Enhancing Mechanical and Thermal Properties of Epoxy

Subscriber access provided by HKU Libraries

Article

Enhancing Mechanical and Thermal Properties of Epoxy Nanocomposites via Alignment of Magnetized SiC Whiskers James Townsend, Ruslan Burtovyy, Pavel Aprelev, Konstantin G Kornev, and Igor Luzinov ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 09 Jun 2017 Downloaded from http://pubs.acs.org on June 13, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Enhancing Mechanical and Thermal Properties of Epoxy Nanocomposites via Alignment of Magnetized SiC Whiskers James Townsend, Ruslan Burtovyy*, Pavel Aprelev, Konstantin G. Kornev, Igor Luzinov* Department of Materials Science and Engineering, 161 Sirrine Hall, Clemson University, Clemson, SC 29626, USA *To whom correspondence should be addressed: [email protected] and [email protected]

Abstract: This research is focused on the fabrication and properties of epoxy nanocomposites containing magnetized SiC whiskers (MSiCWs). To this end, we report an original strategy for fabrication of magnetically active SiCWs by decorating the whiskers with magnetic (iron oxide) nanoparticles via polymer–polymer (polyacrylic acid/poly(2-vinyl pyridine) complexation. The obtained whiskers demonstrated a substantial magnetic response in the polymerizing epoxy resin with application of only a 20 mT (200G) magnetic field. We also found that the whiskers chemically reacted with the epoxy resin causing formation of an extended interphase near the boundary of the whiskers. The oriented with magnetic field SiC whiskers demonstrated positive effects on the behavior of epoxy-based nanocomposites. Namely, the aligned MSiCWs enhanced the thermomechanical properties of the materials significantly above that of the neat epoxy and the epoxy nanocomposite with randomly oriented whiskers.

Keywords: nanocomposites, epoxy, silicon carbide whiskers, magnetic alignment, magnetic nanoparticles, thermo-mechanical properties.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 42

Introduction: Anisotropic polymer composites with aligned fibers have been transforming automotive, aerospace, energy, defense, and construction industries.1, 2 The directionally reinforced composite structures currently used in industrial settings are primarily composed of woven, braided, stitched, or punched together yarns embedded into a polymer matrix.3 These materials originated from the textile industry and were adapted to fabricate anisotropic structures for stress adsorption, electrical/thermal conduction, and wear resistance in macro-scale structural composites.2 The fibers significantly increase the material strength and modulus in the fiber direction because of stress transfer across the fiber-matrix interfaces parallel with the load. It is necessary to point that the fibers currently used industrially are tens micrometers in diameter. Scaling fiber diameter down to a submicron size to obtain fiber reinforced nanocomposites offers dramatic improvement in composite properties.4 The enhancement originates from the large surface area (high aspect ratio) of nanofibers providing extended fiber–matrix stress transferring interfaces at minute volume fractions. To this end, the next natural step in the development of fiber based nanocomposites is to reach directionality of nanofiber orientation in polymer matrices. Current scientific literature provides several examples of the actual orientation of nanoscale fillers in nanocomposites, including nanofibers.1, 5-21 The materials with anisotropic orientation of nanofillers have demonstrated significantly improved (and often unique) mechanical, thermal, electrical, and magnetic properties. To date, orientation of nanocomposite fillers has been induced by magnetic fields5, electrical fields22, mechanical stretching23, shear flow of viscous liquids24, 25, and combinations of shear and magnetic fields.14 The "shear flow" methods presume only an asymmetric geometry of nanofillers, are difficult to control precisely, and are limited to manufacturing methods utilizing controllable flow of the matrix. Magnetically and electrically guided orientation typically assumes the presence of additional active components in the materials capable of interacting with an applied field. In general, orientation of nanoscale fillers/fibers with the magnetic field is the most effective and universal approach, because practically all polymer matrices do not interact with the applied magnetic field. Therefore, the matrices do not interfere with the orientation procedures. Using a magnetic field, it is possible to obtain 3D reinforcement of nanocomposites in specific and multiple directions within the single materials, mimicking structural biological composites.1, 5, 2627

The majority of nanoscale fillers are modified with magnetic nanoparticles11,12, 16-21, 28 to obtain 2 ACS Paragon Plus Environment

Page 3 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


orientable (in relatively low strength magnetic fields) fillers. Deposition of Fe layer on carbon nanotubes was also used for this purpose.15 Without this modification an extremely strong magnetic field must be used to orient nanofibers.10 To date, results have been published for magnetically oriented composites involving the following nanoscale fillers: alumina platelets5, 17, silver nanowires18, boron nitride platelets21, carbon

nanotubes9,

15

,

carbon

nanofibers11,

12

,

lamellar

clay

(montmorillonite)19,

graphene/graphene oxide16, 20, 29 , Fe3O4 nanorods13, CaP rods30, cellulose nanowhiskers,10 and SiC nanoparticles.21 Among these materials, only alumina platelets, lamellar clay, carbon nanotubes/nanofibers, boron nitride platelets, and silicon carbide nanoparticles/whiskers (SiCWs) are produced industrially in commercial quantities and, consequently, are ready to be used in production of nanocomposites with directionality of their orientation. There is no overlap in terms of shape/dimensions and chemical/physical properties in this very limited portfolio. Therefore, every material from this list is principally important for the practical design and fabrication of nanocomposites. With this in mind, this research focused on magnetized SiCWs as a high aspect ratio nanoscale filler capable of significantly enhancing properties of nanocomposites when directionally aligned in a polymer matrix. Specifically, we demonstrated the positive effects that 2D-aligned SiC whiskers have on the behavior of epoxy-based nanocomposites. The attractiveness of employing SiCWs in composite materials resides in their exceptional physical properties, such as a high modulus, thermal and radiation stability, abrasion resistance, a low coefficient of thermal expansion, high thermal conductivity, and low electrical conductivity.25, 31, 32 However, there are rather limited number of published scientific works devoted to composites consisting of the whiskers directionally aligned in polymer matrix.24, 25, 33 In these published works, the alignment was conducted using shear flow orientation. To the best of our knowledge, no published reports exist on nanocomposites containing SiCWs oriented in polymer matrix composites using a magnetic field. To this end, we report an original strategy for fabrication of magnetically active SiCWs (MSiCWs) using decoration of the whiskers with magnetic (iron oxide) nanoparticles (MNPs) via polymer–polymer complexation. The obtained whiskers demonstrated a high magnetic response in epoxy resin with application of a magnetic field of only 20 mT (200G). We fabricated epoxy/MSiCW composites with and without magnetic orientation of the whiskers.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 42

Orientation of MSiCWs within the nanocomposites resulted in significant improvements in thermal and thermo-mechanical properties of the materials.

Experimental section: Iron oxide nanoparticles MNPs were obtained from US Research Nanomaterials, Inc. as a dry powder. According to the manufacturer, they were composed primarily of magnetite. Before use, MNPs were cleaned using a three-step procedure. First, they were dispersed into methanol (VWR International) at 10 wt%. Second, the MNPs were ball-milled by glass balls (1 mm in diameter) on a high setting for 1 h in a ST-20/50 dispersion tube with IKA® Ultra Turrax® tube disperser. Then, a Precision-Durafuge 100 centrifuge was used at 5000 rpm for 15 min to remove large agglomerates from the suspension. Centrifugation was repeated three times on the same suspension. After each centrifugation, the top portion of the supernatant was decanted and retained. Next, an Eppendorf mini-spin plus centrifuge was used at 10000 rpm for three min. The decanted portion of the nanoparticle suspension was centrifuged three more times, and each time the precipitated portion was retained. Prior to further use, the particles were maintained as a concentrated suspension in methanol. Surface modification of MNPs MNPs were dispersed in methanol to prepare a 0.3 wt% colloidal suspension. A 3 wt% solution of polyacrylic acid (PAA, 100,000 g/mol molecular weight, Sigma Aldrich) in methanol was prepared and sonicated using an ultrasonic bath. The nanoparticle suspension was added dropwise in a 1:1 volume ratio to the sonicated PAA solution. Then, the PAA/MNP suspension was shaken for 24 h and cleaned three times in methanol using centrifugation and magnetic separation. The magnetic separation was accomplished with a ~200 G neodymium magnet. Silicon carbide whiskers Silar SR-9M SiCWs were purchased from Advanced Composite Materials, LLC. The SiCWs were cleaned to remove any possible processing impurities. Specifically, a non-glazed ceramic porcelain crucible with SiCWs was placed into the center of a Lindberg/Blue 1000oC ceramic lined oven with a Eurotherm 2416 temperature controller. The oven was heated to 400oC over 20 min and held at 400oC for 1 h. Since SiC is stable up to 1000oC in air32 this temperature does not change structure of the whiskers. Subsequently, the oven was cooled to room temperature at ~3oC/min in 4 ACS Paragon Plus Environment

