Patent Application
20200235245
The work leading to this invention has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreement number 319277.
The present invention relates to the field of textiles and has application, for example, in wearable electronics, smart fabrics and e-textiles. In particular, but not exclusively, it relates to the field of depositing layered materials onto textiles. One example of a suitable layered material is graphene.
Wearable electronics, smart fabrics and e-textiles have the potential to reshape the electronics markets over a wide range of sectors, spanning from biomedical through to fashion-tech. Fabric-integrated components and devices, and innovative textiles which can conduct electricity, and/or guide and/or emit light, and/or regulate temperature are at the centre of a new technical advance in the smart textile industry.
Currently, most wearable electronics are based onto two main technologies. The first technology is that of standard rigid electronic components (e.g. light emitting diodes, transistors, microchips, batteries, etc.) embedded, attached, or simply interconnected to fabrics or textiles using classic metallic (silver, gold, copper, nickel) wire bonding techniques or flexible textile-coated metallic wires. The second technology is that of flexible electronic components integrated into textiles and fabrics by interwoven flexible metal/polymer conductive wires.
Metallic wires and metal-polymer composites are expensive, heavy and require an accurate weaving process to be incorporated into the fabric. Moreover metals tend to oxidize and have a strong dependence of conductivity to temperature and humidity rate, which may affect the reliability and robustness of the wearable circuits and devices. Fibres containing a high density of metal may also limit the mechanical performance of fibres by a reduction in flexibility and a reduction in elongation at break. Use of both metals and organic polymers may affect the bio-compatibility of the fibres, for example, metals such as nickel have shown poor bio-compatibility and allergenic effects that make them not desirable for use in wearable fabrics.
Printing has evolved from a tool for text and graphics [DeGans et al. 2004], to a milestone for rapid manufacturing of plastic electronics [Van Osch et al. 2008] and is now an established technique to print electrodes and interconnections, based on metal nanoparticles [Singh et al. 2010], and electronic devices such as thin film transistors (TFTs) based on organic conducting and semiconducting inks [Sekitani et al. 2009, Sirringhaus et al. 2000]. However, the operation speed of organic TFTs, which is defined in terms of electron mobility, is still much lower than state of the art silicon technology. Several approaches have attempted to coat/print on textiles and fabrics with metallic inks [Jost et al. 2011] and/or organic conducting/semiconducting inks. Printed components based on metal nanoparticle and organic inks are expensive, tend to oxidize and generally require post-printing treatments. Several approaches aiming to improve these results encompass carbon-based inks typically containing particles of amorphous carbon, carbon black or graphite that are suspended in a solvent via binder or surfactant. However the presence of binders or surfactants can affect the final conductivity, requiring further post-printing treatment.
Graphene and related materials (GRMs) are a group of two-dimensional layered materials, including, but not limited to, metals (e.g., NiTe2, VSe2), semi-metals (e.g., WTa2, TcS2), semiconductors (e.g., WS2, WSe2, MoS2, MoTe2, TaS2, RhTe2, PdTe2, black phosphorus), insulators (e.g. h-BN), superconductors (e.g., NbS2, NbSe2, NbTe2, TaSe2), topological insulators and thermo-electrics (e.g., Bi2Se3, Bi2Te3), transition metal dichalcogenides (TMDs) and transition metal oxides (TMOs). GRMs can have unique mechanical, electrical and optical properties. In many cases they also have exceptional environmental stability (low moisture absorption) and potential for low-cost production enabling fully flexible printed flexible electronics and photonics. This places GRMs as prime candidates to play a major role in the wearable electronics and smart textiles sectors, where classical cotton, silk, and other natural or synthetic fabrics can be transformed into advanced active textiles exhibiting electrical, optical and/or smart thermal functions.
WO2014/064432 discloses the manufacture of inks comprising a carrier liquid with a dispersion of flakes derived from a layered material. However, the present inventors have realised, as part of their contribution to the art forming part of this disclosure, that the production of electronic components on a flexible fabric substrate using such ink presents a number of problems. These include including poor adhesion of the ink to the substrate, poor connectivity across layers due to substrate roughness, unwanted absorption of carrier fluid by fabric substrate leading to poor quality deposited layers, and poor durability and washability of the deposited inks.
As discussed above, the inventors have found that simple deposition of graphene or, more generally, GRM (graphene and related material) ink onto a fabric substrate can result in a rather poor quality deposited layer. The inventors consider that this is partly due to a lack of affinity between the fabric and the 2D material. Additionally, there may also be a lack of long-range connectivity between nanoplatelets in the deposited layer caused by the high surface roughness (>50 μm) of typical fabrics used for clothing. This roughness is caused by the weave of the fabric, and/or by the inherent roughness of the fibres and/or yarns of the fabric. The lack of chemical affinity between the fabric and the 2D material can also result in a somewhat random nanoplatelet arrangement within the deposited layer, which may be undesirable.
Furthermore, whilst it is acknowledged that printing of graphene-based inks onto non-fabric substrates is known, binders and surfactants are generally used in such inks. However, the presence of such components, and particularly the presence of substantial quantities of such components in the inks may affect the final properties of a deposited graphene or, more generally, GRM layer. For example, such components may affect the optical, mechanical and electrical properties of the layer, and in some cases may necessitate post-printing treatment, which is disadvantageous. Therefore, it is preferable that no such additives are used, although some small amount may be acceptable.
It is known to treat fibres and/or fabrics to improve dyeability. WO 2014/116230 for example discloses a method of treating a cellulose fibre including the steps of contacting the fibre with a solution, the solution comprising about 0.5 to about 15 g/L of a wetting agent, about 5 to about 150 g/L of an alkaline composition, and about 5 to about 200 g/L of an ammonium salt, wherein a permanent positive charged (cationized) site on the cellulose molecule which can attract an anionic (negatively charged) compound such as an anionic dyestuff. However, this prior art is concerned with dyebility, specifically the ability to achieve a desired colour, along with preventing problems such as colour bleeding and fading, rather than aiming to provide improved deposited functional layers, which is one preferred object of the present invention.
Accordingly, the present inventors consider that there is an unmet need in the prior art, to develop improved functional layers for flexible functional components, such as flexible electrical components. Flexibility of the components here includes flexibility in response to compression, tension, buckling and/or torsion.
The present invention has been devised in order to address at least one of the above problems. Preferably, the present invention reduces, ameliorates, avoids or overcomes at least one of the above problems.
Accordingly, in a first preferred aspect of the invention, there is provided a flexible electronic component comprising a flexible fabric substrate, a smoothing layer formed on the flexible fabric substrate and a coating comprising a deposited layer of nanoplatelets derived from a layered material formed on the smoothing layer.
In a second preferred aspect, the present invention provides a method for producing a flexible electronic component, the method including the steps;
In a third aspect of the invention, there is provided a flexible electronic component or device, obtained or obtainable by a method according to the second aspect.
Any of the first, second and third aspects of the invention may be combined with each other. The first, second and/or third aspects of the invention may have any one or, to the extent that they are compatible, any combination of the following further optional features.
In some embodiments, the method includes the step of treating at least a part of the flexible fabric substrate to provide a treated portion wherein the treated portion is cationized or anionized. In this case, the treated portion at least partly corresponds to the location of the flexible electronic component.
Thus, in some embodiments, at least part of an interface of the fabric on which the coating is deposited is a cationized or anionized treated portion.
