Skip to main content

Graphene-based Electrodes for Application in Supercapacitors

Final Report Summary - ELECTROGRAPH (Graphene-based Electrodes for Application in Supercapacitors)

Executive Summary:
The ElectroGraph project was an effort made to promote efficient and smart energy storage, aiming to provide a positive impact no least in the field of electro-automotive. The investigations in this project revolve around three main points, firstly being the production of high energy and power density supercapacitors, secondly to propose material and components’ production methods that are safe for both the individuals and the environment. Lastly the project also aims to demonstrate to the public with the use of renewable and recyclable the feasibility of sustainable energy management.
In the first point, the researches that took place revolved mostly around the synthesis of the electrode material of choice, graphene, and the processing methods to integrate them into electrodes. The materials were studied both as nano-scaled materials and also as macro components they were fabricated into. Ionic liquids were also looked into during the same phase of the project to provide the supercapacitor with environmentally benign electrolyte that enables its operation at voltages above 2 V.
From electrode material synthesis aspects, the findings within this project suggested that reduced graphene oxide (rGO) and its synthesis route was the most suitable for this application. This outcome was concluded not only with consideration to the quality of the material produced but also its production volume. The various rGOs that have been produced over the run of this project have qualities such as specific surface areas comparable to commercially available ones and also the potential to be processed into supercapacitors with specific capacitances upwards of 100 Fg-1. In view of the graphene based electrodes fabricated, they have demonstrated improved adhesion and specific energies higher than commercial ones composed of activated carbon, however at the expense of its reduced power density due to the high resistance with the current design.
In planning for the end of life of the supercapacitors, recycling routes have been developed to recover and reuse valuable materials such as the metallic collectors, the electrolytes and the electrode. For the latter, positive results have been shown in experiments that the recycled graphene can be applied as functional fillers in polymers for mechanical strength enhancement and bestow the polymers with electrical conductivity.
Through the series of tests conducted over the course of the project, the potential of graphene in replacing activated carbon has been shown through its higher energy storage capability at lower weight. The electrode material synthesis, the electrode’s construct and the electrolyte’s performance will however have to be improved in order to bring forth more motivation for the change. As it stands, life cycle analyses have showed that a scaled up graphene supercapacitor production’s impact is on par with the current technology using activated carbon.
To improve these aspects, the electrochemical exfoliation of graphite can be further investigated as a potential low energy consuming route for the production of graphene. As for the electrodes, an alternative binder matrix would be required to increase the ionic diffusion rate during charging and discharging of the supercapacitor. In these regards, an electrolyte with lower viscosity would also aid in lowering the overall resistance of the supercapacitor.

Project Context and Objectives:
ElectroGraph set off with the goal of producing a device that possesses high energy storage capability and low in weight, which simultaneously charges and discharges within a short period of time, i.e. a device with high energy and power density respectively. Looking at a Ragone plot, capacitors on the far left of the chart are able to be charged and discharged at high rates, while devices like batteries possess higher energy density, however storing/releasing the energy at a lower rate. In the same chart, supercapacitors are illustrated to be sandwiched between capacitors and batteries both in terms of their energy and power to weight performances. Supercapacitors are in fact capacitors that offer manifolds higher energy storage capabilities; this is possible owing to the increased surface area available in the device for charge storage purposes and high dielectric constants of the electrolyte. While the supercapacitors operate under the same principles as a capacitor, this modus operandi also enables the device to charge and discharge faster than electrochemical batteries.

One of the main aims in this project was to employ graphene as the electrode material in the supercapacitor for charge storage purposes. Graphene was considered a novel material which is able to provide high amounts of charge storage surface at low weights. This material consists of a monolayer of carbon and it provides a specific surface area exceeding 2600 m2g-1 as compared to values of up to 1800 m2g-1 obtainable from commonly used activated carbon.
Besides increasing the amount of surface for charge storage, the energy density of the supercapacitor can also be elevated by increasing the device’s charging voltage. For this purpose, ionic liquid electrolytes were also synthesised to allow the supercapacitor to operate at higher voltages without introducing irreversible and destructive redox reactions.
To carry out these tasks, a consortium comprising of specialist from universities, research institutes and small to medium enterprises across the continent was assembled with the specific tasks of realising the goals. The ElectroGraph project follows an integrated, technology driven approach in development of novel materials and components for realization of optimized supercapacitors. The scientific and technical work of ElectroGraph project was divided into three main phases; the first was to synthesize a range of graphene materials and to engineer graphene via multitude of post-treatment processing methods with an objective to develop material with specific and controllable properties.
Along with that suitable electrolytes in the form of ionic liquids were also to be researched and synthesized to compliment the supercapacitor system in order to bring about the optimal energy and power density of the device.
The second phase was to investigate graphene and graphene containing materials and systems with concern of the health and environmental aspects. The third and final phase focused on the development of materials, components and devices for demonstration to industry and exploitation and was targeting the demonstration of the supercapacitor prototype device utilizing graphene based electrodes.
Phase 1: Graphene synthesis and processing
Phase 2: Environmental aspects
Phase 3: Supercapacitor demonstration
Overall the objective of the project was to create the knowledge and be first to market with the cost effective mass production of high quality graphene materials, reply to demands of industry for low cost graphene with controlled properties and related products, and in turn increase high tech exports. ElectroGraph targets the demonstration to the industry the enhanced performance and cost benefits of graphene in its applications and provides the European industry with a practical understanding of the capabilities of nano-sized materials.

The scientific and technical objectives have been specified and are presented in the following parts of the summary report.

Project Results:
The very first tasks before the scientific and technical investigations began were to define the targets achievable in this project. In that, goals were set to develop supercapacitors that could be installed into Delocalized Power Supply Units (DPSU) for an automobile’s rear view mirror which would be charged using renewable energy sources such as solar PV panels. For demonstration purposes, this concept was set to be applied to actuate the external rear view mirror of a car.
The proposed application was also intended to demonstrate an innovative way of generating electrical power in-vehicle, alternative to existing solutions (alternator and batteries) and open new technological opportunities in terms of management of electrical and architectural implementation:
• New management, control and power architectures on the vehicle: the implementation of a system of localized and autonomous power supply allows the realization of a fully distributed wireless architecture, thus eliminating electrical cables for data transmission through the RF communication, and cables for electrical power distribution systems through local generation of energy. Today, on a car there are about four kilometres of electric cables, with a significant impact on the final costs of the product in terms of components and assembly.

