WO2020064440A1

WO2020064440A1 - Biological material electromechanical interaction platform and methods of making and operating thereof

Info

Publication number: WO2020064440A1
Application number: PCT/EP2019/074904
Authority: WO
Inventors: Michael Smith; Sohini Kar-Narayan
Original assignee: Cambridge Enterprise Ltd
Current assignee: Cambridge Enterprise Ltd
Priority date: 2018-09-24
Filing date: 2019-09-17
Publication date: 2020-04-02
Anticipated expiration: 2021-03-24
Also published as: GB201815550D0

Abstract

The present invention relates to a biological material interaction platform, a method of making a biological material interaction platform, and uses of a biological material interaction platform. A biological material interaction platform according to the invention comprises a substrate supporting an array of shear piezoelectric material nanomembers configured to support a biological subject and configured such that in use, application of an electric field across at least a part of the array of nanomembers provides local stimulation of the biological subject supported by the array; and/or shear deformation of one or more of the nanomembers by the biological subject supported by the array of shear piezoelectric material nanomembers causes generation of electric potential across at least part of the array. A method of making a biological material interaction platform according to the invention comprises steps of hot pressing a piezoelectric material and a nanostructured template to thereby form an array of shear piezoelectric material nanomembers. Devices according to the invention have a lower effective stiffness surface compared to similar devices not having such a nanostructured material layer, and may accordingly be particularly suitable for interaction with biological subjects.

Description

Biological material electromechanical interaction platform and methods of making and operating thereof

The project leading to this application has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 639526).

Field of the Invention

The present invention relates to biological material interaction platforms and methods of making thereof and methods of operating thereof, and in particular, although not exclusively, to biological material interaction platforms for investigation of electromechanical interactions of a biological subject.

Background

Mechanobiology is an emergent field of research concerned with the mechanical interactions between biological systems at a cellular level. Although initial ideas around the mechanisms of mechanobiology were first proposed in the early 20th century, the technology required to investigate this thoroughly has only been realised in the past few decades [1 ,2]. Some of the most significant results from this initial research include the fact that cells can sense and respond to the stiffness of a substrate. Skin cells, for example, will alter their morphology and function when grown on chemically equivalent substrates of different rigidities [3,4]. Furthermore, the fate of stem cell differentiation may also be influenced by substrate stiffness. In one example, it was found that mesenchymal stem cells were more likely to differentiate into fat cells when grown on soft substrates, and bone cells when grown on stiff substrates [5,6]. More recently, similar effects have been observed using nanometre amplitude vibrations to stimulate a wide variety of cells including bacteria, stem cells and brain cells [7-9].

There are many tools available to directly investigate biomechanical effects. A lot of these tools require highly specialised techniques, although the measurements can be extremely accurate. Atomic Force Microscopy (AFM) is a good example; it is able to resolve sub-nanonewton forces, however since the technique was not initially designed with biological applications in mind there are a number of experimental challenges in using these measurement techniques to study biological subjects.

These limitations have led to the idea of more’active’ cell culture environments, those which are capable of providing localised stimulation to growing cells with some spatiotemporal resolution. This is very much an emerging field, with current reports detailing the use of shape memory polymers [10, 1 1], soft substrates with embedded magnetic particles [12,13] and opto-thermally triggered materials [14,15]. In the past few years, there has been a huge increase in the popularity of using piezoelectric materials for cell culture applications. For example, [16] proposes piezoelectric PVDF membranes excited with ultrasound to provide electrical stimulation to attached brain-like cells. [17] proposes intracellular boron nitride nanotubes (BNNT) excited with ultrasonic energy to provide electrical stimulation with a brain-like cell.

However, the present inventors have realised that there are a number of issues with present approaches. One significant difficulty in providing appropriate devices for interaction with biological subjects is that the moduli of biological materials span several orders of magnitude. Brain tissue is exceptionally soft, reportedly similar in texture to cottage cheese or tofu, while bones are required to be strong and stiff with Young’s moduli measured in GPa. However, many of the materials proposed for use in such devices are far stiffer and therefore may therefore cause non-optimal or even adverse mechanobiological effects.

The present invention has been devised in light of the above considerations.

Summary of the Invention

The present inventors have realised that, in the context of a cellular environment, it would be desirable to be able to directly sense and actuate extracellular forces. Furthermore, the present inventors have realised that functional and responsive nanomaterials can provide a convenient route towards achieving a bioelectromechanical interfacing platform with advanced functionality.

Accordingly, in a first aspect, the present invention provides a biological material interaction platform comprising:

a substrate supporting an array of shear piezoelectric material nanomembers arranged in a layer on the substrate and configured to support at least one biological subject and configured such that in use:

(i) application of an electric field across at least a part of the array of shear piezoelectric material nanomembers provides local stimulation of the at least one biological subject supported by the array; and/or

(ii) shear deformation of one or more of the shear piezoelectric material nanomembers by the at least one biological subject supported by the array of shear piezoelectric material nanomembers causes generation of electric potential across at least part of the array.

The term“biological subject” is used herein to generally refer to biological materials including but not limited to: biological tissue, cells (including bacterial cells), cellular components (e.g. organelles), biomolecules (e.g. proteins, DNA), and viruses. It is envisaged that the present biological material interaction platform may be useful for application with a wide range of biological subjects. Preferably, the biological subject is a cell.

The term“nanomembers” is used herein to generally refer to nanoscale structures (‘nanostructures’), i.e. structures having at least one dimension smaller than 1 pm, as measurable using standard techniques well known in the art, for example scanning electron microscopy (SEM) or transmission electron microscopy (TEM).

By providing a nanostructured material layer in the form of an array of nanomembers, it is possible to provide a biological material interaction device having a lower effective stiffness surface compared to similar devices not having such a nanostructured material layer. By appropriate selection of the physical properties of the nanomembers, it is therefore possible to create devices tailored for a wide range of biological applications. For example, it is possible to create a device which more accurately mimics selected biological environments.

Furthermore, because many biological subjects are typically micron sized or smaller (typical size of prokaryotes being about 1 - 5 pm, typical size of eukaryotes being about 10 - 100 pm), forces exerted by such subjects are typically in the pico- to nanonewton range. Accordingly, by creating piezoelectric structure of similar dimension or smaller, these forces can be more easily transduced. Forces exerted by the biological subject(s) can cause deformation of at least some of the nanomembers in the

nanomembers array. Due to the piezoelectric nature of the materials, this deformation of the nanomembers causes generation of electric potential across at least a part of the array. This electric potential may then be detected according to a number of well-known techniques. Detection of this electric potential may provide information on behaviour of the biological subject(s) on the device. In this way, the device may act as a sensing device, for example a capacitance sensing device

Alternatively or additionally, the device may be used as an actuating device to provide local stimulation of the at least one biological subject supported by the array. The local stimulation may be mechanical, electrical, or electromechanical. In an actuating mode, application of an electric field across at least a part of the array of nanomembers may cause deformation of at least some of the nanomembers. Such deformation provides local mechanical stimulation of biological subject(s) supported by the nanomember array. Application of an electric field may additionally cause polarisation of at least some of the piezoelectric material nanomembers, and/or development of surface charge on the nanomembers. Accordingly, application of an electric field may provide local electrical stimulation of biological subject(s) supported by the nanomember array. Providing such local mechanical, electrical, or electromechanical stimulation may be desirable to influence behaviour of the biological subject(s) supported on the device. For example, where the biological subject(s) comprise one or more cells, providing local

electromechanical stimulation may promote cell proliferation. The substrate may comprise a piezoelectric material. In some embodiments, the substrate may comprise the same material as the shear piezoelectric nanomembers. The substrate may comprise a film supporting the array of shear piezoelectric material nanomembers. The substrate and the array may be unitary.

The specific size and shape of the nanomembers is not particularly limited, however preferably the nanomembers comprise nanorods, nanowires and/or nanotubes. Nanorods and nanowires may be distinguished by their aspect ratio, with nanorods typically having an aspect ratio of 10:1 or less, and nanowires typically having an aspect ratio of greater than 10: 1. Most preferably, the array of nanomembers comprises a vertically-aligned array of nanorods, nanowires and/or nanotubes. Nanotubes may be particularly preferred, as they may show a greater piezoelectric response to deformation.

Where the nanomembers comprise nanotubes, the nanotubes may have a wall thickness of between 20nm and 70nm. More preferably the nanotube may have a wall thickness of about 50 nm. The present inventors have found that providing a wall thickness of about 50 nm may provide greatest response to an applied electric field.

Such shapes may be particularly preferred due to their typically high aspect ratios, aspect ratio here defined as the characteristic length of the structure divided by its characteristic width. Preferably the nanomembers have an aspect ratio of 5: 1 or more, 10: 1 or more, 50:1 or more, 100: 1 or more or 500:1 or more. The aspect ratio may be selected depending on a number of characteristics of the biological material interaction platform, including the material of the nanomembers. For example, it may be possible to form nanomembers having a higher aspect ratio by selecting a material having a higher bulk stiffness.

Structures having a high aspect ratio may be more susceptible to bending modes of deformation than a bulk layer. The effect of nanostructuring a surface can accordingly be considered by calculating the ‘effective stiffness’ of the surface in comparison to the bulk stiffness. It is found that the effective stiffness is inversely proportional to the square of the aspect ratio of the nanomembers.

Where ^mhaho is the effective stiffness of the nanostructured surface, and f is the aspect ratio of the nanomembers.

As described above, the nanomembers should have at least one dimension smaller than 1 pm, as measurable using standard techniques well known in the art, for example scanning electron microscopy (SEM). Preferably, the nanomembers have at least one dimension smaller than 500 nm. More preferably, the nanomembers have at least one dimension smaller than 300 nm, smaller than 100 nm or smaller than 50 nm. Where the nanomembers are nanorods, nanowires or nanotubes, preferably the diameter of the rods, wires or tubes is smaller than 500 nm, more preferably smaller than 300 nm, smaller than 100 nm or smaller than 50 nm. Preferably, these limits apply to another dimension, orthogonal to the“at least one dimension” mentioned above. Accordingly, preferably the nanomembers have at least two orthogonal dimensions smaller than 1 pm, as measurable using standard techniques well known in the art, for example scanning electron microscopy (SEM). Preferably, the nanomembers have at least two orthogonal dimensions smaller than 500 nm. More preferably, the nanomembers have at least two orthogonal dimensions smaller than 300 nm, smaller than 100 nm or smaller than 50 nm.

The size of the nanomember may be selected based on the intended biological subject for investigation. Preferably, the nanomembers should have at least one dimension smaller than at least one dimension of the biological subject being investigated. The precise size of the nanomembers may be limited by the method of production of the nanomembers. However it may be possible to produce nanomembers having at least one dimension as small as e.g. 10 nm. The nanomember density (defined as density = (total area of nanomembers / total area of device)^*100 may be from 20% to 75%, preferably from 40% to 50%. The area of the nanomembers should be calculated considering the total nanomember footprint - for example, nanotubes and nanowires having the same diameter will have the same effective area. If the nanomember density is too high, deformation of the nanomembers in use may be restricted. If the nanomember density is too low, the nanomembers may not effectively support the biological subject(s).