Page 5 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


air at atmospheric pressure. The controlled heating and cooling rates were used to reduce residual thermal stresses in the SiCWs. After this thermal cleaning, the SiCWs were tested for impurities using thermal gravimetric analysis (TGA). The oven-cleaned SiCWs had no significant weight loss or gain when heated to 700oC. After cleaning, a size separation procedure was conducted to reduce the number of aggregates and small SiC particles in the whisker sample. Specifically, gravity precipitation and centrifugation was used for the separation. First, a 2 wt% suspension of SiCWs in 40 mL of acetone was sonicated for 15 min. The centrifugation vials were then shaken and placed standing upright for 20 s. The suspension was pipetted off the top until 10 mL of the suspension remained. Then, more acetone was added to the vials. The procedure was repeated three more times producing a suspension (90–120 ml) of SiCWs without large aggregates. The SiCWs were re-dispersed in methanol and centrifuged at 5000 rpms for 15 min. The precipitate was retained, but the top solution was decanted. The sediment was re-dispersed in fresh methanol and centrifuged again. This procedure was repeated for 3–6 times to obtain a visually clear methanol layer above the whisker sediment. The obtained whiskers were imaged by scanning electron microscopy (SEM) and found to have diameters in the range of 0.2–0.6 μm (0.42 μm average) and lengths from 2 to16 μm (4.2 μm average). Modification of SiCWs with epoxy silane and poly(2-vinyl pyridine) Epoxy silane [3-glycidyloxypropyl)trimethoxysilane], ES, was obtained from Sigma Aldrich. The modification of SiCWs was conducted using the ES solution in toluene. Namely, 1 wt% suspension of SiCWs in toluene was added dropwise to a 5 wt% ES solution to reach a 1:1 ES/SiCW ratio by weight. The mixture was shaken for 24 h. Then, the SiCWs were centrifuged and rinsed with toluene and methanol three times each to remove any unattached silane. Afterwards, modified SiCWs were transferred to methyl ethyl ketone (MEK) purchased from Acros Organics. Carboxyl terminated poly(2-vinyl pyridine) (P2VP, 53,000 g/mol molecular weight) was purchased from Polymer Source. The SiCWs modified with ES and dispersed in MEK at ~0.2 wt% were added dropwise into a 2 wt% solution of P2VP in MEK at a 1:1 volume ratio. MEK was removed from the mixture in a Yamato RE2000 rotary evaporator operated at 60 rpms at room temperature under a steady flow of nitrogen. The P2VP/SiCW system was annealed at 115oC for 24 h in a vacuum oven. The unreacted polymer was removed with methanol by performing three centrifugation/re-dispersion steps. Prior to further use, the whiskers were kept in methanol.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 42

Fabrication of MSiCWs Two methanol dispersions (SiCWs modified with P2VP and MNPs modified with PAA) were mixed under continuous sonication in 1:10, 1:8.6, 1:5, and 1:2 weight ratios (MNP/SiCW). During the procedure, the dispersion of SiCWs was added dropwise to the dispersion of sonicated MNPs. After the mixing was completed, visual observation clearly indicated that no free nanoparticles remained in the methanol. The magnetic whiskers thus obtained were diluted three times and vigorously shaken for 24 h. To finalize the functionalization, the whisker suspension was added dropwise to a sonicated 2 wt% solution of PAA in methanol and shaken for 24 h. Finally, the MSiCWs were cleaned in methanol by magnetic separation and centrifugation. The magnetic separation was conducted using a ~200 G neodymium magnet. The whiskers modified with MNPs showed an immediate response to the magnetic field. The magnet was held below the solution for 15 min, and then the top portion of the materials was decanted. Then, centrifugation at 5000 rpm was performed for 15 min. The top portion of the solution was decanted, and the sediment of SiCWs was re-dispersed in methanol. Prior to further use, the MSiCWs were maintained in methanol. Fabrication of the composites Araldite 6005 (bisphenol-based) epoxy (Bis-A), dodecyl succinic anhydride (DDSA) hardener, and N-benzyl dimethylamine (BDMA) catalyst were obtained from Electron Microscopy Science. A methanol suspension of MSiCWs was mixed with epoxy resin and dried overnight under a stream of nitrogen to remove the solvent. Then, the MSiCW suspension was mixed with the hardener and catalyst at a 0.41:0.56:0.03 volume (Bis-A/DDSA/BDMA) ratio. The mixture was vacuum pumped to remove any air bubbles created during the mixing. The same procedure (where methanol was initially added to Bis-A) was utilized for the fabrication of the neat epoxy polymer used in this work for comparison. To fabricate the composite and neat epoxy samples, the materials were poured into a Teflon mold. The epoxy resins were cured at room temperature for 48 h and, subsequently, put into an oven for 24 h at 75oC to complete the curing. Calorimetry study of epoxy curing Differential scanning calorimetry (DSC) was used to determine the extent of the epoxy curing from the heat of polymerization.34 Exothermic peaks (representing the amount of uncured epoxy) were recorded for epoxy resin samples cured at room temperature over various times. The samples 6 ACS Paragon Plus Environment

Page 7 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(partially cured at room temperature for 3, 6, 12, 24, and 48 h) were heated in the DCS instrument from 0 to 160oC with a 15oC/min heating rate. The peaks were integrated to obtain the heat of polymerization. The heat values were divided by the heat of polymerization for the "as prepared" epoxy to determine the degree of curing at room temperature. With this procedure we also established that the curing for 48 hours at room temperature with an additional curing for 24 hours at 75 oC leads to the fully cured epoxy resin. To determine the activation energy for the epoxy polymerization, DSC data were used according to the Kissinger methodology (Eq. 2):34

ln (

ℎ𝑒𝑎𝑡𝑖𝑛𝑔 𝑟𝑎𝑡𝑒 𝑇𝑝2

)=−

𝐸𝑎 𝑅𝑇𝑝

+ ln(𝐴)

(Eq. 1)

where ‘heating rate’ is the heating rate used in the DSC experiments, Tp is the peak temperature of the DSC exotherm, Ea is the energy of activation for the curing reaction, R is the gas constant, and A is the pre-exponential factor. Tp was determined for the epoxy resin after 3, 6, 12, 24, and 48 h of curing at room temperature. Ea was determined from the slope of the line formed by plotting ln(heating rate/Tp 2) versus 1/Tp. DSC studies were conducted with the DSC-Q1000 instrument (TA Instruments). Infrared study of epoxy curing: Attenuated Total Reflectance - Fourier Transform Infrared Spectroscopy (ATR-FTIR) was used to study the curing of the epoxy resin. The epoxy samples were drop casted by a glass pipette onto an ATR crystal after 0, 3, 6, 12, 24, and 48 h of curing at room temperature. The IR spectra acquired were used to estimate the degree of polymerization using the 913 cm-1 and 1780 cm-1 IR peaks for the epoxy and anhydride groups, respectively. The recorded peaks were normalized to the 1470 cm-1 methyl IR peak that does not change during the course of reaction. To calculate the degree of reacted functional groups, the area of the normalized peaks for the different curing times were subtracted from the area of the normalized peaks for the "as prepared" epoxy and divided by the latter peak area. ATR-FTIR spectra were acquired with Thermo Nicolet 6700 FTIR spectrometer. Apparent viscosity of epoxy resin After mixing the epoxy with hardener and catalyst, the product was poured into a 20-ml glass test tube with a diameter of 1.5 cm. The test tube was placed upright beside a ruler and a microscope. 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 42

A glass sphere, measuring 2 mm in diameter, was placed on the surface of the epoxy resin. As the sphere dropped into the solution, a meniscus appeared between the glass sphere and the epoxy. A microscope was used to monitor the retraction of the meniscus from the glass sphere. The retraction of the meniscus signified the complete immersion of the sphere in epoxy. As the sphere moved down and away from the resin surface, the speed of the sphere was monitored as the sphere crossed the 2-cm mark and again as the sphere passed the 4–5 cm vertical distance in the resin, where 0 cm was at the surface of the epoxy resin. The ruler was used to measure the distance the glass sphere dropped during a given time. The velocity of the dropping sphere, Vs, was used to measure the apparent epoxy viscosity, η: 35-36.

2r 2 g (  S   L )  9Vs

(Eq. 2)

where ρS and ρL are the density of the glass sphere (~2.6 g/cm3) and the liquid resin (~1 g/cm3), g is gravity (9.8 m/sec2), and r is the sphere radius (1 mm).

Magnetic characterization of MNPs, MSiCWs and nanocomposites An alternating gradient field magnetometer (AGM 2900 Princeton Measurement, Inc.) was used to characterize the magnetic properties of materials. For each sample, the induced magnetic moment was measured as a function of the applied magnetic field. The applied magnetic field was slowly increased to a maximum of 400 kA/m and then decreased to -400 kA/m. The resulting hysteresis loops were analyzed. Samples of MNPs and MSiCWs were prepared by drying at room temperature, weighing, and encapsulating in scotch tape to prevent the loss of material. The dry magnetic samples were mounted on the magnetic probe and their hysteresis curves were measured.

