WO1998033054A1

WO1998033054A1 - Method and apparatus for counting and/or sizing particles in suspension

Info

Publication number: WO1998033054A1
Application number: PCT/GB1998/000208
Authority: WO
Inventors: Peter John Houzego; Sarah Anne Rocca; David Michael Coombs; Nicholas Martin Goy
Original assignee: Pa Knowledge Limited
Priority date: 1997-01-24
Filing date: 1998-01-23
Publication date: 1998-07-30
Also published as: GB9701457D0

Abstract

Disclosed is apparatus for counting and/or sizing particles held in suspension in a conductive liquid, the apparatus comprising: a length of tube of substantially constant bore size; pressure application means by which the conductive liquid can be caused to pass through the substantially constant bore tube; and particle counting/sizing means which detects variations in electrical signals, which variations are caused by the passage of a particle through the tube past the counting/sizing means; wherein the counting/sizing means does not interrupt the substantial constancy of the bore of the tube.

Description

Title Method and Apparatus for Counting and/or Sizing Particles in Suspension

Field of the Invention

This invention relates to a method of and apparatus for sizing and/ or counting particles, especially sub-millimetre diameter particles, held in suspension in a conductive liquid.

Background to the Invention

One well known technique for sizing particles in suspension is the long established Electrical Zone or Coulter particle sizing technique. In this technique two containers with a common wall containing a small orifice are filled with an electrical conductive liquid (such as water and a salt). Platinum electrodes are immersed in each container and a d.c. voltage applied between them. Electrical current is carried from one electrode through the bulk of the liquid to pass through the orifice into the second container and hence to its electrode. The current density through the liquid is low except through the orifice whose small area carrying all of the current produces a high current density. The particles to be measured are placed in one of the containers and stirred to keep them suspended. A pressure difference is induced between the two containers causing liquid and the suspended particles to flow through the orifice. As each particle passes through the orifice it impedes the electrical current flow considerably more than when it was away from the orifice as the current density is so much higher in this region. The increased impedance reduces the current flow and the magnitude of this reduction can be used as a measure of the particle's volume.

Various embodiments of this technique are known. However, all embodiments have a severe limitation in that after passing through the orifice the particle emerges into the large volume of the second container where it is quickly mixed with other particles, so that its identity is lost.

It has become conventional to use small particles as a support for chemical moieties. The particles are normally substantially spherical and typically comprise a substantially inert, porous matrix bead to which certain chemical groups may be attached. Such beads are commonly used in the field of combinatorial chemistry, which involves the use of large numbers of beads, upon which a random collecton of chemical entities is assembled in a step-wise manner, thereby generating a large number of related, but different, chemical entities, each supported on a different bead. The beads are then typically screened in an assay to determine if the chemical entity supported thereon possesses a desired characteristic. Those beads supporting chemical entities with desired characteristics may be retained for further investigation, whilst those beads which are negative for the desired characteristic may be immediately discarded. In this way, starting from a large number of beads, each supporting a chemical entity, repeated rounds of screening can be used to isolate a small number of beads, each supporting a chemical entity of possible interest.

Conveniently, such screening is performed using a microtitre plate. Typically, one would perform an initial assay with perhaps 100 beads per well in the plate. Beads from those wells which returned positive results may be re-tested, with fewer numbers of beads per well, thereby reducing the number of irrelevant beads in each round of screening. Such procedures are particularly useful in the pharmaceutical industry for the discovery of candidate drugs. Similar beads are used in biomedical research, where the chemical entities may be polypeptides or nucleic acid sequences.

Conventional Coulter particle counters are inappropriate for use in connection with counting and/or sizing of such beads, as the beads are lost after counting/sizing. In contrast the present invention relates to a method of and apparatus for sizing and/or counting particles which can, in preferred embodiments, be used to count/size beads of the type described above.

