Freeze Drying Of Target Substances

The present invention relates to freeze drying of a target substance, and particularly, but not exclusively, to a technique and apparatus suitable for freeze drying a reagent such as an electro-active substance in situ in an electrochemical cell of a sample analyser device, such as a biosensor device.

It is desirable to freeze dry target substance in various applications, and typically it is of benefit to minimise cycle times for the process without reducing the effectiveness of the freeze drying process. An improved technique and apparatus have been devised.

According to a first aspect, the present invention provides a process for freeze drying a target substance, the process comprising: providing a target substance carried by a carrier; passing the carrier and target substance into a freeze chamber to reduce the temperature of the target substance; passing the carrier and target substance to a dryer chamber; and removing the carrier and target substance from the dryer chamber, wherein vacuum/reduced pressure is applied during the process: during a final portion of the freeze process when the target substance and carrier are present in the freeze chamber; and/or when the target substance and carrier are transferred from the freeze chamber to the dryer chamber.

In a preferred realisation, the invention comprises: applying vacuum/reduced pressure in the freeze chamber during a final portion of the freeze process; passing the carrier and target substance to the dryer chamber;

applying vacuum/reduced pressure when the target substance and carrier are transferred from the freeze chamber to the dryer chamber; and removing the carrier and target substance from the dryer chamber.

According to a second aspect, the invention provides a method of manufacturing a biosensor device comprising depositing an electro-active reagent substance in solution on a carrier and operating a freeze drying process according to the present invention as described above to freeze dry the electro-active reagent substance.

The invention is particularly suited to small volume applications, preferably in which the target substance is deposited on or in a carrier in liquid form in volumes of nanolitres up to 1000 nanolitres.

The design is driven by the need for a repeatable process when the target substance has a small volume and evaporation losses (typical at ambient and even at lower temperatures) would represent a large proportion of the target substance and therefore affect the final characteristics. In, for example a batch process, where a first carrier in a batch to enter one of the chambers waits a considerable time before the freeze drying process starts compared to a following carrier, this could provide an unacceptable production variation. Additionally, in this method the carrier and target substance are subject to the full temperature cycle without the surrounding chamber and apparatus being subject to the full cycle. This provides substantial savings in time and energy costs.

According to a further aspect, the present invention provides an apparatus for freeze drying a liquid target substance, the apparatus comprising: a freeze chamber in which to reduce the temperature of the target substance; a dryer chamber adjacent the freeze chamber and to which the target substance is passed following exit from the freeze chamber, and in which drying of the target substance is promoted; and a vacuum system operable to apply vacuum/reduced pressure during the process: when the target substance and carrier are present in the freeze chamber; and/or

when the target substance and carrier are transferred from the freeze chamber to the dryer chamber.

It is preferred that vacuum/reduced pressure is also applied during the process when the target substance and carrier are present in the dryer chamber. In the context of this patent 'target substance' refers to any substance which is being freeze dried.

Beneficially, vacuum/reduced pressure is applied during a latter stage or final stage of the freeze process in the freeze chamber.

In one embodiment the vacuum in the freeze chamber may be drawn via the dryer chamber.

In certain embodiments more than one freeze chamber may be provided, generally in series. Additionally or alternatively more than one dryer chamber may be used.

Other preferred features of the invention are presented in the dependent claims and described in relation to the specific embodiments.

The invention will now be further described in specific embodiments, by way of example only, and with reference to the accompanying drawings, in which:

Figure 1 is a schematic side view of an electrochemical cell;

Figure 2 is a schematic plan view of a sensor strip comprising four electrochemical cells.

Figure 3 is a schematic side view of an exemplary apparatus in accordance with the invention for operating the process according to the invention;

Figure 4 is a schematic side view of the freeze chamber of the apparatus of Figure 3;

Figure 5 is a graphical representation of the relationship between temperature and pressure for the system of the invention and a different, previously Mailed system;

Figure 6 is a graphical representation of the effect on the freezing rate of setting the freeze chamber cold plates at different temperatures;

Figure 7 is a representation showing the effect of pressure and inert gas bleed in the dryer chamber;

Figure 8 is a graphical plot of water release (sublimation rate) for the technique of the present invention and a previously trialled technique;

Figure 9 is a graphical plot showing the difference in warming rates between the technique of the present invention and a previously trialled technique;

Figure 10 is a graphical plot showing the profiles for different base sheet card thicknesses for different warm plate temperatures and nitrogen conditions during the transfer from the freeze chamber to the dryer chamber;

Figures 11 to 13 are images of freeze dried deposits (cakes) produced using previously trialled techniques ('old') and the technique of the present invention ('new');

Figure 14 is a graphical plot comparing the thickness of freeze dried deposits ('cakes') for the previously trialled techniques ('old') and the technique of the present invention ('new');

Figure 15 is a graphical plot comparing the brightness and averaged greyscale standard deviation of freeze dried deposits ('cakes') for the previously trialled techniques ('old') and the technique of the present invention ('new');

Figures 16 to 18 are images of freeze dried deposits ('cakes') 5% KCI solution freeze dried using different cold plate temperatures; and

Figure 19 is a plot of calibration current vs TC concentration for a biosensor produced in accordance with the invention in a -35 ⁰ C freeze drying run.