Cationization treatment of the fabric provides a positive charge at the surface of the fabric. Anionization treatment of the fabric provides a negative charge at the surface of the fabric. It has been found that this may enhance uniformity of a nanoplatelet-based ink coating applied to a fabric or textile substrate. Without wishing to be bound by theory, the inventors suggest that this is due to electrostatic interactions between the treated fabric and the nanoplatelets in the ink. For example, the treated portion of the textile surface may attract nanoplatelets which have an opposite charge through electrostatic interactions. For the case of non-functionalised GRMs, these may have varying affinity with the cationized or anionized fabrics depending on their terminal groups. Providing a more uniform ink coating on a flexible fabric substrate has advantages in that it may improve the functionality of the ink coatings by, for example, improving the connectivity of the deposited nanoplatelets and therefore decreasing the sheet resistance of the applied ink coating.
In some embodiments, the nanoplatelets are functionalized. There may be included a step of functionalizing the nanoplatelets, for example, before the ink comprising the dispersion of nanoplatelets is applied to the fabric substrate.
Functionalization of the nanoplatelets may include functionalization to exhibit positive or negative charges on the surfaces of the nanoplatelets. Alternatively it may include functionalization by adding functional groups to the nanoplatelets. Functionalisation of the nanoplatelets can impose a termination with a desired charge polarity. An appropriate functionalization process may be selected according to the selected nanoplatelets materials in the ink, and the intended final properties of the electronic component. Such functionalization of the nanoplatelets may increase electrostatic interaction between the nanoplatelets and the treated portion of the flexible fabric substrate. This can lead to improved uniformity of an ink coating on a flexible fabric substrate, and accordingly improve desirable of the deposited coating. For example, when the nanoplatelets are formed of graphene, these may be functionalised by chemical oxidation and reduction using a modified Hummers method. Alternatively, intercalation of graphite can also lead to formation of functional groups. Treatment with metal salts such as gold chloride, iron chloride or other suitable reagents can generate modification of the in-plane or edge termination of graphene. Another method of achieving chemically functionalised graphene is graphene flake growth by methane cracking in a high temperature furnace.
Preferably, the step of treating the at least a part of a surface of the flexible fabric substrate includes a step of contacting the at least a part of a surface of the flexible fabric substrate with a solution comprising 3-chloro-2-hydroxypropyltrimethylammonium chloride, bis-quaternary ammonium salt, or polymerizable bis-quaternary ammonium salt. More generally, the step of treating the at least a part of the flexible fabric substrate may include a step of contacting the at least a part of flexible fabric substrate with a solution comprising one or more quaternary ammonium salt.
The nanoplatelets may be derived from any suitable layered material. Of particular interest are graphene and related materials (GRMs). This is due to the useful properties of such materials as discussed above, in particular their mechanical, electrical, optical and thermal properties. Graphene and related materials are a group of two-dimensional layered materials, including, but not limited to, metals (e.g., NiTe2, VSe2), semi-metals (e.g., WTa2, TcS2), semiconductors (e.g., WS2, WSe2, MoS2, MoTe2, TaS2, RhTe2, PdTe2, black phosphorus), insulators (e.g. h-BN), superconductors (e.g., NbS2, NbSe2, NbTe2, TaSe2) and topological insulators, thermo-electrics (e.g., Bi2Se3, Bi2Te3), transition metal dichalcogenides (TMDs), transition metal oxides (TMOs), and other two-dimensional materials such as graphite.
In some embodiments, there may be provided sub-layers of different nanoplatelet materials. Additionally or alternatively two or more different nanoplatelet materials may be combined with each other in one layer. For example, the device may include a first sub-layer of a first nanoplatelet material and a second sub-layer of a second nanoplatelet material, different in composition from the first nanoplatelet material. The second sub-layer may be deposited at least in part on the first sub-layer. A third sub-layer of a third nanoplatelet material may be provided, different from at least one of the first and second nanoplatelet materials. The third sub-layer may be deposited at least in part on the second and/or first sub-layer. In this way, electronic devices may be formed, having functionality determined at least in part by the interaction of the first, second and/or third layers at their respective interfaces. Additionally or alternatively, multiple sub-layers of the same nanoplatelet material may be deposited. This can help to ensure that a sufficient thickness is deposited.
In some embodiments, the electronic component is a thermoelectric device.
Preferably, liquid phase exfoliation (LPE) is used as a production method for producing the nanoplatelets used in the ink in this invention. This is because LPE is able to provide nanoplatelets in a convenient form (for example, dispersions, inks or pastes). LPE is also compatible with large scale production (e.g. is capable of producing quantities of nanoplatelets greater than 1 kg). LPE is capable of giving high yields of single layer flakes (up to 80%). It is also a relatively low cost manufacturing process. Whilst LPE is the preferred manufacturing method, any other manufacturing method which provides GRMs of sufficient quality for use in the invention may be used. GRMs produced by LPE, including WS2, MoO3 and BN have diverse properties, e.g. metallic, semiconducting, insulating, electrochemically active, etc. making them suitable functional agents for inks suitable for a wide range of textiles.
Suitable inks of GRMs may be obtained by processes outlined in WO2014/064432, the entire contents of which are here incorporated by reference. The preferred composition of the ink will vary depending on the desired device properties, and different inks containing different GRMs may be used in the manufacture of a single flexible electronic device, for example, for production of a multilayer structure. GRM inks may also be mixed, to enable precise selection of ink properties.
Suitable nanoplatelets derived from a layered material may be considered to be those with lateral size of 1, 2, 3, or 3 or more microns and thicknesses below 100 nm.
Different ink deposition methods (some examples of which include inkjet, flexographic, gravure, spray coating, rod coating, roll to roll coating, slot-dye coating, spin coating, transfer printing, dip coating and screen printing) may be used to produce flexible electronic components or devices of the invention. Different deposition methods may offer different final properties and/or structures of the flexible components or devices. Preferably therefore, the deposition method is selected according to the desired properties of the electronic component or device to be produced.
Different nanoplatelet dispersions for different inks may be selected depending on the deposition process to be used. For example, the viscosity of the ink, the mass % of nanoplatelets in the dispersion, or any other suitable parameters of the ink may be varied, depending on the intended deposition process to be used, and the intended properties of the device to be produced. Typical suitable viscosity ranges for some example printing or coating processes are as follows: Inkjet printing 1-20 mPa s, flexo and gravure printing 150-200 mPa s, spray coating 1-200 mPa s, screen printing 1000 mPa s. For all deposition processes, it is generally preferable to use an ink with higher mass % of nanoplatelets in the dispersion, as this may provide for higher quality deposited layers.
In some preferred embodiments, the ink includes no binder and/or surfactant. However, in some cases, small quantities of one or more such additives may be advantageous, without compromising the performance of the deposited layer. For example, the ink may include 0.1 g/L of such additives. Preferably, the ink includes not more than 10 g/L of such additives.
As defined above, to create a more uniform surface for subsequent printing processes, the fabric is treated with an intermediate smoothing layer. The smoothing layer may also be referred to as a planarization layer. This layer may be applied to the fabric for example by bar coating or screen printing. The smoothing layer may reduce a relatively high surface roughness of a fabric by filling-in the weave of the fabric and offering a surface of reduced roughness to which a GRM ink may then be applied. A particularly appropriate material for this smoothing layer is polyurethane, however any other suitable material may be used, for example silane coupling agent or soft adhesion agent.