The Invention

According to one aspect of the present invention, there is provided a method of counting and/or sizing particles held in suspension in a conductive liquid, the method comprising the steps of: (a) causing the conductive liquid with particles suspended therein to flow through a tube of substantially constant bore size, along at least a section of which an electrical current is caused to flow through the liquid; and (b) detecting variations in electrical signals by a particle counting/sizing means, which variations are caused by the passage of a particle though at least a section of the tube within which the electrical current is flowing; wherein the particle counting/sizing means does not interrupt the substantial constancy of the bore of the tube.

According to another aspect of the invention, there is provided apparatus for counting and/ or sizing particles held in suspension in a conductive liquid, the apparatus comprising: a length of tube of substantially constant bore size; pressure application means by which the conductive liquid can be caused to flow through the constant bore tube; and particle counting/sizing means which detect variations in electrical signals, which variations are caused by the passage of a particle through the tube past the particle counting/sizing means, wherein the counting/sizing means does not interrupt the substantial constancy of the bore of the tube.

Conveniently the apparatus is such that the particle counting/sizing means comprises first electrodes which generate an electrical current in the conductive liquid within the tube, and second electrodes which detect variations in a detectable electrical property, which variations are caused by the passage of a particle through at least that part of the tube within which the current is flowing. The detectable electrical property in which variations are measured or detected is conveniently voltage, but may also be current or resistance in other embodiments.

In preferred apparatus, the particle counting/sizing means comprises first relatively widely spaced electrodes which generate an electrical current along a section of the liquid in the tube, and second relatively closely spaced electrodes which detect and/or measure the electrical pulses produced as the liquid passes along the relatively short segment of tube between the electrodes, within the section along which current is flowing.

Preferably, transportation of the particles is via tubing sized to be comparable with, but larger than, the largest particle and arranged so that there are no connections or bore size changes which might act to slow or trap a particle which could lead to total blockage of the tube.

Typically the particles are carried through the tubing by a carrier fluid whose direction and flow rate is conveniently controlled by the operation of pneumatic controllers to alter the pressure of air (or other gas) in a source container with respect to an output container which is preferably held at atmospheric pressure. In order to avoid delivering fluid when no particles are present, a bypass flow route to a second sealed container containing liquid may be provided, arranged to join the primary delivery pipe near its output end. By appropriate control of the pressure in this second container, liquid can be drawn into or delivered from this container into the primary delivery tube.

The provision of this additional flow path can also enable flow in the primary tube to be reversed to clear any blockages. To prevent any particles from being erroneously sent into the second container the connection of the bypass tube into the delivery tube is preferably made via a filter element which allows liquid to pass but prevents the particles from entering the bypass tube.

As stated, the counting and sizing of the particles is preferably performed by an electrical method in which electrical current is made to flow along a short section of the fluid in the delivery tube through which the particles pass. Within this section of current flow two sensing electrodes, preferably separated by a distance approximately equal to the particle size, are positioned. As a particle passes between these sensing electrodes it partially blocks the current increasing the voltage between the sensing electrodes. It can be shown that the amplitude of the voltage pulse so generated is approximately a linear function of the volume of the particle.

This arrangement provides many performance benefits over conventional electrical particle size measurement techniques.

There are many applications which could utilise this method or apparatus. One application is the sorting and sizing of the beads commonly used for surface bound chemical synthesis or analysis. Typically these beads are in the size range 0.01 to 1.0 mm diameter and sample quantities of beads typically ranging from one hundred to one million in number need to be sized and/or distributed in known numbers into arrays of wells automatically and without losing any beads.

Counter/sizer apparatus in accordance with the invention could also be used for purposes other than counting/sizing beads used in combinatorial processes, such as:-

counting/sizing cells (unicellular micro-organisms, or eukaryotic cells sch as leukocytes or erythrocytes); measurement of the volume of a solid material being carried in the form of a supension; measurement of solids content in water (e.g. for measuring/detecting contamination of water supplies); counting of small objects being carried suspended in a fluid prior to delivery into a package (e.g. packaging of foodstuffs); and generally in the detection of solid contaminants in a fluid where optical or other detection means may bedifficult (e.g. glass contaminants in aqueous liquids) or inappropriate.

One example of the invention is now described with reference to Figure 1 of the accompanying drawings, from which some preferred features of the invention will become apparent.

