WO2014142749A1 - Microneedles integrated with nanotubes - Google Patents

Abstract

The invention relates to a microneedle for transdermal drug delivery comprising, at or adjacent its tip, a nano-filter in the form of a collection of vertical nano-tubes; a composite comprising a plurality of said microneedles; a skin patch comprising a plurality of said microneedles; a method of treatment involving said microneedles or said skin patch and a method for the manufacture of said microneedles.

Description

MICRONEEDLES INTEGRATED WITH NANOTUBES

Field of the Invention

The invention relates to a microneedle for transdermal drug delivery; a composite comprising a plurality of said microneedles; and a method for the manufacture of said microneedles.

Background of the Invention

Generally, drugs can be delivered into a human body via several routes, such as injection and oral delivery. However, administration by injection cannot be easily conducted by patients as it often requires training and, if carried out incorrectly, may result in pain. Additionally, oral delivery may be inappropriate for the administration of certain drugs as the pharmacological action of medicines can be affected by digestive enzymes, with protein and DNA targeting drugs having poor absorption rates when administered in this way.

To overcome these drawbacks, transdermal drug delivery using micro- electromechanical system (MEMS) based microneedles has drawn much attention. MEMS involves the use of very small devices, generally ranging in size from 20 micrometres to a millimetre. In order to penetrate the skin's outmost layer (i.e. stratum corneum; SC) the manufacture of microneedles from various materials, including silicon, stainless steel, titanium, tantalum, nickel and ceramics has been investigated. These materials are used to achieve the requisite mechanical strength, which needs to be capable of being inserted into and penetrating the skin in order to reach the desired tissues. Although microneedles can be fabricated to provide a sharpened end for easier penetration, microneedles may be damaged during the insertion process due to their frangibility. Moreover, most of them have unproven biocompatibility. To overcome this, polymer-based microneedles have been developed. They have favourable properties due to their inherent flexibility and biocompatibility. The fabrication of MEMS devices has evolved from the process technology used in semiconductor device fabrication, i.e. the basic techniques involve the deposition of material layers, patterning by photolithography and etching to produce the required shapes. Lithography in an MEMS context is typically the transfer of a pattern into a photosensitive material by selective exposure to a radiation source such as light. In the context of microneedles, nano- imprint lithography is used, which refers to a method of fabrication on nanometre scale patterns.

For use in drug delivery, nanometer scale filters of high chemical selectivity and high flux are highly desired. However, pore size matching is one of the most critical fabrication challenges. A good pore size that is selective for desired molecules can allow molecular sieving and force interactions between those bound to the pore. Many approaches regarding the fabrication of nano-scale filters of pore size ranging from 1 to 10 nm have been investigated these include functionalized polymer affinity membranes, block copolymers, mesoporous macromolecular architectures and hollow carbon nanotubes (CNTs).

CNTs are allotropes of carbon with a cylindrical nanostructure, possessing unusual strength and flexibility which lends them for use in nanotechnology devices. Vertically aligned CNT forests can be grown from a macro or micro- patterned catalyst layer. We consider such CNT growth techniques, when compared with other nanofabrication technologies, are easier, cheaper and more suitable for mass fabrication. Advantageously the surface of inner channel of CNTs is smooth and uniform at the atomic level over large distance, which is not achievable by other nanofabrication technologies. We consider this atomic level smoothness and uniformity lead to frictionless motion for mass transport.

However, devices utilizing CNTs are difficult to fabricate with limited reported success. Specifically, whilst possessing high tensile strength and elastic modulus, weak shear interactions between adjacent CNTs lead to significant reductions in the effective strength.

This is particularly important in the context of the development of microneedles, where relative effective strength is imperative for safety and effective delivery of drugs into the body.

Herein disclosed are novel microneedles, ideally fabricated from a stretchable, biocompatible and inexpensive polymer-based membrane layer, comprising nano-filter tips based on small arrays of nanotubes such as, in one example only, vertically grown CNT bundles. The microneedles have improved properties for transdermal-based delivery of molecules or drugs, with particular utility for the selective delivery of nano-scale particles and molecules with applications in protein and vaccine therapy.

Statements of Invention

A microneedle for transdermal drug delivery comprising a nano-filter including a plurality of nano-tubes whose longitudinal axis is aligned with that of said needle for the selective delivery of nano-scale molecules or particles.

Reference herein to a nano-filter is reference to a filter that allows the passage of nano-scale or nano-sized molecules or particles typically understood to be in the order of a nm size and/or blocks the passage of molecules or particles with a size larger than the inner tube diameter of said nanotubes. W 201

Reference herein to a drug is intended to include any agent to be delivered by said microneedle, in particular a drug having a therapeutic or even a preparatory property such as an anaesthetic, but it may also include agents of a diagnostic, prognostic or cosmetic nature. Further, reference herein to a drug includes a gas, liquid or a number of small particles.

Reference herein to a drug is reference to any molecular structure that can be administered transdermal^ and so, without limitation, includes reference to conventional pharmaceuticals and therapeutics but also DNA technologies and vaccine technologies.

Those skilled in the art will appreciate that a microneedle contains at least one and, ideally, a plurality of fluidic channels or microtubes. In a preferred embodiment of the invention said nano-filter is located at or adjacent either end of the microneedle and ideally at the needle tip. Alternatively, said nano-filter is located within the microneedle.

In a further preferred embodiment of the invention said nano-filter is integrated with said microneedle, ideally, integration is achieved by said nano-tubes being grown in or on, or assembled on, said microneedle and/or sealed thereto using either adhesive or the use of at least one plastics material and/or at least one polymer and a heat-cooling cycle. The provision, and ideally the integration, of a nano-filter in or on a microneedle is advantageous because the use of a microneedle as a base for the nano-filter enhances the strength of the whole needle and so ensures it does not break during penetration. Moreover, nano-tubes have a limited length and so the provision of same especially at either end of the microneedle, and ideally on the tip of a microneedle, ensures the whole needle is long enough to penetrate deep into the skin and ideally through the stratum corneum of the skin. Additionally, the use of the microneedle with the nano-tubes, the disclosed microneedle achieves particle selectivity such that only particles of nano-scale can traverse. Advantageously, this reduces the incidence of cross-contamination and transfer of pathogens during injection.

Ideally, the nano-tubes are made from carbon. Due to the unique property of carbon nanotubes, which have at the atomic level a smooth inner channel wall, the transportation efficiency of the microneedles disclosed herein is much higher than other nano-scale fluid channel. Moreover, through incorporation of multiple CNTs at the tip, the efficiency of drug delivery is improved.

In a preferred embodiment of the invention said microneedle is made from one or more of the following materials: silicon, stainless steel, titanium, tantalum, nickel, a ceramic, a biodegradable material and a polymer-based material such as SU-8, polymethyl meth-acrylate (PMMA), polycarbonates (PC) and polylactic acid (PLA). Ideally, the microneedle is made from a polymer-based material and so has the advantage of improved whole needle strength and flexibility of the tip. More ideally still, said microneedle is a SU-8 needle. Alternatively, the microneedle is made from a biodegradable material such as starch and so has the advantage of being biodegradable.

Ideally said microneedle is fabricated into a sharp shape for ease of penetration, preferably having an ultra-sharp tip.

As mentioned, in a preferred embodiment of the invention said nano-tubes are made from carbon, however, they may also be made from silicon or indeed any other chosen material provided they have the requisite nano- scale size and so have an internal diameter that restricts the passage of particles or molecules whose size is greater than a nano-scale. In a preferred embodiment of the invention said microneedle has a dissolvable member, ideally, in the form of a dissolvable tip which, preferably is fashioned into a sharp-shape for ease of penetration. Ideally, said dissolvable tip is bio-compatible and, more ideally, is in the form of a carbohydrate-based tip, such as a sugar based tip for example a maltose tip or a glucose tip, or indeed a tip made from any other sugar. Maltose based microtips are preferred because of their unique features i.e. high mechanical strength and biodissolvability. Alternatively said tip is made from a polyvinylpyrrolidone (PVP). The use of a dissolvable tip is a preferred feature of the invention because after the sugar tip has melted, the micro- tubes of the microneedle efficiently allow the flow of a large volume of a desired drug into the nano-filter.