Preferably the shear piezoelectric material is biocompatible. ‘Biocompatible’ is generally defined herein as materials which are not toxic to living tissue, i.e. materials which do not result in catastrophic cell death. Thus, typical piezoelectric materials such as lead zirconium titanate (PZT) are not considered to be suitable. However, inorganic piezoelectric materials such as zinc oxide (ZnO), lithium sodium, or potassium niobate (LNKN) may be appropriate. However, more preferably, the shear piezoelectric material is biologically derived material.

The piezoelectric material may be a non-ferroelectric piezoelectric material. Such materials may be advantageous, as they do not require a poling step in order to exhibit piezoelectric properties. However, this is not essential, as (a) it is possible to perform poling on otherwise-appropriate ferroelectric materials, and (b) some methods of manufacture of a nanomember array (for example, some template wetting methods) can induce‘self-polarisation’ of the nanomember material.

Polymeric piezoelectric materials may be particularly suitable, due to their typically lower stiffness in comparison to non-polymeric piezoelectric materials. Most preferably, the piezoelectric material may be a biologically derived polymer. By using such materials, it may be possible to create a device which more closely mimics selected biological environments.

Specific examples of materials which the shear piezoelectric material nanomembers may comprise include: poly-L-lactic acid (PLLA), poly-D-lactic acid (PDLA), cellulose, collagen, chitin, polypeptides, poly(vinylidene fluoride) (PVDF) or nylon. However, more generally, any polymer having a chiral structure and one or more polar group(s) may be suitable. Of these, poly-L-lactic acid (PLLA) and poly-D- lactic acid (PDLA) may be particularly of interest due to their biocompatibility and biological derivation (for example, PLLA is derivable from bacterial fermentation of carbohydrates such as corn starch and other waste products from agriculture [18]). PLLA may be particularly preferably due to its comparative ease of manufacture relative to PDLA. Additionally, it is not necessary to perform a poling step for PLLA to exhibit piezoelectric properties.

It may be desirable for the piezoelectric material to be at least partly crystalline. Depending on the material involved, high crystallinity can improve the piezoelectric response of the material. Preferably, the piezoelectric material nanomembers have a crystallinity of 20% or more, 30% or more, 40% or more or 50% or more. Crystallinity may be measured according to a wide variety of methods known to the skilled person, including e.g. x-ray diffraction (XRD) or optical techniques. However, it is not essential for the piezoelectric material to be crystalline. For example, the present inventors have found that amorphous PLLA displays a finite piezoelectric response. For polymeric piezoelectric materials, a high degree of chain alignment may provide improved piezoelectric response. Accordingly, where the piezoelectric material is a polymeric material, preferably the material has a Hermans’ orientation factor of 0.5 or greater, where the Herman’s orientation factor is a measurement of chain alignment (0 = no alignment, 1 = perfect chain alignment). The degree of chain alignment may be measuring using standard XRD techniques. Alternatively, chain alignment may also be measured using optical techniques, such as by using polarised light microscopy.

The direction of chain alignment of a polymeric piezoelectric material in relation to the overall shape of the nanomembers may also affect the piezoelectric response of the nanomembers in use. Preferably the polymer chains are generally aligned with the longitudinal axis of the nanomembers i.e. in an (001) orientation. Whilst there may be some measurable piezoelectric response where the polymer chains are generally aligned perpendicular to the longitudinal axis of the nanomembers (i.e. an (010) orientation), the present inventors have found that the piezoelectric response to shear deformation of the nanomembers is far greater when the polymer chains have a (001) orientation.

For polymeric piezoelectric materials, the polymer molecular weight may have an effect on the piezoelectric response. Polymers having a larger molecular weight may display a greater piezoelectric response. However, polymers having a smaller molecular weight may be more facilely manufactured into nanomembers. Where the piezoelectric material is PLLA, preferably the molecular weight M_w is between 80,000 and 150,000. Such a preferred molecular weight range may also apply to other polymeric materials suitable for use in embodiments of the present invention.

The biological material interaction platform may comprise an in plane electrode array in electrical communication with at least a part of the array of shear piezoelectric material nanomembers. The array may be configured for application of and/or detection of an electric field or potential across at least part of the array of shear piezoelectric material nanomembers, preferably in an in-plane direction. ‘In plane’ here refers to the plane of the substrate on which the piezoelectric nanomembers are formed. Preferably, the electrode array is formed beneath the nanomembers. For example, the electrode array may be embedded in the substrate. As the piezoelectric material is a shear piezoelectric material, it is not necessary to provide electrodes above the nanomembers. In other words, preferably the biological material interaction platform comprises a single in plane electrode array which is disposed beneath the nanomembers. In this way, the top surface of the nanomember array (the surface for supporting the biological subject) may be unobstructed by electrodes. This is particularly advantageous for applications such as cell culture and light microscopy, to allow improved access to the top surface of the nanomember array. Furthermore, by embedding the electrodes in the substrate, the biological subject(s) may be preventing from directly contacting the electrodes. The number of electrodes in the electrode array is not particularly limited, and may be selected as appropriate given the desired application of the biological material interaction platform. The electrode density may also be selected as appropriate, although generally it is desirable to have a high electrode density: as the skilled person readily understands, the higher the electrode density, the higher the effective resolution of the biological material interaction platform. The spacing between adjacent electrodes in the electrode array may be from 10pm to 500pm. Preferably, the spacing between adjacent electrodes in the electrode array is 100 14m or less: such spacing may provide appropriate special resolution that is achievable using a variety of methods of production of the electrode array.

The electrode array may be an interdigitated electrode array. Providing such an array can help to create a large electric in-plane electric field over a wide area. However the specific geometry or patterning of the array is not particularly limited, and may be selected as appropriate for the desired application. A‘zigzag’ pattern may be used. Use of a‘zig-zag’ pattern can increase the active area and create a more isotropic electric field over the entire device.

Preferably, each electrode in the electrode array is individually addressable. In this way, when the interaction platform is used in an actuation mode, electromechanical stimulation can be targeted and varied across the device through external electrical signalling. When the interaction platform is used in a sensing mode, signals from each electrode in the array of electrodes can be collated to produce an image corresponding to the electrical response produced by shear deformation of one or more of the shear piezoelectric material nanomembers by the biological subject(s) under investigation.

A matrix addressing scheme may be used to implement individual addressing of electrodes in the electrode array. Matrix addressing allows for an array of m x n elements to be controlled using m + n electrodes. Such matrix addressing schemes are well known in the context of e.g. LCD displays.

In a second aspect, the present invention provides use of a biological material interaction platform according to the first aspect for providing local stimulation to one or more biological subjects supported by the shear piezoelectric material nanomembers. The local stimulation may be mechanical, electrical, or electromechanical.

The local stimulation may be achieved through application of an electric field across at least a part of the array of shear piezoelectric material nanomembers. The applied field may be an AC field. Alternatively, the applied field may be a DC field. Application of an AC field may be particularly advantageous in a cellular environment as it may promote increased cell attachment and/or cell proliferation.

In a third aspect, the present invention provides use of a biological material interaction platform according to the first aspect for sensing micro-mechanical forces produced by one or more biological subjects supported by the shear piezoelectric material nanomembers.

In a fourth aspect, the present invention provides a method of making a biological material interaction platform, the method comprising steps of:

hot pressing a piezoelectric material and a nanostructured template at a temperature above the melting point of the piezoelectric material such that the piezoelectric material at least partly infiltrates the nanostructured template; and

removing the template to leave an array of shear piezoelectric material nanomembers.

By using a nanostructured template, it is possible to more carefully control the shape and arrangement of nanomembers in comparison to processes such as electrospinning. For example, it is possible to form vertically aligned arrays of nanomembers, which is not possible using electrospinning. Additionally, this method may offer a more facile method for production of an array of nanomembers as compared to e.g. solution infiltration techniques using a nanostructure template.

The nanostructured template may be selected as appropriate for the intended application of the biological material interaction platform. For example, where it is desired to produce an array of nanorods, nanowires, or nanotubes, the nanostructured template may be a template having an array of through- holes. Anodic aluminium oxide (AAO) templates (also commonly referred to as“membranes” in the art) are one example of a suitable nanostructured template. Typically, AAO templates have a pore size of 500 nm or less.

Preferably, the step of hot pressing the piezoelectric material and nanostructured template is performed for a time of 5 minutes or less, more preferably 2 minutes or less, most preferably 1 minute or less. By providing a time-limited hot pressing step, it may be possible to control the infiltration regime of the piezoelectric material into the nanostructured template. The mechanisms of template wetting are generally governed by the interfacial relationship between the template material and the infiltrating material. It is generally considered that there are two infiltration mechanisms in template wetting:

precursor wetting and capillary infiltration [19-24]. Precursor films spread rapidly on high energy surfaces such as AAO pore walls - typical timescales for precursor wetting is on the order of seconds to minutes. However, capillary infiltration in nanopores generally occurs over the course of several hours to days. Whilst in principle, the two regimes are not mutually exclusive, the large difference in the rate of each process means that by allowing infiltration to occur only over a short time (i.e. time-limiting the infiltration step), infiltration will predominantly or exclusively occur by precursor film wetting. This may be desirable, as infiltration by precursor wetting can preferentially produce nanotubes, whereas infiltration under capillary infiltration will typically produce nanorods or nanowires. As described above, nanotubes may be preferable to nanorods or nanowires for certain applications due to the larger piezoelectric response which they may exhibit in response to shear deformation.

The hot pressing may be performed at a rate of at least 1 mm/min, for example at about 6 mm/min. Preferably the hot pressing is carried out at a rate of 20 mm/s or less. Performing hot pressing at extremely high rates may cause some system instability in regards to accurate load control.

The hot pressing may be performed at a pressure of 75 kPa to 7.5 MPa. Increasing the pressure used during hot pressing can reduce the thickness of a residual film supporting the array of nanomembers.

The method may include a further step of annealing the piezoelectric material at a temperature below the melting point of the piezoelectric material (T_m) to induce crystallisation of the piezoelectric material.

Preferably, the annealing temperature is from 0.5-0.8 T_m. As discussed above, higher crystallisation may provide for an improved piezoelectric response. The annealing step may be performed for a time or 30 minutes or more. If annealing is carried out for a shorter time, minimal or no crystallisation may occur.

The annealing step may be performed for a time of e.g. up to 30 hours. Annealing for longer than this may be unnecessary and energy-inefficient. The method may comprise a further step of pre-heating the piezoelectric material and/or the nanostructured template before hot pressing. Pre-heating the piezoelectric material and/or the nanostructured template may avoid temperature fluctuations in the molten piezoelectric material as it is contacted with the template during the initial stages of hot pressing - i.e. pre-heating can help to maintain a uniform temperature in the piezoelectric material melt, and thus help to avoid rapid quenching at the polymer-template interface.

The step of removing the template may be performed in any suitable manner known to the skilled person. However, preferably, the template is removed using an etching process. In this way it may be possible to remove the template with minimal or no damage to the nanomembers.

The method may further comprise steps of:

forming an electrode array on a donor substrate;

hot pressing the donor substrate and electrode array with the piezoelectric material and the nanostructured template at a temperature above the melting point of the piezoelectric material such that the electrode array becomes embedded in the piezoelectric material; and

removing the donor substrate.

In this way, it is possible to provide a device having an embedded electrode array. Providing an embedded electrode array may be advantageous for a number of reasons. Firstly, it can provide for better electrical connection between the electrodes and the piezoelectric material nanomembers.

