / burned) attachments, the fibers are heated from 400°C to a holding temperature of 640°C within about 2 min, i.e. with an average heating rate of about 70 K / min. As in experiment 1, the holding temperature has been maintained for a few seconds, then cooled again with an average cooling rate of about 68 K / min.

[0132] It can again be demonstrated that the fibers still have the same cross-section as before and no residues on their surface. This can be seen in Figure 7 , Figure 7 shows the sintered fibers 10, which still have a flat cross-section and show no residues of paraffin oil and / or PVA on their surface.

[0133] Thus, after comparing the three experiments 1 to 3 described above, it can be seen that, in general, no additional cleaning step is required to be able to sinter the fibers to each other without changing their cross-section. However, the additional cleaning step helps to improve the sintering quality, since the resulting clean surfaces can connect to each other better than compared to contaminated surfaces.

Claims

1. A method of assembling a network of fibers, the network of fibers comprising a plurality of metal fibers (10), wherein the method comprises the steps of: - providing a loose network consisting of the plurality of metal fibers (10) at an assembly site (12); - fixing the plurality of metal fibers (10) to each other by forming contact points (14) between individual metal fibers (10) by: heating the plurality of metal fibers (10) to a fixation temperature at a heating rate higher than 50 K / min, the fixation temperature being selected from a range of 50% - 98% of the melting temperature of the plurality of metal fibers (10); and cooling the plurality of metal fibers (10) at a cooling rate higher than 50 K / min; wherein the cooling step is started directly after reaching the fixation temperature during the heating step.

2. The method of assembling a fiber network of claim 1, wherein, heating the plurality of metal fibers (10) to a fixation temperature at a heating rate higher than 100 K / min.

3. The method of assembling a fiber network of claim 1, wherein, heating the plurality of metal fibers (10) to a fixation temperature at a heating rate higher than 200 K / min.

4. The method of assembling a fiber network of claim 1, wherein, heating the plurality of metal fibers (10) to a fixation temperature at a heating rate of 1000 K / min.

5. The method of assembling a fiber network of claim 1, wherein, cooling the plurality of metal fibers (10) at a cooling rate higher than 100 K / min.

6. The method of assembling a fiber network according to any one of claims 1 to 5, wherein, comprising a further step performed before the step of fixing the plurality of fibers to each other, wherein the further step comprises cleaning the plurality of fibers by heating the plurality of fibers to a cleaning temperature, the cleaning temperature being selected from a range of 20% to 60% of the melting temperature of the fibers.

7. The method of assembling a fiber network of claim 6, wherein, the cleaning temperature is selected from a range from room temperature to 60% of the melting temperature of the fibers.

8. The method of assembling a fiber network of claim 6, wherein, the step of cleaning the plurality of fibers comprises applying a gas flow at the assembly site.

9. The method of assembling a fiber network of claim 6, wherein, the step of cleaning the plurality of fibers comprises reducing the atmospheric pressure at the assembly site to a pressure lower than 80 kPa.

10. The method of assembling a fiber network of claim 6, wherein, the step of cleaning the plurality of fibers comprises reducing the atmospheric pressure at the assembly site to a pressure lower than 50 kPa.

11. The method of assembling a fiber network of claim 6, wherein, the step of cleaning the plurality of fibers comprises reducing the atmospheric pressure at the assembly site to a pressure lower than 10 kPa.

12. The method of assembling a fiber network of claim 6, wherein, the step of cleaning the plurality of fibers comprises reducing the atmospheric pressure at the assembly site to a pressure lower than 1 kPa.

13. The method of assembling a fiber network of claim 6, wherein, the step of cleaning the plurality of fibers comprises reducing the atmospheric pressure at the assembly site to a pressure lower than 0.1 kPa.

14. The method of assembling a fiber network of claim 6, wherein, the step of cleaning the plurality of fibers comprises reducing the atmospheric pressure at the assembly site to a pressure lower than 0.0001 kPa or less.