Notably, so far, there are no published data reporting the integration of dissolvable microneedles with microfluidics.

In yet a preferred embodiment of the invention said microneedle further comprises either an integrated or removable electric field driven member and/or a pressure driven member for facilitating the passage of said drug through the nano-filter of said microneedle. Those skilled in the art will appreciate that the former uses an electric field gradient whilst the latter uses a pressure gradient for the purpose of forcing said drug through said microneedle. Typically, but not exclusively, gas, liquid and small particles are passed through said microneedle using said pressure driven member, whereas proteins, nucleic acid e.g. DNA such as double-stranded or single- stranded and charged particles are passed through said microneedle using said electric field driven member. Ideally, a conventional pressure driven member and/or electric field driven member is used in the production of the invention as herein described. More ideally still, where an electric field driven member is used to drive molecules through the nano-tubes electrodes are positioned, or ideally patterned, onto the nano-tubes to enable an electric field to be applied. Thus a microneedle provided with the afore pressure and/or electric field assisted delivery can be used to deliver large molecular drugs such as insulin since this microneedle can apply force in the form of pressure and/or voltage on the drug solution to facilitate the large molecular drug diffusion process. Moreover, continuous delivery with the microneedle is expected to release large volumes. In this way, the device can support those drugs, including vaccines, that require large doses. According to a second aspect of the invention there is provided a composite comprising:

a) a plurality of microneedles for transdermal drug delivery wherein each microneedle comprises a nano-filter including a plurality of nano-tubes whose longitudinal axis is aligned with that of said needle for the delivery of nano-scale molecules or particles; and b) a supporting membrane for supporting said microneedles.

In a preferred embodiment said membrane of said composite is made of two layers:

i) provided beneath said microneedles is a grid layer including a plurality of holes some of which are aligned with the microneedles; and

ii) provided beneath said grid layer is a sealant layer which seals those holes not aligned with said microneedles.

In a preferred embodiment of the invention said nano-filter is located at or adjacent either end of the microneedle and ideally at the needle tip. Alternatively, said nano-filter is located within the microneedle. Most ideally, each microneedle is provided with a dissolvable tip, ideally made of a carbohydrate such as maltose. In a further preferred embodiment of the invention we provide a mironeedles array with approximately a 1000 pm spacing between adjacent microneedles (as this arrangement has been shown to penetrate porcine cadaver skins successfully). The maximum loading force on the individual microneedle can be as large as 7.36±0.48N and after 9 minutes of the penetration, all the maltose tips were dissolved in the tissue. Drugs could then be delivered via these open biocompatible SU-8 microtubes in a continuous flow manner. Most preferably, our microneedle array comprises microtubes of approximately 350pm height and 1000pm spacing between adjacent microtubes.

According to a further aspect of the invention there is provided a skin patch for transdermal drug delivery comprising:

a) a plurality of microneedles wherein each microneedle comprises a nano-filter including a plurality of nano-tubes whose longitudinal axis is aligned with that of said needle for the delivery of nano-scale molecules or particles; and

b) at attached to said microneedles is a skin contact layer for attaching to said skin.

In a preferred embodiment of the invention said skin contact layer is made from a material that sealingly engages with skin, especially when pressure is applied. Additionally, or alternatively, said skin contact layer comprises an adhesive suitable for attaching said patch to said skin.

In yet a further preferred embodiment at least one, ideally the majority, and more ideally still all of said microneedles contain at least one selected drug. In a preferred embodiment of the invention said nano-filter is located at or adjacent the needle tip. Alternatively, said nano-filter is located within the microneedle.

According to a further aspect of the invention there is provided a method of treatment involving the use of said microneedle(s) or the use of said patch wherein said microneedle(s) loaded with a suitable medicament is/are applied to the skin, ideally using pressure, whereby said microneedle(s) is forced through the skin and said medicament is administered.

In particular, the skin patch can advantageously be used for vaccination purposes in either humans or animals. For example, microneedle patches containing influenza vaccine provide a simple patch-based system that enables delivery of said vaccine to the skin's antigen-presenting cells. Most preferably said microneedles are provided with dissolvable tips.

According to a further aspect of the invention there is provided a method for the manufacture of a microneedle for use in transdermal drug delivery comprising:

a) mounting a plurality of vertical nano-tubes on at least one polymer- based substrate to form an assembly;

b) applying a mask to said substrate for producing a number of

microneedles wherein said mask is positioned such that

microneedles are made beneath or above said nano-tubes and the nano-tube are located at or adjacent a microneedle tip with their longitudinal axis aligned with that of the microneedle;

c) etching said substrate to form a plurality of microneedles.

In a preferred method of the invention said nano-tubes are held in a vertical position by the use of a polymer film such as parylene. j More preferably still, said mounted nano-tubes are attached to said substrate by heating the assembly to a temperature that melts said polymer substrate, causing it to flow and so mix with a first contact end of said nano- tubes and then cooling the assembly so that said polymer hardens.

In a preferred method of the invention said nano-tubes are made from carbon or silicon and, more ideally still, said nano-tubes are grown on said substrate. Ideally, said nano-tubes are carbon. In a preferred method of the invention, optionally, a sub-layer is applied to said assembly.

According to a further aspect of the invention there is provided a method for the manufacture of a microneedle for use in transdermal drug delivery comprising fixedly mounting a plurality of nano-tubes on or in at least one microneedle wherein the longitudinal axis of said nano-tubes is aligned with that of said needle.

In a preferred embodiment of the invention said nano-tubes are located at or adjacent either end of the microneedle and ideally at the needle tip. Alternatively, said nano-tubes are located within the microneedle.

In a preferred embodiment of the invention said nano-tubes are made from carbon, however, they may also be made from silicon or indeed any other chosen material provided they have the requisite nano-scale size and so have an internal diameter that restricts the passage of particles or molecules whose size is greater than a nano-scale.

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.

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

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 in greater detail with reference to the Examples below and to the drawings in which:

Figure 1. (a) 3D schematic drawing of detailed layer structure of a composite according to the invention; (b) SEM picture taken from the rear of the SU-8 layer showing a hole under a CNT bundle on SU-8 layer, (c) 3D schematic drawing of the assembled composite (d) SEM picture of the CNT bundle with exposed CNT fibers and a parylene sidewall; (e) 3D schematic drawing of the CNT bundle with exposed CNT fibers and parylene sidewall.

Figure 2. 3D illustration of the composite. The whole composite excluding PDMS bonding part comprises four layers; CNT bundle array [1], SU-8 needle array [2], supporting the CNT bundle array, a grid layer [3] comprising plurality of holes, and a sealing layer [4], which will seal all the grid holes with the exception of those positioned beneath the SU-8 needles. In use, solution is able to pass through these four layers via the CNT bundle-SU-8 channels.

Figure 3. SEM images of the SU-8 needles with CNT tips.

Figure 4. Fabrication process for the manufacture of parylene encased CNT nanotubes.

Figure 5. Fabrication process for stretchable membrane based microneedles using patterned array of vertically grown carbon nanotubes.

Figure 6. (a) An individual CNT bundle grows from catalyst layer, the diameter of the CNT bundle is 50μηη, the height of the CNT bundle is 50pm; (b) An individual CNT bundle reinforced by parylene, the thickness of the parylene layer is 10pm; (c) catalyst layer seals the bottom of the CNT bundle; (d) The catalyst layer is etched by oxygen plasma; (e) TEM picture showing the inner diameter of the CNT is 10nm.

Figure 7. Electric field driven member and pressure driven member, (a) Preloading test solution in the tube and assembling Ti electrode; (b) Connecting long tube with syringe and air pressure sensor for exerting air pressure; (c) Illustration of the ion depletion in the PDMS chamber in the dash line box in (b). Figure 8. (a) The ionic current through the CNTs as a function of time in different NaCI concentrations; (b) The ionic current as a function of time under square wave bias in different NaCI concentrations. The green line and pink line indicate the difference between peak values of ionic current, (c) The pH value of the NaCI solution with different concentrations as a function of time; (d) The pH value of the HCI solution under different driving pressure as a function of time.