Secondly, by embedding the electrodes in the piezoelectric material, the biological subject(s) may be preventing from directly contacting the electrodes during use of the device.

The piezoelectric material may be provided in particulate form between the donor substrate and electrode array on one side and the nanostructured template on the other side. Provision of the piezeoelectric material in particulate form ensures that there are interstitial spaces available between the piezoelectric particles, such that when the piezoelectric material melts, there is provided space for the electrode array to be embedded into the piezoelectric material.

The donor substrate may be e.g. a foil layer, such as Al foil - Al foil is readily available and cheap.

However, a wide range of donor substrates may be used, including but not limited to: glass, Kapton™, acetate, thermoset polymers, and/or ceramic materials (e.g. alumina). The donor substrate may be selected considering the adhesive properties of piezoelectric material, so that the donor substrate can be more easily removed from the piezoelectric material after the hot pressing step.

The step of forming the electrode array may be performed using an aerosol jet printing (AJP) process. Aerosol jet printing may provide a convenient method for high resolution printing of electrodes. The electrodes may be printed using a conductive nanoparticle ink. One preferred ink suitable for use with aerosol jet printing is an ink comprising metal nanoparticles, specifically silver nanoparticles. The size of the nanoparticles may be selected as appropriate for the AJP apparatus used. The method may include a step of curing the printed electrode array prior to hot pressing. Curing the electrode array may be required to remove e.g. unwanted solvents from the printed ink. Additionally curing the electrode array may additionally help to strengthen the electrodes.

The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

Summary of the Figures

Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:

Figure 1 shows (a) surface plots of the simulated potential in a PLLA nanowire and nanotube (40 nm wall thickness), deformed as end-loaded cantilevers by a 100pN force directed as indicated by the grey arrows; and (b) plots showing potential of the end point (indicated by black arrow) and displacement shown as a function of wall thickness.

Figure 2 shows (a) surface plots of the simulated displacement of in a PLLA nanowire and nanotube on application of an electric field as indicated by the black arrow; (b) a plot showing variation in end point displacement with wall thickness for the same applied field.

Figure 3 shows (a) schematic representation of (010) and (001) polymer chain orientation; (b) surface plots of the simulated piezoelectric potentiation as a function of polymer chain orientation on application of a 100pN force directed as indicated by the grey arrows in PLLA nanotubes with 20nm wall thickness; (c) surface plots of the simulated displacement of in a PLLA nanowire and nanotube as a function of polymer chain orientation on application of an electric field as indicated by the grey arrows.

Figure 4 shows schematically a series of steps in a process of producing a biological material interfacing platform.

Figure 5 shows schematically a series of steps in a process of producing a biological material interfacing platform comprising an electrode array.

Figure 6 shows various view of a pristine AAO template captured using a FEI Nova NanoSEM.

Figure 7 shows (a) a typical sample produced via a melt-pressing method as shown in Fig. 4; (b) a top- down view of PLLA nanotubes, displaying their hollow nature; (c) cross-section of a sample, showing the underlying piezoelectric residual film (substrate) and nanotube array on top of this; (d) a sample edge, showing transition between the nanotextured region and surrounding film.

Figure 8 shows the distribution in (a) the outside diameter and (b) the wall thickness of PLLA nanotubes fabricated via a melt-pressing method as shown in Fig. 4. The average diameter matches well with the measured pore size of the AAO template. Figure 9 shows the dependence of film thickness and nanowire length as a function of (a) pressing temperature; (b) force; and (c) time.

Figure 10 shows x-ray diffraction patterns of annealed PLLA nanotubes, annealed at different temperatures.

Figure 11 shows estimated crystallinity as a function of crystallisation temperature, T_c.

Figure 12 shows the evolution of the d 110/200 interplane spacing as a function of T_c.

Figure 13 shows typical DSC scans of film and nanotubes, both as-made and after annealing at T_c = 140°C.

Figure 14 shows typical DSC scans of film and nanotubes under rapid heating (100°C/min).

Figure 15 shows a comparison between deformation of amorphous and crystalline nanotubes as measured by PeakForce quantitative nanomechanical property mapping (QNM).

Figure 16 shows a comparison between properties of amorphous and crystalline nanotubes as measured using piezoresponse force microscopy.

Figure 17 shows the percent cell attachment for four different types of PLLA sample: amorphous film crystalline film, amorphous nanotubes (NTs) and crystalline NTs. Incubation period was 2 hours.

Crystalline samples were prepared by heat treating at 120 for 1 hour. ^* indicates significance to 0.05 level, ^** indicates significance to 0.01 level, as assessed by two-sample T-test assuming unequal variance.

Figure 18 shows SEM images of Human Dermal Fibroblast cells growing on (a, b) amorphous PLLA nanotubes and (c, d) crystalline PLLA nanotubes. Cells were seeded at a density of -25 x 10⁴ cells/ml and incubated for 72 hours.

Figure 19 shows fluorescence microscopy images of HDF cells on (a) standard Tissue Culture Plastic (TCP); (b) PLLA nanotubes (amorphous); and (c), (d) PLLA nanotubes (crystalline).

Figure 20 shows a plan view of an example PLLA nanotube device with three sets of transferred interdigitated embedded electrodes sharing a common ground. The device is supported on an adhesive layer of Kapton™ film.

Figure 21 shows cross-sections through the device of Fig. 20 perpendicular to the direction of the electrodes, showing the nanotubes with embedded silver lines.

Figure 22 shows (a) a combined phase and DAPI image of an entire device assembled from 200 overlaid and tiled individual images, and (b) a magnified view of the portion of the device indicated by the dashed rectangle in Fig. 22(a).

Figure 23 shows a fluorescence microscopy image of DAPI stained nuclei supported by the device shown in Fig. 22. The electrode pattern can be seen weakly fluorescing in the background. Figure 24 shows (a) a heat-map of nuclei count of a PLLA device seeded with a cell density of 50 x 10⁴ cells/ml; (b) a heat-map of nuclei count of a comparative polystyrene device seeded with a cell density of 50 x 10⁴ cells/ml.

Figure 25 shows typical SEM images of AJP printed silver lines on glass, ink Prelect TPS 50 G2 Ag nanoparticle by Clariant. The influence of substrate temperature is demonstrated.

Figure 26 shows (a) SEM image of crystalline isotactic polypropylene (PP) nanotubes; (b) X-ray diffraction of annealed and quenched PP samples, annealed samples being « 40 % crystalline (by volume); (c) Attachment of HDFs to amorphous and crystalline polypropylene (PP) nanotubes, with standard TCP as a control.

Figure 27 shows a schematic outline of a nanotube-based cell capacitance sensor according to the invention.

Figure 28 shows (a) the circuit used by the CapSense® component to monitor changes in the sensor capacitance (Cs) over time; and (b) The response of one device to 800 pi PBS under ambient conditions.

Figure 29 shows the change in capacitance of three nanotube-based cell capacitance sensors, each under different conditions: (i) with growth media and cell suspension (800 pi at 50 x 104 cells/ml); (ii) only growth media, no cells and (iii) completely dry, without cells or media.

Figure 30 is a schematic showing electrode geometry of a nanotube-based cell capacitance sensor according to the invention.

Figure 31 is a schematic exploded view of a device holder used for characterisation of devices disclosed herein.

Detailed Description of the Invention

Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

Nanomember simulation and shape selection

As discussed above, the precise shape and size of nanomembers as used in the present invention is not particularly limited. However, the present inventors have realised that specific shapes may provide for better function of the biological material interaction platform according to preferred embodiments of the present invention. As discussed above, nanowires, nanorods and nanotubes are desirable shapes due to their high aspect ratio. Computational modelling (FEA) was used to investigate the piezoelectric response of PLLA nanostructures (specifically nanowire and nanotubes) subjected to bending. The results of this simulation suggested that nanotubes, rather than solid nanowires, are more suitable for the proposed application and that the molecular orientation within the nanotube may also be important. COMSOL Multiphysics was therefore used to simulate the piezoelectric response of PLLA nanowires and nanotubes when deforming as end-loaded cantilevers. Both structures were 1 mm long and had an outside diameter of 200 nm. The wall thickness of the nanotube was varied between 20 nm and 70 nm. A boundary load of 100 pN was applied along the x direction to the uppermost surface of the structure (z direction parallel to the nanostructure axis). This load was chosen as it is within the range of biologically relevant forces.

Fig. 1(a) shows surface plots of the simulated piezo-potential for a nanowire and nanotube (40 nm wall thickness) as result of the bending force (grey arrow). The average potential and displacement of the end point of the structure (marked with black arrow) are also plotted as a function of wall thickness in Fig.

1 (b). The displacement of the end point demonstrates that the hollow nanotubes are more compliant than the solid nanowire. This is as expected, since removing material from the centre of the structure to create a tube reduces its second moment of area. Both tube and wire develop a piezoelectric potential in response to bending, however when plotted on the same scale the piezoelectric response of the wire is negligible in comparison to the tube. As the tube deforms, the convex face develops a positive potential, while the concave face develops a negative potential of equal magnitude. The magnitude of the potential increases with the distance from the fixed end of the tube. A potential also develops on the inside of the hollow structure, with the same sign and similar magnitude to that generated on the outer surface. The piezoelectric response is found to become more significant as the wall thickness is reduced. The rate at which the piezo-potential increases with decreasing wall thickness is greater than the rate of change in the displacement. This indicates that the increase in piezoelectric output is not solely due to the increased flexibility of thinner walled tubes.

In an end-loaded cantilever with a high aspect ratio, the shear stress in the structure is generally small - the majority of the deformation is accounted for by tensile and compressive forces above and below the neutral axis. However, in thin-walled tubes, the shear stress component becomes far more significant. This explains the observed trend in piezopotential with wall thickness. A lower shear strain occurs in the solid wire, hence the developed potential is low compared to the developed potential in the nanotube. As the wall thickness of the nanotube is reduced, not only does the overall strain increase (for the same applied force) but a greater proportion of this displacement is accommodated by shear strain. Therefore the piezoelectric response is enhanced as the walls become thinner.

Fig. 2 shows surface plots of the simulated displacement of a PLLA nanowire and nanotube (40 nm wall thickness) on application of an electric field as indicated by the black arrow, along with a plot showing variation in end point displacement with wall thickness for the same applied field. It can be seen that there is a maximum displacement at approximately 50 nm wall thickness.

The piezoelectric properties of the simulated nanowires and nanotubes shown in Fig. 1 and Fig. 2 are defined such that the piezoelectric’3’ axis is parallel to the long axis of the nanomember. In PLLA this translates to the polymer chain axis also being parallel to the longitudinal axis of the nanomember. This chain alignment may influence the piezoelectric performance of the nanomember. There are two main cases to consider: the aforementioned case where the chains are parallel to the length of the member, and the case where the chains are radially aligned in the nanomember. Fig. 3 shows the simulated potential for these two cases in a nanotube, labelled (001) and (010) orientation respectively.

A piezoelectric potential is developed in both cases, however the distribution of the potential across the surface differs between to the two orientations. The potential of the (001) orientation is described above. In the (010) case, the potential is developed on the faces along a direction perpendicular to the applied force. Moreover, the potential is evenly distributed along the length of the tube and also extends to its inner surface.

