Field of the Invention

The invention concerns a process for the manufacture of a biosensor; a biosensor manufactured according to said process; and method for taking a biological measurement comprising the use of said biosensor.

Background of the Invention

A biosensor is generally defined as an analytical device which converts a biological or biochemical response into a quantifiable and processable signal. Quantification of biological or biochemical parameters is commonplace in medical, biological and biotechnological applications. Electrochemical biosensors provide an attractive way to analyse the content of a biological sample due to the direct conversion of a biological event into an electronic signal. Electrochemical biosensors are well known and suitable devices for the diagnosis of analytes, especially in complicated and complex samples including blood, urine and serum. However, the signal transduction and the general performance of electrochemical sensors are often determined by the surface architectures that connect the sensing element to the biological sample at the nanometre scale. Moreover, converting the biological information into an easily processed electronic signal is challenging due to the complexity of connecting an electronic device directly to a biological environment. Amongst biosensors, enzymatic amperometric sensors are the most commercially used devices. The use of enzymes predominates, owing to their specific binding capabilities and biocatalytic activity. In amperometric sensors, the reaction under investigation generates a measurable amperometric current when a constant electrical potential is applied between the working and counter electrodes. Amperometric devices continuously measure the current produced from a series of redox reactions between electroactive species. The first generation of amperometric glucose sensors are heavily dependent on oxygen. The glucose oxidase enzyme (GOx) which naturally exists in its oxidised form reduces after reacting with glucose and oxidises again in the presence of oxygen to produce hydrogen peroxide. The hydrogen peroxide produced will then break down to oxygen and water donating electrons at the positively charged working electrode causing an increase in the amperometric signal. The resulting current flow is then proportional to the analyte concentration.

The development of second generation biosensors has led to the introduction of reversible oxidising reagents known as redox mediators, which can react rapidly with the enzyme and act as an electron acceptor/donor therefore removing the need for oxygen in the system (see equations 1 and 2):

ki

GOx _(0X) + glucose ® GOx _(red) + gluconic acid 1

The redox mediators provide three main advantages such as enhancing the selectivity of the sensors, independence of oxygen and less production of hydrogen peroxide which can damage the enzyme.

The redox mediated biosensors also show a better sensitivity than the non- mediated ones due to enhanced electron transfer throughout the sensing layer.

To make sensitive, reproducible and reliable sensor devices, the fabrication of a homogeneous sensing layer of uniformly mixed sensing compounds with good control over the thickness and density of the sensing layer is critical and challenging. Electrochemical growth, chemical binding and physical adsorption are three common methods used for the immobilisation of the sensing material onto the working electrode. Amongst the three methods, chemical binding and physical absorption are less complex and more cost effective for mass production. Moreover, in the case of microneedle sensor arrays the covalent binding of the sensing material within the sensing layer is particularly important where a robust sensing layer is required to stand physical stress due to skin insertion. In terms of sensor fabrication, drop casting is the most commonly used method for the deposition of the sensing compounds in the sensing layer. However, this method may lead to a ‘coffee ring’ effect and non-uniform/non- homogeneous distribution of the drop cast compounds on the surface especially when complex 3D structures are being coated. Further, in many applications the sensor compounds used to form the sensing layers are not suitable for routine manufacture due to their poor solubility in aqueous and organic solvents. Consequently, a uniform/homogeneous mixture of such compounds is not achievable via conventional drop casting or even a spray layering methodology (see figure 1 ).

In this work, we disclose a novel spray technique used to mix sensing compounds into a uniform/homogeneous sensor layer in a reproducible manner. As proof of principle, we demonstrate this by preparing a glucose monitoring sensor comprising tetrathiafulvalene (TTF) mediator and glucose oxidase (GOx) enzyme and show uniform homogenous mixing and layer deposition of these two compounds. Our technique, exemplified using the simultaneous spray deposition of TTF and GOx compounds on a sensor substrate, maximises the current sensitivity response to glucose compared to the use of a conventional drop casting manufactured sensor. Further, advantageously, our spray coating technique is suitable for coating complex 3D sensor structures, such as microneedle arrays, which is not possible via low cost techniques. Also, our spray coating can provide good sensor reproducibility and control over the sensing layer thickness by changing the volume of the sprayed components: e.g. GOx and TTF, and the density of the sprayed layer by varying the concentration of the material used. Advantageously, such parameters can be used to tune the linear range of the sensors.