Flow of beads (particles) through a delivery tube (3) is controlled by air pressure controllers (1) and (2) which control the pressure in the air space in supply and bypass reservoirs ((7) and (8) respectively). Preferably, the pressure can be controlled to any value either side of atmospheric pressure typically within the range lKPa to 300KPa absolute. To enable accurate control of the beads within the delivery tube (3) the rate of change of pressure in the reservoirs (7) and (8) is most preferably fast enough to enable the flow to be stopped in the time a bead takes to pass from a counter/sizer (5) to a T- switch (6). However, the rate of change of pressure is desirably not such as to cause break-up of the liquid in the tube (3) or outgassing leading to air bubbles. The T switch (6) is designed to permit the passage of liquid, but not te passage of particles, into the bypass tubing (4). Conveniently therefore, the T switch will comprise a filter element (the pores of which are less than the average diameter of the particles) or will comprise a junction, the radial gap of which is less than the average diameter of the particles (e.g. about half the average particle diameter) .

Preferably, the supply reservoir (7) has a shape and volume chosen to enable the concentration of beads around the inlet to the delivery tube (3) to be accurately controlled. A conical form, as shown, is one implementation which produces good performance when the beads have a negative buoyancy and sink to the bottom. Controlled agitation of the liquid can be used to disperse the beads through the liquid in a specified distribution. Typically, to enable beads to pass into the tube without jamming at the inlet, the bead concentration at the entry should be less than 0.1 % by volume. The conical form of the supply reservoir (7) can ensure that where only small numbers of beads are present gentle stirring disperses the beads through only a small volume of liquid, whereas when many beads are present vigorous stirring disperses the beads throughout a much greater volume of liquid. In addition, the conical form ensures that beads always settle close to the entry of the delivery tube and smooth internal wall profiles minimise the probability of beads becoming caught in the reservoir.

The supply reservoir is preferably providedwith mixing means so as to disperse the particles evenly throughout the carrier liquid. Such mixing may be by means of stirring or swirling the liquid (e.g. by use of a rotor within the reservoir or by a vortex-inducing device similar to the type known for test tubes and Eppendorf™ tubes) or by introducing jets in the liquid (e.g. introducing pulses of fluid from a nozzle immersed in the carrier liquid). Jetting is generally to be preferred asit is found to meet preferred criteria: particle velocity within the supply reservoir shoule preferably be low; particles should not be forced up the sides of the supply reservoir, where they may be stranded; and the particle concentration around the entrance into the liquid supply should preferably be controllable by the mixing meas.

It is preferred that the supply reservoir is providedwith agitation means to prevent agglomeration or clumping of the particles. Such means may comprise the addition of a surfactant but, in some instances, this may be undesirable if it influences the subsequent analysis of the particles. A suitable alternative isultrasonic agitation. Typically such ultrasonic agitation is required for quite prolonged periods (around 30 minutes) before the apparatus is ready for use. If desired, the ultrasonic agitation may be continued (continuously or intermittetly) whilst the apparatus is in use. The use of ultrasonic agitation is found to have an unexpected advantage, in that it partically degasses the carrier liquid, which reduces the possibility of gas bubble formation in the liquid: gas bubbles can cause erroneous results when certain types of particle detecting means are employed.

The bypass reservoir (8) preferably has a larger volume than the bead reservoir and can be used to replenish the bead reservoir with additional fluid if required. Bypass tubing (4) conveniently connects the bypass reservoir (8) to the T-switch (6). The T-switch (6) is desirably made so that the delivery tube (3) enters and leaves the three-way junction in such a way that the uniform smooth bore is maintained throughout.

The counter/sizer (5) measures the volume of beads as they pass down the tube. In addition to measuring the bead volume it most preferably also has a spatial resolution to discriminate between the signal generated by one large bead and two small beads stuck together. This can be readily achieved using the four electrode electrical conductivity measuring technique already referred to.