The simulation results show the large difference in the magnitude of the piezo-potential between the two cases. As shown in Fig. 3(b), the potential developed in the (001) case is 3 orders of magnitude larger than an equivalent structure with the (010) orientation. Furthermore, as shown in Fig. 3(c), the end point displacement achieved on application of the same electric field is approximately 20 times larger for the (001 ) case as compared with the (010) case.

Whilst the shear strains in the each structure are the same, these strains do not couple to the active piezoelectric mode of the (010) orientation. Accordingly, for best performance, these results show that it is desirable to provide nanomembers having 001 polymer chain orientation.

Fabrication of PLLA nanotubes suitable for use in a biological material interfacing platform

One process for making an array of shear piezoelectric nanomembers is described in work published from the inventor’s own research group:“Direct observation of shear piezoelectricity in poly-l-lactic acid nanowires” Smith, Michael & Calahorra, Yonatan & Jing, Qingshen & Kar-Narayan, Sohini. (2017). APL Materials. 5. 074105. 10.1063/1.4979547. This publication describes a solution template-wetting process for producing solid, rod-like PLLA nanowires. Such nanowires may be suitable for use in a biological material interfacing platform according to the present invention following appropriate device configuration to allow for suitable application of or detection of an electric field or potential across at least a part of the array. However, as discussed above, better device performance may be achieved where the nanomembers comprise or consist of nanotubes.

Accordingly, the present inventors have devised a novel method for the production of nanotubes for use in a biological material interfacing platform, using a‘melt-wetting’ process. Whilst the present results are demonstrated using PLLA, it is theorised that this method is equally applicable to a wide range of other materials, provided that the infiltrating material has lower melting point than the material of the nanostructured template such that deformation of the template during the hot pressing process is avoided.

Fig. 4 shows schematically a series of steps in this method. Pellets of poly-L-lactic acid (Lactel B6002-2, Sigma Aldrich) with molecular weight 85,000 - 160,000 were used. In essence, these polymer pellets were heated and pressed into the AAO template. Whatman Anodise inorganic filter membranes (13 mm, 200 nm pore diameter, Sigma Aldrich) were used as templates. Fig. 6 shows two views of these templates. Filter membranes such as these are often bonded to a polypropylene (PP) support ring to allow them to be handled. However, the high

temperatures involved with the melt template wetting method meant that un-supported membranes were used as otherwise the PP ring would have melted and contaminated the sample.

A custom-built hot stage was used for melt template wetting of PLLA. This consisted of a stainless steel cylinder, 60 mm in diameter, heated with a 375 W band heater [RS Components]. Temperature was controlled via a PID controller (Red Lion PXU, RS Components) and a type K thermocouple, inserted into a small radial pilot hole just beneath the surface. A platinum thermometer (PT 100) was used to calibrate the surface temperature of the stage relative to the thermocouple temperature reading. No significant difference between surface and internal temperature was recorded, thus the temperature reported by the PID controller was used for all subsequent analysis.

The hot stage was mounted into an Instron mechanical testing machine and heated to the desired temperature. A glass slide wrapped in aluminium foil (foil optional, but aids release of the PLLA from the glass slide) was placed onto the hot stage and allowed to heat up. Once hot, two PLLA pellets (of about 8 mg each) were placed onto the aluminium foil. At the same time, a second glass slide (not wrapped in foil) and the AAO template were also placed on the hot stage to heat up. After two minutes the template was positioned on top of the polymer pellets, followed by the second glass slide, and gentle pressure was applied manually to settle and align the stack. Once centred, a PTFE block was placed above the stack and the machine grip was lowered until just touching. The PTFE block was used to insulate the machine grip and load cell from the hot stage. This step is shown schematically in Fig. 4(a).

As shown in Fig. 4(b), the stack was then compressed at 6 mm/min until the desired force set point was reached, where it was then held under load control for the chosen amount of time. After pressing, the sample was removed from the heated stage and cooled to room temperature in air (Fig. 4(c)).

To promote crystallisation of the samples, the pressed stack was placed inside a pre-heated oven (HeraTherm - ThermoScientific) with a small steel weight on top to prevent warping. The sample was annealed for the chosen time and temperature and then allowed to cool slowly in the oven to room temperature. This annealing stage is not shown in Fig. 4, and indeed is optional. However, it may be advantageous, as discussed in further detail below.

After removing the bottom glass slide and gently peeling off the foil from the underside of the polymer film, the upper glass slide was placed in water and left overnight to release the adhered polymer film (now with embedded template) (Fig. 4(d)). Once released, the polymer film was secured to the base of a glass petri dish and submerged for 3 hours in 40 % (v/v) phosphoric acid in H2O to etch the alumina template (Fig. 4(e)). After etching, the sample was washed five times in Dl-water and once in ethanol.

The result of this process is a thin nanotextured PLLA array supported by a relatively thick annular polymer film (substrate), as shown schematically in Fig. 4(f). Samples produced according to this method were then characterised, as discussed below. For comparison, analogous‘film’ samples were prepared in an identical manner, except without the template. Comparison films were manufactured to be between 50 mm and 100 mm thick - large enough to be considered as bulk material.

One of the main advantages of using a method as described in relation to Fig. 4 is that because the nanomembers produced according to this method comprise an array of vertically aligned PLLA nanotubes, there is no need for a top electrode. The piezoelectricity of the nanotubes is sensitive to in plane electric fields. This means the top surface of the device can be open for e.g. cell culture, while electrodes can be incorporated in the plane of the device.

Fabrication of electrode array by Aerosol Jet Printing ( AJP )

Whilst a number of different methods of forming an electrode array will be known to the skilled person, one particularly suitable method for use in the present invention is aerosol jet printing (AJP). AJP may be particularly suitable due to the fine resolution achievable - it is generally desirable to provide electrode space on a similar length scale to the scale of the biological subject(s) of interest, which may typically be in the region of tens of nanometers to tens of microns.

Aerosol Jet Printing (AJP) is a versatile technique suitable for large-area, fine-feature patterning of both rigid and flexible substrates with a multitude of functional inks. In particular, AJP can tolerate ink viscosities between 1 and 1000 cP, with printing resolution of the order of 10 pm, thus making it attractive for flexible and printed electronics.

In AJP, an atomiser is used to create an aerosol of ink droplets typically 1 to 5 mm in diameter. Both ultrasonic and pneumatic atomisation methods are possible. This aerosol is then entrained in a nitrogen carrier gas and fed towards the deposition head (dep-head) at flow rate Q_atm. Within the dep-head, a second gas flow - the‘sheath’ flow, at rate Q_Sh - is introduced and surrounds the atomiser gas flow.

These gases flow co-axially through the dep-head and out of the tip, the opening of which can be chosen between 100 and 300 mm in diameter. The focussing effect of the sheath gas means that features down to 10 mm can be printed under appropriate conditions. By mounting the substrate on an automated positioning system, wide area deposition of these fine features is possible. Typically, the positioning system is not the limiting factor for print resolution.

In AJP, the flow rate of each gas - Qatm and Q_Sh - can be varied independently, allowing for the dynamics of the aerosol jet to be controlled with reasonable precision. Many other aspects of the deposition process can also be varied, such as tip diameter, tip-substrate separation (tip height), substrate temperature, ink temperature and print speed.

Substrate temperature is an important parameter for optimal line quality, and it must be determined individually for each substrate material. Some example SEM images showing AJP printed silver lines, and the influence of substrate temperature, are shown in Fig. 25. One example method of production of an electrode array using AJP is as follows:

A 15 x 15 cm² piece of Al foil as a donor substrate (although other substrates may be used) was fixed to the platen of an Optomec AJ200 aerosol jet printer and held flat using the built-in vacuum system. Once mounted, the foil was cleaned with ethanol.

The electrode array was designed in AutoCAD 2016 and printed onto the foil using Prelect TPS 50 G2 Ag nanoparticle ink (Clariant) in an ultrasonic atomiser (UA Max, Optomec). Printing parameters are shown in Table 1 , below.

Table 1: AJP printing parameters for electrode transfer

Several electrode arrays were tiled to fit onto a single sheet of foil. After printing, the sheet was transferred to an oven (HeraTherm - Thermo Scientific) and cured for 3 hours at 200 °C.

Once cured, individual electrode arrays were cut from the sheet with a scalpel and wrapped around glass slides. From here, the device manufacturing process (shown in Fig. 5) is broadly similar as described previously in relation to Fig. 4. As shown in Fig. 5(a), the glass slide and foil were placed on a hot stage with the polymer, template and a second glass slide, before undergoing hot pressing. During the hot pressing, the molten polymer envelops the printed silver lines. When the Al foil is subsequently peeled off, the printed silver remains embedded in the polymer. The result is a flat surface with the printed silver recessed into the polymer film. The AAO template can then be removed in the same manner as described above in relation to Fig. 4.

Fig. 7 shows a typical sample produced via a melt-pressing method as described above. The result is a thin’nano-textured’ film surrounded by a thick annular supporting ring (where the polymer has spread beyond the template). Fig. 7(a) is a view of the whole sample. The central area appears white due to the nanoscaled dimensions of the structures scattering the incident light. Fig. 7(b) is a top-down SEM image of the PLLA nanotubes, displaying their hollow nature. Fig. (c) shows a sample cross-section, showing the underlying piezoelectric film (substrate) and nanotube array on top of this. Fig. 7(d) shows the sample edge and the transition between the nanotextured region and surrounding film, demonstrating that the polymer does not infiltrate the entire length of the template.

Results from image analysis are presented in Fig. 8 and show (a) the distribution in nanotube outer diameter and (b) the distribution in nanotube wall thickness, with average values 305 ± 24 nm and 56 ±

10 respectively. The outside diameter is larger than expected, since the templates used have a nominal pore size of 200 nm according to manufacturer specifications. However, inspecting the templates with SEM reveals that there is a large variation in pore size, and that the AAO template pore diameter is closer to 300 nm than 200 nm.

The influence of the processing conditions were briefly investigated using cross-section SEM images.

Fig. 9 shows the influence of (a) pressing temperature, (b) force, and (c) time on the thickness of the film beneath the nanotubes (the substrate) and the length of the nanotubes themselves. In all cases, increasing the value of the processing parameter resulted in a decrease in thickness of the supporting film and an increase in the length of nanotubes. The mechanisms governing the height of each are quite different. The thickness of the film is determined solely by the lateral viscous flow of polymer from underneath the template. Higher temperature results in lower viscosity and thus a larger flow rate;

greater force generates a faster rate of flow; longer time allows for a greater volume of material to flow from underneath the template. The length of the nanotubes is dictated by the wetting regimes involved. Given that nanotubes have formed, precursor wetting is the dominant growth mechanism. Higher temperatures will change the relative surface energies of the melt and template, thus creating a greater driving force for the spreading of the precursor film. Although the exact mechanism of this spreading is not known [25], it is reasonable to assume that increased chain mobility at higher temperature also aids in the spreading of the film.

The wetting of the pores is a result of surface energy considerations rather than physical extrusion. It is possible that applying a higher force simply acts to increase the thermal contact between elements of the system, thus increasing the temperature and therefore nanotube length. The applied force is more important for determining the thickness of the supporting film. A longer time allows for the precursor film to spread further along the nanopores. The templates used in this work are nominally 60 14m thick, so it is interesting to note that during the time-scales observed here, the nanotubes do not fill the entire length of the pores of the AAO template.

