WO1995004928A1

WO1995004928A1 - A microelectrode assembly

Info

Publication number: WO1995004928A1
Application number: PCT/AU1994/000466
Authority: WO
Inventors: Stephen Fletcher; Michael David Horne
Original assignee: Commonwealth Scientific And Industrial Research Organisation
Priority date: 1993-08-11
Filing date: 1994-08-11
Publication date: 1995-02-16

Abstract

A microelectrode assembly (4) which includes a substantially uniform dispersion of a plurality of substantially parallel conducting fibres (14), such as carbon fibres, embedded in a non-conducting material (12) such as cured epoxy resin. A method for making a microelectrode assembly is also described, whereby the conducting fibres (14) are added and dispersed in the non-conducting material (12) and introduced into a mould (20) which causes the fibres (14) to align. Upon solidification the mixture (26) is cut to form the assembly (4).

Description

A MICROELECTRODE ASSEMBLY

This invention relates to novel microlectrode assemblies and to novel methods of manufacturing these assemblies.

Microelectrodes are defined as electrodes having one characteristic dimension less than 100 micrometers. When used in electrochemical analysis microelectrodes offer three distinct advantages over ordinary electrodes: 1. The rates of diffusion of reactive species to and from microelectrode surfaces are enhanced relative to the rates or unwanted (non-diffusion controlled) side reactions. That is, signal-to-noise ratios are higher. 2. The distortion of the data due to the resistance of the electrolyte solution is much less. 3. Higher spatial resolution is possible.

The implications of these advantages are significant. Firstly, the enhanced diffusion of reactive species to and from microelectrodes means dilute reactants can be concentrated more rapidly, and faster reactions than normal can be studied. Secondly, the very small ohmic distortion at microelectrodes allows electrochemical measurements to be performed in highly resistive media, something that is impossible using conventional electrodes. Measurements can be made in organic solvents, polymers and solid electrolytes. Thirdly, because the double layer capacitance at a microelectrode is small, the background charging current is also small, and a fifty-fold improvement in sensitivity is possible over macroelectrodes. Furthermore, the low RC time constant at a microelectrode also means very fast scans of applied potential, up to several hundred volts per second, can be used. Finally, the high spatial resolution offered by microelectrodes allows electrochemical measurements in very small volumes. Whilst having all these advantages, single microelectrodes are not commonly used in electrochemical analysis. The main reason is that the absolute magnitude of the electrochemical currents measured at a single microelectrode is extremely small - of the order of picoamps - and to accurately record such small currents without electrical interference requires sophisticated methods. To overcome this problem it has been theoretically proposed to connect a large number of microelectrodes (eg. 400 microelectrodes) in parallel to form an assembly. Such assemblies would possess all the advantages of single microelectrodes yet the total measured electrochemical current would be large.

The critical specifications defining microelectrode assemblies are very difficult to meet and none of the known prior attempts to develop such assemblies is suitable for commercial production.

To make the manufacturing process worthwhile, the diameters of the individual microelectrodes must be less than 10 micrometers, and, because the total measured current at an assembly is the sum of the individual currents, the number of microelectrodes must be in the hundreds. Clearly, a method of manipulating and orienting large numbers of microelectrodes must be developed. Conspiring against this goal, each microelectrode must be separated by about 70 micrometers from its neighbours to ensure that it behaves independently. In addition, a great deal of materials science knowledge is also needed to ensure that the finished product is mechanically robust and chemically inert and may easily be cleaned and reused.

One prior art method for constructing a microelectrode assembly comprises laying carbon fibres across a long narrow strip of copper foil which has been coated with a thin layer of silver-doped conducting epoxy resin. The ends of the carbon fibres protrude over one side of the copper foil. The foil is then rolled up into a cylinder approximately 1 cm in diameter, after which it is placed in an oven to cure the epoxy. (At this stage the rolled up foil looks somewhat like a small shaving brush). Next, the whole workpiece is potted in a non-conducting epoxy resin that has a sufficiently low viscosity for any air bubbles trapped between fibres to be removed under vacuum. After a second cure, the now-rigid workpiece is sectioned at right angles to the fibres to produce the desired assembly of microdisk electrodes. Electrical contact is made to the electrodes by soldering a wire to the copper foil at the rear of the workpiece. Finally, the electrode surface is handpolished using successively finer grades of alumina. The production of such an assembly requires high levels of manual dexterity, is very labour intensive and it is difficult to maintain quality control.

It is an object of the present invention to overcome or alleviate at least one of the problems associated with prior known microelectrode assemblies. In particular it is an object to develop a mechanically robust and chemically inert microelectrode assembly which can be easily cleaned and reused and which can be produced in large numbers.

According to the present invention there is provided a microelectrode assembly including a substantially uniform dispersion of a plurality of substantially parallel conducting fibres embedded in a non-conducting material.

By microelectrode assembly we mean an assembly of microelectrodes in a non-conducting matrix and the term includes the basic electrode without electrical connection(s) between the microelectrodes.

The microelectrode assembly may have a cross-section of any shape. Preferably the microelectrode assembly has a circular, hexagonal or annular cross-section. The microelectrode assembly may have any thickness below about 5 mm.

