Field

[0001] The present disclosure relates to carbon brushes, asynchronous induction generators comprising said brushes and to the use of such devices in wind turbines.

Background of the disclosure

[0002] Wind turbines generally use a three-phase asynchronous induction generator to convert mechanical power from rotation of the wind turbine blades into electrical power. The generator comprises a stator and an electrical rotor arranged in a housing, and a slip ring assembly which is mounted on the rotor shaft and connected to the rotor windings to transfer current between the electrical rotor and a frequency converter. The slip ring assembly comprises contact rings mounted on the electrical rotor, which rotate over static slip ring brushes mounted in a slip ring housing to provide a rotating electrical contact.

[0003] Slip ring brushes (also known as carbon brushes) are usually carbon based, for example, a graphite or carbon silver alloy composite. Although referred to as brushes, slip ring brushes are solid blocks of material. Wear of the slip ring brushes very frequently necessitates unscheduled maintenance that is more expensive than scheduled maintenance and is therefore undesirable.

[0004] Wind turbines are unmanned as it is unsafe to enter the turbine when it is in operation. Wind turbines are intended to run all day, every day. Thus, there are no personnel to check the day to day running of wind turbines, and wind turbine generators are required to operate with higher reliability than other induction generators. Moreover, wind turbines are often located in remote inaccessible locations where maintenance is expensive due to the cost of safely transporting trained personnel to the wind turbines and the time taken to do so. If wind turbines are located in offshore wind farms, maintenance difficulties are exacerbated whilst operating in hostile environments. Wind turbines typically have a scheduled maintenance programme, and any additional maintenance is highly undesirable as it is estimated to double the cost of scheduled maintenance.

[0005] Wearing slip ring brushes have been found to be one of the main causes of wind turbine failures. If the brushes are not replaced once they have worn past a certain point, the turbine must be shut down. When maintenance is performed, the remaining length of the brushes must be measured manually to check that there is sufficient brush material remaining until the next scheduled periodic maintenance.

[0006] We have therefore appreciated that there is a need to provide an induction generator and cost-effective slip ring brush used therein for wind turbines which are able to withstand extreme atmospheric conditions for prolonged periods used in wind energy systems to avoid the need for unscheduled maintenance and to lengthen the time between instances of scheduled maintenance.

Summary of the disclosure

[0007] In a first aspect of the present disclosure there is provided a carbon brush for an electric motor comprising a composite of:

• 30 to 80 wt% metallic phase comprising a major non-silver component and an optional minor silver component, said metallic phase comprising grain boundaries; and

• >0 to 10 wt% silver rich phase;

• 0 to 10 wt% additives; and

• the balance carbon material wherein at least a portion of the silver rich phase is located at or proximal the grain boundaries of said metallic phase.

[0008] The carbon brush of the present disclosure has been found to provide an advantageous combination of oxidation resistance, hardness, and dimensional stability to contribute towards achieving low brush and slip ring wear rates in operation, whilst maintaining an acceptable resistivity. In particular, a relatively small amount of silver concentrated in the silver rich phase at the grain boundaries of the metallic phase has resulted in an unexpected balance of functional properties. These functional properties make the carbon brush particularly suitable for prolonged use in extreme environments, such as within wind turbines.

Metallic phase

[0009] The metallic phase preferably comprises a median grain size (Dso) of at least 10 pm or at least 12 pm or at least 14 pm or at least 16 pm or at least 18 pm or at least 20 pm or at least 22 pm or at least 24 pm. Larger grain sizes correlate to smaller grain boundary areas which are prone to oxidation. Larger grain sizes also promote better electrical and thermal connectivity over the composite. The reduced grain boundary areas enable a smaller amount of the silver rich phase to be required to enhance performance. Additionally, larger copper grains better inhibit crack propagation. The maximum metallic phase grain size is typically less than 50 pm or less than 40 pm or less than 30 pm or less than 25 pm. Excessive grain sizes may affect the homogeneity of the composition and result in localised mismatches of the coefficient of thermal expansion causing crack formation. Larger metallic phase grain sizes are also associated with increased composite wear.