The root mean squared roughness Rq of this smoothing layer is preferably <300 μm, more preferably <100 μm, more preferably <50 μm, even more preferably <10 μm and most preferably <5 μm.
The fabric substrate, optionally including the intermediate smoothing layer, may then be chemically modified as described above, to provide a higher surface energy, increasing the interaction strength between the substrate and the GRM ink flakes, promoting formation of a more uniform deposited layer.
Several flexible interlayers may be used to protect the GRM flexible electronic components, providing improved durability and washability. For example, a flexible polymer interlayer may be bar coated on top of the deposited ink layer to protect the GRM flexible electronic component and assist in preserving the electrical, optical and mechanical properties. This process may be advantageous when it is intended to produce wearable, environmentally stable and durable smart textiles. Such flexible interlayers may be applied by any method as discussed previously. Suitable materials include, for example, polyurethane, or any other material which provide a suitable degree of protection for the GRM printed structures, including, for example, silane coupling agents. One or more additional layers of suitable 2D materials, as listed above, can be used for such a protective function. For example, an h-BN layer is suitable, providing protection to the layers below from oxygen and/or water vapour.
The type of fabric used as a substrate for such flexible electronic device is not particularly limited, however it may be preferable to use cotton, or cotton blended yarns, which has reactive groups, relatively complicated surface morphology, good flexibility, and relatively high porosity, in addition to being a commonly used textile in clothing. The wide use of cotton fibres in diversified outdoor and indoor-applications along with its traditional textile products can be mainly attributed to its economical, eco-friendly, biodegradable and hydrophilic nature (—OH). With an understanding of the structure of cotton, there can be provided control over its modification. The chemical stability of the cotton molecule is considered to be determined by the sensitivity to hydrolytic attack of the β-1,4-glycosidic linkages between the glucose repeating units.
Printed GRM inks on textiles can be used to fabricate flexible, conductive and wearable electronic components and devices in many different forms, some examples being circuits, interconnections, sensors (including, for example, movement, pressure or temperature sensors), capacitors, transistors, displays, antennas, batteries, photodetectors etc.
The invention therefore has a wide range of industrial applications, including fashion dress, military garment devices, high-performance sportswear and personal health monitors, wearable computers, energy harvesting/storage devices directly incorporated into clothes, and many more areas besides.
Preferably the deposited layer of nanoplatelets can be considered to form a first layer of a first nanoplatelet material. There is preferably provided at least a second layer, of a different nanoplatelet material, formed at least in part on the first layer.
There may be additionally provided at least first and second electrodes, in contact respectively with the first and second layers.
In some preferred embodiments, the flexible electronic component may be in the form of a transistor. For example, the flexible electronic component may be in the form of a field effect transistor.
In some embodiments, the first layer may be formed of graphene. As mentioned above, the second layer may be formed of a different material. One suitable different material is h-BN. The first layer may be provided with source and drain electrodes. The second layer may be provided with a gate electrode. Thus, preferably, the source, drain and gate electrodes are separated from the interface between the first layer and the second layer.
Alternatively, in some embodiments, the first layer may be formed of h-BN. The second layer may be formed of graphene. The first layer may be provided with a gate electrode. The second layer may be provided with source and drain electrodes. Thus, preferably, the source, drain and gate electrodes are separated from the interface between the first layer and the second layer.
The flexible electronic component may have a charge carrier mobility of at least 50 cm2/Vs. More preferably, the charge carrier mobility is at least 60 cm2/Vs. Still more preferably, the charge carrier mobility is at least 70 cm2N/s.
As will be understood, if required, many different layers can be deposited in a required structural arrangement in order to form a required electronic device.
With respect to the fabric on which the flexible electronic component is formed, preferably the fabric, before application of the smoothing layer, has a roughness Rq of 35 μm or less. More preferably, Rq is 30 μm or less.
A suitable fabric for use with embodiments of the invention has been found to be polyester satin.
In some embodiments, the smoothing layer is formed from polyurethane. However, different approaches can be taken to the formation of the smoothing layer. For example, the smoothing layer may comprise a first sub-layer of polyurethane and a second sub-layer of h-BN. The layer of h-BN may provide additional functionality, being for example a functional layer of the flexible electronic component.
Preferably, the thickness of the smoothing layer is at least 5 μm. In some embodiments, the thickness of the smoothing layer may be greater, e.g. at least 10 μm. The smoothing layer should preferably not be so thick as to significantly affect the performance of the underlying fabric. For example, the smoothing layer is preferably not more than 100 μm thick, still more preferably not more than 80 μm thick, or not more than 60 μm thick, or not more than 40 μm thick.
The flexible electronic component may further comprise a washable protective layer formed over the device. The washable protective layer may for example be a flexible polymer layer. It is found in some embodiments that the combination of flexibility and the washable protective layer has the effect that the flexible electronic component can survive multiple washing cycles (e.g. typical domestic washing cycles) without significant degradation of the performance of the flexible electronic component.
Preferably, the intermediate smoothing layer applied to the fabric substrate has a surface roughness Rq of less than 10 μm. More preferably, Rq is less than 8 μm. Still more preferably, Rq is less than 6 μm. Still yet more preferably, Rq is less than 5 μm.
In some embodiments, multiple sub-layers of the same nanoplatelet material are deposited, in order to build up a required thickness for the nanoplatelet material layer. Similarly, in some embodiments, the intermediate smoothing layer is formed by deposition of multiple sub-layers, in order to build up a required thickness for the intermediate smoothing layer.
In some embodiments, the smoothing layer may function as an adhesive layer, adhering the subsequent layers with respect to the fabric.
Preferably, the ink is applied to the at least a part of the treated portion of the fabric substrate by inkjet printing.
It is to be understood that this disclosure provides approaches for the formation of different types of flexible electronic device, of different structure and of different degrees of complexity. At a simple level, the disclosure allows the formation of flexible electrical interconnects. At another level, the disclosure allows the formation of a photodetector device (for example). At still another level, the disclosure allows for the formation of a transistor device (for example). Still further, the disclosure allows for the formation of entire or partial electrical circuits, formed of a few or many flexible electronic components. The disclosure therefore permits the formation of integrated printed circuits on textile.
Further optional features of the invention are set out below.
Embodiments of the invention will now be described by way of example with reference to the accompanying drawings in which:
In this detailed description, various specific conditions, starting materials, processing equipment, analytical equipment, etc., are specified. However, it will be understood by the skilled person that different specific conditions, starting materials, processing equipment, analytical equipment, etc., can be used and yet substantially the same result achieved based on the general teaching provided by this disclosure.
In the present invention, two main types of modification techniques are used to modify the fabric substrates to promote adhesion of nanoplatelets in the ink dispersion to the fabric substrates and accordingly improve the quality of deposited ink layers. The first type of modification uses application of a smoothing or planarization layer to decrease the roughness of the fabric substrate. The second type of modification uses cationization or anionisation of at least a part of the fabric substrate to increase the affinity between deposited nanoplatelets and the fabric substrate. These two modification techniques may also be combined; i.e. a fabric substrate may first have a smoothing layer applied, and then may also undergo cationization or anionisation of at part of the substrate to further promote adhesion of the nanoplatelets on deposition.