Preferably the thickness of the assembly is in the - A - range of about 3 mm to about 5 mm.

The conducting fibres may be, for example, metal fibres, metal coated fibres or carbon fibres. The conducting fibres are preferably carbon fibres. Preferably the median nearest-neighbour distance between the fibres is approximately 70 μm or greater.

The non-conducting material may be any insulator. The non-conducting material may be a thermoset resin, a thermoplastic resin or a ceramic. The thermoset resin may be selected from any suitable thermoset resin. The thermoset resin may be selected from phenolic resins, polyester resins, epoxide resins, or silicone resins or elastomers. An epoxy resin is particularly preferred. The thermoplastic resin may be selected from polyethylene, polyvinyl chloride, polypropylene, polystyrene, polyether ether ketone (PEEK), or a fluorinated polymer such as PTFE. The non-conducting material may also be a ceramic insulator.

Further, according to the present invention there is provided a method of making a microelectrode assembly which includes: i) adding a plurality of conducting fibres to a non-conductive flowing material; ii) dispersing the conducting fibres in the non-conductive flowing material; iii) introducing the mixture into a mould such that the fibres are in substantially parallel alignment; iv) inducing solidification of the non-conductive flowing material/fibre mixture in the mould; and v) releasing the solidified mixture from the mould and sectioning it to form at least one microelectrode assembly.

The non-conductive flowing material used in the method of the invention may be a viscous fluid or a powder. The non-conductive flowing material should be capable of forming a solid by, for example, curing, temperature and/or pressure related phase change or sintering. The non-conductive flowing material may be a thermoplastic resin, a thermosetting resin or a ceramic powder.

The thermosetting or thermoplastic resin should have a viscosity such that in step i) of the method of the invention there is inter-fibre penetration by the resin. Where a thermoplastic resin is used it should be heated to a temperature sufficient to allow it to flow. If the thermosetting resin does not have the requisite viscosity at ambient temperature, it may be heated to reduce its viscosity. Where the thermosetting resin requires a hardener or curing agent to cure, it is preferable that step i) be carried out in the absence of the hardener or curing agent. Preferably the thermosetting resin has a viscosity, at room temperature or upon heating, of less than 1000 cP. Preferably the thermosetting resin has a work time of at least 1 hour and has a very low exotherm. It is desirable that the thermosetting resin undergo little or no volume change on curing and have a high chemical resistance when cured. Preferably the product formed by curing the thermosetting resin absorbs little or no solvent and is optically transparent. The thermosetting resin and the resultant cured thermoset resin is preferably an epoxy resin. Where a thermoplastic resin is used, the resin is solidified in step (iv) of the method by cooling the mould. Where a ceramic powder is used it is solidified by heating the mould to cause sintering.

Preferably the conducting fibre is present at a loading of no more than 5% by volume based on the total volume of the assembly. More preferably the conducting fibre is present in an amount less than about 1% by volume. A conductive fibre content of approximately 0.7% by volume is most preferred.

Preferably the conducting fibre has a low electrical resistivity. Preferably the resistivity of the conducting fibre is less than about 20 μΩm. Preferably the conducting fibre has a diameter less than about 50 μm. More preferably the diameter of the fibre is in the range of about 4 to 10 μm with a diameter of about 6 to 7 μm being particularly preferred.

Any suitable number and length of conducting fibres may be used in the method of the invention. Typically, for a microelectrode assembly having a thickness of 3 mm, the length of the fibres should initially be approximately 6 mm.

The length of the conducting fibre is governed by the diameter of the mould in that the length of the fibre should be less than the diameter of the mould. Preferably the length of the conducting fibre used in the method of the invention is less than about 10 mm. A fibre length of about 6 mm is particularly preferred.

The conducting fibre is preferably "sized" (i.e. coated) for the particular thermosetting resin used so as to provide a thin layer of an adherent compound which can be wetted easily by the thermosetting resin. Fibres sized in this way bond strongly to the cured resin and are less susceptible to "pull out" from the cured composite. The conducting fibres may be sized during manufacture. The number of fibres used in the method of the invention depends upon the diameter of the microelectrode assemblies being formed, the volume of the mould, and the final number of fibres desired in the finished product. In a preferred embodiment the number of fibres is selected such that it is in the range of approximately 1600 to 5000 functioning microelectrodes per square centimetre of the finished product.

The invention will hereinafter be described in reference to epoxy resins, however the invention is not limited to this particular non-conducting material.

Any suitable low viscosity epoxy resin may be used in the method of the invention. Ideally the epoxy resin will be chemically inert with respect to the end use of the microelectrode assembly to be formed. It will also be of a suitable viscosity such that it will allow the fibres to become substantially aligned when the epoxy resin/fibre mixture is introduced into the mould. An example of a suitable low viscosity epoxy resin is Epimount from Epirez Australia, Construction Products (Melbourne, Australia). Other examples of suitable epoxy resins are Araldite K99, Araldite LC191 with hardener LC226 and Araldite LC3600 with hardener LC3600. An example of suitable conducting fibre is Sigrafil HM 48 B carbon fibres from Sigri GmbH (Meitingen, Germany) .