Major non-silver component

[0010] The major non-silver component may be selected from the group consisting of copper, aluminium, zinc, iron, nickel steel and tin or combination thereof. In a preferred embodiment, the non-silver component comprises copper. For reasons of convenience and readability, within the specification, a non-silver component are referred to as a copper component. However, it will be understood that other non- silver components may also be substituted for copper.

[0011] The metallic phase comprises a major non-silver metal component, which may be at least 50 wt% or at least 70 wt% or at least 90 wt% of the total weight of the metallic phase.

Optional minor silver component

[0012] The metallic phase may also comprise a minor silver component which may comprise between 0.01 or 0.1 to 30 wt% of the total weight of the metallic phase, but is typically less than 20 wt% or less than 10 wt% or less than 5 wt% or less than 3 wt% of the total weight of the metallic phase. The minor silver component may, for example, be derived from the diffusion of silver into the metallic phase during the heat treatment step above the eutectic temperature of the Cu-Ag (or non-Ag-Ag) system, when Cu is the non-silver metal component.

[0013] The remainder, if any, may be other conductive metals. The amount of metallic phase may vary depending upon the application and performance requirement but is generally between 35 wt% and 75 wt% metallic phase or between 40 wt% and 70 wt% metallic phase or between 45 wt% and 65 wt% metallic phase relative to the total weight of the composite.

Silver rich phase

[0014] The silver rich phase typically comprises at least 90 wt% Ag or at least 95 wt% Ag or at least 98 wt% Ag. The other components, if present, in the silver may be impurities and/or copper defused from the metallic phase. The presence of silver at or proximal to the grain boundaries of the metallic phase has been found to improve the long-term durability of the carbon brushes. The carbon brush may comprise at least 0.01 wt% or at least 0.05 wt% or at least 0.10 wt% or at least 0.15 wt% or least 0.20 wt% or at least 0.30 wt% or at least 0.40 wt% or at least 0.50 wt% or at least 0.80 wt% or at least 1 .0 wt% of the silver rich phase. The upper limit of the silver rich phase is preferably no more than 8 wt% or no more than 7 wt% or no more than 6 wt% or no more than 5 wt% or no more than 4 wt% or no more than 3.5 wt% or no more than 3 wt% or no more than 2.5 wt% or no more than 2.0 wt%. The proportion of the silver rich phase which is at or proximal to the grain boundary is preferably at least 50 wt% or at least 60 wt% or at least 70 wt% or at least 80 wt% or at least 90 wt% or at least 95 wt% of the total amount of the silver rich phase. For the purposes of the present disclosure proximal the grain boundary means that at least a portion of the silver rich phase is within 5 pm or within 4 pm or within 3 pm or within 2 pm of the grain boundary.

[0015] In some embodiments, at least 10 wt% at least 20 wt% at least 30 wt% or at least 40 wt% of the silver rich phase is located at (i.e., in contact with) or proximal the metallic grain boundaries. In some embodiments, the silver rich phase are concentrated (e.g., greater than 50 wt% or greater than 60 wt%) at or proximal the metallic grain boundaries. The silver rich phase is different from silver dissolved in the metallic phase. Additives

[0016] In some embodiments, the carbon brushes comprise additives (e.g., >0 wt%). The additives may be less than 8 wt% or less than 6 wt% or less than 5 wt%. When additives are present, they are usually present in amounts of at least 0. 2wt% or at least 0.5 wt% or at least 1 .0 wt% of the total weight of the composite. In other embodiments, the carbon brushes comprise no additives. The additives may be added to the mixture pre-heat treatment or post treatment (e.g., impregnated) depending upon the heat sensitivity of the additives. A range of different additives may be present, such that are used in the field of sliding electrical contacts including solid lubricants, abrasives, and anti-oxidizing additives.

[0017] In some embodiments, the additives comprise synthetic mechanically resistant material such as silicon nitride and/or boron carbide and/or titanium carbide and/or titanium boride and/or tungsten carbide. Additives for improving sliding properties of the carbon brush may also be present including metal sulphides, metal oxides, polytetrafluoroethylene, graphene, and derivations thereof including graphene oxide (GO), reduced graphene oxide (rGO), graphene nanoplatelets (GNO) and/or carbon nanotubes (CNT).