Samples of

nanocomposite materials were cut into 3-mm squares having a thickness of approximately 0.34 mm. The average difference in length between the sides of the squares was 200 μm. Magnetic hysteresis loops were measured for two samples of nanocomposite with nonoriented embedded nanorods and three samples of nanocomposite with oriented embedded nanorods. For each sample, two magnetic hysteresis measurements were performed: one measurement with the magnetic field oriented along one side of the sample and the other measurement with the magnetic field oriented along the other side of the sample. By comparing the hysteresis loops along each side of the 8 ACS Paragon Plus Environment

Page 9 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


sample, an estimation of magnetic anisotropy was made for oriented and nonoriented nanocomposites. No magnetic hysteresis was observed and the magnetization curves were consistent with the behavior of paramagnetic materials described by the Langevin function37, 38:

  mp H m  Nm p  cot    kBT 

 kBT   mp H

  

(Eq. 3)

where m is the magnetic moment of the sample, H is the magnitude of the external magnetic field measured in Teslas, N is the number of magnetic domains in the sample, mp is the magnetic moment of a single domain, kB is the Boltzmann constant, and T is the absolute temperature. The magnetic moment of the sample was calculated by fitting the magnetization curves with the Langevin equation (1) and extracting the product Nmp37, 39, 40. A comparison between the samples was made after normalizing the saturation magnetic moment by the mass of the sample. Magnetic orientation of MSiCWs Two neodymium magnets (K&J Magnetics) were used to create a quasi-uniform ~ 216 G magnetic field to align the magnetic whiskers in the epoxy matrix. The magnetic field was applied after the epoxy resin was cast into the mold. The magnetic field was measured using a DTM-133 digital teslameter. Materials characterization DSC was used to determine thermal transitions for the MSiCW epoxy nanocomposites and the unfilled epoxy polymer. The DSC samples were prepared from the fully cured (in the Teflon mold) epoxy materials. The weight of the samples was in the range of 5–10 mg. The samples were sealed into an aluminum pan with an aluminum lid. The DSC heating and cooling rate was 15oC/min. The first DSC run was conducted from -10oC to 150oC. Then, the samples were cooled to -10oC and run for a second time to 150oC. Three to four parallel samples were analyzed to obtain the average and standard deviation. A Dynamic Mechanical Analyzer (DMA-Q800, TA Instruments) was used for thermomechanical characterization of the MSiCW/epoxy nanocomposites and unfilled epoxy polymer. The samples for DMA studies were cured in a Teflon mold having a length of ~8.00 mm, a width of ~5.00 mm, and a thickness of ~0.34 mm. DMA testing was performed at 2oC/min from -20oC



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 42

to 150oC using a frequency of 5 Hz while applying a 5 μm amplitude of oscillation. Three to four samples were fabricated and analyzed for the average and standard deviation. TGA was conducted with the TGA Q5000 (TA Instruments). In the TGA experiments, a sample (~5 mg) was heated under an atmosphere of nitrogen or air (gas flow = 20 mL/min) from 25 to 700°C at a heating rate of 20°C/min. Microscopy studies were conducted with Olympus MVX10 (optical microscopy), SEM S4800, Hitachi Inc. (scanning electron microscopy, SEM) and TEM H9500, Hitachi Inc (transmission electron microscopy, TEM). The images were analyzed manually using Gwydion software.

Results and Discussion: Principles of MSiCW design The MSiCWs consisted of several major components (Figure 1). The major constituent is SiCWs, which are fabricated from rice hulls, a renewable material.41 The surface of SiCWs contains SiO2 groups41 and, therefore, can be straightforwardly modified using silane chemistry.42 In fact, in this work, epoxysilane was used for the initial surface modification of the whiskers. We used superparamagnetic (iron oxide) MNPs as a magnetic component to decorate the whiskers. The nanoparticles do not possess a permanent magnetic moment, but are characterized by a relatively large saturation magnetization. These features allow avoiding potential problems with dispersion stability of the whiskers decorated with MNPs (unavoidable in the case of ferromagnetic particles) and ensure operational magnetic response at relatively low magnetic fields. The nanoparticles must be firmly attached to the surface of the whiskers to create robust magnetically responsive fillers. Therefore, formation of insoluble in epoxy resin polymeric complexes between PAA and P2VP was employed for the bonding of the nanoparticles to the surface of the SiCWs.43 This anchoring method is straightforward, effective, and released no byproducts during the process. We selected modification of the nanoparticles with PAA and the whiskers with P2VP polymer layers. In this case, the outer surface of the fillers contained carboxylic groups capable of reacting with epoxy functional groups, thus improving whisker– matrix interfacial bonding.42 The number of nanoparticles should be carefully adjusted. Enough nanoparticles should be attached to create the necessary magnetic moment when MSiCWs are subjected to the magnetic field. On the contrary, the amount of the particles should not be excessive to avoid diminishing the 10 ACS Paragon Plus Environment

Page 11 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


beneficial effect of rigidity of the SiCWs. In addition, significant protrusion of the layer of nanoparticles from the surface of the whiskers would influence the mobility of the whiskers as well as disturb the formation of strong whisker–matrix bonding compromising the mechanical properties of the composite. To regulate the amount of the particles on the whisker surface, MSiCWs were prepared using different MNP/SiCW ratios. The ratio providing the most optimal morphology of the MNP layer was selected.

Functionalization of SiCWs The functionalization of the SiCWs started with attachment of the polyelectrolytes to the surface of the SiCWs and MNPs. Specifically, P2VP was grafted to the SiCWs surface and PAA was anchored to the iron oxide nanoparticles. Prior to grafting with P2VP, the SiCWs were modified with epoxy silane. Surface modification of inorganic materials with epoxy silane is well established surface modification procedure to cover inorganic materials with reactive epoxy groups.42 The groups are used to anchor carboxyl-terminated P2VP macromolecules to the whisker surface via reaction between carboxyl and epoxy groups.44 The result of epoxy silane deposition was monitored by TGA, which indicated that approximately 1.5 wt% of silane was anchored to the SiCWs. Assuming uniform coverage of the average SiCW with the silane, the thickness of the epoxy silane coating was estimated to be ~ 5 nm (see Supporting Information, SI for details). The results indicated formation of a crosslinked epoxy silane coating with a thickness of approximately 5 monolayers.42 The epoxy silane-modified SiCWs were further modified by grafting of the carboxyl-terminated P2VP. The P2VP-grafted SiCWs had ~2 wt% of anchored polymer per TGA measurements. The thickness of the grafted layer was estimated to be approximately 7 nm (see SI for details). TEM observations also confirm formation of a nanoscale (on the level of 10 nm) epoxy silane/polymer-grafted layer evenly covering the SiCWs (Figure 2). The MNPs were modified by PAA adsorption from methanol solution. The bonding of PAA to the MNPs via formation of an insoluble salt was envisioned to create a simple and easily controlled method for surface modification.45 However, the surface of commercially available iron oxide nanoparticles is typically modified. The modification is used to control the surface-tosurface interactions and the size of the particulates.46 Hence, we supposed that the surface modification might hinder the attachment of PAA to the surface of the particles via the formation of an iron salt. Thus, it was necessary to select particles that possessed surface modifiers 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 42

compatible with anchoring of PAA. In fact, the MNPs selected here were covered with a poly(vinyl pyrrolidone) (PVP) nanoscale layer, which can form an insoluble complex with PAA.47 From TEM imaging, the size of the particles used was determined to be 7 ± 4 nm (Figure 3a). To obtain the particle size, more than 600 individual nanoparticles were identified and measured. TGA measurements indicated that the MNPs before PAA anchoring contained about ~6% of PVP. Assuming uniform coverage of the MNP, the thickness of the PVP layer was estimated to be < 0.4 nm (see SI for details), which is significantly lower than a PVP monolayer and cannot screen fully the MNP surface. Therefore, we suggest that combination of the PVP-PAA complex and salt formation are responsible for the attachment of PAA to the MNP surface. The modification of MNPs was monitored by SEM, TEM, and TGA. The surface modification of nanoparticles yielded formation of MNP aggregates and not individually covered particles (Figures 3b and 3c). We associate the formation of aggregates with the high affinity of PAA to PVP due to the complexation where individual PAA macromolecules rapidly "glue" together PVP-covered nanoparticles. At this point in our research, it was decided that the formation of aggregates is a positive phenomenon for this study since more magnetic nanoparticles can be attached to SiCWs to make them more magnetically active. From SEM imaging of SiCWs covered with MNPs (Figure 4 and SI Figure S1), we determined the size of the aggregates. The average agglomerate size was found to be 110 ± 80 nm in diameter. To determine the size, more than 500 MNP aggregates were identified and measured. Figure 3d shows a typical higher magnification TEM image of an agglomerate of the MNPs covered with PAA. It was difficult to estimate accurately the thickness of the PAA layer covering the aggregates from TEM results, but the thickness appeared to be on the level of 5 nm. TGA measurements for the modified MNPs indicated that weight loss associated with PAA was ~3 wt%. Based on the size of the aggregates and presuming that most of the polymer is attached to the surface of aggregate, it was estimated that ~ 3 nm PAA layer covered the surface of the nanoparticle agglomerates (see SI for details). This estimation is in general agreement with the TEM observations. Therefore, only a minor fraction of PAA participated in aggregate formation, while most of the polymer was located at the surface of the aggregates. After anchoring P2VP and PAA to SiCWs and MNPs, respectively, the nanomaterials were mixed at various MNP/SiCW ratios to target high packing density and uniform coverage of the whiskers with the nanoparticles. Specifically, the MNPs and SiCWs were mixed in four different 12 ACS Paragon Plus Environment