Figure 9. (a) IR spectra of ss-DNA in the beaker after applying 25kPa pressure and after applying both 25kPa pressure and 5V bias; (b) IR spectra of haemagglutinin in the beaker after applying 25kPa pressure and after applying both 25kPa pressure and 5V bias.

Figure 10. Dissolvable sharp tip upon a SU-8-CNT microneedle as shown in Fig 2 or 3.

Figure 11. Schematic illustration of a SU-8 microneedle including a carbohydrate tip. Figure 12. (a) Schematic illustration of the stage to ensure flat SU-8 membrane surface, (b) SU-8 membrane bends after development, (c) After bonded with PDMS and clamped in the stage, the membrane becomes flat.

Figure 13. Fabrication process for maltose tips, (a) Expelling water at 140 °C. (b) Immersing micro-tubes into the maltose at 140 °C. (c) Drawing the tips at end of the micro-tubes when the temperature increases up to 60 °C. (d) Increasing drawing speed to form sharp tips.

Figure 14. Optical image of the finished SU-8 microneedles with a carbohydrate tip.

Figure 15. Penetration testing results on a porcine cadaver skin. Figure 16. Maltose tips dissolving process, (a) The original sharp maltose tip. (b) Maltose tip after inserted into skin for 3minutes. (c) Maltose tip after inserted into skin for 6 minutes, (d) Maltose tip after inserted into skin for 9 minutes.

Figure 17. (a) cross-section of an alternative microtube or microneedle for working the invention, (b) The manufacture of a carbohydrate tip using the microneedle of part (a), (c) The manufacture of an alternative carbohydrate tip using the microneedle of part (a).

Figure 18. Schematic illustration of the manufacture of a starch microneedle.

Figure 19. Microfluidic testing of manufactured SU-8 microtubes.

Figure 20. An alternative fabrication process I for making SU-8 microneedle tubes.

Figure 21. (a) 3D schematic drawing of the microneedle device integrated with CNT nanofilters; (b) Optical image of the microneedle array with gold surface electrode, scale bar is 1000pm; (c) SEM image of single SU-8 microneedle with four-beam sidewalls and a sharp tip, scale bar is 80pm; (d) SEM picture of a CNT bundle embedded inside the microneedle, scale bar is 10μιη.

Figure 22. Working principle of the microneedle array integrated with CNT nanofilters for transdermal drug delivery.

Figure 23. Alternative fabrication process II for microneedle array integrated with CNT nanofilters. Figure 24. (a) Process for single step drawing lithography; (b) Process for double drawing lithography.

Figure 25. Insulin delivery test result: (a) IR spectra of insulin by applying different pressure; (b) The peak value of IR spectra by applying different pressure and bias of electric field.

Figure 26. (a) Result of bacterium culture of the solution sampled from the device chamber, bacterial colonies were observed; (b) Result of bacterium culture of the solution loaded in the beaker, no bacterial colony was found.

Figure 27. Schematic illustration of the SU-8 microneedles, (a) Overview of the whole device; (b) SU-8 supporting structures made of 4 SU-8 pillars; (c) Enlarge view of a single SU-8 microneedle.

MATERIALS AND METHODS

Formation of CNT nanofibres

Thermal oxidation of single crystal silicon substrate was used to form an etch stop oxide layer. After the chemical vapour deposition (CVD) of poly-silicon as a sacrificial layer, a 5 nm thickness of pattered Fe film, which acted as the catalyst film for the selective growth of CNTs, was prepared onto the silicon substrate (Figure 4(a)). As illustrated in Figure 4(b), the vertical aligned CNT bundles of 50pm in height were grown via pyrolysis of acetylene at 800°C with an Ar/NH₃ flow for 15 min. Figure 6(a) shows the SEM picture of the topside view of a grown CNT bundle. The CVD parylene-C was then employed to fill into vertically aligned CNTs to reinforce the inter-tube binding at room temperature.

Thus, the top side of CNTs was covered with parylene-C, and the discrete CNTs were bound together by parylene-C (Figure 4(c) and 6(b)). This step was the most critical process for forming the mechanical supporting layer for CNT bundles. The thickness of the flexible parylene-C layer was determined by the CVD process. To achieve reliable mechanical strength for following handling and process, 10μηη thick parylene layer was deployed. The parylene layer was peeled off together with CNT bundles from the substrate. Since the catalyst layer blocked the bottom ends of CNTs, the catalyst layer at the bottom of CNT bundles (Figure 6(c)) was etched by oxygen plasma from the rear side (Figure 4(d)). Figure 6(d) shows the SEM picture taken from the rear side of the parylene layer for the opening of CNT bundle. Then a prebaked SU-8 layer of 200pm thickness on PET was prepared. PET is a kind of transparent soft film having low adhesion with SU-8. The PET film was fixed on glass slide by tapes to provide a rigid substrate for processing.

Fabrication of SU-8 microneedles

The detailed process flow is shown in Figure 5. A layer of UV tape was attached to silicon wafer and exposed to UV to remove its adhesion to silicon. A kapton film was then attached onto the UV tape. Due to the low adhesion between SU-8 and kapton, during the dry releasing process, the silicon wafer could easily be released from the UV tape first followed by release of the SU-8 component from kapton. This permitted one to obtain a large SU-8 film without cracks and damage (figure 5a). We found that when the SU-8 patterned area on the kapton film is large; tearing off the film from the kapton film without damaging the device requires extreme delicacy. This is because both the patterned SU-8 layer and Si substrate are rigid layers. However, in our improved process, the sticky kapton film applied along the edges of samples to tightly fix with PET film can be easily removed after the device developed in SU-8 developer. The PET film with the SU-8 layer is separated from the Si substrate. Then SU-8 layer could be dry released from the PET film just by slightly bending the PET film.

The film was then spun at 2,000rpm for 30 seconds coated and pre-baked at 65°C for 10 minutes and 95°C for 30 minutes to form the first SU-8 layer (figure 5b), of thickness 100pm. This formed the grid layer. This component was then exposed under 40mJ/cm² UV energy and baked 65 °C for 5 minutes and 95 °C for 15 minutes, but not developed (figure 5c). If the first layer is developed to get patterns, the surface will not be smooth enough to achieve a uniform second SU-8 layer on it. A second layer was then spun and pre-baked to form the second SU-8 layer, of thickness greater than 100pm. This layer forms the SU-8 microneedle, and therefore its thickness determines length of needle (figure 5d). Parylene was then attached to the second layer surface, upon which the CNT bundles were present. The film was then re-baked to make the SU-8 reflow, and cooled to permit strong bonding between parylene and SU-8 (figure 5e), and exposed to 650 mJ/cm2 energy with a mask to form the shape of the SU-8 needle (figure 5f). However, the grid underneath the SU-8 needle and within the SU-8 channel is not developed as they are sealed by the parylene layer and sidewall of the SU-8 needle. The material was then post-baked at 65 °C for 10 minutes and 95 °C for 30 minutes, and the SU-8 developed, to generate the shape of the needles and the grid layer. The SU-8 component is then dry-released from the substrate (figure 5g). A layer of SU-8 was then spun on PET film, which has low adhesion to SU-8 that has not been post- baked (figure 5h). The SU-8 layers were then bonded together at 65°C, and cooled to permit bonding (figure 5i).

Following this, well aligned micro-channels under CNT bundles were patterned by UV lithography to form holes under the SU-8 needles (figure 5j). The diameter of the channel was typically 40pm, smaller than the diameter of the CNT bundles, in order to ensure the SU-8 under parylene sidewalls around CNT bundles was exposed. The bonding between SU-8 and parylene sidewalls provided sufficient mechanical bonding strength to fix the CNT membranes on SU-8 substrate. Figure 1 (b) shows the SEM picture taken from rear side of the SU-8 layer for the micro-channel. After UV lithography, the sample was post baked at 65 °C for 10 min and 95 °C for 30 min (figure 5k). Then the PET film together with SU-8 and parylene layer was released from glass slide by removing the tapes. The SU-8 layer together with the parylene layer was dry released from PET by slightly bending the PET film. This was post baked and developed to form the micro-channels on the final layer, the holes on the grid layer and the hollow SU-8 needle. Oxygen plasma etching was used to etch the parylene film between needles. Due to the anisotropic etching, the parylene sidewall around CNT bundles was not etched. By having this parylene sidewall, CNT bundles could still be assembled onto SU-8 needles. A thin PDMS film of 200μιη thickness with a centre space larger than the dimension of the microneedle CNT bundle array cut off was prepared (figure 51). The surface of this PDMS layer was treated with nitrogen plasma to introduce amino groups on one side. When this PDMS surface is contacted with the bottom side of SU-8 surface having epoxy groups on its surface, interfacial amine-epoxide chemical reaction takes place at an elevated temperature (figure 5m) permitting attachment to the SU-8 layer. The sample was therefore baked at 120 °C for 30 min to achieve a permanent bonding between SU-8 and PDMS (figure 5m). In this step, the thickness of PDMS must be kept low enough to avoid the poor quality bonding, which is due to the non-conductivity and the low plasma efficiency of the PDMS layer.