Samples of fabric may be coated with polyurethane or a similar planarization material listed in
Textiles and fibres can be chemically modified to increase the affinity between the fabric and the GRM nanoplatelets, thus aiding the formation of a uniform GRM coating of the textile. For example, the fibres may be positively or negatively charged, increasing the electrostatic attraction between the fibres and the GRM nanoplatelets. Chemical modification of the fibre can be performed by acid treatment using, for example but not limited, to 3-chloro-2-hydropropane-sulfonic acid sodium (CHSAS) and monochloroacetic acid (MCAA) or 3-chloro-2-hydroxypropyl)trimethylammonium chloride. (CHPTAC). Other suitable reagents for cationization modification of textiles include bis-quaternary ammonium salt, or polymerizable bis-quaternary ammonium salt. Suitable reagents for anionization modification of textiles include surfactants with functional terminating groups such as sulfate, sulfonate, phosphate and carboxylates. However any reagent which is able to provide suitable cationization or anionization of the fabric may be used.
A cationization of the fabric may be performed using (3-chloro-2-hydroxypropyl)trimethylammonium chloride (CHPTAC) (35 g/L) (or a suitable replacement material as discussed above) and dissolved in deionized water at a water/cotton weight ratio of 15:1 respectively. The fabric is immersed in the chemical solution and left on a hot plate at 40° C. for 20 min while gentle stirring is applied. The fabric is then removed and lightly hand squeezed to remove excess water. The treated fabric is then sealed between polyethylene film, placed in a plastic bag and stored in an oven at 40° C. for approximately 24 h. After rinsing the treated fabric a few times with deionized water, the fabric is immersed in an aqueous acetic acid solution (1 g/L) for 5 minutes to neutralize the alkalinity. The fabric again rinsed in deionized water and oven dried overnight at 40° C.
The preferred graphene/GRM production method is LPE, however other suitable production methods may be used. LPE involves the production of 2D materials (by ultrasonication, high shear mixing or microfluidic processing) by exfoliation of bulk layered materials. The exfoliation process is generally performed in aqueous solution containing a stabilising agent (surfactant, polymer or other wrapping agent) or an organic solvent whose surface tension substantially matches the 2D material surface energy. After the exfoliation process, the resulting flakes have a thickness and lateral size distribution which may vary depending on the length, power, or type etc. of exfoliation technique used.
The yield of single layer graphene flakes after ultrasonication process has been demonstrated to reach up to 35% [Torrisi et al. 2012] in NMP and up to 80% in aqueous solution. Lower yields (up to 3%) for single layer graphene flakes have been shown in surfactant aided aqueous-based dispersion exfoliated by high shear mixing. Concentrations of GRM nanoplatelets (nanoplatelets here being defined as those with lateral size being a few microns and thicknesses being below 100 nm) up to 50 g/L have been demonstrated by high shear mixing process [Paton et al. 2014]. Graphene and functionalized graphene composed of graphene nanoplatelets (few layer graphene with aspect ratio of 1:200 in thickness:diameter) powder can also be used and dispersed in liquid by solution processing.
Graphene can be produced in solution by liquid phase exfoliating graphite (or graphene powder) via ultrasonication (or shear mixing or microfluidic exfoliation) both in aqueous and/or organic solvents. Preferably, the carrier liquid is selected from one or more of water, ethanol, NMP, chloroform, benzene, toluene, di-chlorobenzene, iso-propyl alcohol, ethanol and/or other organic solvents. Sonication is generally then followed by sedimentation based ultracentrifugation to purify the dispersion. After removing solid powder, the supernatant is obtained as the graphene ink.
We prepare Graphene-Ethanol ink (Gr-Eth) by ultrasonicating (1 hr) 5 mg/ml graphene nanoplatelets (GR1, Cambridge Nanosystems, CNS) in Ethanol. These nanoplatelets are produced by cracking methane and carbon dioxide gases in an enhanced plasma torch. The dispersion is then ultracentrifuged (Beckman Coulter Proteomelab XL-A mounting a SW 32 Ti swinging bucket rotor) at 10 k rpm for 1 hour and the top 70% is collected for the Gr-Eth and further characterization.
Three additional graphene inks were made. The first ink (Gr-Eth-HC) involved adding 10 mg/ml of graphene nanoplatelets (GR1, Cambridge Nanosystems, CNS) to Ethanol and was sonicated for 1 hour. No centrifugation was carried out on this ink. The second ink (Gr-DiW) involved ultrasonicating (Fisherbrand FB15069, Max power 800 W) natural graphite flakes (12 mg/ml) for 9 hours in deionized water with sodium deoxycholate (SDC, 9 mg/ml). The third ink (Gr-NMP) involved ultrasonicating natural graphite flakes (12 mg/ml) in N-Methyl-2-pyrrolidone (NMP) for 9 hours. The last two graphite dispersions are then ultracentrifuged (Sorvall WX100 mounting a TH-641 swinging bucket rotor) at 1 k rpm for 1 hour to remove thick (>10 nm) graphite flakes. The sediment is discarded while the top 70% of ink is re-centrifuged at 32 k rpm for 1 hour. The sediment is collected and it is re-dispersed in the solvent which was used to make the original dispersion. High concentration inks (10 mg/ml) can be made in this fashion.
Other methods for obtaining suitable graphene or GRM inks are outlined in WO2014/064432.
For inks intended for use in inkjet printing, nanoparticles in the ink should be smaller than the inkjet printing nozzle diameter. Typically, it is preferable that the nanoparticles are of the order of 50 times smaller than the nozzle size in order to reduce or avoid printing instability due to clustering of the particles at the nozzle edge which may cause deviation of drop trajectory, or agglomerates, which can cause unwanted blockages of the nozzle.
The surface tension may be measured using the pendant drop method (First Ten Angstroms FTA1000B). The shape of the drop suspended from a needle results from the relationship between the surface tension and gravity. The surface tension is then calculated from the shadow image of a pendant drop using drop shape analysis. A parallel plate rotational rheometer (DHR rheometer TA instruments (Gr-NMP and Gr-SDC inks) and Bohlin C-VOR Rheometer (Gr-Eth ink)) is used to evaluate the viscosity as a function of shear rate, the infinite-rate viscosity is found for the Gr-Eth, Gr-NMP and Gr-SDC inks. Ink density is evaluated from a (Sartorius ME5) microbalance where the density if the mass per unit volume (p=m/V). The viscosity is found to be the similar for each ink.
Rheological measurement of the inks may be beneficial, as rheology of the ink can determine the reliability of drop jetting during inkjet printing.
We derive viscosity (η), surface energy (γ) and density (ρ) as described above and estimate ηGr-Eth ˜2.2 mPa s (
Drops of inks are dispensed on holey carbon transmission electron microscopy (TEM) grids for high resolution transmission electron microscopy (HRTEM) analysis, using a Tecnai T20 high-resolution electron microscope with an acceleration voltage of 200 kV operating in Bright Field mode.
A Bruker Dimension Icon working in peakforce mode was used. For the characterisation of graphene powder, the sample was dispersed in ethanol and bath sonicated for 1 h. The dispersion was then centrifuged for 1 h at 10 krpm and the supernatant was collected, diluted 20 times in ethanol and 4 samples were drop casted on pre-cleaned Si/SiO2 substrates. Each sample was scanned across 3 different areas. Resulting AFM topographical and profile images can be seen in
Optical absorption spectroscopy (OAS) is used to estimate the concentration of the ink via the Beer-Lambert law according to the relation A=α cl, where A is the absorbance, l [m] is the light path length, c [g/L] is the concentration of dispersed graphitic material, and α [Lg−1 m−1] is the absorption coefficient.
