DYNAMIC RESPONSE OF TAILORED CARBONBASED MULTIFUNCTIONAL COMPOSITES

The importance of nanotechnology in the world has dramatically increased in the recent years as miniaturization has become more important in areas such as computing, sensing, biomedicine and many others. Advancements in these disciplines depend largely on the ability to synthesize nanoparticles of various shapes and sizes, as well as to assemble them efficiently into complex architectures. Currently, materials reinforced with nanoparticles have an enormous range of applications owing to their superb mechanical and physical properties. More specifically, recent progress has shown that using inorganic nanomaterials as fillers in polymer/inorganic composites has tremendous application potential in automotive, aerospace, construction and electronics industries. If properly incorporated into a polymeric matrix, conductive nanofillers such as carbon nanotubes (CNTs) and graphene can be assembled into a three-dimensional electrical network. Furthermore, these highly intricate networks can be utilized as an internal sensory mechanism capable of detecting information such as material deformation and various forms of damage. Fundamental investigation into the mechanical and electrical properties of nanomaterial-based polymer composites when subjected to loading is paramount before they can be incorporated into high performance applications. For this purpose, a comprehensive study was conducted to understand the electro-mechanical behavior of both CNT-based and graphene-based composites under static and dynamic loading conditions. Moreover, novel strategies were developed to produce designer graphene-based composites that possess tailored transport and mechanical properties. A series of dynamic compressive experiments were performed to experimentally investigate the electrical response of multi-wall carbon nanotube (MWCNT) reinforced epoxy nanocomposites subjected to split Hopkinson pressure bar (SHPB) loading. Low-resistance CNT/epoxy specimens were fabricated using a combination of shear mixing and ultrasonication. Utilizing the carbon nanotube network within, the electrical resistance of the nanocomposites was monitored using a high-resolution four-point probe method during each compressive loading event. In addition, real-time deformation images were captured using high-speed photography. The percent change in resistance was correlated to both strain and real-time damage. The results were then compared to previous work conducted by the authors (quasi-static and drop-weight impact) in order to elucidate the strain rate sensitivity on the electrical behavior of the material. In addition, the percent change in conductivity was determined using a Taylor expansion model to investigate the electrical response based upon both dimensional change as well as resistivity change during mechanical loading within the elastic regime. Experimental findings indicate that the electrical resistance is a function of both the strain and deformation mechanisms induced by the loading. The bulk electrical resistance of the nanocomposites exhibited an overall decrease of 4065% for quasi-static and drop-weight experiments and 65-85% for SHPB experiments. An experimental investigation was conducted to understand the electromechanical response of graphene reinforced polystyrene composites under static and dynamic loading. Graphene-polystyrene composites were fabricated using a solution mixing approach followed by hot-pressing. Absolute resistance values were measured with a high-resolution four-point probe method for both quasi-static and dynamic loading. A modified split Hopkinson (Kolsky) pressure bar apparatus, capable of simultaneous mechanical and electrical characterization, was developed and implemented to investigate the dynamic electro-mechanical response of the composites. In addition to measuring the change in electrical resistance as well as the dynamic constitutive behavior, real-time surface damage and global deformation was captured using high-speed photography. The real-time damage was correlated to both stress-strain and percent change in resistance profiles. The experimental findings indicate that the bulk electrical resistance of the composite increased significantly due to the brittle nature of the polystyrene matrix and the presence of relative agglomerations of graphene platelets which resulted in micro-crack formations. A novel capillary-driven particle-level templating technique along with hot melt pressing was developed and utilized to disperse few-layer graphene (FLG) flakes within a polystyrene matrix to enhance the electrical conductivity of the polymer. The conducting pathways provided by the graphene located at the particle surfaces through contact of the bounding surfaces allow percolation at a loading of less than 0.01% by volume. This method of distributing graphene within a matrix overcomes the need to disperse the sheet-like conducting fillers isotropically within the polymer, and can be


PREFACE
An experimental investigation has been conducted to investigate the electromechanical behavior of multifunctional materials under static and dynamic loading.
Additionally, novel strategies to develop multifunctional materials with tunable properties have also been developed. Understanding the overall electro-mechanical response of these multifunctional composites will lead to the development of improved smart materials capable of sensing crucial information such as material deformation and damage within the material. Due to the increased demand for high performance materials that possess multi-functionalities, a comprehensive understanding of the dynamic electro-mechanical response of multi-functional composites is pivotal. This dissertation addresses the dynamic electro-mechanical behavior of both carbon nanotube (CNT) and graphene reinforced composites under compressive loadings. This dissertation is prepared using the manuscript format. ix Chapter 2 investigates the electrical behavior of CNT/epoxy nanocomposites subjected to split Hopkinson pressure bar (SHPB) loading. An SHPB apparatus was utilized to load the specimens while the resistance history and high speed deformation images were both captured. In addition, the percent change in conductivity was determined using a Taylor expansion model to investigate the electrical response based upon both dimensional change as well as resistivity change during mechanical loading within the elastic regime. This chapter will follow the formatting guidelines specified by the Journal of Material Science.
Chapter 3 focuses on the electro-mechanical behavior of graphene/polystyrene composites under dynamic loading. An experimental investigation was conducted to understand the electro-mechanical response of graphene reinforced polystyrene composites under static and dynamic loading. Graphene-polystyrene composites were fabricated using a solution mixing approach followed by hot-pressing. Absolute resistance values were measured with a high-resolution four-point probe method for both quasi-static and dynamic loading. A modified split Hopkinson (Kolsky) pressure bar apparatus, capable of simultaneous mechanical and electrical characterization, was developed and implemented to investigate the dynamic electro-mechanical response of the composites. In addition to measuring the change in electrical resistance as well as the dynamic constitutive behavior, real-time surface damage and global deformation was captured using high-speed photography. The real-time damage was correlated to both stress-strain and percent change in resistance profiles. This chapter will follow the formatting guidelines specified by the Journal of Experimental Mechanics.
x Chapter 4 details a novel strategy developed for producing highly conductive graphene-based segregated composites prepared by particle templating. A capillarydriven particle-level templating technique and hot melt pressing was used to disperse few-layer graphene (FLG) flakes within a polystyrene matrix to enhance the electrical conductivity of the polymer. The conducting pathways provided by the graphene located at the particle surfaces through contact of the bounding surfaces allow percolation at a loading of less than 0.01% by volume. This chapter will follow the formatting guidelines specified by the Journal of Materials Science.
Chapter 5 details a novel method for tailoring the electro-mechanical properties of templated graphene/polymer composites by using fixed-angle rotary shear. A capillary-driven particle-level templating technique was utilized to distribute graphite nanoplatelets (GNPs) flakes into specially constructed architectures throughout a polystyrene matrix to form multi-functional composites with tailored electromechanical properties. By precisely controlling the temperature and pressure during a melt compression process, highly conductive segregated composites were formed using very low loadings of graphene particles. Since the graphene flakes form a honeycomb percolating network along the boundaries between the polymer matrix particles, the composites show very high electrical conductivity but poor mechanical strength. To improve the mechanical properties, a novel processing technique was developed that uses rotary shear through fixed angles to gradually evolve the honeycomb graphene network into a concentric band structure over the dimensions of