15. The method of assembling a fiber network of claim 6, wherein, the further step comprises determining a compound to be removed and selecting a reduced pressure and / or an elevated temperature based on a vapor pressure curve of the compound to be removed.

16. The method of assembling a fiber network of claim 15, wherein, the further step further comprises reducing the pressure and / or elevating the temperature stepwise or in a continuous manner based on the vapor pressure curve.

17. The method of assembling a fiber network according to any one of claims 1 to 5, wherein, before fixing the plurality of metal fibers (10) to each other, the method further comprises a step of subjecting the plurality of metal fibers (10) to a predetermined pressure, the predetermined pressure being less than 1 GPa.

18. The method of assembling a fiber network according to any one of claims 1 to 5, wherein, providing a protective gas at the assembly site.

19. The method of assembling a fiber network of claim 18, wherein, The protective gas comprises argon, Ar-W5, Ar-W2, forming gas of N2 or other inert gas; wherein the Ar-W5 is Ar containing 5 vol-% H2, the Ar-W2 is Ar containing 2 vol-% H2, and the forming gas is N2 containing 5 vol-% H2.

20. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The step of heating the metal fibers (10) is performed by an induction furnace, an infrared furnace, a high-temperature ceramic heating element, and / or a zone furnace.

21. The method of assembling a fiber network of claim 20, wherein, The zone furnace is a conveyor furnace.

22. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, The heating step is performed by a continuous furnace.

23. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, The fixed temperature is determined in situ by electron microscopy.

24. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The fixed temperature is selected from the range of 80% to 98% of the melting point temperature of the metal fibers.

25. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The fixed temperature is selected from the range of 90% to 98% of the melting point temperature of the metal fibers.

26. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The step of heating the plurality of fibers and the step of cooling the plurality of fibers are performed within a predetermined time period of less than 30 minutes.

27. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The step of heating the plurality of fibers and the step of cooling the plurality of fibers are performed within a predetermined time period of less than 15 minutes.

28. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The step of heating the plurality of fibers and the step of cooling the plurality of fibers are performed within a predetermined time period of less than 5 minutes.

29. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, The step of heating the plurality of fibers and the step of cooling the plurality of fibers are performed within a predetermined time period of less than 1 minute.

30. The method of assembling a fiber network of claim 26, wherein, The predetermined time period is equally divided between the step of heating the plurality of fibers and the step of cooling the plurality of fibers.

31. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, In the step of cooling the fibers, the cooling rate is maintained above 50 K / min until the fibers are cooled to a temperature of 60% or less of the melting point temperature of the metal fibers.

32. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, In the step of cooling the fibers, the cooling rate is maintained above 100 K / min until the fibers are cooled to a temperature of 60% or less of the melting point temperature of the metal fibers.

33. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, The metal fibers comprise a length of 1.0 mm or more, and / or a width of 100 µm or less, and / or a thickness of 50 µm or less.

34. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The change in width of the fibers along their length is less than 20% compared to the initial width of the fibers prior to performing the step of heating the plurality of fibers.

35. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The change in width of the fibers along their length is less than 10% compared to the initial width of the fibers prior to performing the step of heating the plurality of fibers.

36. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The change in width of the fibers along their length is less than 5% compared to the initial width of the fibers prior to performing the step of heating the plurality of fibers.

37. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, The change in width of the fibers along their length is less than 1% compared to the initial width of the fibers prior to performing the step of heating the plurality of fibers.

38. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10), before and / or after being fixed to each other, show an exothermic event when heated in a DSC measurement, wherein the exothermic event releases an energy of 0.1 kJ / g or more.

39. The method of assembling a network of fibers according to any one of claims 1 to 5, wherein, The metal fibers (10), before and / or after being fixed to each other, show an exothermic event when heated in a DSC measurement, wherein the exothermic event releases an energy of 0.5 kJ / g or more.

40. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10), before and / or after being fixed to each other, show an exothermic event upon heating in a DSC measurement, wherein the exothermic event releases an energy of 1.0 kJ / g or more.

41. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10), before and / or after being fixed to each other, show an exothermic event upon heating in a DSC measurement, wherein the exothermic event releases an energy of 1.5 kJ / g or more.

42. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10) comprise a non-circular cross-section.

43. The method of assembling a fiber network of claim 42, wherein, The non-circular cross-section is a rectangular cross-section, a conic cross-section, a partly circular cross-section, or an elliptical cross-section having a major axis and a minor axis.

44. The method of assembling a fiber network of claim 38, wherein, The ratio of the minor axis to the major axis is in the range of 0.05 to 1.

45. The method of assembling a fiber network of claim 38, wherein, The ratio of the minor axis to the major axis is in the range of 0.1 to 0.

7.

46. The method of assembling a fiber network of claim 38, wherein, The ratio of the minor axis to the major axis is in the range of 0.1 to 0.

5.

47. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10) comprise a circular cross-section.

48. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10) can be obtained by subjecting a molten material of the metal fibers to a cooling rate of 10 2 K min -1 or higher, by a vertical or horizontal melt spinning process.

49. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, At least some of the metal fibers (10) in the plurality of metal fibers are amorphous, or wherein at least some of the metal fibers (10) in the plurality of metal fibers are nanocrystalline.

50. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10) are in electrical contact with each other.

51. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10) are in direct electrical contact with each other.

52. The method of assembling a fibrous network according to any one of claims 1 to 5, wherein, The metal fibers (10) comprise an alloy of one or more of copper, silver, gold, nickel, palladium, platinum, cobalt, iron, chromium, vanadium, titanium, aluminum, silicon, lithium, manganese, boron.

53. The method of assembling a fiber network of claim 52, wherein, The alloy is CuSn8, CuSi4, AlSi1, stainless steel, or a Vitrovac alloy.

54. A network of metal fibers obtained according to the method of any one of claims 1-53, comprising a plurality of metal fibers fixed to each other at contact points (14); wherein, The metal fibers (10) comprise a non-circular cross-section, or The metal fibers (10) comprise a circular cross-section, and The metal fibers (10) comprise a width along the length of the fiber that is substantially constant, such that the variation of the width of the fiber along its length is less than 40%.

55. The network of metal fibers according to claim 54, wherein, The non-circular cross-section is a rectangular cross-section, a conic cross-section, a partly circular cross-section, or an elliptical cross-section having a major axis and a minor axis.

56. The network of metal fibers according to claim 54, wherein, The variation of the width of the fiber along its length is less than 30%.

57. The network of metal fibers according to claim 54, wherein, The variation of the width of the fiber along its length is less than 20%.

58. The network of metal fibers according to claim 54, wherein, The metal fibers (10) in the plurality of metal fibers do not comprise a constriction (16).

59. The metal fiber network of any of claims 54-58, wherein, Individual fibers in the plurality of fibers are sintered to each other.

60. The network of metal fibers according to claim 55, wherein, The ratio of the minor axis to the major axis is in the range of 0.05 to 1.

61. The network of metal fibers according to claim 55, wherein, The ratio of the minor axis to the major axis is in the range of 0.1 to 0.

7.

62. The network of metal fibers according to claim 55, wherein, The ratio of the minor axis to the major axis is in the range of 0.1 to 0.

5.

63. The metal fiber network of any of claims 54-58, wherein, The network is an ordered network or a disordered network.

64. The metal fiber network of any of claims 54-58, wherein, The network has open pores between the metal fibers in the plurality of metal fibers.

65. The metal fiber network of any of the preceding claims 54 to 58, wherein, The contact points (14) between the metal fibers are distributed in a disordered or ordered manner throughout the three-dimensional structure of the network.

Images