• Decoupling of engine operation with respect to electric utilities: the availability of localized and distributed power sources allows separating the generation of energy from the engine. This approach makes the optimization of the control strategies and the management operation of the engine possible, which reduces fuel consumption and pollutant emissions.

The rear-view mirror’s movement was to be controlled by two standard electronic actuators. The battery use for such application is limited due to the high speed movement that produces really short energy discharge time (5 s). The application of a supercapacitor, that offers much higher power density, seemed more reliable than batteries that have relatively slow charge and discharge times.
After the selection of the specific application, studies were performed to identify the following performance requirements of the supercapacitor module:
1. Power output: > 4 W
2. Energy density: >1.5 Wh/kg
3. Power density: > 0.3 kW/kg
4. Stored energy: >70 J (>0.02 Wh)
5. Stored charge: > 5 C
6. Mean-voltage of operation: indicatively 4 V per module
7. 80% of initial capacity after 106 cycles at 100% DoD
8. Temperature range: -20°C to 80°C
9. Weight: 5 to 10 g
10. Ease of use and integration in automotive components
11. Integration of materials and technologies with low impact into existing manufacturing automotive processes.
12. Modularity
13. Possibility of high volumes production
14. Wireless approach for recharge (PV cells)

Furthermore, the specifications of electrolyte that can be used for prototype product based on supercapacitor for automotive applications taken into consideration in further synthesis, processing and testing were:
• high ionic conductivity (at least 5 mS/cm @ -20°C and 15 @ 40°C)
• high dielectric constant (at least 30)
• wide electrochemical window (at least up to 4.5 V)
• operative temperature range between -40 and +85 °C
• boiling point higher than 130°C
• flash point > 65°C
• auto-ignition temperature > 250°C
• low corrosive properties in respect to metallic and polymeric materials

The pouch-type supercapacitor prototype along with its electrode was also designed before further investigations began. The supercapacitor was to consist of several layers of electrodes and separators, in which, the number of layers will depend on the thickness of the electrode, I.e. graphene coating thickness on the aluminium collector.


