Fabrication of Multifunctional Graphenebased Polymer Composite Materials Using Different Fillers

Multi-functional polymer composite materials are of great interest in the field of fabrication of composites owing to their high volume of applications. Different nano and micron sized filler materials are embedded into the matrix to achieve desirable functionalities. Graphene, a sheet of a single atom thick, sp bonded carbon atoms arranged in honeycomb structures is one of the most extensively used filler because of its several exceptional features such as high electrical and thermal conductivities, mechanical and gas barrier properties. Few layer and multilayer graphene sheets are also very promising alternatives of single layer graphene sheet. The biggest challenge in working with graphene is to keep them well dispersed as they always tend to agglomerated due to strong van der Waals force. In this thesis, graphene based polymer composite materials are fabricated and their various engineering properties have been studied. The primary goal of the thesis is to find out different strategies to disperse graphene sheets uniformly in the matrix. A non-conductive, second filler is added to the matrix as a dispersion aid to prevent restacking of graphene sheets. The electrical conductivity of the composites is studied. Several orders of magnitude increase in the electrical conductivity is observed with the addition of non-conductive filler. Different multi-functional polymer composite materials using appropriate fillers are fabricated. Incorporating some fillers can deteriorate the mechanical properties of the system. Suitably selected fillers can act as dispersion aid as well as can enhance the toughness of the composites. Quasi-static compression test and three-point flexural test are performed on these materials. X-ray diffraction and scanning electron microscope are used to study the dispersion of graphene sheets in the matrix. Also Instron Universal testing machine and two-point probe technique are used to examine the electrical and mechanical properties of the composites. The effect of the size of the second filler on the electrical conductivity of the composites is also studied using silica nanoparticles (200 nm) and alumino-silicate ceramic microspheres (12 microns). Smaller particles are found to be more effective in improving the dispersion of graphene compared to the bigger particles.


Thesis Objectives:
The one of the goal of this project is to fabricate the electrically conductive polymer composites. The main principle behind developing these composites is to add conductive filler in an insulating polymer matrix. These conductive fillers would make a connected network which would result in significant increase in the electrical conductivity. The minimum amount of the filler at which the conductivity starts to increase is defined as the percolation threshold of the composites. We use graphene nanoplatelets (GNP) as filler and polystyrene as matrix. Since graphene is essentially a two-dimensional structure, the 'volume' swept by it is equivalent to that of a sphere of diameter corresponding to the lateral dimensions of the graphene sheets, giving a theoretical volume loading at percolation which is significantly lower than that of spheres 16 . If graphene-based sheets are modeled as disks of aspect ratio AR (AR = disk diameter/thickness), the percolation threshold φc is inversely proportional to AR, and is given by Here, φsphere is the percolation threshold for spheres, i.e., φsphere = 0.29. The pre factor is 1.5 for disks, and depends on the geometry of the sheets. Since AR can take on values of the order of 10 4 , the advantage of using high aspect ratio conducting sheets in lowering the percolation threshold becomes apparent. In practice, φc as low as 0.001 has been reported 2 , while for spheres the percolation thresholds reported have been closer to 0.15 2,5,6,17 . Achieving percolation at such a low loading also is also extremely beneficial for mechanical properties, particularly under impact loading conditions, as filler materials can act as nucleation sites for crack growth 18 .
The objectives of this project are to develop strategies for distributing graphene nano platelets (GNP) in a polymer to significantly enhance its functionality. Specific property targets are electrical conductivity, mechanical properties. The combination of the high aspect ratio graphene nano platelets (GNP) and traditional dispersion techniques such as melt extrusion, that are both highly directional, inherently produce composites with the sheets aligned in the flow direction, either in a parallel or perpendicular orientation. In addition, agglomeration of these sheets during processing is a key problem. A key goal of this project is to fabricate graphenepolymer composite materials with enhanced electrical and mechanical properties by developing good dispersion techniques for graphene sheets. But one of the biggest challenge with the processing of these composites is to disperse graphene uniformly in the polymer matrix. Due to its inherent sheet-like two-dimensional structure, graphene sheets always tend to get agglomerated or restacked because of strong van der Waals force 19 . This is very crucial as long as different properties of multifunctional composite materials are concerned. It is very difficult to exploit all interesting, inherent properties of graphene sheets, if they always tend to get agglomerated. As a result of this, keeping graphene sheets dispersed uniformly throughout the matrix is paramount importance to all researchers working with graphene as an electrically conductive additive in the composites. To facilitate the dispersion of graphene in polymer matrix, silica nanoparticles are added in the matrix where they randomly occupy some space of the matrix where graphene sheets are not allowed to enter.
These second, non-conductive fillers can act as posts in the matrix. So the graphene sheets then move around these particles and can make a conductive network in these composites at lower loading. Solvent casting method is used to make these composite materials 17 . Also it is not very convenient to process a large amount of graphene sheets to get desired amount of electrical conductivity as these sheets always get restacked and make the processing condition extremely difficult and as a result of that the effectiveness of using graphene as an electrically conductive additive disappears. So this technique can serve the purpose of utilizing graphene in the polymer composites and helps to keep them dispersed in the matrix without compromising exceptional electrical properties of graphene.
The mechanical properties of GNP-siica/rubber-polystyrene composites are also studied. As we are expecting some significant improvement in the electrical property of the composites after incorporating silica in it, the ductility of the composites would be deteriorated making them more brittle because of the presence of silica in it. So another way to overcome this issue is to replace silica particles with rubber spheres as second filler. In this way, the final composites would be electrically conductive and ductility of the composites would be improved. Finally an interesting study is to do comparative analysis of mechanical properties between GNP -polystyrene composites containing silica particles and rubber particles. Static compression tests and threepoint flexural tests will be performed on these composites and an effort will be made to fabricate a multi-functional polymer composites possessing significant electrical and mechanical properties.