Statements of Invention

According to a first aspect of the invention there is provided a process for the manufacture of a biosensor comprising: a) suspending at least one first sensing component and/or a linker for immobilising said or a component(s) onto the surface of a sensor substrate in a first solution;

b) suspending at least one further sensing component and/or a linker for immobilising said or a component(s) onto the surface of a sensor substrate in a second solution wherein said further sensing component and/or the linker when in solution is immiscible with said first sensing component and/or the linker when in solution;

c) simultaneously spraying said first and second solutions onto the surface of a sensor substrate;

d) optionally, suspending at least one further sensing component and/or a linker for immobilising any one or more of said components onto the surface of a sensor substrate in either said first solution or said second solution or a further solution and spraying same onto the surface of the coated sensor substrate of part c);

e) drying said coated sensor of part c) or part d) to provide a sensor coated with a sensing layer comprising a homogeneous mixture of said components wherein said components are immobilized onto the surface of a sensor substrate,

wherein said linker is provided in at least one of said first and second solutions.

In a preferred embodiment of the invention, said components form part of a biological reaction or a biochemical pathway. Ideally, said components comprises all but at least one component of the biological reaction or the biochemical pathway with the remaining component(s), when present in a sample, participating in the said reaction or pathway, when said sensor is in contact with said sample, to complete the reaction or pathway and so yield a measurable parameter. This parameter is typically, but not exclusively, detected as electrical current flow. Thus, according to a preferred embodiment of the invention said biosensor is an amperometric biosensor. In this instance, said sensor substrate of part c) comprises an electrically conducting part or layer onto which said solutions are coated. Further, said sensor also includes a counter electrode and, ideally a reference electrode. We prefer to use an Ag/AgCI reference electrode and a gold wire counter electrode.

In a further preferred embodiment of the invention part e) involves drying the sensor in a vaporised linker such as GA vapour to further enhance the crosslinking of the sensing layer.

In yet a further preferred embodiment of the invention said first solution is an inorganic solution, preferably water, and said second solution is an organic solution, preferably an alcohol such as ethanol, or vice versa. In some embodiments the inorganic solution is a polar solution and/or the organic solution is a non-polar solution.

Said linker is provided in at least one of said first and second solutions and is ideally provided in said first or second solution, although it can be provided in both. Preferably, said linker is a conventional linker and so one that is known to link any one or more of said components with said substrate or each other with a view to immobilising them on the substrate. Examples of linkers suitable for working the invention include Glutaraldehyde (GA), N-Hydroxysuccinimide (NHS) esters, imidoesters, DSG (disuccinimidyl glutarate), DFDNB (1 ,5- difluoro-2, 4-dinitrobenzene), BS3 (bis(sulfosuccinimidyl)suberate), TSAT (tris- (succinimidyl)aminotriacetate), BS(PEG)5 (PEGylated bis(sulfosuccinimidyl)suberate), BS(PEG)9 (PEGylated bis(sulfosuccinimidyl)suberate, DSP (dithiobis(succinimidyl propionate)), Lomant's Reagent, DTSSP (3,3'-dithiobis(sulfosuccinimidyl propionate)), DST (disuccinimidyl tartrate), BSOCOES (bis(2-

(succinimidooxycarbonyloxy)ethyl)sulfone), EGS (ethylene glycol bis(succinimidyl succinate)), DMA (dimethyl adipimidate), DMP (dimethyl pimelimidate), DMS (dimethyl suberimidate), DTBP (dimethyl 3,3'- dithiobispropionimidat; Wang and Richard's Reagent), and other known protein crosslinking reagents. In a preferred embodiment of the invention, said sensor substrate is selected from the group comprising: silicon, solid polymers such as polyimide, SU-8 type material, injection mouldable polymers, such as polyether ether ketone (PEEK) and conductive acrylonitrile butadiene styrene (ABS), paper, and flexible material such as rubber silicon-based gels or other gel type materials.

In yet a further preferred embodiment of the invention said first and second solutions are sprayed onto said surface at the same rate, alternatively, said solutions are sprayed at different rates and in either case the relative amounts of said components is selectively controlled.

Additionally, or alternatively, the concentration of each component in either said first or second or further solutions is selected so as to provide an amount of said component that is representative of its concentration in the natural biological reaction or the biochemical pathway. Alternatively, where the kinetics of the biological reaction or the biochemical pathway are to be manipulated, to generate an improved measurable parameter, at least one, and possible more than one or all of the components, are provided in said solutions in an amount that is selected so as to maximise the generation of the measurable parameter.

Thus, the relative proportions of the said first and the said further sensing components are adjusted in each of said solutions so that the amount of each component deposited on the sensor substrate is controlled to maximise sensor functionality. For example the amount of mediator to enzyme may be increased to enhance the function of said enzyme; and/or the amount of enzyme to substrate may be increased or vice versa; and/or the amount of linker to mediator or enzyme may be increased. Further step d) may be always part of the method and the solution of step d) may always contain linker as we have discovered that a final coat of linker immobilises the components on the substrate and so improves function. In a preferred method of the invention each solution is sprayed using a different spray gun, alternatively a single gun with multiple nozzles may be used.