In a preferred method, the suspension of beads is caused to flow from the bead reservoir (7) through the delivery tube (3) into the receptor well (9) by raising the air pressure in the supply reservoir (7) using the air pressure controller (1). The flow rate can be adjusted so that liquid flow in the tube is substantially laminar and the beads, having a density close to the carrier fluid, are carried along by the liquid. As the beads pass through the counter/sizer the voltage pulse generated as they pass between the sensing electrodes is analysed for amplitude, width and shape. From this analysis can be determined the volume of the bead and whether it is the correct shape to be a single bead or it is more than one bead passing between the electrodes together.

The beads progress to the end of the tube and preferably drip into the receptor well (9) . When the correct number of beads have been dispensed (known by counting the voltage pulses), the pressure in the bead reservoir (7) is reduced to 1 bar (abs), which stops the flow. The pressure in the bypass reservoir (8) during bead dispensing is preferably held just above atmospheric so that flow into the bypass reservoir (8) is zero. After the flow has been stopped, the pressure in the bypass reservoir (8) can be raised and the pressure in the supply reservoir (7) reduced to cause liquid to flow from the bypass reservoir (8) back into the supply reservoir (7). By setting the correct pressures flow from or into the receptor well (9) will be zero. (It is preferred to cause liquid flow by means of pneumatic controllers to adjust pressures in the reservoir headspace, as this avoids the need to pass the liquid through pumps which could easily be blocked by the particles, or which could damage the particles. Standard pneumatic components are suitable, such as the type IR200 pressure controller obtainable from SMC (Welwyn, UK) in conjunction with the VO 307 type solenoid valve (also obtainable from SMC).) In this way any beads in the tube before the T-switch (6) will be carried back into the supply reservoir.

Preferably the T-switch (6) also incorporates an additional counting/sizing device of the same type as shown at (5) . The additional counter incorporated into the T-switch (6) may be used to confirm that the desirednumber of beads have been delivered to the receptor well (9) and not returned to the bead supply reservoir (7). An additional advantage of having a second counting/sizing device is that analysis of the time delay between signal detection at the first and second counting/ sizing devices during bead delivery provides information to a (conveniently computerised) control system regarding liquid volume flow rate and/or bead velocity.

If more than one receptor well is to be filled, as in a 96 well microwell plate, then the plate is moved to bring the new well under the primary delivery tube outlet and the sequence repeated.

The apparatus can, however, be operated in several different modes to meet other requirements. These include:

1. Bypass flow can be used when bead density is low, the liquid flow being maintained from the supply reservoir (7) to the bypass reservoir (8) by setting the appropriate pressures. Flow is diverted to the receptor well (9) only for the time after a bead has been sensed by the counter/sizer sufficient to ensure it has been dispensed. Operation in this mode minimises the liquid dispensed with each bead, which is important for dispensing into small volume wells.

2. Reverse flow can be used if two beads pass through the counter too close together for the first to be dispensed without the second, when only one bead is required. If this happens the flow is switched into reverse flow (from bypass reservoir (8) to supply reservoir (7)) to send all beads which have not reached the T-switch (6) back to the bead reservoir (7).

3. Blockage clearance in which reverse flow can be used at sufficient pressures to unblock the tubing (3) if beads have become stuck in the tube.

The apparatus shown in Figure 1 and described above is illustrated schematically in greater detail in Figure 9. Generally the apparatus of the invention will be automated, and operated by computerised control means, such as a personal computer.

Description of Embodiments

The invention is further described with reference to the accompanying drawings, in which:

Figure 1 shows the overall system diagramrnatically to enable description of preferred aspects of the invention;

Figure 2 shows a preferred construction of the counter/sizer;

Figures 3(a) and 3(b) show a preferred method of construction of the counter/sizer;

Figures 4, 5 and 6 show some possible constructions of the T-switch;

Figure 7 shows the construction of one embodient of an integrated counter/T-switch; Figure 8 is a schematic illustration of a sensing amplifier suitable for use with the counter; and

Figure 9 is a schematic representation of the system shown diagrammatically in Figure 1.

Figure 1 showing the overall system has already been described.