The conducting fibres may be dispersed in the epoxy resin by any suitable methods. For example the conducting fibres may be dispersed by agitation or stirring. In a preferred embodiment of the invention, the conducting fibres are dispersed in the warmed resin before the hardener is added. In this case the resin may be warmed to a temperature within the range from 40°C to 105°C. It is important that the conducting fibres be sufficiently dispersed such that when the epoxy resin/fibre mixture is introduced into the mould the fibres are in substantial parallel alignment and free of clustering. For example, if 1% by weight of conducting fibres is dispersed in 200 ml of epoxy resin in a rotary evaporator, agitation is preferably continued for approximately 1.5 hours. If desired, the level of dispersion of fibres may be tested by introducing the resin into a mould prior to adding a hardener to ensure that clustering is absent.

Dispersion of the fibre in step (ii) of the method of the invention may include the following steps:

(a) adding the conducting fibres to the resin which has been heated to a temperature about 90°C and stirring the mixture; and

(b) cooling the resin/conducting fibre mixture, if necessary adding hardener or curing agent to the resin, and slowly tumbling the mixture.

Preferably the resin in step (a) is heated to a temperature between about 80°C to 90°C.

More preferably step (ii) of the method includes the following steps:

(a) adding the conducting fibres to the heated resin and stirring the mixture for a period sufficient to disperse the fibres in the absence of a hardener or curing agent; (b) cooling the fibre/resin mixture and rapidly stirring the mixture; and

(c) subjecting the mixture to a tumbling action to maintain the dispersion of the conducting fibres during the addition of a hardener or curing agent. Preferably in step (a) the mixture is heated to a temperature about 90°C, more preferably 80°-90°C.

Preferably in step (b) the mixture is cooled to about 25°C. Preferably the rapid stirring step is carried out for less than 1 minute, and more preferably for about 10 seconds using a protruberance-free stirrer such as a smooth glass rod. In a preferred embodiment the dispersion of the fibres is carried out under vacuum in the rotary evaporator, which removes bubbles and dissolved air. Preferably the viscosity of the mixture poured into the mould is in the range of approximately 700 cP to 20,000 cP. More preferably the viscosity of the mixture is in the range of about 10,000 to 12,000 cP.

The epoxy resin/fibre mixture may be introduced to the mould by any suitable method.

The mould used in the method of the invention preferably induces Poiseuille flow to the resin/fibre mixture. A tube of circular or hexagonal cross-section may be used to induce Poiseuille flow. The diameter of the tube is preferably less than 10mm. Preferably the mould has a portion having sloping sides to induce substantially parallel alignment of the conducting fibres in the resin mixture prior to being subjected to Poiseuille flow. The portion having sloping sides may be a funnel. In a preferred embodiment the mould has an integral funnel at the top through which the epoxy resin/fibre mixture is poured. The angle of the sides of the funnel is preferably about 60°. The junction between the funnel and the mould is smooth-walled.

The mould may be formed from any suitable material. Ideally it is of a transparent material from which the cured epoxy resin/fibre mixture can be easily released. For example, any suitable glass tubing may be used. In a preferred embodiment, the main body of the mould consists of a glass tube of about 8 mm internal diameter and about 500 mm in length. In this case, the volume of the funnel is at least 200 ml. To assist with the release of the cured epoxy resin/fibre mixture from the mould a release agent may be coated on the inside walls of the mould. Suitable release agents include Release Agent from Leco Corporation (St. Joseph, USA). Many difficulties may arise when the epoxy resin/fibre mixture is introduced into the mould. A common difficulty is the entrainment of air bubbles. To overcome this difficulty the mould is preferably flushed with fibre-free deaerated epoxy resin mixture prior to the introduction of the epoxy resin/fibre mixture. The curing may also be carried out under pressure.

Another difficulty experienced during the formation of the microelectrode assemblies is the gradual misalignment of the fibres during the early part of the curing phase of the epoxy resin/fibre mixture. To overcome this difficulty, a vacuum may be briefly applied to one end of the mould (or pressure may be applied to the other end of the mould) which causes flow of the mixture to resume temporarily and thus realigns the fibres. After the epoxy resin/fibre mixture has cured at room temperature, it may be post-cured in an oven. In a preferred embodiment post-curing takes place at 50°C. Thereafter, the casting may be released from the mould by any suitable method. For example, it may be released by gently tapping the casting or by cutting the mould into sections and removing each section from around the casting. Following post-curing of the casting it is typically in the form of a rod which can be sectioned into disks.

These must be thin enough for a plurality of fibres to intersect both faces of each disk.

After cutting the rod into disks, the faces of the disks are polished. It is at this stage that the fibres appear as microdisks embedded in each face.

Electrical contact may then be made to one face - the other face forms the working surface of the assembly.

The electrical contact may be formed by any suitable method. In a preferred embodiment the electrical contact is formed by ion beam sputtering of a noble metal such as platinum.

Finally, the assembly may be cast in non-conducting epoxy resin to insulate any fibres that may be exposed on the sides of the disks.

The invention will now be described with reference to some examples. The examples are useful for illustrating the scope of the invention and the advantages thereof and should not be seen as limiting in any way.