1 Abstract:
Graphene is as an attractive filler material for polymers because of its excellent electrical, mechanical and thermal properties. In this paper, we report a massive increase in the electrical conductivity of a multilayer graphene (MLG)/polystyrene composite following the addition of non-conducting silica particles. The nonconducting filler acts as a highly effective dispersion aid, preventing the sheet-like MLG from restacking or agglomerating during the solvent casting process used to fabricate the composite. The enhanced dispersion of the MLG leads to orders of magnitude enhancement in electrical conductivity compared to samples without this filler.

Introduction:
Defect-free single layer graphene sheets consist of single-atom-thick ,sp 2 -bonded, hexagonally arranged carbon atoms. They display remarkable properties including exceptional in-plane electrical and thermal conductivity, high stiffness and tensile strength, optical transparency, negligible permeability to gases, and van der Waals transparency. [1][2][3][4][5][6][7][8] The scientific and commercial interest in graphene is not restricted to the pristine monolayer, but includes related 2D materials that include few-layer graphene (FLG), multilayer graphene (MLG) and chemically modified forms such as graphene oxide (GO). 2 The essentially 2-dimensional nature of these materials along with their excellent properties makes them important as fillers, imparting useful functionalities into matrices. Polymers that display high conductivity have a variety of uses ranging from bulk applications such as anti-static mats and fuel lines, [9][10][11][12][13][14][15] to specialty applications such as radiation shields, sensors and electrodes for batteries. [16][17][18][19][20][21][22][23][24] While single layer graphene remains expensive and best suited for high-value applications in electronic devices, opto-electronics, and supercapacitors, 1, 25 the much lower cost MLG is a more promising material for applications that seek to impart electrical conductivity to polymers. Therefore, we target MLG/polymer composites in an effort to provide electrical conductivity to the insulating polymer. In this paper, we report an unexpected result, where we observe a massive enhancement in the electrical conductivity of a MLG/polystyrene composite upon the addition of a second, nonconducting filler.
To achieve practical levels of electrical conductivity in an insulating material, a conducting filler must be loaded to a volume fraction beyond the percolation threshold. 16,17 MLG are two-dimensional structures, which if allowed to rotate freely in a matrix, sweep a 'volume' that is a sphere of diameter corresponding to the lateral dimensions of the MLG, giving a theoretical volume loading at percolation that is well below that of spheres. 17 If MLG are modeled as ideally dispersed and randomly rotated disks of aspect ratio AR (AR = disk diameter/thickness), the percolation threshold φc is given by 26 In Equation (1), φsphere is the percolation threshold for spheres, i.e., φsphere = 0.29 (φsphere = 0.29 is for monodispersed spheres; that number is lower if there is polydispersity, but remains of the same order of magnitude). Since AR can take on values of the order of 10 4 for MLG, the advantage of using these high aspect ratio conducting particles in lowering the volume loading at percolation becomes apparent.