In the below discussion of characterisation of PLLA nanotubes, a press temperature of 190 °C, force of 100 N and time of 1 minute were used unless otherwise stated.

Heat treatment/annealing of PLLA nanotubes crystallisation characterisation

X-ray diffraction (XRD) was initially used to investigate the crystalline structure of the nanotubes. It was found that the as-made film and nanotube samples were amorphous. Samples were therefore heat treated post-pressing to induce crystallisation. This was carried out by placing samples in a pre-heated oven at the specified annealing temperatures for one hour. Representative XRD patterns of annealed PLLA nanotubes are shown in Fig. 10, normalised to the height of the main reflection to enable comparison. The two main peaks are the (1 10)/(200) peak at approximately 17 deg, and the (203) peak at approximately 19 deg. Fig. 1 1 shows estimated crystallinity as a function of annealing temperature, T_c.

As can be seen from these figures, performing an annealing step causes the polymer to crystallise into a well-defined crystal structure. The crystallinity was estimated by fitting peak functions to the amorphous and crystalline signals. Increasing the annealing temperature increases the crystallinity of the nanotubes from amorphous to 54 ± 5 %. For similar analysis on comparative films, the maximum crystallinity of the films was found to be 42 ± 4 %. A similar effect was also observed with increasing time, although over the time-scales investigated the change is not particularly significant. An hour was found to be sufficient to produce crystalline samples and was used as the annealing time for all subsequent heat treatments.

There are several crystal phases of PLLA, with a’ and a being the most readily accessible with conventional processing methods. Generally, crystallisation of PLLA at temperatures below 1 10-120 °C will lead to the formation of the a’ phase [26-28] while crystallisation above these temperatures will lead to the a phase. The crystallographic differences between the phases are subtle, and thus changes in the respective XRD patterns are also minor. Nonetheless, the transition between a’ and a can be observed here in at least the sample annealed at 140 °C as a reduction of the lattice parameter and the introduction of the (103), (204), (016), (207) and (0101 ) reflections (marked by vertically dashed lines).

Fig. 12 shows the evolution of the di 10/200 interplane spacing as a function of T_c in both films and nanotubes. The separation shifts to lower values with increasing temperature, however the behaviour of films and nanotubes are subtly different. The majority of the reduction in lattice parameter of the films occurs at low annealing temperatures, while the nanotubes must be annealed at a higher temperature before a similar reduction is seen. As mentioned above, a reduction lattice parameter can be associated with the a’ ® a transition. This result therefore suggests that the disordered-to-ordered transition is suppressed in the nanotubes.

Differential scanning calorimetry (DSC) was used to further investigate the polymer crystalline properties. Typical DSC scans are shown in Fig. 13 for films and nanotubes, both as made and after annealing at T_c = 140° C. Significant cold crystallisation events can be seen in both as-made samples, as well as crystal perfection immediately preceding the main melting endotherm. This suggests a significant amorphous component in the as-made samples. Indeed, analysing the data gives crystallinities of -1.6 % and 4 %, respectively, for the film and nanotubes: i.e. amorphous within the experimental error. The large amorphous content of the as-made samples is also apparent by the presence of glass transition events, as shown in Fig 14 (DSC scans under rapid heating of 100°C/min), for both film and nanotube samples heated at 100 °C/min. This rapid heating is sufficient to suppress cold crystallisation, although a small exotherm can be seen in the nanotube sample.

The conclusion of the above characterisation is that while the melt-pressing method produces amorphous nanotubes as-made, these can be readily crystallised by subsequent heat treatment up to a crystallinity of 54 ± 5 %. The nanotubes can exhibit polymorphism, although the a’ a transition may be slightly suppressed.

PeakForce™ Quantitative Nanomechanical Mapping

PeakForce™ Quantitative Nanomechanical Mapping (PF-QNM) is a scanning probe microscopy technique developed by Bruker Ltd. It can be used to map mechanical properties such as modulus, adhesion, deformation and dissipation while simultaneously recording sample topography.

Fig. 15 shows a comparison between deformation of amorphous and crystalline nanotubes as measured by PeakForce quantitative nanomechanical property mapping (QNM). It can be seen that amorphous nanotubes are more readily deformable than crystalline nanotubes. Accordingly, variation of the crystallinity of the nanotube may be a useful method to adjust e.g. effective stiffness of an array of nanotubes.

Appropriate calibration of the system was performed, and a peak-force set point of 100 nN was chosen for this experiment. This was large enough to create a reliably measurable deformation without causing excessive wear the tip. A scan rate of 0.1 Hz with 512 samples per line was used to collect the QNM data. This is a relatively low scan rate, but was necessary to allow adequate time for the feedback system to accommodate the abrupt changes in sample height. After capture, the data was analysed in Gwyddion 2.45. Processed data was exported for plotting in Origin 2016.

PFM Characterisation

A Bruker Multimode 8 AFM system was used for PFM imaging using Bruker MESP-RCV2 conducting cantilevers. The properties of these cantilevers are shown in Table 3.3 Individual nanostructures were scanned at a rate of 0.1 Hz to reduce tip velocity and therefore the chances of the nanowire being physically moved by the tip. The applied voltage amplitude was varied during the study while the frequency was kept constant at 35 kHz for nanowire samples and 125 kHz for nanotube samples.

Fig. 16 shows a comparison between properties of amorphous and crystalline nanotubes as measured using piezoresponse force microscopy. Fig. 16(a) shows height images for as-made (amorphous) and crystalline PLLA nanotubes acquired during PFM scans. Fig. 16(b) shows uncalibrated lateral PFM signals of individual PLLA nanotubes for an applied oscillating potential of 8V at 125 kHz. Both images are displayed in the same scale. Fig. 16 (c) shows column averages of the height and lateral PFM data. PFM scans were performed with an applied potential of amplitude 0 and 8 V to verify the

electromechanical origins of the signals. It was found that the absolute magnitude of the lateral PFM signal is approximately 1.5 times larger for crystalline nanotubes compared to amorphous nanotubes, but that both crystalline and amorphous nanotubes display some piezoelectric activity. orientation characterisation

Comparison of experimental and simulated XRD patterns predicted that the nanotubes have an axial chain orientation. This molecular alignment was verified conclusively with polarised light microscopy (POM).

POM is an effective way to investigate the birefringent properties of materials. Birefringence occurs when the refractive index of a material is dependent on the direction of propagation and direction of polarisation of incident light. It is characteristic of anisotropic materials and is therefore common in many polymers due to the inherent anisotropy in directions parallel and perpendicular to the chain axis. By comparing the colour of the nanomembers to the colour of drawn films (in which chain orientation will be (001 ) in the drawing direction) using polarised light microscopy, it is possible to qualitatively determine the approximate chain orientation within the nanomembers. For a quantitative optical measurement, a transmission light microscope with a tilting compensator could be used to measure and quantify the value of birefringence of the nanomembers. Alternatively, chain orientation could be determined using TEM with selected area electron diffraction (SAED) analysis.

It is found that in samples made according to the melt-wetting process discussed above, chain orientation is predominately in the (001) direction (i.e. chains aligned with the longitudinal axis of the nanotubes). Without wishing to be bound by theory, the present inventors hypothesise that this is due to the precursor wetting regime under which infiltration of the nanostructured template occurs. Whilst the exact mechanism for precursor film spreading is not well understood, it is reasonable to assume that the elongational flow of the polymer along the pore walls may lead to significant alignment of the polymer chains along the length of the pore. It is proposed the thin wall of the nanotube might confines the chains to this orientation, and these aligned chains then act as nucleation sites for (001) oriented crystals to form upon annealing.

Cell attachment characterisation

The attachment of cells to PLLA films and nanotubes has been assessed, both in an amorphous and crystalline state, without any form of coating. Details of the sample preparation are given below.

Cell culture

Human dermal fibroblasts (HDF) were cultured under standard TC conditions (37 °C, 5 % C02) in T75 flasks with Dulbecco’s modified Eagle’s medium (DMEM, Sigma Aldrich) containing 10% fetal bovine serum (Sigma-Aldrich) and 1 % streptomycin/penicillin (Sigma Aldrich). Cells were prepared for experiments on the materials by detaching from the cell culture flask using 0.05% (w/v) trypsin / 0.02% (w/v) EDTA (Sigma-Aldrich) and re-suspended in growth media at a density of 5 x 10⁴ cells ml ¹

(measured with a haemocytometer). Cell attachment assay

A lactate dehydrogenase (LDH) assay was used to assess the adhesive properties of the PLLA samples. Samples were prepared as described above in relation to Fig. 4. Crystalline samples were prepared by heat treating at 120 °C for 1 hour. Three 5 mm circles were cut from each sample using a biopsy punch so that the assay could be run in triplicate. Each piece was washed in ethanol, allowed to dry and then placed into a separate well of a 48 well plate (CytoOne). Tissue culture plastic (TCP) acted as a positive control for the experiment. When growth media was added, it was often found that samples would float as a result of gas bubbles nucleating on the rough sample surface. To prevent this, a thin polymerferrite magnet was fixed to the underside of the plate and small gold coated neodymium magnets were placed on top of the samples in each well (1.5 x 0.75 mm N52 gold plated disc, Gaussboys, USA). The small magnetic attraction through the base of the plate was sufficient to prevent the samples from floating.

To ensure that the presence of the magnets did not influence the cell behaviour, ferrite/neodymium magnet pairs were also added to wells containing only tissue culture plastic (i.e. without sample material) and compared to the TCP positive control. Cells were prepared as described above. The assay was carried out as follows. A 600 ml volume of cells was added to each well and the entire plate was incubated at 37°C and 5% C02 for a set amount of time. After this incubation period, the growth media was removed and the samples were gently washed with 2 x 500ml phosphate buffered saline (PBS) solution. The samples were then transferred to a new 48 well plate. Cells were lysed using a 200 ml volume of 2% Triton X-100 in de-ionised (Dl) water at room temperature over a period of 2 hours. A 100 ml volume of the lysate was then transferred to a new 96 well plate, where 100 ml of LDH substrate (Sigma Aldrich) was also added to each well. This was left for 20 minutes at room temperature, after which the plate was analysed spectrophotometrically using a plate-reader (SpectroStar Nano, BMG Labtech). Converting the values of absorbance back to percentage attachment was achieved by including a set of calibration wells on each plate. In these wells, a known percentage of cells was added - 0%, 10%, 25%, 50% and 100 % - with the remainder of the volume being made up with plain growth media (DMEM) to 600 ml. The values of absorbance were then plotted against the percentage of cells added and fitted to linear model. This linear relationship was then used to convert values of absorbance from the samples into a percentage attachment.

The results of this assay are shown in Fig. 17. Fig. 17 shows the percent cell attachment for four different types of PLLA sample: amorphous film crystalline film, amorphous nanotubes (NTs) and crystalline NTs. Both amorphous and crystalline PLLA films are non-adhesive; no cells attached to either type of film.

Nano-texturing the PLLA material results in significantly greater cell attachment. After two hours, approximately 7.5 % of the cells added had attached to amorphous nanotubes. Over the same time period, roughly 14 % of the cells attached to a crystalline nanotubes. These values of attachment are relatively low in comparison to standards such as Tissue Culture Plastic (TCP), which was measured to be around 80% under the same conditions. However, the relative values of attachment are more interesting than the absolute values. Both conditions of nanotubes display some degree of cell attachment, which was not observed in equivalent films. This difference is significant to at least a 0.05 level as assessed by a two-sample T-test.