[0018] The addition of Zn, graphene, Bi, Mg, Si, Sn and/or Al, in elemental form or as a compound thereof, has been found to reduce brush expansion.

[0019] In one embodiment, the additive is an alkaline earth metal, an alkaline earth metal salt, a metal halide or sulphide, or metallic lead. In a preferred embodiment, the additive comprises an alkaline earth metal salt, such as calcium carbonate. In some embodiments, the composite comprises in the range of 0.5 to 4 wt% alkaline earth metal salt or in the range 1 .0 to 3.0 wt% alkaline earth metal salt. This class of additives has been found to exhibit improved durability through reducing porosity during the sintering process, when applied to the composite mixture.

Carbon material

[0020] Typically, the composite comprises at least 1 wt% or at least 5 wt% or at least 10 wt% carbon material. In some embodiment, the composite comprises between 10 and 70 wt% carbon material or between 20 wt% and 60 wt% carbon material or between 25 wt% and 55 wt% carbon material. The balance between carbon material and other components of the composite will depend upon the desired properties of the carbon brush required.

[0021] The carbon material may be any carbon material comprising good electrical properties including, but not limited to, graphite, coke, carbon black, carbon fibre and/or carbonised pitch or carbonised resin (e.g., phenolic resin). In one embodiment, the carbon material comprises graphite and carbonised resin.

[0022] The long-term durability of carbon brushes may be influenced by dimensional brush expansion which accelerates the oxidation mechanism and in turn the wear rate of the carbon brushes. Through controlling the composite’s microstructure (including porosity, grain boundaries and silver distribution) wear mechanisms may be inhibited. [0023] The composite preferably comprises a porosity of no more than 13.0% v/v or no more than 12.0% v/v or no more than 11 .0% v/v or no more than 10.0 % v/v or no more than 9.0 %v/v or no more than 8 % v/v. The lower the porosity, the lower the propensity for the ingress of moisture and/or oxygen, which may adversely affect the durability of the composite.

[0024] The average pore size of the composite may be less than 0.5 pm or less than 0.45 pm or less than 0.40 pm or less than 0.38 pm or less than 0.36 pm. The pore size may be determined by a mercury intrusion porosimetry technique whereby the pores of a composite sample are filled with non-wetting mercury under applied pressure following standard methods known in the art (e.g., ISO 15901 ).

[0025] The composite may have a resistivity of less than 0.75 pom or less than 0.70 pom or less than 0.65 pom or less than 0.60 pom or less than 0.55 pom or less than 0.50 pom.

[0026] The composite may have a hardness greater than 12 HBN or greater than 13 HBN or greater than 14 HBN or greater than 15 HBN.

[0027] The strength of the composite may be greater than 20 MPa or greater than 25 MPa or greater than 30 MPa or greater than 32 MPa or greater than 35 MPa or greater than 37 MPa.

[0028] The carbon brushes may have an expansion ratio of less than 0.02 upon cyclic heating at 140°C in dry or humid air. [0029] The operating temperature of the carbon brush may be less than 100 oC or less than 90 oC or less than 80 oC. Lower operating temperatures reduce brush overheating, contact drop, and reactivity in dry or humid air, thereby reducing brush wear.

[0030] Dry or humid air conditions refer to exposure to air at humidity levels ranging from 0 to 100%. In some embodiments, 100% humidity levels can be achieved by immersing the brush in boiling water prior to exposing it to operating conditions.

[0031] In one embodiment, the composite of the carbon brush comprises at least one or at least two or at least three or at least 4 or all of the following properties:

• resistivity of less than 0.75 pom;

• hardness greater than 12 HBN;

• strength greater than 20 MPa;

• average pore size of less than 0.5 pm;

• operating temperature of less than 100 °C; and

• expansion ratio of less than 0.02 upon cyclic heating at 140 °C in dry or humid air.