Page 13 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


weight ratios 1:2, 1:5, 1:8.6, and 1:10. In our procedure, the whisker suspension was added dropwise to the suspension of the nanoparticles to avoid agglomeration of the whiskers by bridging them with MNPs. SEM studies demonstrated that SiCWs can be readily covered with MNP aggregates to form MSiCWs via formation of a P2VP–PAA complex (Figure 4 and SI Figure S1). From the SEM images, we estimated the apparent thickness of the MNP layer on the surface of the whiskers for the different MNP/SiCW ratios. The thickness was measured as the distance from the whisker surface to the outer edge of the MNP layer. From each batch 6–8 whiskers were analyzed. For each whisker ~50 equally spaced measurements were conducted. The results of the measurements are presented in Figure 5a. Based on the results, the 1:5 MNP/SiCW weight ratio was selected for further investigation due to a significant level of anchoring of the nanoparticles with an average thickness of 111 ± 47 nm). This thickness was near the average size of the aggregates and, therefore, indicated formation of the aggregate monolayer on the whisker surface. SEM imaging also indicated that the MNP layer appeared to have a relatively low bulk density with a significant level of porosity. We confirmed the low density of the layer using Z contrast TEM imaging. In fact, the image in Figure 4c clearly illustrates that the MNP layer attached to the SiCWs had significant porosity originating from pores inside and between the nanoparticle aggregates. Most of the pores appear to be open. Therefore, they can be impregnated with an epoxy matrix and can serve as a nanocomposite interphase. We have foreseen that this interphase must possess a modulus value that is different than the modulus of the epoxy polymer and the modulus of the SiCWs. Magnetic Characterization of the MSiCWs We determined the parameters of the Langevin equation (Eq. 3) – mp and Mg – for the MNPs and MSiCWs. The parameter Mg was used to analyze the amount of magnetic material attached to the whiskers. Additionally, the parameter mp was compared between MNPs and MSiCWs to investigate the effect of magnetic interaction between individual nanoparticles on total magnetization of material. A typical plot of MNPs and MSiCWs magnetic moment vs. applied field along with their Langevin fit is presented in Figure 5b. The magnetization curve shows that there is no significant ferromagnetic coercivity or remnant magnetization of either MNPs or MSiCWs at the fields investigated. This is consistent with the expectation and is due to the superparamagnetic properties of iron oxide nanoparticles used in this research. Moreover, the average magnetic moment of each nanoparticle in the MNP samples and in the MSiCW samples 13 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 42

are practically the same, mp = (2.2±0.1)*10-19 Am2 and mp = (2.2±0.2)*10-19 Am2 , respectively. This indicates that the particle-to-particle interactions caused by the attachment of the MNPs to each other and to the SiCWs do not significantly affect the magnetic properties of the MNPs. The results also confirmed a significant level of attachment of the magnetic nanoparticles to the whisker surface. Since the magnetic properties of the nanoparticles are not affected by the attachment, the ratio of the Mg values for MNPs and MSiCWs can serve as a verification of the mass ratio between the nanoparticles and the silicone-carbide whiskers in the MSiCWs sample. The saturation mass magnetization for MNPs and SiCWs are Mg = 0.064±0.01 Am

2

g

and Mg = 0.0055±0.002

Am 2

g

,

respectively. From this measurement, it follows that the mass concentration of magnetic nanoparticles in MSiCWs sample is 9±4%. Because an ~110 nm layer of the nanoparticles covers the whisker that is ~ 420 nm in diameter, geometrical considerations imply that anchored to the whiskers layer possess high porosity on the level of 95% (see SI for details). Therefore, the magnetic measurements confirmed the significant porosity of the MNP layer.

Magnetic orientation of MSiCWs in epoxy resin Although the fabricated MSiCWs are highly magnetic, the epoxy system can present challenges for the orientation of the whiskers during resin curing. These challenges can arise due to local stoichiometry at the surface, whisker-to-whisker cross-linking, and variation in cross-linking density.48, 49 These events are directly related to polymerization of the epoxy. Therefore, in our study, the polymerization of the resin was monitored by following the viscosity change and reaction kinetics. The ball drop method was used to determine how the apparent viscosity of the epoxy resin changes in the course of curing at room temperature. Specifically, the change in drag was monitored as a glass sphere moved through the epoxy resin.35, 36 Figure 6a illustrates that the epoxy resin reached the gel-point at ~ 45 h, when the measured viscosity increased exponentially. To complement the viscosity measurement, DSC was used to monitor the extent of the curing reaction by the heat emitted from the polymerization of the epoxy. The degree of epoxy curing is plotted against the time of reaction in Figure 6b. The reaction progression exhibited a linear trend over time. The resin reached gelation at room temperature with around ~20% of the material 14 ACS Paragon Plus Environment

Page 15 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reacted after 45 h. In general, the viscosity of the epoxy resin increased significantly during the polymerization. In fact, it increased ~ 10 times within the first 10 h of the reaction (Figure 6a). Therefore, we decided to conduct the orientation of MSiCWs in the epoxy resin within the first 5 hours of the curing process. To identify the most effective way to create directional alignment of the whiskers, a magnetic field was applied using three different methods. The first method of orientation was the 'singular' method in which a magnetic field was applied once for 15 min at 1, 2, or 3.5 h (one application per sample) during the curing of the epoxy/MSiCWs composite. A second method of orientation was the 'pulsed' method that included a pulsed application of the magnetic field to a single sample for 15 min, four times at 1, 2, 3.5, and 5 h, during the epoxy polymerization. The third method of orientation was the 'continuous' alignment method, which involved the continuous application of the magnetic field to the composite for 4 h, from 1 h to 5 h, into the epoxy curing process. The 'singular,' 'pulsed,' and 'continuous' magnetic field applications were used to determine the best methodology for orientation of MSiCWs at 3 vol% in the epoxy matrix. After the samples were fully cured, they were placed in liquid nitrogen and broken in the longitudinal direction. The fracture surface was imaged with SEM (Figure 7). First, SEM imaging indicated that the MSiCWs were well dispersed in the epoxy resin without significant aggregation. The good dispersivity of the whiskers observed in the composite matrix is a necessary requirement for fabrication of nanocomposites with effective matrix–whisker stress transfer. Second, SEM results showed that the SiCWs were quite well oriented in the epoxy resin by the application of the magnetic field. We plotted a histogram of the level of the whisker orientation for the different methods of magnetic field applications (SI Figure S2). To obtain the histograms, the directionality of 450–600 individual whiskers from two to three images for each sample were analyzed. The orientation was analyzed by drawing a line through the whisker and employing software to determine the average orientation from the set net director. The comparison between the methods of magnetic alignment indicated that the 'pulsed' method of alignment resulted in the highest degree of orientation. For this method about 30% of the whiskers are oriented within 10o from the net detector, while for the ‘singular’ and ‘continuous’ methods less than 25% of the whiskers has such orientation (SI Figure S2). If larger population of the whiskers is considered for the ‘pulsed’ and ‘continuous’ methods more than 70% of the whiskers has orientation within 40o from the net detector, while for the ‘pulsed’ method the 15 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 42

orientation is on the level of 60%. In general, the orientation for the 'pulsed' alignment had an average orientation of 24o from the net director. In contrast, the average orientation for the 'singular' and 'continuous' methods were 36o and 30o, respectively. We suggest that in the 'singular' alignment method, local internal stresses developed during the curing of the epoxy and the associated matrix relaxation decreased the level of the orientation after the magnetic field was released. In addition, the total orientation time in this method was significantly shorter than for the 'pulsed' and 'continuous' methods. In the 'continuous' method the sample was influenced by the magnetic field for an extended period enabling MSiCWs to attain higher level of alignment and preventing possible relaxation of the whiskers. However, with an extended application of the magnetic field the migration of the whiskers to the magnets was initiated. Therefore, we concluded that the migration could disrupt uniformity of the distribution of the whiskers in the epoxy matrix. Conversely, the short-term application of the magnetic field in the 'pulsed' method allowed high orientation levels to be attained due to the sufficiently long cumulative time of interaction with the magnetic field and prevention of excessive relaxation resulting from repetitive application. Hence, the 'pulsed' method was selected for fabrication of the nanocomposite materials in this research.

Magnetic characterization of the nanocomposites Magnetization curves were measured for composites consisting of nonoriented and magnetically oriented MSiCWs with a pure epoxy sample as a control. The magnetic response of a pure epoxy sample was measured to be negligibly small, while the nanocomposites were observed to be significantly magnetic with superparamagnetic characteristics. To characterize the magnetic anisotropy, two magnetic measurements were performed for each sample: one measurement with the in-plane magnetic field oriented along one side of the sample and the other measurement with the in-plane magnetic field oriented along the other side of the sample. Thus, for the oriented nanocomposites, the measurements were conducted along the direction of orientation of the nanorods (mp|| , Mg|| ) and perpendicularly to the direction of orientation of the nanorods, (mp , Mg ). Results of the magnetic measurements are presented in Table 1. Typical plots of the magnetization curves for nonoriented and oriented samples are presented in Figure 8, however, the graphical representations do not allow one to distinguish the results.


Page 17 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For quantitative evaluation of these differences, one can introduce the order parameters as  m p  m p||  m p  m p|| and  M g  M g ||  M g  M g|| . These parameters provide a metric for the

nanorod orientation in the sample. As a reference, an estimation of magnetic anisotropy was made for nonoriented samples. Following the same procedure, we obtained  m p = 0.02 and  M g = 0.03. These values define the accuracy of our measurements of anisotropy. For oriented samples, the order parameters were measured as  m p = 0.05±0.02 and  M g = 0.0±0.03. This suggests that the nanorod orientation does not affect the magnetic properties of the composite at high fields when the magnetic moments are co-aligned with the field, but does demonstrate an anisotropy at the lower fields when the magnetic moments are not completely aligned with the field. These weak fields correspond to the Langevin dependence shown in Figure 8 at the fields roughly below 0.4 Tesla.

