Field of the Invention

The present invention relates to fabricating a flexible pressure sensor array. The sensor array can provide a pressure map in addition to the magnitude of pressure; hence providing tactile information of the pressure exerted on the sensor array.

Background

Flexible pressure sensors have drawn a great deal of attention because of their applications in electronic skins and wearable electronics. Piezoresistive pressure sensors that transduce any deformation to an electrical resistance are attractive due to the easy and simple fabrication method, low cost and fast processing.

The main two components of flexible pressure sensors are conductive elements and elastic polymers. However, traditional piezoresistive sensors, based on semiconductors and metal foils, although cost- effective, are rigid and have low measurement ranges, so there is a limit to their wearable electronics applications. Carbon nanomaterials, such as graphene, graphene oxide (GO), and carbon nanotubes, are alternative materials with superior flexibility and a wider working range. GO is a single layer of carbon atoms containing hydroxyl, epoxy, diol, and carbonyl functional groups and has been produced by rough oxidation using Hummer's method. These functional groups make it very hydrophilic and easily dispersible in water. A more stable dispersion can also be obtained compared to that with graphene. GO is usually converted into fibers, foams, woven and non-woven fabrics to reduce its brittleness in achieving highly sensitive and flexible sensors. Three-dimensional (3D) conductive foams have attracted great consideration in fabricating stretchable and flexible pressure sensors enabling low density, mechanical flexibility and electrical conductivity. Dip coating of GO on a polymer template is a convenient, inexpensive and facile approach that provides an efficient method to control the structure of the 3D foam. The conductivity of GO is low due to oxygen-containing groups. It needs to be reduced to decrease its oxygen groups and restore its electrical conductivity to some extent.

Thermal and chemical reduction methods have been employed to convert it into reduced-GO (rGO). However, using hazardous chemicals, complicated and long processes, and expensive equipment are some drawbacks of the aforementioned methods.

Summary of Invention

In a first aspect the present invention consists in a pressure sensor comprising a reduced graphene oxide foam layer and printed circuit board, adhered together.

Preferably the foam layer and PCB are attached together to form a sensing surface that is capable of producing a pressure map and/or magnitude of a force applied to the surface.

Preferably the foam layer and PCB are encapsulated in a flexible material. Preferably the flexible material is an elastomeric material.

Preferably the foam layer and PCB are adhered together using an adhesive.

Preferably the adhesive is a conductive paste.

Preferably the conductive paste is a silver conductive paste.

Preferably the printed circuit board may be flexible, rigid or flexible and stretchable.

Preferably the printed circuit board comprises two electrode layers.

Preferably the sensor is a sensor array.

In a second aspect the present invention consists in a method of producing a sensor comprising the steps of: a. dispersing a graphene oxide powder in water to form a suspension, b. dip coating a piece of polyurethane foam into the suspension to coat it, c. drying the foam, d. attaching the foam to a printed circuit board.

Preferably the resulting sensor is preloaded so as to break weak bonds within the foam.

Preferably the preloading is to subject the sensor to up to 30 kPa pressure for a period of time.

Preferably the predetermined time is at least 30 minutes.

In a third aspect the present invention consists in a method of producing a sensor comprising the steps of: a. dispersing graphene oxide powder in deionised water to form a suspension, b. dip coating a piece of polyurethane foam into the suspension to coat the foam with graphene oxide, c. drying the dipped foam, d. ethanol flame treating the dipped foam, such that the foam is burnt away, and the graphene oxide is converted to reduced graphene oxide, leaving a reduced graphene oxide film, and e. attaching the reduced graphene oxide film onto the printed circuit board.

Preferably the method including the step of: f. Pouring an elastomeric material onto the reduced graphene oxide film and printed circuit board, and oven curing it. Preferably step e includes adhering the reduced graphene oxide film to the printed circuit board using a conductive paste.

Preferably the resulting sensor is preloaded to break weak bonds within the foam.

Preferably the preloading is to subject the sensor to up to 30 kPa pressure for a period of time.

Preferably the predetermined time is at least 30 minutes.