In yet a further preferred embodiment of the invention said method further includes insulating at least a part of the coated sensor with an electrically insulating material, ideally, by the application of an insulating material to said substrate. This application of an insulating material serves to electrically isolate the backplane of the sensor. Preferably, said insulating material is an organic material that is solid at room temperature, such as a wax, ideally paraffin wax. Thus the method further involves, heating a wax until it forms a liquid and then dipping a part of said sensor in said wax or applying said molten wax to a part of said sensor whereby at least a part of said sensor is provided with an insulating material. Typically, said wax is applied to a rear part of said sensor i.e. the part not participating in the biological reaction or the biochemical pathway.

In this work paraffin wax was applied to electrically isolate the backplane of the sensor. Paraffin wax is low cost, and easy to process due to its relatively low viscosity in a liquid state. Also, the low melting point of wax is greatly advantageous compared to many other insulating agents that may require high curing temperatures for long periods which would denature biological enzymes, whereas wax melts and hardens at low temperatures in seconds. Further, other insulating agents normally possess high viscosity making them difficult to handle. Finally, in its solid state paraffin wax is a great electrical isolator, water resistant and non-hazardous when in contact with skin. These characteristics make wax an ideal candidate as a backplane isolator for the mass production of sensors.

In yet a further preferred embodiment said sensor comprises a plurality of microneedles which are coated with said components. More preferably said microneedles comprise an electrically conducting part or layer onto which said solutions are coated. Preferably, an electrically insulating material is applied to the coated sensor at the base of said plurality of microneedles to electrically isolate the backplane of the sensor.

In yet a further preferred embodiment of the invention said at least one first sensing component comprises at least one reversible oxidising reagent also known as a mediator. Alternatively, in yet a further preferred embodiment of the invention said at least one first sensing component comprises at least one enzyme.

In yet a further preferred embodiment of the invention said at least one further sensing component comprises at least one reversible oxidising reagent also known as a mediator. Alternatively, in yet a further preferred embodiment of the invention said at least one further sensing component comprises at least one enzyme.

Most suitably said first solution comprises said mediator and said second solution comprises said enzyme.

As would be appreciated by the skilled artisan, the enzyme component of the first and/or second solution is selected depending on the intended use of the biosensor end product. In preferred embodiments, GOx is selected as the enzyme for a glucose biosensor. However, alternative enzymes suitable for working the invention include, but are not limited to, lactate oxidase (for the detection of lactate), lactate dehydrogenase (for the detection of lactate), uricase (for the detection of uric acid), glutathione oxidase (for the detection of glutathione), glutamate oxidase (for the detection of glutamate) and glutaminase (for the detection of glutamine), or the like.

As would also be appreciated by the skilled artisan, the mediator component of the first and/or second solution can be any conventional reversible oxidising reagent that can react with the enzyme and act as an electron acceptor/donor. The use of a combination of such conventional mediators is also envisaged. Examples of mediator components that are suitable for working the invention include, but are not limited to, tetrathiafulvalene (TTF), ferrocene or derivatives thereof such as, for example, dimethylferrocene or ferrocenecarboxaldehyde; Tetracyanoquinodimethane (TCNQ); Conducting salts such as, for example, Tetrathiafulvalene-Tetracyanoquinodimethane (TTF-TCNQ), N- methylphenazinium-tetracyanoquinodimethane (NMP-TCNQ), and N- methylacridinium-tetracyanoquinodimethane (NMA-TCNQ); quinones; Ferrcyanide; and Ferrocyanide. In a preferred embodiment, the mediator component is tetrathiafulvalene (TTF).

Said first and/or said second solution also comprise(s) a linking agent whereby said at least one first component and said at least one further components are immobilised onto said sensor substrate.

More preferably still, said first solution comprises mediator tetrathiafulvalene (TTF) and said second solution comprises glucose oxidase (GOx) enzyme. Further, said linker preferably comprises glutaraldehyde (GA).

In a more particular embodiment of the invention said first solution comprises a mixture of said TTF and GA in an organic solution such as an alcohol e.g. ethanol, preferably said second solution comprises GOx in an inorganic solution such as water. Most suitably, said first solution comprises 300 to 600 pL TTF (40mM) in ethanol plus 1 .8 ml_ pl_ GA (5%) in ethanol to make a total of 2.1 to 2.4 ml_. Most suitably also, said second solution comprises 1200 mI_ GOx (10 mg/mL) in deionised (Dl) water.