Brief Description of Drawings

Figure 1 is a schematic cross section of a mould for use in the method of the invention; Figure 2(a) is a schematic section view of an electrode arrangement including a microelectrode assembly in accordance with one form of the invention;

Figure 2(b) is a schematic plan view of a microelectrode assembly in accordance with the invention; Figure 3 is topography of part of a surface of a microelectrode assembly in accordance with the invention revealed by Atomic Force Microscopy. Figure 4 is a graph showing electrochemical currents recorded in response to triangular scans of applied potential in an aqueous solution of 10 —3M K₄Fe(CN)_fi and 10^~3M K₃Fe(CN)₆ in 10^-1M KNO„. The scan limits used were -180 mV to 620 mV (vs Ag/AgCl/KCl) .

EXAMPLES

Example 1

Sigrafil HM 48 B fibres from Sigri GmbH (Meitingen, Germany) are cut into 6 mm lengths, and the weight of cut lengths needed to give the fibre density chosen for the final product is added to the appropriate weight of Epimount mounting resin from Epirez Australia Construction Products (Melbourne, Australia) . The mixture is heated to 80°C and stirred briskly for 15 min using a low shear stirrer which does not cut or break the carbon fibres.

The resulting partially dispersed mixture is transferred to a round-bottom flask which is then mounted onto a rotary evaporator, and held between 80°C and 90°C under vacuum for about lh to deaerate. The temperature is then reduced to 25°C and the mixture tumbled slowly under vacuum until the fibres are fully dispersed. This latter procedure takes about 1.5h. The hardener is then added and the epoxy resin/fibre mixture is returned to the rotary evaporator and is tumbled slowly under vacuum at 25°C to thoroughly mix and deaerate again.

A suitable mould is shown in figure 1. The mould 20 consists of a tubular glass stem 22 500 mm in length having a funnel 24 smoothly joined at the top. The angle of the sides of the funnel is 60°. The junction between the funnel and the mould is smooth-walled.

The moulds are cleaned, and their internal surfaces are coated with a thin film of Release Agent from Leco Corp. (St. Joseph, USA). A valve for regulating fluid flow in the tube is fitted to the bottom. The epoxy resin/fibre mixture is allowed to partially cure in the rotary evaporator until its viscosity reaches about 15,000 cP. This takes about 1.5h to 2.Oh after adding the hardener. At this stage the mould is filled with a fibre-free, bubble-free, deaerated epoxy resin mixture and, when the walls of the mould are thoroughly wet, it is drained out. The epoxy resin/fibre mixture is then immediately poured into the funnel taking care not to entrain any air bubbles, and the flow 26 is allowed to continue until the mould is filled with the mixture and the fibre alignment is good. At 20 min to 30 min intervals therafter, a vacuum is applied to the bottom end of the mould and the flow is temporarily restarted to realign fibres which have become misaligned, and this process is continued until the viscosity of the mixture is too high for flow to occur, even with vacuum assistance. The final mixture is then allowed to cure at room temperature for 12h to 15h, and then is postcured at 50°C for 36h to 48h. The cast product is released from the mould and fibre alignment is checked using a strong light source. If the alignment is good, a slow-speed wafering saw is used to cut 4mm thick disks from the rod. The disks are temporarily mounted with one face down onto brass keepers using double-sided Sellotape from Wrightcel (Melbourne, Australia), and the exposed faces are polished using successively finer grades of Hypres Five Star diamond polishing compound from Engis Ltd, (Maidstone, England) finishing with 1 micrometer grit size. The polished faces are then immediately coated with a conducting layer of platinum using an ion-beam sputter coating apparatus.

After removal from the mould, the electrodes are machined to the final dimensions, and the sidewalls are polished to a mirror finish so that they become hydrophobic. The hydrophobicity can be enhanced with a variety of proprietary reagents. Alternatively the hydrophobic agent may be incorporated into the epoxy resin used to form the assembly so as to impart hydrophobicity to the assembly. This minimises the possiblity for current to bypass the microdisks and leak along the sides of the electrodes. Finally the working surfaces of the electrodes are polished in exactly the same manner as the surfaces of the disks prior to coating with platinum.

Stainless steel connectors for making electrical contact to the platinum film on the disk faces are prepared by grit blasting and degreasing, then they are mounted on to the disks using E-Solder silver-doped conducting epoxy resin from Acme Chemicals and Insulation Co., (Connecticut, USA). The mechanical adhesion of the contact is maximised by curing at 60°C for 12h. The assemblies are then removed from the brass keepers, and the disks are machined concentric with and to the same diameter as the metal connectors. The resulting workpieces are embedded in Epimount mounting resin, leaving a short section of each stainless steel connector exposed at the rear for electrical contact, and when the resin has cured they are machined to size. Finally, the working face of each assembly is polished to expose the conducting carbon fibres as disks.

Figure 2(a) shows one form of an electrode arrangement incorporating a microelectrode assembly according to the invention. The electrode 2 is made up of microelectrode assembly 4 which contains a plurality of carbon fibres 14 located in a cured epoxy matrix 12. At least a portion of the carbon fibres extend between faces 16 and 18 of the assembly with each end of the fibre exposed at one of the faces. An electrically conducting layer 6 provides electrical connection to current collector 10 which may be a conductive metal such as stainless steel.