Without wishing to be bound by theory, the inventors theorise that the lower effective stiffness of the nano-textured surface in comparison to the bulk film may be influencing cell attachment here, as it is well known that the modulus of a surface is important in determining the behaviour of cells [3,4,29]. However, possibly other factors such as relative hydrophilicity of the surfaces could also be the reason for this difference.

Crystalline nanotubes display greater cell attachment compared to amorphous nanotubes. The difference is significant to a 0.01 level. As discussed previously, crystallisation of the nanotubes leads to an increased modulus, greater surface potential and increased piezoelectric activity, each of which may promote cell attachment in relation to amorphous films.

Cell-nanotube interaction

It is useful to investigate exactly how the cells interact with the nanotubes. Scanning Electron Microscopy (SEM) was used to directly observe this interaction using fixed and dehydrated samples. Dehydration was performed after fixing the cells using 5 % gluteraldehdye (Sigma Aldrich) in PBS. Samples were immersed in a serial dilution of ethanokwater in the ratios: 25:75, 50:50, 75:25, 100:0. Samples were then immersed in a serial dilution of hexamethyldisilazane (HMDS, Sigma Aldrich) in the ratios 25:75, 50:50, 75:25, 100:0. Samples were then left overnight for the remaining HMDS to evaporate. Once thoroughly dry, samples were sputter coated with palladium ready for SEM observation.

The cell shape is notably different to that observed when cultured on standard substrates such as TCP. On these rigid substrates, fibroblasts are typically long and thin with a stretched morphology. Softer substrates tend to yield more spherical cells with fewer projections [3]. The high aspect ratio PLLA nanotubes provide an effectively’soft’ surface, hence the cells exhibit more of a spherical shape, yet still with multiple projections and attachments. This is perhaps due to the increased surface roughness and possibly the stimulating influence of piezoelectric potentials in the individual nanotubes. The nanotubes can be seen bending under the influence of the cellular contractile forces.

Fig. 18 shows Human Dermal Fibroblast cells growing on (a, b) amorphous PLLA nanotubes and (c, d) crystalline PLLA nanotubes. Cells were seeded at a density of -25 x 10⁴ cells/ml and incubated for 72 hours.

The nanotube surface was also observed to influence how the cells interact with each other. Florescent cell staining was used to visualise the cells on the sample surface. A rhodamine phalloidin dye (Life Technologies) was used to stain the actin filaments within each cell, and a DAPI stain (Sigma Aldrich) was used to mark the cell nuclei. This allowed the cell number to be counted as well as the cell morphology to be observed. Cells were prepared as described above and incubated on the materials for a predetermined amount of time using the magnet method also described above. Following incubation, the cells were washed with 3 x 500ml PBS and fixed using 5 % gluteraldehdye (Sigma Aldrich) in PBS. The samples were then lysed using 0.5 % Triton X-100 solution in PBS for 5 minutes. The lysate was removed and 100 ml of 0.1 % rhodamine phalloidin 0.1 in PBS added and left at room temperature for 45 minutes. The samples were then washed with 3 x 500 ml PBS. A 100 ml volume of 0.01 % DAPI solution in Dl-water was added to each well and left for 5 minutes. Following this the samples were washed 3 x 500ml in de-ionised water. Florescence microscopy was performed using a Zeiss Axio Observer Z1 phase contrast microscope fitted with a Zeiss Axiocam 503 mono camera (Carl Zeiss Ltd, Cambridge, UK).

The cell behaviour between a nanotube surface and a standard TCP substrate is shown in Figure 19. In Fig. 19 (a) and (b), the cell body is stained with rhodamine phalloidin, nuclei with DAPI. Fig. 19(a) demonstrates the typical elongated phenotype typically associated with fibroblasts when cultured on stiff substrates like standard Tissue Culture Plastic (TCP). Note also how the cells are somewhat isolated from each other, growing independently. Fig. 19(b) shows that when cultured on PLLA nanotubes (a softer substrate than the TCP), the cells are far more likely to cluster together and integrate with each other. Some of the nuclei also appear to have irregular shapes. Fig. 19(c) shows live/dead staining of HDFs on crystalline PLLA nanotubes. Live cells are stained green, dead cells are stained red. Fig. 19(d) shows a higher magnification image of the live/dead staining in Fig. 19(c). These images demonstrate the biocompatibility of the surface.

Device characterisation and cell stimulation

The piezoelectric contribution to the cellular behaviour observed in the previous sections arises as a result of the direct piezoelectric effect: potential due to an applied strain. However, the piezoelectric effect is reversible and thus the indirect effect can also be used, as applying an electric field across the diameter of a nanomember will cause the structure to bend, providing local dynamic electromechanical stimulation to a biological subject in contact with or supported by the nanomember. The amplitude of the nanomember displacement is likely to be small, however research into dynamic mechanical stimulation of cells suggests that even vibrations of nanometre amplitude can have an influence on cell behaviour [9].

As described above in relation to Fig. 5, devices were made incorporating in-plane electrode arrays. Various electrode geometries were trialled as part of or this investigation, but the result discussed below are all from a zig-zag pattern with a separation of 75 mm. This pattern can be seen faintly in the background of Fig. 23. All the devices used in the subsequent investigations included a shear piezoelectric material nanomember array which had been annealed for 1 hour at 120 °C.

Fig. 20 shows an example PLLA nanotube device with three sets of transferred interdigitated embedded electrodes sharing a common ground. The device is supported on an adhesive layer of Kapton™ film. Three sets of electrodes were used so that each device could effectively be considered as a single experiment, with two variables and a control. Figure 21 shows cross-sections through the device perpendicular to the direction of the electrodes, showing the nanotubes with embedded silver lines.

A piezoelectric material in an electric field has a greater apparent stiffness [30] and will also become electrically polarised, as will any dielectric material. Therefore, a static electric field was applied across one set of the electrodes to see if the apparent changes in material electromechanics induced by an external potential had an influence on cell growth.

An oscillating potential was applied to another set of electrodes to determine the influence of dynamic stimulation.

Electric fields in isolation are also known to have stimulating effects on cell function and this too must be considered when analysing the results [31]. The third set of electrodes was grounded on both sides to act as a control against any effects of stimulation.

A 1 V potential was initially used, mainly as this was easily sourced from a microprocessor control board (FreeSoC2, SparkFun Electronics). This potential results in a field of 1.3 x 10⁴ V/m over the interdigitated electrodes. However, greater voltages are possible with more refined equipment. Signals were passed to the device using copper wires fed through a port at the rear of the incubator. As described above, the device had three banks of interdigitated electrodes sharing a common ground. One bank was set at a constant voltage (Vo), another had a sinusoidally varying voltage (Vosincut) and the third had no applied potential. A frequency of 1 Hz was initially used for the oscillating potential as this is within the realm of biologically relevant frequencies, although it should be noted that many researchers have found that electronic and mechanical stimulation applied at much higher’non-biological’ frequencies (of the order of kHz) can have a beneficial effect.

Initially, devices were seeded with 1 ml of cells at a density of 5 x 10⁴ cells/ml and incubated for 72 hours with the stimulating voltages applied from the beginning of the incubation period. Following incubation the devices were washed, fixed, stained with DAPI to mark the positions of the nuclei and imaged across the entire surface. A combined phase and DAPI image of an entire device assembled from 200 overlaid and tiled individual images is shown in Fig. 22(a), and a magnified view of a portion of the device is shown in Fig. 22(b), where the‘zig-zag’ electrode pattern can more clearly be seen.

Fig. 23 shows a fluorescence microscopy image of DAPI stained nuclei from the dashed rectangle indicated in Fig. 22. The electrode pattern can be seen weakly fluorescing in the background.

The experiment was repeated with a cell density of 50 x 10⁴ cells/ml. The number of nuclei in each individual image were counted using the Nuclei Counter plug-in for ImageJ and plotted as heat-maps across the device surface.

A result is shown in Fig. 24(a) - The colour range is fitted to an interpolated model based on the number of nuclei in each of the images that tiled the surface. The total cell number at the end of the incubation period was 210000 cells, significantly more than the 50000 cells seeded at the beginning of the experiment. This suggests that the cells are proliferating. There was a significantly greater cell number over the area stimulated with an AC voltage, as compared to the DC and Ground portions of the device.

A comparative device was fabricated from polystyrene (Sigma Aldrich), an amorphous non-piezoelectric polymer. The procedure to make this device is identical to that for the PLLA device, expect for a higher press temperature (250 °C) to account for the higher melting point. The device was run under the same conditions as described above for the PLLA device.

A nuclei heat-map for a polystyrene device is shown in Fig. 24(b) - there was no significant increase in cell number in the AC stimulated region. In the particular device shown, there is a region of greater cell density but this does not correlate to any of the set of electrodes. This suggests that the hot-spots in the AC stimulated PLLA devices may well be due to the local mechanical stimulation provided by the nanomembers.

Fabrication and cell attachment characterisation of Polypropylene (PP) nanotubes

To investigate the effect of piezoelectricity on cell attachment, nanotubes were also fabricated from isotactic polypropylene (PP) - a semi-crystalline but non-piezoelectric polymer.

Polypropylene (PP) nanotubes were fabricated using the same melt-press template wetting method discussed above for production of PLLA nanotubes except with a higher pressing temperature of 250 °C. Pellets of isotactic PP (Goodfellow) were heated to 250 °C and pressed into the AAO membranes.

Isotactic PP crystallises readily and cooling in air after pressing is insufficient to produce amorphous samples. To avoid crystallisation, samples were quenched to room temperature with a water bath.

Crystalline samples were produced via a heat treatment at 160 °C for 1 hour.

Wide-angle X-ray diffraction (WAXD) of the samples was performed with a Bruker D8 diffractometer in Bragg-Brentano geometry with Cu Ka radiation. X-ray Diffraction (XRD) patterns of the quenched and annealed samples are shown in Fig. 26(b).

Amorphous and crystalline samples of PP NTs were prepared and used to perform the same HDF attachment assay as discussed above in relation to PLLA NTs. Crystalline samples had « 40 % (volume) crystalline fraction, while amorphous samples were less than 5% crystalline. The results, shown in Fig. 26(c) show that there is no significant difference in attachment between amorphous and crystalline PP nanotubes. This is contrary to the case of PLLA nanotubes where crystalline nanotubes displayed significantly greater cell attachment. Significance was assessed with two sample t-test (N = 4 for each condition, ns = not significant; ^* = p < 0.05; ^**

The outcome of the PP control experiment suggests that piezoelectricity has an influence on the level of cell attachment in PLLA nanotubes. Only in the case of PLLA will crystallisation result in a change in piezoelectric properties, and only in the case of PLLA does crystallisation result in an increase in cell attachment. Fabrication and characterisation of nanotube-i sensors

Capacitance-based sensors offer a qualitative, accurate and non-invasive measurement of live cells with minimal hardware requirements. Several‘lab-on-a-chip’ style devices have been reported which monitor the capacitance of insulated electrodes as cells grow across their surface. Arrays of these electrodes can provide information about the attachment, proliferation and mobility of cells in real time with a degree of spatiotemporal resolution. However, those devices reported in literature are fabricated using standard CMOS techniques and therefore consist of hard, rigid materials.

