WO2001072424A1

WO2001072424A1 - Heating specimen carriers

Info

Publication number: WO2001072424A1
Application number: PCT/GB2001/001284
Authority: WO
Inventors: Ian Alan Gunter
Original assignee: Bjs Company Ltd.
Priority date: 2000-03-24
Filing date: 2001-03-23
Publication date: 2001-10-04
Also published as: GB0007219D0; EP1265712B1; JP2003527873A; CA2403973C; DK1265712T3; ES2329009T3; DE60139246D1; AU774195B2; JP4965786B2; PT1265712E; AU3942601A; CA2403973A1; EP1265712A1

Abstract

A method and apparatus for heating specimens in wells of a metallic specimen carrier. The specimen carrier is heated by applying resistive heating directly to the carrier. An AC source and transformer may be used where the specimen carrier is in series with a single turn secondary winding of the transformer. A magnet may be placed in each well and excited by the heating current to provid a stirring effect.

Description

"Heating specimen carriers "

The present invention relates to heating and more particularly to the thermal

cycling of specimen carriers.

In many fields specimen carriers in the form of support sheets which may have a multiplicity of wells or impressed sample sites, are used for various processes

where small samples are heated or thermally cycled.

A particular example is the Polymerase Chain Reaction method (often referred to as PCR) for replicating DNA samples. Such samples require rapid and accurate thermal cycling, and are typically placed in a multi-well block and cycled between several selected temperatures in a pre-set repeated cycle. It is important that the temperature of the whole of the sheet or more particularly the temperature in each well be as uniform as possible.

The individual samples are normally liquid solutions, typically between lμl and

200μl in volume, contained within individual sample tubes or arrays of sample tubes that may be part of a monolithic plate. It is desirable to minimise

temperature differentials within the volume of an individual sample during thermal processing. The temperature differentials that may be measured within a liquid sample increase with increasing rate of change of temperature and may limit the maximum rate of change of temperature that may be practically employed. Previous methods of heating such specimen carriers have involved the use of attached heating devices such as wire, strip and film elements and Peltier effect

thermoelectric devices, or the use of indirect methods where separately heated fluids are directed into or around the carrier

The previous methods of heating suffer from the disadvantage that heat is

generated in a heater that is separate from the specimen carrier that is required to be heated.

The thermal energy must then be transferred from the heater to the carrier sheet, which in the case of an attached heater element occurs, through an insulating barrier and in the case of a fluid transfer mechanism occurs by physically moving fluid from the heater to the sheet.

The separation of the heater from the block introduces a time delay or "lag" in the temperature control loop. That is to say that the application of power to the

heating elements does not produce an instantaneous or near instantaneous increase in the temperature of the block. The presence of a thermal gap or barrier between

the heater and the block requires the heater to be hotter than the block if heat energy is to be transferred from the heater to the block. Therefore, there is a further difficulty that cessation of power application to the heater does not instantaneously stop the block from increasing in temperature. The lag in the temperature control loop will increase as the rate of temperature change of the block is increased. This can lead to inaccuracies in temperature control and limit the practical rates of change of temperature that may be used.

Inaccuracies in terms of thermal uniformity and further lag may be produced when

attached heating elements are used, as the elements are attached at particular

locations on the block and the heat produced by the elements must be conducted from those particular locations to the bulk of the block. For heat transfer to occur

from one part of the block to another, the first part of the block must be hotter than the other.

Another problem with attaching a thermal element, particularly a Peltier effect device, is that the interface between the block and the thermal device will be subject to mechanical stresses due to differences in the thermal expansion

coefficients of the materials involved. Thermal cycling will lead to cyclic stresses that will tend to compromise the reliability of the thermal element and the integrity of the thermal interface.

More recently our PCT application GB97/00195 has disclosed a novel method where the specimen carrier is metallic and direct electrical resistive heating is applied to the metallic specimen carrier. The Specification of the aforesaid PCT application discloses various features of heating the carrier and the whole of that disclosure is deemed to be part of this Specification.

Now, a problem with heating samples in sample wells of such a carrier is that

agitation or stirring is sometimes desirable. The present invention aims to solve

that problem.

Accordingly the invention provides a method of heating a specimen carrier in the form of a metallic sheet and in which a matrix of sample wells is incorporated in

the sheet, which method includes applying an alternating current to said sheet to provide heating of the samples in the wells, and a magnet is loosely contained within at least one well and is arranged to be

agitated by the alternating current so as to provide a stirring action during the heating. Usually, but not necessarily always, each well will contain a magnet.