[0032] The carbon brush may be used with a current density of at least 15 A/cm ² or at least 17 A/cm ² or at least 20 A/cm ² or at least 25 A/cm ² or at least 30 A/cm ² or at least 35 A/cm ² or at least 39 A/cm ² . It has been found that the wear properties of the carbon brushes of the present disclosure are particularly advantageous in applications requiring higher current densities.

Structural configuration

[0033] In some embodiments, the carbon brush comprises a monolithic composite body. A monolithic body has the advantage of a simpler manufacturing process free of lamination joints which may be prone to oxidation. In other embodiments the carbon brush may comprise an outer wear layer and an inner core. In embodiments comprising an outer wear layer and an inner core, the amount of silver is preferably greater in the outer layer than in the inner core.

[0034] In a second aspect of the present disclosure, there is provided a process for manufacturing a carbon brush according to the first aspect comprising: a. Providing a mixture of carbon material, a non-silver metal component, a silver component (or precursors thereof) and optional additives, said non-silver metal component comprising grains; b. Compressing the mixture; c. Heat treating the compressed mixture at a temperature above an eutectic temperature of a silver and non-silver metal composition for sufficient time for the silver component and the non-silver metal component to sinter together to form a metallic phase comprising grain boundaries; and a silver rich phase at or proximal to said grain boundaries of the metallic phase.

[0035] The non-silver metal component may be selected from the group consisting of copper, aluminium, zinc, iron, nickel steel and tin or combinations thereof. In some embodiments, the non-silver metal component comprises copper.

[0036] The process of the present disclosure uses a silver component (e.g. silver or compounds of silver) as a sintering agent and to promote a contiguous network of copper (or other non-silver metal) particles/grains, with the heat treatment controlled such that a silver rich phase is present at or proximal to the grain boundaries, thereby providing an oxidation inhibitor at the sites most vulnerable to oxidation. The presence of silver during the heat treatment has also been found to promote larger particle/grain sizes thereby reducing the grain boundary surface area, further minimising the propensity of oxidation in the composite.

[0037] In one embodiment, the metallic phase grains have an average grain size of at least 10% or at least 20% greater than the average grain size of a metallic phase grains without the addition of a silver component to the mixture. In one embodiment, the metallic phase grains have an average grain size of at least 10% or at least 20% or at least 40% or at least 50% or at least 60% greater than the Dso grain size of a metallic phase grains without the sintering temperature being above the eutectic temperature of a silver and copper system.

[0038] Surprisingly, the use of elevated sintering temperatures, normally associated with increased porosity and poor resistance to oxidation, has been able to be used on the formulations of the current disclosure to produce durable carbon brushes. [0039] The mixture may also comprise a binder, such as a phenolic resin or coal tar pitch. The proportion of binder is typically between 1 wt% and 25 wt%. The binder may become a carbon material after its carbonisation, during the sintering process. [0040] Typically, the mixture comprises particulates, although the metal raw materials may be in precursor form, such as oxides, salts or solutions. The sizes of the particulates will be those known in the field of carbon brush manufacture. In some embodiments, the silver may be in the form of a fine powder or solution to enable homogeneous mixing of this minor component within the copper. In some embodiments, the metal components (e.g., copper and silver) are pre-mixed prior to the addition of the carbon material.

[0041] The mixture is then compressed (typically cold pressed), optionally in an appropriate mould having the shape of the desired brush, then the green (i.e., unsintered) material obtained is sintered above the eutectic temperature of the silver/copper system, thereby producing at least some alloyed material. The sintering time and/or the cooling rate of the sintered material is controlled such that a portion of the silver remains or precipitates at the copper grain boundaries.

[0042] While some natural solid-state diffusion of the copper into the silver phase may occur over time, the rate of diffusion is accelerated by elevating the temperature. To better control copper grain growth rate and silver distribution, the sintering temperature may be between >0 to 400 °C above the eutectic temperature of the silver and copper (or non-silver metal) system. In some embodiments, the sintering temperature at least 20 °C or at least than 50 °C or at least 100 °C or at least 150°C above the eutectic temperature of the silver and copper (or non-silver metal). For example, a 78 wt% silver, 22 wt% copper composition has an eutectic temperature of about 779 °C and a 75 wt% silver, 25wt% aluminium composition has an eutectic temperature of about 562 °C. The sintering temperature is preferably sufficiently high for the silver and non-silver component to form a liquid sintering phase at the interface between the particles and more preferably sufficiently high for the silver and non-silver phases to transform into said silver rich and non-silver component phases.