Work was done to produce devices capable of capacitance measurement, by using an array of polymeric nanomembers. Fig. 27 shows a schematic outline of the nanotube-based cell capacitance sensor. The vertical array of nanotubes presents as a soft surface to the cell, whose behaviour is monitored by the embedded electrodes via changes in capacitance.

The device contains a central sensing electrode (9 mm diameter) and an annular shield electrode (10 mm ID, 13 mm OD) divided into three sections. Both electrodes are filled with a hatch to reduce the print time and the parasitic capacitance of the final device. The sensing electrode is filled with an isometric net with 250 pm spacing between vertices, while each section of the shield electrode is filled with an orthogonal grid of 500 pm spacing. A schematic of the electrode geometry is shown in Fig. 30. The central region is the sensing electrode, hatched with an isometric net. The surrounding areas are the shield electrodes. The sensing electrode has three contact pads to allow the quality of the electrical connection to be assessed while the device is mounted and to provide a degree of redundancy should one contact fail. A driven shield is used to reduce the coupling of the sensor field lines to ground, thereby reducing the parasitic capacitance of the device. Inactive sensors are also connected to the shield signal which helps to reduce cross-talk between devices.

A bespoke device holder was used to perform both characterisation and testing. A schematic exploded view of the device holder is shown in Fig. 31. The holder comprises a base, a printed connector circuit and a‘well-segment’. The well-segment bolts onto the base, clamping the device against the connecting circuit. O-rings within the well segment form a water-tight seal around the device while pressing the contact pads in the device against another set of contact pads on the connector circuit, thus forming an electrical connection. The connector is wired to a solderless breadboard so that standard jumper wires can be easily inserted. When assembled, the outside dimensions of the holder are equivalent to a standard well plate. The well segment contains 3 wells, each 12 mm deep and 12 mm in diameter to give a maximum working volume of 800 pi. Using this holder, the same set-up can be reused multiple times and many devices can be tested without the need to individually solder to each connection. Furthermore, devices can easily be removed after testing to perform other assays or microscopy. The well segment, base and lid were machined and laser cut from acrylic. The Kapton sheet forming the substrate for the connector circuit was also laser cut. The toolpath to print the circuit was then aligned and printed onto this piece using fiducial markers. A PSoC microcontroller was used to monitor the capacitance values for each sensor via an embedded CapSense® component. In this component, capacitance is determined using a‘switched capacitors’ method, as illustrated in Fig. 28(a). A pair of non-overlapping switches (Si and S2) periodically connect the sensor either to a charging current from a modulating capacitor (Cmod), or to ground. On average, the sensor capacitor (Cs) behaves as an equivalent resistor, sinking a current proportional to its capacitance. If Cs increases, for example, the sensor will draw a larger current from Cmod and cause the modulating capacitor to discharge at a greater rate. The voltage across Cmod is monitored by a Sigma- Delta analogue-to-digital converter (SD ADC). Based on this voltage level, the SD ADC controls switch 3 (S3) to recharge Cmod from a constant current source (Lod) such that the average voltage across Cmod is maintained at a reference voltage (V_ref). The raw count output of the SD ADC is a bit stream that represents the duty cycle of S3. This is related to the capacitance of the sensor via the equation:

(1 ) RaW COUnt— (2^^®— 1 ) Vre_fF swCs/lmod

The total capacitance of the sensor, as measured by the microcontroller, can be written as:

(2) Cs = C_P +Cc

C_P is the parasitic capacitance and Co is the capacitance due any cells in the vicinity of the electrode. The parasitic capacitance is a generic term that includes all sources of additional capacitance that do not contribute to the useful signal - i.e. the baseline capacitance. This arises from many sources including the capacitance of the microcontroller itself and any coupling of the sensor components to other conductors in the system. Maximum sensor sensitivity is achieved when C_P is minimized, so that Cc is a larger proportion of the total measured capacitance. The shield electrode and hatched fill described above are efforts to reduce the value of C_P.

When mounted in the device holder, the sensors have the parasitic capacitance between 2-5 pF. The variation is caused by subtle differences in each device from the fabrication processes, differences in the length of each connecting trace and the differences in capacitance of each microcontroller input pin. This variation means that absolute values of sensor capacitance cannot reliably be compared. Instead, it is the change in capacitance (DO) that is of interest.

Fig. 28(b) demonstrates the change in capacitance observed when 800 pi of PBS solution is added to a device under ambient conditions. A slight increase in DO can be seen during the remainder of the test. This is attributed to the wetting behaviour of the nano-textured surface and to the polymer itself absorbing water. This drift is small and does not influence the device function. The PBS solution was removed at approximately 13.5 hours into the test, causing a fluctuation in the capacitance values at that time. Even though the majority of the liquid has been removed, the capacitance value is relatively unchanged. This is due to a thin layer of PBS solution trapped within the nanotextured surface. Approximately one hour after the solution was removed, the remaining liquid film evaporates and the capacitance value of the sensor returns to the initial value. Fig. 29 shows the capacitance changes observed from three devices over a 48-hour time period. One device was left completely dry to act as a control against any fluctuations in the measurement circuitry (part (iii) in Fig. 29). As can be seen from the data, there is very little change in the capacitance value throughout the test, indicating that placing the microcontroller board inside the incubator does not compromise the stability of the measurement. Minor perturbations can be seen at the same time as the other devices under test are manipulated. This is due to the inevitable coupling between sensors and connecting traces.

Another device contained only plain growth media was also tested - part (ii) in Fig. 29. This device had been pre-wetted before the test, so the initial increase in capacitance of 150 fF over the first « 4 hours is difficult to fully explain, although the inventors theorise that it is possible that the wetting phase was not complete at the beginning of the experiment, and thus the surface continued to wet as the fresh media was added. After this initial period, the capacitance slowly decreases by 25 fF until T = 24 hours. At this point the media was changed, causing a shift in the capacitance values and a slow increase in AC over the remainder of the test.

The final device contained the suspension of Human Dermal Fibroblasts (HDFs) (part (i) in Fig. 29). The capacitance reading from this device increases from the point at which cells were added and continues to rise for the next 30 hours before reaching a maximum AC = 300 fF. At approximately 45 hours into the test, a trypsin/EDTA solution (TE) was added to both devices containing media. This solution acts to break the focal adhesions of adherent cells and release them from a substrate. Around 30 minutes after adding this solution, a large decrease of AC = 100 fF is observed only in the device which contains cells. This demonstrates that the sensed capacitance is, at least in part, due to the presence of adhered HDFs.

It is interesting to note that the capacitance does not return to the initial value after the addition of TE.

This behaviour has also been observed by others using conventional CMOS sensors and is sometimes attributed to residual protein layers on the device surface and/or to drifts in the sensor baseline.

Nonetheless, the sensor is capable of resolving the presence of Human Dermal Fibroblasts and confirms that capacitive cell sensing can be combined with nanotextured and polymeric surfaces.

The inventors theorise that device performance could be improved by use of a bespoke Printed Circuit Board (PCB) could be created to replace the FreeSoC2 development board used here. This board could contain both the contact pads needed to make connection to the device as well as the PSOC

microcontroller, eliminating the need for lengthy connecting traces. Furthermore, the specifications of the PCB could be selected to optimize the performance for this specific application. Using a dedicated PCB such as this would reduce the parasitic capacitance of the device, making the capacitance change due to the presence of cells a larger proportion of the total sensed capacitance. This would enable the area of each sensor to be reduced, allowing multiple sensors to be included within each device to provide a degree of spatial resolution. Complex electrode patterns are straightforward to print and transfer, the only limitation being the resolution of the Aerosol Jet printer. The reliability and stability of the system can also be improved with a dedicated electronic circuit. Furthermore, refinements to the device holder may also facilitate greater consistency. Scannina Electron Characterisation

Unless otherwise stated, all scanning electron microscopy (SEM) images were captured using a Hitachi TM3030 desktop microscope. The vast majority of samples were nonconducting and thus were sputter coated with palladium before imaging to prevent charging. Samples were mounted on aluminium SEM stubs with adhesive carbon tape.

Example calculation of effective stiffness for cylindrical nanomembers

The force F required to give a displacement x of a bulk material is given by

G A x

F =

l

where G is the shear modulus of the material, A is the surface area over which the force acts and I is the thickness of the material. Defining the effective stiffness of the bulk material (pbui_k) as the ratio of force to displacement gives:

Where Y and v are the Young’s modulus and Poisson ratio of the material respectively. A similar treatment can be carried out with the surface of nanostructures. Assuming deformation by beam bending:

3ixYr⁴

21³

Where r is the diameter of the nanostructures, for nanostructures assumed to be cylindrical in shape. From this expression, the nanostructured stiffness follows as:

For cylindrical nanomembers, the area is A = nr² , and the aspect ratio is defined by f =

Substituting these relationships in previous equations:

The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof. While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

Throughout this specification, including the claims which follow, unless the context requires otherwise, the word“comprise” and“include”, and variations such as“comprises”,“comprising”, and“including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

It must be noted that, as used in the specification and the appended claims, the singular forms“a,”“an,” and“the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from“about” one particular value, and/or to“about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and/or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent“about,” it will be understood that the particular value forms another embodiment. The term“about” in relation to a numerical value is optional and means for example +/- 10%.

References

A number of publications are cited above in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Full citations for these references are provided below. The entirety of each of these references is incorporated herein.

[1] T. Iskratsch, H. Wolfenson, M.P. Sheetz, Appreciating force and shape— the rise of

mechanotransduction in cell biology., Nat. Rev. Mol. Cell Biol. 15 (2014) 825-33.

doi: 10.1038/nrm3903.

[2] U.S. Schwarz, Mechanobiology by the numbers: A close relationship between biology and

physics, Nat. Rev. Mol. Cell Biol. 18 (2017) 71 1-712. doi: 10.1038/nrm.2017.109.

[3] J. Solon, I. Levental, K. Sengupta, P.C. Georges, P.A. Janmey, Fibroblast adaptation and stiffness matching to soft elastic substrates, Biophys. J. 93 (2007) 4453-4461.

doi: 10.1529/biophysj.106.101386.

[4] P.A. Janmey, J.P. Winer, M.E. Murray, Q. Wen, The hard life of soft cells, Cell Motil. Cytoskeleton. 66 (2009) 597-605. doi:10.1002/cm.20382.

[5] J. Fu, Y.-K. Wang, M.T. Yang, R.A. Desai, X. Yu, Z. Liu, C.S. Chen, Mechanical regulation of cell function with geometrically modulated elastomeric substrates, Nat. Methods. 7 (2010) 733-736. doi: 10.1038/nmeth.1487.

[6] A. Buxboim, D.E. Discher, Stem cells feel the difference, Nat. Methods. 7 (2010) 695-697.

doi: 10.1038/nmeth0910-695.

[7] Y. Ito, T. Kimura, K. Nam, A. Katoh, T. Masuzawa, A. Kishida, Effects of vibration on differentiation of cultured PC12 cells., Biotechnol. Bioeng. 108 (201 1 ) 592—9. doi: 10.1002/bit.22961.