2.2 Electrode material synthesis

The next series of investigations carried out were with concerns to the electrode material graphene. Three graphene synthesis methods producing materials of different grades were studied for their application as the electrode material. Two of which were relatively mature methods being i) chemical vapour deposition (CVD), ii) chemical oxidation route (rGO) and finally a comparatively newer method iii) Electrochemical exfoliation of graphite.
Graphene synthesis through CVD
Several processes were investigated, with the CVD method. Firstly, catalysed CVD growths of graphene using metal substrate such as nickel (Ni) and copper (Cu) were looked into. The results indicated that while Ni often yielded multiple layers, growth on Cu yielded higher amounts of single layer graphene.
Further on, a variety of different Cu foils were tested as graphene growth substrates. From the results, it was determined that the quality of foil used was of vital importance. Rough foils produced by cold rolling yielded defective and non-uniform films whereas smoother films produced by electro-deposition yielded uniform films consisting primarily of monolayer domains.
With the aim of increasing active surfaces available for electrochemical processes in supercapacitor applications, atmospheric DC plasma discharge CVD method was developed and implemented to produce a structure known as “carbon nanowalls” (CNWs). The morphology of graphene can be modified through varying a number of parameters. In changing carbohydrate precursor a variation in density of CNWs was observed where graphene tilts vertically from the surface of the substrate. With ethanol, sparse CNWs grow on the surface was observed. Its Raman spectrum showed a prominent D-mode that was related to the edges of the walls in direction of the Raman spectrometer’s laser beam of. SEM image demonstrated that dense CNWs were formed with hexan as the precursor. Even though the image does not clearly present the vertically oriented wall-like morphology, its Raman spectra proved the layer contained compact CNWs. However, a flat graphitic layer covering the substrate resulted from the regions beyond the vicinity of the DC plasma. Its Raman spectra reflected a relatively well ordered graphite structure with a small D-line intensity relative to the G-line.
Synthesis of reduced graphene oxide (rGO)
The second route involved the oxidation of graphite with strong oxidizing agents to produce graphite/graphene oxide (GO) before they were reduced again for the recovery of the material’s electrical conductivity.
As described in the following, 5 variants of chemical oxidising methods was looked into within the frame of this project for the production of GO.
1. Preparation of GO using modified (Fugetsu) Hummers method (GO1)
Graphite (2g), sodium nitrate (1g), concentrated sulphuric acid (50ml) and potassium permanganate (KMnO4) (6g) were consistently mixed in ice bath for 2 hours until the mixture gradually becoming pasty and black-greenish. Next, the mixture was placed in a 35°C water bath for 30 minutes, followed by slow addition of distilled water (100ml) to keep the mixture from effervescing. The temperature of the mixture was controlled at well below 100°C for 3 hours until the colour turned a little yellowish. Further treatment was performed with H2O2 (5%, 100ml) to reduce the residual KMnO4, the filtered cake was then washed with distilled water and dried at 75°C.
2. Preparation of GO using modified (Bangal) Hummers method (GO2)
Graphite powder (2g) was added to a 250 ml baker containing sodium nitrate (1g). Sulphuric acid (46ml) was added subsequently under stirring in ice bath. Then KMnO4 (6g) was thereafter added slowly to the system and the temperature was controlled to under 20°C. After 5 minutes the ice bath was removed, and mixture was heated at 35°C for 30 minutes. Next 92 ml of water was added drop wise and stirred for 15 minutes. Then 80 ml of hot water and 3% H2O2 aqueous solution were added to reduce the residual KMnO4 until the bubbling Stopped. The suspension was filtered off through nylon filter and washed with ample distilled-water until the pH of the filtrate approached 7. Finally the drying process was carried out at 75°C.
3. Preparation of GO using modified (Jeong) Hummers method (GO3)
Graphite (2g) was mixed with sulphuric acid (350ml) at 0-5°C in three-necked 500ml round bottom flask equipped with thermometer for 15 minutes. KMnO4 (8g) and sodium nitrate (1g) were added portion wise at 0°C and stirred for 30 minutes at 0°C, followed by 30 minutes at 35°C. Water (250ml) was added via dropping funnel and reaction mixture was heated to 98°C for 3 hours. Reaction was terminated by adding of 500 ml of water deionized and 40ml of 30% H2O2. Mixture was filtered off through nylon filter, washed with diluted HCl (10%) to remove metal ions before washing with distilled water until pH of filtrate approached 7, and dried at 75°C.
4. Preparation of GO using Staudenmaiers method (GO4)
Graphite (1g) was added portion wise into a three-necked round bottom flask equipped with thermometer and containing a mixture of sulphuric acid (17,5ml) and fuming nitric acid (9ml) at 0-5°C within 15 minutes. The mixture was further stirred for 30 minutes at 0°C. Potassium chloride oxide (11g) was added at 0-5°C portion wise followed by further stirring for 30 minutes at 0°C. During the process, the flask was capped loosely to allow the escape of ClO2 gas and left to stir vigorously 96 hours at room temperature. Reaction mixture was poured into deionized water (1L) and filtered off. Further washing was carried out firstly with diluted HCl (5%) to remove sulphate ions and then with water until pH of filtrate approached 7, and dried at 75°C.
5. Preparation of GO using Brodie method (GO5)
Fuming nitric acid (40ml) and was mixed with graphite (2g) at 0°C in 250 ml round bottom flask for 15 minute. Sodium chloride oxide (17g) was added portion wise at 0°C and the mixture was stirred for 24 hours. Reaction mixture was poured to deionized water (350ml), filtered through nylon filter, washed with water until pH of filtrate is about 7, and dried at 75°C.
From the following SEM images proved that the different oxidising condition resulted in GO of different morphologies. Therein, GO3 was determined to have higher interlayer distances where a homogenous structure was seen and the graphene layers can no longer ne distinguished. Whereas in GO4 and GO5 particles with layered structure of partly exfoliated (intercalated) graphite was observed due to its lower interlayer distance.
This statement is supported by the X-ray diffraction (XRD) measurements. There the measurement results of GO1 – GO5 are being compared to graphite’s. In all cases of the oxidation methods GO1 – GO5, the typical diffraction peak for 002 crystal orientation of graphite at ?????????completely disappeared and instead new sharp peaks at lower angles from 12o to 15o appear, corresponding to the lattice distance from 5.9 Å to 7.6 Å. This indicated that, after oxidation of graphite, the inter-layer distance almost doubled. The relatively sharp peaks indicate that new interlayer distance was homogenously present in the structure. The most dramatic increase in the lattice distance was achieved with the method GO3 (d=7.6Å). This difference in interlayer spacing was also reflected in the material’s bulk electrical conductivity; GO3, with its highest interlayer spacing, possessed the lowest electrical conductivity, while GO5, the most compact amongst, had the highest.
For the recovery of electrical conductivity, the obtained GO from different sources were reduced chemically (N2H4, NaBH4) and thermally (thermal annealing) in the processes described below to produce reduced graphene oxide (rGO).
Preparation of reduced GO (rGO) using hydrazine (N2H4)
GO (1g) was placed into a beaker with deionized water (150ml) added and vigorously stirred for 24 hours at room temperature. The suspension was then sonicated in ultrasonic bath cleaner for 3 hours, followed by sonication finger for 30 minutes and finally for 1 hour in bath sonicator. The mixture was placed in a 250ml round bottom flask with ammonia (1.5ml) and hydrazine monohydrate (3ml) added and stirred vigorously at 85°C for 24 hours under reflux condenser. After cooling, suspension was filtered off through nylon filter, washed with deionized water (500ml) and with methanol (50ml). The cake was dried at 75°C in oven for 24 hours.
Preparation of reduced GO (rGO) using sodium borohydride (NaBH4)
GO (1g) was placed into a beaker, deionized water (150ml) was added and vigorously stirred for 24 hours at room temperature. Suspension was then sonicated in ultrasonic bath cleaner for 3 hours, followed by sonication finger for 30 minutes and finally for 1 hour in bath sonicator. The mixture was placed to 250ml round bottom flask with NaBH4 (2g) added and stirred vigorously at room temperature for at least 5 hours. Suspension was filtered off through nylon filter, washed with deionized water (500ml) and with methanol (50ml). The cake was dried at 75°C in oven for 24 hours.
Preparation of reduced graphene oxide (rGO) through thermal annealing
GO powders have been annealed in a quartz tube at ambient pressure and a steady flow of 25 sccm of argon. The temperature increased from the room temperature to the set temperature in several steps. First, the temperature was elevated to 140°C at a rate of 1.5 °C/min followed by 30 min of dwelling. After that, the temperature increased at 0.3°C/min to 350°C and maintained for 30 min. The last step was heating the powder up to the set temperature, 700°C or 1000°C, again at a rate of 1.5°C/min. The powders were kept at the final temperature for at least one hour.
The efficiencies of the three reduction methods were determined by measuring the electrical conductivities of the various materials after being processed. Through the results, the thermal reduction method appeared to be the most effective. The electrical conductivities of the GOs reduced with NaBH4 and N2H4 respective are shown, their values lie within the same order of magnitude as they were before treatment. However through the thermal treatment, the electrical conductivities of the rGOs increased at least by one order of magnitude. The conductivities of two other rGO3 variants, rGO3M200 and rGO3M100, produced with raw graphite of larger grain sizes were also determined. Their larger grain sizes proved to inhibit the full oxidation during the GO production process, thus giving them higher conductivities than rGO3.

Plasma reduction and doping of GO

As an alternative, a process has been also developed for the direct reduction and functionalization of GO with a downstream plasma source. This plasma can be described as “chemical plasma” and functionalises graphene surfaces in a non-destructive manner.