[0043] In some embodiments, the sintering temperature is less than 300 °C or less than 200 °C or less than 100 °C or less than 80 °C or less than 60 °C or less than 50 °C or less than 25 °C above the eutectic temperature of the silver and copper system. It should be understood that higher temperatures may be used if there is sufficient control over the temperature profile of the sintered material with respect to time. [0044] While not wanting to be constrained by theory, it is thought that the silver functions to promote copper (or non-silver metal) grain growth and thereby is concentrated at the copper grain boundaries. Thus, the function of the silver is both to promote copper grain growth and to concentrate the anti-oxidising properties of silver at an oxidising prone region of the composite. This mechanism enables small amounts of silver to provide significant functional benefits.

[0045] Furthermore, it is proposed to use the brush obtained in an electric machine used to transfer power, and in particular a generator such as a wind turbine.

[0046] Moreover, the disclosure relates to the use of the brush according to the disclosure in an application characterized by electrical currents lying between 1 and 1000 A, and by voltage drops across the contact of between 1 and 1000 V, typically a signal transfer application. In some embodiments, the carbon brushes are operated with an electrical current of at least 100 A or at least 200 A and/or a voltage drop of at least 100 V or at least 200 V.

[0047] The disclosure is not limited to a given application; however, mention may especially be made of: o applications relating to the transfer of electrical power, for example in the fields of wind turbines, special machines, etc.; o applications relating to signal transfer, for example in the fields of tachometers, of the sensing of measurement currents such as with thermocouples and thermometer probes, in small precision motors for timepieces, medical applications, etc.; and o applications in very dry atmospheres, for example in the aeronautic or aerospace field.

[0048] The carbon brush’s durability and ability to withstand extreme environments makes the carbon brush particularly suited to extreme environments where nonroutine maintenance can be difficult and expensive, such as those associated with wind turbine applications.

[0049] In a third aspect of the present disclosure there is provided an asynchronous induction generator comprising the carbon brush of the first aspect of the present disclosure or a carbon brush produced by the second aspect of the present disclosure.

[0050] In a fourth aspect of the present disclosure, there is provided the use of the carbon brush of the first aspect of the present disclosure or a carbon brush produced by the second aspect of the present disclosure, in a wind turbine. In some embodiments, the wind turbine comprises an asynchronous induction generator.

[0051] Unless otherwise indicated, reference to elemental components such as silver or copper are references to the elemental components in their elemental form (e.g., metallic form).

[0052] ASTM E7-17 standard terminology is used for definitions, with a grain defined as an individual crystallite in metals; and a grain boundary defined as an interface separating two grains or at the periphery of the grain. When the boundary is between two interfacing grains, the orientation of the lattice changes from that of one grain to that of the other. Particles may comprise one or more grains.

Brief Description of the Figures

[0053] Figure 1 is a graph illustrating the correlation between composite porosity and expansion of carbon brush composites of the present disclosure;

[0054] Figure 2 is a graph illustrating the distribution of pore sizes of carbon brush composites of the present disclosure;

[0055] Figure 3 is an EDS scan of silver content in a metallic phase of a carbon brush composites of Example 1 of the present disclosure;

[0056] Figure 4 is a magnified EPMA analysis image of the metallic phase of Figure 3; [0057] Figure 5(a) is a SEM of a carbon composite brush of comprising 2 wt% silver of Example 2;

[0058] Figure 5(b) is an EDS scan of the copper content corresponding to the SEM image of Figure 5(a);

[0059] Figure 5(c) is an EDS scan of the silver content corresponding to the SEM image of Figure 5(a);

[0060] Figure 6(a) is an EDS scan of the copper content of a carbon composite brush of Comparative Example 7 (C-7); [0061] Figure 6(b) is an EDS scan of the silver content of a carbon composite brush of Comparative Example 7 (C-7);