[8] P.G. Childs, C.A. Boyle, G.D. Pemberton, H. Nikukar, A.S.G. Curtis, F.L. Henriquez, M.J. Dalby,

S. Reid, Use of nanoscale mechanical stimulation for control and manipulation of cell behaviour, Acta Biomater. 34 (2016) 159-168. doi: 10.1016/j.actbio.2015.1 1.045.

[9] S.N. Robertson, P. Campsie, P.G. Childs, F. Madsen, H. Donnelly, F.L. Henriquez, W.G. Mackay, M. Salmeron-Sanchez, M.P. Tsimbouri, C. Williams, M.J. Dalby, S. Reid, Control of cell behaviour through nanovibrational stimulation: nanokicking, Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 376 (2018) 20170290. doi:10.1098/rsta.2017.0290.

[10] K.A. Davis, K.A. Burke, P.T. Mather, J.H. Henderson, Dynamic cell behavior on shape memory polymer substrates, Biomaterials. 32 (201 1 ) 2285-2293. doi: 10.1016/j.biomaterials.2010.12.006.

[1 1] D.M. Le, K. Kulangara, A.F. Adler, K.W. Leong, V.S. Ashby, Dynamic topographical control of mesenchymal stem cells by culture on responsive poly(lp-caprolactone) surfaces, Adv. Mater. 23 (201 1 ) 3278-3283. doi: 10.1002/adma.201 100821.

[12] J.D. Kiang, J.H. Wen, J.C. Del Alamo, A.J. Engler, Dynamic and reversible surface topography influences cell morphology, J. Biomed. Mater. Res. - Part A. 101 A (2013) 2313-2321.

doi: 10.1002/jbm.a.34543.

[13] N.J. Sniadecki, A. Anguelouch, M.T. Yang, C.M. Lamb, Z. Liu, S.B. Kirschner, Y. Liu, D.H. Reich, C.S. Chen, Magnetic microposts as an approach to apply forces to living cells., Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 14553-8. doi: 10.1073/pnas.061 1613104.

[14] Z. Liu, Y. Liu, Y. Chang, H.R. Seyf, A. Henry, A.L. Mattheyses, K. Yehl, Y. Zhang, Z. Huang, K.

Salaita, Nanoscale optomechanical actuators for controlling mechanotransduction in living cells, Nat. Methods. 13 (2016) 143-146. doi:10.1038/nmeth.3689.

[15] A. Sutton, T. Shirman, J.V.I. Timonen, G.T. England, P. Kim, M. Kolle, T. Ferrante, L.D. Zarzar, E.

Strong, J. Aizenberg, Photothermally triggered actuation of hybrid materials as a new platform for in vitro cell manipulation, Nat. Commun. 8 (2017) 14700. doi:10.1038/ncomms14700.

[16] M. Hoop, X.-Z. Chen, A. Ferrari, F. Mushtaq, G. Ghazaryan, T. Tervoort, D. Poulikakos, B. Nelson, S. Pane, Ultrasound-mediated piezoelectric differentiation of neuron-like PC12 cells on PVDF membranes, Sci. Rep. 7 (2017) 4028. doi: 10.1038/s41598-017-03992-3. [17] G. Ciofani, S. Danti, D. D’Alessandro, L. Ricotti, S. Moscato, G. Bertoni, A. Falqui, S. Berrettini, M. Petrini, V. Mattoli, A. Menciassi, Enhancement of Neurite Outgrowth in Neuronal-Like Cells following Boron Nitride Nanotube-Mediated Stimulation, ACS Nano. 4 (2010) 6267-6277.

doi:10.1021/nn101985a.

[18] M. Jamshidian, E.A. Tehrany, M. Imran, M. Jacquot, S. Desobry, Poly-Lactic Acid: Production, Applications, Nanocomposites, and Release Studies, Compr. Rev. Food Sci. Food Saf. 9 (2010) 552-571. doi: 10.1 1 1 1/j.1541 -4337.2010.00126.x.

[19] M. Zhang, P. Dobriyal, J.-T. Chen, T.P. Russell, J. Olmo, A. Merry, Wetting Transition in

Cylindrical Alumina Nanopores with Polymer Melts, Nano Lett. 6 (2006) 1075-1079.

doi:10.1021/nl060407n.

[20] J. Chen, C. Lee, M. Chi, I. Yao, Solvent-Annealing-Induced Nanowetting in Templates: Towards Tailored Polymer Nanostructures, Macromol. Rapid Commun. 34 (2013) 348-354.

doi: 10.1002/marc.201200640.

[21] M. Steinhart, J.H. Wendorff, R.B. Wehrspohn, Nanotubes a la Carte: Wetting of Porous

Templates, ChemPhysChem. 4 (2003) 1 171-1 176. doi: 10.1002/cphc.200300733.

[22] J. Martin, C. Mijangos, Tailored Polymer-Based Nanofibers and Nanotubes by Means of Different Infiltration Methods into Alumina Nanopores, Langmuir. 25 (2009) 1 181-1 187.

doi: 10.1021/la803127w.

[23] J. Martin, J. Maiz, J. Sacristan, C. Mijangos, Tailored polymer-based nanorods and nanotubes by “template synthesis”: From preparation to applications, Polymer (Guildf). 53 (2012) 1 149-1 166. doi: 10.1016/j.polymer.2012.01.028.

[24] C. Mijangos, R. Hernandez, J. Martin, A review on the progress of polymer nanostructures with modulated morphologies and properties, using nanoporous AAO templates, Prog. Polym. Sci. 54- 55 (2016) 148-182. doi:10.1016/j.progpolymsci.2015.10.003.

[25] M. Steinhart, R.B. Wehrspohn, U. Gosele, J.H. Wendorff, Nanotubes by Template Wetting: A

Modular Assembly System, Angew. Chemie Int. Ed. 43 (2004) 1334-1344.

doi: 10.1002/anie.200300614.

[26] M. Yasuniwa, K. lura, Y. Dan, Melting behavior of poly(l-lactic acid): Effects of crystallization

temperature and time, Polymer (Guildf). 48 (2007) 5398-5407.

doi: 10.1016/j.polymer.2007.07.012.

[27] J. Zhang, K. Tashiro, H. Tsuji, A.J. Domb, Disorder-to-Order Phase Transition and Multiple Melting Behavior of Poly(l -lactide) Investigated by Simultaneous Measurements of WAXD and DSC, Macromolecules. 41 (2008) 1352-1357. doi:10.1021/ma0706071.

[28] P. Pan, B. Zhu, W. Kai, T. Dong, Y. Inoue, Polymorphic Transition in Disordered Poly( l-lactide) Crystals Induced by Annealing at Elevated Temperatures, Macromolecules. 41 (2008) 4296-4304. doi: 10.1021/ma800343g. [29] A. Petersen, P. Joly, C. Bergmann, G. Korus, G.N. Duda, The Impact of Substrate Stiffness and Mechanical Loading on Fibroblast-Induced Scaffold Remodeling, Tissue Eng. Part A. 18 (2012) 1804-1817. doi:10.1089/ten.tea.201 1.0514.

[30] R.B. Karabalin, L.G. Villanueva, M.H. Matheny, J.E. Sader, M.L. Roukes, Stress-Induced

Variations in the Stiffness of Micro- and Nanocantilever Beams, Phys. Rev. Lett. 108 (2012)

236101. doi:10.1 103/PhysRevLett.108.236101 .

[31] G. Thrivikraman, S.K. Boda, B. Basu, Unraveling the mechanistic effects of electric field

stimulation towards directing stem cell fate and function: A tissue engineering perspective, Biomaterials. 150 (2018) 60-86. doi: 10.1016/j. biomaterials.2017.10.003.

Claims

Claims:

1. A biological material interaction platform comprising:

2. The biological material interaction platform according to claim 1 wherein the nanomembers comprise nanorods, nanowires and/or nanotubes.

3. The biological material interaction platform according to claim 2 wherein the nanorods, nanowires and/or nanotubes have an aspect ratio of 5:1 or more, 10:1 or more, 50:1 or more, 100:1 or more or 500:1 or more.

4. The biological material interaction platform according to any one of the preceding claims wherein the nanomembers have at least one dimension less than 500nm.

5. The biological material interaction platform according to any one of the preceding claims wherein the nanomembers comprise an array of substantially aligned nanotubes upstanding from the substrate.

6. The biological material interaction platform according to any one of the preceding claims wherein the shear piezoelectric material nanomembers have a crystallinity of 20% or more.

7. The biological material interaction platform according to any one of the preceding claims wherein shear piezoelectric material nanomembers comprise a biocompatible polymer.

8. The biological material interaction platform according to claim 7 wherein the shear piezoelectric material nanomembers comprise poly-L-lactic acid (PLLA), optionally having (001) polymer chain orientation, optionally having a molecular weight Mw between 80,000 and 150,000.

9. The biological material interaction platform according to any one of the preceding claims comprising an in plane electrode array in electrical communication with at least a part of the array of shear piezoelectric material nanomembers and configured for application of and/or detection of an electric field across at least part of the array of shear piezoelectric material nanomembers.

10. The biological material interaction platform according to claim 9 wherein the electrode array is an interdigitated electrode array.

11. The biological material interaction platform according to claim 9 or claim 10 where the spacing between adjacent electrodes in the electrode array is from 10pm to 500pm.

12. The biological material interaction platform according to any one of claims 9 to 1 1 wherein each electrode in the electrode array is individually addressable.

13. The biological material interaction platform according to any one of the preceding claims wherein the biological subject is a cell.

14. Use of a biological material interaction platform according to any one of claims 1 to 1 1 for providing local stimulation to one or more biological subjects supported by the shear piezoelectric material nanomembers.

15. Use of a biological material interaction platform according to any one of claim 1 to 1 1 for sensing micro-mechanical forces produced by one or more biological subjects supported by the shear piezoelectric material nanomembers.

16. A method of making a biological material interaction platform, the method comprising steps of: hot pressing a piezoelectric material and a nanostructured template at a temperature above the melting point of the piezoelectric material such that the piezoelectric material at least partly infiltrates the nanostructured template;

and removing the template to leave an array of shear piezoelectric material nanomembers.

17. A method according to claim 16 wherein the step of hot pressing the piezoelectric material and nanostructured template is performed for a time of 5 minutes or less.

18. A method according to claim 16 or 17 further including a step of annealing the piezoelectric material at a temperature below the melting point of the piezoelectric material to induce crystallisation of the piezoelectric material.

19. A method according to any one of claims 16 to 18 further comprising a step of pre-heating the piezoelectric material and/or the nanostructured template before hot pressing.

20. A method according to any one of claims 16 to 19 wherein the hot pressing is performed at a rate of between 1 mm/min and 20 mm/s.

21. A method according to any one of claims 16 to 20 wherein the step of removing the template is performed by etching.

22. A method according to any one of claims 16 to 21 , the method further comprising steps of: forming an electrode array on a donor substrate;

removing the donor substrate.

23. A method according to claim 22 wherein the electrode array is formed using an aerosol jet printing process.

24. A method according to claim 23 including a step of curing the printed electrode array prior to hot pressing.

25. A method according to any one of claims 16-24 further including the step of supporting at least one biological subject on the array of shear piezoelectric material nanomembers.

26. A cell capacitance sensor comprising a biological material interaction platform according to any one of claims 1 to 15.