With this treatment which was carried out with NH4 plasma, the graphene lattice could be simultaneously reduced and N-doped. Such doping is of great significance for supercapacitors and other energy applications as it is widely accepted that doping heteroatoms into the graphitic lattice can tailor its chemical and physical properties. N-doped carbon nanostructures show n-type or metallic behaviour and are expected to display greater electron mobility whilst introducing chemically active sites.
The characterisation results in show that oxygen moieties were reduced and removed from the material while nitrogen is incorporated into the material. The XPS analysis supports this showing a reduction in oxygen content and the addition of nitrogen following plasma treatment.

Electrochemical exfoliation of graphite
The final graphene production method studied involved applying an electrical potential across two graphitic electrodes to compel the electrolyte’s ions into intercalating the electrode’s graphitic structure. The aim of this process was to increase the distance between the graphene layers in the graphite electrode, thereby reducing the Van der Waals forces holding them together and allow them to be exfoliated.
The basic setup required for this reaction is an electrochemical cell which consist of a graphite working electrode, a counter electrode, a suitable electrolyte and a power supply to supply the necessary potential for the reaction to occur.
In separate processes, the electrochemical exfoliation was executed both the anode and the cathode. For the anodic exfoliation, a constant voltage was applied across the electrodes to allow the oxidative intercalation reactions to occur. Several electrolytes of different classes were experimented for the anodic exfoliation and they are:
i) Ionic liquids: BMIMBF4, OMIMBF4, EMIMBF4, EMIMPF6, BMIMPF6, and DMIM-I
ii) Polyelectrolytes: Poly(styrene sulphonate) sodium salt and Poly(acrylic acid) sodium salt, and
iii) Bases: Potassium hydroxide (KOH) and sodium hydroxide (NaOH).

As for the cathodic exfoliation, it was performed with a different principle without oxidising the graphite electrode. Rather, the exfoliation was induced by increasing the local pressure between the graphene layers through cycles of intercalation and de-intercalations of the electrolyte’s ions on the graphite electrode. The cathodic intercalation with ethyl-methylimidazolium bis (trifluoromethylsulfonyl)-imide (EMI-TFSI) as the electrolyte began at a potential of about -1.9 V for the first scan. During the reverse scan, an anodic current indicates the de-intercalation process. For the successive scans, it was observed that an increase (anodic shift) of the threshold potential (or critical potential) for the cathodic intercalation and in the same time an increase of the de-intercalation current. These results indicate that the graphite electrode expansion increased the electrochemically active area due to the intercalation process.
The following is an SEM image of the materials exfoliated from the anode with NaOH as electrolyte and 9 V applied across. The materials produced under these conditions were mostly thick stacks instead of thin layers synthesised through the previous two methods. To the right of the SEM image, the exfoliated material’s corresponding Raman spectrum is seen to have a high D-band, representative of the presence of sp3 bonds from the oxidative process. The broad and modulated 2D band is reflective of the turbostratic structure, i.e. the layers/stacks were arranged on top of each other in a disordered fashion

2.3 Characterisation of the materials

From the list of materials produced through the different routes, the reduced graphene oxides were deemed as the most suitable for application as supercapacitor electrodes; in making this choice, both the quality of the materials and the production volumes were considered. The highlighted specific surface areas of GO3 and GO5 after their respective reductive treatments presented values on par or even better to those characterized from various similar commercial graphene/graphene nano-platelets. In pertinence, rGOs such as those were also determined to have the potential in providing the supercapacitor with high energy storage capability of more than 100 Fg-1 in specific capacitance.
2.3 Prototype supercapacitor fabrication

The steps following the synthesis of the electrode materials were to assemble them into supercapacitors. This section will consist of 2 parts describing the preparation of the two main components in a supercapacitor: firstly, the electrode fabrication process and secondly, the synthesis of an ionic liquid 1-ethyl-2, 3dimethylimidazolium bis ((trifluoromethyl)sulfonyl) imide (EdMI-TSFI) to be employed as the electrolyte.
It is to be mentioned that the electrode materials used for the experiments in processing of electrodes were obtained commercially. Due to the investigation works for electrode material synthesis and electrode processing occurred concurrently, it was deemed advisable to study the electrode preparation process with an electrode material that is in abundance before the technique was transferred to other graphene materials.

Electrode fabrication
The electrodes fabricated in the initial phases were with accordance to industrial standards used in the production of battery electrodes. Through characterising these initial samples, some of the main challenges were highlighted as: i) improving the electrode material’s adhesion on the collector, ii) improve electrical connectivity of the active layer to increase power performances and iii) create porous structures accessible to the electrolyte during charging and discharging phases.
The filtration technique was thus adopted as for the production of electrodes without the need of commonly used polymer binders (e.g. PTFE). In the place of these binders, carbon nanotubes (CNT) were used to contain the commercial graphene nano-platelets (GNP) and simultaneously providing active surface area for charge storage. The electrode build up consisted of layers of GNP as the primary active material enveloped between the more mechanical stable and conductive CNT membranes.
The processes leading to the fabrication of such an electrode started with the preparation of water based CNT and GNP dispersions. These dispersions were thereafter filtered sequentially with the assistance of pressurised air to form filter cakes such as those shown in. These filter cakes were then cut into shape and assembled together with an aluminium collector between the 2 layers. Finally the cut out electrodes were wrapped between paper towels and placed under the press for 24 hours of drying to prevent deformation/wrinkling before they were further dried in a vacuum oven.

Electrolyte synthesis

For the synthesis of the ionic liquid, EdMI-TFSI, the following procedures were followed. Firstly an EdMIBr solution was mixed with equimolar of LiTFSI at an elevated temperature of 70 °C, over 24 hours before being extracted with dichloromethane (CH2Cl2). Washing process was carried out with water for the removal of Li+ salts. Further purification was done by filtration through activated carbon and drying over MgSO4 followed by further drying in a rotary evaporator to reduce the water contents below 20 ppm.

Performance characterisation of supercapacitor prototypes

While the electrodes an electrolyte functioned adequately as individuals, their combinations as supercapacitor did not performed as they were supposed to. The causes of the malfunction were identified as i) high mass transport resistance experienced by the ions of the electrolyte which was brought about by the CNT membrane layers of the electrode and ii) the high viscosity of the electrolyte. In combination the two factors brought about high equivalent series resistance of the device (>100 ?) and poor wetting of the electrodes.