[0062] Figure 7(a) is an EDS scan of the copper content of a carbon composite brush of Comparative Example 8 (C-8);

[0063] Figure 7(b) is an EDS scan of the silver content of a carbon composite brush of Comparative Example 8 (C-8);

[0064] Figure 8 is a SEM image of a carbon composite brush comprising no silver (Comparative Example 1 (C-1 ));

[0065] Figure 9 is a SEM image of a carbon composite brush comprising 0.5 wt% silver of Example 3; and

[0066] Figure 10 is a graph illustrating the impact of silver addition on the average size of copper grains in carbon brush composites of the present disclosure.

Detailed Description

[0067] All samples are made by conventional powder metallurgy processes. In a typical process, milled and/or sized raw materials are mixed, cold-pressed, and baked in a non-oxidizing environment at elevated temperatures. The baking temperatures used to prepare the samples are listed in Table 1 . After baking, the samples are allowed to cool by controlled convection.

[0068] The time that the composite is held at the target temperature may be between 3 hours and 24 hours or between 6 hours and 12 hours, although shorter or longer times are possible.

[0069] The cooling rate is generally no more than 100°C/minute or no more than 50°C/ minute or no more than 20°C/ minute or no more than 10°C/ minute or no more than 5°C/ minute or no more than 1 °C/ minute. Higher cooling rates may result in cracking and/or inhibit the precipitation of Ag.

[0070] The metallic particles used are 100 wt% silver particles and 100 wt% copper particles, except for CE-7, in which alloyed particles with a composition of 2 wt% silver and 98 wt% copper are used as starting materials. The alloyed particles were sourced from Morgan Technical Ceramics (Hayward, CA) and produced via an atomisation process, with particles of less than passing a 325 mesh sieve used (<45 pm).

[0071] The method of the disclosure is not limited by conventional processing parameters employed in the art for carbon brush composite manufacture. Examples

[0072] In a representative example, six parts of weight of a graphitic carbon material are mixed with 1 part by weight of a phenolic binder resin dissolved in ethanol. The mixture is stirred and dried under mild heat to remove the alcohol. The dried product is milled, sized, and premixed with CaCOa, prior to being mixed with copper and silver powders of average sizes <100 pm, cold-pressed to make blocks, baked to the target heat treatment temperature of 955°C, which is maintained for about 12 hours, and cooled by convection in non-oxidative environment to obtain the carbon brush composite with final composition listed in Table 1 .

Table 1

[0073] Table 1 also lists the heat treatment temperatures and final compositions of the carbon brush composites made following the same processing guidelines as noted for the representative example, namely Comparative Example 1 (C-1), Comparative Example 2 (C-2), Comparative Example 3 (C-3), Comparative Example 4 (C-4), Comparative Example 5 (C-5), Comparative Example 6 (C-6) and Comparative Example 8 (C-8) , except for Comparative Example 7 (C-7), where alloyed Cu-Ag particles are used.

Methodology

Wear

[0074] The resistance of the carbon composites to sliding wear was tested under experimental conditions designed to resemble those of carbon brush performance in asynchronous induction generators utilized in conventional wind turbines. The tests were carried out using a custom-built testing machine following the guidelines provided in standard method ASTM G-77.

Resistivity

[0075] The composite resistivity is determined by passing an electric current between two contact points of individual composite specimens of dimensions prescribed according to standard test methods known in the art (e.g. ASTM D7775) and measuring the voltage drop in order to calculate the average resistivity (p) according to the formula p = R.A/L, where the resistance R is calculated from the applied current and measured voltage drop for specimens of cross sectional area A and distance L between voltage contacts.

Contact drop

[0076] Brush contact drop, usually reported in volts measured using a conventional voltmeter, refers to the voltage drop expressed as the sum of voltage drop across the anodic brush plus that across the cathodic brush.

Friction coefficient

[0077] The coefficient of friction of a brush is defined as the ratio of tangential force to normal force at the face of the brush using standard procedures described in ASTM G77.