INTRODUCTION AND LITERATURE REVIEW
The importance of nanotechnology in the world has dramatically increased in the recent years as miniaturization has become more important in areas such as computing, sensing, biomedicine and many others. Advancements in these disciplines depend largely on the ability to synthesize nanoparticles of various shapes and sizes, as well as to assemble them efficiently into complex architectures. Currently, materials reinforced with nanoparticles have an enormous range of applications owing to their superb mechanical and physical properties. More specifically, recent progress has shown that using inorganic nanomaterials as fillers in polymer/inorganic composites has tremendous application potential in industries such as automotive, aerospace, construction and electronics [1][2][3][4][5][6][7]. If properly incorporated into a polymeric matrix, conductive nanofillers such as CNTs and graphene can be assembled into a threedimensional electrical network. Furthermore, these highly intricate networks can be utilized as an internal sensory mechanism capable of detecting crucial information such as material deformation and various forms of damage. Despite recent progresses on the mechanical and electrical characterization of nanomaterial-based polymer composites, little results have been published regarding the electro-mechanical behavior of such composites when subjected to dynamic loading conditions. To meet this need, CNT-based composites as well as graphene-based composites will be investigated to study their electro-mechanical behavior under dynamic loading.
Extraordinary mechanical properties and excellent transport properties make CNTs and graphene a promising addition to the future of smart composite materials [1][2][3][4][5][6][7]. CNTs are a promising addition to the future of developing novel materials capable of self-sensing and active response due to their extraordinary electrical conductivity and excellent transport properties. Unlike other smart materials, CNTs can provide both structural and functional capabilities simultaneously and have been used in many applications including actuation, sensing, and power generation [8][9][10][11].
Carbon nanotube-based polymeric nanocomposites have found new applications in various fields such as future spacecraft, anti-meteorite/anti-ballistic shields for satellites, anti-ballistic vests, explosion-proof blankets for aircraft cargo bays, and safety belts [12,13].
Significant research has been performed to fundamentally understand the enhancement of mechanical properties due to CNT reinforcement of polymers [14,15]. Allaoui et al. [16] investigated the influence of multi-walled carbon nanotubes (MWCNTs) in a rubbery epoxy matrix, and found that the addition of up to 4 wt.% MWCNTs could lead to a significant increase in the strength and Young's modulus [17]. Given the practical potential applications of CNTs in electromechanical devices, specifically as piezoresistive sensors, the effect of mechanical deformation on the electrical properties of individual CNTs has been studied theoretically [18][19][20][21] and experimentally [19,21].
An increased amount of research has been conducted in studying the electrical response of CNT/polymer composites under mechanical loading. Alexopoulos et al. [18], Nofar et al. [20], and Gao et al. [22,23] have studied damage detection and health monitoring of composites reinforced with CNT-embedded glass fibers.
Alexopoulos et al. [18] performed various incremental tensile loading-unloading steps 3 as well as three-point bending tests on specimens with CNT fibers in the tensile region. Results indicated that CNT fibers provide unquestionable advantages for sensing and damage monitoring of non-conductive composites, when compared to the competitors, e.g. the embedded carbon fibers and a modified (doped) conductive network. Sensing ability for the investigated specimens with CNT fibers in the compressive region was also reported. Gao et al. [23] studied the sensing of damage development in composites using CNT networks utilizing two sensing techniques: electrical resistance and acoustic emission. Resistance change and acoustic emission counts showed a bi-linear relation in detecting damage in quasi-static and cyclic experiments which can be used to give additional insight toward damage evolution.
Thostenson et al. [24] performed tensile experiments on CNT/epoxy samples and demonstrated a highly linear relationship between the specimen deformation and the electrical resistance. This result suggests that CNT networks formed in an epoxy polymer matrix could be utilized as highly sensitive sensors for detecting the evolution of damage in advanced polymer-based composites [24]. Later, Lim et al. [25] experimentally investigated the mechanical and electrical response of CNT-based fabric composites to Hopkinson bar loading, further demonstrating the effectiveness of a percolating carbon nanotube network being capable of sensing damage caused by impact. In their previous work, the authors experimentally investigated the electrical response of multi-walled carbon nanotube reinforced nanocomposites under quasistatic and dynamic loading. The results indicated that the electrical resistance of the nanocomposite decreased under both quasi-static and dynamic loading due to the 4 formation of more efficient carbon nanotube networks caused by the compression of the epoxy matrix [25,26].
Although carbon nanotubes (CNTs) possess excellent physical and mechanical properties, graphene amazingly possesses superior electrical and thermal properties, as well as a higher specific surface area [4,27]. Its reinforcement can also offer exceptional properties in future high performance novel composites. In recent years, graphene based composites have become a topic of significant academic and industrial interest. While a number of studies have shown that the presence of graphene within polymers can enhance the mechanical properties of the bulk composite [27][28][29][30][31][32][33], other studies have shown that graphene can also have adverse effects on the mechanical properties [34][35][36]. Fang et al. [28] investigated the effect of low concentrations of graphene on the mechanical strength of graphene/polystyrene composites. The results showed a substantial increase in tensile strength as graphene loadings were increased from 0.1 wt. % to 0.9 wt. % in comparison to pristine polystyrene. The increase in strength was attributed to effective load transfer between the graphene and polymer.
Alternatively, the addition of certain filler materials can also have adverse effects on the mechanical properties of the resulting composite due to factors such as reinforcement phase concentration, dispersion quality, interface bonding, aspect ratio, surface-to-volume ratio of filler, etc. [27,[34][35][36][37][38][39]. Wang et al. [39] compared the use of graphite nanosheets to carbon black as a filler material in high density polyethylene.
They reported a gentle increase in both tensile strength and impact strength of the composite with low loadings of graphene (0.5 to 2.0 wt. %) but a sharp reduction when the graphene content was greater than 2 wt. %. Due to the high surface energy of 5 graphene, as well as the weak interaction between the graphene and polyethylene, an inhomogeneous dispersion in the polymer matrix was formed when the content of graphene was high, leading to adverse effects on the properties of the composites.
Due to the exceptional electrical properties of graphene, several researchers in the past have also studied the utilization of graphene as an electrically conductive additive in composites [3,38,40,41]. The electrical conductivity of graphene-based composites has been studied theoretically [42] as well as experimentally [38,41,43,44]. Studies have shown remarkable increases in composite electrical properties with graphene reinforcement. More recently, Qi et al. [41] demonstrated a substantial enhancement of electrical properties of polystyrene (PS) with the addition of graphene. The conductivity of the graphene/polystyrene composite was shown to be ~ 2-4 orders of magnitude higher than that of multi-walled-carbon-nanotube/polystyrene composites.
The combination of the remarkable mechanical properties and the exceptional electrical properties make graphene another ideal candidate for use as a filler material in fabricating multi-functional composites capable of sensing material behavior. Many reports demonstrate the effectiveness of utilizing graphene in providing strain sensing functions [45][46][47]. Eswaraiah et al. [45] demonstrated the real time strain response of functionalized graphene-polyvinylidene fluoride (f-G-PVDF) composites on the macro-scale under tensile loads and the use of the composite as a strain sensor. The analysis of the change in voltage of various composite films revealed that the graphene-based composite showed better strain sensing performance compared to carbon nanotube-based polymer composites. In their previous work, the authors 6 experimentally investigated the electrical response of multi-walled carbon nanotube reinforced nanocomposites under quasi-static and dynamic loading. The results indicated that the electrical resistance of the nanocomposite decreased under both quasi-static and dynamic loading due to the formation of more efficient carbon nanotube networks caused by the compression of the epoxy matrix [26,48].
Apart from this, it was identified that the practical use of graphene has been heavily restricted because current polymer processing technologies distribute graphene in a highly anisotropic fashion within polymer matrices, undermining some of the key advantages of using graphene as a filler material. The predicted percolation threshold for randomly aligned and uniformly dispersed 2-dimensional sheets such as graphene (aspect ratio ~ 4000) in a matrix is 0.01 % by volume [49]. Achieving this threshold is difficult, because strong van der Waals interactions between these sheets lead to aggregation [50,51]. In addition, most processing techniques, especially at the pilot and commercial scales result in highly anisotropic flows, which tend to align sheets along the direction of flow, thereby inhibiting the formation of a percolating network.
Achieving the theoretical percolation limit for scalable techniques has therefore been difficult. Because of the energy demand for removing solvents, and sometimes their potentially hazardous nature, melt processing is often chosen over solvent based mixing of filler and polymer, despite the increased viscosity of a melt. Dispersing high aspect ratio sheets isotropically in a melt of high viscosity is a major challenge.
An alternate method for creating a connected pathway for conductive particles is to make segregated composites. The conductive particles within segregated composites are only permitted to reside on the surfaces of the polymer matrix 7 particles. When consolidated into a monolith, these conductive particles become connected in a three-dimensional network, dramatically increasing the conductivity of the composite [52][53][54][55]. Sheets do not have to be distributed isotropically throughout a matrix to achieve percolation, overcoming a major limitation. This way of achieving three-dimensional connectivity of the particles also decreases the contact resistance between the particles [52]. Du et al. prepared multi-walled carbon nanotube (MWCNT)/high density polyethylene (HDPE) and graphene nanosheets (GNS)/HDPE composites with a segregated network structure by alcohol-assisted dispersion and hot-pressing. The electrical properties of the GNS/HDPE and MWCNT/HDPE composites were compared and it was found that the percolation threshold of the GNS/HDPE composites (1 % v/v) was much higher than that of the MWCNT/HDPE composites (0.15 % v/v) while the MWCNT/HDPE composite showed higher electrical conductivity than the GNS/HDPE composite at the same filler content. They concluded that, due to crimp, rolling and aggregation of the GNSs in the HDPE matrix, the two dimensional GNSs were not as effective as MWCNTs in forming conductive networks. Later, Hu et al. prepared graphene/polyethylene segregated composites using a two-step process. A combination of sonication and mechanical mixing was used to first coat the ultrahigh molecular weight polyethylene (UHMWPE) with graphene oxide (GO) sheets. The excess solvent was removed from the system and then the coated powders were added to a hydrazine solution and stirred at 95 º C to reduce the GO to graphene. All coated powders were compressively molded and hot pressed to form composite sheets. This two-step process was shown to effectively prevent aggregation, leading to composites exhibiting high electrical 8 conductivity at a very low percolation threshold (0.028 % v/v). To date, there have been no studies reported on the electro-mechanical behavior of either CNT-or graphene-reinforced polymers when subjected to dynamic loading.
14 Abstract A series of dynamic compressive experiments were performed to experimentally investigate the electrical response of multi-wall carbon nanotube (MWCNT) reinforced epoxy nanocomposites subjected to split Hopkinson pressure bar (SHPB) loading. Low-resistance CNT/epoxy specimens were fabricated using a combination of shear mixing and ultrasonication. Utilizing the carbon nanotube network within, the electrical resistance of the nanocomposite was monitored using a high-resolution fourpoint probe method during each compressive loading event. In addition, real-time deformation images were captured using high-speed photography. The percent change in resistance was correlated to both strain and real-time damage. The results were then compared to previous work conducted by the authors (quasi-static and drop-weight impact) in order to elucidate the strain rate sensitivity on the electrical behavior of the material. In addition, the percent change in conductivity was determined using a Taylor expansion model to investigate the electrical response based upon both dimensional change as well as resistivity change during mechanical loading within the elastic regime. Experimental findings indicate that the electrical resistance is a function of both the strain and deformation mechanisms induced by the loading. The bulk electrical resistance of the nanocomposites exhibited an overall decrease of 40-65% and 65-85% during quasi-static/drop-weight and SHPB experiments respectively.
Keywords: Electrical response, carbon nanotube/polymer composites, dynamic response, four-point probe method