A plan view of the microelectrode assembly portion of the electrode is shown in figure 2(b) which shows the random distribution of the substantially parallel carbon fibres 14 in the cured epoxy matrix 12. Figure 2(a) is not to scale and the carbon diameters of the carbon fibres as shown have been enlarged for the purpose of more clearly showing the random nature of their location.

Figure 3 shows the topography of a surface region 34 which is part of the surface of an electrode assembly of the invention revealed by Atomic Force Microscopy. The depressions 32 7 μm in diameter are the exposed ends of the carbon fibres.

In Figure 4 there is shown the electrochemical currents recorded for a carbon macrodisk (lower trace) and for a carbon microdisk assembly in accordance with the invention (upper trace) in response to triangular scans of applied potential in a solution of 10 M K.Fe(CN)_ and 10 -3M K_Fe(CN)_c in 10-1M KNO_. The scan limits used were -180mV to + 620mV (vs Ag/AgCl/KCl) .

The shape of the 'upper trace in Fig. 4 shows that the reduction and oxidation currents recorded on the assemblies quickly reach their diffusion-limited values as the potential is varied around +0.22V. However, once they have reached their diffusion-limited values, then the currents vary very little, regardless of whether the potential is held or whether the scan continues, indicating that steady-state currents are being measured. By contrast, the shape of the lower trace shows a peak in both the reduction and oxidation currents. The observation of steady-state currents on the assemblies proves that they behave as assemblies of independent microdisks, and not as pseudo-macroelectrodes.

Example 2

Fibre dispersion begins when shear is introduced in the epoxy resin/fibre mixture. However, if the shearing process is continued too long the fibres can be fractured, in which case their average length is decreased to the point where the mixture is useless for making the electrode assemblies. A further complication is that the viscosity of the mixture increases significantly during the cure and there is a strong tendency to trap air bubbles. To cope with these problems a three step process has been found to successfully disperse Sigrafil HM 48 B fibres in Epimount resin at fibre concentrations below 0.7% w/w.

In the first step the resin monomer only (no hardener) is heated on a hotplate to approximately 80 to 90°C and stirred using a mechanical stirrer. A cylindrical stirrer is used because, unlike more conventional stirrers, it introduces shear into the mixture without breaking the fibres. The fibres are added in batches over a period of about 5 min, and the stirring is continued for a further 15 to 20 min to remove the larger tangles only. A magnetic stirrer bar is not suitable for this step as it grinds the fibres to shorter and shorter lengths, as does the stirring cylinder if it touches any part of the container during use.

In the second step the hot fibre/resin mixture is cooled to 25°C and briskly stirred by hand using a glass stirring rod for 5 s only. Care should be taken during this step not to grind the rod against the walls or the bottom of the container as this breaks the fibres. At the end of this step, the mixture should look well dispersed. In the third step, the mixture is transferred to a suitable vessel which is attached to a rotary evaporator (rotavap) and a water bath is used to keep the temperature of the mixture at 60°C to 80°C. The vessel is then evacuated to deaerate the mixture, and during this stage the mixture should be tumbled. Different batches of resin do behave differently, and some may take longer to deaerate than others, but generally this step takes 15 to 30 min. Full dispersion of the fibres is maintained by reducing the temperature of the mixture to 25°C using the water bath, adding the hardener, then tumbling the mixture under vacuum for 15 to 20 min. The degree of dispersion may be judged by viewing the mixture through the walls of the vessel as it rotates. Full dispersion has been achieved when no fibre tangles are visible, and at this point the mixture should not be tumbled any longer, although it should be left under vacuum until it is ready to be poured.

The mould used in this example had a funnel diameter of 100 mm, apex angle of 60°, stem length of 500 mm (to produce a cylindrical cross section) and 130 mm (to produce a hexagonal cross section) . The internal dimensions of the stem were 8 mm diameter for the cylindrical cross section and 8 mm diameter inscribed circle for the hexagonal cross section.

Good results were obtained with both moulds.

The moulds used in this example were made from glass, although any material which may be formed into shape and which will release cured epoxy resin is suitable. It is important that the join between the funnel and the stem be smooth and not interfere with the flow of the fibre/epoxy resin mixture. All internal surfaces of the mould were cleaned and then coated with a thin layer of a release agent from the Leco Corporation (Michigan, U.S.A). A short length of rubber tube was attached to the outlet of the stem, onto which was fastened a clamp to control the flow of epoxy resin.

Before the fibre/epoxy resin mixture was poured, the mould was primed by filling it with degassed, fibre-free epoxy resin. The hardener was added to the priming epoxy resin at the same time as the hardener was added to the fibre-containing mixture, so their viscosities were similar at the time of pouring. This precaution reduced the formation of wall-bound bubbles during the pour, and also minimised fibre tangles forming at the stem walls as the fibre mixture flowed down the mould. Approximately 25 mis of resin was required for the moulds having a hexagonal stem, and 50 mis for the moulds with a cylindrical stem. Pumping the rubber tube at the base of the mould when the mould was primed dislodged any small bubbles on the walls of the mould and forced them to the centre of the column of liquid resin. When the pour commenced these bubbles drained with the priming epoxy resin.