Providing such a low loading at percolation also has a significant benefit for mechanical properties, particularly under impact conditions, as filler materials can act as nucleation sites for crack growth. [27][28][29] While the volume loading at percolation is small for sheet like materials, van der Waals attraction between these sheets causes rapid agglomeration, degrades dispersion, and enhances restacking. The restacking reduces the aspect ratio and typically prevents achieving the performance predicted by Equation (1). Thus dispersing these high aspect ratio sheets in a polymer remains a major challenge. We hypothesized that the addition of a second filler could overcome this issue, because this filler would act as spacers and prevent agglomeration of MLG during processing.
In addition, if the second fillers were dispersed homogeneously throughout the polymer, they would guide the sheet-like MLG into a more random orientation in the polymer, enhancing the probability of MLG percolation at low loadings as shown in

4 Fabrication of Composites:
7g of the polystyrene pellets are dissolved in 42ml of N,N-dimethylformamide (DMF) and the solution is stirred magnetically for 12 hours. 30 The silica particles are then added and the mixture sonicated for 1.5 hrs. MLG at a concentration of 0.001gm/ml are dispersed in DMF and sonicated for 1.5hrs. Both particle-containing suspensions are then mixed in a 1:1 ratio and magnetically stirred for 2hrs. This mixed suspension is then poured into methanol, an antisolvent for PS. The PS precipitates rapidly, creating the composite. The excess methanol is withdrawn, and the composite is dried in an oven for 18 hr at 90ºC. The sample is then hot pressed at 120ºC to get rid of all entrapped air bubbles, and to create a sample with a disk-like shape that is amenable for electrical conductivity measurements. All reported loadings are based upon the volume percent in the final composite.

Characterization and Electrical Conductivity Measurement Technique:
The surfaces of specimens are coated with silver paint to reduce contact resistance. A standard two-point probe using a constant current source (Keithley Instruments Model 6221) is used to obtain bulk volumetric electrical conductivity. The voltage drop across the specimen is recorded, and the resistance of the sample calculated from this measurement. This is normalized with the dimensions of the sample to produce the electrical conductivity. The surface morphology of the composites is observed using scanning electron microscopy (Zeiss SIGMA VP FE-SEM) in backscatter mode. A Rigaku Ultima IV diffractometer with Cu Kα radiation is used for the X-ray diffraction (XRD) measurements.

6 Results and Discussions:
We    Figure 3(b) shows better dispersion of the MLG at this silica concentration..
As the silica loading is increased to 12 vol%, the conductivity rises further to 1S/m, and the MLG are dispersed more uniformly throughout the sample (Figure 3(c)).
Beyond 12 vol% silica, there is a decrease in electrical conductivity of the composite, which is then nearly constant over the remaining range of feasible silica loadings. The excessive silica particles at these concentrations starts to break the connectivity of the MLG network, as seen in Figure 3(d) at 20 vol% silica.
We measure the full width at half maximum of the graphite (0 0 2) diffraction peaks (Figure 3(e)), and use Scherrer's analysis to determine an average 'crystallite' size for the MLG as an indicator of restacking (Figure 3(f)). The average crystallite size decreases as the silica loading goes to 12 vol%, and then rises again. This indicates a suppression of restacking and also suggests improved dispersion of the MLG at concentrations up to 12 vol% silica, followed by increased MLG agglomeration as the silica content is increased further.