Interphase formation Thermal transitions and mechanical behavior of nanocomposite materials have been reported to significantly deviate from the values recorded for the unfilled polymer matrix.4, 50-56 The observed positive or negative deviations are associated with the formation of an interphase in the vicinity of the nanoscale filler. The polymer chains involved in the interphase demonstrated fundamental changes in mobility (relaxation time) due to interaction with the filler surface. In the nanocomposites (possessing a high ratio between the surface of the filler and the volume of the matrix), extended interphase zones were found, which can percolate through the entire nanocomposite and dominate the properties of the material.48 In addition, for thermoset materials (like epoxy resin) synthesized in the presence of the nanoscale filler, chemical reaction and/or interaction between the monomers and the surface can also change the chemical structure of the polymer materials, such as cross-linking density. In general, formation of extended interphases has been reported for epoxy nanocomposites. Using microscopy and nanoindentation methods the extent of the interfacial region in the composites was determined to be between tens of nanometers and several microns.48, 50, 52, 53, 57-59 If the surface of the filler was interacting with the epoxy matrix, formation of interphases with thicknesses between hundreds of nanometers and several microns were reported.57, 60, 61 In terms of mechanical properties, the interphase region can demonstrate a higher or lower modulus in 17 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 42

comparison to the bulk polymer. Within a single interphase, variation of the modulus from lower to higher values compared to the modulus for the bulk material was reported as well.57 We note that in some examples the glass transition temperature (Tg) of the epoxy composite material increased through incorporation of the nanoscale filler, indicating formation of an interphase possessing a higher glass temperature.48,

50

Conversely, many examples of depressed Tg are

reported for epoxy composites.78 In this research, formation of the epoxy network near the MSiCWs is affected by the presence of PAA inside and on the surface of the MNP layer. Indeed, carboxyl groups of the PAA react with epoxy groups in the presence of the basic catalyst, which is used to cure the epoxy resin in this research.62 This reaction is capable of initiating alternative chemical pathways for the formation of the epoxy network in the composite. Therefore, the degree of curing of the composite was monitored with DSC and compared with the curing of an unfilled epoxy resin (SI Figure S3a). The obtained results demonstrated that in the presence of the whiskers, the epoxy resin was cured ~1.2 times faster than the neat epoxy resin. We also determined the activation energy for the epoxy curing reaction using the Kissinger method (SI Figure S3b). Ea was determined to be 78.2 J/mol and 72.5 J/mol for the curing of the neat epoxy and the MSiCW/epoxy composite, respectively. Finally, ATR-FTIR was used to evaluate if addition of the MSiCWs influenced the chemical mechanism of the epoxy curing.63 As primary components of the epoxy resin system, epoxy and anhydride IR peaks associated with these substances were followed during the course of curing (SI Figure S5). The neat epoxy resin was determined to have a higher rate of conversion of epoxy groups than the composite material during curing at room temperature for 48 h. During the same time of curing, the neat epoxy exhibited a lower consumption of anhydride than the composite system. Hence, the functionalized whiskers caused the curing process to change from its original chemical pathway toward a higher level of reaction of the DDSA anhydride groups with the acrylic acid and/or epoxy groups64. These results confirmed that the whiskers chemically reacted with the epoxy resin and can be considered to be an active reagent and accelerant for the curing reaction. Because the influence of the MSiCWs on the curing process is significant, we assume the formation of interphases in the nanocomposites with thicknesses between hundreds of nanometers and several microns. We can identify at least three distinct structural domains in the nanocomposite material (Figure 9a): (A) an MNP layer impregnated with epoxy polymer with the thickness of 18 ACS Paragon Plus Environment

Page 19 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


approximately 110 nm and the epoxy matrix near the layer, (B) a domain located next to the domain A, and (C) an epoxy network unaffected by the presence of the MSiCWs. The domains A and B constitute the interphase in the nanocomposite. From straightforward geometrical considerations (see SI for details), we evaluated how the volume fraction of the interphase in the polymer matrix of the MSiCW/epoxy composite will vary as a function of its thickness for a perfectly oriented composite (Figure 9b). It is obvious that in our system, a significant fraction of the epoxy matrix is part of the interphase zone where the mobility of polymer chains is altered.

Thermal characterization of the nanocomposites We conducted DCS studies of the nanocomposites to observe the behavior of the epoxy materials in the Tg region. Figure 10a illustrates that the behavior of the "as fabricated" neat epoxy and nanocomposite materials are quite different in the Tg region. The neat epoxy and nonoriented composite clearly demonstrated enthalpic relaxations associated with the transition from a nonequilibrium conformational state of the polymer chains to an equilibrium state upon annealing.50 The midpoint Tg value of the neat epoxy was observed to be ~ 60oC and is approximately 13oC higher than that of the nonoriented composite. The observed result indicated formation of an extended interphase with a lower Tg value near the MSiCWs. The magnetic orientation of the whiskers dramatically changes the behavior of the material upon heating. The oriented composite has a Tg transition where the midpoint Tg value is only about 2oC less than that of the neat epoxy. The DSC measurements also revealed that the oriented composite had no visible enthalpic relaxation and had an exceptionally extended (~30oC) Tg region. The region starts at the similar to nonoriented composite temperature and ends at the temperature higher than that for the neat epoxy. The observed results point to the formation of an extended interphase where both higher Tg and lower Tg zones are present. Based on our DSC data we propose a coarse grain model where the low Tg region (domain A with polymer chains possessing higher mobility than bulk epoxy) is located closer to the whisker surface and is associated with epoxy network disruption caused by the highly porous layer of nanoparticles and the high concentration of carboxylic groups of PAA (Figure 9a). Domain B has a higher Tg value than the neat epoxy with chains having lower mobility caused by connectivity of the cross-linked chains to the solid surface via the chains involved in domain A. Such behavior of 19 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 42

the interfacial region in epoxy composites has been experimentally observed via nanomechanical measurements.57 It has appeared that the magnetic orientation of MSiCW causes formation of more extended domain B. Therefore, in the oriented composite, where the higher Tg value of domain B compensates for the lower Tg value of domain A, the glass transition temperature approaches that of the neat epoxy material. In the nonoriented nanocomposite, having a smaller extent of domain B, the Tg compensation does not occur. Figure 10b demonstrates the DSC results for the second run, in which the fabrication history of the materials is erased via annealing. The DSC traces now show no visible signs of enthalpic relaxations indicating that polymer chains in the epoxy approach their equilibrium conformations. In general, the midpoint Tg values are similar to the ones obtained for "as fabricated" materials. The Tg region of the oriented composite materials is reduced in breadth and is comparable to the Tg value of the neat epoxy matrix, whereas the nonoriented composites again demonstrated lower Tg values. The obtained results indicated that the proposed coarse-grain model for the composite materials (Figure 9a) can be applied for the annealed samples as well, and the differences in Tg values of different materials are not artefacts of the fabrication procedure. Thermomechanical properties of the nanocomposites The neat epoxy, the epoxy with randomly orientated MSiCWs, and the directionally oriented MSiCW nanocomposites were tested by tensile DMA to evaluate their mechanical properties at different temperatures. The dependence of the storage modulus, E/ for the materials is presented in Figure 11a. It is evident that below Tg the composites exhibited higher modulus than the pure epoxy matrix pointing to the efficient stress transfer from the matrix to the whiskers in the mechanical measurements. The room temperature value of the modulus for the neat epoxy was found to be 1.52 ± 0.02 GPa. The randomly oriented nanocomposite had a tensile modulus of 1.9 ± 0.1 GPa. The material with aligned MSiCWs had a tensile modulus of 2.2 ± 0.1 GPa measured in the direction of the fiber orientation. These results reveal approximately 25% improvement was attained by just adding the filler and an approximately 45% increase in the modulus was realized by adding the MSiCWs and orientating them. We compared the obtained storage modulus values with those predicted by the Halpin-Tsai model that is commonly used for estimation of nanocomposite mechanical properties.17, 56, 65 The Halpin-Tsai model is a semi-empirical equation for prediction of the modulus of a discontinuous 20 ACS Paragon Plus Environment

Page 21 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


composite structure. The model is based on the assumptions that only the matrix sustains the axial load and transmits stress to the fiber through shear stress and that all fibers are well dispersed in the matrix.66 The Halpin-Tsai model equations express the longitudinal and transverse engineering moduli in the following form:

1  f  Eo  Em    1  f 



(Eq. 4)

E f / Em  1 E f / Em  

(Eq. 5)

where, Eo is the predicted composite modulus; Em is the matrix modulus; Ef is the fibrous inclusions modulus (450 GPa for SiCW25); f is the fiber volume fraction; η fiber-to-load transfer parameter for the Halpin-Tsai equation; and ξ is the shear coefficient (~ two times the aspect ratio of a fiber for the longitudinal, fiber direction modulus; and 2 for the transversal, perpendicular to fiber direction, modulus). The average aspect ratio for the whiskers used in this research is approximately 10. This equation, therefore, predicts the longitudinal modulus of our composite material with aligned whiskers to be 2.4 GPa, while the result from our experimentation is 2.2 GPa. The calculated theoretical and experimental results are in a good agreement. We associate the slightly lower values for the experimental modulus with non-ideal (24o instead of the ideal 0o) orientation of MSiCWs in the material. The modulus for randomly oriented fibers can be estimated using Eq. 6:67