The sheet may be of silver or similar material of high thermal and electrical conductivity and will generally have a thin section in the region of 0.3mm

thickness, where the matrix of sample wells is incorporated in the sheet. The sample wells may incorporate samples directly or may carry sample pots or test tubes shaped to closely fit within the wells.

The sheet may have an impressed regular array of wells to form a block and a basal grid or perforated sheet may be attached to link the tips of the wells at their closed ends to form an extremely rigid three-dimensional structure. In some

applications the mechanical stiffness of the block is an important requirement.

Where a basal grid is used, heating current is also passed through the metal of the

grid. The basal grid is preferably made of the same metal as the block.

While the metallic sheet may be a solid sheet of silver (which may have cavities forming wells) an alternative is to use a metallised plastic tray (which may have

impressed wells), in which deposited metal forms a resistive heating element.

Another alternative is to electro form a thin metal tray (which again may have impressed wells), and to coat the metal with a bio-compatible polymer.

These measures enable intimate contact to be achieved between the metallic

heating element and the bio-compatible sample receptacles. This gives greatly improved thermal performance in terms of temperature control and rate of change of temperature when the actual temperatures of the reagents in the wells is measured.

The plastic trays are conventionally single use disposable items. The incorporation of the heating element into the plastic trays may increase their cost, but the reduction in cycling time for the PCR reaction more than compensates for any increased cost of the disposable item.

The bottom of the composite tray should be unobstructed if fan cooling is

employed. If sub-ambient cooling is required at the end of the PCR cycles, either

with a composite tray or a block, chilled liquid spray-cooling may be employed. The boiling point of the liquid should be below the low point of the PCR cycle

so that liquid does not remain on the metal of the tray or block to impede heating. This also allows for the latent heat of evaporation of the liquid to increase the

cooling effect.

The heating current may be an alternating current supplied by a transformer system wherein the heating power is controlled by regulating the power supplied

to the primary winding of the transformer. The sheet to be heated may be made part of the transformer secondary circuit. The secondary winding may be a single

or multiple loop of metal that is connected in series with the sheet. By these means, the high current, low voltage power that is required to heat the highly conductive sheet may be simply controlled by regulating the high voltage, low current power supplied to the primary winding of the transformer.

The transformer may comprise a toroidal core having an appropriate mains primary winding and a single bus bar looped through the core and connected in series with the metallic sheet to form a single turn secondary circuit. Generally the sample wells may be conical in shape. This helps any stirring action of each magnet within the respective well.

More specifically, in direct resistance heating using alternating current, an

oscillating magnetic field is produced at each well by the heating current. A small

bar magnet, (typically 5mm long by 1mm diameter), may be placed in each

sample tube and the heating current will cause oscillating forces to be applied to the magnet. The geometry of the conical section of the sample tube will then

constrain the bar to spin about an axis that is not coaxial with, or normal to, the axial dimension of the bar. The stirring action is then similar to that which would

be produced by vigorously stirring each individual tube with a manual stirring rod.

The magnets may be made of readily available materials, in particular hard magnetic alloys such as Alnico 4. Rare earth magnets (for example

iron-neodymium-boron or samarium-cobalt) may also be used. To prevent contamination of the liquid sample, the magnet may be given an inert coating. Such a coating may be of a bio-compatible polymer such as polypropylene or polycarbonate, or a noble metal such as gold. A noble metal coating has the

advantage that it adds no significant volume to the magnet when applied in a coating of sufficient thickness to ensure that the coating is not porous. When using gold a 5μm thickness is sufficient to provide a pore-free coating, and adds a volume of 0.08μl to the magnet. The magnets cost much less than the typical reagent mix to be placed in a sample tube, and may therefore be regarded as consumable items. However the magnets

may clearly be easily sorted from the waste reagents for cleaning and re-use.

The magnets may be small. In particular embodiments, for a lOOμl liquid sample,

a magnet 1mm in diameter and 5mm long may be employed. Such a magnet has

a volume of 3.9μl. A 0.5mm diameter by 3mm long magnet may be provided for use in smaller tubes and would have a volume of 0.58μl. The approximate masses

of these magnet examples would be 31mg and 4.5mg respectively.

In certain embodiments, a magnet is placed in each of the wells to be agitated. In standard practice the shape of the individual wells is conical and the magnet

length is chosen such that the long axis of the bar magnet is constrained to be within a range of between 5 and 30 degrees of the axis of the well. Such

orientation ensures that the agitation magnet will spin eccentrically and will not jam in the well. The diameter of the magnet should be as small as is practical, in order to minimise the volume of the magnet. The passage of the alternating

heating current through the block gives rise to an alternating magnetic field circling the block in a plane normal to the direction of current flow. The alternating magnetic field causes alternating forces to be applied to the bar magnets as they try to align themselves with the magnetic field. The conical shape of the wells constrains the movement of the magnets, which then spin eccentrically in each well.