Hardness

[0078] The composite Brinell hardness (HBN) is determined by specimen indentation tests as prescribed by standard methods known in the art (e.g., ASTM E10). Expansion Ratio

[0079] The average expansion ratio or the composite is determined by subjecting composite specimens to repeated cycles of heating at 140°C and periodic cooling once per week for a 52-week period and calculating the change in brush dimension upon completion of the cycles. Brush expansion tests were carried out in both dry conditions (without exposure to humidity) and humid conditions (after exposing the specimens to 100% humidity and pre-drying them).

Strength

[0080] The composite strength is determined by standard 3-point flexural test method ASTM C651.

Brush temperature

[0081] Unless noted otherwise, the temperature of the composite brush during operation was determined by a standard technique using thermocouples placed 3 mm away from the contact surface. In selected cases, the brush temperature was determined using a non-contact method (infrared camera) which provided a good correlation with the standard technique. Lower operating temperatures are desirable in order to extend the brush life by minimizing their cyclic expansion and consequent deterioration during their use. Appropriate brush composition and microstructure aid in reducing brush operating temperature by providing a more effective dissipation of heat through enhanced composite thermal conductivity and network connectivity effects. Grain size distribution

[0082] Optical micrographs were taken of the etched microstructure and the metallic grains were analysed using Imaged™ software. Approximately 200 grains from each sample were assessed to form the grain size distribution curve (Figure 10).

Table 2 current density: 17 A/cm ²

[0083] As indicated by the results in Table 2, for carbon brush composites of similar copper content, an increase in silver content results in a decrease in resistivity. This decrease is expected on the basis of the higher conductivity of silver compared to copper. In contrast, when comparing composites of wider compositions and heat treatment temperatures, an overall increase in resistivity and a decrease in hardness may be noted. Table 2 also shows that the presence of small amounts of silver also inhibits the expansion of the carbon brush composites in both dry and humid conditions.

[0084] It will be appreciated that brush temperature may also be dependent upon a number of compositional factors and hence the most relevant comparison is between Comparative Example 4 (C-4) and Example 7, where the difference in the composites relates to their heat treatment. The lower brush temperature of Example 7 is indicative of differences at a microstructure level which enables the brush temperature of composites under the present disclosure to remain cooler. Comparative Example 1 (C-1) failed performance testing at higher current densities (Table 4 &5) with the brush overheating (>200°C) with the composite breaking down and/or wearing quickly.

While not wanting to be bound by theory, it is thought that this is due to an increase in particle/grain size and interconnectivity between the metal phases within the composite.

[0085] With reference to tables 3 to 5, Examples 2 & 3 exhibit improved wear compared to the comparative examples with no or high levels of silver content. Example 1 , with an elevated copper content (65 wt%) had lower performance, but still superior to that the silver free comparative example. Although C-2 had the lowest brush temperature at the current density of 39 A/cm ² (Table 5), it’s wear rate was higher than Examples 1 -3. This apparent contradiction was due to the fact that silver has a lower heat capacity and higher conductivity than copper and, as a result, is able to dissipate heat more readily.

Table 3 current density: 17 A/cm ²

Table 4 current density: 26 A/cm ²

Table 5 current density: 39 A/cm ²

[0086] Effect of processing route on carbon composite microstructure [0087] With reference to Figures 5(a), (b) and (c), the microstructure of the carbon composite of Example 2 clearly highlights a concentration of a silver rich phase proximal or at the grain boundaries, which are emphasised with a white dashed lines around the grains 500, 510, 520, 530, 540, 550 in Figure 5(a) and corresponds to the concentration of silver in the EDS spectra of Figure 5(c). It is estimated that at least 30wt%, for example, at least 50wt% of the silver is concentrated proximal or at the grain boundaries, with the remainder of the silver appearing as a minor component within a copper rich phase. This may be observed through the relatively low intensity (i.e., low level of white pixels) within the boundaries of grains 500, 510, 520, 530, 540 in Figures 5(c) compared to the corresponding grains in 5(b)). In some embodiments, there is a contiguous concentration of silver at the grain boundaries, wherein, according to some embodiments, "contiguous" refers to the silver connecting without a break along the grain boundaries and/or forming an edge around a portion of a grain boundary.