Printable GRM inks could be chemically modified/functionalised to be positively or negatively charged by the use of chemical oxidation/reduction steps or functionalization by molecules with charged chemical bonds. For example positively charged graphene oxide (GO) ink can be synthesized by adding DDAB (30 mg) into a GO solution (10 mg/10 mL) in acidic surrounding followed by sonication.
A range of methods used for characterisation of the flexible components or devices are set out below.
The conductive graphene fabrics are washed with water containing soap and sodium carbonate. Copper tape is added to the edges of the fabric where necessary in order to preserve the position of the electrical contacts on the fabric.
A Rockwell indenter (100 μm) is used to apply a normal force to the sample from an initial load of 0.03N to 0.5N at a loading rate of 0.10 N/min while the friction force, acoustic emission (AE) was recorded. The cantilever is moved across the sample at a speed of 0.64 mm/min.
Raman measurements are collected with a Reinshaw 1000 InVia micro-Raman spectrometer at 514.5 nm and a ×50 objective, with an incident power of ˜0.3 mW.
The sample stripes or bundles can be placed between the machine grippers and a strain of 0.3 N/m{circumflex over ( )}2 is applied and stress measured until fracture.
The electrical resistance of printed 1 mm wide films was characterised using a 2-point probe across a distance of 1 cm.
Sheet resistance of dip-coated samples may be measured using a 4-point probe and reading off a Keithley meter
SEM imaging may be used to image the surface morphology of the fabric substrate, before and/or after deposition of a GRM nanoplatelet layer.
The inkjet printed circuits were prepared using a (Fujifilm Dimatix, DMP-2800) inkjet printer. Firstly a cartridge (Fujifilm DMC 11610) was filled with the prepared Gr-Eth ink and was deposited at an inter-drop spacing (i.e the centre to centre distance between two adjacent deposited droplets) of 25 μm onto cotton fabric coated with 1 layer of polyurethane. Once a droplet gets ejected it falls under the action of gravity until it contacts the substrate and spreads according to Young's equation, γSV−γSL−γLV cos θc=0, (where γSV [mJ m−2] is the solid-vapour surface energy, γSL the solid-liquid interfacial tension, and γLV the liquid-vapour surface tension). The drop then dries through solvent evaporation (the platen was kept 60° C. throughout printing) and the resulting thickness depends on the number of droplets delivered per unit area, the drop volume and the concentration of nanoplatelet material in the ink. Consequently a stripe of graphene ink is printed to our desired pattern as shown in
Raman spectroscopy (see methods) was undertaken on the printed conductive strip. The resulting Raman spectrum for this Example is shown in
The electrical resistance of the printed 1 mm wide films was characterised using a 2-point probe across a distance of 1 cm where it was found that the films reached percolation after 30 layers (i.e. printing passes) as shown in Table 1. As the number of the ink-jet printing layer increases, more and more flakes are deposited onto the surface. As a result, a stripe with flakes is gradually formed. Once the flakes connect with each other, the film becomes conductive.
Electrically conducting e-fabrics were fabricated by dip-coating of fabric (poplin 100% cotton) into graphene ink directly. No smoothing layer was used. Dip coating allows the ink to infiltrate deeper into the fabric than comparative surface coating techniques may allow. The cotton fabric first undergoes a chemical functionalization treatment as described in the method section above, in order to cationize the fabric before application of the ink. Some fabric samples do not undergo chemical modification, to provide comparative samples. The fabric samples are then respectively dip-coated into one of three respective inks: Gr-Eth-HC, Gr-DiW, and Gr-NMP, the formulation of each of which is discussed above.
Two types of cotton fabric were used in the following coating process, type 1 is a dense cotton fabric (7.4 tex) while type 2 (13.8 tex) has less threads of fibers per unit area. Modified and control cotton fabrics are then dipped into 20 mL of graphene ink of choice, the immersed fabric is then removed and dried at room conditions (21° C.) overnight. After drying the fabric is turned over and once again immersed in the ink and left to dry once more. The resulting fabrics are labeled the Gr-NMP-F, Gr-DiW-F and Gr-Eth-HC-F depending on the graphene ink which was used as a coating. In order to identify the most effective chemical functionalization treatment the Gr-Eth-HC ink was applied to fabric modified with three different cations: 3-chloro-2-hydroxypropyltrimethylammonium chloride, bis-quaternary ammonium salt and polymerizable bis-quaternary ammonium salt. These samples were be labeled Gr-Eth-HC-F-1 to Gr-Eth-HC-F-8 indicating the sample number.
The resulting dip-coated fabric samples (Gr-Eth-HC-F-1 to Gr-Eth-HC-F-8) are characterized by Raman spectroscopy as shown in
The Gr-Eth-HC-F-8 sample was also characterized with scanning electron microscopy both before (
The fabric electrical resistance of the samples was then measured using a 2-point probe across a distance of 1 cm, using 1 cm2 pieces of cloth. Silver paint (agar scientific) was used to paint on contacts. The fabrics had an electrical resistance of 0.43±0.35 kΩ, 18±4 kΩ and 51±18 kΩ for the Gr-NMP-F, Gr-DiW-F and Gr-Eth-HC-F fabrics respectively. The washability of these fabrics was then tested by covering both side contacts of the samples with copper tape to avoid damage, and implementing the fabric washing process (see methods). A layer of polyurethane protective coating is laminated on top of the graphene-coated fabric, while uncoated graphene-fabrics also undergo the same treatment as control samples.
In order to investigate the effect of different types of chemical modification on the fabric samples, the samples were subjected to a number of mechanical tests. Scratch tests with an atomic force microscope (AFM) were undertaken (see methods) on the samples in order to determine the extent of the graphene adhesion to the fabrics. The investigation assumes that the same cotton fibre has been used (i.e of equal linear density) and that only the coating chemistry has been varied. Furthermore it is assumed that the thickness of the coatings of the different samples are in close precision to one other. As shown in
The fabrics were also subjected to tensile testing (see methods), the dip coated samples were tested as strips (cutting rectangular parts of the dip coated samples) and also as bundles (i.e a collection of fibrils taken from each of the fabric samples). From
We can see from the above figures that the strain to break of the bundles is higher (about 5%) as a consequence of the modification while the strength of the fibers remains approximately consistent. Without wishing to be bound by theory, the inventors speculate that this could be due to the increased graphene pickup as a result of the fabric cationization. The fabrics are induced with a positive charge due to the cationization while the negatively charged OH groups on the edges of the graphene flakes result in a net attraction between the two materials resulting in an increase in pickup.
At the time of writing, it is considered that dip-coating is a less preferred approach to the formation of the embodiments of the invention, compared for example with inkjet printing. One reason for this is that inkjet printing can be carried out at high resolution, forming the material layers in a desired pattern in one process. Another reason for this is that inkjet deposition appears to form superior quality layers of deposited GRM material.
Two inks were manufactured. The first was a graphene ink, the second was a MoS2 ink. Each used a particle size of <30 microns mixed at a concentration >50 g/L with sodium deoxycholate surfactant (SDC) at a concentration 9 g/L in water and stir bar mixed for 5 min. Then each dispersion was exfoliated by high shear mixer for 1 hour. The final product of the exfoliated material was collected. Cellulose (CMC) was continuously added whilst stirring to adjust the viscosity to the required value. CMC was slowly added until fully dissolved.