The effect of the eccentric spinning of the magnets is to vigorously stir the liquid

sample in each of the wells to which a magnet has been introduced. The stirring

effect almost completely eliminates any of the temperature differentials that may

be observed in a static sample during thermal cycling.

Preferably, the bottom of the sheet, even if a basal grid is attached, has an open

structure with a large surface area. Such a surface is ideal for forced-air cooling. Moreover, preferably there are no attached elements to impede free and full contact between the metal of the sheet and moving air.

Ducting of the air may be provided to encourage even cooling effects over the extent of the sheet. To allow for controlled cooling rates, the air movement may be under proportional control. The control response time of a device that imparts movement to air, for instance a mechanical element such as a fan, is slow

compared to the fast electronic control response of the heating system. The heating system may therefore be used together with the fan to control the temperature

changes of the sheet during cooling.

The secondary winding in series with the sheet may have more than one loop through the core of the transformer. The power supply means and control for the heating current may be a high frequency AC power supply permitting a reduction in the amount of material in

the transformer core.

The thermal uniformity of the sheet will be dependent on the heating power dissipation at any point in the sheet being matched to the thermal characteristics

of that point. For instance, a point around the centre of the sheet will be surrounded by temperature controlled metal, whereas a point at the edge of the

sheet or block will have temperature controlled metal on one side and ambient air on the other. The geometry of the sheet may be adjusted with the aim of achieving thermal uniformity. In general practice the geometry of sample sites or wells of a sheet or block will be a standardised regular array. The industry standard arrays consist of 48, 96 or 384 wells in a 110 X 75mm rectangular plate or block. These layouts are arbitrary and larger arrays of 768 and 1536 wells are appearing.

Typically, the geometric factors that may be varied comprise the thickness of the

metal from which the sheet is formed, and if a basal grid is used, the geometry of the webs in the plane of the grid.

Embodiments of the invention will now be described by way of example with reference to the accompanying diagrammatic drawings in which: Figure 1 is a side elevation of a heating apparatus;

Figure 2 is a plan view of the apparatus of Figure;

Figure 3 is a side view of sample tubes incorporating magnets and located in wells of a sheet of the heating apparatus of Figure 1 ;

Figure 4 is a top plan view showing the magnet location, and

Figure 5A to 5C shows a perspective, plan and side view of the block specimen carrier of the apparatus shown in Figure 1.

A metallic sheet specimen carrier in the form of a multi-well block (1) measuring 110mm x 75mm and having 96 wells (2) disposed in a grid layout is formed in

silver nominally 0.3mm thick. This is attached to bus bars (3) of substantial cross-sectional area. The bus bars loop once through a transformer (toroidal or square), core (4). The core (4) has a primary winding (5) appropriate for the mains voltage employed. The bus bars (3) also act as a structural member

supporting the block (1). The transformer primary current is controlled using a triac device (6). The triac device receives current from an AC source and is

controlled by a temperature control circuit (7) which uses at least one fine wire thermocouple (8) soldered to a central underside region of the block to sense the temperature of the block. The temperature control circuitry may be operated manually or by a personal computer (9). More specifically, the heating power may

be controlled by proportional phase angle triggering of the triac (6) in response to

signals from the thermocouples (8) combined with programmed temperature / time

information entered to describe the required thermal behaviour of the apparatus.

Cooling of the block is by means of a fan (10) mounted under the block, passing

ambient air over the protruding well forms (2), the air being directed by the

enclosure in which the block is mounted. The fan is controlled by the same temperature control circuitry that drives the heater triac. Although not shown in detail, the airflow is guided to give even cooling of the block (1) by means of multiple shaped air inlets on the top, sides and bottom of the apparatus enclosure. The fan extracts air from the inside of the enclosure. The negative pressure within the case is varied proportionally by proportionally controlling the fan speed.

It will be appreciated that the rear surface of the block (1) has a large surface area which is ideally suited to the dissipation of heat.

The measured performance of the example apparatus gives rates of change of temperature in excess of 6 degrees per second and over/under shoots of less than

0.25 degrees within the typical PCR working range of 50-100 degrees. The thermal uniformity of the block is such that within 10 seconds of any temperature transition, even at rates of change of temperature in excess of 6 degrees Celsius per second, the range of temperatures that may be measured in wells around the

block does not vary more than +/- 0.5 degrees from the mean temperature.