In a fourth aspect the present invention consists in a method of producing a sensor comprising: a. ethanol flame treating graphene oxide powder to produce reduced graphene oxide powder, b. dispersion of reduced graphene oxide powder in deionised water to form a suspension, c. dip coating a piece of polyurethane foam into the suspension, thereby to coat the foam with reduced graphene oxide, d. oven drying the PU reduced graphene oxide foam, and e. attaching the PU reduced graphene oxide foam onto a printed circuit board.

Preferably the method includes the step of: f. pouring an elastomeric material onto the PU reduced graphene oxide foam and printed circuit board, and oven curing it.

Preferably the resulting sensor is preloaded so as to break weak bonds within the foam.

Preferably the preloading is to subject the sensor to up to 30 kPa pressure for a period of time.

Preferably the predetermined time is at least 30 minutes.

In any of the methods described above for making a sensor the water and graphene oxide suspension may be obtained by mixing using ultrasonic treatment and a magnetic stirrer.

Preferably the ultrasonic treatment is for 1 hour.

Preferably the suspension is stirred with a magnetic stirrer for 4 hours.

Preferably the flame treatment is above 440°C, but more preferably 600°C.

The disclosed subject matter also provides method or system which may broadly be said to consist in the parts, elements and features referred to or indicated in this specification, individually or collectively, in any or all combinations of two or more of those parts, elements or features. Where specific integers are mentioned in this specification which have known equivalents in the art to which the invention relates, such known equivalents are deemed to be incorporated in the specification.

Further aspects of the invention, which should be considered in all its novel aspects, will become apparent from the following description.

Drawing Description

A number of embodiments of the invention will now be described by way of example with reference to the drawings as follows.

Figure 1 is a Schematic of a 2-layer PCB embodiment for a 4 x 4 sensei array of the sensor of the present invention.

Figure 2 shows a fabrication process flow of one embodiment of a sensor of the present invention, a tactile pressure sensor with a flexible PCB.

Figure 3 show graphs of the Thermogravimetric Analysis (TGA) and FTIR spectrum results where a) are the TGA results, and b) are the FTIR spectrum of GO, PU foam, and PU foam coated with GO.

Figure 4 shows scanning electron microscopy (SEM) images of PU foam, PU foam coated with GO and rGO film.

Figure 5 shows graphs of signal to noise ratios for a) GO amount, b) Ethanol flame time, c) Ecoflex amount.

Figure 6 shows Pareto ANOVA analysis for contribution share of different parameters when arriving at minimum resistance for those parameters noted in Figure 5.

Figure 7 shows a picture of a sensor of the present invention a) before applying load, and b) after preloading and releasing.

Figure 8 shows the setup of a test rig fortesting the sensor of the present invention, a) shows a schematic representation of the characterization setup, (b) shows the actual test apparatus to conduct electromechanical characterisation, (c) shows a cross-section of the sensor showing four senseis.

Figure 9 shows a schematic view of a sensor of the present invention after applying a load to prestress it.

Figure 10 shows graphs as a result of electro-mechanical tests on the sensor of the present invention, where (a) shows relative resistance change of the sensor for different pressures, (b) shows 10 cycles relative resistance change of the sensor subjected to cyclic pressure 30 kPa, (c) shows the sensitivity of the sensor at different pressure range, (d) shows 10 cycles Hysteresis performance of the sensor, inset shows the hysteresis of the sensor.

Figure 11 shows graphs comparing the relative resistance of the sensor where the sensor (a) 1 mg/ml GO concentration in DI water and (b) 4 mg/ml GO concentration in DI water.

Figure 12 shows the block diagram of the proposed electronics that interfaces the sensor to Bluetooth, as an example.

Figure 13 shows resultant pressure maps for different objects and weights placed on top of a sensor of the present invention. Figure 14 shows a graph of 10 cycles of relative resistance change of two sensors, each made by different methods, subjected to a cyclic pressure of 30kPa.

Figure 15 shows the sensitivity (S) values of each of the sensors from Figure 14 in the 0-30 kPa pressure range.