Where part d) of the invention is practised, said further sensing component comprises a linking agent and so is ideally 1 .2 ml_ GA (5%) in Dl water.

According to a further aspect of the invention there is provided a sensor manufactured according to the above process. Such sensors comprise a sensing layer that comprises a homogeneous mixture of sensing components immobilized onto the surface of a sensor substrate. According to yet a further aspect of the invention there is provided a sensor comprising a homogenous mixture of sensing components immobilized onto the surface of a sensor substrate, manufactured according to the above process for use as a biosensor, for example, as a glucose sensor.

According to a yet further embodiment of the invention there is provided a microneedle array comprising a sensing layer that comprises a homogeneous mixture of sensing components immobilized onto the surface of said array, adapted to perform or function as a biosensor wherein said array has been exposed to the manufacturing process of the invention.

Most preferably, said sensor or said microneedle array has coated on at least a part thereof said components whereby said components are mixed there together homogeneously so as to maximise the measurement of said measurable parameter. Ideally, said components are mixed as herein described and so have the properties herein disclosed.

According to a yet further aspect of the invention there is provided a method for taking a biological measurement comprising the use of said biosensor or said microneedle array.

Preferred features of each aspect of the invention may be as described in connection with any of the other aspects.

Throughout the description and claims of this specification, the words “comprise” and“contain” and variations of the words, for example“comprising” and“comprises”, mean“including but not limited to” and do not exclude other moieties, additives, components, integers or steps. Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise. All references, including any patent or patent application, cited in this specification are hereby incorporated by reference. No admission is made that any reference constitutes prior art. Further, no admission is made that any of the prior art constitutes part of the common general knowledge in the art.

Other features of the present invention will become apparent from the following examples. Generally speaking, the invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including the accompanying claims and drawings). Thus, features, integers, characteristics, compounds or chemical moieties described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein, unless incompatible therewith.

Moreover, unless stated otherwise, any feature disclosed herein may be replaced by an alternative feature serving the same or a similar purpose.

The Invention will now be described by way of example only with reference to the Examples below and to the following Figures wherein:

Figure 1. Shows a schematic diagram of different sensor surface coating techniques: the drop cast in layers (DL), sprayed in layers (SL), and spray mixing (SM) of the sensing compounds;

Figure 2. Shows scanning electron microscope (SEM) images (left) and energy-dispersive X-ray spectroscopy (EDX) maps of sulphur (right) showing the planar surface of the DL, SL and SM fabricated sensors;

Figure 3. Shows the cross-sectional EDX analysis for a surface coated using a) SM and b) SL methods;

Figure 4. Shows the amperometric response to glucose using the a) DL b) SL and c) SM sensors and d) the average Lax and KM values for DL, SL, and SM sensors; Figure 5. Shows the sensitivity of a sensor with standard deviations (a), typical selectivity via cyclic voltammetry (b), typical selectivity using amperometric response at 0.2V working potential (c), and the average selectivity using amperometric response at 0.2V with n=2;

Figure 6. Shows a schematic diagram of the sprayed layers on a typical working electrode;

Figure 7. Shows an image of a carbon sprayed electrode of a microneedle array (left) compared to a full working electrode with waxed backplane (right);

Figure 8. Shows the sensitivity of a glucose sensor manufactured in accordance with the invention when using different TTF volumes as well as an increased amount of GA as a measure of the sensor performance;

Figure 9. Shows the sensitivity of a glucose sensor manufactured in accordance with the invention when using different GOx volumes as well as an different amounts of GA as a measure of the sensor performance .

Figure 10. Shows the sensitivity of a glucose sensor manufactured in accordance with the invention when part d) is practised using two different GA volumes in the part d) spray.

Table 1. Shows the performance of the sensors of the invention when compared to a number of TTF mediated enzymatic glucose sensors disclosed in the literature that were manufactured using the conventional drop cast method.

MATERIALS AND METHODS

Materials

Aspergillus Niger GOx (270 U/mg) was purchased from BBI solutions Ltd, UK. TTF, 0.01 M sterile phosphate buffer solution (PBS used as the supporting electrolyte for the electrochemical studies), D glucose anhydrous, and extra pure deionised water were purchased from Fisher Scientific, UK. Glutaraldehyde (25%) and diacetone alcohol were purchased from Sigma Aldrich, UK. Graphene based carbon ink (HDPIas® IGSC02002) was purchased from Haydale, UK. Iwata Eclipse CS spray guns were purchased from Airbrushes UK. All the reagents were of analytical grade and used without further purification.