The block (1) of the present embodiment will have an electrical resistance of

around 0.00015 Ohms. To obtain the levels of heating desired, a current in the

order of 1600A is supplied to the block. The order of this required current is

easily calculable on the basis of the size of the block and the innate properties of silver. The current in the primary winding (5) might be up to around 3 A at 240N or 7A at 110V. Thus even though high current is supplied across the block (1), the voltage across the block remains low, say 0.25N. Further, the block (1) and bus bars (3) are isolated from mains power and may be connected to ground to enhance safety further.

The described example uses a silver block with cavities, but metallised plastic tray

inserts, or electro formed thin metal trays, as previously described, may also be used.

The system as described has several important advantages.

1.1 The block is heated directly with no requirement for heat transfer from an attached heat source. This is very efficient and taken together with the very low specific heat capacity of silver allows very rapid temperature changes.

1.2 Direct heating means that there is no thermal lag at all. Temperature control functions are immediate so that the block may be cycled in temperature

with little or no over or undershoot. Temperature control is therefore inherently precise.

1.3 Since there are no obstructions or thermal barriers attached to the

block, simple forced-air cooling of the back of the block provides rapid and controllable cooling.

1.4 The fine wire thermocouple is soldered directly to the block so as to provide close temperature measurement and control. Any other temperature measurement device may be used as long as it does not introduce significant sensor lag.

1.5 The temperature distribution around the surface of the block is

dependent on the evenness of heating and the thermal conductivity of the block. The thermal conductivity of silver is very high, and the distribution of heat energy around the block is dependent upon the distribution of the heating current. This

may be regulated by varying the geometry of the multi-well block. The variation in geometry will typically be achieved by spatial variation in the thickness of the block (1) such that, (for instance), the minimum metal thickness (of about

0.25mm), may be found at the middle of the block surface and the maximum

metal thickness (of about 0.4mm), may be found along the edges of the block (1) parallel to the longer axis. The variations in metal thickness are used to maintain

thermal uniformity across the area of the block during thermal cycling by compensating for the differing thermal environments experienced by different

points in the block (1).

The variations in metal thickness are produced whilst manufacturing the block by electroforming. During the electroforming process the distribution of the electrodepositing current is modulated such that the depositing current is higher in areas where a greater thickness of metal is required.

The overall geometry of the block is standardised to accept liquid samples of 20-1 OOμl contained in either individual 200μl sample tubes or arrays of samples contained in a 96 well microplate.

The large currents required may be easily produced and controlled since the block becomes part of a heavy secondary circuit of the transformer. The cross-sectional area of the winding bars is made considerably larger than the cross-sectional area of the block so that significant heat generation only occurs in the block. The current can be easily controlled in the primary winding (where the current is small), using thyristors, triacs or other devices. Alternatively, the primary winding may be driven by a high frequency, switch mode, controllable power supply. This

allows the same degree of control of the current induced in the secondary winding incorporating the block, but the high frequency allows the use of a more compact

core in the transformer.

Referring now to Figures 3 and 4, a novel stirring arrangement is shown. A

sample carrier (l)(which is equivalent to the block (1) described above) has conical cavities (12) carrying 200μl sample tubes (13). Then, within each tube is loosely carried a magnet (14).

Each is a small bar magnet, (typically 5 mm long by 1mm diameter), which is placed in each sample tube and the heating current is then able to cause oscillating

forces to be applied to the magnet. The geometry of the conical section of the sample tube will then constrain the bar to spin about an axis that is not coaxial with, or normal to, the axial dimension of the bar. The stirring action is then similar to that which would be produced by vigorously stirring each individual

tube with a manual stirring rod.

The magnets can be made of readily available materials such as Alnico 4 and coated with non-reactive materials such as polypropylene or PTFE or nobel metals such as gold, for example a 5μm layer of acid hard gold plating may be used. The magnets cost much less than the typical reagent mix to be placed in a sample tube,

and may therefore be regarded as consumable items. However the magnets may clearly be easily sorted from the waste reagents for cleaning and re-use.

The magnets are small, 1mm diameter by 5mm long which gives a volume of

3.92μl for use in a 200μl sample tube. A 0.5mm diameter by 3mm long magnet for use in smaller tubes has a volume of 0.58μl. The approximate masses of these

magnets are 1mg and 4.5mg respectively.

The action of the agitation magnets not only removes measurable temperature differentials from the lOOμl liquid samples used, but also increases the overall rate

of heat transfer from the block to the sample. Thus the programmed temperature/time profile is more accurately reproduced in the thermal processing

experienced by the liquid sample.