The textile fibre was chemically modified as follows. After washing with deionized water, the fabrics were cationized using the exhaust method at room temperature in a weight ratio of 17:1. The cationization was performed with CHPTAC concentration 35 g/L. A 60 g fabric sample was first immersed in the solution of CHPTAC. Following this, NaOH was added to the solution to achieve a CHPTAC/NaOH ratio of 2.33. The fabric was gently stirred and left for 20 min, then removed and hand squeezed to remove excess water. The wet pick-up was approximately 100%. The treated fabric was then placed in a plastic bag to prevent chemical migration and water evaporation and stored at room temperature for approximately 24 h. After rinsing 5 times with tap water, the treated fabric was immersed in an acetic acid solution (1 g/L) for 3-5 min to neutralize the alkalinity.
Graphene ink was flexographically printed (or printed with any other suitable printing/coating technique as described previously) to deposit a thin film, of thickness approximately 500 nm, acting as electrode on the functionalised textile. Subsequently a MoS2 ink was flexographically printed to produce an equally-thick MoS2 film. A graphene film was flexographically printed on the top of the stack.
The graphene-MoS2-graphene heterostructure was protected by applying a protective polymer (e.g. polyurethane) coating on top of the conductive graphene interconnection, generally by bar-coating or screen printing. This heterostructure device represents a wearable and washable GRM-printed photodetector on textiles.
In this part of the disclosure, we demonstrate the fabrication of flexible and washable fully inkjet printed graphene/hexagonal-boron nitride field effect transistors (FETs) on polyethylene terephthalate (PET) film and on polyester fabric. The devices have a charge carrier mobility of as high as μh=150±18 cm2 V−1 s−1 on polyethylene terephthalate (PET) film and μe=73±23 cm2 V−1 s−1 on polyester fabric, at low operating voltages (<5 V). In the preferred embodiment described here, the FET is fabricated by inkjet printing heterostructures of graphene and h-BN inks prepared by scalable liquid phase exfoliation and microfludization production techniques, respectively. The devices remained operational and maintained their performance even under strain of bending radius 4 mm. The printed FETs show stable operation for periods up to 2 years, indicating the two-fold role of the h-BN layer as a dielectric and encapsulant. Finally, we demonstrated that the hexagonal-boron nitride textile FETs are washable up to 20 cycles using an encapsulation layer (formed in this embodiment from polyurethane) which is ideal for applications in wearable and textile electronics. The FET is sometimes referred to here as a thin film transistor (TFT).
FET—Field effect transistor
TFT—Thin film transistor
h-BN—Hexagonal Boron Nitride
NMP—N-Methyl-2-pyrrolidone
CMC—Carboxymethylcellulose sodium salt
SEM—Scanning electron microscopy
EDX—Energy-dispersive X-ray spectroscopy
rGO—Reduced graphene oxide
CNT—Carbon nanotube
PVA—poly(vinyl alcohol)
PEDOT: PSS—poly(3,4-ethylenedioxythiophene) polystyrene sulfonate
SAA—Sodium alga acid
PQAS—Polyurethane, Polymerizable quaternary ammonium salt
HAADF-STEM—High angle annular dark field scanning transmission electron microscopy
NMF—Non-negative matrix factorization
Metal oxide semiconductor technology has dominated the electronics industry for the last century, however this technology is incompatible with printed electronics due to poor tensile performance metals and metal oxides have with flexible substrate materials such as polymers and textiles [De and Coleman (2011)]. The discovery and development of electrically conductive organic polymers Hideki et al (1977); Heeger (2001)] advanced the field of printed electronics allowing the manufacture of flexible devices with solution processibility, enabling large scale manufacture [Sirringhaus et al (2000)]. However both metal oxides and organic polymers have low charge mobility (μ) (˜0.01-10 cm2/Vs), which has limited their prospects in specific applications such as RFID tags and control electronics for displays [Nathan et al (2012)]. The exfoliation of graphene [Novoselov et al (2005), (2005) and (2004)] has driven a surge of exploration of novel two-dimensional (2D) materials with unique properties [Geim and Grigorieva (2013); Ferrari et al (2014)]. Graphene has shown great potential in the field of printed electronics owing to its mechanical flexibility [Gomez De Arco et al (2010)], stretchability [Lee et al (2008)], thermal conductivity [Yao et al (2016)], high electrical conductivity [Novoselov (2005)], environmental stability [Liu et al (2015)] and compatibility with low cost large scale manufacturing [Paton et al (2014)]. Moreover solution processed graphene field effect transistors (FETs) can have a high carrier mobility (about 100 cm2/Vs) [Torrisi et al (2012)] coupled with an ambipolar behaviour which make it an attractive material for radio frequency applications [Akinwande (2014)].
Atomically thin 2D materials, identified by their intralayer covalent bonding and interlayer van der Waals bonding, such as graphene [Novoselov et al (2005), (2005) and (2004)] and boron nitride (BN) [Novoselov et al (2005); Geim and Grigorieva (2013)], can be arranged into heterostructures, thus creating structures with novel properties which are different from those of the individual components [Novoselov (2011)]. The combination of conducting, insulating and semiconducting 2D materials in different combinations allows for a practically infinite number of different heterostructures with precisely tailored properties with multiple functionalities and improved performance for novel applications [Novoselov (2011); Wither et al (2015)]. These 2D materials can be exfoliated in solution by liquid phase exfoliation (LPE) or microfluidization and developed into inks [Nicolosi et al (2013); Lotya et al (2009); Hernandez (2008); Karagiannidis et al (2017)]. Consequently, layered structures of 2D material ink can then be printed in part or as whole by means of different printing technologies such as inkjet [Torrisi (2012); Kelly et al (2016)], spray [Kelly et al (2016)], screen {Gualandi (2016)], gravure [Lau et al (2013)] and flexographic printing [Yan et al (2009)]. These printed techniques offer a competitive advantage over conventional silicon based electronics as the high-vacuum equipment, subtractive processes and lithography add to the number of processing steps and the overall cost involved Baeg et al (2013)]. Thus, there have a myriad of printed electronics applications which been developed over the past two decades, such as organic light-emitting diodes (OLED) [Kopola et al (2009)], photovoltaic devices [Krebs et al (2009)] and transistors [Sirringhaus (2000)]. Perhaps even more interestingly is the adaptation of many printed devices for applications in wearable electronics such as thermoelectric power generators [Kim et al (2014)], sensors [Gualandi et al (2016)], RFID [Lakafosis et al (2010)], energy storage [Chen et al (2010)] and antennas [Chauraya et al (2013)] which enhance the users ease of integration with external electronics while providing analytical information to the wearer by monitoring functions such as movement [Ren et al (2017)].
In this disclosure inkjet printing is chosen to print FETs on polyethylene terephthalate (PET) and polyester as it is a non-contact, well controlled one step deposition and patterning of inks on any substrate and at room temperature, moreover it is a scalable technique amenable for mass production [Krebs review article (2009)]. Inkjet printing also offers reduced material wastage when compared to other printing due to the small amount of material it uses (typically about 3 ml) and has excellent control over the deposition of ink which can be used to create very complex patterns with high resolution (about 20 μm) [Krebs review article (2009)]. Graphene and BN inks are formulated though LPE and microfluidization respectively and are subsequently inkjet printed with a commercially available silver and PEDOT: PSS inks to fabricate FET heterostructures in arrays at room temperature and ambient pressure. The devices achieved an exceptionally high mobility up to 150 cm2/Vs on PET and up to 73 cm2/Vs on polyester fabric which was coated with a polyurethane planarization layer. The flexibility and washability of the devices was also examined to establish their applicability in real world applications.