Introduction
Carbon nanotubes (CNTs) are a promising addition to the future of developing novel materials capable of self-sensing and active response due to their extraordinary electrical conductivity and excellent transport properties. Unlike other smart materials, CNTs can provide both structural and functional capabilities simultaneously and have been used in many applications including actuation, sensing, and power generation [1][2][3][4]. Carbon nanotube-based polymeric nanocomposites have found new applications in various fields such as future spacecraft, anti-meteorite/anti-ballistic shields for satellites, anti-ballistic vests, explosion-proof blankets for aircraft cargo bays, and safety belts in industries [5,6].
Significant research has been performed to fundamentally understand the enhancement of mechanical properties due to CNT reinforcement of polymers [7,8].
Allaoui et al. [9] investigated the influence of multi-walled carbon nanotubes (MWCNTs) in a rubbery epoxy matrix, and found that the addition of up to 4 wt.% MWCNTs could lead to a significant increase in the strength and Young's modulus [10]. Given the practical potential applications of CNTs in electromechanical devices, specifically as piezoresistive sensors, the effect of mechanical deformation on the electrical properties of individual CNTs has been studied theoretically [11][12][13][14] and experimentally [11,12].
An increased amount of research has been conducted in studying the electrical response of CNT/polymer composites under mechanical loading. Alexopoulos et al. [13], Nofar et al. [14], and Gao et al. [15,16] have studied damage detection and health monitoring of composites reinforced with CNT-embedded glass fibers. as well as three-point bending tests on specimens with CNT fibers in the tensile region. Results indicated that CNT fibers provide unquestionable advantages for sensing and damage monitoring of non-conductive composites, when compared to the competitors, e.g. the embedded carbon fibers and a modified (doped) conductive network. Sensing ability for the investigated specimens with CNT fibers in the compressive region was also reported [13]. Gao et al. [15] studied the sensing of damage development in composites using CNT networks utilizing two sensing techniques: electrical resistance and acoustic emission. Resistance change and acoustic emission counts showed a bi-linear relation in detecting damage in quasi-static and cyclic experiments which can be used to give additional insight toward damage evolution. Thostenson et al. [17] performed tensile experiments on CNT/epoxy samples and demonstrated a highly linear relationship between the specimen deformation and the electrical resistance. This result suggests that CNT networks formed in an epoxy polymer matrix could be utilized as highly sensitive sensors for detecting the evolution of damage in advanced polymer-based composites [17]. Later, Lim et al. [18] experimentally investigated the mechanical and electrical response of CNT-based fabric composites to Hopkinson bar loading further demonstrating the effectiveness of a percolating carbon nanotube network being capable of sensing damage caused by impact. In their previous work, the authors experimentally investigated the electrical response of multi-walled carbon nanotube reinforced nanocomposites under quasi-static and dynamic loading. The results indicated that the electrical resistance of the nanocomposite decreased under both quasi-static and dynamic loading due to the formation of more efficient carbon nanotube networks caused by the compression of the epoxy matrix [18,19].
The aforementioned studies revealed that when properly dispersed within a given matrix, an internal sensory network can be formed and utilized to detect important information such as strain and damage within the material. To further this investigation, it is crucial to understand the electrical response of the nanocomposites under dynamic loading conditions. The present paper experimentally investigates the electrical response of MWCNT reinforced nanocomposites subjected to dynamic split Hopkinson pressure bar loading. A modified four-point probe method, using line and face contacts rather than point contacts, was implemented to measure more consistent and accurate results during mechanical loading [19]. Fabricated nanocomposites were loaded using a split Hopkinson pressure bar equipped with solid steel bars. The history between the electrical resistance change, the mechanical loading, and the highspeed deformation photography are correlated to characterize the electrical response of CNT reinforced epoxy under compressive loading. For the sake of completeness, results corresponding to static and drop weight impact experiments previously performed by the authors will be presented for comparison purposes [19].
It was observed that the overall bulk resistance decreased significantly when subjected to SHPB loading, demonstrating a similar electrical response as seen during quasi-static and drop weight loading. A 65 -85% decrease of resistance was observed. This is due to a more efficient carbon nanotube network forming under compressive loading caused by the compression of the epoxy matrix.

Material Fabrication
Due to the simplicity of casting and low curing temperature, a two-part epoxy, consisting of bisphenol-A resin (Buehler Epothin 20-8140-032) and an epoxy hardener (Buehler Epothin 20-8142-016) with a mixing ratio of 50g/18g, was chosen as a polymeric matrix. CNTs used for this study were multi-wall carbon nanotubes The general procedure of material fabrication is shown in Fig. 1. High surface energy of carbon nanotubes causes the agglomeration of nanotubes when dispersed, adversely affecting the electrical transport properties of the material [20]. In order to address the agglomeration issue and effectively disperse the CNTs, the present work implemented high-intensity ultrasonication and high-speed shear mixing. Premeasured amounts of Epothin Part-A Resin and carbon nanotubes were mechanically stirred for 5 minutes in a copper beaker. The mixture was then placed into a shear mixer (Ika Werke RW 16 Basic) outfitted with a 3-blade propeller stirrer (R1381 Propeller stirrer) and shear-mixed at 600 RPM for 30 minutes. Following shear mixing, the ultrasonication process was applied for one hour on pulse mode, 4.5 sec on 9 sec off, at 100 kHz (Sonics & Materials Inc. VCX750). The mixture was then placed into a vacuum chamber to remove any trapped air bubbles generated during the mechanical mixing process [21]. A pre-measured amount of Part B epoxy hardener, in a separate container, was also placed inside the vacuum chamber. Both solutions were placed under vacuum for 1 hour. Once all air was removed from both solutions, they were combined and mechanically stirred for 2 minutes. The mixture was once again placed back into the vacuum chamber for 5 minutes. Finally, the CNT/epoxy solution was slowly poured into pre-manufactured wax molds and allowed to cure for 3 days under ambient conditions. It is critically important to control the temperature of the mixture during the sonication process for the quality of the fabricated specimens. The sonication process generates substantial heat that may damage CNTs and deteriorate the electrical properties of the final composite [22]. Moreover, too much heat could cause the epoxy to reach the flash point. To control the temperature of the mixture during sonication, a cooling apparatus was designed and built as shown in

20
the mixture temperature was maintained in between 18°C -30°C, depending on the sonication duration. A percolation study on the material was previously performed by the authors which resulted in an electrical percolation threshold occurring between 0.1 and 0.2 wt.% of CNTs for the particular CNTs and polymer used [19]. Nanotube concentration greater than 0.2 wt.% does not provide better electrical conduction.
Therefore, the concentration of CNTs was set to 0.2 wt.% for all experiments in the present study.

Electrical Characterization
In order to effectively capture the change in electrical resistance of the cylindrical specimen, a novel approach utilizing the four-point probe method was implemented. resistance of the individual specimens varies slightly. Therefore, the initial resistance of each specimen served as the baseline for each experiment. Based on previous studies [19], this method better provides the means to detect changes in the resistance caused by strain and damage mechanisms in the material as compared to the classical four-point probe method. Since the current uniformly flows through the cross sectional area, the measured resistance is an estimation of the bulk resistance of the inner section. By using this average voltage measurement technique, more consistent and accurate results were obtained during a wide range of mechanical loading schemes and consequent specimen deformations.
A sketch of the experimental setup used to capture the electrical resistance change during the dynamic compressive experiments is shown in Fig. 4 drop between the two inner probes was measured by a differential amplifier (Tektonix ADA 400A) and recorded by a digital oscilloscope (Tektronix TDS 3014).

Split Hopkinson Pressure Bar Loading
A split Hopkinson bar, equipped with a solid maraging steel incident and transmission bar, was used to apply a dynamic load to the specimens. A sketch of the SHPB device and typical pulse profiles are given in Fig. 4(a) and (b), respectively.
The incident and transmission bar were 12.7 mm in diameter and measured 1220 mm in length. For the SHPB dynamic loading, the incident pulse length is related to the projectile length. To induce sufficient strain in the specimen, a 355 mm striker was used. A layer of electrical tape was placed at the ends of the incident and transmission 23 bars to insulate the specimen from the loading apparatus. A high-speed camera (Photron SA1) captured high-speed deformation images at a frame rate of 100,000 fps.
The high-speed images were used to calculate the strain history of the specimen under dynamic loading as well as capture major damage mechanisms. The stress within the specimen monotonically increases to ~75 MPa at 4% engineering strain and then gradually decreases. As the compressive strain increases to ~12%, the electrical resistance decreases ~56%. The electrical resistance of the specimen is

Drop Weight Impact
Impact experiments, using a drop weight apparatus, were performed to study the electrical response of the nanocomposites under intermediate strain rate loadings (10 1 /s). A typical electrical response and the real-time deformation images of a rectangular CNT/epoxy nanocomposite under high mass, low velocity impact loading are shown in Fig. 6. In general, the response of the nanocomposites under drop weight loading is more brittle than under quasi-static loading due to the increased strain rate.
The overall resistance change follows a similar trend as under quasi-static loading. A strain threshold, lasting approximately 250 µs, is evident. Once surpassed, the resistance monotonically decreases as the specimen deforms uniformly. At 600 µs, it can be observed that the right and left sides of the specimen demonstrate an expansion.
As the material reaches this critical strain value, the decrease in electrical resistance quickly arrests and shows very little change due to the combination of the material compression and spreading. Damage then quickly initiates and propagates throughout the specimen resulting in a sharp increase in resistance. Overall, the resistance of the material decreases approximately 65% during drop weight impact loading.

Split Hopkinson Pressure Loading
To investigate the electrical response of the nanocomposites under higher strain rates, a SHPB apparatus was utilized to load the specimens. A typical result of the actual resistance, as a function of engineering strain during a SHPB experiment, is shown in Fig. 7. The initial resistance of the inner section of the specimen can be seen to be approximately 13.5 kΩ prior to loading. As the specimen undergoes dynamic compression at a strain rate of approximately 2000/s, the electrical resistance change is inversely proportional to the change in strain. As the stress of the specimen monotonically increases to 160 MPa at 8% engineering strain, a small decrease in electrical resistance from 13.5 kΩ to 11 kΩ is observed. Once the material begins to yield, the resistance begins to decrease at a higher rate up until the strain reaches 12 % engineering strain. As the compressive strain increases, the electrical resistance begins to decrease at a slower rate due to the spreading of the material (larger cross-sectional area). The electrical response of the nanocomposites under SHPB loading shows a 27 similar response to that seen during quasi-static compression as well as drop weight impact compression [19]. When considering the negligible change in CNT geometry during compression, the resistance change is primarily attributed to the rearrangement of the CNT networks present within the material, thus causing new tube-to-tube contacts and a decrease in electrical resistance.
A typical electrical response along with the real-time deformation images of a CNT/epoxy nanocomposite subjected to SHPB loading are shown in Fig. 8. The time frames used in the loading event are chosen in a manner such that they can be correlated to the time at which certain deformation mechanisms were first observed. A schematic representing the deformed configurations of the specimen and damage mechanisms induced by mechanical compression is shown in Fig. 9. During the first 60 µs, the specimen undergoes a uniform compression. This axial compression, shown in Fig. 9 (I), causes more efficient electrical pathways by decreasing the inter-tube gaps between the CNTs present within the matrix and increasing the number of new  30 attributed to the variability and complexity of the CNT networks present within each specimen. As the material compresses, new tube-to-tube contacts were made with nonlinear deformation of the epoxy matrix, which increases the electrical conductivity of the embedded CNT network. As seen in Fig. 7, specimens subjected to dynamic compressive loading using a split Hopkinson pressure bar apparatus showed a 65 -85% decrease in material resistance during compression. Hence there exists rate sensitivity of electrical response of embedded CNT network with matrix deformation.

