Formation of Layers of Amphiphilic Molecules
In one aspect, the present invention relates to the formation of layers of amphiphilic
molecules such as lipid bilayers. It is particularly concerned with the formation of high quality layers
suitable for applications requiring measurement of electrical signals with a high degree of sensitivity,
for example single channel recordings and stochastic sensing for biosensor or drug screening
applications. In one particular aspect, it is concerned with applications employing arrays of layers of
amphiphilic molecules, for example lipid bilayers. In another aspect, the present invention relates to
the performance of an electrode provided in a recess, for example for conducting electro-
physiological measurements.
The potential for using cellular proteins for biosensing and drug discovery applications has
long been appreciated. However there are many technical challenges to overcome in developing this
technology to fully realise the potential. There is a wealth of literature on using fluorescent and
optical approaches, but the focus of this document is on the measurement of electrical signals to
recognise analytes in biosensing.
In one type of technique, a layer of amphiphilic molecules may be used as the layer separating
two volumes of aqueous solution. The layer resists the flow of current between the volumes. A
membrane protein is inserted into the layer to selectively allow the passage of ions across the layer,
which is recorded as an electrical signal detected by electrodes in the two volumes of aqueous
solution. The presence of a target analyte modulates the flow of ions and is detected by observing the
resultant variations in the electrical signal. Such techniques therefore allow the layer to be used as a
biosensor to detect the analyte. The layer is an essential component of the single molecule biosensor
presented and its purpose is two-fold. Firstly the layer provides a platform for the protein which acts
as a sensing element. Secondly the layer isolates the flow of ions between the volumes, the electrical
resistance of the layer ensuring that the dominant contribution of ionic flow in the system is through
the membrane protein of interest, with negligible flow through the bilayer, thus allowing detection of
single protein channels.
A specific application is stochastic sensing, where the number of membrane proteins is kept
small, typically between 1 and 100, so that the behaviour of a single protein molecule can be
monitored. This method gives information on each specific molecular interaction and hence gives
richer information than a bulk measurement. However, due to the small currents involved, typically a
few pA, requirements of this approach are a very high resistance seal, typically at least 1 GΩ and for
some applications one or two orders of magnitude higher, and sufficient electrical sensitivity to
measure the currents. While the requirements for stochastic sensing have been met in the laboratory,
the conditions and expertise required limit its use. In addition, the laboratory methods are laborious
and time-consuming and are not easily scalable to high-density arrays, which are desirable for any
commercial biosensor. Furthermore, the fragility of single bilayer membranes means that anti-
vibration tables are often employed in the laboratory.

By way of background, existing techniques for forming layers of amphiphilic molecules such
as lipid bilayers will be reviewed.
Several methods for forming planar artificial lipid bilayers are known in the art, most notably
including folded bilayer formation (e.g. Montal & Mueller method), tip-dipping, painting, patch
clamping, and water-in-oil droplet interfaces.
At present, the bulk of routine single ion channel characterisation in research labs is
performed using folded bilayers, painted bilayers or tip-dip methods. These methods are used either
for the ease of bilayer formation, or for the high resistive seals that can be formed (eg 10-100GΩ).
Tip-dip bilayers and bilayers from patch-clamping of giant unilamellar liposomes are also studied as
they can be formed in a solvent free manner, which is thought to be important for the activity of some
protein channels.
The method of Montal & Mueller (Proc. Natl. Acad. Sci. USA. (1972), 69, 3561-3566) is
popular as a cost-effective and relatively straightforward method of forming good quality folded lipid
bilayers suitable for protein pore insertion, in which a lipid monolayer is carried on the water/air
interface past either side of an aperture in a membrane which is perpendicular to that interface.
Typically, the lipid is added to the surface of the aqueous electrolyte solution by first dissolving it in
an organic solvent, a drop of which is then allowed to evaporate on the surface of the aqueous
solution on either side of the aperture. Once the organic solvent has been evaporated, the solution/air
interfaces are physically moved up and down past either side of the aperture until a bilayer is formed.
The technique requires the presence of a hydrophobic oil applied as a pre-treatment coating to the
aperture surface. The primary function of the hydrophobic oil is to form an annulus region between
the bilayer and the aperture film where the lipid monolayers must come together over a distance
typically between 1 and 25 µm.
Tip-dipping bilayer formation entails touching the aperture surface (e.g. a pipette tip) onto the
surface of a test solution that is carrying a monolayer of lipid. Again the lipid monolayer is first
generated at the solution/air interface by evaporating a drop of lipid dissolved in organic solvent
applied to the solution surface. The bilayer is then formed by mechanical actuation to move the
aperture into/out of the solution surface.
For painted bilayers, the drop of lipid dissolved in organic solvent is applied directly to the
aperture, which is submerged in the aqueous test solution. The lipid solution is spread thinly over the
aperture using a paint brush or equivalent. Thinning of the solvent results in formation of a lipid
bilayer, however, complete removal of the solvent from the bilayer is difficult and consequently the
bilayer formed is less stable and more noise prone during measurement.
Patch-clamping is commonly used in the study of biological cell membranes, whereby the cell
membrane is clamped to the end of a pipette by suction and a patch of the membrane becomes

attached over the aperture. The method has been adapted for artificial bilayer studies by clamping
liposomes which then burst to leave a lipid bilayer sealing over the aperture of the pipette. This
requires stable giant unilamellar liposomes and the fabrication of small apertures in glass surfaced
materials.
Water-in-oil droplet interfaces are a more recent invention in which two aqueous samples are
submerged in a reservoir of hydrocarbon oil containing lipid. The lipid accumulates in a monolayer at
the oil/water interface such that when the two samples are brought into contact a bilayer is formed at
the interface between them.
In any of these techniques, once the bilayer has been formed, the protein is then introduced to
the bilayer either by random collision from the aqueous solution, by fusion of vesicles containing the
protein, or by mechanically transporting it to the bilayer, for example on the end of a probe device
such as an agar tipped rod.
There have been great efforts recently to increase the ease of bilayer formation using micro
fabrication. Some techniques have attempted essentially to miniaturise standard systems for folded
lipid bilayers. Other techniques include bilayer formation on solid substrates or directly on electrode
surfaces, through either covalent attachment or physical adsorption.
A large proportion of the devices that are capable of performing stochastic sensing form a
bilayer by using a variant of the folded lipid bilayers technique or the painted bilayer technique. To
date most have concentrated either on novel methods of aperture formation or on utilising the
emerging technologies in micro fabrication to miniaturise the device or to create a plurality of
addressable sensors.
An example is Suzuki et al., "Planar lipid bilayer reconstitution with a micro-fluidic system",
Lab Chip, (4), 502-505, 2004. Herein, an aperture array is created by etching a silicon substrate,
followed by a surface treatment to encourage the bilayer formation process, although the disclosed
rate of successful bilayer formation is very low (two out often).
A more recent example is disclosed in Sandison, et al., "Air exposure technique for the
formation of artificial lipid bilayers in microsystems", Langmuir, (23), 8277-8284,2007. Herein the
device fabricated from polymethylmethacrylate) contains two distinct aqueous chambers. Problems
with the reproducibility of bilayer formation are attributed to the difficulty in removing the excess
hydrophobic material from the aperture, and tackled by using a period of air exposure to aid the
bilayer formation process to thin the pre-treatment.
The devices of both Sandison et al. and Suzuki et al. are both miniaturised versions of a
standard painted bilayer technique with two distinct fluidic chambers separated by a septum
containing an aperture across which the bilayer is formed, one chamber being filled before the other.
This presents a number of difficulties for scaling up the system to a large number of individually

addressable bilayers, as at least one of the aqueous chambers must be a distinct chamber with no
electrical or ionic connectivity to any other chamber. Sandison et al. created a device with three fluid
chambers, each with separate fluidics, an approach which would be difficult to scale to large numbers
of bilayers. Suzuki et al. tried to address this problem by using a hydrophobic photoresist layer to
create small aqueous chambers on top of the aperture containing substrate. In this case, it is difficult
to control the flow of solution across the aperture containing interface and the use of small volumes
exposed to air makes the apparatus susceptible to evaporation effects. In both cited examples, the
need for the individual aqueous chambers for each bilayer means that a large sample volume must be
used to fill all the chambers.
An example of biosensor device using a supported lipid bilayer is disclosed in US-5,234,566.
The device is capacitive. A gated ion channel responds to an analyte, the binding of this analyte
causes a change in the gating behaviour of the ion channel, and this is measured via the electrical
response of the membrane capacitance. To support the lipid bilayer, there is used a monolayer of
alkane-thiol molecules on a gold electrode, which provides a scaffold for a lipid monolayer to self-
assemble onto. This monolayer can incorporate ion channels such as gramicidin which are used as the
sensing element of the device. Variations on this method have been used to create a tethered lipid
bilayer onto an electrode surface to incorporate other membrane proteins. However, the approach has
a number of drawbacks, the first is that the small aqueous volume present under the lipid bilayer,
typically of the order of lnm to 10nm thick, does not contain enough ions to perform a direct current
measurement for any useful period of time. This is an effect common to nearly all tethered bilayer
systems on solid supports. For recordings of any meaningful duration, an alternating current
measurement must be used to overcome the ionic depletion at the electrode, but that limits the
sensitivity of the device.
An example of a biosensor device using a supported lipid bilayer is disclosed in Urisu et al.,
"Formation of high-resistance supported lipid bilayer on the surface of a silicon substrate with
microelectrodes", Nanomedicine, 2005, (1), 317-322. This device exploits the strong surface adhesion
between phospholipid molecules and a SiO2 surface to form a supported bilayer. A silicon oxide
surface is modified, using etching techniques common in silicon chip production, to expose small
channels to an electrode surface. A bilayer is then formed on the silicon oxide surface, resulting in an
electrical resistance of a few MΩ. In this system, the wells created by this process could not be
individually addressed.
In both of the cited examples using a supported lipid bilayer, it is very difficult to form a high
resistive seal using these methods. Although the resistance may be sufficient to observe a change
arising from a large number of ion channels, single channel or stochastic measurements, which are
inherently more sensitive, are incredibly challenging using this methodology.

There are a number of problems with the supported bilayer approach in these documents and
in general, which makes this system unsuitable. The first problem lies with the resistance of the
bilayer membrane which is typically about 100MΩ While this may be suitable for examining protein
behaviour at large protein concentrations, it is not sufficient for a high-fidelity assay based on single
molecule sensing, typically requiring a resistance of at least 1GΩ and for some applications one or
two orders of magnitude higher. The second problem is the small volume of solution trapped in the
short distance between the bilayer and the solid support, typically of the order of lnm. This small
volume does not contain many ions, affecting the stability of the potential across the bilayer and
limiting the duration of the recording.
A number of methods have been proposed to overcome the problems with solid supported
bilayers. One option is to incorporate a chemical linkage between the bilayer and the surface, either a
small polyethylene glycol layer is introduced (polymer cushioned bilayers), or the lipid is chemically
modified to contain a small hydrophilic linkage and reacted with the surface providing a scaffold for
vesicle deposition (tethered bilayers). While these methods have increased the ionic reservoir beneath
the lipid bilayer, they are inconvenient to implement and have done little to decrease the current
leakage across the bilayer.
The techniques used in the silicon chip industry provide an attractive technology for creating
a large number of electrodes that could be used in biosensor applications. This approach is disclosed
in the related applications US-7,144,486 and US-7,169,272. US-7,144,486 discloses a method of
fabricating a microelectrode device containing microcavities etched into layers of insulator material.
The devices are said to have a wide range of electrochemical applications in which electrodes in the
cavities measure electrical signals. It is stated that thin films may be suspended across the cavities.
Several types of film are mentioned, including being a lipid bilayer. However this is merely a proposal
and there is no disclosure of any technique for forming the lipid bilayer, nor any experimental report
of this. Indeed the related application US-7,169,272, which does report experimental formation of
lipid bilayers in the same type of device, discloses the supported lipid bilayers being chemically
attached directly on the electrodes. This uses similar techniques to those presented in Osman et al.
cited above and suffers from the same drawbacks relating to the lack of a sufficiently high resistive
seal for stochastic measurements and the lack of an ionic reservoir for recording ionic flow across the
bilayer system.
To summarise, the known technologies summarised above either present methods of bilayer
formation which can not reproducibly achieve high resistance, or suffer from low ionic reservoirs and
are not capable of high duration direct current measurements, or require a separate fluidic chamber for
each array element, limiting the scale up of that device to a high-density array. It would be desirable
to reduce these problems.