Figures 16a and b show the spread of relative resistance changes of various sensor samples at different pressures.

Detailed Description

The present invention is a method of fabricating a flexible pressure sensor array and a sensor or sensor array. A sensor array can provide a pressure map in addition to a magnitude of pressure; hence providing tactile information of the pressure exerted on the sensor array. More particularly, the invention relates to the combination of a foam, preferably an rGO foam, and a PCB where the PCB is at the bottom, and rGO foam attaches to it. The PCB can be a typical PCB or flexible PCB. This new combination prevents resistance change of the electrodes after applying pressure. Importantly, the electrodes are all on the PCB and so all connectivity is only on one side of the sensor, providing ease of use and eliminating the electrical connection (or electrodes) exposed to physical contact with the object or load, over traditional 2-sided sensors.

To avoid the use of hazardous chemicals and complicated and long processes, for the fabrication method of the present invention, we used an ethanol flame treatment to reduce graphene oxide (GO), proposed by Du et al ¹ . It is an ultrafast, simple, convenient, and cost-effective method for reducing GO. In the preferred method of the present invention, a printed circuit board (PCB) is attached at the bottom of the reduced graphene oxide (rGO) foam to make electrical contacts (electrodes) with the rGO foam as sensing elements to realise a sensor array; hence eliminating any physical contact with the electrodes when used as a tactile pressure sensor array. The change in resistance for each sensing element (sensei) provides the magnitude of the local pressure; hence with many senseis a pressure map showing the pressure distribution can be generated.

1. Materials

Common kitchen polyurethane (PU) foam (sponge) obtained from a supermarket was used as the template or scaffold for the GO. However, any polyurethane may be used with the method of the present invention. GO with the particle size of around 15 pm was supplied from Graphenea Inc., Spain. However, any graphene from any supplier may be used, the preference being that the graphene should be between 10-20um in size. Silver conductive paste, elastomer, copper tape (or PCB copper traces are used in electrodes), deionized (DI) water and ethanol are the other materials used to fabricate the sensor.

¹ X. Du, H.-Y. Liu, Y.-W. Mai, Ultrafast synthesis of multifunctional N-doped graphene foam in an ethanol flame, ACS Nano. 10

(2016) 453-462 htt s_ 7dci_._orTg/_1_0_._1_02_1 /acsnan_o._5b0537_3. 2. Fabrication of flexible tactile pressure sensor

The sensor of the present invention is fabricated by dispersion of either a reduced graphene oxide powder or graphene oxide powder in water to form a suspension. Dip coating a piece of foam in the suspension and oven drying the dipped foam. The foam may then be reduced (in the case of use of graphene oxide powder). Then the resulting foam or film is attached to a printed circuit board.

A first preferred method of producing the sensor of the present invention will now be described. Experiments with different GO materials is also described. GO was dispersed in distilled water (DI) water with a concentration of 1 mg/ml was well mixed with one-hour ultrasonic treatment followed by 4 hours magnetic stirrerto create a DI/GO suspension. After cutting the PU foam into the desired size; in this case 40mm (length) x 40mm (width) x 3 mm (height) was preferred, but other sizes appropriate to the proposed use may be used, it was washed with DI water and ethanol three times and then dried in the oven (85°C) for 4 hours to remove any remaining water.

The PU foam was then dipped coated in the DI water solution with GO and then squeezed and resoaked three times to ensure sufficient GO was absorbed into the foam and finally dried in the oven at 85°C. The dip coating process was repeated twice to ensure adequate amount of GO was absorbed into the foam. Finally, the sample was exposed to an ethanol flame for 30 seconds to reduce the GO and also burn off the PU foam. The resulting sample was a conductive rGO film. The density of the rGO film, made by this first method was tested at approximately 4.15 mg/cm ³ .

The fabrication process of a tactile pressure sensor of the present invention, made according to the first method described above, and shown in the flow diagram below, with 4 x 4 sensei array on a PCB is summarised visually in Figure 2. The final tactile pressure sensor is shown with a rGO film 1 and PCB 2.