The general function of the counter/sizer and its general mode of operation has been described previously. Figure 2 shows the basic arrangement of one possible embodiment of the particle counter/sizer. The advantages of this arrangement are best described by considering how to implement the conventional method in an in-line configuration. The major difference between this approach and conventional approaches is the use of four annular electrodes in the walls of a constant bore insulating tubing, as compared to the use of two large electrodes situated in large volume vessels either side of a small linking orifice. In the conventional approach the current density away from the sensing orifice is so low that beads only influence the current as they pass through the high current density region of the sensing orifice. However, in an in-line implementation the current density would be substantially uniform at all points between the electrodes. Thus, if only two annular electrodes were to be used, then the signal generated would last for all the time a bead is between the electrodes. If the electrodes are several tube diameters apart the signal amplitude would be small and the duration long, giving poor signal to noise ratio and poor discrimination between closely spaced beads. If, however, the two electrodes were placed very close together, then current would be non-uniform across the tube and electrode-liquid interface interactions occur both of which produce unreliable performance. Such an arrangement would therefore not be practical.

One possible method to improve performance would be to reduce the tube bore between the two electrodes. This would improve the electrical performance by concentrating the current density midway between the electrodes. However it would also increase the probability of beads jamming in the orifice or becoming caught in circulating currents. This arrangement would therefore also be impractical. The present invention, in a preferred aspect, overcomes all these problems, as shown in Figure 2, by using two widely spaced annular electrodes (11) to impose a uniform current flow axially down a short section of the tube through which the beads flow. Midway between these electrodes (11) are located two narrow sensing electrodes (12) separated by an insulator of less than the tube diameter thickness. These electrodes (12) sense the voltage in the liquid at their relative positions. A differential amplifier (13) connected between these electrodes (12) produces an output voltage that increases markedly when a bead is between these electrodes (12). It can be shown that the height of the voltage pulse generated is approximately linearly proportional to the volume of the bead over a wide size range and that the signal has good signal to noise and excellent discrimination between beads close together.

In a typical implementation for measuring beads around 100 microns in diameter in a 300 micron bore tube, the current electrodes would be 4 mm apart and the supply voltage in the range 1 to 10 volts. The sensing electrodes would be centrally positioned separated at 100 μm spacer. Thus, without beads present the voltage generated between the sensing electrodes by the current flow through the liquid would be 0.25V. As a 100 micron bead passes through the voltage would momentarily increase. The current through the fluid would depend upon the conductivity of the fluid, but in practice v/ould be in the range 0.0 lμA to 10μA. Low current is necessary to eliminate the generation of bubbles by electrolytic action. An additional advantage of this design is that the magnitude of the signal is independent of the position of the bead in the tube as the separation of the current electrodes is sufficient to ensure axial current flow at the sensing electrodes, whereas in a conventional approach current flow is highly curved into and out of the orifice in the partition wall between the two containers. In order to minimise the effects of electrode to liquid interface effects it is advantageous to ensure there is always some current flow from the liquid into the sensing electrodes rather than to allow them to float using a high impedance amplifier. A preferred arrangement is to use bias resistors to provide a path such that approximately 1 % of the current flows into each sensing electrode.

By selection of the appropriate voltage and current levels it is possible to minimise the effect of the electrical properties of the beads in the same way as is used in conventional techniques.

It is preferable to use platinum or gold electrodes to minimise electrochemical reactions with the liquid manufacture and under some conditions it may be necessary to continuously switch the polarity of the current electrodes to avoid polarisation effects at the electrodes.

Whilst many methods of manufacture can be used to implement the design, one method has been demonstrated to be particularly suitable for applications where the tube bore size is less than 0.5 mm. At these dimensions it is difficult both to manufacture the components with required tolerances and to assemble them with the necessary degree of alignment of the components. Figure 3a shows a preferred method of construction in which all the electrodes and spacers are encapsulated in a casting resin to hold them precisely positioned.