According to a first aspect of the present invention, there is provided a method of forming a
layer separating two volumes of aqueous solution, the method comprising:
(a) providing an apparatus comprising elements defining a chamber, the elements including a
body of non-conductive material having formed therein at least one recess opening into the chamber,
the recess containing an electrode;
(b) applying a pre-treatment coating of a hydrophobic fluid to the body across the recess;
(c) flowing aqueous solution, having amphiphilic molecules added thereto, across the body to
cover the recess so that aqueous solution is introduced into the recess from the chamber and so that a
layer of the amphiphilic molecules forms across the recess separating a volume of aqueous solution
introduced into the recess from the remaining volume of aqueous solution.
Such a method allows the formation of layers of amphiphilic molecules which are of
sufficiently high quality for sensitive techniques such as stochastic sensing whilst using apparatus and
techniques which are straightforward to implement.
The apparatus used is relatively simple, involving most importantly a body of ionically non-
conductive material having formed therein at least one recess. It has been demonstrated, surprisingly,
that it is possible to form a layer of the amphiphilic molecules across such a recess simply by flowing
the aqueous solution across the body to cover the recess. To achieve this a pre-treatment coating of a
hydrophobic fluid is applied to the body across the recess. The pre-treatment coating assists formation
of the layer. The layer is formed without any need for a complicated apparatus involving two
chambers separated by a septum and requiring a complicated fluidics arrangement to achieve separate
filling. This is because the method does not require the recess to be pre-filled prior to introducing
aqueous solution into the chamber above. Instead, the aqueous solution is introduced into the recess
from the chamber. Despite this, it is still possible to form the layer by mere control of the aqueous
solution flowing into the chamber. Such flow control is a straightforward practical technique.
Importantly, the method allows the formation of layers of amphiphilic molecules which are
suitable for high sensitivity biosensor applications such as stochastic sensing and single channel
recording. It has been demonstrated possible to form layers of high resistance providing highly
resistive electrical seals, having an electrical resistance of 1GΩ or more, typically at least 100GΩ.
which, for example, enable high-fidelity stochastic recordings from single protein pores. This is
achieved whilst trapping a volume of aqueous solution in the recess between the layer and the
electrode. This maintains a significant supply of electrolyte. For example, the volume of aqueous
solution is sufficient to allow stable continuous dc current measurement through membrane proteins
inserted in the layer. This contrasts significantly with the known techniques described above using
supported lipid bilayers.
Furthermore, the simple construction of the apparatus allows the formation of a miniaturised

apparatus having an array of plural recesses and allowing the layer across each recess to be
electrically isolated and individually addressed using its own electrode, such that the miniaturised
array is equivalent to many individual sensors measuring in parallel from a test sample. The recesses
may be relatively densely packed, allowing a large number of layers to be used for a given volume of
test sample. Individual addressing may be achieved by providing separate contacts to each electrode
which is simple using modem microfabrication techniques, for example lithography.
Furthermore, the method allows the formation of multiple layers of the amphiphilic molecules
within a single apparatus across the plural recesses in an array using a very straightforward technique.
In most applications, one or more membrane proteins is subsequently inserted into the layer.
Certain membrane proteins that can be used in accordance with the invention are discussed in more
detail below.
According to further aspects of the invention, there is provided an apparatus suitable for
implementing such methods of formation of a layer of amphiphilic molecules.
Further details and preferred features of the invention will now be described.
The amphiphilic molecules are typically a lipid. In this case the layer is a bilayer formed from
two opposing monolayers of lipid. The lipids can comprise one or more lipids. The lipid bilayer can
also contain additives that affect the properties of the bilayer. Certain lipids and other amphiphilic
molecules, and additives that can be used in accordance with the invention are discussed in more
detail below.
Various techniques may be applied to add the amphiphilic molecules to the aqueous solution.
A first technique is simply to add the amphiphilic molecules to the aqueous solution outside
the apparatus before introducing the aqueous solution into the chamber.
A second technique which has particular advantage is, before introducing the aqueous
solution into the chamber, to deposit the amphiphilic molecules on an internal surface of the chamber,
or elsewhere in the flow path of the aqueous solution, for example in a fluidic inlet pipe connected to
the inlet. In this case, the aqueous solution covers the internal surface during step (c) whereby the
amphiphilic molecules are added to the aqueous solution. In this manner the aqueous solution is used
to collect the amphiphilic molecules from the internal surface. Such deposition of the amphiphilic
molecules has several advantages. It allows the formation of layer of amphiphilic molecules in the
absence of large amounts of organic solvent, as would typically be present if the amphiphilic
molecules were added directly to the aqueous solution. This means that it is not necessary to wait for
evaporation of the organic solvent before the layer can be formed. In addition, this means that the
apparatus is not required to be made from materials that are insensitive to organic solvents. For
instance, organic-based adhesives can be used and screen-printed conductive silver/silver chloride
paste can be used to construct electrodes.

Advantageously, the deposited amphiphilic molecules can be dried. In this case, the aqueous
solution is used to rehydrate the amphiphilic molecules . This allows the amphiphilic molecules to be
stably stored in the apparatus before use. It also avoids the need for wet storage of amphiphilic
molecules. Such dry storage of amphiphilic molecules increases shelf life of the apparatus.
Several techniques may be used to insert a membrane protein into the layer of amphiphilic
molecules.
A first technique is simply for the aqueous solution to have a membrane protein added
thereto, whereby the membrane protein is inserted spontaneously into the layer of amphiphilic
molecules. The membrane protein may be added to the aqueous solution outside the apparatus before
introducing the aqueous solution into the chamber. Alternatively the membrane protein may be
deposited on an internal surface of the chamber before introducing the aqueous solution into the
chamber. In this case, the aqueous solution covers the internal surface during step (c), whereby the
membrane protein is added to the aqueous solution.
A second technique is for the aqueous solution to have vesicles containing the membrane
protein added thereto, whereby the membrane protein is inserted on fusion of the vesicles with the
layer of amphiphilic molecules.
A third technique is to insert the membrane protein by carrying the membrane protein to the
layer on a probe, for example an agar-tipped rod.
To form the layer of amphiphilic molecules, the aqueous solution is flowed across the body to
cover the recess. Formation is improved if a multi-pass technique is applied in which aqueous solution
covers and uncovers the recess at least once before covering the recess for a final time. This is thought
to be because at least some aqueous solution is left in the recess which assists formation of the layer
in a subsequent pass.
The pre-treatment coating is a hydrophobic fluid which assists formation of the layer by
increasing the affinity of the amphiphilic molecules to the surface of the body around the recess. In
general any pre-treatment that modifies the surface of the surfaces surrounding the aperture to
increase its affinity to lipids may be used. Certain materials for the pre-treatment coating that can be
used in accordance with the invention are discussed in more detail below.
To assist in the spreading of the pre-treatment coating, surfaces including either or preferably
both of (a) the outermost surface of the body around the recess and (b) at least an outer part of the
internal surface of the recess extending from the rim of the recess may be hydrophobic. This may be
achieved by making the body with an outermost layer formed of a hydrophobic material.
Another way to achieve this is for the surfaces to be treated by a fluorine species, such as a
fluorine radical, for example by treatment with a fluorine plasma during manufacture of the apparatus.
The application of the pre-treatment coating may leave excess hydrophobic fluid covering

said electrode contained in the recess. This potentially insulates the electrode by reducing ionic flow,
thereby reducing the sensitivity of the apparatus in measuring electrical signals. However various
different techniques may be applied to minimise this problem.
A first technique is to apply a voltage across the electrode in the recess and a further electrode
in the chamber sufficient to reduce the amount of excess hydrophobic fluid covering said electrode
contained in the recess. This produces a similar effect to electro-wetting. The voltage is applied after
flowing aqueous solution across the body to cover the recess so that aqueous solution flows into the
recess. As the voltage will rupture any layer formed across the recess, subsequently the aqueous
solution is flowed to uncover the recess, and then aqueous solution, having amphiphilic molecules
added thereto, is flowed across the body to re-cover the recess so that a layer of the amphiphilic
molecules forms across the recess.
A second technique is to make an inner part of the internal surface of the recess hydrophilic.
Typically this will be applied in combination with making the outer part of the internal surface of the
recess hydrophobic. This may be achieved by making the body with an inner layer formed of a
hydrophilic material and an outermost layer formed of a hydrophobic material.
A third technique is to provide on the electrode a hydrophillic surface, for example a
protective material, which repels the hydrophobic fluid applied in step (c) whilst allowing ionic
conduction from the aqueous solution to the electrode. The protective material may be a conductive
polymer, for example polypyrrole/ polystyrene sulfonate. Alternatively, the protective material may
be a covalently attached hydrophilic species, such as thiol-PEG.
In general, a wide range of constructional features may be employed in the apparatus to form
the body of non-conductive material, the at least one recess formed therein and the other elements
defining the chamber. Examples are described in further detail below.
According to a second aspect of the present invention, there is provided a method of
improving the performance of an electrode in a recess in conducting electro-physiological
measurements, the method comprising depositing a conductive polymer on the electrode.
Further according to a second aspect of the present invention, there is provided an apparatus
for conducting electro-physiological measurements, the apparatus comprising, a body having a recess
in which an electrode is located, wherein a conductive polymer is provided on the electrode.
It has been discovered that the providing a conductive polymer on an electrode in a recess can
improve the performance of the electrode in conducting electro-physiological measurements. One
advantage is to improve the electrode's performance as a stable electrode for conducting electro-
physiological measurements. A further advantage is to increase the charge reservoir available to the
electrode within the recess without increasing the volume of aqueous solution contained in the recess.
To allow better understanding, an embodiment of the present invention will now be described

by way of non-limitative example with reference to the accompanying drawings, in which:
Fig. 1 is a perspective view of an apparatus;
Fig. 2 is a cross-sectional view of the apparatus of Fig. 1, taken along line II-II in Fig. 1, and
showing the introduction of an aqueous solution;
Fig. 3 is a cross-sectional view of the apparatus, similar to that of Fig. 2 but showing the
apparatus full of aqueous solution;
Fig. 4 is sequence of a cross-sectional, partial views of the recess in the apparatus over an
electrochemical electrode modification process;
Figs. 5 is an SEM image of a recess formed by CO2 laser drilling;
Fig. 6 is an OM image of a recess formed using photolithography;
Figs. 7a and 7b are 3D and 2D LP profiles, respectively, of a recess formed using
photolithography;
Figs. 8a and 8b are 3D and 2D LP profiles, respectively, of a recess formed using
photolithography, after electoplating;
Fig. 9 is a cross-sectional, partial view of the recess in the apparatus with a pre-treatment
coating applied;
Figs. 10a to lOe are a sequence of cross-sectional, partial view of the recess in the apparatus
during a method of removing excess pre-treatment coating;
Fig. 11 is a cross-sectional, partial view of the recess in the apparatus having plural further
layers in the body;
Fig. 12 is a diagram of an electrical circuit;
Fig. 13 is a perspective view of the apparatus and electrical circuit mounted on a printed
circuit board;
Fig. 14 is a diagram of an electrical circuit for acquiring plural signals in parallel;
Fig. 15 is a graph of the applied potential and current response for a dry apparatus;
Fig. 16 is a graph of the applied potential and current response for a wet apparatus;
Fig. 17 is a graph of the applied potential and current response on electro-wetting of the
apparatus;
Fig. 18 is a graph of the applied potential and current response on formation of a layer of
amphiphilic molecules;
Figs. 19 to 22 are graphs of the applied potential and current response for various different
apparatuses;
Figs. 23 to 25 are plan views of a further layer in a modified apparatus having plural recesses;
Figs. 26 to 28 are plan views of the substrate in the modified apparatuses having plural
recesses;

Figs. 29 and 30 are graphs of the current response for two different apparatuses having plural
recesses;
Fig. 31 is a cross-sectional view of a portion of a modified apparatus;
Fig. 32 is a cross-sectional view of another modified apparatus;
Fig. 33 is a flow chart of a method of manufacture of the apparatus;
Figs. 34a and 34b are 3D and 2D surface profiles of a recess having an electrode modified by
electropolymerisation of polypyrrole, measured by profilometry;
Fig. 35 is a graph of current recorded on an array of recesses having an electrode modified by
electropolymerisation of polypyrrole.
An apparatus 1 which may be used to form a layer of amphiphilic molecules is shown in Fig.
1.
The apparatus 1 includes a body 2 having layered construction as shown in Figs. 2 and 3
comprising a substrate 3 of non-conductive material supporting a further layer 4 also of non-
conductive material. In the general case, there may be plural further layers 4, as described further
below.
A recess 5 is formed in the further layer 4, in particular as an aperture which extends through
the further layer 4 to the substrate 3. In the general case, there may be plural recesses 5, as described
further below.
The apparatus 1 further includes a cover 6 which extends over the body 2. The cover 6 is
hollow and defines a chamber 7 which is closed except for an inlet 8 and an outlet 9 each formed by
openings through the cover 6. The lowermost wall of the chamber 7 is formed by the further layer 4 in
Fig. 2, but as an alternative the further layer 4 could be shaped to provide side walls.
As described further below, in use aqueous solution 10 is introduced into the chamber 7 and a
layer 11 of amphiphilic molecules is formed across the recess 5 separating aqueous solution 10 in the
recess 5 from the remaining volume of aqueous solution in the chamber 7. The apparatus includes the
following electrode arrangement to allow measurement of electrical signals across the layer 11 of
amphiphilic molecules.
Use of a chamber 7 which is closed makes it very easy to flow aqueous solution 10 into and
out of the chamber 7. This is done simply by flowing the aqueous solution 10 through the inlet 8 as
shown in Fig. 2 until the chamber 7 is full as shown in Fig. 3. During this process, gas (typically air)
in the chamber 7 is displaced by the aqueous solution 10 and vented through the outlet 9. For
example, a simple fluidics system attached to the inlet 8 may be used. This may be as simple as a
plunger, although more complicated systems may be used to improve the control. However, the
chamber 7 is not necessarily closed and may be open, for example by forming the body 2 as a cup.
The substrate 3 has a first conductive layer 20 deposited on the upper surface of the substrate

3 and extending under the further layer 4 to the recess 5. The portion of the first conductive layer 20
underneath the recess 5 constitutes an electrode 21 which also forms the lowermost surface of the
recess 5. The first conductive layer 20 extends outside the further layer 4 so that a portion of the first
conductive layer 20 is exposed and constitutes a contact 22.
The further layer 4 has a second conductive layer 23 deposited thereon and extending under
the cover 6 into the chamber 7, the portion of the second conductive layer 23 inside the chamber 7
constituting an electrode 24. The second conductive layer 23 extends outside the cover 6 so that a
portion of the second conductive layer 23 is exposed and constitutes a contact 25.
The electrodes 21 and 24 make electrical contact with aqueous solution in the recess 5 and
chamber 7. This allows measurement of electrical signals across the layer 11 of amphiphilic
molecules by connection of an electrical circuit 26 to the contacts 22 and 25. The electrical circuit 26
may have basically the same construction as a conventional circuit for performing stochastic sensing
across a lipid bilayer formed in a conventional cell by the Montal & Mueller method.
An example design of the electrical circuit 26 is shown in Fig. 12. The primary function of the
electrical circuit 26 is to measure the electrical current signal developed between the electrodes 21
and 24 to provide a meaningful output to the user. This may be simply an output of the measured
signal, but in principle could also involve further analysis of the signal. The electrical circuit 26 needs
to be sufficiently sensitive to detect and analyse currents which are typically very low. By way of
example, an open membrane protein might typically pass current of lOOpA to 200pA with a 1M salt
solution.
In this implementation, the electrode 24 in the chamber 7 is used as a reference electrode and
the electrode 21 in the recess 5 is used as a working electrode. Thus the electrical circuit 26 provides
the electrode 24 with a bias voltage potential relative to the electrode 21 which is itself at virtual
ground potential and supplies the current signal to the electrical circuit 26.
The electrical circuit 26 has a bias circuit 40 connected to the electrode 24 in the chamber 7
and arranged to apply a bias voltage which effectively appears across the two electrodes 21 and 24.
The electrical circuit 26 also has an amplifier circuit 41 connected to the electrode 21 in the
recess 5 for amplifying the electrical current signal appearing across the two electrodes 21 and 24.
Typically, the amplifier circuit 41 consists of a two amplifier stages 42 and 43.
The input amplifier stage 42 connected to the electrode 21 converts the current signal into a
voltage signal.
The input amplifier stage 42 may comprise transimpedance amplifier, such as an electrometer
operational amplifier configured as an inverting amplifier with a high impedance feedback resistor, of
for example 500MΩ, to provides the gain necessary to amplify the current signal which typically has a
magnitude of the order of tens to hundreds of picoamps.