A stock solution of 40 mg/mL GOx was prepared in deionised (Dl) water by weighing GOx powder at room temperature. This was further diluted to 10mg/ml_ so that it could be sprayed without blockage at the nozzle. Glutaraldehyde stock solutions of 5% were prepared in Dl water and ethanol. All of the above-mentioned solutions were stored at 4°C. 40 mM TTF stock solutions were prepared in ethanol and stored at -20°C. Before use, the TTF solution was ultrasonicated for 10 min and vortexed re-dissolving the TTF crystals in the solution. 3M glucose stock solution was made in PBS and left for at least 24 hours at 4°C to allow mutarotation.

Fabrication of working electrodes

The glucose sensors were fabricated on flexible polyimide substrate by spray coating a conductive carbon layer and subsequently depositing the sensing compounds.

To produce the conductive layer of the working electrodes, 0.15 mm thick polyimide sheets were cut into 1 .4 x 0.7 cm rectangular shapes; 5 g of the carbon paste was mixed in 20 g of diacetone alcohol to achieve a 4: 1 solvent to carbon paste ratio and spray deposited to form the conductive layer for the working electrodes. After spray deposition, carbon electrodes were annealed on a hot plate at 250°C for 10 min. Resistance of the conductive layer was measured using a standard multimeter which was approximately 200 W across 1 .4 cm along the electrodes. The conductive carbon electrodes were coated with TTF and GOx sensing compounds via three different fabrication methods (Figure 1 ): drop casting TTF and GOx in subsequent layers (DL method); spraying TTF and GOx in subsequent layers (SL method); and spray mixing of the TTF and GOx compounds simultaneously (SM method). In all three cases the last layer, involving GA, was applied to ensure that the GOx was well cross-linked especially at the surface; consequently, this layer was not studied as a part of the sensing layer in the current study. Distance between the spray outlet and the samples in all the three cases was 13.2 cm. Spray rate for GOx and GA (both in Dl) was found to be optimum around 1 pL/sec which was the fastest spray rate without forming any visible droplets (wetting) on the surface. Spray rate of TTF/GA (in ethanol) was adjusted to allow simultaneous spray of both compounds over the same period of time.

DL Sensors

A 3x3 mm window was created on the carbon electrodes using paraffin wax. Subsequently, 1 .13 pl_ TTF (4 mM) in ethanol, 1 .42 mI_ GOx (20mg/ml_) in Dl water, and 2.47 mI_ GA (5%) in Dl water were drop-cast one at a time onto the carbon window under intensive ventilation. These volumes were chosen to achieve equal quantities of GOx and TTF for all three types of sensors to allow direct comparison between them. Sensors were allowed to completely dry between each layer and were then left overnight at 4°C to maximise cross- linking of the GOx and GA. Before performing the amperometric measurements these sensors were rinsed thoroughly with Dl water to remove any loosely bound compounds and blow dried with nitrogen.

SL Sensors

Regarding the SL sensors, the carbon electrodes were spray coated with 1 .2 mL TTF (20mM) in ethanol followed by 450 pL GOx (13.33 mg/mL) in Dl water and finally 1 .05 pL GA in Dl water. Fabricated sensors were then left at 4°C overnight for GOx and GA to crosslink and for the sensing layer to dry. Sensors were then rinsed with Dl water and blow dried with nitrogen. A 3x3 mm window was then created using paraffin wax before sensor testing. SM Sensors

Fabrication of the SM sensors was carried out in two steps. In the first step, two solutions (i.e. solutions A and B) were simultaneously spray coated onto the carbon electrode from two spray guns. Solution A consisted of 600 pL TTF (40mM) in ethanol, 450 pl_ GA (5%) in ethanol and 150 mI_ of ethanol to make a total of 1 .2 ml_. Solution B consisted of 450 mI_ GOx (40 mg/mL) in Dl water. In the second step 1 .05 ml_ GA (5%) in Dl water was sprayed. The same amount of GA was deposited as the outer layer for all three cases (i.e. DL, SL and SM).

The quantities of GOx, TTF, and GA used in SL and SM sensors were determined via UV Vis absorption and used to fabricate the DL sensors. Based on the UV results 9 pg TTF, 8 active units GOx and 2.47 pL GA (5%) were deposited within the active region of each sensor for the SM, SL, and DL sensors. An extra 450 pL GA (5%) was mixed with TTF to enhance the immobilisation of the GOx and TTF for the SM case.

Backplane Electrical Isolation of Sensors

It is very important to electrically isolate the backplane of the (e.g. microneedle) sensor. This is to avoid interference with the amperometric signal due to perspiration at the surface of the skin. Here we disclose the use of paraffin wax to cover the backplane area of the (e.g. 3D microneedle) sensor. Paraffin wax is low-cost and easy to process due to its low viscosity (below 3 mpa.s at 75 °C) and relatively good wetting (contact) on the sensing materials.