Referring now to Figure 1 of the accompanying drawings, an electrochemical cell 1, illustrated in a cross-sectional side view, comprises a base layer 2 formed from a non-conducting material. The base layer 2 preferably has a thickness of 50 - 2000 μm, preferably around 125 μm.

A non-conducting supporting layer 3 is formed on the base layer 2. The supporting layer 3 is preferably formed from PET and has a thickness in the range of 50 μm to 500 μm, preferably 250 μm, 150μm, or 50μm.

The supporting layer 3 forms a support on which a working electrode 4 is formed. The working electrode 4 is preferably in the form of a continuous band around the wall(s) of the cell 1. The thickness of the working electrode 4, which is its dimension in a vertical direction when the cell 1 is placed on the base 2, is typically from 0.01 to 50 micrometers. Preferred and possible thicknesses of the working electrode are as described in our co-pending application WO 03/056319.

The working electrode 4 is preferably formed from carbon, for example in the form of a conducting ink. A preferred carbon based conducting ink comprises a suspension of carbon dispersed in a resin solution. The working material may be formed of other materials and inks as detailed in WO 03/056319. Furthermore, two or more layers of the same or different materials may be used to form the working electrode.

A dielectric layer 5 comprising an insulating material, typically a polymer, a plastic, or a ceramic, again as detailed in WO 03/056319, is formed on and insulates the working electrode 4 from a pseudo-reference electrode 6. Typically, the dielectric layer 5 has a thickness of 1 to 1000 μm. The dielectric layer could be formed of more than one layer.

The cell 1 is formed to have one or more wells 7.

Well diameters of 0.1mm to 5mm maybe utilised dependent upon the particular application. Where non-circular wells are used, the length or width dimension will typically be in the range 0.1mm to 5mm (more typically 0.9 to lmm). Typically the well depth will be in the range 50μm to lOOOμm, preferably 50 μm to 500 μm, more preferably about 150μm or 50μm.

The base layer 2 forms the bottom of the well 7 and may take the form of a porous membrane.

The open end of the cell maybe covered with a membrane 9 that is permeable to components of the sample to be tested, for example blood or plasma. The membrane may also be used to filter out components of the sample that should not enter the cell, for example red blood cells.

Referring now to Figure 2 of the drawings, there is illustrated in a schematic plan view, layers of a sensor strip 10 comprising four electrochemical cells of the type and made as described above.

The sensor strip 10 comprises an insulating substrate sheet 11. Formed on the insulating substrate sheet 11 is a patterned layer 12 of a material that forms four working electrodes 12a, 12b, 12c, and 12d, one for each of the respective four cells and four conductive tracks 12e, 12f, 12g, and 12h each of which is in electrical contact with a respective one of the four working electrodes 12a, 12b, 12c, and 12d.

It will be appreciated that for ease of viewing the various layers, the dielectric layer 13 and the pseudo-reference electrode layer 14 are each illustrated shifted laterally sideways from their true positions in the strip 10.

An electro-active substance 8 is contained within the well 7. The electro-active substance 8 is freeze dried in accordance with the invention to form a porous deposit. On introduction of a measurement sample (not shown) into the well 7, the electro-active substance 8 dissolves and an electrochemical reaction may occur and a measurable current, voltage, or charge may occur in the cell. Electro-active substances are discussed in more detail in, for example, our co- pending application WO 03/056319.

The sensor strips 10 are formed on a base sheet 30 that acts as a substrate for a large number of strips 10. The substrate base sheet 30 may comprise the PET base layer 3 of the respective cells when the strips 10 are eventually divided from the base sheet 30.

The electro-active substance 8 is introduced in to the wells 7 of the strips 10 supported on the base sheet 30 in liquid form (aqueous solution). The well is typically about lmm diameter and a measured dose (for example 0.4 microlitres) of liquid is introduced into each well 7. The liquid is then subjected to a freeze drying process in accordance with the present invention. The technique of the present invention is particularly suitable for freeze drying an array of wells containing micro-volumes of liquid, typically in the range 1 nanolitre to 1000 nanolitres, more typically in the range 200 nanolitres to 700 nanolitres, most typically in the range 200 nanolitres to 400 nanolitres.

The freeze drying apparatus of the present invention, as shown in Figure 3 , comprises a freeze chamber 31 and a dryer chamber 32. A warm discharge chamber 33 may be provided downstream of the dryer chamber 32, dependent upon specific processing requirements. An in-feed conveyor 35 is positioned upstream of the freeze chamber 31 and conveyors are provided internally of chambers 31 and 32 (and also chamber 33, where present). An out-feed conveyor 47 is provided to collect the sheet 30 exiting the apparatus. Slit valves 38a and 38d are provided on entry and exit from chambers 31 and 33. A sealed shroud passage 37 connects the freeze chamber 31 with the dryer chamber 32. The sealed shroud passage 37 is provided with a slit valve 38b. A sealed shroud passage 57 connects the dryer chamber 32 with the discharge chamber 33. The sealed shroud passage 57 is provided with a slit valve 38c.