Alternatively, the input amplifier stage 42 may comprise a switched integrator amplifier. This
is preferred for very small signals as the feedback element is a capacitor and virtually noiseless. In
addition, a switched integrator amplifier has wider bandwidth capability. However, the integrator does
have a dead time due to the necessity to reset the integrator before output saturation occurs. This dead
time may be reduced to around a microsecond so is not of much consequence if the sampling rate
required is much higher. A transimpedance amplifier is simpler if the bandwidth required is smaller.
Generally, the switched integrator amplifier output is sampled at the end of each sampling period
followed by a reset pulse. Additional techniques can be used to sample the start of integration
eliminating small errors in the system.
The second amplifier stage 43 amplifies and filters the voltage signal output by the first
amplifier stage 42 . The second amplifier stage 43 provides sufficient gain to raise the signal to a
sufficient level for processing in a data acquisition unit 44. For example with a 500MΩ feedback
resistance in the first amplifier stage 42, the input voltage to the second amplifier stage 43, given a
typical current signal of the order of 100pA, will be of the order of 50mV, and in this case the second
amplifier stage 43 must provide a gain of 50 to raise the 50mV signal range to 2.5V.
The electrical circuit 26 includes a data acquisition unit 44 which may be a microprocessor
running an appropriate program or may include dedicated hardware. The data acquisition unit 44 may
be a card to be plugged into a computer 45 such as a desktop or laptop. In this case, the bias circuit 40
is simply formed by an inverting amplifier supplied with a signal from a digital-to-analog converter 46
which may be either a dedicated device or a part of the data acquisition unit 44 and which provides a
voltage output dependent on the code loaded into the data acquisition unit 44 from software.
Similarly, the signals from the amplifier circuit 41 are supplied to the data acquisition card 40 through
an analog-to-digital converter 47.
The various components of the electrical circuit 26 may be formed by separate components or
any of the components may be integrated into a common semiconductor chip. The components of the
electrical circuit 26 may be formed by components arranged on a printed circuit board. An example of
this is shown in Fig. 13 wherein the apparatus 1 is bonded to a printed circuit board 50 with
aluminium wires 51 connecting from the contacts 22 and 25 to tracks 52 on the printed circuit board.
A chip 53 incorporating the electrical circuit 26 is also bonded to the printed circuit board 50.
Alternatively the apparatus 1 and the electrical circuit 26 may be mounted on separate printed circuit
boards.
In the case that the apparatus 1 contains plural recesses 5, each having a respective electrode
21, then the electrical circuit 26 is modified essentially by replicating the amplifier circuit 41 and A/D
converter 47 for each electrode 21 to allow acquisition of signals from each recess 5 in parallel. In the
case that the input amplifier stage 42 comprises switched integrators then those would require a

digital control system to handle the sample-and-hold signal and reset integrator signals. The digital
control system is most conveniently configured on a field-programmable-gate-array device (FPGA).
In addition the FPGA can incorporate processor-like functions and logic required to interface with
standard communication protocols i.e. USB and Ethernet.
Fig. 14 shows a possible architecture of the electrical circuit 26 and is arranged as follows.
The respective electrodes 21 of the apparatus 1 are connected to the electrical circuit 26 by an
interconnection 55, for example the aluminium wires 51 and the printed circuit board in the
arrangement of Fig. 13. In the electrical circuit 26, the amplifier circuits 41 may be formed in one or
more amplifier chips 56 having plural channels. The signals from different electrodes 21 may be on
separate channels or multiplexed together on the same channel. The outputs of the one or more
amplifier chips 56 are supplied via the A/D converter 47 to a programmable logic device 57 for
receiving the signal on each channel. For example to handle signals from an apparatus having 1024
recesses, the programmable logic device 57 might operate at a speed of the order of 10Mbits/s. The
programmable logic device 57 is connected via an interface 58, for example a USB interface, to a
computer 59 to supply the signals to the computer 59 for storage, display and further analysis.
During use the apparatus 1 may be enclosed in a Faraday cage to reduce interference.
Various materials for the components of the apparatus 1 will now be discussed. The materials
for each component of apparatus 1 are determined by the properties required to enable the component
to function correctly during operation, but the cost and manufacturing throughput are also considered.
All materials should be chosen to provide sufficient mechanical strength to allow robust handling, and
surfaces compatible with bonding to the subsequent layers.
The material of the substrate 3 is chosen to provide a rigid support for the remainder of the
apparatus 1. The material is also chosen to provide a high resistance and low capacitance electrical
insulation between adjacent electrodes 21 when there are plural recesses 5. Possible materials include
without limitation: polyester (eg Mylar), or another polymer; or silicon, silicon nitride, or silicon
oxide. For example, the substrate may comprise a silicon wafer with a thermally grown oxide surface
layer.
The material of the further layer 4 (or in the general case layers) are chosen to provide
a high resistance and low capacitance electrical insulation between the electrodes 21 and 24 and also,
when there are plural recesses 5, between the electrodes 21 and 24 of adjacent recesses 5. Also the
surface of the further layer 4 should be chemically stable both to the pre-treatment coating applied
before operation (as discussed below) and to the aqueous solution 10. Lastly, the further layer 4
should be mechanically robust in order to maintain its structural integrity and coverage of the first
conductive layer 20, and should be suitable for subsequent attachment of the cover 6.
The following is a list of possible materials for the further layer 4, together with thicknesses

which have been successfully employed experimentally, although these thicknesses are not limitative:
photoresist (eg SU8 photoresist or Cyclotene) with a variety of thicknesses; polycarbonate, 6um thick
film; PVC, 7µm thick film; polyester, 50µm thick film; adhesive backed polyester, 25um and 50um
thick film; thermal laminating films, eg Magicard 15um thick and Murodigital 35um; or a screen-
printed dielectric ink.
Advantageously, surfaces including (a) the outermost surface of the body 2 around the recess
and (b) the outer part of the internal surface of the recess 5 extending from the rim of the recess 5 are
hydrophobic. This assists in the spreading of the pre-treatment coating and therefore also formation of
a lipid bilayer. One particular way to achieve this is to modify these surfaces by a fluorine species.
Such a fluorine species is any substance capable of modifying the surfaces to provide a fluorine-
containing layer. The fluorine species is preferably one containing fluorine radicals. For example the
modification may be achieved by treating the body 2 with a fluorine plasma, for example a CF4 during
manufacture.
The conductive layers 20 and 23 will now be discussed further.
The material of the electrodes 21 and 24 should provide an electrochemical electrode in
contact with the aqueous solution 10, enabling measurement of low currents, and should be stable to
the pre-treatment coating and aqueous solution 10. The material of the remainder of the conductive
layers 20 and 23 (usually but not necessarily the same as the electrodes 21 and 24) also provides
electrical conductance from the electrodes to the contacts 22 and 25. The first conductive layers 20
will also accept bonding of the further layers 4. The conductive layers 20 and 23 can be constructed
with plural overlapping layers and/or an appropriate surface treatment. One possible material is
platinum, coated with silver at the area exposed to the test solution and then silver chloride formed on
top of the silver. Possible materials for the first conductive layer 20 include without limitation:
Silver/silver chloride electrode ink; silver with or without a surface layer, for example of silver
chloride formed by chloridisation or of silver fluoride formed by fluoridisation; gold with or without
redox couple in solution; platinum with or without redox couple in solution; ITO with and without
redox couple in solution; gold electrochemically coated with conductive polymer electrolyte; or
platinum electrochemically coated with conductive polymer electrolyte. Possible materials for the
second conductive layer 23 include without limitation: silver/silver chloride electrode ink; silver wire;
or chloridised silver wire.
Some specific examples of include: the substrate 3 being silicon and the conductive layer 20
being a metal conductor (diffusion or polysilicon wires are poor methods) buried in a silicon oxide
insulating layer (e.g. using typical semiconductor fabrication technology); the substrate 3 being glass
and the conductive layer 20 being metal conductors (e.g using typical LCD display technology); or the
substrate 3 being a polymeric substrates and the conductive layer 20 being an ablated metal or printed

conductor (e.g. using typical glucose biosensor technology).
The requirements for the material of the cover 6 are to be easily attached to create a seal for
the chamber 7, to be compatible with both the pre-treatment coating and the aqueous solution 10. The
following are possible materials, together with thicknesses which have been successfully employed
experimentally, although these thicknesses are not limitative: silicone rubber, 0.5,1.0, 2.0mm thick;
polyester, 0.5mm thick; or PMMA (acrylic) 0.5mm to 2mm thick.
Various methods of manufacturing the apparatus 1 will now be discussed. In general terms,
the layered construction of the apparatus 1 is simple and easy to form by a variety of methods. Three
different fabrication technologies which have actually been applied are: lamination of polymer films;
printed circuit board manufacture with high resolution solder mask formation and photolithography
using silicon wafers or glass.
An example of a lamination process is as follows.
The substrate 3 is a 250um thick polyester sheet (Mylar), and the first conductive layer 20 is
deposited by either: screen printing silver/silver chloride electrode ink; adhesion of metal foil; or
vapour deposition (sputtering or evaporation). The further layer 4 is then laminated onto the substrate
3 by either: a pressure-sensitive adhesive; a thermally activated adhesive; or using the wet silver/silver
chloride ink as the adhesive painted directly onto the dielectric before lamination (referred to as
"painted electrodes")- The aperture in the further layer 4 that forms the recess 5 is created with 5-
100µm diameter either before or after lamination to the substrate 3 by either: electrical discharge
(sparking); or laser drilling, for example by an excimer, solid state or CO2 laser. An apparatus created
by lamination of polymer films sometimes requires an additional sparking step to activate the
electrodes prior to use. The second conductive layer 23 is formed on top of the further layer 4 by
screen printing. The cover 6 is laminated on top using pressure sensitive adhesive.
An example of a process employing photolithography using silicon wafers is as follows.
The substrate 3 is a silicon wafer with an oxide surface layer. The first conductive layer 20 is
formed by gold, silver, chloridised silver, platinum or ITO deposited onto the substrate 3. Photoresist
(eg SU8) is then spin-coated over the substrate 3 to form the further layer 4. The recess 5 is formed
with 5-100um diameter by removal of the photoresist following UV exposure using a mask to define
the shape of the recess 5. The second conductive layer 23 is formed on top of the further layer 4, for
example by screen printing. The cover 6 is laminated on top using pressure sensitive adhesive.
The ability to use this type of process is significant because it allows the apparatus to be
formed on silicon chips using standard silicon wafer processing technology and materials.
The electrodes 21 and 24 will now be discussed further.
For stable and reliable operation, the electrodes 21 and 24 should operate at the required low
current levels with a low over-potential and maintain their electrode potential value over the course of

the measurement. Further, the electrodes 21 and 24 should introduce a minimum amount of noise into
the current signal. Possible materials for the electrodes 21 and 24 include without limitation:
Silver/silver chloride electrode ink; silver with or without a surface layer, for example of silver
chloride formed by chloridisation or of silver fluoride formed by fluoridisation; gold with or without
redox couple in solution; platinum with or without redox couple in solution; ITO with and without
redox couple in solution; palladium hydride, gold electrochemically coated with conductive polymer
electrolyte; or platinum electrochemically coated with conductive polymer electrolyte.
Silver is a good choice for the material of electrodes 21 and 24 but is difficult to incorporate
in a silicon wafer manufacturing process due to its tendency to undergo oxidation on exposure to
light, air and high temperatures. To avoid this problem it is possible to manufacture the apparatus
with an inert conductive material (eg Pt or Au) in the recess, and then change the surface type or
properties of the inert conductive material using methods including but not limited to electroplating,
electropolymerisation, electroless plating, plasma modification, chemical reaction, and other coating
methods known in the art.
Electroplating of silver may be achieved, for example, using a modified version of the method
of Polk et al., "Ag/AgCl microelectrodes with improved stability for microfluidics", Sensors and
Actuators B 114 (2006) 239-247. A plating solution is prepared by addition of 0.41g of silver nitrate
to 20ml of 1M ammonium hydroxide solution. This is rapidly shaken to avoid precipitation of the
insoluble silver oxide, and to facilitate the formation of the diammine silver complex. The solution is
always fresh to avoid fall in plating efficiency. The plating is performed using conventional
equipment, connecting the electrode 21 as the cathode and using a platinum electrode is used as the
anode. For example in the case of plating on Pt electrodes, a potential of-0.58V is applied to the
cathode, with the anode being held at ground potential, whereas in the case of plating on Au
electrodes, the potential is held at -0.48V with respect to ground. A target charge of 5.1x103C/m2 has
been found empirically to result in a silver deposition of between 1µm and 2µm for a 100µm diameter
eletrode, typically taking of the order of 60s.
In performing such plating it is desirable to achieve uniform penetration of the aqueous
plating solution to the bottom of the recess 5. In the case that the layer 4 is formed from a naturally
hydrophobic material (eg SU8 photoresist) and in order to ensure uniform wetting of the recess,
desirably the degree of hydrophilicity can be increased. Three methods to achieve this are as follows.
A first method is application of a lipid to the surface of the layer 4, so that the lipid acts as a
surfactant, facilitating the entry of the plating solution. A second method is exposure of the layer 4 to
oxygen plasma which activates the material of the layer and produces hydrophilic functional groups.
This produces a well defined hydrophilic and clean surface. A third method is to add ethanol to the
plating solution.