[0088] In contrast, Figures 6(a) and (b) highlight the copper and silver distribution from an EDS scan of Comparative Example 7 (C-7), in which the silver is homogeneously distributed throughout the copper/silver alloy phase.

[0089] Figure 7(a) and (b) highlight the copper and silver distribution from an EDS scan of Comparative Example 8 (C-8), in which the microstructure is composed of discrete particles of silver and copper, with no alloy formation.

[0090] Whilst the carbon composite composition is the same in Example 2, in Comparative Example 7 (C-7) and Comparative Example 8 (C-8), the processing route used in Example 2 promotes the formation of a silver rich phase at or proximal to the grain boundaries of a copper rich phase. While not wanting to be held by theory, it is thought the sintering of discrete silver and copper particles in Example 2 results in localised eutectic compositions leading to alloy formation, in which a portion of the silver component precipitates out at the metallic phase boundary upon cooling. [0091] This phenomenon is not observed when using Cu-Ag alloyed particles Comparative Example 7 (C-7) as the homogeneous alloy composition does not promote the formation of an eutectic composition (e.g., about 78% Ag and 22wt% Cu), given the 2wt% silver content of the alloy. Whereas the silver particles (with a melting point of 962°C) at the interface with the copper particles are able to form a liquid phase with the copper able to diffuse into the silver phase and thereby promote further liquid phase formation as the solidus/liquidus temperature of the composition decreases.

[0092] Similarly, the particles in Comparative Example 8 (C-8) remain as discrete particles as they are not exposed to a sintering temperature above the copper/silver eutectic temperature. As a result, the microstructure of the metallic phase particles in Comparative Example 7 (C-7) and Comparative Example 8 (C-8) remain substantially unchanged from their microstructure prior to the sintering step used in the formation of the carbon composites.

[0093] The effect of having the silver concentrated at the grain boundaries is that the copper is better protected against oxidative corrosion compared to the microstructures of Comparative Example 7 (C-7) and Comparative Example (C-8). This is highlighted with the reduced expansion levels observed, particularly under humid conditions (Table 2). The lower material expansion results in the mechanical integrity of the composite being maintained for longer, which thereby results in improved wear rates, particularly at higher current densities (Tables 3 to 5).

Effect of sintering temperature on performance

[0094] With reference to Figure 3, an EDS scan of the silver distribution within the composite of Example 1 , reveals a low level of silver evenly distributed within the metallic phase 100, which was predominately comprised of copper. The brighter particles/grains of the silver rich phase 110 were predominately located at the grain boundary of the metallic phase. As indicated in the EPMA analysis image (Figure 4), the silver rich phase 200 is located at the grain boundary of a copper rich phase 210. [0095] The difference in microstructure is illustrated in Figures 8 (Comparative Example 1 ) & 9 (Example 1 ), with the effect of 0.5 wt% silver with heat treatment above the eutectic temperature of the Cu-Ag system resulting in a significant increase in grain size, as shown in the cumulative grain size distributions of Figure 10, with the D50 grain size being 8 pm for C-1 compared to 23 pm for Example 1 .

[0096] As indicated in Figures 1 & 2, in addition to promoting larger metallic grain sizes, the addition of a small amount of silver at an evaluated baking temperature has the effect of reducing pore size, porosity and consequently the propensity of the composite to expand, as indicated by the expansion test. By reducing the expansion properties of the composite, the composite is less prone to the corrosive effects of oxidation and moisture ingress. It is counter-intuitive that the addition of a small amount of silver at high baking temperatures can result in a composite microstructure with a reduced propensity to expand after repetitive exposure to extreme temperature variations. [0097] For the avoidance of doubt, it should be noted that in the present specification the term “comprise” in relation to a composition or a particle/grain size range is taken to have the meaning of include, contain, or embrace, and to permit other ingredients or other particle/grain sizes to be present. The terms “comprises” and “comprising” are to be understood in like manner. Many variants of the carbon brushes of the present disclosure will be apparent to the person skilled in the art and are intended to be encompassed by this disclosure.