Two variations of the electrodes measured with 1MTEABF4/acetonitrile as electrolyte. In (a) the electrode was constructed with the active material, GNP, split in 2 layers. The absolute capacitance measured of a 2 electrode setup was about half a farad with cycling rate of 10 mVs-1; this can be translated to each electrode having 65 Fg-1. In (b) the electrodes were optimised by reducing the number of CNT membranes and increasing the amount of GNP. Under similar conditions to (a), the absolute capacitance measured was about one farad with specific capacitance of 61 Fg-1 for each electrode.

However it is worth mentioning that the specific capacitances measured in both cases are about 3 times higher than the 20 Fg-1measured of electrodes from commercial supercapacitors. These results showed that graphene possesses the potential to be made into more efficient supercapacitors with higher charge storage to weight ratio.





2.5 Recycling

In preparation for the end of life (EOL) of these supercapacitors, a recycling route was developed to recover materials of reusable values. These materials included the aluminium used as collector, the electrode material graphene and the electrolyte.

Following the collection and shredding of the EOL supercapacitors a pyrolysis technology was developed in this project for recovery of valuable materials from mixture of shredded materials. The pyrolysis was performed under nitrogen with a temperature first at 350oC to recover electrolyte acetonitrile, thereafter at 600oC to break down the PTFE binder and dielectric materials such as paper that bind the graphene nanoparticles. After the second stage of pyrolysis, the residue, which is mainly graphene on the current collector material, is dispersed in a polymeric material at a high shear rate. The pyrolysis temperatures were determined using thermogravimetric analysis (TGA). The dispersion of graphene in the polymer can be used directly as a moulding compound for fabrication of various polymer composites. The following are two examples.

1. Application of recovered graphene as mechanical reinforcement in polymers

In the first application for recycled graphene from the electrodes a comparatively low weight percentage of the recycled material (= 1 w%)l was added into an epoxy resin to reinforce its mechanical properties. Test coupons were made from these composite resins for the characterisations of their mechanical properties.


The mechanical properties of the recycled graphene reinforced epoxy resin composite have been measured. The results compare the properties of pure epoxy resin to the composites’ with graphene from an electrode with and without pyrolysis treatment. The graphene from pyrolysis treatment had a significant toughening effect on the epoxy resin. The graphene that was not treated by pyrolysis had a deleterious effect on the mechanical properties of the polymer due to the difficulty in dispersing the graphene bonded by PTFE binder. The addition of 0.8% of graphene recovered from an electrode by pyrolysis shows significant increases in tensile modulus, strength and elongation at break.

2. Application of recovered graphene as conductive filler in polymers

In the second application, the recycled graphene content was increased to 5 w% to exploit its conductivities. Two different polymer matrices were tested for compounding with the recycled materials to produce flexible conductive polymers. The polymers selected were:
1. Polypropylene copolymer - (PP1)
2. Polypropylene homopolymer - (PP2)

As a reference, fresh multi-walled carbon nanotubes (MWCNT) were also compounded with the above two polymers and characterised.

The results show that recovered graphene can be compounded with thermoplastics to obtain conductive polymers to be used for the realization of components with electrostatic discharge properties. Although the CNT/polymer composites lead to better performances in terms of lower electrical resistance with the same percentage of filler loading, these recovered graphene materials still possess the potential in their application as functional filler for polymers. However more research would be needed to achieve results comparable to the state of the art.


2.6 Life cycle analysis (LCA)

This section of the report aims at assessing and comparing the environmental impacts of supercapacitor technologies. The comparison was between the state of the art commercially available supercapacitors produced by Maxwell Technologies Inc. using activated carbon for the electrodes and a prototype that uses graphene.
The supercapacitors were placed in a car door mirror to power the motors to adjust it over 20 seconds (4 movements of 5 seconds each). The motors work at 13.5V and need 0.35A therefore the supercapacitors should be able to deliver 80 J of energy.
As this LCA follows the ILCD recommendations, it includes all suggested impact categories adopting a cradle-to-grave approach, therefore from raw material extraction to the recycling activity.
To scale up the process from laboratory scale to mass production process, modelling software was used to increase volumes and minimise inefficiencies inevitably occurring in a lab.
The simulation results showed that if the graphene synthesis process were to be fully scaled up to mass production, the impacts generated by the graphene based supercapacitor overall production would be similar to those generated by the activated carbon based supercapacitors. The global warming potential for the two best technologies measured within the Electrograph project. It is possible to appreciate how, when including the use phase and recycling, the graphene GO2T based supercapacitor (GO2H is chemically oxidised material and chemically reduced, GO2T is chemically oxidised material and thermally reduced) perform better than the activated carbon one.
It is important to say that the LCA, should not take into consideration costs, therefore this study should be interpreted more as a feasibility study than a real case study as assumptions are used when scaling up the process.
2.6 Health & Safety and Risk Assessment

The objective of this work package was to carry out a risk assessment of exposure to particles from the production and processing of graphene. That should include a review of exposure to graphene and associated substrates used in electrodes, a comprehensive list of exposure scenarios in graphene production and processing and exposure assessment for key scenarios identified across production, processing and recycling activities. A basic risk assessment identifying the safe practice and control measures for workers in the exposure scenarios should be considered.
Moreover, the objective will be to use the commensurate life-cycle thinking approach with the target to underpin the exposure assessment activities of the safety work package.
Exposure characterization

Firstly, a conceptual ‘life-cycle’ map of the materials used across the processes in the project has been constructed. Four principal stages have been identified: i) synthesis, ii) manufacture of the product (electrodes), iii) use and service life and iv) waste treatment. For each stage, several activities have been identified. For example, the synthesis stage is divided into three activities: a) chemical vapour deposition (CVD) synthesis, b) electrochemical synthesis and c) synthesis via graphene oxide. Each of these activities is divided into one or more tasks.

Although the use of graphene is recent, interest and applications of graphene are growing fast and attention to potential health and safety concerns about this material and related materials is emerging but publications are sparse and focused on hazard, and not exposure assessment, at this time.