Where the electrode 21 is made of silver (or indeed other metals), the outer surface of the
electrode is desirably converted to a halide, in order for the electrode 21 to function efficiently as a
provider of a stable reference potential. In common usage, the halide used is chloride, since the
conversion of silver to silver chloride is relatively straightforward to achieve, for example by
electrolysis in a solution of hydrochloric acid. Alternative chemical methods avoiding the use of a
potentially corrosive acid which may affect the surface condition of the layer 4 include a) sweeping
voltammetry in 3M sodium chloride solution, and b) a chemical etching by immersion of the electrode
21 in 50mM ferric chloride solution.
An alternative halogen for the halidisation is fluorine. The choice of fluorine has the
significant advantage that the silver fluoride layer can be formed in the same step as modification of
surfaces (a) and (b) of the body 2 to make them hydrophobic, as discussed above. For example this
may be achieved during manufacture of the apparatus 1 by treatment of the body 2 by a fluorine
plasma for example a CF4 plasma. This is effective to modify the surfaces of the body 2, particularly
in the case that the layer 4 is a photoresist such as SU8 to achieve a sufficient degree of
hydrophobicity to support the formation of a stable lipid bilayer. At the same time exposure to the
fluorine plasma converts the metal of the electrode 21 into an outer layer of metal fluoride.
There will now be discussed some possible adaptations of the electrode 21 in the recess 5 as
alternatives to the use of a fluorine plasma as discussed above.
The electrode 21 may be electrochemically modified to change the surface-type. This allows
use of additional materials with good bulk properties but poor surface properties, such as gold.
Possible electrochemical surface modifications include without limitation: silver electroplating;
electrochemical chloridisation of silver; electropolymerisation of a polymer/ polyelectrolyte.
By way of further example, one possible sequence of modification is shown in Fig. 4 in which
a coating 37 of silver is formed on the electrode 21 formed of gold or platinum by electrochemical
deposition. Electroplating may typically be performed in 0.2M AgNO2, 2M KI, 0.5mM Na2S2O3 at
-0.48V using a standard single liquid junction Ag/AgCl reference electrode and a platinum counter
electrode. A typical thickness of the coating 37 is estimated to be 750nm with a deposition time of
about 50s and about 50µC charge passed. Subsequently a chloridised layer 38 is formed by
chloridisation, typically at +150mV in 0.1M HC1 for 30s.
Another possible surface modification is to apply a conductive polymer. The conductive
polymer may be any polymer which is conductive. A suitable conductive polymer will have mobile
charge carriers. Typically such a conductive polymer will have a backbone having delocalised
electrons which are capable of acting as charge carriers, allowing the polymer to conduct. The
conductive polymer may be doped to increase its conductivity, for example by a redox process or by
electrochemical doping. Suitable conductive polymers include, without limitation: polypyrroles,

polyacetylenes, polythiophenes, polyphenylenes, polyanilines, polyfluorenes,
poly(3-alkylthiophene)s, polytetrathiafulvalenes, polynaphthalenes, poly(p-phenylene sulfide)s,
polyindoles, polythionines, polyethylenedioxythiophenes, and poly(para-phenylene vinylene)s.
One possible conductive polymer is a polypyrrole, which may be doped, for example with
polystyrene sulfonate. This may be deposited, for example, on an electrode 21 of gold by
electrooxidizing an aqueous solution of 0.1M pyrrole + 90mM polystyrene sulfonate in 0.1M KC1 at
+0.80V vs. Ag/AgCl reference electrode. The estimated thickness of polymer deposited is 1µm at
30uC, based on a assumption that 40mC/cm2 of charge produces a film of thickness around 0.1 µm.
The polymerization process can be represented as follows, where PE stands for polystyrene sulfonate:

One advantage of using a conductive polymer deposited on an inert electrode, such as
polypyrrole doped with polystyrene, electropolymerised onto gold or platinum, is to improve the
electrode's performance as a stable electrode for conducting electrophysiological measurements. A
further advantage is to increase the charge reservoir available to the electrode within the recess
without increasing the volume of aqueous solution contained in the recess. These advantages are
generally applicable when conducting electrophysiological measurements using an electrode in a
recess, such as the electrode 21 in the apparatus 1.
Other advantages of using a conductive polymer on the electrode 21 in the recess 5 of the
apparatus 1 include but are not limited to control of the hydrophilic nature of the electrode surface to
aid wetting of the electrode surface by the aqueous buffer solution and similarly prevention of
blocking of the electrode by the chemical pre-treatment prior to bilayer formation.
Figs. 34a and 34b are 3D and 2D surface profiles of an example electrode modified by
electropolymerisation of polypyrrole, measured by profilometry. The thickness of electrochemically
deposited polymer film in this example is about 2µm. Fig. 35 shows the current recorded on an array
of recesses modified by electropolymerisation of polypyrrole, showing stable lipid bilayers and single
molecule detection of cyclodextrin from inserted protein pores.
In all embodiments, an alternative to the second conductive layer 23 is to form an electrode in
the chamber 7 simply by insertion through the cover 6 of a conductive member, such as a chloridised
silver wire.
In order to characterise the electrodes 21, visualisation of recesses 5 formed in a body 2 has
been conducted using optical microscopy (OM), scanning electron microscopy (SEM), and laser
profilometry (LP).
Figs. 5 shows an SEM image of a recess 5 formed by drilling with a CO2 laser in an apparatus

1 formed by lamination of polymer layers, with subsequent application of electrical discharge to
activate the electrode 21. The image illustrates that the geometry of the recess 5 is poorly defined
using this method of formation, with considerable surface damage therearound and variability in
diameter, although it is hoped this may be improved through optimisation of the laser characteristics.
Fig. 6 shows an OM image of a recess 5 formed using photolithography of a further layer 4 of
SU8 photoresist over an electrode 21 of vapour deposited gold on a substrate 3 of silicon. Similarly,
Figs. 7a and 7b are 3D and 2D LP profiles of a similarly manufactured recess 5. Figs. 8a and 8b are
3D and 2D LP profiles of the same recess 5 after electroplating to form a coating 38 of silver. These
images show that photolithography provides a high degree of control of the geometry and dimensions
of the recess.
Excimer laser methods also produce a controlled geometry similar to photolithography.
There will now be described an example of a method of manufacture of the apparatus 1, as
shown in Fig. 33. The rationale of this method is to provide high throughput manufacture. This is
achieved by processing a wafer of silicon which forms the substrate 3 of plural apparatuses 1 and
which is subsequently diced. The wafer is prepared with an insulating layer, for example a thermally
grown silicon-oxide.
First the wafer is prepared. In step S1, the wafer is cleaned. In step S2, the wafer is subjected
to a HF dif to improve adhesion of metals and resist. Typical conditions are a 3 minute dip in 10:1
buffered oxide etch. In S3, the wafer is subjected to a bake as a dehydration step. Typical conditions
are baking for 1 hour at 200°C in an oven.
Next, the wafer is metallised to provide the first conductive layer 20 of each apparatus 1. In
step S4, photoresist is spun onto the wafer which is then subjected to UV light to form the desired
pattern. In step S5, the conductive layers 20 are deposited, for example consisting of successive layers
of Cr and Au. Typically of respective thicknesses 50nm and 300nm. In step S6 the resist is removed
for example by soaking in acetone.
Next, the layers 4 and recesses 5 are formed. In step S7, photoresist adhesion is improved by
the application of an O2 plasma and dehydration bake for example in an oven. In step S8, the wafer
has applied thereto photoresist which is then subjected to UV exposure to form the layers 4 and
recesses, for example SU8-10 with a thickness of 20m. In step S9 an inspection and measurement of
the recesses is performed.
Next, the electrodes 21 are plated. In step S10, the surface is prepared for plating by
performing an 02 plasma descum. In step S11, silver plating of the electrode is performed, as
described above, for example to form a plating thickness of 1.5µm.
In step S12, the wafer is diced to form the bodies 2 of separate apparatuses 1.
Lastly, the bodies 2 are treated by a CF4 plasma which modifies the surfaces of the body 2 and

the electrode 21 as discussed above. A typical exposure is for 12 minutes at 70W and 160mTorr.
In practice with an apparatus 1 manufactured using this method, the results of bilayer
formation and pore current stability have been comparable to those achieved with bodies plated and
chloridised by wet chemical means.
The method of using the apparatus 1 to form a layer 11 of amphiphilic molecules will now be
described. First the nature of the amphiphilic molecules that may be used will be considered.
The amphiphilic molecules are typically a lipid. In this case, the layer is a bilayer formed
from two opposing monolayers of lipid. The two monolayers of lipids are arranged such that their
hydrophobic tail groups face towards each other to form a hydrophobic interior. The hydrophilic head
groups of the lipids face outwards towards the aqueous environment on each side of the bilayer. The
bilayer may be present in a number of lipid phases including, but not limited to, the liquid disordered
phase (fluid lamellar), liquid ordered phase, solid ordered phase (lamellar gel phase, interdigitated gel
phase) and planar bilayer crystals (lamellar sub-gel phase, lamellar crystalline phase).
Any lipids that form a lipid bilayer may be used. The lipids are chosen such that a lipid
bilayer having the required properties, such as surface charge, ability to support membrane proteins,
packing density or mechanical properties, is formed. The lipids can comprise one or more different
lipids. For instance, the lipids can contain up to 100 lipids. The lipids preferably contain 1 to 10
lipids. The lipids may comprise naturally-occurring lipids and/or artificial lipids.
The lipids typically comprise a head group, an interfacial moiety and two hydrophobic tail
groups which may be the same or different. Suitable head groups include, but are not limited to,
neutral head groups, such as diacylglycerides (DG) and ceramides (CM); zwitterionic head groups,
such as phosphatidylcholine (PC), phosphatidylethanolamine (PE) and sphingomyelin (SM);
negatively charged head groups, such as phosphatidylglycerol (PG); phosphatidylserine (PS),
phosphatidylinositol (PI), phosphatic acid (PA) and cardiolipin (CA); and positively charged
headgroups, such as trimethylammonium-Propane (TAP). Suitable interfacial moieties include, but
are not limited to, naturally-occurring interfacial moieties, such as glycerol-based or ceramide-based
moieties. Suitable hydrophobic tail groups include, but are not limited to, saturated hydrocarbon
chains, such as lauric acid (n-Dodecanolic acid), myristic acid (n-Tetradecononic acid), palmitic acid
(n-Hexadecanoic acid), stearic acid (n-Octadecanoic) and arachidic (n-Eicosanoic); unsaturated
hydrocarbon chains, such as oleic acid (cis-9-Octadecanoic); and branched hydrocarbon chains, such
as phytanoyl. The length of the chain and the position and number of the double bonds in the
unsaturated hydrocarbon chains can vary. The length of the chains and the position and number of the
branches, such as methyl groups, in the branched hydrocarbon chains can vary. The hydrophobic tail
groups can be linked to the interfacial moiety as an ether or an ester.
The lipids can also be chemically-modified. The head group or the tail group of the lipids may