CNT Sensitivity in the Elastic Region
To better understand the electrical response of the material, the change in electrical conductivity of the nanocomposite was determined. To calculate the change in conductivity of the material from the obtained results, it was assumed that the conductivity c of the nanocomposite with a length L, cross-sectional area A, and a resistance R is given by c= l⁄RA. Since large changes in resistance were measured during experimentation, higher order terms are necessary in order to account for the large changes in material conductivity. Taylor Expansion for multiple variables was applied.
Replacing dl = Δl, dR = ΔR, dA = ΔA, Eq. (1) can be written as The change in electrical conductivity, Δc/c, is then The change in area is related to the change in length by dA/A = -2ʋε where ʋ is the Poison's ratio of the material. Thus, Eq. (4)

Conclusions
The obtained results provide further insights on the electrical behaviors of Due to the exceptional electrical properties of graphene, several researchers in the past have also studied the utilization of graphene as an electrically conductive additive in composites [7,18,[22][23]. The electrical conductivity of graphene-based composites has been studied theoretically [21] as well as experimentally [18,23,[25][26].

Material Fabrication
The graphene platelets used in this study were xGnP TM Nanoplatelets (XG Sciences). These unique nanoparticles consist of short stacks of one or more graphene sheets having a lateral dimension of ~25 µm. The edges of these sheets are sites for functionalization, which may help facilitate bonding within the polymer matrix. An SEM image of these platelets is shown in Fig. 1. The specific polymeric matrix chosen for this study was polystyrene (PS) (Crystal PS 1300) purchased from Styrolution. The PS had an average molecular weight of about 265,000 g/mol. Graphene's strong intrinsic van der Waals forces of attraction between sheets and high surface area make graphene very difficult to disperse uniformly within polymer materials [20]. In order to disperse the platelets throughout the PS matrix, a solution mixing process was 44 employed [23]. The general procedure used to disperse the graphene platelets is shown in Fig. 2 The resulting composite was then filtered and dried in an oven at ~ 80 ºC for ~ 18 h.
Finally, the dried graphene-PS composites were hot-pressed using a heated steel mold (~ 190 ºC) and a hydraulic press. face. The channels were used to implement a modified four-point probe method [29] in order to effectively measure the change in electrical resistance of the specimen during loading. The loading is exerted in the longitudinal direction along the length of the specimen. The left face, right face, and the two inner channels of the specimen served as four probes to obtain a four point probe measurement. All four probes were

Electrical Characterization
In order to effectively capture the change in electrical resistance of the cylindrical specimen, a novel approach previously developed by the authors utilizing the four-point probe method was implemented [29]. The four probes consisted of the left face, right face, and the two inner channels. To allow a constant current flow through the entire bulk of the specimen, a constant current was supplied through the right and left faces of the specimen. The two inner channels served as the two peripheral electrodes that measure the voltage drop across the middle section of the specimen. The electrical resistance of the middle section can be easily determined from the input current and the voltage drop across the inner probes. As the specimen underwent deformation, the instantaneous resistance of the middle section changed.
Percent change in the resistance was calculated for each experiment. Due to the complex dispersion pattern of graphene inside, the initial resistance of the individual specimens varies slightly. Therefore, the initial resistance of each specimen served as the baseline for each experiment. Based on previous studies [29], this method better Since the current uniformly flows through the cross sectional area, the measured resistance is an estimation of the bulk resistance of the inner section. By using this average voltage measurement technique, more consistent and accurate results were obtained during a wide range of mechanical loading schemes and consequent specimen deformations.

Quasi-static Electro-Mechanical Characterization
The quasi-static loading was implemented by a screw-driven testing machine.
A modified four-point probe method was utilized to measure the resistance change during the compression tests [29]. The experimental setup used to capture the resistance change of the composites under quasi-static loading is shown in Fig. 4. A constant current source was used to supply a DC current flow through the specimen.
The graphene/PS specimen was sandwiched between two aluminum plates to establish uniform current flow through the specimen during the compressive loading. Silver paint was applied to the top and bottom of each specimen to minimize the contact resistance between the specimen and the plates. Each loading head was insulated from the electrical measurement system. Two electrometers were used to measure the voltage at each of the two individual inner probe rings. The difference between the two voltage readings, which corresponds to the voltage drop across the two inner probes, was measured using a digital multimeter and recorded using a LabView system.

Dynamic Electro-Mechanical Characterization
A modified split Hopkinson (Kolsky) pressure bar apparatus, capable of simultaneous mechanical and electrical characterization, was developed and implemented to investigate the dynamic electro-mechanical response of the graphene-PS composites.
A typical SHPB consists of a striker bar, a solid incident bar and a solid transmission bar. The striker bar is propelled using an air-operated gun. A pulse shaper is commonly placed at the impact end of the incident bar with a thin layer of lubricant to improve force equilibrium conditions at the specimen-bar interfaces. The theoretical details of SHPB can be obtained from Kolsky [34]. The specimen is sandwiched between the incident bar and the transmission bar. A lubricant is applied between the specimen and the bar interfaces to minimize friction.
When the striker bar impacts the incident bar, an elastic compressive stress pulse, referred to as the incident pulse, is generated and then propagates along the incident bar towards the specimen. When the incident pulse reaches the specimen, part of it reflects back into the incident bar (reflected pulse) in the form of a tensile pulse due to the impedance mismatch at the bar-specimen interface and the remaining pulse is transmitted (transmission pulse) to the transmission bar. Axial strain gages mounted on the surfaces of the incident and transmission bar provide time-resolved measures of the elastic strain pulses in the bars. The amplitude and length of the incident pulse is related to the projectile velocity and projectile length which allows for variation in achievable strain rates.
Using one-dimensional wave theory, the engineering stress and engineering strain in the specimen can be determined from the reflected and transmitted strain pulses respectively, as given in Eqs. 1 and 2.
The above equations were suitably modified to obtain the true stress and true strain in the specimen. The expressions for the forces at the specimen incident bar interface and at the specimen transmission bar interface are given in equations Eqs. 3 and 4 respectively.  Fig. 6 shows typical pulses obtained from the strain gages for the two cases when the incident bar is in contact with the transmission bar without any specimen in between. Since there is no impedance mismatch at the bars interface, the entire incident pulse is transmitted to the transmission bar. It can be clearly seen that there is no effect on the pulses when a current of 1 mA was supplied through the bars.

52
A striker length of 406 mm was used in all experiments to achieve large strains in the specimen. A high-speed digital camera (Photron SA1) was used to capture the real time deformation of the specimen at a frame rate of 100,000 fps.

Quasi-Static Compressive Response
A typical electro-mechanical response of a 5 vol. % graphene-PS composite under compressive loading is shown in Fig. 7. During the quasi-static compression, the stress in the specimen monotonically increases to 47 MPa at 5 % strain and then gradually decreases. Initially, no significant change in resistance is observed up until ~ 1 % strain. Taking the initial resistance as a baseline, the percent change in electrical resistance increases proportionally with strain. Since the electrical resistance of the matrix material is very high, the graphene particles exclusively conduct the electrical current within the material. Due to the brittle nature of the PS matrix, small micro- As shown in Fig. 9, evidence of micro-cracks located primarily around small agglomerates of FLG. The small agglomerates of the graphene sheets appear to serve   properties when graphene is used as a filler material within various polymers [8][9][10][11].
Generally, the enhancement of strength and modulus is attributed to high aspect ratio       The dynamic true stress-strain curves for pristine PS and PS containing 5 vol. % graphene is shown in Fig. 15. The high strain-rate yield stresses were much higher than the quasi-static ones for both pristine PS as well as graphene reinforced PS.
Similar to static loading, the graphene-PS composites demonstrated a reduced strength

Conclusions
The present paper describes the electro-mechanical response of graphene reinforced polystyrene composites under quasi-static and dynamic compressive loading. Graphene-PS composites with low resistance were fabricated using a solution mixing approach followed by hot-pressing. A modified four-point probe method, using line and face contacts rather than point contacts, was implemented to accurately monitor the bulk electrical resistance of the composites. Moreover, a modified split Hopkinson (Kolsky) pressure bar apparatus, capable of simultaneous mechanical and electrical characterization, was developed and implemented to investigate the dynamic electro-mechanical response of the composites. In addition to measuring the change in electrical resistance as well as the dynamic constitutive behavior, real-time damage was captured using high-speed photography. The real-time damage was correlated to both stress-strain and percent change in resistance profiles. Due to a high [20] Wang L, Hong J, Chen G.