After the SU-8 membrane is released from the PET film, it bends during the developing process (Fig. 12 (b)), due to residue stress gradients in the SU-8 membrane. Here, a homemade stage is used to reduce this effect. As shown in Figure 12 (a), two akryl plates are applied to clamp SU-8 membrane bonded with PDMS layer. Two other PDMS layers are used as soft buffer layers to protect the device. This stage offers a flat surface and uniform pressure on the SU-8 membrane. When put into the 120 °C oven, the SU-8 membrane will slightly deform because the bonded PDMS layer has a thermal expansion at this high temperature. Such effect relieves the previous bended layers and leads to a more flat surface (Figure 12 (c)). This flat surface is important for the following maltose tips drawing process. Then O2 plasma is performed on the opposite side of the first PDMS layer and one side of the second PDMS layer. After attaching these two surfaces together, two PDMS layers are bonded firmly together to form a fluidic chamber for a drug to flow into microneedles.

After the bonding, the parylene layer was etched by oxygen plasma to open the sealed top ends (figure 5m). In this process, the whole parylene layer was etched away except the parylene sidewall around the CNT bundles due to the anisotropic property of RIE (Figure 1 (d) and 1 (e)). The adhesion between parylene and SU-8 was good enough to hold suspended CNT bundles as mentioned above. In the RIE etching process, the plasma generated large amount of heat. The temperature of SU-8 layer would rise to several hundred °C during the long period of the 10μιη thick parylene etching. This high temperature would degenerate the bonding between SU-8 and PDMS. Thus, the whole etching process was divided into several periods to avoid over heating for samples.

After etching away parylene on CNT bundles, the product array was bonded with a thick PDMS layer (figure 5n). This process is a conventional bonding between PDMS surfaces. Both PMDS surfaces were treated with oxygen plasma and bonded together. This thick PDMS layer allows us to connect tubes for further fluidic tests.

Alternative fabrication process I

As shown in Fig. 20, SU-8 microtubes fabrication starts from a layer of Polyethylene Terephthalate (PET, 3M USA) film pasted on the Si substrate by sticking the edge area with kapton tape (Fig. 2 (a)). The PET film, a kind of transparent film not sticky on both sides, is used as a sacrificial template to dry release the final device from the Si substrate because of the poor adhesion between PET film and SU-8. [The sticky kapton film just applied along the edges of samples to tightly fix with PET film can be easily removed after the device developed in SU-8 developer]. The PET film with the SU-8 layer is separated from the Si substrate. Then SU-8 layer could be dry released from the PET film just by slightly bending the PET film.

A 140 μηι thick SU-8 layer is deposited on the fixed PET film (Fig. 20(b)). To ensure a smooth SU-8 surface, this deposition is conducted in two steps of coating with 70 μΐ layer each. In each step, SU-8 2050 is spun at 2000 rpm for 30 seconds, followed by prebaking steps at 65 °C for 10 minutes and 95 °C for 30 minutes. After the prebaking steps, this SU-8 layer is exposed under 450 mJ/cm2 ultraviolet (UV) energy to define the membrane structure on this layer (Fig. 20(c)). After exposure baking steps at 65 °C for 5 minutes and 95 °C for 15 minutes, another 350 m SU-8 layer is directly deposited on this layer in two steps without development (Fig. 20(d)). If the first layer is developed to get patterns, the surface will not be smooth enough to achieve a uniform second SU-8 layer on it. With careful alignment, an exposure of 650 mJ/cm2 energy is performed on the second layer to get the pattern of SU-8 microtubes, which are precisely above the holes patterned on the first layer (Fig. 20(e)). After post exposure baking steps at 65 °C for 10 minutes and 95 °C for 30 minutes, then the SU-8 device with PET film is released from the silicon substrate by the same method described before (Fig. 20(f, g)). After soaking in an ultrasonic cleaner for 30 minutes, the SU-8 microtubes array on the membrane is developed (Fig. 20(h)).

Alternative fabrication process II

The microneedle array is fabricated by an innovative double drawing lithography process.

The design of the microneedle array integrated with CNT nanofilters is shown in Fig. 21. An array of SU-8 microneedles was patterned above a SU-8 membrane(Fig. 21 (a)). Every SU-8 microneedle has two parts: four-beam sidewalls at the bottom and a sharp tip at the top as shown in Fig. 21 (c). The four-beam sidewalls are patterned by photo lithography. The gaps along the sidewalls are the outlets of the microneedles. The sharp tips above the four- beam structure (shown in Fig. 21 (a)) are assembled and patterned by double drawing lithography. Above them, a layer of gold surface electrode was deposited onto the whole surface. This surface electrode allows us to apply electric field in the test. Inside the four-beam structures, vertical grown CNT bundles (black parts in Fig. 21 (a)) were embedded in the SU-8 membrane to form the CNT nanofilters. Fig. 21 (d) shows the SEM image of the CNT bundle. Underneath the SU-8 membrane, there is a SU-8 chamber layer to support the SU-8 membrane layer and form a solution chamber. PDMS layers (the base part in Fig. 21 (a))are bonded to the SU-8 chamber layer for tubing in the test. Solution could be loaded in the chamber under the CNT bundles, pass through the CNT bundles and finally through the SU-8 microneedle into the tissue. The optical image of the microneedle array with gold surface electrode is shown in Fig. 21(b). For applying the electric field across the CNT nanofilters in the test, one electrode will be bonded onto the surface electrode and another electrode will be inserted into the PDMS chamber as shown in Fig. 22. When solution is loaded in the drug reservoir and flows through the CNT nanofilters, two electrodes would be connected by solution and an electric field is generated across the CNT nanofilters.