A base sheet 30 provided with the printed layer structures forming a number of electrode strips 10 in a matrix array is fed from a well filling station (not shown) immediately upstream of the in- feed conveyor 35. As a result, when in position on the in- feed conveyor 35 the wells contain the measured dose of the electro-active substance 8 in liquid form, ready to be freeze dried.

Heat transfer into and out of the carrier layer can be modified in a number of ways, for example, holding the base sheet 30 on supports or adding a layer to form a barrier between the base sheet 30 and the cooling plate.

Before passing into the freeze chamber 31 and dryer chamber 32, the base sheet 30 may have a thermal moderator means applied thereto in order to alter the processing characteristics in the chambers. Attaching the thermal moderator means to the base sheet has been shown to enable

the processing characteristics to be modified in ways that may be beneficial for freeze drying of certain reagents and substances. The thermal moderator means may be achieved by means of use of a backing and/or facing for the base sheet 30. Metallic sheets or foam insulation (such as PE foam sheets) have been found to give good results. The provision of such a thermal moderator means for the base sheet 30 (or even the sensor strips/devices per se) has several benefits in terms of processing as will be described. For example:

1. the thermal moderator means can be applied to the base sheet or the devices prior to entering the freeze drying system or to the cooling plates themselves; and 2. the thermal moderator means separates the sheet containing the target substance from the cooling plates.

If attached to the base sheet 30 the thermal moderator means may be able to be removed following processing.

The thermal moderator means is described as such because it tends to control the effect that environmental exposure conditions, especially temperature, have on the base sheet or devices . The thermal moderator means may therefore be heat conductive (such as a metallic heatsink) or non-heat conductive such as PE foam. Accordingly, the thermal moderator means maybe a shielding, insulating, or heatsink means. The nature and purpose of the thermal moderator means is further described later in this document.

A vacuum system 70 is provided in order to enable evacuation of the chambers. The vacuum system 70 may typically include an oil free pump and booster arrangement 71 and a vacuum line extending via respective branches and regulator control valves 73 to the respective chambers. As shown in Figure 3 in an exemplary system, the vacuum line extends to the dryer chamber 32, enabling the dryer chamber 32 to be evacuated during the process. It is also required to evacuate the freeze chamber 31 during the process and this may be achieved by drawing the vacuum via the dryer chamber 32 when the slit valve 38b is opened. In this case the freeze chamber 31 is evacuated via the shroud 37 and dryer chamber 32. In an alternative embodiment a separate branch of the vacuum line (shown in dashed line in Figure 3) can be provided to extend directly to the freeze chamber 31 , providing an evacuation route by-passing the dryer chamber 32. The arrangement also includes a dry gas purge/bleed system 80

(typically using nitrogen as the purge or bleed gas) including a gas delivery line extending via respective branches and regulator control valves 83 to the chambers.

In order to feed the base sheet 30 from the in-feed conveyor 35 into the freeze chamber 31 , the slit valve 38a moves from a closed position to an open position (as shown in Figure 4) in which the base sheet 30 can be fed through the horizontally aligned slit passage 40 of the slit valve 38a into the interior of the freeze chamber 31. Passing into the freeze chamber 31 the base sheet 30 is received on a chamber internal conveyor arrangement comprising separate peripheral conveyor bands 43 wrapped around respective pulleys. The respective conveyor bands 43 are provided to underlay respective opposed longitudinally running marginal portions of the base sheet 30. The conveyor feeds the base sheet 30 until it is contained wholly within the freeze chamber 31 and the conveyor is deactivated when a reference mark on the base sheet 30 breaks a light beam limit switch.

The freeze chamber 31 contains an upper refrigerated plate 47 and a lower refrigerated plate 44. A refrigeration unit is situated outside the freeze chamber 31 and supplies a heat transfer fluid (typically silicone oil) to cool the plates 44, 47 via conduit connections into the freeze chamber 31 through the outer walls of the freeze chamber 31. The plates 44, 47 are cooled to a temperature below the collapse temperature of the substance, preferably to at least -25 ⁰ C, more preferably to at least -35°C. In one operational embodiment, the plates 44, 47 are cooled to -58°C. The lower plate 44 is normally stationed below the level of the conveyor bands 43 (as shown by a bold line in Figure 4). When the respective base sheet 30 is positioned centrally with respect to the plates 44, 47, the lower plate 44 is raised up (as shown by a dashed line in Figure 4), for example by means of a pneumatic ram and cylinder arrangement 42, lifting the base sheet 30 from the conveyor and carrying it into close proximity with the upper refrigerated plate 47 (typically to within 3mm of the upper plate). By positioning the two refrigerated plates 44, 47 in close proximity in this way, controlled freezing is achieved. One or both of the refrigerated plates 44, 47 may be provided with ridge, projection, or other proud standing formations (for example formations 49) to contact the base sheet 30 at zones not printed with the layer structure electrodes (i.e., at neutral zones) in order to inhibit bowing of the base sheet 30 during the freezing stage.