Introduction
Electrical conductivity in polymers that are traditionally insulating can be achieved by dispersing conducting particles within the non-conducting matrix. The predicted percolation threshold for randomly aligned and uniformly dispersed 2dimensional sheets such as graphene (aspect ratio ~ 4000) in a matrix is 0.01 % by volume [1]. Achieving this threshold is difficult, because strong van der Waals interactions between these sheets lead to aggregation [2][3][4]. In addition, most processing techniques, especially at the pilot and commercial scales, result in highly anisotropic flows, which tend to align sheets along the direction of flow and inhibit the formation of a percolating network. Achieving the theoretical percolation limit for scalable techniques has therefore been difficult. Because of the energy demand for

Material and methods
The few-layer graphene flakes used in this study were xGnP TM Nanoplatelets (XG Sciences, USA). These nanoparticles consist of short stacks of graphene layers having a lateral dimension of ~ 25 µm and a thickness of ~ 6 nm. The polymeric material chosen for this study was polystyrene (Crystal PS 1300, average molecular weight of 121,000 g/mol) purchased from Styrolution, USA. The PS pellets (~ 2 mm) used were elliptical prisms with a total surface area of 1.03 ± 0.01 cm2.
A two-step process was utilized to produce the FLG/PS segregated composites.
First, the desired amount of graphene platelets were measured and added to 7 g of dry 72 PS pellets. The FLG spontaneously adheres to the dry polymer particles by physical forces, which may be van der Waals forces or electrostatic attraction associated with surface charges. Figure 1 shows PS pellets coated with various amounts of FLG using this dry coating process. This coating process works well for FLG loadings below 0.2 % v/v. However, at higher FLG loadings, this dry method leaves behind excess FLG because the charge on the pellets is neutralized after the initial coating.
To provide a means of temporarily attaching larger quantities of the FLG to the surface of the PS, an additional step is implemented during the fabrication procedure, shown in Fig. 2. The PS is first soaked in a methanol bath and the excess methanol is drained from the PS pellets. FLG is added, and the mixture is then shaken vigorously, creating a dense coating of graphene on each PS pellet. The methanol temporarily moistens the polymer pellets forming small liquid bridges. The capillary pressure

Results and discussion
As seen in Fig

Conclusions
We demonstrate a simple, inexpensive and commercially viable technique that can be used to disperse conductive sheet-like particles, such as graphene, into a highly organized pattern within polymeric materials on either the micro-or macro-scale.
Utilizing capillary interactions between polymeric particles and few-layer graphene particles, liquid bridges on the surface of a polymeric material allows for coating of graphene onto the polymer surfaces. By precisely controlling the temperature and pressure during the melt compression process, highly conductive composites are formed using very low loadings of graphene particles. Applications for such composites could include sensing devices, coloring mechanisms, as well as barrier mechanisms.
Significant research has shown that carbon-based polymer nanocomposites have proven to demonstrate remarkable physical and mechanical properties by incorporating very small amounts of filler material [13][14][15][16][17][18]. One of the most compelling features of polymer nanocomposites is the ability to create a new class of materials with attributes that come both from the filler and the matrix. Having the ability to manipulate the degree and nature of the dispersion is key to the development of these types of novel composites [19]. Many studies have documented enhancement of properties such as stiffness and strength, thermal stability, electrical and thermal conductivities, dielectric performance and gas barrier properties of polymer composites with the incorporation of fillers [20][21][22][23][24][25].
Owing to its extraordinary mechanical and physical properties, graphene appears to be a very attractive filler material for the next generation of smart materials in batteries, supercapacitors, fuel cells, photovoltaic devices, sensing platforms and other devices [13,14]. Although significant research has been performed to develop strategies to effectively incorporate nanoparticles into polymers, ability to control the dispersion and location of graphene-based fillers to fully exploit their intrinsic properties remains a challenge [26][27][28][29].
Along with the aspect ratio and the surface-to-volume ratio, the distribution of the filler in a polymer matrix has been shown to directly correlate with its effectiveness in improving material properties such as mechanical strength, electrical and thermal conductivity, and impermeability [19]. The critical content of a filler material that characterizes a drastic increase in composite properties, such as electrical conductivity, is commonly termed the percolation threshold. From a physical standpoint, the predicted percolation threshold for randomly aligned and uniformly dispersed 2-dimensional sheets such as graphite nanoplatelets (aspect ratio ~ 4000) in a matrix is 0.01 % by volume [15]. Achieving this threshold is difficult, because strong van der Waals interactions between these sheets lead to aggregation, especially in the face-to-face configuration [4,16,17]. In addition, most processing techniques, especially at the pilot and commercial scales, result in highly anisotropic flows, which tend to align sheets along the direction of flow and inhibit the formation of a percolating network. Achieving the theoretical percolation limit for scalable techniques has therefore been difficult. Because of the energy demand for removing solvents, and sometimes their potentially hazardous nature, melt processing is often chosen over solvent based mixing of filler and polymer, despite the increased viscosity of a melt. Dispersing high aspect ratio sheets isotropically in a melt of high viscosity is a major challenge.
An alternate method for creating a connected pathway for conductive particles is to make segregated composites. The conductive particles within segregated composites are specially localized on the surfaces of the polymer matrix particles.
When consolidated into a monolith, these conductive particles form a percolating three-dimensional network that dramatically increases the conductivity of the composite [30][31][32][33][34][35]. Sheets do not have to be distributed isotropically throughout a matrix to achieve percolation, overcoming a major limitation. This way of achieving three-dimensional connectivity of the particles also decreases the contact resistance 83 between the particles [30]. Du et al. [30] prepared multi-walled carbon nanotube (MWCNT)/high density polyethylene (HDPE) and graphene nanosheets (GNS)/HDPE) composites with a segregated network structure by alcohol-assisted dispersion and hot-pressing. The electrical properties of the GNS/HDPE and MWCNT/HDPE composites were compared and it was found that the percolation threshold of the GNS/HDPE composites (1 % v/v) was much higher than that of the MWCNT/HDPE composites (0.15 % v/v) while the MWCNT/HDPE composite showed higher electrical conductivity than the GNS/HDPE composite at the same filler content. They concluded that, due to crimp, rolling and aggregation of the GNSs in the HDPE matrix, the two-dimensional GNSs were not as effective as MWCNTs in forming conductive networks. Later, Hu et al. [32] prepared graphene/polyethylene segregated composites using a two-step process. A combination of sonication and mechanical mixing was used to first coat the ultrahigh molecular weight polyethylene (UHMWPE) with graphene oxide (GO) sheets. The excess solvent was removed from the system and then the coated powders were added to a hydrazine solution and stirred at 95 ºC to reduce the GO to graphene. All coated powders were compressively molded and hot pressed to form composite sheets. This two-step process was shown to effectively prevent aggregation, leading to composites exhibiting high electrical conductivity at a very low percolation threshold (0.028 % v/v). In their previous work, the authors demonstrated a simple, inexpensive and commercially viable technique that can be used to disperse conductive sheet-like particles, such as graphene, into a highly organized pattern within polymeric materials [31]. By utilizing capillary interactions between methanol, polystyrene (PS) particles and few-layer graphene flakes, highly conductive segregated composites were produced. The conducting pathways provided by the graphene located at the particle surfaces through contact of the bounding surfaces allow percolation at a loading of less than 0.01% by volume.
These studies revealed that highly conductive composites can be created when graphene is segregated into organized networks throughout a matrix material.
Although the highly segregated networks provide excellent transport properties throughout the composite, they inevitably result in poor mechanical strength, since fracture can occur easily by delamination along the continuous segregated graphene phase. Since most multi-functional materials are required to provide excellent transport properties while maintaining sufficient mechanical strength, alternative methods of distributing graphene need to be developed. Despite recent progresses on the electrical characterization of graphene-based segregated composites, no results have been published yet regarding the combined electro-mechanical behavior of these highly conductive materials. In this work, a novel capillary-driven, particle-level templating technique was utilized to distribute graphite nanoplatelets (GNPs) into specially constructed architectures throughout a polystyrene (PS) matrix to form multi-functional composites with tailored electro-mechanical properties. By precisely controlling the temperature and pressure during a melt compression process, highly conductive composites were formed using very low loadings of graphene particles. To

Material
The graphite nanoplatelets used in this study were xGnP TM Nanoplatelets (XG Sciences, USA). These nanoparticles consist of short stacks of graphene layers having a lateral dimension of ~ 25 µm and a thickness of ~ 6 nm. This thickness corresponds to approximately 18 graphene layers at a typical graphite interlayer spacing. It has been proposed that materials of this thickness (> 10 layers) be referred to as exfoliated graphite, or graphite nanoplatelets for scientific classification [8]. The same materials are sometimes marketed by suppliers as "graphene nanoplatelets". The polymeric material chosen for this study was polystyrene (Crystal PS 1300, average molecular weight of 121,000 g/mol) purchased from Styrolution, USA. The PS pellets used were elliptical prisms with an average diameter of 2.76 mm and a length of 3.21 mm.

Particle Templated Composites
A two-step process was utilized to produce the GNP/PS segregated composites [9]. For composites consisting of less than 0.2 % v/v, the desired amount of graphene platelets were measured and added directly to 7 g of dry PS pellets. The GNP spontaneously adheres to the dry polymer particles by physical forces, which may be van der Waals forces or electrostatic attraction associated with surface charges. This coating process works well for GNP loadings below 0.2 % v/v. However, at higher GNP loadings, this dry method leaves behind excess GNP because the charge on the pellets is neutralized after the initial coating.
To provide a means of temporarily attaching larger quantities of the GNP to the surface of the PS, an additional step is implemented during the fabrication procedure as shown in Fig. 1. For GNP loadings greater than 0.2 % v/v, the PS is first soaked in a methanol bath. The excess methanol is drained from the PS pellets. GNP is added, and the mixture is then shaken vigorously, creating a dense coating of graphene on each PS pellet. The methanol temporarily moistens the polymer pellets forming small liquid bridges between the GNP and the pellet surface. The capillary pressure created through these bridges allows the GNPs to stick easily to the surface of the pellets. During the subsequent hot melt pressing, the temperature and mold pressure are precisely controlled allowing the pellets to be consolidated into a 87 monolith while maintaining boundaries. The methanol evaporates during the molding cycle. In our experiments, a stainless steel mold consisting of a lower base and a plunger was heated to 125 ºC. The graphene flakes coated PS was placed inside the cavity of the lower base and the plunger was placed on top. The temperature of both the plunger and the base mold was maintained for 20 min at which point it was hotpressed at 45 kN using a hydraulic press. By precisely controlling the temperature and pressure during a melt compression process, highly conductive composites were formed. This method of distributing graphene within a matrix overcomes the need to disperse the sheet-like conducting fillers isotropically within the polymer, and can be scaled up easily.

Particle Templated Composites with Shear Manipulation
Modified particle-templated composites were fabricated by incorporating a shearing technique during the melt compression process. Following the same coating process as discussed earlier, the graphene coated pellets were placed inside a modified steel mold, which was equipped with guide pins to ensure that the base remained stationary. The plunger was then placed on top of the material and heated to 160 ºC while the lower base mold was heated to 125 ºC and maintained for 20 min. Next, 20 MPa was applied to the plunger and then rotated to various predetermined angles.
Once the desired rotation was achieved, 45 MPa was applied and held for 5 minutes.
All shear-modified composites were fabricated with 0.3 % v/v graphene platelets. A schematic of the compression molding process used to produce both types of segregated composites is shown in Fig. 2(a) and Fig.2(b). By applying such a strain in the azimuthal direction on the top surface of the material, as shown in Fig 2(b), a gradient of graphene organization/orientation in the axial direction is formed which results in a composite possessing unique properties.