be chemically-modified. Suitable lipids whose head groups have been chemically-modified include,
but are not limited to, PEG-modified lipids, such as l,2-Diacyl-sn-Glycero-3-Phosphoethanolamine-N
-[Methoxy(Polyethylene glycol)-2000]; functionionalised PEG Lipids, such as 1,2-Distearoyl-sn-
Glycero-3 Phosphoethanolamine-N-[Biotinyl(Polyethylene Glycol)2000]; and lipids modified for
conjugation, such as l,2-Dioleoyl-sn-Glycero-3-Phosphoethanolamine-N-(succinyl) and 1,2-
Dipalnutoyl-sn-Glycero-3-Phosphoethanolamine-N-(Biotinyl). Suitable lipids whose tail groups have
been chemically-modified include, but are not limited to, polymerisable lipids, such as l,2-bis(10,12-
tricosadiynoyl)-sn-Glycero-3-Phosphocholine; fluorinated lipids, such as l-Palmitoyl-2-(16-
Fluoropalmitoyl)-sn-Glycero-3-Phosphocholine; deuterated lipids, such as l,2-Dipalmitoyl-D62-sn-
Glycero-3-Phosphocholine; and ether linked lipids, such as l,2-Di-0-phytanyl-sn-Glycero-3-
Phosphocholine.
The lipids typically comprise one or more additives that will affect the properties of the lipid
bilayer. Suitable additives include, but are not limited to, fatty acids, such as palmitic acid, myristic
acid and oleic acid; fatty alcohols, such as palmitic alcohol, myristic alcohol and oleic alcohol;
sterols, such as cholesterol, ergosterol, lanosterol, sitosterol and stigmasterol; lysophospholipids, such
as l-Acyl-2-Hydroxy-sn- Glycero-3-Phosphocholine; and ceramides. The lipid preferably comprises
cholesterol and/or ergosterol when membrane proteins are to be inserted into the lipid bilayer.
However, although lipids are commonly used to form bilayers, it is expected that in general
the method is applicable to any amphiphilic molecules which may form a layer.
As to the aqueous solution 10, in general a wide range of aqueous solutions 10 that are
compatible with the formation of a layer 11 of amphiphilic molecules may be used. The aqueous
solution 10 is typically a physiologically acceptable solution. The physiologically acceptable solution
is typically buffered to a pH of 3 to 11. The pH of the aqueous solution 10 will be dependent on the
amphiphilic molecules used and the final application of the layer 11. Suitable buffers include without
limitation: phosphate buffered saline (PBS); N-2-Hydroxyethylpiperazine-N'-2-Ethanesulfonic Acid
(HEPES) buffered saline; Piperazine-l,4-Bis-2-Ethanesulfonic Acid (PIPES) buffered saline; 3-(n-
Morpholino)Propanesulfonic Acid (MOPS) buffered saline; and Tris(Hydroxymethyl)aminomethane
(TRIS) buffered saline. By way of example, in one implementation, the aqueous solution 10 may be
10mM PBS containing 1.0M sodium chloride (NaCl) and having a pH of 6.9.
The method of using the apparatus 1 is as follows.
First, a pre-treatment coating 30 is applied to the body 2 across the recess 5, as shown in Fig.
9. The pre-treatment coating 30 is a hydrophobic fluid which modifies the surface of the body 2
surrounding the recess 5 to increase its affinity to the amphiphilic molecules.
The pre-treatment coating 30 is typically an organic substance, usually having long chain
molecules, in an organic solvent. Suitable organic substances include without limitation: n-decane,

hexadecane, isoecoisane, squalene, fluoroinated oils (suitable for use with fluorinated lipids), alkyl-
silane (suitable for use with a glass membrane) and alkyl-thiols (suitable for use with a metallic
membrane). Suitable solvents include but are not limited to: pentane, hexane, heptane, octane, decane,
and toluene. The material might typically be 0.1µl to 10µl of 0.1% to 50% (v/v) hexadecane in
pentane or another solvent, for example 2µl of 1% (v/v) hexadecane in pentane or another solvent, in
which case lipid, such as diphantytanoyl-sn-glycero-3-phosphocholine (DPhPC), might be included at
a concentration of 0.6mg/ml.
Some specific materials for the pre-treatment coating 30 are set out in Table 1 by way of
example and without limitation.

The pre-treatment coating 30 may be applied in any suitable manner, for example simply by
capillary pipette. The pre-treatment coating 30 may be applied before or after the cover 6 is attached
to the apparatus 1.
The precise volume of material of the pre-treatment coating 30 required depends on the size
of the recess 5, the formulation of the material, and the amount and distribution of the when it dries
around the aperture. In general increasing the amount (by volume and/or by concentration) improves
the effectiveness, although excessive material can cover the electrode 21 as discussed below. As the
diameter of the recess 5 is decreased, the amount of material of the pre-treatment coating 30 required
also varies. The distribution of the pre-treatment coating 30 can also affect effectiveness, this being
dependent on the method of deposition, and the compatibility of the membrane surface chemistry.
Although the relationship between the pre-treatment coating 30 and the ease and stability of layer
formation is complex, it is straightforward to optimise the amount by routine trial and error. In
another method the chamber 7 can be completely filled by pre-treatment in solvent followed by
removal of the excess solvent and drying with a gas flow.
The pre-treatment coating 30 is applied across the recess 5. As a result and as shown in Fig. 9,

the pre-treatment coating 30 covers the surface of the body 2 around the recess 5. The pre-treatment
coating 30 also extends over the rim of the recess 5 and desirably covers at least the outermost portion
of the side walls of the recess 5. This assists with formation of the layer 11 of amphiphilic molecules
across the recess 5.
However, the pre-treatment coating 30 also has a natural tendency during application to cover
the electrode 21. This is undesirable as the pre-treatment coating 30 reduces the flow of current to the
electrode 21 and therefore reduces the sensitivity of measurement of electrical signals, in the worst
case preventing any measurement at all. A number of different techniques may be employed to reduce
or avoid this problem, and will be discussed after the description of forming the layer 11 of
amphiphilic molecules.
After application of the pre-treatment coating 30, the aqueous solution 10 is flowed across the
body 2 to cover the recess 5 as shown in Fig. 3. This step is performed with the amphiphilic molecules
added to the aqueous solution 10. It has been demonstrated that, with an appropriate pre-treatment
coating 30 this allows the formation of the layer 11 of amphiphilic molecules across the recess 5.
Formation is improved if a multi-pass technique is applied in which aqueous solution 10 covers and
uncovers the recess 5 at least once before covering the recess 5 for a final time. This is thought to be
because at least some aqueous solution is left in the recess 5 which assists formation of the layer 11 in
a subsequent pass. Notwithstanding this, it should be noted that the formation of the layer 11 is
reliable and repeatable. This is despite the fact that the practical technique of flowing aqueous
solution 10 across the body 2 through the chamber 7 is very easy to perform. Formation of the layer
11 may be observed by monitoring of the resultant electrical signals across the electrodes 21 and 24,
as described below. Even if a layer 11 fails to form it is a simple matter to perform another pass of the
aqueous solution 10. Such reliable formation of a layer 11 of amphiphilic molecules using a simple
method and a relatively simple apparatus 1 is a particular advantage of the present invention.
Furthermore, it has been demonstrated that the layers 11 of amphiphilic molecules are of high
quality, in particular being suitable for high sensitivity biosensor applications such as stochastic
sensing and single channel recording. The layers 11 have high resistance providing highly resistive
electrical seals, having an electrical resistance of 1GΩ or more, typically at least 100GΩ. which, for
example, enable high-fidelity stochastic recordings from single protein pores.
This is achieved whilst trapping a volume of aqueous solution 10 in the recess 5 between the
layer 11 and the electrode 21. This maintains a significant supply of electrolyte. For example, the
volume of aqueous solution 10 is sufficient to allow stable continuous dc current measurement
through membrane proteins inserted in the layer.
Experimental results demonstrating these advantages are set out later.
There are various techniques for adding the amphiphilic molecules to the aqueous solution 10,

as follows.
A first technique is simply to add the amphiphilic molecules to the aqueous solution 10
outside the apparatus 1 before introducing the aqueous solution 10 into the chamber 7.
A second technique which has particular advantage is, before introducing the aqueous
solution 10 into the chamber 7, to deposit the amphiphilic molecules on an internal surface of the
chamber 7, or on an internal surface elsewhere in the flow path of the aqueous solution 10 into the
chamber 7, for example in a fluidic inlet pipe connected to the inlet. The amphiphilic molecules can
be deposited on any one or more of the internal surfaces of the chamber 7, including a surface of the
further layer 4 or of the cover 6. The aqueous solution 10 covers the internal surface during its
introduction, whereby the amphiphilic molecules are added to the aqueous solution 10. In this manner,
the aqueous solution 10 is used to collect the amphiphilic molecules from the internal surface. The
aqueous solution 10 may cover the amphiphilic molecules and the recess 5 in any order but preferably
covers the amphiphilic molecules first. If the amphiphilic molecules are to be covered first, the
amphiphilic molecules are deposited along the flow path between the inlet 8 and the recess 5.
Any method may be used to deposit the lipids on an internal surface of the chamber 7.
Suitable methods include, but are not limited to, evaporation or sublimation of a carrier solvent,
spontaneous deposition of liposomes or vesicles from a solution and direct transfer of the dry lipid
from another surface. An apparatus 1 having lipids deposited on an internal surface may be fabricated
using methods including, but not limited to, drop coating, various printing techniques, spin-coating,
painting, dip coating and aerosol application.
The deposited amphiphilic molecules are preferably dried. In this case, the aqueous solution
10 is used to rehydrate the amphiphilic molecules . This allows the amphiphilic molecules to be stably
stored in the apparatus 1 before use. It also avoids the need for wet storage of amphiphilic molecules.
Such dry storage of amphiphilic molecules increases shelf life of the apparatus. Even when dried to a
solid state, the amphiphilic molecules will typically contain trace amounts of residual solvent. Dried
lipids are preferably lipids that comprise less than 50wt% solvent, such as less than 40wt%, less than
30wt%, less than 20wt%, less than 15wt%, less than 10wt% or less than 5wt% solvent.
In most practical uses, a membrane protein is inserted into the layer 11 of amphiphilic
molecules. There are several techniques for achieving this.
A first technique is simply for the aqueous solution 10 to have a membrane protein added
thereto, whereby the membrane protein is inserted spontaneously into the layer 11 of amphiphilic
molecules after a period of time. The membrane protein may be added to the aqueous solution 10
outside the apparatus 1 before introducing the aqueous solution 10 into the chamber 7. Alternatively
the membrane protein may be added after formation of the layer 11.
Another way of adding the membrane protein to the aqueous solution 10 is to deposit it on an

internal surface of the chamber 7 before introducing the aqueous solution 10 into the chamber 7. In
this case, the aqueous solution 10 covers the internal surface during its introduction, whereby the
membrane protein is added to the aqueous solution 10 and subsequently will spontaneously insert into
layer 11. The membrane proteins may be deposited on any one or more of the internal surfaces of the
chamber 7, including a surface of the further layer 4 or of the cover 6. The membrane proteins can be
deposited on the same or different internal surface as the amphiphilic molecules (if also deposited).
The amphiphilic molecules and the membrane proteins may be mixed together.
Any method may be used to deposit the membrane proteins on an internal surface of the
chamber 7. Suitable methods include, but are not limited to, drop coating, various printing techniques,
spin-coating, painting, dip coating and aerosol application.
The membrane proteins are preferably dried. In this case, the aqueous solution 10 is used to
rehydrate the membrane proteins. Even when dried to a solid state, the membrane proteins will
typically contain trace amounts of residual solvent. Dried membrane proteins are preferably
membrane proteins that comprise less than 20wt% solvent, such as less than 15wt% , less than 10wt%
or less than 5wt% solvent.
A second technique is for the aqueous solution 10 to have vesicles containing the membrane
protein added thereto, whereby the membrane protein is inserted on fusion of the vesicles with the
layer 11 of amphiphilic molecules.
A third technique is to insert the membrane protein by carrying the membrane protein to the
layer 11 on a probe, for example an agar-tipped rod, using the techniques disclosed in WO-
2006/100484. Use of a probe may assist in selectively inserting different membrane proteins in
different layers 11, in the case that the apparatus has an array of recesses. However, this requires
modification to the apparatus 1 to accommodate the probe.
Any membrane proteins that insert into a lipid bilayer may be deposited. The membrane
proteins may be naturally-occurring proteins and/or artificial proteins. Suitable membrane proteins
include, but are not limited to, P-barrel membrane proteins, such as toxins, porins and relatives and
autotransporters; membrane channels, such as ion channels and aquaporins; bacterial rhodopsins; G-
protein coupled receptors; and antibodies. Examples of non-constitutive toxins include hemolysin and
leukocidin, such as Staphylococcal leukocidin. Examples of porins include anthrax protective antigen,
maltoporin, OmpG, OmpA and OmpF. Examples of autotransporters include the NalP and Hia
transporters. Examples of ion channels include the NMDA receptor, the potassium channel from
Streptomyces lividans (KcsA), the bacterial mechanosensitive membrane channel of large
conductance (MscL), the bacterial mechanosensitive membrane channel of small conductance (MscS)
and gramicidin. Examples of G-protein coupled receptors include the metabotropic glutamate
receptor. The membrane protein can also be the anthrax protective antigen.