3 5 E  E  E o 8 l 8 t

(Eq. 6)

where El and Et are the longitudinal and transverse moduli, respectively. We have already estimated the longitudinal modulus, and the transverse modulus was calculated by equations (4) and (5) assuming ξ equals 2.65 Next, using Eq. 6, the modulus for the composite with randomly distributed whiskers was found to be approximately 1.9 GPa in excellent agreement with the DMA results. In general, the comparison between the calculated and experimental values for oriented



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 42

and nonoriented epoxy/MSiCW composites clearly indicated that efficient matrix–whisker stress transfer existed in the nanocomposites. The temperature dependence of tan δ is shown in Figure 11b. The position of the tan δ peak indicates the glass transition temperature of the material when variable stress is applied to the sample. As in DCS measurements, the lowest Tg value (67oC) was observed for the epoxy composite with nonoriented MSiCWs, which is significantly lower than that of the neat epoxy material (Tg = 77oC). However, the Tg value measured by DMA for the nanocomposite with oriented MSiCWs (83oC) is higher indicating that mobility of polymer chains is restrained in the whisker direction in comparison with the bulk epoxy. We also note that the width of the tan δ peak is significantly greater for the composites with asymmetry into the higher temperature region. The width of the transition is related qualitatively to the homogeneity of the epoxy network (variation in cross-linked density) and, therefore, to the variation in mobility of polymer chains.68-70 Therefore, as was already determined from the DSC measurements, zones with noticeable differences in network density are present in the nanocomposites. The asymmetry of the dependence of the tan δ peak on temperature corroborates formation of interphase domains possessing chains with mobility changed by the presence of the whiskers. Noticeable differences in the height of the tan δ peaks are observed for the materials as well. The peak height is a measure of interfacial stress transfer in composite materials, where smaller values point toward a higher degree of elastic response to the stress applied.68 The oriented composite has the lowest tan δ peak, and the nonoriented composites have the highest tan δ peak. This behavior of the oriented nanocomposite materials is indicative of lower loss (more elastic) material, if compared with the neat epoxy and nonoriented composite. In general, DMA results confirm our suggestion that an extended interphase with variable properties is formed in the nanocomposites and that the interphase formation is affected by the orientation of the MSiCWs in a positive way leading to a nanocomposite with superior mechanical properties. Conclusions: We demonstrated effective methodology for fabrication of MSiCWs exhibiting a substantial magnetic response in a polymerizing epoxy resin with application of only ~ 20mT (200G) magnetic field. The whiskers chemically reacted with the epoxy resin causing formation of an extended interphase near the boundary of the whiskers. The 2D aligned SiC whiskers demonstrated positive effects on the behavior of epoxy-based nanocomposites. Namely, MSiCWs—aligned in 22 ACS Paragon Plus Environment

Page 23 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the epoxy matrix by the magnetic field—enhanced the thermo-mechanical properties of the materials significantly above that of the neat epoxy and the epoxy nanocomposite with randomly oriented whiskers. In general, we expect that the magnetic orientation will permit fabrication of advanced SiCW-based nanocomposite materials including multilayered materials with whiskers oriented in variable directions. The described methodology possesses a high degree of flexibility allowing for tuning the magnetic properties, surface roughness, and surface energy of the whiskers.

Acknowledgments. This project was funded by the Air Force Office of Scientific Research, Contract FA9550-12-1-0459. The authors would like to express their appreciation for the helpful suggestions and support of their contract monitor, Dr. Ali Sayir. The authors thank to Prof. Olga Kuksenok (Clemson University) for helpful discussion. Supporting Information Available. (S1) Calculations of the thickness of (macro)molecular layer attached to the surface of whiskers and nanoparticles from TGA results. (S2) SEM images of MSiCW prepared using 1:10 and 1:2 MNP/SiCW weight ratio. (S3) The level of MSiCWs orientation in epoxy composite for the different methods of magnetic field application. (S4) The degree of epoxy curing versus the time of the reaction and determination of activation energy (Ea) of the epoxy curing using DSC. (S5) Conversion of epoxy and anhydride functional groups in course of the epoxy curing as determined with ATR-FTIR. (S6) Calculation of MNP volume fraction in the NMP nanoparticle layer covering SiCWs from magnetic measurements. (S7) Calculation of interphase volume fraction for composite with perfectly oriented in epoxy matrix MSiCWs as a function of the interphase thickness. This material is available free of charge via the Internet at http://pubs.asc.org.

References: 1.

Erb, R. M.; Martin, J. J.; Soheilian, R.; Pan, C. Z.; Barber, J. R., Actuating Soft Matter

with Magnetic Torque. Adv. Funct. Mat. 2016, 26, 3859-3880. 2.

Hull, D.; Clyne, T. W., An Introduction to Composite Materials. Second ed.; Cambridge

University Press: Cambridge, 1996; p 326. 3.

Freitas, G.; Magee, C.; Dardzinski, P.; Fusco, T., Fiber Insertion Process for Improved

Damage Tolerance in Aircraft Laminates. J. Adv. Mater. 1994, 25, 36-43. 23 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4.

Page 24 of 42

Hussain, F.; Hojjati, M.; Okamoto, M.; Gorga, R. E., Review Article: Polymer-Matrix

Nanocomposites, Processing, Manufacturing, and Application: An Overview. J. Compos. Mater. 2006, 40, 1511-1575. 5.

Erb, R. M.; Libanori, R.; Rothfuchs, N.; Studart, A. R., Composites Reinforced in Three

Dimensions by Using Low Magnetic Fields. Science 2012, 335, 199-204. 6.

Marins, J. A.; Mija, A.; Pin, J.-M.; Giulieri, F.; Soares, B. G.; Sbirrazzuoli, N.; Lançon,

P.; Bossis, G., Anisotropic Reinforcement of Epoxy-Based Nanocomposites with Aligned Magnetite–Sepiolite Hybrid Nanofiller. Compos. Sci. Technol. 2015, 112, 34-41. 7.

Gungor, S.; Bakis, C. E., Anisotropic Networking of Carbon Black in Glass/Epoxy

Composites Using Electric Field. J. Compos. Mater. 2014, 49, 535-544. 8.

Denver, H.; Heiman, T.; Martin, E.; Gupta, A.; Borca-Tasciuc, D.-A., Fabrication of

Polydimethylsiloxane Composites with Nickel Nanoparticle and Nanowire Fillers and Study of Their Mechanical and Magnetic Properties. J. Appl. Phys. 2009, 106, 064909. 9.

Malkina, O.; Mahfuz, H.; Sorge, K.; Rondinone, A.; Chen, J.; More, K.; Reeves, S.;

Rangari, V., Magnetic Alignment of SWCNTs Decorated with Fe3O4 to Enhance Mechanical Properties of SC-15 Epoxy. AIP Adv. 2013, 3, 042104. 10.

Li, D.; Liu, Z.; Al-Haik, M.; Tehrani, M.; Murray, F.; Tannenbaum, R.; Garmestani, H.,

Magnetic Alignment of Cellulose Nanowhiskers in an All-Cellulose Composite. Polym. Bull. 2010, 65, 635-642. 11.

Das, A.; Raffi, M.; Megaridis, C.; Fragouli, D.; Innocenti, C.; Athanassiou, A., Magnetite

(Fe3O4)-Filled Carbon Nanofibers as Electro-Conducting/Superparamagnetic Nanohybrids and their Multifunctional Polymer Composites. J. Nanopart. Res. 2015, 17, 1. 12.

Pech-May, N. W.; Vales-Pinzon, C.; Vega-Flick, A.; Cifuentes, A.; Oleaga, A.; Salazar,

A.; Alvarado-Gil, J. J., Study of the Thermal Properties of Polyester Composites Loaded with Oriented Carbon Nanofibers Using the Front-Face Flash Method. Polym. Test. 2016, 50, 255261. 13.

Wang, M. S.; He, L.; Zorba, S.; Yin, Y. D., Magnetically Actuated Liquid Crystals. Nano

Lett. 2014, 14, 3966-3971. 14.

Moaseri, E.; Karimi, M.; Baniadam, M.; Maghrebi, M., Improvements in Mechanical

Properties of Multi-Walled Carbon Nanotube-Reinforced Epoxy Composites Through Novel


Page 25 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Magnetic-Assisted Method for Alignment of Carbon Nanotubes. Compos. Part A-Appl. S. 2014, 64, 228-233. 15.

Yamamoto, N.; Manohara, H.; Platzman, E., Magnetically Anisotropic Additive for

Scalable Manufacturing of Polymer Nanocomposite: Iron-Coated Carbon Nanotubes. Mater. Res. Express 2016, 3, 025004. 16.

Yan, H. Y.; Tang, Y. X.; Long, W.; Li, Y. F., Enhanced Thermal Conductivity in

Polymer Composites with Aligned Graphene Nanosheets. J. Mater. Sci. 2014, 49, 5256-5264. 17.

Libanori, R.; Erb, R. M.; Studart, A. R., Mechanics of Platelet-Reinforced Composites

Assembled Using Mechanical and Magnetic Stimuli. ACS Appl. Mater. Interfaces 2013, 5, 10794-10805. 18.