At a predetermined point in the freezing procedure, the freeze chamber 31 is evacuated. In one embodiment, this is achieved by means of opening the slit valve 38b between the freeze chamber 31 and the dryer chamber 32, and prior to this, simultaneously, or subsequently, opening the regulator control valve 73 in the vacuum line. This applies the vacuum to the freeze chamber 31 , via the dryer chamber 32. In operating in this way the evacuation time is the time required to effect evacuation to the required pressure the volume of both of the chambers 31 and 32. As an alternative the freeze chamber 31 could be evacuated by its own, dedicated vacuum line (shown by a dashed line in Figure 3), separate from the vacuum line for evacuating the dryer chamber 32.

Following holding at the raised position sandwiched between the two refrigeration plates 44, 47 for a predetermined period, the base sheet 30 is lowered on plate 44, and replaced on the conveyor bands 43. When the desired vacuum pressure is reached (typically lO^mbar) the base sheet 30 is transferred between the freeze chamber 31 and the dryer chamber 32, by means of operation of the in-chamber conveyor 43 so that the base sheet 30 is fed through the horizontally aligned exit slit valve 38b into the dryer chamber 32 to be received on the conveyor in the dryer chamber 32.

The procedure is microprocessor controlled to ensure operation of the various process steps at the desired point in the procedure. For example, a process timer determines the freeze time, evacuation rate and time, the rate and timing of raising of the lower plate 44. Alternatively, these timings can be controlled by determination of the temperature or pressure in the chambers, etc. The process parameters are consequently controlled and variable to tailor the freezing conditions to meet certain requirements, hi particular it may in certain instances be beneficial to modify, alter, or tailor the freezing rate of the substance. Typically, a balance needs to be struck between rapid freezing (and hence low overall time spent in the freeze chamber 31) and slow freezing to ensure appropriate control of the sublimation process.

The temperature of the freeze chamber 31 is maintained substantially at the refrigeration temperature (which is below the collapse temperature, and preferably between -25 ⁰ C to -60 ⁰ C, more preferably between -35 ⁰ C to -6O ⁰ C) before, during, and after each pass through cycle for each respective base sheet 30. As one base sheet 30 exits the freeze chamber 31 , the exit slit valve 38b closes and seals the freeze chamber 31 ready for the next successive base sheet 30 to

enter via the entry slit valve 38a. The gas purge/bleed system 80 is subsequently operated to backfill the freeze chamber 31 to a small positive pressure. The base sheet 30 is held in the cold environment of the freeze chamber 31 typically for a period in the range 1 to 7 minutes with the final stages being at applied vacuum.

It is important that the freeze dried deposit is cooled in the freeze chamber 31 to below its collapse temperature. The collapse temperature is defined as the point at which the material softens to the point of not being able to support its own structure.

On exiting the freeze chamber 31 through the exit slit valve 38b, the base sheet 30 passes through a sealed shroud duct 37 and into the dryer chamber 32, via the inlet slit valve 38c, which is at that juncture positioned to receive the base sheet 30 passing through its horizontally orientated slit passage 40. The base sheet 30 is received on a conveyor arrangement positioned internally of the dryer chamber 32. The conveyor activates to position the base sheet 30 entirely within the chamber 32 and is then de-activated. The conveyor feeds the base sheet 30 until it is contained wholly within the chamber 32 and the conveyor is deactivated when a reference mark on the base sheet 30 breaks a light beam limit switch.

The slit valve 38b is then closed in order to seal the dryer chamber 32 from the freeze chamber 31. Total transfer time from being sealed in the freeze chamber 31 to being sealed in the dryer chamber 32 is kept to less than 30 seconds, more preferably less than 20 seconds or less. During the transfer procedure the vacuum is maintained in the freeze chamber 31 , the dryer chamber 32, and the connecting shroud 37.