The membrane proteins preferably comprise a-hemolysin or a variant thereof. The a-
hemolysin pore is formed of seven identical subunits (heptameric). The polynucleotide sequence that
encodes one subunit of a-hemolysin is shown in SEQ ID NO: 1. The full-length amino acid sequence
of one subunit of a-hemolysin is shown in SEQ ID NO: 2. The first 26 amino acids of SEQ ID NO: 2
correspond to the signal peptide. The amino acid sequence of one mature subunit of a-hemolysin
without the signal peptide is shown in SEQ ID NO: 3. SEQ ID NO: 3 has a methionine residue at
position 1 instead of the 26 amino acid signal peptide that is present in SEQ ID NO: 2.
A variant is a heptameric pore in which one or more of the seven subunits has an amino acid
sequence which varies from that of SEQ ID NO: 2 or 3 and which retains pore activity. 1, 2, 3,4, 5, 6
or 7 of the subunits in a variant a-hemolysin may have an amino acid sequence that varies from that of
SEQ ID NO: 2 or 3. The seven subunits within a variant pore are typically identical but may be
different.
The variant may be a naturally-occurring variant which is expressed by an organism, for
instance by a Staphylococcus bacterium. Variants also include non-naturally occurring variants
produced by recombinant technology. Over the entire length of the amino acid sequence of SEQ ID
NO: 2 or 3, a variant will preferably be at least 50% homologous to that sequence based on amino
acid identity. More preferably, the subunit polypeptide is at least 80%, at least 90%, at least 95%, at
least 98%, at least 99% homologous based on amino acid identity to the amino acid sequence of SEQ
ID NO: 2 or 3 over the entire sequence.
Amino acid substitutions may be made to the amino acid sequence of SEQ ED NO: 2 or 3, for example
a single amino acid substitution may be made or two or more substitutions may be made.
Conservative substitutions may be made, for example, according to the following table. Amino acids
in the same block in the second column and preferably in the same line in the third column may be
substituted for each other:

Non-conservative substitutions may also be made at one or more positions within SEQ ID
NO: 2 or 3, wherein the substituted residue is replaced with an amino acid of markedly different
chemical characteristics and/or physical size. One example of a non-conservative substitution that
may be made is the replacement of the lysine at position 34 in SEQ ID NO: 2 and position 9 in SEQ

ID NO: 3 with cysteine (i.e. K34C or K9C). Another example of a non-conservative substitution that
may be made is the replacement of the asparagine residue at position 43 of SEQ ID NO: 2 or position
18 of SEQ ID NO: 3 with cysteine (i.e. N43C or N17C). The inclusion of these cysteine residues in
SEQ ID NO: 2 or 3 provides thiol attachment points at the relevant positions. Similar changes could
be made at all other positions, and at multiple positions on the same subunit.
One or more amino acid residues of the amino acid sequence of SEQ ID NO: 2 or 3 may
alternatively or additionally be deleted. Up to 50% of the residues residues may be deleted, either as a
contiguous region or multiple smaller regions distributed throughout the length of the amino acid
chain.
Variants can include subunits made of fragments of SEQ ID NO: 2 or 3. Such fragments
retain their ability to insert into the lipid bilayer. Fragments can be at least 100, such as 150, 200 or
250, amino acids in length. Such fragments may be used to produce chimeric pores. A fragment
preferably comprises the P-barrel domain of SEQ ID NO: 2 or 3.
Variants include chimeric proteins comprising fragments or portions of SEQ ID NO: 2 or 3.
Chimeric proteins are formed from subunits each comprising fragments or portions of SEQ ID NO: 2
or 3. The p-barrel part of chimeric proteins are typically formed by the fragments or portions of SEQ
ID NO: 2 or 3.
One or more amino acid residues may alternatively or additionally be inserted into, or at one
or other or both ends of, the amino acid sequence SEQ ID NO: 2 or 3. Insertion of one, two or more
additional amino acids to the C terminal end of the peptide sequence is less likely to perturb the
structure and/or function of the protein, and these additions could be substantial, but preferably
peptide sequences of up to 10,20, 50,100 or 500 amino acids or more can be used. Additions at the
N terminal end of the monomer could also be substantial, with one, two or more additional residues
added, but more preferably 10, 20, 50, 500 or more residues being added. Additional sequences can
also be added to the protein in the trans-membrane region, between amino acid residues 119 and 139
of SEQ ID NO: 3. More precisely, additional sequences can be added between residues 127 and 130
of SEQ ID NO: 3, following removal of residues 128 and 129. Additions can be made at the
equivalent positions in SEQ ID NO: 2. A carrier protein may be fused to an amino acid sequence
according to the invention.
Standard methods in the art may be used to determine homology. For example the UWGCG
Package provides the BESTFIT program which can be used to calculate homology, for example used
on its default settings (Devereux et al (1984) Nucleic Acids Research 12, p387-395). The PILEUP and
BLAST algorithms can be used to calculate homology or line up sequences (such as identifying
equivalent residues or corresponding sequences (typically on their default settings)), for example as
described in Altschul S. F. (1993) J Mol Evol 36:290-300; Altschul, S.F et al (1990) J Mol Biol

215:403-10. Software for performing BLAST analyses is publicly available through the National
Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/).
The membrane proteins can be labelled with a revealing label. The revealing label can be any
suitable label which allows the proteins to be detected. Suitable labels include, but are not limited to,
fluorescent molecules, radioisotopes, e.g. 1251, 35S, enzymes, antibodies, polynucleotides and linkers
such as biotin.
The membrane proteins may be isolated from an organism, such as Staphylococcus aureus, or
made synthetically or by recombinant means. For example, the protein may be synthesized by in vitro
transcription translation. The amino acid sequence of the proteins may be modified to include non-
naturally occurring amino acids or to increase the stability of the proteins. When the proteins are
produced by synthetic means, such amino acids may be introduced during production. The proteins
may also be modified following either synthetic or recombinant production.
The proteins may also be produced using D-amino acids. In such cases the amino acids will
be linked in reverse sequence in the C to N orientation. This is conventional in the art for producing
such proteins.
A number of side chain modifications are known in the art and may be made to the side
chains of the membrane proteins. Such modifications include, for example, modifications of amino
acids by reductive alkylation by reaction with an aldehyde followed by reduction with NaBH4,
amidination with methylacetimidate or acylation with acetic anhydride.
Recombinant membrane proteins can be produced using standard methods known in the art.
Nucleic acid sequences encoding a protein can be isolated and replicated using standard methods in
the art. Nucleic acid sequences encoding a protein can be expressed in a bacterial host cell using
standard techniques in the art. The protein can be introduced into a cell by in situ expression of the
polypeptide from a recombinant expression vector. The expression vector optionally carries an
inducible promoter to control the expression of the polypeptide.
Thus the apparatus 1 can be used for a wide range of applications. Typically a membrane
protein is inserted in the layer 11. An electrical signal, typically a current signal, developed between
the electrode 21 in the recess 5 and the further electrode 24 in the chamber 7 is monitored, using the
electrical circuit 26. Often a voltage is also applied between the electrodes 21 and 24, whilst
monitoring the electrical signal. The form of the electrical signal, and in particular changes therein,
provide information about the layer 11 and any membrane protein inserted therein.
Some non-limitative examples of uses will now described. One use is in vitro investigation of
membrane proteins by single-channel recording. An important commercial use is as a biosensor to
detect the presence of a range of substances. The apparatus 1 may be used to detect an analyte
molecule that binds with an inserted membrane protein, or another stimulus, using stochastic sensing

by detecting a change in the current flow indicating the presence of the anlayte molecule or other
stimulus. Similarly, the apparatus 1 may be used to detect the presence or absence of membrane pores
or channels in a sample, by detecting a change in the current flow as the pore or channel inserts. The
lipid bilayer may be used for a range of other purposes, such as studying the properties of molecules
known to be present (e.g. DNA sequencing or drug screening), or separating components for a
reaction.
Some techniques to reduce or avoid the problem of the pre-treatment coating 30 covering the
electrode 21 will now be discussed.
A first technique is, after application of the pre-treatment coating 30 to apply a voltage across
the electrode 21 in the recess 5 and the further electrode 24 in the chamber 7 sufficient to reduce the
amount of excess hydrophobic fluid covering the electrode 21 in the recess 5 . This is produces a
similar effect to electro-wetting.
This technique is illustrated in Figs. 10a to 10e. First, as shown in Fig. 10a, the pre-treatment
coating 30 is applied as shown in Fig. 10a where the pre-treatment coating 30 covers the electrode 21.
Next, as shown in Fig. 10b, aqueous solution 10 is flowed across the body 2 to cover the recess 5 so
that aqueous solution 10 flows into the recess 5. Next, a voltage is applied which removes the pre-
treatment coating 30 covering the electrode 21, as shown in Fig. 10c. This voltage will rupture any
layer of amphiphilic molecules formed across the recess 5. Therefore, next, as shown in Fig. 10d, the
aqueous solution 10 is flowed out of the chamber 7 to uncover the recess 5. Typically an amount of
aqueous solution 10 will remain in the recess 5. Lastly, as shown in Fig. lOe, aqueous solution 10,
having amphiphilic molecules added thereto, is flowed across the body 2 to re-cover the recess 5 so
that the layer 11 of the amphiphilic molecules forms.
This is most simply performed by flowing the same aqueous solution 10 in and out of the
chamber 7. However, in principle, the aqueous solution 10 flowed into the chamber 7 to re-covering
the recess 5 (in Fig. 10e) could be different from the aqueous solution 10 flowed into the chamber 7 to
first cover the recess 5 (in Fig. 10b) before applying the voltage. Similarly, there could be no
amphiphilic molecules added to the aqueous solution 10 flowed into the chamber 7 to first cover the
recess 5 (in Fig. 10b) before applying the voltage.
A second technique is to make an inner part of the internal surface of the recess 5 hydrophilic.
This may be achieved by making the body 2 with two (or in general more) further layers 4a and 4b as
shown in Fig. 11, of which the innermost further layer 4a (or layers) formed of a hydrophilic material,
for example SiO2. Typically but without limitation, the innermost further layer 4a might have a
thickness of 2µm.
The outermost further layer 4b (or layers) is formed of a hydrophobic material and as a result
both of (a) the outermost surface of the body 2 around the recess and (b) the outer part of the internal

surface of the recess 5 extending from the rim of the recess 5 is hydrophobic. This assists in the
spreading of the pre-treatment coating. Indeed this property of these surfaces of the body 2 is
desirable even if there is not an inner further layer 4a formed of a hydrophilic material. Typically but
without limitation, outermost further layer 4b might have a thickness of 1µm, 3µm, 5µm, 10µm, 20µm
or 30um.
A third technique is to provide a hydrophillic surface on the electrode 21 which repels the
applied pre-treatment coating 30, whilst allowing ionic conduction from the aqueous solution 10 to
the electrode 2. This may be achieved by depositing a protective material on the electrode 21. A range
of protective materials may be used. One possibility is a conductive polymer, for example
polypyrrole/ polystyrene sulfonate as discussed above. Another possibility is a covalently attached
hydrophilic species, such as thiol-PEG.
The apparatus 1 described above has been made and used experimentally to demonstrate
formation of a layer 11, in particular being a lipid bilayer, and insertion of a membrane protein, in
particular a-hemolysin. The following procedure was followed after manufacture of the apparatus 1:
1) apply pre-treatment coating 30 to body 2;
2) introduce aqueous solution 10 into chamber 7 to cover recess 5;
3) electro-wet the electrode 21;
4) remove aqueous solution 10 to un-cover recess 5 and re introduce aqueous solution 10 into
chamber 7 to cover recess 5 and form the layer 11;
5) add a-hemolysin free into aqueous solution 10 and monitor insertion into layer 11.
m step 1), the pre-treatment coating 30 was hexadecane dissolved in pentane. The quantity
and volume of the pre-treatment coating 30 was varied for each test to obtain the optimum conditions
for formation of the layer 11. Insufficient pre-treatment coating 30 prevented formation of the layer
11 while excess pre-treatment coating 30 caused blocking of the recesses. However routine variation
of the amount allowed optimisation.
The amphiphilic molecules were a lipid, in particular l,2-diphytanoyl-sn-glycero-3-
phosphocholine. The lipid was dissolved in pentane and then dried onto the surface of the cover 6
defining an internal surface of the chamber 7 before attaching the cover 6 on top of the body 2. In step
2), the aqueous solution 10 collected the lipid.
Step 3) was performed by application of a large potential to across the electrodes 21 and 24.
This removed excess pre-treatment coating 30 from the electrode 21. Although not required in every
case, when performed this stage helped to condition the recess 5 for formation of the layer 11 and
assisted subsequent measurement of electrical signals.
By monitoring of the electrical signals developed across the electrodes 21 and 24, in steps 4)
and 5), formation of the layer 11 and insertion of the membrane protein was observed.

The procedure was successfully performed for an apparatus 1 of the type described above
formed by lamination onto a polymer substrate 3. Formation of the layer 11 and insertion of the
membrane protein was observed using all the fabrication variables described above, albeit with
varying degrees of repeatability and signal quality.
An example will now be described for a typical apparatus 1, in which the first conductive
layer 20 was formed by a silver foil strips (25µm thick, from Goodfellow) thermally laminated onto
the substrate 3 using a 15µm thick laminating film (Magicard) to form the further layer 4. A circular
recess 5 of diameter 100µm was created further layer 4 using an excimer laser, exposing a circular
silver electrode 21 of diameter lOOum. The exposed silver was chloridised electrochemically as
described previously. The second conductive layer 23 was a screen printing silver/silver chloride ink
printed on the top side of the body 2.
The pre-treatment coating 30, comprising 0.5ul of 1% heaxadecane +0.6 mg/ml DPhPC in
pentane, was then applied to the body 2 and dried at room temperature.
The cover 6 comprised a 1mm thick silicon rubber body with a 250µm thick Mylar lid. Lipid
(4µl of 10mg/ml DPhPC in pentane) was applied to the inside of the cover 6 and allowed to dry at
room temperature before attachment to the body 2 with self-adhesive.
A typical successful test proceeded as follows.
The dry contacts 22 and 25 were attached to the electrical circuit 26 enclosed in a Faraday
cage and a 20mV 50 Hz triangular potential waveform applied. Fig. 15 shows the applied waveform
and the resultant current signal which is indicative of the expected capacitive response.
Addition of the aqueous solution 10 creates an "open circuit" connection between the
electrodes, such that the current response to the applied potential waveform is large, typically
saturating the current amplifier. A typical trace is shown in Fig. 16, involving a current response
greater than 20,000pA to the 20mV potential. This corresponds to a resistance of less thanlMΩ,
which is sufficiently small for use in conjunction with bilayer formation and pore current
measurement.
In the event that the electrode 21 does not initially form a proper electrical connection with
the aqueous solution 10 , application of a -IV DC potential can be used to increase in the available
active electrode area. This is illustrated in Fig. 17, in which the electrode begins partially active and is
then fully activated after around 4s of the applied potential.
Following open-circuit connection between the aqueous solution 10 and the electrode 21, the
aqueous solution 10 is removed from the chamber 7 and reintroduced. On re-introduction, a layer 11
of the lipid collected from the internal surface of the chamber 7 is formed across the recess 5. The
formation is observed by an increase in the capacitive squarewave current response to just under 500
pA, for example as shown in Fig. 18. This value is consistent with the capacitance expected for a