Fig. 23 illustrates the fabrication process. The process began with thermal oxidation of single crystal silicon substrate to form a etch stop oxide layer. After the CVD of polycrystalline silicon as a sacrificial layer, a 5 nm thickness of pattered Fe film, which acted as the catalyst film for the selective growth of CNTs, was prepared onto the silicon substrate (Fig. 23(a)).As illustrated in Fig. 23(b), the vertical aligned CNT bundles of 50 pm in height were grown via pyrolysis of acetylene at 800°C with an Ar/NH3 flow for 15 min. As illustrated in Fig. 23(c), the CVD parylene-C was employed to fill into vertically aligned CNTs and then to reinforce the inter-tube binding at room temperature. Thus, the top side of CNTs was covered with parylene-C, and the discrete CNTs were bound together by parylene-C. This step was the most critical process for forming the mechanical supporting layer for CNT bundles. The thickness of the flexible parylene-C layer was determined by the CVD process. To achieve reliable mechanical strength for following handling and use, 10 pm thick parylene layer was deployed. The parylene layer was peeled off together with CNT bundles from the substrate. As shown in Fig. 23(d), the parylene film was attached onto a thin glass slide. Then a layer of 50pm SU-8 was deposited onto the parylene layer. The thickness of this SU-8 layer was the same as the height of CNT bundles. Due to the transparency of the glass slide and parylene layer, the SU-8 layer was exposed from the back side of glass slide. The catalyst layer under the CNT bundles could act as mask in this lithography step. The SU-8 above the CNT bundles would not be exposed. After development of SU-8, the parylene top of CNT bundles would not be covered by SU-8 as shown in Fig. 23(e). Such SU-8 layer deposited above the parylene layer could act as hard mask for plasma etching. The sealed parylene top of CNT bundles could be etched by oxygen plasma as shown in Fig. 23(f). Then release the parylene layer together with the SU-8 cover layer was separated from the glass slide and bonded onto an unexposed SU-8 layer deposited on another thin glass slide as shown in Fig. 23(g). In this process, a layer of SU-8 was spin coated and pre-baked on a thin glass slide first. After cooling, attach the released parylene layer onto the SU8 layer then re-bake the SU-8 layer to make it melted. After cooling, a good bonding would form between the parylene layer and the SU-8 layer. Expose the sample from the backside of the glass slide to form a drug reservoir under the parylene layer as shown in Fig. 23(h). The size of the drug reservoir should be slightly larger than the dimension of the CNT bundle array. Etch off the catalyst layer at the backside of CNT bundles by oxygen plasma and bond it with a thin PDMS layer as shown in Fig. 23(i). For the bonding between PDMS and SU-8, the PDMS layer should be treated with nitrogen plasma then be attached onto the SU-8 layer and baked at 120°C for 30 minutes. Then a 30pm thick SU-8 membrane was patterned on the front layer of sample as shown in Fig. 23(j) to reinforce the structure. On this membrane layer, holes aligned with the CNT bundle array were patterned. As shown in Fig. 23(k), array of SU-8 four-beam sidewalls array was further aligned and patterned above the membrane layer. As shown in Fig. 23(i), a thick PDMS layer with a center hole was bonded at the backside. This PDMS layer was used for tubing purpose. The centre hole was for the insertion of tube. Then SU-8 sharp tips were assembled onto the four-beam sidewalls array by double drawing lithography as shown in Fig. 23(m). Finally, a gold surface electrode was deposited onto the whole surface by evaporation as shown in Fig. 23(n). The detailed structure of a single microneedle integrated with CNT nanofilter is shown in Fig. 23(o).

Double drawing lithography to assemble microneedles upon CNT nanofilters

The process for drawing lithography is as shown is Fig24(a). A pre-baked SU-8 layer was prepared on Si substrate. Then a four beam structure is mounted above the SU-8 layer and the SU-8 layer is baked to make it melt as shown in step (I). Then the four beam structure is inserted into the melted SU-8 layer to a depth d as shown in step(ll). The four beam structure is drawn out of the melted SU-8 layer. Some Su-8 will attach onto the top of the four beam structure and a Su-8 bridge will be formed between the melted SU-8 layer and the four beam structure as shown in step(lll). The four beam structure is further drawn out the pillar to break the SU-8 bridge and form the sharp tip as shown in step(IV). We then developed an innovative double steps drawing lithography process as shown in Fig 24(b). We conducted first time stepwise controlled drawing lithography and got hollowed tips as shown in step(l) . Then the whole device was baked in an oven at 120°C to melt the hollowed SU-8 tips as shown in step(ll). Melted SU-8 reflowed into the gaps between four-beam sidewalls and the tips became domes. Then a second drawing process was conducted on the top of melted SU-8 to form sharp and solid tips as shown in step(lll) and step(IV).

The new double drawing lithography is developed to create sharp SU-8 tips on the top of four SU-8 pillars for penetration purpose. Drugs can flow through the sidewall gaps between the pillars and enter into the tissues. The experiment results indicate that the new device can have larger than 1 N planar buckling force and be easily penetrated into skin for drugs delivery purpose. By delivering glucose solution inside the hydrogel, the delivery rate of the microneedles can be as high as 71% when the single microneedle delivery speed is lower than 0.002mL/min.

An array of 3x3 SU-8 supporting structures was patterned on a 140 μιτι thick, 6mmx6mm SU-8 membrane (Fig. 27(a)). Each SU-8 supporting structure included four SU-8 pillars and was 350 μηη high. The four pillars were patterned into a tube like shape on the membrane (Fig. 27(b)). The inner diameter of the tube was 150 μιη, while the outer diameter was 300 μιη. SU- 8 needles of 700 μιη height were created on the top of SU-8 supporting structures to ensure the ability of transdermal perforation. Considering that the total SU-8 supporting structure is only 350 μιη high, we choose 125 °C as baking temperature for proper SU-8 flow-in speed and easier SU-8 flow-in depth control. In our new device with the four pillars supporting structure, the SU-8 could flow inside the sidewall gaps between the pillars to form anchors. These anchors could enhance the microneedles' mechanical strength and overcome any planar shear force problems. Two PDMS layers were bonded with SU-8 membrane to form a sealed chamber for storing drugs from a connection tube. Once the microneedles penetrated into the tissue, drugs could be delivered into the body through the sidewall gaps between the pillars (Fig. 27(c)).

To safe guard against drugs leaking into the skin surface through the sidewall gaps of the four pillars especially under pressure or electric field assisted delivery Hydrogel can be used with eth device.

Microneedle Design Considerations

An array of 5x5 SU-8 microtubes is patterned on a 140μιη thick, 2.5cmx2.5cm SU-8 membrane (Figure 11 (a)). Each SU-8 microtube is 350μηπ high. The inner diameter of the SU-8 microtube is 150μιτι, while the outer diameter is 300μιη (Figure 11 (d)). Maltose needles of around Ι ΟΟΟμιη height are integrated on the SU-8 microtubes to ensure the ability of transdermal perforation. Two PDMS layers and the 2.5cmx2.5cm SU-8 membrane are bonded together to form a sealed chamber for retaining drugs from the connection tube during delivery process. The 2.5cmx2.5cm size of the device is designed on purpose in order to conceptualize a skin patch kind of drug delivery device. However, the critical area comprising the SU-8 microneedles at the centre is only 6mmx6mm. The large marginal space offers sufficient area to achieve good bonding between SU-8 layer and PDMS layer, i.e., tolerating higher pressure to drive drugs into tissues during the delivery process.

Drawing Process of Maltose Tips

Maltose needles are integrated on the SU-8 micro-tubes by a drawing lithography technology (figure 10). In our device, drugs are delivered through SU-8 microtubes and the maltose is just used as sharp tips for skin penetration.

As shown in Figure 13, fabrication of maltose tips on top of SU-8 micro-tubes is divided into four steps. Firstly, concentrated maltose solution (in this instance containing Methylene blue) is dripped on a glass slide. The slide is kept at 140 °C on a hotplate until the water inside maltose solution completely vaporizes (Fig. 13 (a)). Secondly, a composite of SU-8 microneedles is fixed on a precision stage which can control the position of SU-8 microneedles in three-dimensions. Then the SU-8 microneedle composite is immerse in the liquid maltose at 140 °C and maltose liquid is coated on the SU-8 microneedles' surface (Fig. 13 (b)). Thirdly, the temperature of the liquid maltose is gradually increased and we start drawing SU-8 microneedle composite away from the liquid maltose and air interface (Fig. 13 (c)). Finally, when the temperature rises up to 160 °C, the drawing speed is increased. Since the maltose liquid is less viscous at higher temperature, the connection between the SU-8 microneedle composite and the surface of the liquid maltose creates individual maltose bridges which shrink gradually, and then break. The end of each shrunken maltose bridge forms a sharp tip on top of each SU-8 microneedle (Fig. 13 (d)). Fig. 14 shows the final drug delivery device containing SU-8 microneedles integrated with maltose tips and microfluidics, e.g. channels and the chamber formed by PDMS.

An alternative embodiment of a Maltose tip is shown in Figure 17a-b. In this embodiment the microneedle comprising a number of up-standing but disconnected pillars is made from our preferred polymer, SU-8. As above, a maltose tip is added to the microneedle using a first lithography drawing, the microneedle and maltose tip are then heated until the maltose tip dissolves and so flows into the central cavity of the microneedle and into the side gaps between adjacent pillars. The microneedle is then subjected to a second lithography drawing, as above, to provide a final maltose tip on the microneedle. We have found that a microneedle made in this fashion has a maltose tip with improved adhesion to the remainder of the microneedle.