4

Pouring Ecotlex 6Q-3<> onto rGO fihu sad curing

A second preferred method of producing the sensor of the present invention may also be used. This will now be described. In this method the GO material is first burnt with ethanol flame to convert into rGO powder. The rGO powder is then dissolved into DI water. The PU foam is then soaked into the rGO/DI water solution and then let to dry and repeated several times as required to increase the amount of rGO soaked by the PU foam/film. When dried, the foam/film's underside is coated with silver paste and placed on top of the PCB and then elastomer such as Ecoflex 00-30 is poured about the film and PCB. The elastomer is then cured in the oven at 50C for 4 hours. The fabrication process of the second preferred method is shown in the flow diagram below.

Essentially it produces a sensorthat appears the same as that in the first method, see Figure 2a, where the rGO film is denoted 1 ’ and the PCB 2’.

Preferably a two-layer PCB with electrodes is then added to the rGO film (made by either of the methods above) to make an array of 4 x 4 sensei. In other embodiments, such as n x 1 or 1 x n, a single layer PCB can be used but if the configuration is n x m or m x n, a double layer PCB is required. A schematic of the film and PCB is shown in Figure 1 . Each sensei can be referenced to with its <column><row> such as sensei A1 , A2, etc. Preferably, the rGO film is attached to the PCB using a conductive silver paste, however, other appropriate fixing methods or materials may be used, such as other conductive glues. Finally, Ecoflex (00-30) was poured onto the rGO film and PCB and then cured in the oven at 50°C for 4 hours. The Ecoflex will act as an elastomeric material holding the rGO film together, making it flexible and stretchable. The fabrication process of a tactile pressure sensor with 4 x 4 sensei array on a PCB is summarised in Figure 2.

While Ecoflex has been described above, any elastomeric material could be used to encapsulate the sensor. The hardness of the elastomer chosen contributes to the sensing pressure range of the sensor. The higher the hardness the higher is the sensing pressure range. The described use of elastomer with Shore hardness of 30 using Ecoflex 00-30 allows the pressure range of up to 30 kPa.

The PCB used in the sensor of the present invention may be a standard rigid PCB or may be a flexible PCB. Alternatively, the PCB may be made from other materials, such as carbon black and elastomer - resulting in a soft, flexible and stretchable PCB.

3. Characterisation

The GO, PU foam or sponge, and rGO foam thermal characteristics were analysed using thermogravimetric analysis (TGA) (Q5000, TA instrument, USA). Fourier transform infrared (FTIR) spectroscopy (Nicolet FTIR iS 50) was used to evaluate the chemical characteristics of the materials. The surface structure of rGO foams was observed using scanning electron microscopy (SEM) (HITACHI SU-70). The completed sensor's performance was characterised using a self-made apparatus with an LCR meter (Agilent 4263B) to record the resistance. 4. Results and Discussion

Thermogravimetric analysis (TGA) results of GO, PU foam, and PU foam coated with GO are shown in Figure 3a.

The curve for PU foam shows that thermal decomposition of pristine PU foam occurs in two main stages: in the ranges 200-320 °C and 320-440 °C. At the temperature range 200-320°C, PU foam was decomposed through the cleaving of urethane bonds, followed by degradation of the remaining polyol chains above 320°C. it also indicates that at a temperature higher than 440°C, PU is mostly decomposed.

The TGA curve of GO shows that it starts to lose mass below 200°C as a result of the vaporization of absorbed water and accelerates at 200°C due to the removal of the oxygen-containing functional groups.

According to the results of TGA illustrated in Figure 3a, the temperature of the ethanol flame (600°C) is sufficient to burn-off the PU foam completely and reduce the GO, leaving behind the 3D conductive rGO film.

Fourier transform infrared (FTIR) spectra of the PU foam, GO, and rGO film are shown in Figure 3b. FTIR spectrum of the PU foam displays its characteristic peaks, which is consistent with the previous studies.