In previously trialled arrangements a vacuum was not applied in the freeze chamber or during transfer to the dryer chamber. In the previously trialled process the base sheet was first frozen in the freeze chamber and then transferred to the dryer chamber where, once sealed in the dryer chamber, a vacuum was applied to dry the electro-active substance present on the base sheet. Consequently, in the previously trialled arrangements somewhat narrow constraints are placed on the temperature of the freeze chamber and dryer chamber and the temperature moderation characteristics of the carrier. These constraints are required in order to minimise the warming of the base sheet and electro-active target substance during the transfer between the two chambers. Typically, the target substance and carrier reached a temperature between -25 ⁰ C

and -2O ⁰ C, before sublimation began in the dryer chamber. Such temperature in the target substance during drying limits the use of the machine to substances with collapse temperatures warmer than this. Significant process benefits have been realised by moving to a technique as now described in which a vacuum is applied during a part of the freezing process and maintained during transfer between the freeze chamber 31 and the dryer chamber 32. A major benefit is that it is possible to enhance control of the temperature at which the freeze drying/sublimation begins, allowing the process to be conducted at the most efficient temperature, i.e., at just below the collapse temperature of the target substance, while ensuring that the collapse temperature of the substance is not exceeded during the process. Because transfer between the freeze chamber 31 and the dryer chamber 32 is effected at vacuum pressure (e.g., typically 1 O^mbar) the heat transfer by gaseous conduction and/or convection is almost entirely non-existent. The warming rate is thus significantly slower than in the previously trialled embodiment. There is consequently the facility to tailor the process to vary the warming rate over a greater range than previously. A graphic comparison of pressure on temperature for the present system and the previously trialled system is shown in Figure 5. The results of the previously trialled system are identified as 'P before' and '-58 before'. The effect on the freezing rate of setting the cold plates at a higher temperature is shown in Figure 6. Additionally, for example, a required temperature profile of the target substance may be achieved with a reduced thermal moderation effect in the carrier that may allow a faster overall processing time. Further, the cold plates in the freeze chamber 31 can be set at a higher temperature than in the previously trialled embodiment, thus enabling the freezing rate to be reduced.

Conditions in the drying chamber 32 are such that following sealing and at ambient pressure the ambient temperature is in the region of, for example, 2O ⁰ C to 25°C. The dryer chamber 32 includes a heater arrangement including spaced warming plates above and below the conveyor.

It is important to be able to control the temperature during the drying stage in the dryer chamber 32. To tune or tailor the warming rate, the nitrogen bleed to the dryer chamber 32 can be varied or the temperature of the heater plates in the dryer chamber 32 can be altered. For practical purposes there is an upper limit to which the temperature of the plates can be operated without detrimentally affecting enzyme stability. The vacuum pressure (Pi _ow ) in the dryer chamber 32 can be tailored by the rate of the purge gas bleed into the dryer chamber 32 and has been found to work well at Pi _ow set to 0.05mbar. Introducing nitrogen and increasing

the pressure (Pi _ow ) speeds up the warming rate. Similarly, the warming rate maybe controlled by other process characteristics, for example, thermal moderation. For example, insulation may be provided to decrease the rate of warming. Therefore tailoring of the process conditions in the dryer chamber 32 can be used to adjust the warming rate to reach optimum results. The correct balance needs to be found to produce the desired freeze dried target structure at a commercially viable rate of production. A fast warming rate might shorten the process but can run the risk of collapse of the structure. A slow warming rate lengthens the process and might not remove sufficient solvent/water. It is believed that the inert gas (nitrogen) bleed into the chamber also operates to increase rate of water removal from the deposited, dried substance. The effect of pressure and inert gas (nitrogen) bleed for two volumes of Pi _ow is shown in Figure 7.

Following operating the reduced pressure regime in the dryer chamber 32 for the required period, the dryer chamber 32 is back filled with nitrogen until atmospheric pressure is achieved once again within the dryer chamber 32. At this point the slit valve 38c is operated to open the dryer chamber 32 and a conveyer is activated to pass the base sheet 30 out of the dryer chamber 32.

It has been found that the temperature at which the sublimation occurs can be tailored by thermal moderation of the base sheet 30. The addition of a thermal moderator means (for example on the bottom of the base sheet 30) can have a number of effects. The thermal moderator means can slow warming, firstly as the sheet is passed between the freeze chamber

31 and the dryer chamber 32, and secondly, when present in the dryer chamber 32. It has been found that insulating the base sheet 30 may produce a lowering in the pressure at which sublimation occurs as a result of the decrease in the actual temperature of the base sheet 30.

Thus the use of appropriate thermal moderator means arrangements for the sensors and base sheet 30 can ensure that the freeze drying process and sublimation process parameters (including temperature) can be tailored to produce enhanced effect and result in a dried deposit of superior characteristics.

The heating plates in the dryer chamber 32 may in certain circumstances alternatively be operated to cool the chamber environment, by being operated at a temperature below the dryer

chamber 32 temperature or ambient environment temperature. In this context they may be more accurately described as temperature control means, provided within the dryer chamber 32.

Whilst the purge and bleed scenarios have been described in relation to the use of nitrogen, it should be appreciated that other gases, particularly inert gases or dry air, maybe suitable. It is understood that in this patent, inert gas means a gas that does not react with the target substance and includes dry air.