Electrical Characterization
Electrical conductivity measurements were made on the GNP/PS composites

Mechanical Characterization
A series of 3 point bend experiments were carried out to investigate the influence of graphene content on the flexural properties of the composites. A screw-driven testing machine (Instron 3345) was implemented to load the specimens in a three point bending configuration. Specimens were cut into 5 x 6 x 38 mm rectangular prisms. A support span of 30 mm was used and the loading was applied at a rate of 0.1 mm/min.

Particle Templated Composites
As seen in Fig. 1     The coupled electro-mechanical behavior of the GNP-PS organized particle templated composite, when loaded parallel to the pressing direction, is shown in Fig.   7. The flexural strength and electrical conductivity is normalized with respect to the flexural strength (σ 0 ) and electrical conductivity (k 0 ) of the pristine PS particle templated composite (0 % v/v GNP), respectively. It can be seen that the highly segregated GNP network, although very efficient for electron transfer, causes a significant decrease in flexural strength. While the conducting pathways provided by the graphene, located at the particle interfaces of the PS, allow percolation at a graphene loading less than 0.01 % v/v GNP, they also cause the flexural strength of the composite to decrease by ~ 40 %. As the GNP loading is further increased, the

Shear-modified Particle Templated Composites
Images of a 0.3 % v/v GNP/PS shear modified specimen exhibiting a graphene network that is functionally graded in the axial direction is shown in Fig. 8. The top surface of the composite exhibits a chaotic and disorganized pattern of GNP while the bottom maintains a highly organized segregated structure of GNP.

95
The effect of azimuthal strain on the top surface on the electrical conductivity of the shear-modified GNP/PS composite is shown in Fig. 9. The electrical conductivity decreased from ~ 3 S⋅m -1 to ~ 4x10 -2 S⋅m -1 when the plunger was rotated 90 ○ during the compression process. Although, the electrical conductivity decreased by two orders of magnitude, the value of 4x10 -2 S⋅m -1 is still very high and acceptable for many applications. The decrease in electrical conductivity can be attributed to the partial disruption of the GNP networks within the polymer, as shown in Fig. 8 (c).
Further rotation of the plunger resulted in only a slight decrease in conductivity.   mechanical properties of composite materials and therefore can be used to intelligently optimize materials for specific target applications.

Summary
We demonstrate a simple, inexpensive, and commercially viable technique that can be used to disperse conductive 2D (sheet-like) materials, such as graphene, into specifically constructed hybrid architectures within polymeric materials on either the micro-or macro-scale. Utilizing capillary interactions between polymeric particles and few-layer graphene particles, liquid bridges on the surface of the polymeric material allows for the coating of graphene onto the polymer surfaces. By precisely controlling the temperature and pressure during the melt compression process, highly conductive composites are formed using very low loadings of graphene particles.
Since the graphene particles are localized at the boundaries between the polymer matrix particles, the composite exhibited poor mechanical strength. To improve the mechanical properties of the composite, a controlled amount of rotary shear was

Introduction
Advancements in technology have and continue to drive the evolution of composite materials, making them lighter, stronger and more advanced for use in a vast range of industries [1][2][3]. In recent decades, polymer nanocomposites have shown tremendous potential in becoming the next generation high performance materials that provide multifunctional capabilities [4][5][6][7][8][9][10].
Significant research has shown that, in particular, carbon-based nanocomposites have proven to demonstrate remarkable physical and mechanical properties by incorporating very small amounts of filler material [11][12][13][14][15][16][17][18][19][20][21][22][23]. One of the most compelling attributes of polymer nanocomposites is the ability to create a new class of materials with properties that come both from the filler and the matrix. Having the ability to manipulate the degree and nature of the dispersion is key to the development of these types of novel composites [24]. Many studies have documented enhancement of properties such as stiffness and strength, thermal stability, electrical and thermal conductivities, dielectric performance and gas barrier properties of polymer composites with the incorporation of fillers [25][26][27][28][29][30].
Although significant research has been performed to develop strategies to effectively incorporate nano-particles into polymers, most techniques rely on solvent based mixing of filler and polymer to disperse particles at the micro-and nano-scale.
Due to high cost and the potentially hazardous nature of solvents, such techniques fail to be commercially viable and therefore limit the implementation of this novel technology into many of the potential applications. Despite recent progresses in the development of more advanced polymer composites by graphene reinforcement, a very limited number of commercially viable fabrication techniques have shown efficient incorporation of graphene-based materials into host polymers [31].
As one of the most important thermoplastic elastomers (TPE), styrene/butadiene/styrene block copolymer (SBS) merges good balance of mechanical property along with favorable processability and recyclability, which can be used in various fields, such as modifiers and adhesives [32]. Moreover, SBS can serve as a host polymer with high flexibility to accommodate various conductive fillers to produce electrically conductive composites [33]. SBS electroconductive composites have been successfully prepared using multi-walled carbon nanotubes (MWCNTs) [34], carbon black [35], as well as graphene [32,36]. Although significant research has been conducted in producing flexible electroconductive materials, the fabrication techniques used still rely on the use of solvents, thus preventing these types of composites from being manufactured on the commercial scale.

Material
The graphite nanoplatelets used in this study were xGnP TM Nanoplatelets (XG Sciences, USA). These nanoplatelets consist of short stacks of graphene layers having a lateral dimension of ~ 25 µm and a thickness of ~ 6 nm. This thickness corresponds to approximately 18 graphene layers at a typical graphite interlayer spacing. It has been proposed that materials of this thickness (> 10 layers) be referred to as exfoliated graphite, or graphite nanoplatelets [39] for scientific classification. These same materials are sometimes marketed by suppliers as "graphene nanoplatelets". The polymeric material chosen for this study was a styrene/butadiene thermoplastic elastomer (Asaprene T-439) acquired from Asahi Kasei Chemical Corp., Japan. The copolymer contains a total polystyrene weight fraction of 0.45 and came in pellet form (~ 3 mm in diameter).

Preparation of SBS-GNP Templated Composites
The SBS-GNP templated composites were prepared using the fixed-angle rotary shear technique [37,38]. This technique, shown in Fig. 1, consists of utilizing capillary interactions between the conductive filler material, such as graphene, and a polymeric material. In this case, the technique is adapted to temporarily attach GNP to the surfaces of the SBS pellets during the melt compression process.
The GNP coated SBS pellets were placed into a stainless steel mold consisting of a lower base and a plunger was heated to 140 ºC. The mold was equipped with guide pins to ensure that the base remained stationary. The plunger was then placed on top of the material and heated to 160 ºC and both temperatures were maintained for 20 min. For GNP/SBS composites exhibiting a fully organized segregation of GNP, the material was hot-pressed at 45 kN using a hydraulic press. For GNP/SBS composites exhibiting a functionally graded organization of GNP, the coated pellets were placed inside the mold cavity and pressed to 5 metric tons and rotated to various predetermined angles. Once the desired rotation was achieved, the composite was pressed to 10 metric tons and held for 5 minutes. Finally, the mold is placed into a cooling bath. By applying such a strain in the azimuthal direction on the top surface of the material, a gradient of graphene organization/orientation in the axial direction is formed which results in a composite possessing unique properties. A schematic of the

Electrical Characterization of SBS/GNP Templated Composites
Electrical conductivity measurements were made on the GNP/PS composites

Mechanical Characterization of SBS/GNP Templated Composites
A series of uniaxial tensile experiments were carried out to investigate the influence of graphene content on the tensile properties of the composites. A screwdriven testing machine (Instron 3366) was utilized to load the specimens in uniaxial tension. Specimens were cut into 40 x 12 x 5 mm rectangular prisms and the loading was applied at a rate of 50 mm/min.

Experimental Results & Discussion
As seen in Fig. 3, the composite (with 0.3 % v/v GNP) has a foam-like structure in which the dark wall-like structures are GNPs while the lighter domains are the SBS. The percolation threshold observed is comparable to the percolation achieved using RGO/HO-SBS [24]. It is important to note that no harsh solvents were needed to fabricate the particle template composites, while still achieving comparable conductivities. Additionally, the melt compression process can easily be scaled up for future commercial application where a solution blending approach cannot. The normalized electro-mechanical behavior of the organized GNP/SBS material is shown in Fig. 6. It can be seen that the highly segregated GNP network, although efficient for electron transfer, causes a significant decrease in elongation at Vol.% Graphene break. While the conducting pathways provided by the graphene, located at the pellet interfaces of the SBS, allow percolation at the graphene loading less than 0.15 % v/v GNP, they also cause the total elongation achieved by the composite to decrease ~ 48 %. As the GNP loading is further increased, the electrical efficiency of the networks continues to increase while the maximum elongation is decreased.

Shear-modified SBS/GNP Templated Composites
The effect of azimuthal strain on the top surface on the electrical conductivity of the shear-modified GNP/SBS composite is shown in Fig. 7. The electrical conductivity decreased from ~ 2 S⋅m -1 to ~ 2x10 -2 S⋅m -1 when the plunger was rotated 15° during the compression process. The decrease in electrical conductivity can be attributed to the partial disruption of the GNP networks within the polymer. Further rotation of the plunger resulted in only a slight decrease in conductivity (~ 2-3 orders of magnitude).

Conclusions
The first part of the research conducted experimentally investigated the dynamic electro-mechanical behavior of multifunctional composites when subjected to static and dynamic mechanical loadings. When properly dispersed within a polymeric matrix, conductive nanofillers such as CNTs and graphene can be assembled into a three-dimensional internal sensory network. These highly intricate electrical networks can be utilized to detect important information such as material deformation as well as various forms of damage. The main objective of this investigation was to characterize the electro-mechanical behavior of multifunctional composites when compressively deforming under low and high strain rates. Two types of multifunctional materials were investigated: carbon nanotube/epoxy and graphene/polystyrene composites. The knowledge obtained from this study will further the development of novel "smart" materials that could be used in many applications where compressive loading may be present.
The second part of the research focused on the development of novel strategies to effectively incorporate graphene into polymeric host materials. To further the development of more advanced multifunctional materials, novel techniques were developed that overcame many of the major limitations and issues associated with incorporating two-dimensional conductive filler materials into polymers matricies.
Moreover, the techniques developed allow for the fabrication of multifunctional composites with tunable properties.