After the rinse step the 3D microneedle working electrodes are slowly lowered into hot paraffin wax (about 75° C) until the backplane is completely covered with wax while the needles remain unwaxed. At this point the sensor is gently moved out of the wax and cooled down until the backplane wax is in solid state. At this point the working electrodes are ready for sensor testing. Figure 6 shows all the layers sprayed on a microneedle (complex 3D) working electrode. In Figure 7 a typical carbon sprayed electrode and a full working electrode with waxed backplane are compared together.

Sample preparation for SEM and EDX studies

SEM imaging and EDX analysis were performed using a Hitachi S-4800 SEM equipped with an Oxford instruments EDX detector for topographical and elemental analysis respectively. UV-Vis absorption was performed using a Hitachi U2900 to determine the quantity of the spray deposited compounds.

The uniformity of the TTF and GOx in the fabricated sensors was studied using SEM and EDX techniques. All studies were performed on the sensing layer deposited onto carbon coated silicon substrates, instead of polyimide substrate to avoid interference in the EDX carbon peak from the polyimide and to minimise charging effects. GOx and GA quantities for all samples were kept the same in the DL, SL and SM sensors. For the cross-sectional SEM and EDX studies, the deposited TTF was increased by 10-fold to enhance the EDX signal from sulphur in TTF for both SM and SL samples to a detectable range in order to study the uniformity throughout the sensing layer. Also this allowed a much thicker TTF layer which could be observed more clearly in the SEM images. The carbon ink was not deposited for the cross sectional studies to avoid interference in the EDX signal on the carbon especially in the SM case.

Electrochemical cell setup

Amperometric measurements were performed using an Ivium CompactStat from Alvatek and an electrochemical cell consisting of a fabricated working electrode (area 0.09 cm ² ), an Ag/AgCI reference electrode and a gold wire counter electrode at 0.3 V which is the established oxidation potential for TTF mediator [1,2] All three electrodes were immersed in 10mL of sterile PBS buffer. After applying the electrical potential, the cell was allowed to stabilise electrochemically until the change in the electrical current was smaller than 100 pico amps/sec. The amperometric measurements were performed whilst stirring at 350 rpm at which rate the noise associated with the magnetic stirrer was minimised to <5 nA.

RESULTS

Sensing Layer Characterisation

The following section investigates the uniformity of TTF throughout the sensing layer to support the postulations made in the section above. The distribution of the sensing elements, i.e. TTF and GOx, at the surface of the electrodes (planar surface) and through the sensing layer (cross section) was investigated. In the following figures, the presence of TTF in the sensing layer is indicated by observing the sulphur EDX peak. The biological components, GOx and GA are indicated by the presence of carbon and oxygen. TTF also contains carbon in its molecular structure; however, the EDX results on the SL sample in Figure 3 b shows negligible carbon peak in the pure TTF region on the right-hand side of the image. Also, there is a good correlation between the carbon and oxygen was observed in Figure 3 b which confirmed the carbon peak was dominated by the presence of GOx and GA rather than TTF. The Si peak is due the underlying silicon substrate in all EDX results.

Planar Surface Analysis

The uniformity of the deposited sensing elements across the surface for DL, SL, and SM sensors was studied via SEM imaging and EDX analysis. Figure 2 shows the SEM images and EDX mapping analysis for sulphur (i.e. associated with TTF) at the planar surface in typical DL, SL and SM samples. The EDX map for the DL samples exhibits much stronger Sulphur signal intensity near the edges of a waxed window which indicates the presence of a ‘coffee ring’ effect. In addition, the formation of the TTF crystals, especially around the edges of a silicon substrate, was observed by an optical microscope. Around the edges of the sensing window, TTF crystals are highlighted by a red circle in the EDX map in figure 2.

On the contrary to the DL samples, a uniform distribution of the TTF is observed for the SL and SM samples as shown in figure 2; which explains the significantly improved performance of the SL and SM sensors compared to the DL sensors. The random formation of the TTF large crystals at the surface would significantly decrease the active interface between TTF and GOx, hence restricting the redox reaction between the two elements. From this, the lack of uniformity and hence proximity between the TTF and GOx in DL sensors has resulted in the significantly small Lax values observed in figure 4a. Furthermore, the significant sensor to sensor variation could be explained by the random formation of the large crystals in the sensing layer.

The planar distribution of carbon and oxygen was also studied and showed a corresponding decrease in carbon at the edges where the TTF was accumulated.