In certain embodiments the base sheet 30 will exit the dryer chamber 32 and directly pass for onward processing (such as cutting out of the strips 10) and sealed packaging. In other certain embodiments, prior to this, a warming stage will be utilised in which the base sheet 30 passes from the dryer chamber 32 into a warm discharge chamber 33, which is maintained at a temperature above the dew point of the factory. The warm discharge chamber 33 includes a conveyor similar to the conveyor in the dryer chamber 32. An out- feed conveyor 47 is provided at the downstream end of the apparatus. The warm discharge chamber 33 may be purged with an inert gas (such as nitrogen) in a similar manner, and for similar reasons, as the dryer chamber 32. The heating plates in the warm discharge chamber 33 may in certain circumstances alternatively be operated to cool the warm discharge chamber 33, by being operated at a temperature below the chamber or ambient environment temperature. In this context, this may be more accurately described as a temperature control means, provided within the warm discharge chamber 33.

An important advantage of the invention is that efficacious and rapid freeze drying of the liquid target substance is able to be achieved. It is particularly of benefit to have the ability to rapidly reduce the pressure in the dryer chamber 32 to the desired level and, according to the preferred regime, in a chamber separate and distinct from the chamber in which the freeze process step is conducted. This enables the liquids in the target substance to sublime effectively, resulting in a high quality dried solid residue remaining. Because the carrier is moved from a freeze chamber to a separate dryer chamber the environment can be more closely controlled and the cycling of freezing to drying can be more rapid. In this way the base sheet carrier and target substance are subject to the full temperature cycle without the

surrounding chamber and apparatus being subject to the foil cycle. This reduces time and energy costs.

The technique and apparatus of the present invention enables the freezing and vacuum drying of target substances (held on base sheets or otherwise) to be achieved in a continuous or quasi- continuous manner, in which separate freeze and dryer chambers are utilised. It is also possible to have a warming chamber. The system of the invention enables additional freeze, dryer, or warming chambers to be added in circumstances where this is beneficial.

In certain embodiments it is envisaged that there could be multiple chambers at different temperatures and pressures to enable control of product structure.

This freeze drying process enables the control of a number of variables that define the structure of the target substance when freeze dried, and the optimization of the formed material. The factors that make an optimal freeze dried material depend on the chemical composition and the precise requirements of the particular use to which the target substance is put and hence change depending on that use and composition. The factors of general importance are crystal size and structure, porosity, moisture content, residual strength, degree and type of cracking, ease of solubility, and volume. However, other parameters maybe of importance in specific circumstances.

Li alternative arrangements it is envisaged that a number of base sheets could be fed simultaneously (or sequentially) into, and passed out of, the freeze chamber and/or the dryer chamber. Further, the liquid reagent could be held in containers, wells, or vessels rather than presented on base sheets. The process and apparatus is envisaged as having applications in other situations in which rapid and accurate freeze drying of liquids (particularly small dosed quantities) is required.

An important aspect of the technique is that a vacuum system exists to apply reduced pressure in the dryer chamber and at the final stage of the freeze process in the freeze chamber. This is in contrast to the previously trialled arrangement as disclosed in WO2007/066132 (sometimes referred to herein as 'old' mode, technique, or protocol), where the vacuum was applied in the second chamber only. The application of a vacuum in the first chamber means that the

deposited cakes no longer warm up significantly when they are transferred between the two chambers, hi the previously trialled arrangement based on the disclosure of WO2007/066132 the freeze process needs to commence at -58 ⁰ C (which is around 25 ⁰ C lower than the target substance's collapse temperature) to ensure that the cakes remain below their collapse temperature during transfer, whereas the modified machine can start at a warmer temperature, say -26°C, which is just a few degrees below the target substance's collapse temperature, resulting in a number of benefits. Firstly, it is no longer necessary to put the cakes onto a surface with a high thermal mass to ensure that they remain cool during transfer. Secondly, it is possible to use formulations with lower collapse temperatures - i.e., a greater range of freezing temperature is accessible. Thirdly, the system can now run close to the collapse temperature of the formulation. This means that the freezing process occurs in a more controllable manner, enabling control of the size of the ice crystals. It is believed, from experimental testing, that this may offer significantly better properties.

The system as described in WO2007/066132, and trialled ('old' mode or protocol) has been able to adequately freeze dry formulations with collapse temperatures above -20 ⁰ C, however, below this temperature difficulties have been encountered. The present invention enables freeze drying of formulations with lower collapse temperatures more easily and/or the system to be run in a more efficient manner, for example, by enabling the system to be run without using so much energy in cooling.

The previously trialled mode of system operation involved freezing the card, base sheet, or other carrier in the freeze chamber and then transferring to the dryer chamber, where the temperature is warmer, and the vacuum was pulled immediately. The new mode of system operation involves freezing the card in the freeze chamber, then opening the valve between the freeze and dryer chambers and pulling a vacuum in both chambers simultaneously. Thus, in the new mode, sublimation begins while the card is still on the cold plate. This sublimation causes evaporative cooling which keeps the cakes cool as the card is then transferred to the warmer dryer chamber under vacuum, with the increase in temperature as soon as the card leaves the cold plate increasing the rate of sublimation. The differences in the two protocols are summarised below.