circular lipid bilayer of diameter of order 100pm and varies predictably for different geometries.
Subsequent addition of a-hemolysin to the aqueous solution 10 creates a current response
typical of pore insertion under an applied potential of 100mV. For example Fig. 19 is a typical
example with cyclodextrin present in the aqueous solution 10 and shows an expected current response
with binding events confirming that the current is through the pores.
Although the example above shows data for the thermally laminated apparatus 1, the other
systems investigated also produced successful formation of the layer 11 and pore insertion. For
example, this was also successfully demonstrated for an apparatus 1 formed by lamination using
pressure-sensitive adhesive bonding of the further layer 4. However, the adhesive layer was found to
complicate formation of the recess 5 both in terms of the resulting aspect ratio and spreading of the
adhesive across the electrode 21. This problem was overcome by electrical sparking to "activate" the
electrode 21.
The impact of the quality of the recess 5 is evident by comparing results from recesses formed
by a CO2 laser and an excimer laser, as shown in Figs. 20 and 21, respectively. In both cases
formation of the layer 11 and pore insertion is successful and evident in the response, but more
reproducible apertures were produced using the excimer laser. Recesses 5 formed by the COz laser
tended to form relatively leaky layers 11 with more noisy pore signals and were also susceptible to
blocking. Recesses 5 formed by the excimer laser produced well sealed layers 11 with good pore
signals.
Formation of the layer 11 and pore insertion was similarly observed with an apparatus 1
formed as described above using high definition printed circuit board manufacture. Im this case, to
form apparatus 1, the first conductive layer 20 was formed by etching the copper foil on an FR4
substrate typically used in printed circuit board manufacture. The board was then screen printed with
a Ronascreen SPSR™ photoimageable solder mask to a depth of 25 µrn and exposed to UV light on an
Orbotech Paragon 9000 laser direct imaging machine and developed with KaCO3 solution to create
100µm circular apertures over the electrodes 21.
Formation of the layer 11 and pore insertion was similarly observed with an apparatus 1
formed as described above using photolithography. In this case, to form the apparatus 1, the first
conductive layer 20 was formed by gold vapour deposited using clean-room facilities onto the
substrate 3 and a further layer 4 of SU8 photoresist of thickness 12.5µm was spin-coated on top.
Recesses 5 were formed by curing of the photoresist by UV exposure with a mask and subsequent
removal of the uncured photoresist. The recesses 5 had a diameter of 100µm, exposing an electrode
21 of diameter lOOum. After baking to set the photoresist, the wafer was diced to form separate
substrates each with a single recess 5. The electrodes 21 were electroplated with silver and then
chloridised electrochemically as described previously. The second electrode 24 was screen printed

silver/silver chloride ink printed on the top side of the body 2.
The pre-treatment coating 30, comprising 0.5 µl of 0.75% hexadecane in pentane, was then
applied to the body 2 and dried at room temperature.
The cover 6 comprised a 1mm thick silicon rubber body with a 250um thick Mylar lid. Lipid
(4µl of 10mg/ml DPhPC in pentane) was applied to the inside of the cover 6 and allowed to dry at
room temperature before attachment to the body 2 with self-adhesive.
Testing was performed as described above and successful formation of the layer 11 and pore
insertion was observed. For example, Fig. 22 shows a typical current trace showing cyclodextrin
binding events with wild-type a-hemolysin pores.
These results generally show the ease with which the method of formation of the layer 11 may
be performed. In particular formation of the layer 11 is achieved with a wide range of materials of the
apparatus 1, dimensions (width and depth) of the recess 5, and methods of manufacture. Some
variation in success rate is evident but in general this can be optimised by routine testing of different
apparatuses 1. In particular the formation of the layer 11 is not overly dependent on the width of the
recess 5. Formation has been demonstrated over widths from 5µm to 100µm and in view of the ease of
formation it is expected that formation is possible at higher widths up to 200µm, 500um or higher.
Also in view of this ease of formation of the layer 11, it is expected that variations of the shape of the
recess 5 could also be accommodated.
There will now be discussed modifications to the apparatus 1 to include plural recesses 5,
commonly referred to as an array of recesses 5. The ability to easily form an array of layers 11 across
an array of recesses 5 in a single apparatus 1 is a particular advantage of the present invention. By
contrast to traditional methods of formation of lipid bilayers, the apparatus 1 has a single chamber 7,
but creates the layer 11 in situ during the test and captures a reservoir of electrolyte in the recess 5
under the layer 11 which allows continuous stable measurement of current passing through protein
pores inserted in the layer 11. Further the layer 11 formed is of high quality and is localised to the
area of the recess 5, ideal for high-fidelity current measurements using membrane protein pores.
These advantages are magnified in an apparatus 1 which forms an array of layers 11 because this
allows measurements to be taken across all the layers 11 in parallel, either combining the current
signals to increase sensitivity or monitoring the current signals separately to perform independent
measurements across each layer 11.
Apparatuses having an array of recesses 5 have been tested and demonstrated successful
formation of an array of layers 11, showing the possibility of creating a miniaturised array of close
packed individually addressable layers recording current signals in parallel from a test sample.
Essentially an apparatus 1 having an array of recesses 5 can be formed simply using the
manufacturing techniques described above but instead forming plural recesses 5. In this case, the first

-
conductive layer 20 is divided to form a separate electrode 21, contact 22 and intermediate conductive
track 27 in respect of each recess 5. The apparatus 1 has a single chamber 7 with a single electrode 24
common to all the recesses 5.
Figs. 23 to 25 show first to third designs in which the apparatus 1 is modified by providing,
respectively, four, nine and 128 recesses 5 in the further layer 4. In each of the first to third designs,
the first conductive layer 20 is divided, as shown, respectively, in Figs. 26 to 28 being plan views of
the substrate 3. The first conductive layer 20 provides, in respect of each recess 5: an electrode 21
underneath the recess 5; a contact 22 exposed for connection of the external circuit 26 and a track 27
between the electrode 21 and the contact 22. Thus each electrode 21, and its associated track 27 and
contact 22, is electrically insulated from each other allowing separate measurement of current signals
from each recess 5.
Manufacture of the apparatus 1 may be performed using the techniques described above using
lamination of polymer films or photolithography using silicon wafers.
Apparatuses 1 having plural recesses 5 have been made and used experimentally to
demonstrate formation of a layer 11, in particular being a lipid bilayer, and insertion of a membrane
protein, in particular a-hemolysin. The experimental procedure was as described above for an
apparatus 1 having a single recess 5, except that formation of the layer 5 and membrane protein
insertion was observed at plural recesses 5. Some examples are as follows.
An apparatus of the first design having four recesses 5 was manufactured by the technique
described above of lamination onto a polymer substrate 3. The first conductive layer 20 was silver
vapour deposited on a polyester sheet substrate 3. The further layer 4 was a 15um thick laminating
film thermally laminated on top. The four recesses 5 of 100µm diameter were formed at a pitch of
300µm by an excimer laser.
For recording of from each recess 5 simultaneously in parallel, multiple Axon current
amplifier devices were operated in parallel with a single silver/silver chloride electrode 24 in the
chamber 7 as the ground electrode common to all channels. Formation of layers 11 and insertion of
membrane proteins at plural recesses 5 was successfully recorded in parallel. Often this occurred at
each recess 5 although sometimes a layer 11 failed to form at one or more recesses 5. For example
typical current traces are shown in Fig. 29 demonstrating simultaneous formation of four layers 11,
each having one or two a-hemolysin pores inserted, with cyclodextrin binding events. Notably there is
no cross-talk between the signals. This confirms that the layers 11 are operating independently and
can produce meaningful measurements in parallel while being individually addressed and using a
common second electrode 24.
An apparatus of the second design having nine recesses 5 was manufactured by the technique
described above of photolithography using silicon wafer substrates 2. The further layer 4 was 5µm

thick SU8 photoresist. The nine circular recesses 5 were formed at a pitch of 300µm by
photolithography. In this case, the recesses 9 had different diameters, in particular of 5µm, 10µm,
15µm, 20µm, 20µm, 30µm, 40µm, 50µm, and 100µm. The substrate 3 was bonded to a printed circuit
board with separate tracks connected to each contact 22 and 25. Epoxy was added across the contacts
22 and 25 for protection
In order to control the applied potential and record the current response in parallel, a
multichannel electrical circuit 26 was created with corresponding software. Testing was computer
automated using a syringe pump to provide fluidics control of the repeated application and removal of
the aqueous solution 10.
Formation of layers 11 and insertion of membrane proteins at plural recesses 5 was
successfully recorded in parallel. Often this occurred at each recess 5 although sometimes a layer 11
failed to form at one or more recesses 5. For example, typical current traces for recesses 5 constructed
with gold electrodes and operating without a redox couple in solution are shown in Fig. 30
demonstrating simultaneous formation of eight layers 11, each having one or two a-hemolysin pores
inserted, with cyclodextrin binding events. Again there is no cross-talk and this confirms that the
layers 11 are operating independently and can produce meaningful measurements in parallel.
Furthermore the apparatus 1 demonstrates successful formation of a layer 11 across the recess
5 of each diameter in the range of 5µm to 100µm. Accordingly the apparatus 1 was used to investigate
the role of the diameter of the recess 5 and the quantity of pre-treatment coating applied, by
experimentally testing the percentage success rate of forming a layer 5 with three different
concentrations of pre-treatment coating 30, namely 0.5%, 1.0%, and 2.0% hexadecane in pentane. The
results showed that in the case of too little pretreatment coating 30, it was not possible to form the
layer 11 across the range of diameters of recess 5. Furthermore in the case of too much pretreatment
coating 30, it was not possible to wet the electrode 21 and formation of the layer 11 could not be
observed. In this particular configuration, the yield of formation of layers 11 was greater than 60% for
the range of diameters 15µm to 100µm. Factors affecting layer formation, some of which were
investigated in this experiment include, but are not limited to, pretreatment coating 30, diameter of
recess 5, depth of recess 5, aspect ratio of recess 5, surface properties of the recess 5, surface
properties of the surfaces around the recess, fluid flow within the chamber 7, the amphiphilic
molecules used in the layer formation and the physical and electrical properties of the electrode 21
within the recess 5. Subsequent experiments have demonstrated yield of formation of layers 11,
verified by stochastic binding signals of inserted membrane channels, greater than 70% using the 128
recesses, each 100µm in diameter, of the device of Fig. 28.
In the apparatus 1 described above, the conductive tracks 27 from the electrode 21 to the
contact 22 is formed on a surface of the substrate 3 under the further layer. This may be referred to as

a planar escape route for the conductive track 27. As previously described the separate conductive
tracks 27 allow each electrode 21 to be connected individually to a dedicated low-noise high- input
impedance picoammeter in the circuit 26 whilst minimising the signal deterioration due to noise and
bandwidth reduction. Such planar conductive tracks 27 are ideal for an apparatus 1 having a small
number of recesses 5 and a thick layer between the tracks 27 and the aqueous solution 10.
However, for uses where high sensitivity is required, the electrical connection between the
electrodes 21 and the amplifier circuit desirably has low parasitic capacitance and low leakage to the
surroundings. Parasitic capacitance causes noise and hence signal deterioration and bandwidth
reduction. Leakage also increases noise, as well as introducing an offset current. In the apparatus 1,
the conductive tracks 27 experience some degree of parasitic capacitance and leakage, both between
tracks 27 and between track and aqueous solution 10. As the number of recesses in the array
increases, the number of electrical connections to escape increases and with a planar escape route, a
practical limit is reached where the density of the conductive tracks 27 creates too much parasitic
capacitance and/or leakage between tracks. Furthermore as the thickness of the layer 4 decreases the
capacitance and/or leakage between the tracks 27 and the aqueous solution 10 increases.
By way of example, typical figures may be obtained by modelling the lipid bilayer as a
capacitive element with a typical value for the capacitance per unit area of 0.8µF/cm2. The parasitic
capacitance between track 27 and aqueous solution 10 can be crudely modelled as a capacitative
element with the area of track 27 exposed, through the layer, to the aqueous solution. Typical values
for the track 27 may be 50um wide with 2mm exposed and a relative permittivity (dielectric constant)
of the layer around 3. For a 100µm diameter bilayer and 20µm deep recess the capacitance is 63pF
with a track-solution parasitic capacitance of 0.13pF. However scaling to smaller bilayers of 5µm
diameter and 1 µm deep the capacitance is 0.16pF with parasitic capacitance 0.53pF. For smaller
bilayers and thinner layers the parasitic capacitance dominates.
To reduce this problem, a modification shown in Fig. 31 is to replace the conductive track 27
by a conductive path 28 which extends through the body 2 to a contact 29 on the opposite side of the
body 2 from the electrode 21. In particular, the conductive path 28 extends through the substrate 3. As
this substrate 3 provides a thicker dielectric between the conductive paths 28 than is possible between
the planar conductive paths 27, a much lower parasitic capacitance is achieved. Also, the leakage is
low due to the thickness and dielectric properties of the substrate 3. Consequently, the use of the
conductive paths 28 effectively increases the number of recesses 5 which may be accomodated in the
body 2 before the practical limits imposed by parasitic capacitance and/or leakage are met. This form
of interconnect can be attached to a low-capacitance multi-layer substrate 61, which allows a far
greater number of electrical escape routes by virtue of the number of layers and the low dielectric
constant of the material. In addition the use of solder bump technology (also known as "flip chip"

technology) and a suitable connector allows the apparatus 1 shown in Fig. 31, excluding the substrate
61, to be made as low cost disposable part.
The conductive path 28 may be formed using known through-wafer interconnection
technology. Types of through-wafer interconnects which may be applied to form the conductive path
include without limitation:
on substrates 3 of silicon, through-wafer interconnects formed by producing a via through the
silicon wafer, isolating the internal surface of via and filling the via with a conducting material, or
alternatively the conductive path 28 is formed by producing a semiconductor PN junction in the form
of a cylindrical via through the silicon substrate;
on substrates 3 of glass , through-wafer interconnects formed by methods including laser
drilling, wet etching and filling vias with metal or doped semiconductor material; and
on substrates 3 made of polymers, through-wafer interconnects formed by methods including
laser drilling, laser ablation, screen printed conductors and known printed circuit board techniques.
As the opposite side of the body 2 from the electrode 21 is dry, an electrical point contact
array can be used to make connections to the electrical circuit 26. By way of example, Fig. 31
illustrates the use of solder bump connections. In particular, deposited on each contact 29 are
respective solder bumps 60 on which a circuit element 61 is mounted so that the solder bumps 60
make electrical contact with a track 62 on the circuit element 61.
The circuit element 61 may be a printed circuit board for example as shown in Fig. 13.
Alternatively, the circuit element 61 could be an integrated circuit chip or a laminate, for
example a low temperature cured ceramic package. Such an integrated circuit chip or laminate may be
used as a method of spreading out connections, connecting to a further solder bump array on the
opposite side of the integrated circuit chip or laminate with a greater pitch. An example of this is
shown in Fig. 32 in which the circuit element 61 is an integrated circuit chip or a laminate providing
connections from the solder bumps 60 deposited on the body 2 to further solder bumps 63 arrayed at a
greater pitch and used to connect to a further circuit element 64, for example a printed circuit board.
The circuit element 61 being an integrated circuit chip or laminate may also be used to escape
connections sideways in a multi-layer format.
In the case of a substrate 3 of semiconductor material such as silicon, two types of through-
wafer interconnect which may be applied to make the conductive path 28 are Metal-Insulator-
Semiconductor (MIS), and a PN junction type. In MIS, a hole is drilled through the silicon chip by
Deep Reactive Ion Etching (DRIE) process and this hole is coated with insulator and then filled with
metal to forma the conductive path 28. The PN junction type of through-wafer interconnect is a
semiconductor junction formed into a cylindrical via through a silicon chip. Each type of through-
wafer interconnection is formed on silicon wafers that have been thinned down to less than 0.3mm to

save DREE processing time in making the holes. The important feature of PN junction type through-
wafer interconnects is the low capacitance provided by having a large depletion region compared to
the MIS type of interconnect. This is partially helped by increasing the reverse-bias of the junction.