In this preferred method of manufacture the tubing used for the delivery tube (21) is commercially available PTFE or similar polymer tubing, typically with bore diameters in the range 0.15 to 0.5 mm and an outer diameter of 1.6 mm. The electrodes (22) for introducing the current into the liquid are manufactured from platinum foil cut into rectangles with a side ratio of 3 to 2 with the smaller side just larger than the outside diameter of the tube. A hole is formed in the platinum foil with a diameter precisely matched to the bore diameter of the tube. There are several methods available to achieve this, including mechanical drilling, laser drilling, punching and electrodischarge machining. Where mechanical drilling is used it is possible and preferred to laminate many electrode foils together sandwiched between end plates using a rigid wax. This ensures a sharp edged uniform profile hole through the thin foil with a well controlled diameter.

To ensure that the diameters are matched, high magnification comparative measurements are necessary using either optical or Scanning Electron Microscopy.

The thickness of the current supply electrodes needs to be sufficient to ensure that the effects of high current densities at the liquid-electrode interface do not cause adverse heating or gas generation. Typical electrode thicknesses can range between 0.3 x to 3. Ox the mbe bore diameter.

The sensing electrodes can be much thinner as they carry less current and to ensure good spatial resolution it is preferable that they are thin compared to the bead diameter. Typically this is in the range 0.03x to 0.3x the tube diameter.

The sensing electrodes are separated by a thin insulated spacer, typically in the range 0. Ix to 0.5x the bore diameter. For a 0.3 mm tube a separation of 0.1 mm is preferred. It is preferable to manufacture the two sensing electrodes bonded directly to the spacer and then to form the hole through the electrode-spacer-electrode laminate (23). The three layer laminate may be formed by deposition of platinum by evaporation or sputtering on to the spacer or by laminating platinum foil to the spacer using a suitable adhesive. One convenient method is to produce a 5 layer spacer-electrode-spacer-electrode-spacer laminate using 0.01 mm platinum foil for the electrodes and 0.1 mm Mylar (Trade Mark) sheet for the spacer bonded together with a resin adhesive. The outer layer protects the electrodes during assembly and drilling. This assembly is fabricated as a strip of, for example, 10 units long and several strips are then laminated together sandwiched between end-plates using a rigid wax for drilling of the holes to match the tube bore. After drilling the strips are separated by melting the wax and the individual sensing electrode assembly can be cut from the strip.

The reduced height of the outer spacer layer exposes the sensing electrode to enable electrical connection to be made.

To accompany the current and sensing electrodes two short sections of the tube (24) in the range 3x to 30x the bore diameter are cut to act as spacers between the sensing electrodes and the current electrodes. It is important that the end faces of the tube are flat and square as they are required to seal against the electrodes.

These five components now have to be assembled between two longer lengths of the tube so that all the holes are aligned to ideally an accuracy of +/- 1 % of the tube bore and then held so that they form a rigid uniform bore assembly with the electrodes appropriately spaced. The preferred method to achieve this as shown in Figure 3b, is to use a guide wire (25) which has, over a short section of its length, a diameter accurately matched to the mbe bore. Either side of this region the wire diameter is reduced to enable it to pass easily through the mbe. All components are threaded on this guide wire, in the appropriate order, and it is then tensioned on a support frame (26) to hold it straight. The counter/sizer components are then slid over the thicker section of the guide wire to align them and spring pressure applied to the tubing to compress all components together to form a continuous mbe. As illustrated in Figure 3a, electrical contact wires are then attached to the platinum electrodes and connector pins (27) and the components potted together into a solid block (28) to form a robust and rigid assembly. Cold casting resins have been found to provide the high degree of electrical resistivity and mechanical stability required. After the resin has set the guide wire is carefully removed from the mbe.

Typically the device will be used with an electronic sensing amplifier mounted adjacent to it and the whole unit may be electrically shielded if it is to be used in an electronically noisy environment.

The other important component required by the preferred system is the T-switch. As described previously, this consists of a connector which allows liquid to enter and leave the delivery mbe from the bypass mbe without letting beads into the bypass mbe. It does this in such a way that the smooth bore of the delivery mbe is not disturbed.

There are various ways to implement this component, one of which is shown in Figure 4. This uses a cylinder of a porous material with an axial hole through it of the same bore as the mbe. Providing the pore size of the porous material is small compared to the beads this maintains the smooth inner bore requirement.