The difference and benefits between the old and the modified CFD have been characterised in a number of experiments in which the system has been used with the same cold plates temperature as the previously trialled system as disclosed in WO2007/066132, operating having cold plates temperatures of -58 ⁰ C, and in the new protocol having new cold plates temperatures of -35°C.

Examples

Experimental data results summary

Experiments have been completed to demonstrate the benefits of the modified technique over the previously trialled system as generally disclosed in WO2007/066132. The improvements are summarised below.

The pressure/time graphs enable calculation of rate of sublimation and water removal. This data shows significant difference for the new -35 ⁰ C data. The peak in the sublimation rate is a very clear difference between the previously trialled technique and the new system and technique of the present invention.

hi the case of operation of the system of the present invention at -35 ⁰ C, since the warming rate is slower and is maintained below the collapse temperature throughout the process, the core of the cake is not collapsed and so water can be more easily removed than under the regime of the previously trialled system. A general difference between the techniques is that the old drying process was controlled by pressure whereas the new one is controlled by temperature.

With the previously trialled system, it was necessary to use a fast freezing rate. However, a major advantage of the system and technique of the present invention is that the freezing rate may be controlled as a separate process allowing the freezing rate to be adjusted according to the desired process speed.

With the system and technique of the present invention the warming rate is much more controlled and can be varied by the user by changing the warm plates' temperature, the presence of a nitrogen bleed, and the thickness of the substrate base sheet or card.

The visual assessment data showed that the old -58 ⁰ C and the new -35 ⁰ C give very similar cakes in terms of height and brightness - in fact the new procedure appears to give a slight improvement in both cases. The visual assessment also indicated that, because of the slower freezing rate, cakes with larger ice crystals were produced.

Calibration performance

The process of the present invention gives improved gradients over the previously trialled process.

Strip Construction

Sheets of electrodes were manufactured in accordance with generally known prior art arrangements. Screen printed electrode strips with laser drilled wells were used as disclosed, for example, in WO200356319.

Enzyme solutions for a Standard Total Cholesterol sensor (TC) were used, these being:

0.1 M Tris pH 9

0.05 M MgSO ₄

5% glycine l% ecotine 1% inositol

80 mM ruthenium hexaamine trichloride

8.8 mM thio nictotinamide dinucletotide

5% CHAPS

5% deoxy-big CHAPS 3.3 mg/ml cholesterol esterase 4.2 mg/ml putidaredoxin reductase 66 mg/ml cholesterol dehydrogenase

Ln producing the biosensors on the strips, the enzyme solutions were dispensed into the electrode fragments using a 4-head dispensing machine, used to accurately dispense 350nl of TC solution into every well on each sheet, which totals 176.4μl per sheet (4 wells per electrode, 126 electrode strips per sheet). These sheets were then freeze dried. Each sheet was subjected to a different freeze dry protocol.

CFD Protocol Old -58 ^{C 1} C Old -35 ^{C 1} C New -35 ⁰ C

Sheet T0050 / 0118 T0050 / 0121 T0050 / 0122

Freezing -58 ⁰ C for 5 -35 ⁰ C for 5 -35°C for 5 mins mins mins

Freeze - Vacuum NA NA 1 min

Warm - Vacuum 5mins 5mins 5mins

After freeze drying, the sheets were immediately removed to a low humidity environment.

The strips on the sheets were tested by chronoamperometry using an Autolab (PGSTAT 12) and a multiplexer (MX452, Sternhagen Design). At T=O seconds the chronoamperometry test was initiated using the multiplexer attached to the Autolab. 15 repeat oxidations at + 0.15V for 1 second were performed, followed by a final reduction current at -0.45V for 1 second. There was a 15 second delay between oxidations which resulted in oxidations at approximately 0, 14, 28, 42, 56, 70, 84, 98, 112, 126, 140, 154, 168, 182, 196, and 210 seconds. Data was analysed for current values at 1 second on the transient using an in-house procedure. Plasma samples were used for all the testing. The plasma was analysed for its reference TC and TRG values on a spACE analyser.

The temperature profile through the freeze dryer system was measured using an I-button DS1922L-F50 temperature sensor fitted on a fully assembled card.

Results

The results centre on comparisons between the previously trialled system as generally described in WO2007/066132 and the system and technique of the present invention. Two differences of the present invention over the previously trialled technique and system are the temperature of the cold plates and the presence of a vacuum during the card/base sheet transfer. It should be noted that because in the modified case the volume to evacuate is twice as large (because both of the two chambers are evacuated) it takes longer to reach 10 ^"1 mbar (i.e., vapour pressure at -4O ⁰ C, pressure below which it is safe to transfer the card). In these experiments it was found to take 40 s as opposed to 26 s in the previously trialled system as generally described in WO2007/066132. Hence the freeze- vacuum time (time during which the vacuum is applied but the card is still in the freeze chamber) is clearly an additional variable parameter.