In this study we used a drop-on-demand ink jet printer (Fujifilm Dimatix DMP-2800). The viscosity, η[mPa s], surface tension, γ[mN m−1], density, ρ[g cm−3] and nozzle diameter, a [μm] influence the jetting of individual drops from a nozzle [Derby and Reis (2003)]. During droplet ejection a primary drop may be followed by secondary (satellite) droplets which need to be avoided during printing [Dong et al (2006); Jang et al (2009)]. The inverse Ohnesorge number is used as a figure of merit, Z=(γρa)1/2/η and is commonly used to characterize the drop formation, stability and assess the jettability of an ink from a nozzle [Derby and Reis (2003); Dong et al (2006); Fromm (1984)]. A range of 2<Z<24 has been identified as an optimal range which minimizes the number of satellite droplets and improves stability [Torrisi et al (2012); Fromm (1984)]. Additionally, nozzle clogging can be an issue unless the particles have diameter of about 1/50 or less times the nozzle diameter [Torrisi et al (2012)]. Therefore we used a 21 μm diameter nozzle (Fujifilm DMC-11610) where the volume of individual droplets from this nozzle is about 10 μL. When inkjet printing, the ejected drop falls under the action of gravity until it contacts the substrate and spreads according to Young's equation, γSV−γSL−γLV cos θc=0, (where γSV is the solid-vapor surface energy, γSL the solid-liquid interfacial tension, and γLV the liquid-vapor surface tension) [Ryntz and Yaeneff (2003)]. The drop then dries through solvent evaporation (the platen was kept 20° C. throughout printing) and the resulting thickness will depend on the number of droplets delivered per unit area (controlled by the interdrop spacing, i.e the centre to centre distance between two adjacent deposited droplets), the drop volume and the concentration of material in the ink.
Suitable inkjet printable formulations which are produced by liquid phase exfoliation [Lotya et al (2009); Hernandez et al (2008)] typically contain surfactants or polymer stabilization agents which can act as a source of contamination which can hinder device performance however they can also positively impact the ink by acting as an adhesion or rheology modifier [Karagiannidis et al (2017)]. High boiling point solvents (>150° C.), such as N-Methyl-2-pyrrolidone (NMP) can stabilize 2D materials without stabilization agents due to a matching of the Hansen solubility parameters [Lotya et al (2009); Hernandez et al (2008); Hansen (2007)]. However, they are still far from ideal as they are based on toxic and expensive solvents which require high annealing temperatures (>150° C.) to remove residual solvent [McManus et al (2017)]. Low boiling point inks (<150° C.) are a suitable alternative, due to their fast evaporation at room temperature and have been reported though two solvent formulation where the mixture is tuned to improve the affinity of the solvent to the 2D crystals [Zhou et al (2011)]. However the different evaporation rate of the two solvents can result in rheological instabilities and particle aggregation over time. An alternative ink formulation route is though solvent exchange whereby 2D materials can be exfoliated effectively in a high boiling point solvent and subsequently transferred to a low boiling point solvent and concentration as desired [Zhang et al (2010)].
We prepare the 2D crystal-based inks are prepared as follows. The graphene ink is prepared by dispersing graphite flakes (10 mg/ml, Sigma-Aldrich No. 332461) and ultrasonicating (Fisherbrand FB15069, Max power 800 W) for 9 hours in NMP [Hernandez et al (2008)]. The graphene ink in NMP then undergoes a solvent exchange to ethanol (see methods, described below). The h-BN ink is prepared by mixing h-BN powder (10 mg/ml, Goodfellows <10 μm, B516011) with deionized water and carboxymethylcellulose sodium salt (CMC, Average Molecular Weight MW=700,000, Aldrich No. 419338) (3 mg/ml), a biocompatible and biodegradable stabilization agent and rheology modifier [Karagiannidis et al (2017); Lin et al (2015)]. The h-BN/CMC mixture is then processed with a shear fluid processor, (i.e. a microfluidizer, M-110P, Microfluidics International Corporation, Westwood, Mass., USA) with a Z-type geometry interaction chamber with microchannels about 87 μm wide for 50 cycles, at 207 MPa system pressure and room temperature (20° C.) [Karagiannidis et al (2017)]. We use the microfluidic process to disperse and exfoliate h-BN while the high shear rate generated (about 9.2×107 s−1) helps to achieve high concentration dispersions [Karagiannidis et al (2017)]. The h-BN and graphene dispersions are then ultracentrifuged (Sorvall WX100 mounting a TH-641 swinging bucket rotor) at 3 k rpm (20 min) and 10 k rpm (1 hour) respectively to remove thick flakes which would clog printer nozzles. Subsequently, the supernatant (i.e top 70%) is decanted for further characterization. The rheological parameters (viscosity η, surface tension γ, density ρ) for both inks are determined as ηBN of about 1.7 mPa s, γBN of about 72 mN/m, ρBN of about 1.01 g cm−3; ηGR of about 1 mPa s, ρGR of about 30 mN/m, ρGR of about 0.82 g cm−3, consistent with previous reports [Torrisi et al (2012); Lotya et al (2009); Hernandez et al (2008)]. Consequently, we find a Z number for the h-BN (Z of about 19.4) and graphene (Z of about 22) inks, which are within the optimal Z range [Torrisi et al (2012); Fromm (1984)].
Optical absorption spectroscopy (OAS) can estimate the flake concentration [Lotya et al (2009); Hernandez et al (2008)] via the Beer-Lambert law which correlates the absorbance A=αcl, with the beam path length I [m] the concentration c [g/L] and the absorption coefficient α [L g−1 m−1].
The average lateral size and thickness of the graphene and h-BN flakes are estimated by atomic force microscopy (AFM).
Inkjet Printed h-BN Capacitors:
We investigate the dielectric properties of the h-BN ink in a Ag/h-BN/Ag parallel plate capacitor configuration.
We first investigate bottom-gate top-contact (inverted staggered) and top-gate top-contact (coplanar) TFT structures and optimize the inkjet printed graphene/h-BN heterostructures on a PET substrate (Novele, Novacentrix) before moving to the technology onto polyester textile. The inverted staggered TFT structure is built up as shown through the schematic in
Raman spectroscopy (Reinshaw 1000 InVia micro-Raman) is used to monitor the quality of materials used in the heterostructure.