A second implementation is shown in Figure 5. In this a small radial gap is made in the wall of the delivery mbe. Providing this gap is small compared to the diameter of the bead then the smooth bore requirement is maintained. A method of carrying out this implementation has been developed which uses the assembly approach described for the counter/sizer.

This method is shown in Figure 6e. It comprises five laminations of thin sheet material. The two outer layers (31, 32, shown in Figures 6a and 6c respectively) and the central layer are provided with holes to match the bore sizes of the delivery and bypass tubes. The three intermediate layers (34, shown in Figure 6d, two of which are present in the appartus, and 33, shown in Figure 6b) have similar holes but with a passage linking between them. Laminations 31, 32 and 33 are formed from a material with a thickness sufficient to provide mechanical rigidity but are sufficiently thin to facilitate manufacture. A thickness in the range 50-500μm is generally suitable. Lamination 34 is typically formed with a thickness between one quarter and one half that ofthe average diameter of the beads/particles being counted/sized. The five layers are assembled together by aligning a dowel pin through the delivery mbe holes. The laminations are held tightly together by pressure and the edges bonded together and sealed using resin (35).

The assembly so formed thus provides channels the same thickness as the lamination linking the delivery mbe to the bypass mbe. The thickness of the lamination is chosen to be less than 50% of the bead diameter to prevent beads passing along it.

The lamination assembly (36) is assembled on the guide wire described previously between sections of the delivery mbe (37) with the bypass mbe (38) held against the hole at the other end of the laminations and the assembly is potted into a rigid block (39) using cold casting resin.

The T-switch may be made with one or two bypass tubes. The additional bypass mbe can be used to assist initial air removal on filling or for the inclusion of a pressure sensor at this point. If neither of these are required it can be left out or blanked off.

If required, the T-switch and counter/sizer modules can easily be constructed together, to form a single integrated component, one embodiment of which is shown in Figure 7. In this embodiment the counter electrode arrangement as described previously is fabricated by casting platinum sheets of the appropriate thickness (and with wires attached) in resin blocks of a thickness required to give the necessary spacing (40). Holes of the appropriate size are then driled through the resin and the electrodes to match the bore of the tube. The resin blocks are then assembled on a guide wire along with the T-switch lamination assembly (36). Plastic tubing of the bore required for the delivery mbe is then assembled on the guide wire along with fittings which grip the wire, to assist the retention of the mbe when the whole assembly is cast in resin (43). In the embodiment illustrated, brass ferrules (42) are used and additional fluid sealing is provided by O-rings (41) between the ferrules (42) and the counter/T-switch (36).

The construction of both a counter/sizer and a combined counter/T-switch should ensure that there is no electrical connection between the supply electrodes and the sensing electrodes, other than through the conductive fluid in the mbe. Where the system is using only weakly conducting fluid to carry the beads, the sensing current is measured in nanoamps and the insulation between the electrodes must provide an impedance in excess of 1MΩ.

In addition, when such small currents are being measured, the amplifier should preferably be co-located with the counter. In one embodiment a sensing amplifier, of the type illustrated schematically in Figure 8, is mounted on a printed circuit board on which the counter is also mounted, and the two are electrically connected via connector pins on the board. One important aspect of the circuit are the bias resistors R2 and R3 of 100MΩ value. These resistors maintain a small current flow through the sensing electrodes at all times, which is important in establishing the correct current flow in the liquid in the tube.

While the preferred method of practising this invention is a four electrode arrangement as above-described in detail, it would be practical, especially in the case of opaque beads (particles), to employ optical sensing at the short segment of uniform bore mbe through which the liquid is flowing.

Thus, the principal challenge to be faced in sizing and/or counting particles in suspension is the technique required to be able to extract the beads from the suspension into a delivery mbe so that they are entrained in a liquid flow as individual beads well spaced apart and to be able to deliver the beads in a minimal amount of liquid, or to divert or return the beads if they are not to be delivered, out of the end of the tube. The arrangement of pumping, valving connectors and sensors above-described gives excellent performance for the flexible control and manipulation of individual sub millimetre beads held as a suspension in a fluid.