The FTIR of GO shows the following characteristic peaks: the stretching vibration of O-H groups at 3338 cm- ¹ , C=O at 1727 cm- ¹ , C=C at 1620 cm- ¹ , O-H deformation from C-OH at 1392 cm- ¹ , C-O from the C-OH at 1223 cm ^_1 and C-O from the C-O-C at 1047 cnr ¹ . This spectrum indicates that GO has hydroxyl and carboxyl groups. The spectrums of rGO film shows that the characteristic peaks of PU foam are absent, implying the pyrolysis of the PU template during the ethanol flame exposure. Moreover, the peaks at 3338, 1727, 1392, and 1223 cm ^_1 for GO were not presented in rGO film, which confirmed the reduction of GO in the ethanol flame process. A peak at 1606 cnr ¹ is observed for the rGO film, and relates to the stretching vibration of C=C. The scanning electron microscopy (SEM) images of PU foam (Figure 4a-b) shows a porous 3D scaffold structure with pore sizes ranging from 300-800 pm with a very smooth surface and no impurities observed. The PU foam will act as a template scaffold for the GO. After dip coating of PU foam into DI solution with GO, the foam has many wrinkles and folds, which changes the surface of the foam from smooth to rough, and the wrinkles implied that the GO had been successfully absorbed around the PU scaffold (Figure 4c-d). After the ethanol flame process in which the PU foam as a template scaffold was burnt-off and the GO is reduced (Figure 4e-f), the structure of rGO film follows that of pure PU foam, as the scaffold or template is scarified during the ethanol treatment step, which also converts the GO into rGO. 5. Taguchi Design of Experiment

Taguchi design of experiment was used as parameter design for the fabrication of the rGO film. A L9 Taguchi table with three parameters and three levels was designed, and nine sets of experiments were performed. For each set, five samples were tested. Table.1 shows the parameters and their levels for the Taguchi method, and Table.2 shows the L9 Taguchi trials. After preparing all samples, their resistance were measured using a LCR meter. The aim is to achieve the lowest rGO resistance, using the signal to noise ratio for each test as shown below in Equation 1 :

^ = -101og ) (D where y is the resistance of each sample and n is the number of samples which is 5 in these experiments.

Table 1. Parameters and their levels used in the Taguchi method

Table 2. Taguchi method L9 trials Table 3, shows the S/N ratios for each trial. The average amount of S/N ratio for each level was calculated, and the graphs plotted (Figure 5) to show the parameters to achieve the minimum resistance.

Table 3. S/N ratios of the trials

The results from the Taguchi design of experiment, that is, the parameters to obtain the minimum resistance are: 8mg/ml concentration of GO in DI water, 3 minutes Ethanol flame and 1 ml of Ecoflex (00-30).

The Pareto ANOVA was also conducted to study the contribution share of each parameter and the results are summarized in Table 4. From Figure 6, it could be concluded that the concentration of GO in DI water has the most significant contribution (72.24 %) in getting the minimum resistance as the rGO film is the conductive network that depends on the amount of GO being reduced.

Table 4. Results of pareto ANOVA

6. Preloading to Precondition the sensor

Experiments have determined that preloading the fabricated sensor is preferred to ensure that the sensor has a repeatable characteristic. Some weak bonds in the rGO network need to be broken, and the preloading of all fabricated sensors at 30 kPa pressure for 30 minutes to ensure that all weak bonds are broken before use. However, the preloading pressure depends on the thickness of the senor. In this particular example, the sensor was 3mm requiring a 30 kPA pressure. After preloading, the initial zero load resistance will change as all weak bonds are broken, and a steady-state is achieved. Figure 7 illustrates the 3D rGO scaffold structure encapsulated by Ecoflex (00-30) after preloading. The red circles show the weak bonds broken after preloading.

7. Electromechanical Performance of The Tactile Sensor

After applying the preloading conditions, an electromechanical characterization of the sensor was conducted with a test apparatus shown in Figure 8a. A 3D printer fitted with a plunger attached to a load cell applies a compressive force on sensei A1 of the sensor (Figure 8b). The resistance data was recorded with a LCR meter. As shown in the schematic illustration of the sensor (Figure 8b), the PCB is attached at the bottom side of the rGO film and the resistance between the electrodes A and 1 that forms sensei A1 will change when compressed due to the distortion of the rGO scaffold structure (film) above it.