We then characterise the output and transfer electrical characteristics of both coplanar (
The field effect mobility (μ) of the coplanar and inverted staggered devices are derived from the slope of the transfer characteristic according to p=(L/W*C*Vds)/(dld/dVgs), where L [μm] and W [μm] are the channel length and width, respectively, and C dielectric capacitance [Schwierz (2010)]. We use the previously calculated dielectric capacitance of 8.7 nF/cm2 at a drain voltages of 1 Vds. The hole mobility (μh) and electron mobility (μe) of the coplanar devices are calculated to be 150±18 cm2 V−1 s−1 and 78±10 cm2 V−1 s−1 respectively while having an on/off current ratio (defined as the maximum Id divided by the minimum Id) of about 2.5±0.1. For the inverted staggered devices we find an ON/OFF ratio of about 1.5±0.2, μh=32±5 cm2 V−1 s−1 and μe=10±4 cm2 V−1 s−1 which is one magnitude lower than the non-inverted structure field effect mobility on PET, we attribute this decrease in mobility to the rougher surface of the h-BN layer (Rq=68 nm, determined by AFM) in contrast to the PET film (Rq=15.2 nm) which could affect the stacking quality of the graphene flakes. Such difference between hole and electron mobility corresponds to a preferential hole conduction over electron conduction, which may be due in part to the unintentional extrinsic doping [Lemme at al (2008); Liang et al (2010)]. Such preferential hole conduction has been reported for various sources of graphene, including graphene synthesized by CVD[Suk et al (2013)] and mechanical exfoliation [Lemme at al (2008)]. The field effect mobility is higher than printed carbon nanotube TFT's (p of about 20 cm2 V−1 s−1, on/off of about 104) [Ha et al (2010)] and is about 15 times higher than the best organic (p of about 10.5 cm2 V−1 s−1, on/off of about 106) [Li et al (20120)] and oxide transistors (μe of about 9 cm2 V−1 s−1, on/off of about 107) [Huang at al (2016)] while comparable to inkjet printed graphene TFT's (p=95 cm2 V−1 s−1, on/off of about 10) [Torrisi et al (2012)] and reduced graphene oxide (rGO) transistors (p of about 210 cm2 V−1 s−1, on/off of about 3) [Su et al (2010)]. However the on/off ratio is lower than that of organic, oxide and CNT transistors [Ha et al (2010); Li et al (20120); Huang at al (2016)], this is however consistent with the on/off measured on previously reported TFTs from graphene [Torrisi (2012); Su et al (2010)]. The flexibility of the coplanar device was tested as a function of bending radius using metal rods (
The decrease in field effect mobility resulting from the small (about 50 nm) increase in surface roughness between the coplanar and inverted staggered heterostructure on PET emphasizes the importance of roughness minimisation for the implementation of high performance devices on textile where Rq is typically in the range of about 30 μm. Therefore before transferring the inverted staggered heterostructure to textile, we adopt an additional solution to improve performance in our textile devices through the use of a planarization layer. Typically building components on the weave of the textile requires the use of a planarization layer such as Polydimethylsiloxane (PDMS) [Khan et al (2012)], polyimide [Sekitani et al (2010)], polyurethane [Kim et al (2013)] or poly(vinyl alcohol) (PVA) [Kim et al (2015)] to decrease the rms roughness and thus improve performance of devices [Peng and Change (2014)]. For example, Kim et al (2013) used laminated polyurethane (t of about 20-50 μm) on polyester reducing the rms roughness from 10 μm to <5 μm, while Sekitani et al (2010) used spin coated polyimide (t of about 500 nm) on polyimide, reducing the rms roughness from 2.5 nm to 0.3 nm. Here, we choose to use polyester satin fabric as a substrate for our wearable graphene-h-BN TFTs because it is very durable and represents about about 80% of the 2016 synthetic fibre market [Krifa and Stewart-Stevens (2016)]. To determine a suitable planarization layer we rod coat (K202 RK coating machine) the polyester with eight different materials; sodium alga acid (SAA), gelatin, arabic gum, guar gum, xanthan gum, sodium carboxymethylcellulose (CMC), polyurethane, polymerizable quaternary ammonium salt (PQAS) and measured their rms roughness using a profilometer (DektakXT, Bruker) (
In addition, wearable electronic devices require not only flexibility, but to preserve the same stretchability of the fabric with little or no effect on the electrical and optical performances. Hence, we replace the printed silver electrodes with a stretchable polymer such as poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) (Sigma-Aldrich, 739316, 0.8 w/v in H2O) (Z of about 30) [Vosgueritchian et al (2012)].
Wearable textile devices will normally undergo naturally occurring tensile strain as well as washing steps [Ren et al (2017)]. We then investigate the effect of bending (
In
Additional work is now reported on Graphene/h-BN/Graphene fabric capacitors. A cotton or polyester fabric (but not limited to) in size of 1 cm×2 cm is cleaned by deionized water and then are dried in oven at 60° C. The fabric can optionally be treated with a cationic or anionic modification agent to improve adhesion of the 2d material. The cleaned polyester fabrics are immersed into a graphene dispersion for 3 min with continuous stirring. Then the soaked fabric is stuck on glass slide and dried at 60° C. for 5 min. This ‘dip and dry’ procedure can be marked as one cycle and repeated for several cycles to put more graphene into the fabric. Then the graphene fabrics are processed by hot pressing at 200° C. for several minutes. This can be repeated for a h-BN dispersion to create h-BN fabrics. The graphene/h-BN/graphene structure can then be assembled together by using PVA glue at the edges of the fabrics. The structure is then hot pressed again to improve adherence between the layers.
As indicated above, it is also possible to fully inkjet print flexible electronic components, including complete circuits, according to embodiments of the invention.
We have demonstrated fully inkjet printed graphene FETs on PET and polyester fabric, and more complex electronic components. Both LPE and microfluidization inks are ideal low-cost production techniques to engineering printable inks for heterostructure devices. These inks can be easily deposited by inkjet creating FET heterostructures on demand. We show that the mobility of these devices decreases significantly as the channel roughness increases. The devices are flexible and maintain their functionality over time even over periods of 2 years. Moreover the FETs on textile are demonstrated to be washable for up to 20 cycles, enhancing their lifetime which can cut replacement costs and improve compatibly with current textile industry technologies. These transistors present a new application for 2D inks in active devices with the competitive advantage over conventional silicon based electronics as they are fully printed at room temperature minimising the number of processing steps and the overall cost involved.
We refer to the experimental methods set out earlier. Here, certain additional methods, applicable to the reported work on inkjet printed electronic devices, are set out.
First (˜20 ml) of graphene/NMP ink is passed through a PTFE membrane (Merck Millipore, 0.1 μm). The process is hastened with the use of Büchner flask which is attached to a vacuum pump. The membrane is then placed into 5 ml of ethanol and bath sonicated (Fisherbrand FB15069, Max power 800 W) for 10 min to redisperse the flakes into the ethanol.
Films of each ink and a Gr/h-BN heterostructure are inkjet printed on Si/SiO2 substrate and the Raman spectra are acquired with a Reinshaw 1000 InVia micro-Raman spectrometer at 457, 514.5, and 633 nm and a ×20 objective, with an incident power of below ˜1 mW to avoid possible thermal damage. The G peak dispersion is defined as Disp(G)=ΔPos(G)/ΔλL, where λL is the laser excitation wavelength.
Scanning electron microscopy images were taken with a high resolution Magellan 400 L scanning electron microscope (SEM). The field emission gun was operated at an accelerating voltage of 5 KeV and gun current of 6.3 pA. Images were obtained in secondary electron detection mode using an immersion lens and TLD detector.
A Bruker Dimension Icon working in peakforce mode was used. From the centrifuged graphene and BN dispersions samples were collected and after 10 times dilution they were drop casted onto pre-cleaned (with acetone and isopropanol) Si/SiO2 substrates wafer substrates. For the graphene and BN inks, 150 flakes were counted to determine the statistics for the lateral size and thickness. For the rms roughness measurements areas of 50 μm2 were scanned.
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.
All references referred to above are hereby incorporated by reference.
| Number | Date | Country | Kind |
|---|---|---|---|
| 1616140.8 | Sep 2016 | GB | national |
| 1705742.3 | Apr 2017 | GB | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2017/073827 | 9/20/2017 | WO | 00 |