Conclusions:
We see that addition of a non-conductive filler can significantly reduce the loading of MLG required for percolation. While we have used 200nm silica particles as dispersion aids in this work, we recognize that there is a range of materials, morphologies and sizes of fillers that can be exploited to impart desirable properties to a composite. Optimization of this novel second-filler concept will be the subject of future work.

Acknowledgements:
This work was supported by a grant from the Rhode Island Science and Technology Advisory Council.

Introduction:
Graphene Nanoplatelets (GNP) have attracted the attention of many researchers due to its exceptional electrical, thermal , mechanical and barrier properties 1-3 . GNP is  then hot pressed at 120ºC to get rid of all entrapped air bubbles, and to create a sample with a disk-like shape that is amenable for electrical conductivity measurements. All reported loadings are based upon the volume percent in the final composite. The surfaces of specimens are coated with silver paint to reduce contact resistance.

Measurement:
A standard two-point probe using a constant current source (Keithley Instruments

Results and Discussions:
Core-shell rubber particles induce percolation in the matrix by maintaining a well connected network of GNP.     We study quasi-static compression stress-strain behavior of GNP-rubber/silicapolystyrene composites under static compression loading. Figure 6 shows the stressstrain behavior of GNP-silica/rubber-polystyrene composites. In figure 6  To exploit the best possible inherent properties of the filler, uniform dispersion of the filler is very essential which on the other hand can affect the electrical and mechanical properties significantly 29 . In our study, we see that rubber particles can perform better as filler materials due to better interaction of shell structures of rubber particles with the polystyrene matrix. For electrical property of the composites, lesser amount of rubber particle is necessary to induce percolation in the system compared to the silica particles. Also the flexural strength of the composites increases with rubber fillers due

Conclusion:
We see that incorporating appropriate filler materials into the polymer matrix enables

Acknowledgements:
I would like to thank department of chemical engineering, University of Rhode Island for supporting my work. We thank Dr. Arun Shukla for several insightful discussions.

Abstract:
The size of the second, non-conductive filler plays a significant role in order to enhance the electrical conductivity of the system. Two different sized second fillers are chosen to compare the effectiveness of the particles in increasing the conductivity of the composites. In the work, it has been found that smaller particles can act as better dispersion aids in preventing the restacking of GNP sheets in compared to bigger particles. The distribution of GNP networks is studied using back-scattered scanning electron microscope (SEM) and the orientation of the non-conductive filler along with GNP sheets is analyzed using charge-contrast scanning electron microscope (SEM).

Introduction:
Graphene nanoplatelets (GNP) are considered to be one of the most widely used filler in the field of fabrication of composite materials [1][2][3][4] . Electrically conductive polymer composites have been accepted for numerous applications in a broad range of areas from specialty uses such as electromagnetic interference (EMI) shielding materials, electrodes for batteries,sensors to bulk applications which include anti-static plastic mats, fuel lines [5][6][7][8][9][10][11][12][13][14][15][16][17] . Electrically conductive fillers are added into non-conductive matrix to impart conductivity in the system. In order to achieve the conductivity, a minimum amount of conductive filler needs to be present which is called the percolation threshold of the system [18][19][20] . In our previous work we have shown that incorporating a second, non-conductive filler can induce percolation with improved dispersion of GNP 21 . One of the biggest problem in working with GNP is that they always tend to get agglomerated due to strong van der Waals force. Introducing a second filler can remarkably prevent restacking of GNP sheets resulting in increase in electrical conductivity at lower loading of GNP. Several properties of the second filler such as size, shape, interaction of the filler with the matrix can significantly affect the electrical conductivity of the final composites [22][23][24]