In Figure 17c there is shown an alternative method for manufacturing a microneedle with a maltose tip. In this method a microneedle comprising a number of up-standing but disconnected pillars is made from our preferred polymer, SU-8. As above, a maltose tip is added to the microneedle using a first lithography drawing, the microneedle and maltose tip are then heated until the maltose tip dissolves and so flows into the central cavity of the microneedle and into the side gaps between adjacent pillars and also over the entire outer surface of the microneedle. The microneedle is then subjected to a second lithography drawing, as above, to provide a final maltose tip on the microneedle. Those skilled in the art will appreciate that in this alternative method sufficient maltose must be deposited on the microneedle, relative to the size of the microneedle, for this method to be undertaken.

In Figure 18 there is illustrated a method for the manufacture of a microneedle made from a biodegradable material such as starch. The manufacture of a starch microneedle involves the deposition of starch on a substrate using conventional techniques, the use of a mask (shown as a solid image in the figure but in reality having the requisite features to enable a microneedle to be manufactured) followed by the deployment of a suitable etching process such as etching with oxygen plasma.

Optimization of Spacing Between Maltose Tips

We have investigated various spacing between two adjacent maltose tips to get the minimum spacing. Microtubes 150μιη to 350 μηι in height require a 100μηη interval spacing whereas microtubes 300pm to 1200pm in height require a 300pm interval spacing. Table 1 shows the observed data for the ratio of formed individual maltose tips:formed clustered tips in 48 samples. We concluded that individual maltose tips are successfully derived from samples of microtubes with the height beyond 350pm and the spacing larger than 900pm. As a result, we prepared many samples of SU-8 microtubes of 350pm height and 1000pm spacing for further maltose integration experiment in our study.

TABLE. I. Ratio of individual maltose tips to clustered maltose tips among a 5x5 microneedles array

Mechanical Strength of the microneedles

To ensure both the adequate adhesion property between the maltose tip and SU-8 microtubes, and the sufficient stiffness of the SU-8 microtubes for successful skin penetration, the mechanical strength of the microneedle was studied. We discovered the SU-8 microtubes are strong enough to withstand considerable pressure: after characterization of 20 samples, the average pressure threshold value is 7.36±0.48N for the microneedles (300pm apart at the microneedle base and 1000 m high). Since the minimal force required for a successful penetration is reported to be less than 1 N for a similar microneedle array, the device is reliable during the penetration process.

Pressure and Electrode Driven Ion Exchange

The test setup is shown in Figure 7. A short tube was inserted in the hole on the PDMS layer. A Ti electrode penetrated the PDMS layer and connected the solution inside the PDMS chamber. Bias could be applied via this electrode. Test solution of very small volume was pre-loaded in the PDMS chamber and part of the short tube. This preloading is necessary to get rid of the air in the PDMS chamber and make the solution contact the surface of CNT bundles. Then a longer tube and a syringe were connected to the short tube. The air in the long tube and syringe were compressed when the syringe was driven by a syringe pump to supply a constant air pressure on the solution in the short tube. Further there was another tube connected to the long tube and a pressure sensor. This pressure sensor was used to calibrate the level of air pressure applied by the syringe pump. Based on our observation, the CNTs were able to withstand a pressure level of 40kPa, but for some samples, leakage between the interface of PDMS and SU-8 was detected at a pressure level higher than 25kPa. Thus, the pressure level was kept within 25kPa in our experiments. The composite was placed in the beaker. The liquid surface in the beaker should be kept lower than the exposed part of the electrode inserted in PDMS. This was to avoid the two electrodes being shorted by the solution connection. Another Ti electrode was immersed in the solution in the beaker. The two electrodes were connected to a semiconductor characterization system (KEITHLEY 4200) which could apply bias and measure the current simultaneously. Resultant solution was sampled from the solution inside the beaker.

In the study of driving ions by electric field, NaCI solution was loaded in both the beaker and the tube. The solution volume in the beaker was 60 ml and in the tube was about 0.2 ml. A bias of 5 V was applied between the two Ti electrodes for 160 min. The electrode in the beaker was grounded and the electrode in the tube was positively biased. No air pressure was applied. No obvious electrolysis of water was observed. Two concentrations of NaCI solution, 1mol/L and 0.1 mol/L, were employed to study the relation between the ion concentration and ionic current. The experiments were carried out at room temperature (~26°C). The solution in the beaker was sampled for pH measurement at 10, 20, 40, 80 and 160 min. NaCI and HCI solutions were prepared using deionized (Dl) water from a water purification system (Millipore SAS 67120 MOISHEIM). DNA synthesis

Single strand DNA (ss-DNA) was prepared by the following steps: Bordetella pertussis genomic DNA was used as a template to synthesize a PCR fragment of 805 base pair using Go Taq Green Master Mix (Promega) kit protocol. Primers used for the PCR were from Sigma Aldrich, HPLC purified vipC forward primer (5TTGAATTCGAGTTCGAGCCGGTGCTGG3') and vipC reverse primer (5'TTAAGCTTTTGCTGGTAAGGAATGCGCTG3'). Annealing temperature of 64.5°C was used and the denaturation, annealing and extension cycle was repeated 25 times. Final elongation step was carried out at 72°C for 10 min. The ds-DNA generated was then kept at 95°C for 5 min to separate the two DNA strands to generate ss-DNA. PCR reaction was set up using Bio-Rad iCycler-thermal cycler PCR. Haemagglutinin was purchased from Zuellig Pharma. Other chemical reagents were purchased from Sigma Aldrich and used without a further purification.

RESULTS

Microneedle design overview

Patterned CNT bundles were transferred and assembled on SU-8 substrate and micro-channels were patterned under CNT bundles as shown in Figure 1(a). Normally, CNTs are grown vertically from silicon substrates, making it very difficult to pattern well aligned fluidic channels under patterned CNT bundles. Herein, patterned CNT bundles grown on silicon substrate were reinforced by parylene first. Then the parylene layer together with the CNT bundles were peeled off from a silicon substrate and transferred to the SU-8 substrate. SU-8 is reminiscent of a negative photoresist commonly used during the fabrication of microfluidic devices. Well aligned channels were created under the CNT bundles on SU-8 (Figure 1(a) and (b)). For opening the sealed top ends of CNT bundles, oxygen plasma was used to etch the parylene layer. Due to the anisotropic property of reactive ion etching (RIE), all the parylene layer was etched except the sidewalls around the CNT bundles (Figure 1(d) and (e)). These parylene sidewalls around the CNT bundles have a good bonding with SU-8 to fix the suspended CNT bundles (Figure 1(c)).

Figure 2 shows the 3D illustration of the device. The whole device excluding PDMS bonding part has four layers. The top layer is CNT bundle array. Beneath the CNT bundle array is the SU-8 needle array supporting the CNT bundle array. The third layer is a grid layer. On this grid layer there are many holes. The last layer is the sealing layer, which will seal all the grid holes but just let holes under the SU-8 needles open. Then solution could pass through these four layers. SEM images of the SU-8 needles with CNT tips can be seen in Figure 3.

Permeability of CNT microneedles Patterned CNT bundles were used as nano-filters to test the permeability of small ions and large molecules by using pressure driven and electric field driven methods. For tests of electric field driven small ions, NaCI solution was employed as the test solution. The ionic current was measured to prove that ions could pass through CNTs under an external electric field. For tests of pressure driven small ions, HCI solution was employed as the test solution. The change of the pH value of the solution was measured to prove that ions could pass through CNTs under external pressure. For the tests of permeability of larger molecules, we use ss-DNA and haemagglutinin, a flu vaccine, as test chemicals. It is shown that DNA and ss-DNA could pass through CNTs by external electric field. In this study, we explored the possibility of driving ss- DNA and haemagglutinin by high pressure and influence on permeability by applying an additional electric field.

The change of ionic current was recorded as shown in Figure 8(a). The ionic current quick dropped down as a function of time. The peak values of the ionic current of NaCI concentration 0.1 mol/L and 1mol/L were 9.5x10^"5A and 2.4x 0^"5A.