Trotsenko, O.; Tokarev, A.; Gruzd, A.; Enright, T.; Minko, S., Magnetic Field Assisted

Assembly of Highly Ordered Percolated Nanostructures and their Application for Transparent Conductive Thin Films. Nanoscale 2015, 7, 7155-7161. 19.

Lv, F. Z.; Xu, L. N.; Xu, Z. X.; Fu, L. L.; Zhang, Y. H., Fabrication and Characterization

of the Orientated MMT/PI Composite Films via Relatively Low Magnetic Field. J. Appl. Polym. Sci. 2015, 132, 41224. 20.

Ding, L.; Liu, L. P.; Li, P. G.; Lv, F. Z.; Tong, W. S.; Zhang, Y. H., Dielectric Properties

of Graphene-Iron Oxide/Polyimide Films with Oriented Graphene. J. Appl. Polym. Sci. 2016, 133, 43041. 21.

Kim, K.; Ju, H.; Kim, J., Vertical Particle Alignment of Boron Nitride and Silicon

Carbide Binary Filler System for Thermal Conductivity Enhancement. Compos. Sci. Technol. 2016, 123, 99-105. 22.

Park, C.; Wilkinson, J.; Banda, S.; Ounaies, Z.; Wise, K. E.; Sauti, G.; Lillehei, P. T.;

Harrison, J. S., Aligned Single-Wall Carbon Nanotube Polymer Composites Using an Electric Field. J. Polym. Sci. Part B Polym. Phys. 2006, 44, 1751-1762. 23.

Escobar, S. A. M.; Merino, C. A. I.; Meza, J. M. M., Mechanical and Thermal Behavior

of Polyvinyl Alcohol Reinforced with Aligned Carbon Nanotubes. Materia-Rio De Janeiro 2015, 20, 794-802. 24.

Yamaguchi, K.; Horaguchi, I., Development of Directionally Aligned SiC Whisker

Wheel. Precis. Eng. 1995, 17, 5-9.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

25.

Page 26 of 42

Compton, B. G.; Lewis, J. A., 3D-Printing of Lightweight Cellular Composites. Adv.

Mater. 2014, 26, 5930-5935. 26.

Barthelat, F.; Li , C. M.; Comi, C.; Espinosa, H. D., Mechanical Properties of Nacre

constituents and their Impact on Mechanical Properties. J. Mater. Res. 2006, 21, 1977-1986. 27.

Studart, A. R., Additive Manufacturing of Biologically-Inspired Materials. Chem. Soc.

Rev. 2016, 45, 359-376. 28.

Yan, H. Y.; Wang, R. R.; Li, Y. F.; Long, W., Thermal Conductivity of Magnetically

Aligned Graphene-Polymer Composites with Fe3O4-Decorated Graphene Nanosheets. J. Electron. Mater. 2015, 44, 658-666. 29.

Le Ferrand, H.; Bolisetty, S.; Demirors, A. F.; Libanori, R.; Studart, A. R.; Mezzenga, R.,

Magnetic Assembly of Transparent and Conducting Graphene-Based Functional Composites. Nat. Commun. 2016, 7, 12078. 30.

Martin, J. J.; Riederer, M. S.; Krebs, M. D.; Erb, R. M., Understanding and Overcoming

Shear Alignment of Fibers during Extrusion. Soft Matter 2015, 11, 400-405. 31.

Yuan, Z.; Yu, J.; Rao, B.; Bai, H.; Jiang, N.; Gao, J.; Lu, S., Enhanced Thermal

Properties of Epoxy Composites by Using Hyperbranched Aromatic Polyamide Grafted Silicon Carbide Whiskers. Macromol. Res. 2014, 22, 405-411. 32.

Harris, G. L., Properties of silicon carbide. INSPEC, Institution of Electrical Engineers:

London, 1995. 33.

Afaghani, J. E. D.; Yamaguchi, K.; Horaguchi, I.; Nakamoto, T., Whisker Behaviors and

Tool Wear in Cutting of Unidirectional SIC Whisker-Reinforced Plastic. Wear 1996, 195, 223231. 34.

Abenojara, J.; Encinas, N.; del Real, J. C.; Martínez, M. A., Polymerization Kinetics of

Boron Carbide/Epoxy Composites. Thermochim. Acta 2014, 575, 144-150. 35.

Stokes, G. G., On the Effect of the Internal Friction of Fluids on the Motion of

Pendulums. Trans. Cambridge Phil. Soc. 1851, 9, 8–106. 36.

Lamb, H., Hydrodynamics. Cambridge University Press (reprinted by Dover, 1945):

1932; Vol. 6. 37.

Gu, Y.; Burtovyy, R.; Townsend, J.; Owens, J. R.; Luzinov, I.; Kornev, K. G., Collective

Alignment of Nanorods in Thin Newtonian Films. Soft Matter 2013, 9, 8532-8539.


Page 27 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


38.

Chikazumi, S. O.; Graham, C. D., Physics of ferromagnetism. 2nd ed.; Oxford University

Press: Oxford ; New York, 2009; p 655. 39.

Gu, Y.; Kornev, K. G., Ferromagnetic Nanorods in Applications to Control of the In-

Plane Anisotropy of Composite Films and for In Situ Characterization of the Film Rheology. Adv. Funct. Mat. 2016, 26, 3796-3808. 40.

Aprelev, P.; Gu, Y.; Burtovyy, R.; Luzinov, I.; Kornev, K. G., Synthesis and

Characterization of Nanorods for Magnetic Rotational Spectroscopy. J. Appl. Phys. 2015, 118, 074901. 41.

Han, H. W.; Liu, H. S., Characterization of Vapour Deposited Products in Furnace Tube

During SIC Synthesis from Carbonized Rice Hulls. Ceram. Int. 1999, 25, 631-637. 42.

Luzinov, I.; Julthongpiput, D.; Liebmann-Vinson, A.; Cregger, T.; Foster, M. D.;

Tsukruk, V. V., Epoxy-Terminated Self-Assembled Monolayers: Molecular Glues for Polymer Layers. Langmuir 2000, 16, 504-516. 43.

Fujimori, K.; Trainor, G.; Costigan, M., Complexation and Rate of Polymerization of

Acrylic Acid and Methacrylic Acid in the Presence of Poly(4-Vinylpyridine) in Dilute Methanol Solution. J. Polym. Sci. 1984, 22, 2479-2487. 44.

Zdyrko, B.; Klep, V.; Li, X. W.; Kang, Q.; Minko, S.; Wen, X. J.; Luzinov, I., Polymer

Brushes as Active Nanolayers for Tunable Bacteria Adhesion. Mat. Sci. Eng. C-BIOMIM 2009, 29, 680-684. 45.

Cheng , Y.; Chen, Z.; Zhu, M.; Lu, Y., Polyacrylic Acid Assisted Assembly of Oxide

Particles and Carbon Nanotubes for High-Performance Flexible Battery Anodes. Adv. Energy Mater. 2015, 5, 1401207. 46.

Teja, A. S.; Koh, P.-Y., Synthesis, Properties, and Applications of Magnetic Iron Oxide

Nanoparticles. Prog. Cryst. Growth Charact. Mater. 2009, 55, 22-45. 47.

Lau, C.; Mi, Y., A study of blending and complexation of poly(acrylic acid)/poly(vinyl

pyrrolidone). Polymer 2002, 43, 823-829. 48.

Putz, K. W.; Palmeri, M. J.; Cohn, R. B.; Andrews, R.; Brinson, L. C., Effect of Cross-

Link Density on Interphase Creation in Polymer Nanocomposites. Macromolecules 2008, 41, 6752-6756. 49.

Kim, B.; Choi, J.; Yang, S.; Yu, S.; Cho, M., Influence of Crosslink Density on the

Interfacial Characteristics of Epoxy Nanocomposites. Polymer 2015, 60, 186-197. 27 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

50.

Page 28 of 42

Lu, H. B.; Nutt, S., Restricted Relaxation in Polymer Nanocomposites Near the Glass

Transition. Macromolecules 2003, 36, 4010-4016. 51.

Karl W. Putz; Marc J. Palmeri; Rachel B. Cohn; Rodney Andrews; L. Catherine Brinson,

Effect of Cross-Link Density on Interphase Creation in Polymer Nanocomposites. Macromolecules 2008, 41, 6752-6756. 52.

Rittigstein, P.; Priestley, R. D.; Broadbelt, L. J.; Torkelson, J. M., Model Polymer

Nanocomposites Provide an Understanding of Confinement Effects in Real Nanocomposites. Nat. Mater. 2007, 6, 278-282. 53.

Bansal, A.; Yang, H. C.; Li, C. Z.; Cho, K. W.; Benicewicz, B. C.; Kumar, S. K.;

Schadler, L. S., Quantitative Equivalence Between Polymer Nanocomposites and Thin Polymer Films. Nat. Mater. 2005, 4, 693-698. 54.

Zhou, Y. X.; Hosur, M.; Jeelani, S.; Mallick, P. K., Fabrication and Characterization of

Carbon Fiber Reinforced Clay/Epoxy Composite. J. Mater. Sci. 2012, 47, 5002-5012. 55.

Putz, K. W.; Mitchell, C. A.; Krishnamoorti, R.; Green, P. F., Elastic Modulus of Single-

Walled Carbon Nanotube/Poly(Methyl Methacrylate) Nanocomposites. J. Polym. Sci. Part B Polym. Phys. 2004, 42, 2286-2293. 56.