When the fibres had been thoroughly dispersed in the epoxy resin and the mixture had been degassed, the mixture was left under vacuum until its viscosity increased to about 10,000 to 12,000 cP. This normally takes 1.5 to 2.0 hours after the hardener has been added, but the time varies with different batches of epoxy resin. At such viscosities the mixture flows only slowly when it is poured. Once this viscosity is reached, the priming epoxy resin is allowed to drain from the mould until only the stem is filled, and the fibre containing mixture is then poured carefully into the funnel in such a way that no air is entrained. The valve at the end of the stem is closed throughout the pour. The ideal volume of the mixture to prepare depends on the mould used. For the moulds described above the ideal volumes are:

100 mm diameter funnel, 500 mm cylindrical stem: 100 mis 100 mm diameter funnel, 130 mm hexagonal prism stem: 75 mis.

To force the mixture to flow down the mould and align the carbon fibres a suitable apparatus is joined to the mould via the rubber tubing at the stem outlet. A Buchner funnel connected to a vacuum pump and sealed with a rubber bung through which a short length of glass tubing has been fitted, is ideal for this. The tube clamp at the stem outlet is opened, and the flow is allowed to continue until the first of the fibre-containing mixture exits the mould. The tube clamp is then closed, and, at this point, the fibre alignment should be excellent for the full length of the tube. The mould is left for 10 to 30 minutes to allow the viscosity to increase further, and during this time it is not uncommon to observe the degree of fibre alignment to degrade slightly. Preferably, the flow should be resumed to restore fibre alignment only when the viscosity of the mixture has risen to a point where flow is very slow, even under full vacuum. This guarantees that any further degradation in the degree of alignment is negligible.

Three important points should be noted regarding this step:

(i) the flow should be allowed to proceed only long enough to establish good fibre alignment, otherwise fibre tangles develop at the stem walls and can build up excessively,

(ii) if the viscosity of the mixture is too low at the time of pouring, the rate at which the fibres become misaligned on standing can be appreciable, and (iϋ) some care should be taken to ensure that the volume of mixture remaining in the funnel is no greater than about 50 to 70 mis, otherwise thermal runaway of the epoxy resin may occur as it cures.

When the desired alignment in the mould has been achieved and the viscosity of the mixture has risen above approximately 20,000 cP the mould is left for about 24 hours at room temperature (25°C) and elevated pressure to cure. In this work the castings were cured in a vessel that was pressurised to 3 MPa (435 psi) with nitrogen. Pressure curing is desirable because voids can form in epoxy resin castings that are cured at ambient pressure, and also air bubbles tend to re-dissolve at high pressure.

Most epoxy resins will achieve 90% (or more) of their ultimate strength if they are cured at room temperature for 24 h. However the casting should be post-cured at an elevated temperature to achieve the maximum performance from the epoxy resin. The optimum conditions for post-curing will depend upon the resin used, but in this work the castings were left for a further 24 h at 50°C. Experience has shown that a cured casting can be released without damaging the mould if the release agent has been applied uniformly and if the bore of the stem does not vary along its length. If either of these conditions is not met then the mould will have to be destroyed to recover the casting. Of course, if glass moulds are used, then particular caution needs to be taken during this step. Whether or not the mould is destroyed during this step, the casting should be recovered in one piece, and the rod (the section recovered from the stem of the mould) should be checked for fibre alignment and the absence of voids. If necessary, the rod is subjected to a preliminary sectioning process to recover the best-aligned lengths using a slow speed, water cooled, precision diamond saw, such as a Buehler Isomet (Buehler, Illinois, U.S.A.). Then these are further sectioned into disks of precise thickness - 2.60 mm was chosen for the present work. The disks are inspected to ensure they are void free and that the carbon fibres are well dispersed, and those passing this test are mounted onto a stainless steel keeper using double-sided adhesive tape in preparation for polishing. The keeper used in this work was 45 mm in diameter onto which 22 disks could be mounted.

The front surfaces of the disks mounted on the stainless steel keepers are polished to a mirror finish using successively finer grades of polishing abrasive, and an automated polishing machine such as a Kent 2A (Engis Ltd., Maidstone, U.K.) is used for this step. If a slow speed diamond saw is used to cut the disks initially then it is necessary to commence polishing with 25 μm abrasive or P1200 silicon carbide paper. Care must be taken when using this abrasive not to grind away too much of the disks. The finer grades of abrasive found suitable in this work were 14 μm, 6μm, 3μm, and lμm diamond paste, and 0.3 μm alumina, all obtained from Kemet Australia Pty. Ltd. (East Bentleigh, Victoria, Australia) . Inspection of the disk surfaces under the microscope is the most effective way for judging the effectiveness of each polishing step, and it is important that the polish with each grade be perfect before proceeding to the next.