The large peak value difference suggests that the main transport mechanism under electric field is electrophoresis. The drop of the ionic current is due to the depletion of the ions at the PDMS chamber. The high peak values of the ionic currents represent fast ion transport. This fast ion transport may cause depletions which cannot be compensated by the ion diffusion from other regions. As shown in Figure 7(c), the geometry of PDMS chamber connecting to the CNT bundles is very limited. We hypothesized that ions cannot efficiently diffuse from other places to this region where ions quickly pass through the CNT bundles. Thus the depletion of ions in the DPMS chamber causes the drop of the ionic current. We carried out an experiment to prove this. The 5 V bias was applied in a square wave mode with duty cycle of 1 min on and 1 min off. This square wave mode bias was applied for several cycles and the recorded ionic current is shown in Figure 8(c). The ionic current dropped when bias was on. When bias was not off, the solution in the PDMS chamber was replenished with ions by diffusion from other regions. Thus, when bias was on again, the peak value of ionic current recovered to a certain level. But after a 1 min time duration was again depleted. The green line shows the drop of peak values from about 1.5*10^"5A to about 1.4*10^"5A for 1 mol/L NaCI concentration.

In the study of driving ions by pressure, 0.2 ml 3.7% HCI solution was preloaded in the tube. 60 ml Dl water was loaded in the beaker. Air pressure of 10kPa and 20kPa was applied for 80 min to study the relation between pressure and permeability with a zero biasing voltage. Solution in the beaker was sampled at 10, 20, 40 and 80 min. The pH value of solution samples was measured by pH meter (CORNING Pinnacle, Model: 530) as shown in Figure 8(d). The line of 20kPa pressure shows a greater pH value drop than that of 10kPa. At 80 min, the pH values of solution after applying 10kPa and 20kPa were 4.25 and 3.97. According to the definition of pH value:

pH=-log[H⁺]; ([H⁺] refers to the H⁺ concentration.).

The ratio of H⁺ concentration is [H⁺]₂okp_a/[H⁺]iokPa =10^{4 25}/10^{3 97}=1 .9. The ratio of H⁺ concentration was almost the same as the ratio of pressure level. This result suggests a linear relation between mass transport rate and pressure level, which is consistent with the reported results.

Large molecule transfer

In the study of large molecule translocation through CNTs, ss-DNA and Haemagglutinin were employed to study CNT bundles' function as a filter. The inner diameter of the CNT in this study is 10 nm (Figure 6(e)). The ss- DNA whose cross sectional dimension is smaller than 10nm was expected to pass through CNTs just by applying pressures. Haemagglutinin, whose shape is like a cylinder and whose widest dimension is approximately 13.5nm, was expected not to pass through CNTs by applying pressure. A combination of pressure and electric field driven methods could drive ss-DNA through CNTs. In the experiment of ss-DNA, 60 ml Dl water was loaded in the beaker, a droplet of ss-DNA solution was preloaded in the tube. Pressure of 25kPa was applied for 7 hours. The solution in the beaker was sampled after the experiment. The absorbance spectra were shown by FTIR (Fourier transform infrared spectroscopy Cary 660-FT-IR). Dl water was used as background to show the absorbance in Figure 9(a). The green line shows that ss-DNA could pass through CNT membranes just by applying pressure. The absorbance shown in Figure 9(a) suggests that more ss-DNA could pass through CNTs by the additional electric field. In the case of Haemagglutinin under pressure driven condition, no absorbance observed in spectra of samples carried out at applied pressure of 10, 15 and 20kPa for 7 hours. Figure 9(b) shows the measured spectrum under 25kPa for the above samples with additional 15 hours. Again there is no absorbance observed. Then a pressure level of 25kPa and a bias of 5V was applied together. The measured spectrum indicates a weak absorbance as shown in Figure 9(b). It suggests that Haemagglutinin could pass through CNTs when pressure and electric field were applied together. But the translocation rate was relatively low compared to that of ss-DNA. This is because the dimension of Haemagglutinin is much larger than that of ss-DNA. The test result shows that permeability decreases with the increase of dimension of molecule. We conducted experiments to see if insulin could be delivered using our system. Insulin is a peptide hormone and central for regulating carbohydrate and fat metabolism in the body. Due to the poor absorption or enzymatic degradation of insulin in the gastrointestinal tract and liver, the transdermal delivery has been so far the preferred method of insulin administration. The molecular radius of insulin is 1.34nm which is smaller than the inner diameter of the CNTs in the device. It could pass through the CNTs just by applying pressure. And also the insulin molecules are positively charged in solution, and so the transport rate could be tuned by applying electric field.

The insulin solution of 1mg/ml concentration was preloaded in the drug reservoir. Air pressure levels range from 5kPa to 20kPa were applied for 30mins. The resultant solution samples were analyzed by FTIR as shown in Fig 25(a). The peak value indicates the concentration of insulin in sampled solutions. From the test results, the concentration of insulin is proportional to the pressure level which means the transport rate of insulin through CNTs is linear to the pressure level.

Then the test was repeated by applying bias ranges from -10V to +10V and air pressure ranges from 5kPa to 20kPa. The peak value of IR spectra at 4.7pm wavelength was recorded in Fig 25(b). The positive bias could facilitate the transport of insulin and negative bias could decrease the transport rate. For the line of -7.5V and -10V, when the pressure was lower than 10kPa, the IR spectra at 4.7pm wavelength was lower than the noise level thus no insulin was detected. This result indicates that the CNT nanofilters could be used as both pressure valves and electric switches for the delivery of molecules such as insulin. And a sufficient reverse bias could balance the air pressure, realizing a zero delivery of insulin.

Exclusion of micro-scale particles

To confirm CNTs nanofilters could block all micro-scale substance, we further tried passing bacterium through CNTs nanofilters. In this experiment, PBS and BPSM (Bordetella pertussis, streptomycin resistant) are employed. 0.1 ml mixed solution was loaded in the device and PBS was loaded in a beaker in fluid communication with the device tips. A medley of pressure and electric field driven method was used to drive the PBS and BPSM through the CNTs nanofilters. The pressure is 20Kpa and the electric bias is 5V. To detect the bacterium at very low concentration, the solution should be sampled for cell culture and enrichment. To ensure the bacterium were viable in ambient temperature during the experiment, the test was conducted for 2 hours. The solution sampled in the beaker was cultured on blood agar plates for 5 days to see whether bacteria passed through the CNTs nanofilters. Solution with bacteria loaded in the device chamber was also cultured as control group for comparison. Figure 26(a) shows the result of bacterial culture of the solution sampled from the device chamber. After 5 days of culture, bacterial colonies were observed. Figure 26(b) shows the result of bacterial culture of the solution sampled in the beaker. No bacterial colony was found which confirms a perfect bacterium blockage.

Optimization of Spacing between Maltose Tips

Dissolvable microneedles have been shown to encapsulate bioactive molecules and deliver their cargo into skin when microneedles are dissolved in body fluid. Such results showed that dissolvable microneedles offer an attractive and effective mean to administer drugs while providing safety and immunogenicity. However, so far, there are no published data reporting the use of dissolvable microneedles integrated with microfluidics. To continuously provide a large volume of drugs via the perforated skin using a dissolvable microneedle system, a novel dissolvable microneedles device comprising fluidic channels or micro-tubes connected with individual dissolvable tips is highly desired.

Conventionally, sharp micro-needles are required to create micro-channels in the SC layer for further drug administration. Sharp micro-tips made of maltose have been reported as a method for forming holes in the SC layer through which a drug solution could subsequently pass. These reports show that maltose micro-tips have enough mechanical strength to be stably inserted into the SC layer. Fabrication of such sharp maltose micro-tips has been demonstrated by moulding and draw lithography based approaches. More importantly, maltose is a well-known dissolvable material in a moist environment, with it reported that maltose micro-tips are dissolvable in body fluid, i.e., bio-dissolvable materials. Therefore, we leveraged the unique features of maltose based micro-tips, i.e. high mechanical strength and bio- dissolvability, to validate the feasibility of integrating sharp maltose tips on top of SU-8 microneedles in the present study (figure.10).

To ensure the stronger and shorter maltose tips of a microneedle array for an ease of skin penetration, experiments were conducted to optimize parameters in the maltose drawing process. The temperature of melted maltose during the drawing step and the drawing speed were identified as the key parameters to get up to 1000pm tall maltose sharp tips on the SU-8 microneedles. Although the maltose tip can be easily and routinely formed on top of SU-8 microneedles at 140 °C and optimized drawing speed, the formation of uniform maltose tips array also depended upon the spacing between two adjacent SU-8 microneedles.