123
The specific deliverables of the project can be summarized as follows: (1) The electrical resistance of the CNT-reinforced epoxy composites was highly affected by material strain and deformation mechanisms induced by the applied loading. The electrical response observed during SHPB loading demonstrated a similar response as previously observed during both quasi-static and drop weight loadings, where the bulk electrical resistance decreased during compression and then increased as damage initiated and propagated. The bulk electrical resistance of these nanocomposites decreased ~ 85% during SHPB experiments.
(2) The change in electrical conductivity of the material due to the CNT arrangement for small strains was determined using a Taylor expansion model to better characterize the electrical response demonstrated by the material.
(3) It was observed that the changes in CNT networks within the nanocomposite contributed approximately 64% to the overall resistance change of the material while only 36% was due to dimensional changes. This phenomenon differs from a typical strain gage measurement where the change in electrical resistance is based primarily on the dimensional changes rather than the change in material conductivity.
(4) The sensitivity of the CNT/epoxy composite, or the rate of increase in electrical conductivity as a function of strain, increased as the material became closer to yield for both quasi-static and dynamic loading. However, the rate of change of conductivity as a function of strain is higher during quasi-static loading than dynamic loading. It can be postulated that the difference between the two curves is due to an increase in matrix stiffness occurring over a very brief period of time in (7) A simple, inexpensive and commercially viable technique that can be used to disperse conductive sheet-like particles, such as graphene, into a highly organized pattern within polymeric materials on either the micro-or macro-scale was developed. Utilizing capillary interactions between polymeric particles and fewlayer graphene particles, liquid bridges on the surface of a polymeric material allows for coating of graphene onto the polymer surfaces. By precisely controlling the temperature and pressure during the melt compression process, highly conductive composites were formed using very low loadings of graphene particles.

Future Work
The current research is a step forward in understanding the electro-mechanical Understanding the static and dynamic electro-mechanical response would be highly beneficial to further the development of novel, more robust sensing devices.
(4) Since the capillary-driven particle templating technique was successful in producing highly conductive composites using polymeric pellets, this process could be extended to coat fibers / yarns. The coated fibers / yarns can then be woven into highly intricate structures throughout a polymer thus producing a novel multifunctional composite. Examples of smart fiber reinforced composites are schematically represented in Fig. 1. The rectangular specimen, shown in Fig 1.(a), has conductive fibers intricately woven into a three dimensional network throughout the polymer. Fig. 1(b) shows a hollow cylindrical fiber reinforced specimen with the conductive fibers weaved throughout the thickness of the tube.
The investigation would include using certain chemical or mechanical processes to enhance and/or optimize the coating process. These novel composites would possess unique properties that may be useful for many future applications. The electro-mechanical behavior could also be investigated when the composites are subjected to various types of loading schemes (i.e. compression, tension, flexure, torsion, blast loading).

129
(5) The electro-mechanical behavior of CNT/epoxy and GNP/PS was studied under dynamic compressive loading using an SHPB apparatus. These investigations revealed fundamental information about the electrical behavior of such composites when subjected to compressive loading. Using this information, these novel sensors could be incorporated into structures optimized for energy absorption applications, thus creating a novel smart sandwich structure. Figure 2 shows a schematic of a smart composite structure that could be used in a blast application.
The electro-conductive polymeric sensing material can be sandwiched between two stiff facesheets (E composite << E facesheet ). A shock tube apparatus could be used to generate a controlled shock wave directed at the sandwich structure while the electrical resistance of the composites could be monitored. Along with high-speed photography, a 3-D digital image correlation could be implemented to obtain the real-time in-plane strains and out-of-plane displacements of the structure. The change in electrical resistance could be correlated to deformation. Once the fundamental physics and electro-mechanical behaviors are understood, the smart  6. Next, the solution is dropped into a large volume of vigorously stirred methanol to coagulate the PS nanocomposites.
7. The composite is then filtered and dried in an oven at ~80 °C for ~ 25 hrs.
8. Finally, the PS-G nanocomposites are compression melted using a hot press

Melt Compression Procedure
A schematic of the stainless steel mold fabricated is shown in Fig. 1 8. Press to 10 metric tons and maintain pressure for 20 seconds. Release, then cool.

Place ball bearing (sandwiched between two brass plates) on top of piston and
press to 5 metric tons.
3. Rotate piston to desired angle, and press to 10 metric tons. Hold pressure for 5 min.
A schematic of the stainless steel mold fabricated is shown in Fig. 1 A schematic of the stainless steel mold fabricated is shown in Fig. 1. The mold consists of a base plate, lower insert, outer shell, piston and a cartridge heater.

BEHAVIOR OF GNP/PS PREPARED BY SOLVENT CASTING
Prior to studying the electrical response of the GNP/PS composites under load, it was essential to understand the electrical behavior of these composites under static conditions. In order to obtain an electro-conductive composite, a conductive filler material must be properly dispersed throughout the polymer. The level of dispersion directly correlates with the resulting composite properties. For this reason, two separate studies were carried out to determine (1) which size of graphene is best to use and (2) which solvent will better disperse the GNP in polystyrene.

< 2µm vs. 25 µm xGnP Nanoplatelets
Two types of graphite nanoplatelets (GNPs), < ~ 2µm and 25 µm in diameter, were studied to determine which would provide a higher electrical efficiency. Briefly, To evaluate the dispersion of the graphene nanoplatelets inside the polystyrene matrix, each disk was sliced into approximately 12 rectangular prisms, as shown in Fig. 1. A two-point probe measurement technique was utilized to measure the conductivity of the material. A schematic representation of the measuring technique is shown in Fig. 2. Silver paint was used to reduce contact resistance between the specimen and measurement device. Currents ranging from 1 nA to 1 mA was passed through the sample to obtain the resulting resistance measurements.   146 obtain an estimate of material conductivity, each composite disk was sliced into approximately 12 rectangular prisms, as shown in Fig. 1. When CHCl 3 was used as a solvent, only (7) out of (20) specimens exhibited a resistance that was measureable (<52 GΩ). On the other hand, (23) out of (23) specimens that were made using DMF were all conductive. For this reason, DMF was chosen over CHCl 3 as the solvent of choice to use in the composite fabrication.  heating. On the contrary, for a composite of lower resistance (40 Ω) , the resistance remains constant for all currents supplied, as shown in Fig. 9. Further studies were carried out and sufficient evidence was found to support the theory that joule heating of the GNP/PS composites that had electrical resistances in either the kilo-ohm or mega-ohm range was occurring. According to the classical percolation theory, the electrical conductivity of the composite can be described as where σ is the electrical conductivity of the composite, σ o is a scaling factor, φ is the volume fraction of the conductive filler, φ c is the volume fraction filler at percolation, and t is a critical exponent which is related to system dimensionality. The conductivity data was fitted by plotting log σ vs log (φ-φ c ) and incrementally varying φc until the best fit was achieved.  [Kogut et al, 1979]. During the melt compression process, the graphene particles are templated over the polymer pellets, and forced into a conducting network through the sample. Due to the anisotropic conductivity of graphene, the orientation of the graphene sheets will have a significant effect on the electron transport through the conductive networks.

Effect of Particle Size on GNP -Polystyrene Particle Templated Composites
A small study was conducted to investigate the effect of the polymer particle size on the electrical behavior of the composites. Two different particle sizes were experimentally tested. The standard pellet size consisted of PS pellets having an average size between 2.23 mm -3.18 mm while the crushed pellet size consists of pellet sizes ranging from 590 µm -1400 µm. To obtain the smaller pellets, an ordinary coffee grinder was used and the crushed material was assorted according to size using a set of sieves. Again, for GNP loadings less than 0.1 % v/v, the dry electrostatic adsorption technique was sufficient to coat the PS pellets. For loadings greater than and equal to 0.1 % v/v, the capillary-driven particle templating technique was used.
A volumetric two-point probe measurement technique was implemented to evaluate the material conductivity, as shown in Fig. 2. The conductivity measurements were made in the direction perpendicular to the molding direction. Figure 3 shows the particle size effect on the electrical behavior of the GNP/PS composites. Both the standard and crushed particle composites exhibited excellent electrical conductivity. However, the standard size particles exhibited a percolation threshold at a volume fraction less than 0.1 while the crushed particles showed a percolation to occur below 0.3 % v/v. As the GNP content was further increased from 0.5 % v/v to 1 % v/v, the conductivity remained unchanged and showed similar magnitude for both particle sizes. The variation in the percolation threshold for these composites may be attributed to the particle size difference but may also be attributed to the non uniform coating of GNP on the crushed pellets. When handling the smaller 155 pellets, the capillary driven coating technique must be carefully executed so that the small polymeric particles do not stick to other polymeric particle via capillary forces.
It is important to note that careful attention must be paid to using a critical amount of methanol, no more and no less, as well as the mixing procedure used.