The next requirement is to provide a method of counting and sizing the beads as they move along the delivery mbe.

A well known problem with implementing such a sensor is that any discontinuity or change in the internal dimensions of the mbe can act to trap or slow down a bead. If this occurs beads may remain trapped in the tubing and if several beads are trapped a blockage will quickly occur. The essential requirement for this invention is a counter/sizer which can be implemented in the delivery mbe without significant change to the internal form. An alternative possibility to electrical measurement is therefore an optical technique to look through the walls of a transparent mbe to measure the size of the bead, although it will be appreciated that accurate sizing of such particles through the highly refracting cylindrical section of the tubing is more expensive and more complex to implement than the preferred system described in detail above.

Claims

1. Apparatus for counting and/or sizing particles held in suspension in a conductive liquid, the apparatus comprising: a length of tube of substantially constant bore size; pressure application means by which the conductive liquid can be cased to pass through the substantially constant bore tube; and particle counting/sizing means which detects variations in electrical signals, which variations are caused by the passage of a particle through the tube past the counting/sizing means; wherein the counting/sizing means does not interrupt the substantial constancy of the bore of the tube.

2. Apparatus according to claim 1, wherein the particle counting/sizing means comprises first electrodes which generate an electrical current in the conductive liquid within the tube, and second electrodes which detect variations in a detectable electrical property, which variations are caused by the passage of a particle through at least part of the tube within which the current is flowing.

3. Apparatus according to claim 2, wherein the second electrodes comprise a pair of sensing electrodes disposed between the first electrodes, which sensing electrodes detect pulses of increased voltage caused by the passage of a particle.

4. Apparatus according to claim 2 or 3, wherein the first and/or second electrodes comprise a pair of annular electrodes located around the tube.

5. Apparatus according to any one of claims 2, 3 or 4, wherein the first electrodes are relatively widely spaced, and the second electrodes are relatively closely spaced.

6. Apparatus according to any one of claims 2-5, wherein the current flow within the conductive liquid is substantially axial along the tube at the second electrodes.

7. Apparatus according to any one of claims 2-6, wherein the current flowing through the conductive liquid within the section of tube between the first electrodes is in the range

8. Apparams according to any one of the preceding claims, for counting and/or sizing particles with a diameter in the range 0.01 to 1.0mm.

9. Apparams according to any one of the preceding claims, wherein conductive liquid with particles suspended therein may be supplied to the mbe from a supply reservoir, and wherein the mbe is connected via bypass ducting to a bypass reservoir, the pressure in the headspace of the supply and bypass reservoirs being regulatable by means of pressure application means, such that liquid can flow from the mbe to the bypass reservoir or from the bypass reservoir to the mbe.

10. Apparams according to claim 9, wherein the bypass ducting is connected to the mbe by means of a T switch, in which the constant bore of the tube is not interrupted.

11. Apparams according to claim 10, wherein the T switch comprises a filter element which allows the passage therethrough of the conductive liquid but does not permit passage of the particles suspended in the liquid.

12. Apparams according to any one of the preceding claims comprising two or more particle counting/sizing means, each particle counting/sizing means being disposed to count/size particles at a different location within the mbe.

13. Apparams according to any one of the preceding claims, comprising particle counting/sizing means and a T-switch combined in a single integrated component.

14. A method of counting and/ or sizing particles held in suspension in a conductive liquid, the method comprising the steps of:

(a) causing the conductive liquid with particles suspended therein to flow through a mbe of substantially constant bore size, along at least a section of which an electrical current is caused to flow through the liquid; and (b) detecting variations in electrical signals by a particle counting/sizing means, which variations are caused by the passage of a particle through at least a section of the mbe within which the electrical current is flowing; wherein the particle counting/sizing means does not interrupt the substantial constancy of the bore of the mbe.

15. A method according to claim 14, comprising the use of apparams in accordance with any one of claims 1-13.

16. Apparams substantially as hereinbefore described and with reference to the accompanying drawings.

17. A method substantially as hereinbefore described and with reference to the accompanying drawings.