Cross-Sectional Analysis

The cross-sectional EDX analysis allowed the study of the mixing of the biological component and the mediator throughout the sensing layer for the SM sample in comparison to the SL sample (figure 3a and 3b, respectively). The DL sample in contrast led to formation of TTF large crystals and non- uniform distribution of TTF across the surface. The EDX results for SM sample in figure 3 a show a uniform distribution of sulphur (representing TTF), carbon and oxygen (both representing the biological components) from top to bottom of the sensing layer (i.e. left to right of the image). For the SL sample, figure 3 b shows strong EDX signal for carbon and oxygen (associated with the GOx and GA) at the top of the sensing layer (i.e. left hand side of the image). As the line profile reached the TTF layer, the EDX signal for carbon and oxygen dropped almost to zero while the signal for sulphur reached maximum. This confirmed the layer by layer formation with GOx and GA compounds as the top layer and TTF as the bottom layer for the SL sample. This indicates a limited contact area between the GOx and TTF compared to the uniform mixture of the TTF and GOx in the SM sample. For both SL and SM samples as the line profile reaches the silicon substrate on the right hand side of the images, in the EDX signal for C, 0, and S decreases and Si increases. The uniform mixing of the TTF mediator and the biological components (i.e. GOx and GA), which are typically immiscible, throughout the sensing layer makes the SM fabrication process ideal for high performance sensors. In addition the technique is relatively simple, low cost and scalable to allow market demands for such devices to be met.

Amperometric sensor performance

The amperometric response of DL, SL, and SM glucose sensors shown in figure 4 was studied against successive additions of glucose until the current response reached a plateau.

In figure 4 all sensors response curves reach a plateau at saturation point showing the characteristics of the Michaelis-Menten kinetics. Previous work has showed the Michaelis-Menten equation can represent the amperometric response of the enzymatic sensors. Figure 4 illustrates the amperometric response of the DL, SL, and SM glucose sensors by means of the average Lax and KM parameters representing maximum amperometric current and the Michaelis-Menten constant respectively.

The DL sensors exhibited the smallest amperometric response to glucose compared to both the SL and SM sensors. The weak performance of the DL sensors (especially the average Lax) shown in figure 4a would suggest the lack of redox reactions between the GOx and the TTF mediator given their quantities were the same as SL and SM sensors. The SL sensors in figure 4b show a larger average Lax (6.5 times) and KM (250 times) than the DL sensors. This significant increase in the performance of the SL sensors (especially the average Lax) could indicate the enhanced redox reaction as a result of better proximity between the enzyme and the mediator due to the spray coating technique. The SM sensors in figure 4c show a much larger average Lax (20 times) and KM (12 times) than the DL sensors. The average Lax of the SM sensors was also about 3 times larger than the SL sensors. Furthermore, the SM sensors exhibited the smallest sensor to sensor variation as compared to the DL sensors which showed the largest variation for Lax and KM parameters. This is evident in the standard deviations of the mean Lax and KM values shown in figure 4d. The KM parameter is (21 times) larger for the SL sensors compared to the SM sensors (figure 4d).

Two phenomena could account for the observed increase in the Lax for the SM spray coated sensors. Firstly, the closer proximity between the TTF and GOx enzyme as a result of better mixing enhances the redox reaction and the amperometric signal. Secondly, the improved electron transfer via the TTF throughout the sensing layer increases the current signal. Without wishing to be constrained by theory, we consider the total mass of the sprayed mediator and the sprayed area was similar for both SM and SL sensors, the concentration (mass/volume) of the immobilised mediator that actively interacted with each enzyme molecule would be proportional to the mixing of the enzyme and mediator. Consequently, the concentration of the redox mediator in reaction with the enzyme, [M] ^* would be much larger for the SM sensors as compared to the SL sensors which in theory would result in a larger current response. This is in good agreement with the amperometric response observed in figure 4b, c for the SM and SL sensors.

Unlike previous studies that have failed to improve the sensor I max by conventional mixing and drop casting methods, this study shows a significant improvement in the Lax parameter due to spray mixing of the mediator and enzyme. Without wishing to be bound by theory we consider the significantly better Lax observed in the SM sensors is due to a vastly improved enzyme- mediator interface as a result of uniform distribution of TTF/GOx throughout the sensing layer.

Sensitivity, selectivity and stability

The sensitivity of the SM glucose sensors were studied and compared to the SL and DL sensors. The selectivity and stability of the SM sensors were also studied. As observed in figure 4 the DL sensors showed the least performance in terms of lma _X as well as a very large sensor to sensor variation. In figure 5a sensitivity of the SL sensors was very small compared to the SM sensors and the smallest r ² value with the largest standard deviation was observed. From both studies we can confirm that the DL sensors are not suitable for biosensing applications mainly due to their inconsistency between the sensors and they will no longer be discussed in this study.