Because in the modified system of the present invention, the cold plates can be set at a higher temperature, the freezing rate can be reduced. Experimentally it is found to almost half from a maximum of 200°C/min to 100°C/min when the temperature is changed from -58 ⁰ C to -35 ⁰ C (see Figure 6). The freezing rate is still rapid but the reduction might induce the crystallisation of larger ice crystals.

The system can be used with a Pirani gauge attached to the dryer chamber which monitors the pressure in that chamber only. The pressures recorded in the dryer chamber for each of the runs can be compared with those of an empty card (and knowing the efficiency of the pump and water vapor pressure), and so the amount of water released from each sheet can be estimated. The pressure changes and subsequent calculated water release for the modified and the unmodified freeze drier are given in Figure 8 and Table 1.

In the modified freeze dryer system of the present invention, the base sheet card is transferred when the pressure is below 10 ^"1 mbar, and hence heat transfer by gaseous conduction and/or convection is nearly non-existent. Consequently the warming rate is much slower than in the previously trialled technique in which the vacuum is drawn in the dryer chamber only.

Figure 9 shows the differences in warming rates between the two configurations. It is clear in

the previously trialled technique that a decrease in the warming rate is observed between -3O ⁰ C and -2O ⁰ C. For the system as described in WO2007/066132, this is when the vacuum is first applied. Using the modified freeze dryer it is possible to vary the warming rate in a controlled manner by varying both or either the temperature of the warming plates and the presence and rate of nitrogen bleed. It can be important to be able to control the rate of warming because though a rapid warming rate clearly shortens the process it could put the cakes at risk of collapse while a slow warming rate lengthens the process and may result in insufficient water removal within an acceptable time. Hence control of this process is vital.

An additional variable that can be used to control the warming rate is the thickness of the substrate card or base sheet, because a thinner card or base sheet has a lower thermal mass it warms up more rapidly. Examples of controllability of the warming rate by varying the warm plates' temperature, the presence of a nitrogen bleed, and the thickness of the card are shown in Figure 10. Details of the different conditions are given in Table 2. Experimentally, the nitrogen bleed is introduced into the process by setting the minimum pressure to reach (Pi _ow ), and when the pressure reaches a value lower than Pi _ow , the nitrogen bleed switches in to stabilize the pressure at Pi _ow .

The cakes were photographed using an 'OGP Smartscope' camera system under standard, controlled lighting (Figures 11, 12, and 13). This camera system has a calibrated focal length gauge which was used to determine the height of each of the cakes. Each picture was then analysed using the 'Image J' program that selects a predetermined area within the cake and counts the number of pixels with the same greyscale value, from 0 = black to 255 = white. A standard deviation of these greyscale values can be calculated for each cake measured allowing the homogeneity of the cake to be assessed (Figures 14 and 15). These results clearly show that the least dense and brighter freeze dried cakes (more porous) were achieved using the technique of the present invention at -35 ⁰ C.

Using the cold plates at a higher temperature decreases the freezing rate, which favours growth of large ice crystals. Experimentally, a 5% KCl solution was freeze-dried using different cold plates temperatures (-20, -40, and -58 ⁰ C). The resulting cakes exhibit different structures, with the cakes that were frozen slowly at -2O ⁰ C and -4O ⁰ C exhibiting large crystals, a few hundreds of microns in size (Figures 16 and 17), while the cake frozen at -58 ⁰ C shows a much denser

structure (Figure 18). Figures 16 to 18 were recorded using a Canon Powershot A80.

Some of the electrodes were constructed with top fill 'X-type' flow cells and calibrated against the spACE analyser. The electrodes were tested with 20μl of thawed plasma (Figure 19, Table 3).

Specific advantages have been identified over prior art techniques and also the previously trialled ('old') technique as generally described in WO2007/066132.

1. The cold plates can be conditioned to be near the collapse temperature.

2. An increased range of freezing temperatures is accessible. This means that it is not necessary to reformulate enzyme mixes so that they have high collapse temperatures. Instead, the system working parameters can simply be varied (this is a large commercial advantage).

3. A lower freezing rate is facilitated, enabling greater control of the frozen cake structure.

4. The system gives the ability to freeze at a higher temperature and still obtain satisfactory results. Thus processing times can be reduced.

5. At low temperatures, very little energy is brought to the cakes and so sublimation is slow. What was found particularly surprising is that the higher freezing temperature that can be used by the system in accordance with the invention, results in a lower temperature during the sublimation in the drying chamber.

The system enables significantly more control over the warming rate and temperature of the base sheet card throughout the process. This provides improved temperature control when the card moves between chambers. Also, the thermal insulation of the cakes is no longer required (i.e., the base sheet cards do not necessarily need to be placed on high thermal mass plates).

A key feature of the present invention is the ability to start the freezing process at temperatures close to the collapse temperature of the cake and to maintain that temperature while

compensating for evaporative cooling, thus giving the fastest and most efficient freeze drying cycle, resulting in a slower cooling rate producing cakes with structures that are optimal for use.