Figures 5 A to 5C show the sample carrier sheet (block) (1) of Figures 1, 2 and 4 in more detail. As desribed above this metallic specimen carrier is in the form of a multi-well block (1). This block (1) measures 110mm X 75mm and has an 8 X 12 array of standardised conical wells 12mm deep and is formed in silver having

an average metal thickness of 0.33mm. An attached basal grid may also be provided which ties together to exterior bottoms (101) of the wells. It will be seen that the wells in the sheet (1) have a significant depth and thus include side walls (102) and have an overall generally frustoconical shape. The wells are arranged to accept and surround a significant portion of any sample

tubes positioned in the wells. This can help in the efficient transfer of heat into

and/or out of samples. A large surface area of tube is in contact with the sheet (1).

Furthermore, in cooling it will be noted that this large area of tube is in direct

contact with a portion of the sheet, ie the exterior or underside of the wells, over which ambient air is fed.

Similar considerations also apply if samples are placed directly in the sheet rather than in a sample tube.

It has been found that mains frequency currents eg 50Hz provide a good stirring effect.

The fact that the rear of the carrier sheet is exposed can lead to various other advantages, in particular other apparatus may be located behind the sheet and/or access to the rear of the sheet is easy to obtain. In a particular alternative, a

method and apparatus for realtime analysis or monitoring of reactions occurring in the sample sites during heating and/or stirring can be provided. This may be implemented by providing a optical probe in each sample site or well, typically this probe will be the tip of a optical fibre which is located in an aperture towards the base of the well. The fibre in each well will lead away from the rear (or underside) of the sheet to suitable transmitter, receiver and analysis equipment.

The monitoring will typically make use of the fact that the fluorescing

characteristics of the reagents change as the reaction progresses. Thus an exciting

frequency of light will be fed from the transmitter along the fibres to each well.

This exciting frequency will cause fluorescence in the reagents and the emitted

light will travel back along the fibres to the receiver and analysis equipment where

the fluorescence or changes in fluorescence will be analysed to give an indication of the state of the reaction.

Claims

CLAIMS:

1. A method of heating a specimen carrier in the form of a metallic sheet

and in which a matrix of sample wells is incorporated in the sheet, which method includes applying an alternating current to said sheet to provide

heating of the samples in the wells, and wherein a magnet is loosely contained within at least one well and is arranged

to be agitated by the alternating current so as to provide a stirring action during the heating.

2. Apparatus for thermally cycling and stirring samples, the apparatus

comprising, a specimen carrier in the form of a metallic sheet, in which sheet a matrix of sample wells is incorporated,

means for applying an alternating current to said sheet to provide heating of the samples in the wells, and

a magnet loosely contained within at least one well which magnet is arranged to be agitated by the alternating heating current so as to provide a stirring

action during heating.

3. A method according to claim 1 or an apparatus according to claim 2 in which each well contains a magnet.

4. A method according to claim 1 or claim 3 or apparatus according to claim 2 or claim 3 in which the sheet is of a metal having a thermal and

electrical conductivity.

5. A method according to any one of claims 1, 3 and 4 or an apparatus

according to any one of claims 2 to 4 in which the sample wells are arranged

to incorporate samples directly.

6. A method according to any one of claims 1, 3 and 4 or an apparatus according to any one of claims 2 to 4 in which the sample wells are arranged to carry sample pots or test tubes shaped to closely fit within the wells.

7. A method according to any one of claims 1 and 3 to 6 or an apparatus according to any one of claims 2 to 6 in which the sample wells are conical in shape.

8. A method or apparatus according to claim 7 in which the or each magnet is a bar magnet and the geometry of the conical section of the sample tube constrains the bar to spin about an axis that is not coaxial with, or normal to, the axial dimension of the bar.

9. A method according to any one of claims 1 and 3 to 8 or an apparatus according to any one of claims 2 to 8 in which the or each magnet is coated

with a non-reactive material.

10. A method according to any one of claims 1 and 3 to 9 or an apparatus according to any one of claims 2 to 9 wherein the thickness of the material of

the sheet is varies as a function of position in the sheet in such a way as to promote even heating.

11. A method or apparatus according to claim 10 wherein the thickness of the sheet is thinner than an average thickness towards the centre of the sheet and thicker than an average thickness along edges of the sheet running

generally parallel to the direction in which current flows during operation.

12. A method or apparatus according to claim 10 or claim 11 in which the

sheet is formed by an electrodeposition process and the differences of thickness are acheived by controlling the electrodeposition process.