WE CLAIM
1. A method of forming a layer separating two volumes of aqueous solution, the method
comprising:
(a) providing an apparatus comprising elements defining a chamber, the elements including a
body of non-conductive material having formed therein at least one recess opening into the chamber,
the recess containing an electrode;
(b) applying a pre-treatment coating of a hydrophobic fluid to the body across the recess;
(c) flowing aqueous solution, having amphiphilic molecules added thereto, across the body to
cover the recess so that aqueous solution is introduced into the recess from the chamber and a layer of
the amphiphilic molecules forms across the recess separating a volume of aqueous solution introduced
into the recess from the remaining volume of aqueous solution.
2. A method according to claim 1, wherein step (c) comprises:
(c1) flowing aqueous solution across the body to cover the recess so that aqueous solution
flows into the recess;
(c2) flowing the aqueous solution to uncover the recess, leaving some aqueous solution in the
recess; and
(c3) flowing aqueous solution, having amphiphilic molecules added thereto, across the body
and to re-cover the recess so that a layer of the amphiphilic molecules forms across the recess
separating a volume of aqueous solution inside the recess from the remaining volume of aqueous
solution.
3. A method according to claim 2, wherein
the apparatus is provided with a further electrode in the chamber outside said recess,
in step (c1), the aqueous solution is flowed also to contact the further electrode, and
step (c) further comprises, between steps (cl) and (c2):
(c4) applying a voltage across said electrode contained in the recess and said further electrode
sufficient to reduce the amount of excess hydrophobic fluid covering said electrode contained in the
recess.
4. A method according to claim 2 or 3, wherein the aqueous solution caused to flow in steps (cl)
and (c2) is the same aqueous solution.
5. A method according to any one of the preceding claims, wherein surfaces including one or

both of (a) the outermost surface of the body around the recess, and (b) at least an outer part of the
internal surface of the recess extending from the rim of the recess, are hydrophobic.
6. A method according to claim 5, wherein the body comprises an outermost layer formed of a
hydrophobic material, the recess extending through the outermost layer and said outer part of the
internal surface of the recess being a surface of the outermost layer.
7. A method according to claim 5, wherein an inner part of the internal surface of the recess
inside the outer part is hydrophilic.
8. A method according to claim 7, wherein the body comprises an outermost layer formed of a
hydrophobic material and an inner layer formed of a hydrophilic material, the recess extending
through the outermost layer and inner layer, said outer part of the internal surface of the recess being a
surface of the outermost layer, and said inner part of the internal surface of the recess being a surface
of the inner layer.
9. A method according to claim 5, wherein said surfaces are modified by a fluorine species.
10. A method according to claim 9, wherein said surfaces are modified by a fluorine species by
treatment with a fluorine plasma.
11. A method according to any one of the preceding claims, wherein the electrode contained in
the recess is provided on the base of the recess.
12. A method according to any one of the preceding claims, wherein the body comprises a
substrate and at least one further layer attached to the substrate, the recess extending through the at
least one further layer.
13. A method according to any one of the preceding claims, wherein the electrode has provided
thereon a hydrophillic surface which repels the hydrophobic fluid applied in step (c) whilst allowing
ionic conduction from the aqueous solution to the electrode.
14. A method according to claim 13, wherein the hydrophillic surface is the surface of protective
material provided on the electrode.

15. A method according to claim 14, wherein the protective material is a covalently-attached
hydrophillic species or a conductive polymer.
16. A method according to any one of the preceding claims, wherein the electrode has a
conductive polymer provided thereon.
17. A method according to any one of the preceding claims, wherein the elements defining the
chamber further include a cover extending over the body so that the chamber is a closed chamber.
18. A method according to claim 17, wherein the cover comprises at least one inlet and at least
one outlet, the aqueous solution being introduced into the chamber through the inlet in step (c) and the
outlet venting fluid displaced by the aqueous solution thus introduced.
19. A method according to any one of the preceding claims, wherein the internal surface of the
recess has no openings capable of fluid communication.
20. A method according to any one of the preceding claims, wherein the at least one recess
comprises plural recesses.
21. A method according to any one of the preceding claims, wherein the layer of the amphiphilic
molecules is a bilayer of the amphiphilic molecules.
22. A method according to claim 21, wherein the amphiphilic molecules are lipids.
23. A method according to any one of the preceding claims, wherein the layer of the amphiphilic
molecules has an electrical resistance of at least 1GΩ
24. A method according to any one of the preceding claims, further comprising, before step (c),
depositing the amphiphilic molecules on an internal surface of the chamber or on an internal surface
in the flow path of the aqueous solution into the chamber, the aqueous solution covering the internal
surface during step (c) whereby the amphiphilic molecules are added to the aqueous solution.
25. A method according to any one of the preceding claims, further comprising inserting a
membrane protein into the layer of amphiphilic molecules.

26. A method according to claim 25, wherein the aqueous solution has a membrane protein added
thereto, whereby the membrane protein is inserted spontaneously into the layer of amphiphilic
molecules.
27. A method according to claim 25, further comprising, before step (c), depositing the membrane
protein on an internal surface of the chamber, the aqueous solution covering the internal surface
during step (c) whereby the membrane protein is added to the aqueous solution.
28. A method according to any one of the preceding claims, wherein the at least one recess
comprises plural recesses and the method comprises inserting different membrane protein into the
layers of amphiphilic molecules formed in different recesses.
29. A method according to any one of claims 25 to 28, wherein the apparatus is provided with a
further electrode in the chamber outside the recess, and the method further comprises applying a
potential across the electrode in the recess and the further electrode and monitoring an electrical signal
developed between the electrode in the recess and the further electrode.
30. An apparatus for supporting a layer separating two volumes of aqueous solution, the apparatus
comprising:
elements defining a chamber, the elements including a body of non-conductive material
having formed therein at least one recess opening into the chamber; and
an electrode contained in the recess.
31. An apparatus according to claim 30, wherein surfaces including either or both of (a) the
outermost surface of the body around the recess, and (b) at least an outer part of the internal surface of
the recess extending from the rim of the recess, are hydrophobic.
32. An apparatus according to claim 31, wherein the body comprises an outermost layer formed of
a hydrophobic material, the recess extending through the outermost layer and said outer part of the
internal surface of the recess being a surface of the outermost layer.
33. An apparatus according to claim 31, wherein an inner part of the internal surface of the recess
inside the outer part is hydrophilic.
34. An apparatus according to claim 33, wherein the body comprises an outermost layer formed of

a hydrophobic material and an inner layer formed of a hydrophilic material, the recess extending
through the outermost layer and inner layer, said outer part of the internal surface of the recess being a
surface of the outermost layer, and said inner part of the internal surface of the recess being a surface
of the inner layer.
35. An apparatus according to claim 31, wherein said surfaces are modified by a fluorine species.
36. An apparatus according to claim 35, wherein said surfaces are modified by a fluorine species
by treatment with a fluorine plasma.
37. An apparatus according to any one of claims 30 to 36, wherein the electrode contained in the
recess is provided on the base of the recess.
38. An apparatus according to any one of claims 30 to 37, wherein the body comprises a substrate
and at least one further layer attached to the substrate, the recess extending through the at least one
further layer.
39. An apparatus according to claim 38, wherein the at least one further layer is: polycarbonate;
poly-vinyl chloride; polyester; a thermal laminating film; a photoresist; or an ink.
40. An apparatus according to claim 38 or 39, wherein the substrate comprises at least one of
silicon, silicon oxide, silicon nitride or a polymer.
41. An apparatus according to any one of claims 30 to 40, wherein the body has a conductive path
extending from the electrode in the chamber to a contact allowing connection to an electrical circuit.
42. An apparatus according to claim 41, wherein the conductive path extends through the body to
a contact disposed on the opposite side of the body from the recess.
43. An apparatus according to claim 41, wherein the conductive path extends across a surface of
the substrate under the at least one further layer.
44. An apparatus according to any one of claims 30 to 43, wherein the electrode has provided
thereon a hydrophillic surface which repels the hydrophobic fluid applied in step (c) whilst allowing
ionic conduction from the aqueous solution to the electrode.

45. An apparatus according to claim 44, wherein the hydrophillic surface is the surface of
protective material provided on the electrode.
46. An apparatus according to claim 45, wherein the protective material is a covalently-attached
hydrophillic species or a conductive polymer.
47. An apparatus according to any one of claims 30 to 46, wherein the electrode has a conductive
polymer provided thereon.
48. An apparatus according to any one of claims 30 to 47, further comprising a further electrode
in the chamber outside said recess.
49. An apparatus according to any one of claims 30 to 48, wherein the elements defining the
chamber further include a cover extending over the body so that the chamber is a closed chamber.
50. An apparatus according to claim 49, wherein the cover comprises at least one inlet and at least
one outlet, the aqueous solution being introduced into the chamber through the inlet in step (c) and the
outlet venting fluid displaced by the aqueous solution thus introduced.
51. An apparatus according to any one of claims 30 to 50, wherein the internal surface of the
recess has no openings capable of fluid communication.
52. An apparatus according to any one of claims 30 to 51, wherein the recess has a width of at
most 500um.
53. An apparatus according to any one of claims 30 to 52, wherein the at least one recess is plural
recesses.
54. An apparatus according to any one of claims 30 to 53, further comprising amphiphilic
molecules deposited on an internal surface of the chamber.
55. An apparatus according to claim 54, wherein the amphiphilic molecules are lipids.
56. An apparatus according to any one of claims 30 to 55, further comprising a membrane protein
deposited on the on an internal surface of the chamber.

57. An apparatus according to any one of claims 30 to 56, further comprising a pre-treatment
coating of a hydrophobic fluid applied to the body across the recess.
58. An apparatus according to claim 57, wherein the recess and the chamber contain aqueous
solution.
59. An apparatus according to claim 58, further comprising a layer of amphiphilic molecules
extending across the opening of the recess.
60. An apparatus according to claim 59, wherein the layer of the amphiphilic molecules has an
electrical resistance of at least 1GΩ.
61. An apparatus according to claim 59 or 60, wherein the amphiphilic molecules are lipids.
62. An apparatus according to any one of claims 59 to 61, the layer of amphiphilic molecules
having a membrane protein inserted therein.
63. A method of using an apparatus according to any one of claims 59 to 62, wherein the
apparatus is provided with a further electrode in the chamber outside the recess, and the method
comprises applying a potential across the electrode in the recess and the further electrode and
monitoring an electrical signal developed between the electrode in the recess and the further electrode.
64. A method of improving the performance of an electrode in a recess in conducting electro-
physiological measurements, the method comprising depositing a conductive polymer on the
electrode.
65. A method according to claim 64, wherein the electrode is made of metal.
66. A method according to claim 65, wherein the electrode is made of silver, gold or platinum.
67. A method according to any one of claims 64 to 66, wherein the conductive polymer is a
polypyrrole.
68. An apparatus for conducting electro-physiological measurements, the apparatus comprising, a
body having a recess in which an electrode is located, wherein a conductive polymer is deposited on

the electrode.
69. A method according to claim 68, wherein the electrode is made of metal.
70. A method according to claim 69, wherein the electrode is made of silver, gold or platinum.
71. A method according to any one of claims 68 to 70, wherein the conductive polymer is a
polypyrrole.

To form a layer separating two volumes of aqueous solution, there is used an apparatus
comprising elements defining a chamber, the elements including a body of non-conductive material
having formed therein at least one recess opening into the chamber, the recess containing an electrode.
A pre-treatment coating of a hydrophobic fluid is applied to the body across the recess. Aqueous
solution, having amphiphilic molecules added thereto, is flowed across the body to cover the recess so
that aqueous solution is introduced into the recess from the chamber and a layer of the amphiphilic
molecules forms across the recess separating a volume of aqueous solution introduced into the recess
from the remaining volume of aqueous solution.