Information on current practices when handling nanomaterials was gathered through an online questionnaire using the SurveyMonkey platform. The questionnaire was structured to provide information at activity and task level on the exposure determinants (physical form of the substance, amounts handled, task duration and frequency, energy applied to the process, exposure controls and use of PPE) and organisational safety measures (safety documents, safety practices and availability of measurements).
A total of 22 responses were received from all partners on the following activities:
? Synthesis: CVD (5 responses)
? Synthesis: Electrochemical deposition (6 responses)
? Functionalization (1response)
? Characterization (6 responses)
? Dispersions & Inks development (1 response)
? Recycling (2 responses)
? Supercapacitor development (1 response)

No information was provided in the following task:

? Synthesis: Exfoliation of graphite oxide
? Other synthesis route
? Electrode & Electrolyte development

Three of the activities (dispersions & inks development, recycling and super-capacitor development) had not started and the respondents did not have information for some of the questions. However the information provided, although it might change during the performance of the activity, was found useful to develop the exposure scenario. The information has been analysed in a qualitative way as the main aim was to build exposure scenarios and inform the monitoring campaign rather than comparing current practices between partners and activities. Exposure risk is a function of the exposure probability, which is mostly influenced by the substance and activity emission potential and the severity of the exposure (amount and time in contact with the substance).

Overall the activity that carries a mayor risk of exposure is functionalization, since graphene is handled as powder. At a task level, unpacking of the material was identified as having a higher risk of exposure as large amounts are handled usually without engineering controls.
Exposure measurement
The purpose of this task was to undertake an assessment of the potential for release(s) and exposure(s) to aerosols containing nanomaterials for the various scenarios in the production, processing and recycling activities involving graphene and graphene-containing electrodes. The activities, their processes and specific tasks across the lifecycle included material handling (e.g. weighing, dispersion), primary production (e.g. electrochemical exfoliation), secondary production (e.g. recovery of graphene-film electrodes), assembly of the supercapacitor and recycling of graphene-film electrodes.
The analysis of the data from the monitoring campaigns led to the development of general conclusions and recommendations for the safe handling of nanomaterials, across the production of supercapacitors and for the synthesis of graphene. These helped to form an overall general view, over and above the individual site-specific conclusions, that is relevant to manufacturers and researchers both within the industries specific to ElectroGraph and beyond. The quantitative and contextual data gathered, observations of the engineering controls and use of personal protective equipment (PPE), particular to individual sites and particular locations and activities within a given site, were used to develop typical case examples and recommendations for the safe handling of nanomaterials in the manufacture of graphene-based supercapacitors. These practical illustrations of risk assessment considerations and recommendations for good working practices to prevent exposure to potentially hazardous materials are included in the Deliverable 7.6 Risk Assessment. This document can be made available upon request from the respective scientist at: Steve.Hakin@iom-world.com
The analysis of the data from the monitoring campaign has led to general conclusions across the production of supercapacitors and for the synthesis of graphene, over and above the individual site-specific conclusions, that are relevant to manufacturers and researchers both within the industries specific to ElectroGraph and beyond. Conclusions are expressed in the context of impact for graphene production, which draws together the main aspects in relation to:
• Particle release from graphene synthesis
• Particle release from graphene dispersions
• Particle release from graphene-film spraying
• Particle release from supercapacitor assembly
• Particle release from recycling
• Particle persistence in the working environment
• Aerodynamic behaviour of platelet materials
• Control measures (engineered and PPE- site specific)

At this time there remains a lack of definitive workplace studies reported in the literature with regards to adverse health effects in workers producing or using these products, and so their occupational health effects is not established clearly. However enough uncertainty remains regarding their hazardous nature, based on toxicology and other general considerations of the mechanisms of inhaled particles, that it is important to limit workers’ exposure, and various guidance documents have been published to help people work safely with these products.
In summary, from the observations made during the ElectroGraph project, it is evident there is a spectrum of potential for release from processes and demonstration of practical steps (and their importance) for controlling exposure to respirable particles such as graphene platelets and carbon nanotubes.
Recommendations have been made based on the measurements and observations made during the exposure assessment activities of the ElectroGraph project with the aim of raising awareness of where releases can occur and of good practice with appropriate control measures proportionate to the scale and nature of the activities involved in the supercapacitor production.
Risk assessment
The risk assessment was developed using data and observations from across all site visits to provide a comparison of the potential for releases, the effectiveness of controls and generically applicable risk management recommendations to be considered. It also incorporates the screening-level hazard assessment being conducted.

A. Screening-level Hazard Assessment

Four GFNs were tested for the ElectroGraph project, a graphene oxide (GO5), reduced GO5 (rGO5), and two graphene nano-platelets (XGnP and MS3 ). Carbon black nanoparticles (CBNP) are well characterised carbonaceous materials which were also included in the study to act as a benchmark.
The potential hazard status of the materials was determined by carrying out physicochemical analysis of the materials, and then linking those characteristics to any potential toxicity outcomes. Given that the most likely exposure route for these materials is via inhalation and that the lung is sensitive to particle exposure (therefore demonstrating a potentially, worst case scenario), these studies were carried out in two well characterised lung cell models. These in vitro cell models represent pulmonary macrophages, which are responsible for the clearance of invasive material into the lung, and the airway epithelium.
Overall, in terms of in vitro toxicity in alveolar cells, the compounds could be ranked as follows:
Reduced graphene oxide > graphene oxide and XGnP > CBNP > MS3
This information was gathered from in vitro experiments and literature review, and further investigation should be performed prior to drawing any final conclusions. In particular in vivo or more advanced in vitro models (e.g. multi-cellular, microfluidics) should be used for further testing, to validate the proposed toxicological profiles of the materials tested, and to confirm the target sites within the body following real-life occupational exposure. In addition, it should be considered that in relation to rGO, the basis of observed toxicity may be the chemical reduction of the GO using hydrazine. Whilst the rGO is washed as part of the reduction process to remove chemical components, to rule out the role of this in toxicity a comparative analysis with thermally reduced rGO (based on the same GO precursor) is recommended yet out with the scope of this project. However, this study does provide detailed information on the toxicity of industrially-relevant GFNs, and how that is modulated according to the physicochemical characteristics of each material. This should enable more accurate development of risk assessment strategies, and facilitate the development of “safe by design” products with lower hazard and/or risk.
B. Control Banding
Risk evaluation used results from the exposure characterisation and measurement tasks, with available hazard data, to determine risk levels, in the context of exposure control measures to minimise the risk to human health.
In summary, the two control banding tools provide very similar estimates of risk, based on different methods and information, for the majority of tasks.
Although the tools attempt to make use of the hazard and exposure information making an adequate conservative estimate of risk when there are so many unknown variables can be difficult. It is likely that given the level of uncertainty for a number of the input parameters that the estimates of risk may be overly protective but in the absence of the information required to make a good estimate of risk this is the most sensible option.
In a series of Case Examples relevant to the research and industrial activities of the manufacture of graphene-containing supercapacitors, the outcomes from the Control Banding assessment are supplemented with a comparison of the exposure controls recommended by the tools with observations and measurement data gathered. This has enabled the development of good practice recommendations with the aim of raising awareness of where releases can occur and of appropriate control measures which are proportionate to the scale and nature of the activities involved in the supercapacitor production.