4 Fabrication of Composites:
7g of the polystyrene pellets are dissolved in 42ml of N,N-dimethylformamide (DMF) and the solution is stirred magnetically for 12 hours. 25 The silica particles/alumino-silicate ceramic microspheres are then added and the mixture sonicated for 1.5 hrs.
GNP at a concentration of 0.001gm/ml are dispersed in DMF and sonicated for 1.5hrs.
Both particle-containing suspensions are then mixed in a 1:1 ratio and magnetically stirred for 2hrs. This mixed suspension is then poured into methanol, an antisolvent for PS. The PS precipitates rapidly, creating the composite. The excess methanol is withdrawn, and the composite is dried in an oven for 18 hr at 90ºC. The sample is then hot pressed at 120ºC to get rid of all entrapped air bubbles, and to create a sample with a disk-like shape that is amenable for electrical conductivity measurements. All reported loadings are based upon the volume percent in the final composite.

Characterization and Electrical Conductivity Measurement Technique:
The surfaces of specimens are coated with silver paint to reduce contact resistance. A

Results and Discussions:
Electrical Conductivity of graphene/ alkali alumino silicate ceramic/ polystyrene composites: We have investigated the effect of the size of the non-conductive filler on the electrical conductivity of the composites. These alumino silicate ceramics (12 micron in size) act as posts in the matrix, preventing agglomeration of graphene sheets which in turn helps in building the connected pathways of graphene sheets which will increase the electrical conductivity.
The electrical conductivity of these composites is lower than the electrical conductivity of graphene/ silica particles/ polystyrene composites at the same loading of graphene and the non-conductive filler. These results are summarized in figure 1.
Smaller particles get distributed more evenly These results can be explained by the fact that the smaller particles get dispersed more easily and randomly in the matrix during processing whereas it is more challenging to keep these bigger ceramic microspheres dispersed due to their larger size as they always have a tendency to settle    In figure 5, the TEM images of GNP-polystyrene composites show the orientation of silica/ alumino-silicate-ceramic spheres and graphene sheets in the matrix. Figure   9.(a) shows the distribution of silica nanoparticles and graphene sheets in the matrix.
Energy dispersive X-ray spectroscopy (EDS) shows the presence of silicon peak which confirms the presence of silica nanoparticles in figure 9.(b). Figure 9.  2) The electro-mechanical response of these multi-functional composite materials can be studied. This feature is very useful for various sensors applications.
3) Composites with improved electrical and thermal properties can be developed with graphene and good thermally conductive second fillers such as gold, silver, aluminum nanoparticles. Electronic devices, thermal pastes, heat-actuated, shape-memory polymers have enormous demands for these type of composites.
4) The stress-strain response of these multi-functional composites can be studied under dynamic loading condition. The dynamic responses of these composites have many applications in blast loading, impact during crash and impulse loading.
5) The combination of two conductive fillers such as graphene with carbon nanotube might be very promising area to work on. These electrically conductive polymer composites can be used as anti-static mat, anti-static coating, conducting paint, electromagnetic interference (EMI) shielding materials.
6) The effect of the shape of the second filler can significantly affect the electrical conductivity of the composites. Different shaped second filler can alter the network of graphene network which will result in variation in the conductivity. Investigating the best possible shape of the second filler in terms of improving the electrical conductivity would be a very interesting topic.

7)
Varying the type of polymer for the matrix can help in fabricating a completely different type of composites where incorporating graphene with any second filler will impart remarkable electrical and mechanical properties in the final composites. For example, graphene and polyisobutylene spheres can be used as fillers in the polybutadiene matrix to make a very flexible, electrically conductive polymer composites which can be used as automobile fuel line injector applications.