Putz, K.; Krishnamoorti, R.; Green, P. F., The Role of Interfacial Interactions in The

Dynamic Mechanical Response of Functionalized SWNT-PS Nanocomposites. Polymer 2007, 48, 3540-3545. 57.

Steffan, C. E.; Liu, H.; Winter, R. M.; Cabibil, H.; Lozano, J.; Celio, H.; White, J. M.;

Houston, J. E., The Relationship Between Nanomechanical and Chemical Properties in Polymer Composites. In Proceedings of the Society of Experimental Mechanics IX International Congress on Experimental Mechanics, 2000; pp 920-923. 58.

Diez-Pascual, A. M.; Gomez-Fatou, M. A.; Ania, F.; Flores, A., Nanoindentation

Assessment of the Interphase in Carbon Nanotube-Based Hierarchical Composites. J. Phys. Chem. C 2012, 116, 24193-24200. 59.

Yedla, S. B.; Kalukanimuttam, M.; Winter, R. M.; Khanna, S. K., Effect of Shape of the

Tip in Determining Interphase Properties in Fiber Reinforced Plastic Composites Using Nanoindentation. J. Eng. Mater. Technol.-Trans. ASME 2008, 130, 041010.


Page 29 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


60.

Williams, J. G.; Li, F. P.; Miskioglu, I., Characterization of the Interphase in

Epoxy/Aluminum Bonds Using Atomic Force Microscopy and a Nano-Indenter. J. Adhes. Sci. Technol. 2005, 19, 257-277. 61.

Downing, T. D.; Kumar, R.; Cross, W. M.; Kjerengtroen, L.; Kellar, J. J., Determining

the Interphase Thickness and Properties in Polymer Matrix Composites Using Phase Imaging Atomic Force Microscopy and Nanoindentation. J. Adhes. Sci. Technol. 2000, 14, 1801-1812. 62.

Zdyrko, B.; Hoy, O.; Kinnan, M. K.; Chumanov, G.; Luzinov, I., Nano-Patterning with

Polymer Brushes via Solvent-Assisted Polymer Grafting. Soft Matter 2008, 4, 2213-2219. 63.

Chu, F. J.; Robertson, R. E., The Effect of Adherends on the Curing of an Epoxy

Adhesive. J. Adhes. 1995, 53, 149-166. 64.

Kolár, F.; Svítilová, J., Kinetics and Mechanism of Curing Epoxy/Anhydride Systems.

Acta Geodyn. Geomater. 2007, 4, 85-92. 65.

Fornes, T. D.; Paul, D. R., Modeling Properties of Nylon 6/Clay Nanocomposites Using

Composite Theories. Polymer 2003, 44, 4993-5013. 66.

Halpin, J. C.; Kardos, J. L., Halpin-Tsai Equations - Review. Polym. Eng. Sci. 1976, 16,

344-352. 67.

Nielsen, L. E., Mechanical properties of polymers and composites. Vol. 2. Dekker: New

York, 1974; p 462. 68.

Vennerberg, D.; Rueger, Z.; Kessler, M. R., Effect of Silane Structure on The Properties

of Silanized Multiwalled Carbon Nanotube-Epoxy Nanocomposites. Polymer 2014, 55, 18541865. 69.

Miyagawa, H.; Misra, M.; Drzal, L. T.; Mohanty, A. K., Biobased Epoxy/Layered

Silicate Nanocomposites: Thermophysical Properties and Fracture Behavior Evaluation. J. Polym. Environ. 2005, 13, 87-96. 70.

Miyagawa, H.; Rich, M. J.; Drzal, L. T., Amine-Cured Epoxy/Clay Nanocomposites. I.

Processing and Chemical Characterization. J. Polym. Sci. Part B Polym. Phys. 2004, 42, 43844390.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 42

Table 1. Magnetic properties of nanocomposites. Type of sample and measurements

Magnetic moment of each particle, mp [ Am2 ]

Mass magnetization, 2 Mg [ Am g ]

Nonoriented average – direction 1

1.92*10-19

3.78*10-4

Nonoriented average – direction 2

1.96*10-19

3.68*10-4

Oriented average – parallel

1.92*10-19

4.84*10-4

Oriented average – perpendicular

1.83*10-19

4.86*10-4


Page 31 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. The schematic representation of MSiCW.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 42

Grafted layer

(a)

(b)

Figure 2. (a) TEM images of SiCWs modified with P2VP (scale bar- 300 nm). (b) Higher modification image of the area marked on the image (a), (scale bar- 100 nm).


Page 33 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(a)

(b)

(c)

(d)

Figure 3: (a) TEM image of the bare MNPs; (b) SEM image of MNP aggregate modified with PAA; (c) TEM image of MNP aggregate modified with PAA; (d) Higher modification image of the area marked on the image (c).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(a)

Page 34 of 42

(b)

(c)

Figure 4. SEM (a and b) and TEM (c) images of MSiCW prepared using 1:5 weight ratio between MNPs and SiCWs. Scale bars: (a, c) 1 µm and (b) 600 nm.


1:2

200 1:5

150 1:8.6

100

1:10

50

(a) 0 0.05

0.10

0.15

0.20

0.25

0.30

0.35

Weight fraction of MNPs Magnetic Moment (*10-5 Am2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Thickness of MNP layer, nm

Page 35 of 42

1.0 MNPs Data Fit

0.5 MSiCWs Data Fit

0.0

-0.5

(b) -1.0 -0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

Magnetic Field (T) Figure 5. (a) The average thickness of the MNPs layer covering SiCW as a function of the targeted weight fraction of the particles in MSiCWs. The MNP/SiCW weight ratio used for the MSiCWs fabrication is marked on the graph; (b) magnetization curves for MNPs and MSiCWs (prepared with the 1:5 ratio). 35 ACS Paragon Plus Environment


Apparent Viscosity (Pa*s)

1000 (a)

100

10

1 0.5

1

5

10

50

Time of epoxy curing (hrs)

Degree of epoxy curing (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 42

(b)

25 20 15 10 5 0 0

10

20

30

40

50

Time of epoxy curing (hrs) Figure 6. The apparent viscosity of the epoxy resin (a) and degree of epoxy curing (b) versus the curing time at room temperature.


Page 37 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(a)

(b)

(c)

Figure 7. SEM images of fracture surface of epoxy/MSiCW composite materials: (a) without the magnetic orientation. (b and c ) with the ‘pulsed’ magnetic orientation of the whiskers. Scale bar for (c) is 20 microns. 37 ACS Paragon Plus Environment

Page 38 of 42

Nonoriented MSiCWs 2 Direction 1 Direction 2 1

0

-1

-2 -0.8

-0.4

0.0

0.4

0.8

Magnetic Field (T) Oriented MSiCWs

Magnetic Moment (*10-6 Am2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Magnetic Moment (*10-6 Am2)


1.0 Parallel Perpendicular 0.5

0.0

-0.5

-1.0 -0.8

-0.4

0.0

0.4

0.8

Magnetic Field (T) Figure 8. Magnetometer measurements of the epoxy composites with nonoriented and magnetically oriented MSiCWs.


Page 39 of 42

(a)

Domain A

SiCW Domain B Domain C

Interphase volume fraction

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.0

(b)

0.8

0.6

0.4

0.2

0.0 0.0

0.2

0.4

0.6

0.8

1.0

Interphase thickness (microns) Figure 9. (a) Schematic representation of the structural domains in the composite material. Domain A: MNP layer impregnated with epoxy polymer; Domain B: domain located next to the domain A; Domain C: epoxy network unaffected by the presence of the MSiCWs. (b) Interphase volume fraction (domains A and B) for composite with perfectly oriented in epoxy matrix MSiCWs as a function of the interphase thickness. Concentration of the whiskers: 3 vol%.



0.0

Heat Flow (W/g)

(a)

Neat Epoxy Nonoriented MSICWs Oriented MSiCWs

-0.1 -0.2 -0.3 -0.4 -0.5 -0.6 30

40

50

60

70

80

90

o Temperature ( C) 0.0

(b)

Heat Flow (W/g)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 42

Neat Epoxy Nonoriented MSiCWs Oriented MSiCWs

-0.1 -0.2 -0.3 -0.4 -0.5 -0.6 30

40

50

60

70

80

o Temperature ( C)

90

Figure 10. DSC traces for neat epoxy and epoxy/MSiCW nanocomposites. (a) is the first run and (b) is the second run.


2500

(a)

2000

(1) Neat epoxy (2) Nonoriented MSiCWs (3) Oriented MSiCWs

3 2 1

1500

1000

500

0 -40 -20

0

20

40

60

80

100 120 140 160

o

Temperature ( C)

3.0 Neat epoxy Nonoriented MSiCWs Oriented MSiCWs

(b) 2.5

Tan d (E"/E')

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Storage modulus (MPa)

Page 41 of 42

2.0 1.5 1.0 0.5 0.0 0

20

40

60

80

100

120

140

160

Temperature (oC)

Figure 11. DMA results: (a) the storage modulus, E/ and (b) tan δ data for neat epoxy and epoxy/MSiCW composites.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 42

For Table of Contents Only:

Epoxy Magnetic field


Enhancing Mechanical and Thermal Properties of Epoxy

Recommend Documents