As illustrated in Figure 9, due to the low Young's modulus of the Ecoflex elastomer encapsulation, the side edges of foam extend sideways as it is being compressed, and rGO pores elongate sideways, resulting in longer conductive paths between the two electrodes of the sensei. Therefore, the electrical resistance for that particular sensei increases. With a further increase in pressure, the foam is further squeezed sideways, increasing the conductive path and increasing the resistance. After releasing the pressure, the deformed bonds recover, and resistance returns to its initial zero load value. The deformed bonds can quickly recover after the load is removed due to the elasticity of the Ecoflex, which is repeatable. Figure 10 shows electromechanical results for the sample with 1 mg /ml GO concentration in DI water. As mentioned earlier, when the sensor is loaded with different pressures (0-30 kPa), resistance increases with an increase in load (Figure 10(a)). The relative resistance is defined as: relative resistance = AR/ R _o AR = R — R _o (2) where R0 is the sensor's resistance before loading and R represents the sensor's instantaneous resistance. Some noises in the resistance plot can be filtered with the appropriate filtered setting in the LCR meter. Furthermore, the sensor shows an overshoot when loaded or change in loading, and then the resistance decays to a stable value overtime. These abrupt changes could be due to the acceleration and viscoelastic nature of the elastomer. In general, repeatability under cyclic deformation is a fundamental requirement for wearable pressure sensors. Thus, loading and unloading at 30 kPa were repeated for 10 cycles with a 5 mm/min compression rate to observe its repeatability, and the results are shown in Figure 10(b). After 10 cycles, the change in the zero-load resistance is insignificant and demonstrates good stability and robustness for daily life applications. The sensitivity of a sensor is another critical parameter. The sensitivity of this rGO-based piezo-resistive sensor can be calculated with: where the R relates to the resistance under the pressure P and the RO refers to the initial zero load resistance of the piezo-resistive sensor. Figure 10(c) shows the sensor's relative resistance at different pressure. The slope of the graph indicates the sensitivity of the sensor. The plot shows two linear regions with the sensitivity of 0.13 kPa-1 at the range of < 10 kPa and 0.05 kPa-1 for > 10 kPa. The sensor's hysteresis performance is shown in Fig. 8(d) with loading and loading cycle of pressure up to 30 kPa at a 5 mm/min compression rate. The small hysteresis could be due to the nature of the viscoelastic and creep characteristics of the Ecoflex. However, the cyclic test of hysteresis indicates good repeatability of the sensor with relatively slight plastic deformation. Figure 11 shows the relative resistance of the sensor at different pressures for GO concentrations in DI water (1 and 4 mg/ml). Here, increasing the amount of GO does not necessarily increase the sensitivity, and hence 1 mg/ml of GO in DI water was used for the rest of the remaining experiments.

8. Tactile sensor setup

Electromechanical characterisation allows the relationship of the resistance and pressure to be established. With this relationship the sensor of the present invention can be used to determine the pressure in each sensei and hence the creation of a pressure maps realising the tactile feature of this sensor. Figure 12 shows the block diagram of the proposed electronics that could interface the sensor to Bluetooth, as an example. This is merely one possible way to set up the electronics and alternative configurations could be used.

When objects of different shapes and weights were placed on top of the tactile pressure sensor, the resistance of each sensei will change according to the pressure. As the pressure on the sensor depends on the deformation of the rGO film on top of each sensei, a pressure map that corresponds to the shape of an object placed on top of it can be produced. Figure 13 shows the pressure map of various shapes and objects. The pressure map allows the object to be recognised. Preliminary results shows that a force of as low as 0.08 gf on a sensei can be detected.

9. Comparison of the two methods of fabricating the sensors

A comparison of each sensor, made by the first (Process A) and second (Process B) preferred methods, was made, particularly to look at each sensor's performance, but electromechanical tests were also performed. There are weak bonds in the network of the rGO films, so each of the sensors were preloaded before further testing to break all weak bonds and improve repeatability.