2.7 Demonstration

In the last phase of the project, a demonstrator consisting of an external rear view mirror powered by a supercapacitor module as its DSPU was assembled. As the name implies, the DSPU was not connected to a central power source, but rather to a photovoltaic cells on the rear side of the mirror to recharge the supercapacitors module. The initial targets set for this demonstrator were to be actuated 4 times per day and 5 seconds of actuation each time. Charging and actuation tests were conducted with the demonstrator to determine that it outperformed the expectations.
It is however to be mentioned that the DSPU consisted of supercapacitors supplied by Maxwell Technologies Inc. instead of the planned graphene/ionic liquid supercapacitors investigated in this project as further improvements were needed before they could be integrated into the system.
The charging rate under direct sunlight was ca. 35% higher and allowed periods of actuation. In comparison to the pre-set targets of 4 actuations and 5 seconds each, the fully charged demonstrator was able to actuate 26 times consecutively for 5 seconds under direct sunlight and 12 times of 5 seconds indoor.
In conclusion, although the demonstrator was powered with commercial supercapacitors instead of those fabricated within the framework of this project, the applicability of supercapacitors in DSPU for storage of energy from renewable energy resources was further proven and promoted through this demonstration.

Potential Impact:
Strategic impact
Positioning of Europe on Supercapacitor/Energy Storage Market
The developement is a supercapacitor for the automotive technology sector. Furthermore, the developed materials and processes within project ElectroGraph can be further used and optimized for supercapacitors in applications requiring high storage capabilities like wind turbines.
An overall situation on the supercapacitor market can be reflected in the global perspective. The market of consumer electronic is thriving in Asia, especially in China, and as a result producers of energy storage devices also benefit from this development. A global Sources report that due to low raw material costs, mature production technology and R&D programmes the market of supercapacitors will grow in China by 50%. These facts mean tremendous changes for this sector in Europe where companies such as Epcos and Montena Components dominated the market in 2002, generating more than 60% of the revenues. The number of large companies involved has not changed much since then. The only chance to keep pace with this global development for Europe is to tailor new technological solutions that can be then implemented into competitive products. These solutions will probably not change the overall market situation in the short run, but Europe can thereby participate in the growth of this market.
Supercapacitor cost and materials depending on capacity, producer and number of items the prices for supercapacitors range from 2.57€ for 10F (30x10mm) to 17.18€ for 100F (35x35mm).
As the pure form of graphene is exceeding all price expectations, within the framework of project ElectroGraph developed a cost effective up-scalable production method for graphene that also ensures high reliability of the graphene performance. There is a common industry trend toward higher purity and consistency in specifications for some specialized and high-tech applications. The predicted increase in manufacturing and sales of hybrid and electric vehicles is expected to increase demand for high purity graphite in fuel-cell and battery applications. One prediction is that the demand for high-quality, high-carbon graphite can increase to more than 100,000 metric tons per year (t/yr) for fuel-cell and battery applications alone and the global demand for graphite used in batteries may increase to more than 25,000 t/yr in the next 4 to 5 years [45]. Graphene as target material together with its application as electrodes in supercapacitors will benefit from those trends as they will ensure high quality and broad availability of base material for production of graphene.
Additionally the developed graphene-based supercapacitor will further trigger development and commercialization of hybrid and electric vehicles.
Impact in automotive sector
Supercapacitors can play an important role for transport vehicles. In effect, electric vehicles may incorporate a battery as energy storage device, the latest being the Li-ion battery or the fuel cells. Supercapacitors can be integrated with a battery in electric cars or with the fuel cell to provide the power peaks required during the operation of the vehicle and improve efficiency and energy economy under variable power driving conditions. In the case of fuel cell, a supercapacitor is essential for the energy recovery during breaking. An important market acceptance feature of the electric vehicle is high acceleration performance, provided by the supercapacitor. By including both a battery and a supercapacitor in the design of the power system of a vehicle, it is possible to decouple the specific energy and the specific power requirements, so that a storage battery may provide the best storage capability for the longest life while the supercapacitor provides the power requirement. For high voltage applications such as electric vehicles, series of capacitors are used to avoid exceeding the working voltage of individual capacitors, what reduces the effective capacity of the assembly. As a result, a supercapacitor pack may be designed for a transport vehicle, including supercapacitor cells connected in various in series and in parallel combinations.
Environmentally friendly technology
Supercapacitors show potential to be the most important device for storage and recovering of energy in hybrid vehicles. Commercialization of those vehicles, which will be triggered by the development of suitable supercapacitor technology, will have a direct influence on our environment by reducing the amount of fossil fuels being in use as well as decreasing the number of combustion engines and thus the amount of CO2 released to the atmosphere.
The supercapacitors are also suited for used in unbreakable power supplies what makes emergency generators with petrol obsolete. Additionally, in comparison to conventional batteries, supercapacitors generate less waste and those produced are much more convenient to handle. The supercapacitors do not make use of toxic materials like cadmium or rare natural resource as lithium. Resignation from fluid electrolyte and use of solid electrolyte, or as in this case ionic liquids, which are also known under the name “green solvents”, will prevent the loss of fluid in case of damage and the risk of polluting the environment or causing a health risks to the user.
Scientific impact
Within the framework of project ElectroGraph the scientists developed an innovative graphene production technology. This includes the synthesis process of graphene nanoplatelets, their processing and the manufacturing of electrodes for energy storage technologies.