As shown in figure 5, the sensitivity of the SM TTF mediated glucose sensors of the invention are better than some of the sensors developed by other groups (some of which involve the use of nanomaterials) as stated in Table 1 . The r ² value for the linear regression was larger than 0.9 for the SM and SL cases. The SM sensors possess an average sensitivity of 38.7uA/mM/Cm ² , which compares well with the previously reported sensors in the literature. The KM value on the other hand is smallest for the SM sensors showing high affinity for the enzyme glucose reaction.

Sensing layer optimisation (for SM sensors)

Key:

V is volume or part expressed in spray volume

1V GOx: 600uL, 10mg/ml GOx

1V TTF: 600ul, 40mM TTF

1V GA: 600ul, 5% GA

Figure 8 shows that an increase in sensitivity (slope of the curve at zero mM of glucose) was achieved as a result of increasing TTF deposition during the simultaneous spraying of a solution of GOx and a solution of TTF when 1V, 2V, and 3V TTF was deposited keeping the GOx and GA constant. Also it was observed that an increase in GA mixed with TTF increased the sensitivity for the 3Vol TTF case. Any more TTF would cause the sensing layer to become too fragile to withstand the skin insertion process. Figure 9 shows the increase in sensitivity as a result of increasing GOx deposition during the simultaneous spraying of GOx and TTF where 0.5V, 1V, and 2V GOx & GA are mixed with 1 V of TTF. As can be seen the introduction of more GOx affects the performance of the sensors but not as much as varying the amount of mediator i.e. the TTF. Also, increasing the amount of GOx causes the sensors response to slow down which is a down fall. The response time at 2V GOx is about 2 mins which is on the border of acceptable. Therefore, more GOx was not investigated here. This consideration is especially important for in-vivo microneedle glucose sensors where there is a natural lag of 8 to 10 minutes between blood and interstitial fluids on average (https://www.ncbi.nlm.nih.gov/pubmed/9699080).

In Figure 10 the change in sensitivity as a result of a final deposition of linker i.e. GA after the simultaneous spraying of GOx and TTF was studied at two amounts of GA, 1V and 2V GA. It was observed that 2V GA improves the sensitivity but any more than that did not have a significant effect on the sensor performance.

All of the spray deposition volumes are reported specifically for the spray setup reported in this work. A change in the spray distance (from the nozzle to the sample surface), and the air pressure will require re-optimisation of the spray.

Based on these studies the optimum spray parameters are 2V GOx sprayed simultaneously with a mixture of 3V TTF and 2V GA. Additionally, 2V GA is sprayed after the simultaneous spraying of components to ensure an excess of GA is present on the surface for covalent crosslinking of the GOx based sensing layer.

Summary

This work has shown the successful use of a novel spray coating technique for the simple, low cost and mass reproducible fabrication process to produce a uniformly mixed sensing layer from immiscible compounds. As proof of principle, this was demonstrated using GOx and TTF to maximise the sensitivity of a mediated glucose sensor. The SEM and EDX studies showed that the two compounds were distributed evenly throughout the sensing layer in both planar and cross-sectional trajectories. The amperometric measurement studies confirm that uniform mixing of the two compounds enhanced the L _ax significantly compared to other conventional deposition methods with also an advantageous improved sensitivity. The sensitivity of a sensor, when processed in this manner, is significantly improved compared to sensors fabricated by other means. The uniform distribution of the TTF and GOx throughout the sensing layer correlated well with the L _ax in other words, the uniformly distributed mediator and enzyme led to enhanced amperometric response to glucose.

Table 1.

References:

1. Chaubey, A. and B.D. Malhotra, Mediated biosensors. Biosensors and Bioelectronics, 2002. 17: p. 441.

2. Kausaite-Minkstimiene, A., et al. , Evaluation of Some Redox Mediators in the Design of Reagentless Amperometric Glucose Biosensor.

Electroanalysis, 2014. 26: p. 1528.

3. German, N., et al., The use of different glucose oxidases for the development of an amperometric reagentless glucose biosensor based on gold nanoparticles covered by polypyrrole. Electrochimica Acta, 2015. 169: p. 326.

4. Gierwatowska, M., et al., Multifunctional Mediating System Composed of a Conducting Polymer Matrix, Redox Mediator and Functionalized Carbon Nanotubes: Integration with an Enzyme for Effective Bioelectrocatalytic Oxidation of Glucose. Electroanalysis, 2013. 25: p. 2651.

5. Campuzano, S., et al., Preparation, characterization and application of alkanethiol self-assembled monolayers modified with tetrathiafulvalene and glucose oxidase at a gold disk electrode. Electroanalytical Chemistry, 2002. 526: p. 92.