The final surface preparation involves two steps: a mechanical polish and a chemical treatment, and once they have been carried out the surfaces of the disks are coated immediately with platinum. The final polish can be completed by hand on a napped polishing cloth with a water-dispersed slurry of the 0.3 μm alumina, and the polishing detritus can be removed by rinsing the disks with a jet of distilled water before wiping the disk surfaces on clean, napped polishing cloth wetted with distilled water. The chemical treatment involves immersing the disks in a 10% v/v aqueous solution of nitric acid for 10 s, then rinsing them thoroughly with a jet of distilled water. The water is completely removed from the disks before placing them into the coating apparatus. At this stage no scratches should be visible on the surfaces of the disks to the naked eye, and only a few small ones should be visible at 400 times magnification under the microscope. No polishing detritus should be present.

Electrical contact with the exposed carbon fibres is made by coating the polished surfaces of the disks with an electrically conductive layer of platinum. An effective method for doing this is to sputter platinum from a target onto the disks using an ion beam. Although slower than other forms of sputtering, ion beam sputtering is preferred because it produces a particularly durable film which is capable of withstanding the mechanical stresses produced when the electrode is machined. Further electrical connection can then be made via silver-doped epoxy resin.

The platinum film is deposited until the electrical resistance across the electrode disk (8 mm diameter) has fallen below 20 Ω. A film several hundred nanometers thick is necessary for this condition to be met, which may take several hours to deposit. The platinum coat should have a mirror finish with no blemishes present.

Electrical connectors of some suitable material are prepared and attached to the platinum coated disks as soon as they are recovered from the coating apparatus. It is recommended that connectors be used which have the same diameters as the final diameters of the fibre containing disks, and not ones which are attached directly to fractions of the disk only. Full size connectors ensure that electrical currents are conducted directly to or from every fibre through the platinum and silver-doped epoxy film, rather than laterally along some part of this film as would occur if undersized connectors were used. In this work, stainless steel connectors were used, which had been grit blasted, sonicated in distilled water to remove any remaining grit, and degreased in a bath of 2-propanol (isopropyl alcohol) vapour for 30 min. These connectors were glued to the disks using a thin layer of silver-doped electrically conducting epoxy resin from Insulating Materials Incorporated, Acme Division (New Haven, Connecticut, U.S.A.), and the assemblies were cured at 50° for 12 h to maximise the adhesion of the joints. During this procedure the connectors were handled using gloves to protect them from contamination which would otherwise compromise the strength of the bond formed between the connector and the epoxy resin used for embedding the assemblies.

The assemblies are machined so that the fibre containing disks are concentric with and the same diameter as the connector. The connectors are degreased again to remove any contamination from the machining step by immersion in 2-propanol for 15 min.

The assemblies are embedded in the same epoxy resin as was used for making the fibre containing disks. The variety of shapes which may be formed in this step is limited only by the standard requirements for casting with thermosetting resins. However, it is strongly recommended that the epoxy resin used for embedding be degassed, be poured at a low viscosity, and be pressure cured. A suitable material for making the mould to be used for this step is high molecular weight poly(ethylene) (HMWPE), which is inexpensive and releases cured epoxy resin easily. The final step in preparing the electrodes is to polish their front surfaces. These are polished in exactly the same manner as were the back surfaces of the disks prior to coating with platinum. Normally, polishing commences using 25 μm diameter abrasive compound, or P1200 silicon carbide papers. Care must be taken when using this abrasive not to grind away too much of the electrode. Successively finer grades of abrasive must then be used to obtain a mirror finish. The finer grades of abrasive found suitable in this work were 14 μm, 6 μm, 3 μm, and 1 μm diamond paste, and 0.3 μm alumina, all obtained from Kemet Australia Pty. Ltd. (East Bentleigh, Victoria, Australia). In this work the last polish with 0.3 μm alumina was carried out by hand on clean, napped polishing cloth. An optional step that may be applied to the finished product is to polish the side walls of the electrodes to a mirror finish so that they become hydrophobic. The hydrophobicity can also be enhanced with a variety of proprietary reagents. This minimizes the possibility for current to bypass the microdisks and leak along the sides of the electrodes.

TYPICAL USES FOR MICRODISK ASSEMBLIES

Microdisk assemblies of the present invention are useful in any situation requiring electrochemical detection with good signal-to-noise ratio. For example: as electrodes for anodic stripping voltammetry, as detectors in liquid chromatography, as viscosity-sensing devices.

Also a large market for microdisk assemblies has been identified in environmental monitoring. A number of industry segments, all of which are potential end users, are the following: environmental testing agencies, water utilities, metals production and metallurgy, mining, medical/clinical, education, pharmaceutical, agriculture and aquaculture, research and private laboratories, food and beverages, process waters, and petroleum, petrochemical and chemical.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as described without departing from the spirit and scope of the invention. The present examples and specific details are therefore to be considered in all respects as illustrative of the invention and not restrictive.

Claims

CLAIMS :

1. A microelectrode assembly including a substantially uniform dispersion of a plurality of substantially parallel conducting fibres embedded in a non-conducting material.

2. A microelectrode assembly according to claim 1 wherein the conducting fibres are present in an amount up to about 5% by volume based on the total volume of the non-conducting material and the conducting fibres.