Given the maltose melted at 140°C, the planar extensional viscosity in maltose easily leads adjacent maltose tips to clusters. We investigated various spacing between two adjacent maltose tips to get the minimum spacing. The height of microneedles changes from 150pm to 350pm while the spacing changes from 300pm to 1200pm. Microtubes 150pm to 350 pm in height require a 100pm interval spacing whereas microtubes 300pm to 1200pm in height require a 300pm interval spacing. We concluded that individual maltose tips are successfully derived from samples of microneedles with the height beyond 350pm and the spacing larger than 900pm. As a result, we prepared many samples of SU-8 microtubes of 350pm height and 1000pm spacing for further maltose integration experiment in our study.

Characterization of Maltose tip SU-8 microneedle penetration

Figure 15 (a) shows the insertion result of a 5*5 microneedles array into a porcine cadaver skin. After the insertion, maltose tips were rapidly dissolved once inserted in the tissue. Methylene blue was added into the maltose for inspection purpose. Ten minutes after insertion, 25 blue traces were easily found, which matched the pattern of the microneedle array. The optical microscope image in Figure 15 (b) shows a hole perforated in the skin after the skin surface was cleaned. During the insertion experiment, it was important to avoid the shear force influence caused by deformed skin surface on the individual maltose tip in order to get successful microneedle penetration for the whole array. We used precision stages to hold the microneedle device substrate and control the relative position of device substrate and skin sample. Dissolving of Maltose Tips and Demonstration of Injection via SU-8 Microneedles

In order to check whether the maltose tips could be dissolved once inserted in the tissue, four chips with the same maltose tips height were inserted into the skin and taken out one by one with 3 minutes interval. Maltose tips were gradually dissolved versus increased time as shown in Fig.16. After 9 minutes, the maltose tips were totally dissolved and the lumens of SU-8 microneedles were observed from the top view. Unlike traditional dissolvable needles which encapsulates drugs within the needles, this microneedle array allows large volume of drugs to pass through via the remaining SU-8 microneedles inside the skin.

Moreover, blue dyed water was ejected through the lumens of SU-8 microneedles to a beaker containing fresh water to demonstrate the hollow tubes formed in each individual microneedle during the fabrication process. This microfluidic testing is performed in fresh water as shown in Figure 19. Visual monitoring of the ejection proved that there was no blockage inside the microneedles. Because of the good bonding quality between each layer, there was no obvious damage to the device in this experiment even though the syringe pump is increased to its maximum speed at 3.3mL/min for 6mL syringe.

Summary We propose a new microneedle and its fabrication process based on a stretchable membrane or a biodegradable material and a nano-filter comprising vertically grown nanotubes. The device can endure high pressure. The functionality is optionally improved by pumping drugs through the nano-tubes either by applying pressure or applying an electric field. The test results prove that the nano-tubes can be deployed as a nano-filter working at high pressure. The nano-filters provided on or in a flexible polymer or biodegradable material enables integration with other microfluidics for chemical and pharmaceutical applications. Further, the microneedles may be provided with dissolvable tips to enhance their performance, in particular their penetration function.

Claims

1. A microneedle for transdermal drug delivery comprising a nano-filter that includes a plurality of nano-tubes whose longitudinal axis is aligned with that of said needle for the selective delivery of nano- scale molecules or particles.

2. A microneedle according to claim 1 wherein said nano-filter is located at, or adjacent, either a first or second end of the microneedle.

3. A microneedle according to claim 2 wherein said nano-tubes are mounted on the tip of said microneedle.

4. A microneedle according to claims 1 or 2 wherein said nano-filter is located within the microneedle.

5. A microneedle according to any one of claims 1-4 wherein said nano-filter is integral with said microneedle.

6. A microneedle according to claim 5 wherein said nano-filter is grown on or in, or assembled on or in, said microneedle.

7. A microneedle according to any one of the preceding claims wherein said microneedle is made from one or more of the following materials: silicon, starch, stainless steel, titanium, tantalum, nickel, a ceramic and a polymer-based material.

8. A microneedle according to any one of the preceding claims wherein said microneedle is fabricated into a sharp shape for ease of penetration.

9. A microneedle according to any one of the preceding claims wherein said microneedle is a SU-8 needle or a starch needle.

10. A microneedle according to any one of the preceding claims wherein said microneedle has a further tip which is dissolvable.

11. A microneedle according to claim 10 wherein said further tip is fashioned into a sharp-shape for ease of penetration.

12. A microneedle according to claims 10 or 11 wherein said dissolvable tip is a carbohydrate-based tip.

13. A microneedle according to claim 12 wherein said tip is a sugar tip.

14. A microneedle according to claim 12 wherein said tip is a starch tip.

15. A microneedle according to claim 11 wherein said tip is a polyvinylpyrrolidone (PVP) tip.

16. A microneedle according to any one of the preceding claims wherein said microneedle further comprises an associated, integrated or removable electric field member and/or a pressure member for facilitating the passage of said drug through said microneedle.

17. A microneedle according to claim 16 wherein said microneedle further comprises an associated, integrated or removable electric field member and said nano-filter comprises at least one electrode.

18. A microneedle according to any one of the preceding claims wherein said nano-tubes are carbon nano-tubes.

19. A composite comprising:

a) a plurality of microneedles for transdermal drug delivery wherein each microneedle comprises a nano-filter including a plurality of nano-tubes whose longitudinal axis is aligned with that of said needle for the delivery of nano-scale molecules or particles; and b) a supporting membrane for supporting said microneedles.

20. A skin patch for transdermal drug delivery comprising:

a) a plurality of microneedles wherein said microneedles comprise nano-filters including a plurality of nano-tubes whose longitudinal axis is aligned with that of said needle for the delivery of nano-scale molecules or particles; and

b) attached to said microneedles is a skin contact layer for attaching to said skin.

21. A skin patch according to claim 20 wherein said skin contact layer is made from a material that sealingly engages with skin.

22. A skin patch according to claim 20 or 21 wherein said skin contact layer comprises an adhesive suitable for attaching said patch to said skin.

23. A skin patch according to any one of claims 20-22 wherein at least one, or the majority, or all of said microneedles contain(s) at least one selected drug, vaccine or DNA molecule.

24. A method of treatment involving the use of said microneedle according to claims 1-18 or the use of said composite according to claim 19 or the use of said patch according to claims 20-23 wherein said microneedle(s) when loaded with at least one suitable medicament is/are applied to the skin, ideally using pressure, whereby said microneedle(s) is forced through the skin and said medicament is administered.

25. A method for the manufacture of a microneedle for use in transdermal drug delivery according to claims 1-18 comprising:

a) mounting a plurality of vertical nano-tubes on at least one

polymer-based substrate to form an assembly;

b) applying a mask to said substrate for producing a number of

microneedles wherein said mask is positioned such that microneedles are made beneath or above said nano-tubes and the nano-tube are located at or adjacent a microneedle tip with their longitudinal axis aligned with that of the microneedle c) etching said substrate to form a plurality of microneedles.

26. A method according to claim 25 wherein said nano-tubes are

carbon nano-tubes.

27. A method according to claim 25 or 26 wherein said nano-tubes are held in a vertical position by the use of a polymer film.

28. A method according to claim 27 wherein said film is parylene.

29. A method according to any one of claims 25-27 wherein said

mounted nano-tubes are attached to said substrate by heating the assembly to a temperature that melts said polymer substrate, at least in the region of said nano-tubes, causing it to flow and so mix with a first contact end of said nano-tubes and then cooling the assembly so that said polymer hardens.

30. A method according to any one of claims 25-29 wherein a sub-layer is applied to said assembly

31.A method for the manufacture of a microneedle for use in

transdermal drug delivery according to claims 1-18 comprising fixedly mounting a plurality of nano-tubes on or in said microneedle wherein the longitudinal axis of said nano-tubes is aligned with that of said needle.

32. A method according to claim 31 wherein said nano-tubes are

mounted on an end of said microneedle.

33. A method according to claim 32 wherein said nano-tubes are

mounted on the tip of said microneedle.

34. A method according to claim 31 wherein said nano-tubes are

mounted in said microneedle.