The results for both methods are shown in Figures 14 and 15. In Figure 14, 10 cycles of relative resistance change of the sensor subjected to a cyclic pressure of 30kPa is shown. As can be seen in Figure 14 both sensors show an increase in resistance with increasing load. Due to the flexibility of Ecoflex 00-30, the sensor extends from the sides with applied compressive loads, resulting in a longer conductive path between the bottom electrodes, thus increasing the electrical resistance. When the load is removed, a sensor recovers to its initial shape, and its resistance returns to the initial value. Figure 15 shows the sensitivity (S) values of each of the sensors in the 0-30 kPa pressure range. The sensitivity was calculated by dividing the relative resistance change of the sensor by pressure. For process A the sensitivity is much higher than that for process B. It shows a remarkable sensitivity value of 0.13 kPa-1 in the pressure range <10 kPa and 0.05 kPa-1 in the range of >10 kPa for process A. For process B, the sensitivity is 0.017 kPa-1 in the pressure range <10 kPa and 0.005 kPa-1 in the range of >10 kPa. One reason could be that the dispersion of rGO in DI water is more limited than that of GO in DI due to its functional groups, making GO more hydrophilic. Therefore, a better and more uniform conductive network could be created when PU is soaked in GO suspension than the rGO suspension.

Three samples were tested, and the spread of relative resistance changes of those samples at different pressures were compared together in Figures 16a,b. The comparison of the two methods reveals that the spread of data prepared with process B is considerably lower than that in process A. It may be that PU is not removed in process B compared to process A and is still a part of the sensor. The PU foam improves the robustness of the conductive rGO network and improves manufacturing repeatability. In summary, both techniques can be used to manufacture flexible-piezo-resistive pressure sensors, one with higher sensitivity (process A) while the other with a more reliable manufacturing process (process B). In conclusion, the piezo-resistive pressure sensors prepared with two unique and facile manufacturing methods have high potential as flexible sensors.

10. Conclusions

The present invention is a sensor that acts as a pressure senor array, which has been developed to be flexible. The pressure sensor array (4 x 4 senseis) consists of an rGO foam, and flexible PCB, encapsulated in an elastomer. Ecoflex 00-30 was used as the elastomer to encapsulate the rGO foam to provide mechanical flexibility and ease of use. TGA and FTIR were used to evaluate the chemical characteristics of the materials, and SEM was used to observe the surface structure of samples. Electromechanical tests were conducted to characterise the sensors. An L9 table Taguchi method was used to analyse the effect of different parameters on the sensor’s resistance, and the most significant factor was identified by Pareto ANOVA analysis. The sensor with 1 mg/ml GO concentration in DI water and Ecoflex 00-30 shows notable sensitivity (0.13 kPa- ¹ for pressure <10 kPa) and a wide working range (0-30 kPa), long-term stability and low hysteresis. Different objects were placed on top of the sensor array, and by plotting the pressure map with the resistance change of each sensei, the sensor could be used for object recognition.

It is believed that the highly sensitive pressure sensor arrays of the present invention with the simple structure and facile construction method are promising candidates in making cost-effective artificial intelligence, prosthetics and human-machine interface machine.

Throughout the description like reference numerals will be used to refer to like features in different embodiments. Unless the context clearly requires otherwise, throughout the description, the words “comprise”, “comprising”, and the like, are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense, that is to say, in the sense of “including, but not limited to”. Although this invention has been described by way of example and with reference to possible embodiments thereof, it is to be understood that modifications or improvements may be made thereto without departing from the scope of the invention. The invention may also be said broadly to consist in the parts, elements and features referred to or indicated in the specification of the application, individually or collectively, in any or all combinations of two or more of said parts, elements or features. Furthermore, where reference has been made to specific components or integers of the invention having known equivalents, then such equivalents are herein incorporated as if individually set forth.

Any discussion of the prior art throughout the specification should in no way be considered as an admission that such prior art is widely known or forms part of common general knowledge in the field.