POLYSTYRENE PARTICLE TEMPLATED COMPOSITES
A brief study was performed to investigate the electrical behavior of the particle templated composites using carbon black (CB) as a filler material. The CB used was acquired from Cabot Corp (LITX 50, Cas No. 1333-86-4). The average particle size of was ~ 50 nm. The polymeric material chosen for this study was polystyrene (Crystal PS 1300, average molecular weight of 121,000 g/mol) purchased from Styrolution, USA. The PS pellets (~ 2 mm) used were elliptical prisms with a total surface area of 1.03 ± 0.01 cm2. All CB/PS particle templated composites were fabricated using the same capillary driven-particle templated procedure used to produce the GNP/PS composites. The electrical behavior of the CB/PS and GNP/PS is shown in Fig. 1. Similar to the GNP/PS, it can be observed that the capillary driven fabrication procedure was successful in creating highly conductive CB/PS also. The percolation threshold was < 0.05 % v/v carbon black for the CB/PS composites. This value is slightly higher than the GNP/PS composites (< 0.01 % v/v GNP).
Furthermore, the electrical conductivity at 0.5 % v/v filler for CB/PS was ~ 2-3 orders of magnitude less than GNP/PS. This difference can be attributed to the variation in particle shape size since the CB particles were ~ 50 nm whereas the GNP sheets were ~ 25 microns in length. where  is the density, c is the wave speed and A is the area.
2. Then select the pressure bars (steel or Aluminum) closer to the impedance of the specimen. We also have different diameters for the pressure bars.
Note: The basic thumb rule is that we use steel bars for the harder materials (l) Place the pulse shaper at the impact end of the incident bar with a thin layer of KY jelly grease (if you are using lead pulse shaper) and align it with respect to bar center. We generally use clay and lead pulse shapers. These give us very good results for harder materials, but for the softer materials, you can try different pulse shapers. These include paper, copper etc.
(m) Release the nitrogen gas from the gas tank into the gas gun chamber until the required pressure level is achieved.
(n) Arm the oscillation to capture the strain gage voltage signals and make sure the arm holds until you release the projectile. If the arm is not holding, adjust trigger levels. (Note: if you are getting high noise in your signals more than 20mv, turn off the tube lights before the experiment).
(o) Once again, ensure that the specimen is well aligned between the bars and verify the status of the trigger hold before pressing the solenoid valve release button.
(p) Press solenoid valve control box button to release the projectile.
(q) Save captured voltage pulses onto a USB drive for further analysis of the data.
(r) A MATLAB program is written to read the data from the pulses and analyze the pulses using the one-dimensional wave theory stress and strain equations. After the experiments are performed, the pulses are used along with the MATLAB program to determine the equilibrium and true stressstrain plots of the specimen.
(s) After the experiment is completed, turn off the cylinder and make sure all the left over nitrogen gas in the gas chamber is released.
(t) After the data is transferred from the oscilloscope to USB drive, verify that in your computer and turn off the amplifier and oscilloscope.
Analyzing the results: 1. There are two MATLAB codes to analyze the data. 1. Verify_Equilibrium and 2. Steel/aluminum_SHPB. Use the appropriate codes to analyze the data.
Depending on the bars you used, the respective code has to be used.

166
2. Make sure the code has the right properties and dimensions of the pressure bars you used. These include diameter, wave speed, and diameter.
3. If we use hollow tubes, make sure you have the right dimensions in the code.
For solid bars, dimensions for the hollow tube should be zero.
4. First run the verify equilibrium code. Make sure the data you get from the oscilloscope has the following names for the four channels. TEK00000, TEK00001, TEK00002,and TEK00003. The code recognizes these names.
Make sure the codes and the data are in the same folder. 8. When you run the code, you get two figures. Figure 1 gives the incident and reflected pulses. Figure 2 gives the transmitted pulse. pulses by pressing 'zoom in' button at the top to get the right times. Then go to MATLAB main window and press 'ENTER'. 10. Input the values you found out and press 'ENTER'. 11. Now you will get 3 more Figures. Figure 3 shows the incident, reflected and transmitted pulses. Figure 4 shows the incident, reflected and transmitted pulses you picked on before. Figure 5 shows the force ratio. Front face represents the forces calculated on the incident and reflected pulses. Back face represents the force calculated on the transmitted pulse. Ideally, these two fronts and back face should match perfectly.
12. Various factors decide the equilibrium. These include type of material tested, strain rate etc.. 13. Make sure the incident and reflected pulses start at the same time on 16. Now open the SHPB code and make sure you have the same value for filter as in the verify_equilibrium code.
17. Enter the specimen thickness and diameter in inches. 18. Again, you get two figures. Figure 1 gives the incident and reflected pulses. 24. Go to MATLAB main window and you can see the strain rate. Note down this value. Next you will end up with final figure (Figure 6). This is eng. strain rate vs. time. 25. Be careful when you pick up the strain rate points. Consider the following points a. Make sure the region you pick is in the equilibrium. 2. Selecting the bar is same as explained before.
3. Experimental procedure is also similar to the above. Here, you place the pulse shaper on the flange. You can use paper, clay or lead. 4. Different striker bars can be used to perform experiments at different strain rates. Make sure the striker bar slides freely on the bars. 5. The specimen will be threaded at both ends to the pressure bars. There is no need to use the lubricant. 6. The connections remain the same as explained before. You can use the same amplifier and oscilloscope, and same settings. 2. Lead wires should be securely attached to each bar to provide a means of supplying a DC current flow through the specimen during loading.
3. Apply a conductive lubricant (i.e. AI Technology Inc. ELGR8501) to the specimen faces to minimize contact resistance as well as frictional forces at the specimen-bar interfaces. 4. A pulse shaper must be carefully chosen to provide good force equilibrium conditions while still ensuring complete isolation between the striker bar and incident bar throughout the experiment. An example of a pulse shaper used is 1 layer of electrical tape and clay (~ 2 mm thick). 5. The voltage across the specimen can be measured one of two ways. Additional lead wires can be attached to the incident and transmitter bar. This approach provides a better volumetric measurement and has less risk of having the measurement probes disturbed/damaged during the experiment. The other option is to implement the concentric ring method where two very small concentric rings are machined into the surface of the specimen, filled with silver paint and lead wires are attached. This method is more accurate in the sense that the contact resistance between the bar and specimen is avoided.
However, validation experiments were completed to ensure very minimal difference between the two measurement techniques.

A constant current source with high frequency response (Keithley Instruments
Model 6221) was used to supply the constant DC current flow under the high rate deformation while the voltage drop between the two inner probes was measured by a differential amplifier (Tektonix ADA 400A) and recorded by a digital oscilloscope (Tektronix TDS 3014).
7. It is important to note that proper strain gage selection is critical in preventing any electrical interference in strain measurements while conducting these types of experiments. The particular strain gages chosen (Micro-Measurements C2A-13-250LW-350) consists of an encapsulated gage mounted on a thin highperformance laminated polyimide film backing. The polyimide film backing provides a layer of insulation between the actual gage and the bar surface and therefore prevents any voltage interference. A series of experiments were performed with and without supplying current through the bars, validating that the strain gages bonded to the bars remain unaffected. 8. Compression SHPB at elevated temperatures: 1. The tungsten carbide inserts will be used. The specimen will be sandwiched between these inserts.
2. The diameter of the specimen should be smaller than the inserts. The below figure shows the set up. 3. For experiments at elevated temperatures, the SHPB apparatus in conjunction with the induction coil heating system will be utilized as shown in Fig. 2. 4. A special fixture is used to load the specimen. 5. The inserts were used to eliminate the temperature gradient in the bars and thus protect the strain gages mounted on them.
6. The impedance of the inserts was matched with the bars; hence they do not disturb the stress wave profiles in the bar. The impedance matching requires the diameter of these tungsten carbide inserts to be smaller than the main Holding Fi t

Incident Thermocouple
Transmitted bar

Induction
Tungsten carbide inserts Specimen Coil pressure bars. This is the reason for the specimen diameter for high temperature testing being smaller than that for room temperature testing. 7. By varying the power, higher temperatures can be achieved. 8. The induction coil heating system has a power control box, remote to start and stop, a cooling unit and cooling supply (blue box) to reserve water. Make sure the blue box has sufficient distilled water. The copper coils are connected to the cooling unit and it is places around the inserts. 9. First turn ON the blue box, then the power supply. The power supply needs the larger output in the DPML lab. 10. Make sure the wheel on the cooling unit is pinning smoothly and fast. If not, do not do the experiment. Increase the power supply, to heat the specimen. 11. When the regulator is turned ON, it should give a click sound after around 30s.
If it does not, turn it off and try again. If the problem persists, turn off the regulator and the problem can be determined.
12. Turning ON the power supply regulator, it will read 'cycle continuous' on the remote (smallest one), which is desired. 13. The system should already be set to manual power output again, which will allow to control the power. If it is not set, you can do by using the switch located to the immediate right of the dial on the regulator.
14. Make sure the dial on the regulator is zero, so there will be no immediate power output. 15. Now press 'start' button on the remote (small one that reads the diaplay). 16. The bars were kept apart initially, later the specimen and carbide inserts were heated in isolation to the desired temperature (usually about 20-50°C higher than the test temperature) and soon after the bars were brought manually into contact with the specimen. The temperature of the specimen was monitored using a 0.127mm chromel-alumel thermocouple, which was spot welded onto the specimen. 173 17. In most of the experiments, it takes less than two minutes to heat the specimen to the required temperature and it takes less than 10 seconds to bring the pressure bars into contact with the tungsten inserts and fire the gun. 18. Once the temperature is reached, hit 'stop' on the display and turn off the regulator and the induction heater. Now trigger the oscilloscope. If you trigger the oscilloscope before, due to magnetic fields from the induction heater, you will see lot of noise. 19. Allow the cooling unit to run for some time son that it reaches room temperature.
20. All other experimental procedure, data capturing, and analyzing the results remain the same as explained in compression SHPB section.

Note:
1. Always make sure the yield strength of the material you are testing is never beyond the yield strength pressure bars.
2. For testing ceramics of high strength, we need to use inserts so as to protect the bars from plastic deformation. . Make sure to yell "firing" when experiment is about to be run and SHPB is being pressurized, keep outside doors closed so no one walks in 7. Make sure everyone in the lab, helping or not with the experiment, is aware an experiment will be taking place

8.
Never ride in an elevator with liquid nitrogen! When using passenger elevators, use an elevator key to prevent the door from being opened by unauthorized persons. If a key is not available, then station a person at each floor to ensure no one enters.

9.
Note that outside of normal working hours (M-F, 8:00 a.m. -5:00 p.m.), no one is allowed to transfer liquid nitrogen from the Dow loading dock area without a second person present. Failure of a container or a large spillage could result in asphyxiation at a time when you are unlikely to be found or able to get assistance.
10. Always fill warm dewars slowly to reduce temperature shock effects and to minimize splashing.
11. Always make sure that containers of liquid nitrogen are suitably vented and unlikely to block due to ice formation.
12. Do not fill cylinders and dewars to more than 80% of capacity, since expansion of gases during warming may cause excessive pressure buildup. save modifed_resistance.txt res_modified /ascii save modifed_force.txt force_modified /ascii save modifed_strain.txt strain_modified /ascii