The synthesis process of graphene nanoplatelets is called electrochemical exfoliation. The work for this project enabled a deep understanding for the process which is a clear development beyond the state of the art. The main aim is to make the very high theoretical surface area accessible. The more surface is accessible for electrolyte ions, the more energy can be stored. The application is recuperation for electromobility. This development significantly improves the technology and can be improved and progress further.
The developed process enables electrodes for energy storage purposes with an improvement of specific capacitance of 75% compared to state of the art commercial electrodes. The innovative electrodes achieve better adhesion of the active material to the matrix material. One of the main advantages of the process is, that the green technology consummated less energy than other ways of synthesizing graphene.
Main dissemination activities
Within the framework of project ElectroGraph, there was a total of 19 peer reviewed publications. The titles of the five most important ones are:
• Optimization of the characterization of porous carbons for supercapacitors
• The influence of pore size and surface area of activated carbons on the performance of ionic liquid based supercapacitors
• Pore size distribution and capacitance in microporous carbons
• Electroanalytical Sensing Properties of Pristine and Functionalized Multilayer Graphene
• Inkjet-defined field-effect transistors from chemical vapour deposited grapheme
For the full list please see table A1: list of scientific (peer reviewed) publications.
Furthermore we identified a total of 28 dissemination activities in the form of presentations, posters, websites and workshop. For example Fraunhofer IPA gave a presentation to a European audience from the scientific community at GRAPHENE 2020 – Opportunities for Europe in Brussels on 21/03/2011. Fraunhofer IPA also gave the presentation “Ionic liquid based graphene electrochemical exfoliation” at Stuttgart Nanodays 2011 for the international scientific community on 29/09/2011. Regarding health risks, the institute of occupational medicine gave the oral presentation “Current risk management measures for controlling exposure to nanomaterials in European workplaces” on SENN 2012 in Finland. CSIC displayed a poster about “Graphene-type materials from carbon fibers for supercapacitors” at the 3rd International Symposium of Enhanced Electrochemical Capacitors ISEECAP, Taormina, Italy on 05/06/2013. In addition,
Dissemination workshops
Workshops dedicated to graphene nanomaterials where all partners had the opportunity to present their results were held. They were open to public and to presentations from other leading researches in the graphene field from Europe and in the World. The use was for dissemination of the project results to the academic community, but also to industrial partners and potential investors. The Workshops provided a forum for facilitated discussions about range of topics related to novel graphene materials, such as synthesis, applications, standards, etc. There were several dissemination workshops fot the project ElectroGraph held. The information from the workshop in 2011 can be found here: http://www.electrograph.eu/downloads. The data from the workshop in 2014 is accessible on: http://www.electrograph.eu/node/45.
Dissemination material for the public is also made available on: http://www.electrograph.eu/public-documents.
Patent and Literature Study
In the progress of ElectroGraph project it was important for the participating members that the carried out investigations and developments were complimentary and progressing beyond the state-of-the-art. For this reason a continuous scoping of the scientific publications and patents relevant to the objectives of the project was undertaken. The “Patent and Literature Study” report resulting from these activities is made available here: http://www.electrograph.eu/public-documents
For the full list please see A2: list of dissemination Activities.
Exploitation of results
There are the following exploitable project outputs:
Power storage materials and manufacturing processes: They are made of carbon and graphene based material systems with high power or energy storage capability. Furthermore, ways to connect these into systems were researched.
Structural power storage systems: The material solutions were introduced into system conceptual designs. These systems are developed for efficient packaging and power management and connectivity. Methods for their inspection, repair and disposal were also researched.
Power storage car components: The developed supercapacitor was implemented in a cars rear view mirror for technology demonstration. The components were designed, manufactured and tested according to the OEMs specifications. This resulted in ready to use solutions for structural energy storage components, which can be introduced in today’s assembly line.
Experimental data: Material data and empirical solutions for design and, in conjunction with the microscopy results, provide information to validate predictive and simulation models. These data enabled critical parameters to be identified, and build a better understanding of these materials.
Ultimately, this offers European industry lower costs and improves design expertise associated with these potentially rewarding materials.
Two consortium partners have the main role in the exploitation of the project results. These are Maxwell Technologies and CRF within the FIAT Group. Maxwell Technologies Inc. is a company that knows supercapacitors and advanced energy storage technologies and the markets that those technologies serve very well. As a pioneer in the development, commercialization and utilization of supercapacitors the strengths and weaknesses of the current devices are well understood. There are relatively few things that supercapacitor users really want by way of improvement in the performance of the devices however the list is very commonly shared among those users as the need is really universal. First more energy in the same device is a welcome improvement. Higher energy density is desired as well. This was achieved through higher specific capacitance.


List of Websites:
Public Website: www.electrograph.eu

No. Participants Representatives Contact information
1 Fraunhofer IPA (Coordinator) Carsten Glanz carsten.glanz@ipa.fraunhofer.de

2 Danubia Nanotech s.r.o Viera Skakalova skakalova@danubiananotech.com

3 Institute of Occupational Medicine Steve Hankin steve.hankin@iom-world.org

4 Trinity College, Dublin Georg Duesberg duesberg@tcd.ie

5 Centro Ricerche Fiat SCPA Antonino Veca antonino.veca@crf.it

6
Agencia Estatal Consejo Superior de Investigaciones Cientificas Teresa A. Centeno teresa@incar.csic.es

7 University of Nottingham Stephen Pickering stephen.pickering@nottingham.ac.uk

8
Université Paris Diderot - Paris 7
Hyacinthe Randriamahazaka hyacinthe.randria@univ-paris-diderot.fr

9 Maxwell Technologies S.A Martin Angeloz mangeloz@maxwell.com

10 The University of Exeter Yanqiu Zhu y.zhu@exeter.ac.uk