3. A microelectrode assembly according to claim 2 wherein the conducting fibres are present in an amount less than 1% by volume based on the total volume of the non-conducting material and conducting fibres. 4. A microelectrode assembly according to any one of the preceding claims wherein the non-conducting material is selected from a thermoset resin, a thermoplastic resin or a ceramic.

5. A microelectrode assembly according to claim 4 wherein the non-conducting material is a thermoset resin.

6. A microelectrode assembly according to claim 5 wherein the thermoset resin is an epoxy resin.

7. A microelectrode assembly according to any one of the preceding claims wherein the conducting fibres are carbon fibres.

8. A microelectrode assembly according to any one of the preceding claims wherein the conducting fibres each have a diameter less than about 10 μm.

9. A microelectrode assembly according to claim 8 wherein the carbon fibres each have a diameter in the range of about 4 to 10 μm.

10. A microelectrode assembly according to claim 9 wherein the conducting fibres each have a diameter of about 7 μm. 11. A microelectrode assembly according to any one of the preceding claims wherein the microelectrode assembly has a thickness up to about 20 mm. 12. A microelectrode assembly according to claim 11 wherein the microelectrode assembly has a thickness in the range of about 3 to about 5 mm.

13. A microelectrode assembly according to any one of the preceding claims wherein the sides of the assembly have been hydrophobized.

14. A microelectrode assembly according to any one of the preceding claims further including a conducting layer on one face of the microelectrode assembly making electrical contact with the exposed ends of at least a portion of the conducting fibres.

15. A microelectrode assembly according to claim 14 further including an electrical connector attached to the conducting layer. 16. A method of making a microelectrode assembly which method includes: (i) adding a plurality of conducting fibres to a non-conducting flowing material; (ii) dispersing the conducting fibres in the non-conducting flowing material;

(iii) introducing the non-conducting flowing material/ fibre mixture into a mould such that the fibres are in substantially parallel alignment; (iv) inducing solidification of the non-conductive flowing material/fibre mixture in the mould; and

(v) releasing the solidified mixture from the mould and sectioning the solidified mixture to form at least one microelectrode assembly.

17. A method according to claim 16 wherein the non-conductive flowing material is selected from a thermoset resin, a thermoplastic resin or a ceramic powder.

18. A method according to claim 17 wherein the non-conductive flowing material is a thermosetting resin.

19. A method according to claim 18 wherein the thermosetting resin is an epoxy resin.

20. A method according to any one of claims 16 to 19 wherein the conducting fibres are carbon fibres. 21. A method according to any one of claims 16 to 20 wherein the diameter of each of the conducting fibres is in the range of about 4 μm to 10 μm.

22. A method according to claim 21 wherein the diameter of each of the conducting fibres is about 7 μm.

23. A method according to any one of claims 16 to 22 wherein the conducting fibres are present in an amount up to about 5% by volume based on the total volume of the non-conductive material and the conducting fibres. 24. A method according to claim 23 wherein the conducting fibres are present in an amount of less than about 1% by volume.

25. A method according to any one of claims 16 to 24 wherein the microelectrode assembly has a thickness up to about 20 mm.

26. A method according to any one of claims 16 to 25 wherein the mould includes a funnel, the geometry of which causes a substantial alignment of the conducting fibres, communicating with a stem portion suitable to induce Poiseuille flow in the resin/fibre composition passing therethrough.

27. A method according to claim 26 wherein the angle of the sides of the funnel is about 60°.

28. A method according to claim 26 or 27 wherein the stem portion is a tube of circular or hexagonal cross section.

29. A method according to claim 28 wherein the tube has an internal diameter less than about 10 mm.

30. A method according to claim 29 wherein the tube has an internal diameter of about 8 mm and a length of about 500 mm

31. A method according to any one of claims 17 to 30 wherein the viscosity of the thermosetting resin in step (i) is such as to permit inter-fibre penetration by the thermosetting resin. 32. A method according to any one of claims 17 to 31 wherein step (ii) includes the steps of

(a) adding the conducting fibres to a heated thermosetting resin, in the absence of a hardener or a curing agent, stirring the conducting fibre/thermosetting resin mixture for a period sufficient to partially disperse the conducting fibres; (b) cooling the resin and rapidly stirring the conducting fibre/thermosetting resin mixture; and

(c) adding a hardener or curing agent, and subjecting the mixture to a tumbling action to maintain the dispersion of the conducting fibres in the thermosetting resin.

33. A method according to claim 32 wherein in step (a) the thermosetting resin is heated to a temperature in the range of about 80°C to 90°C and in step (b) the resin is cooled to a temperature of about 25°C. 3 . A method according to claim 32 to 33 wherein the rapid stirring in step (b) is carried out for about 10 seconds.

35. A method according to any one of claims 16 to 34 further including the step of depositing a conductive medium on one face of the microelectrode assembly.

36. A method according to claim 35 wherein the conductive medium is deposited by vapour deposition or sputtering.

37. A method of claim 35 or 36 further including the step of attaching an electrical connector to the conductive metal coating on said one face of the microelectrode assembly.

38. A method according to any one of claims 16 to 37 wherein the side walls of the microelectrode assembly are hydrophobized to impart hydrophobicity.