High-Temperature Mechanics of Boron Nitride Nanotube âBuckypaper

Article Cite This: ACS Appl. Nano Mater. 2019, 2, 4402−4416

www.acsanm.org

High-Temperature Mechanics of Boron Nitride Nanotube “Buckypaper” for Engineering Advanced Structural Materials Pranjal Nautiyal, Cheng Zhang, Archana Loganathan, Benjamin Boesl, and Arvind Agarwal* Nanomechanics and Nanotribology Laboratory & Plasma Forming Laboratory, Department of Mechanical and Materials Engineering, Florida International University, Miami, Florida 33174, United States

Downloaded via BUFFALO STATE on July 30, 2019 at 07:21:33 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Boron nitride nanotube (BNNT) is an attractive load-bearing nanomaterial with excellent mechanical properties and high-temperature stability. In this study, the mechanics of BNNT buckypaper assembly is examined at elevated temperatures (up to 750 °C). In situ mechanical investigations are performed inside the electron microscope for real-time visualization of deformation. The deformation characteristics are examined at multiple hierarchical levels to understand the role of defects in a single nanotube, stresstransfer between entangled nanotubes, and interactions between multiple nanotube layers of the buckypaper. The ultralightweight buckypaper is flexible and damage-tolerant and withstands repeated loading/unloading/reloading indentation cycles with an elastic modulus ∼0.8−1.2 GPa. Digital image correlation analysis of the real-time videos indicates excellent strain re-distribution in the buckypaper, which prevents localized stress-concentration. In situ high-speed camera imaging during tensile deformation reveals crack-deflection and crack-bridging due to nanotube entanglements, providing failure-resistance. The buckypaper has energy-dissipation ability, with loss tangent (tan δ) at room temperature as high as 0.5. This study attests to the ability of BNNT macroassembly to bear mechanical stresses up to 750 °C. The application of this macroassembly for developing a polymer-based nanocomposite with superior stiffness (1170% improvement) is also demonstrated. The findings in this work can be applied for engineering BNNT-based advanced structural materials. KEYWORDS: boron nitride nanotube, buckypaper, mechanical properties, in situ mechanics, high-temperature deformation

1. INTRODUCTION Boron nitride nanotube (BNNT), a structural analogue of carbon nanotube (CNT), has emerged as a remarkably promising candidate for developing the next generation of advanced, multifunctional materials.1−5 BNNT exhibits brilliant mechanical properties, with reported elastic modulus and tensile strength as high as 1 TPa and 60 GPa,6−8 respectively, making the nanotubes attractive for engineering strong structural materials. These nanotubes are characterized by high thermal conductivity, ranging from 300 to 3000 W m−1 K−1,9−11 which is important for thermal management applications. It has been demonstrated that BNNT is piezoelectric in nature12−14 and has neutron radiation shielding capability,14,15 which has implications for self-powered electronics and space vehicle bodies, respectively. In contrast to CNT, BNNTs display superior thermal stability as they can withstand oxidative conditions up to temperatures as high as 700−1000 °C without structural degradation16,17 (as opposed to CNTs which start degrading at 400 °C18). This makes BNNT a suitable reinforcement candidate for lightweight metal matrix composites (such as Al, Mg, and Ti) since the metalworking operations typically involve severe thermal and chemically reactive conditions.19−22 Several studies have shown the remarkable potential of BNNT for © 2019 American Chemical Society

augmenting the mechanical strength and stiffness of lightweight metals,23−29 which is promising for aerospace and automotive applications. Structural stability at elevated temperatures also opens up applications such as oxidation-resistant heat shields, thermal management of hybrid electric propulsion systems, and load bearing under high in-service temperature conditions. Intrigued by the promise of BNNT for advanced applications, development of the macroscopic nanotube assemblies has gained attention.30−33 Some of the examples of macroscale nanotube assemblies are yarn, thin film, mat, rope, and aerogel.16,33−36 Solution processing technique is employed to develop 3D structures with controlled structural features and dimensions.33,34 Kim and co-workers fabricated flexible and foldable BNNT sheets by direct deposition synthesis and vacuum infiltration methods.34 This BNNT buckypaper was characterized by very low mass density < 0.5 mg cm−3, Young’s modulus of 0.37 GPa, and an ultimate tensile strength of 2 MPa. Construction of cellular foams of BNNT has been demonstrated by chemical vapor deposition technique.32 The foam displayed Received: April 30, 2019 Accepted: June 25, 2019 Published: June 25, 2019 4402

DOI: 10.1021/acsanm.9b00817 ACS Appl. Nano Mater. 2019, 2, 4402−4416

Article

ACS Applied Nano Materials

Figure 1. (a) Illustration of buckypaper structure and composition at multiple length scales: macroscopic view of freestanding foldable sheet, microscopic view of nanotube network, and individual nanotube with alternating B and N atoms in a honeycomb-like structure. SEM micrographs showing: (b) high-magnification micrograph of a network of nodes/intertube junctions and nanotube struts and (c) cross-section of the buckypaper comprised of multiple layers of BNNT meshes with interlayer nanotube linkages providing structural support.

superelasticity as the authors reported shape recovery even after cyclic compressions with 90% deformation. More recently, Adnan and co-workers demonstrated a solution-extraction route using chlorosulfonic acid (CSA) to develop high-purity thin films, mats, and aerogels of BNNT.33 Solubility of BNNT in CSA was exploited to remove byproducts such as h-BN generated during nanotube synthesis. Furthermore, suspension of BNNT in CSA provides a viable route for molding aerogels in desired 3D shapes and dimensions. The interconnected network of nanotubes provides these 3D macrostructures with the ability to bear loads without fracturing, which makes them easy to handle. These “freestanding” assemblies can be easily integrated in metal, ceramic, or polymer matrices to develop high-

performance nanocomposites. Use of macroassemblies eliminates the requirement to disperse the nanotubes in the matrix, which is a key challenge during composite processing. It also facilitates engineering of composites with higher volume fraction of nanotubes, with uniform distribution. For instance, BNNT buckypaper was used to develop epoxy- and polyurethane-based composites, with BNNT loading as high as 32 and 93 wt %, respectively.34 Despite the fanfare around the excellent thermomechanical properties of BNNT, the current understanding of the hightemperature deformation characteristics of the nanotubes is very limited. Chen and co-workers examined the structural and mechanical properties of annealed nanotube fibrils after 4403


Article

ACS Applied Nano Materials exposing them to temperatures up to 900 °C.37 In situ Raman investigation revealed a shift of the G band peak from 1369 cm−1 at room temperature to 1339 cm−1 at 900 °C, indicating softening of B−N bonds at elevated temperatures. Atomic force microscopy compression of the annealed BNNT (after the nanotubes were cooled to room temperature) demonstrated that the radial modulus does not deteriorate after annealing up to 850 °C. This is ascribed to the retention of structural integrity of BNNT up to 900 °C. However, these investigations were performed postcooling (at room temperature) and do not provide insight into the real-time deformation of nanotubes at elevated temperatures. In addition to the elastic properties, examination of high-temperature deformation in the plastic and failure regime is very important to engineer high-temperature composites based on BNNT. Although freestanding BNNT assemblies are promising candidates for advanced high-temperature composites, there are several fundamental questions that need to be answered for better understanding of the mechanics of BNNT assemblies:

2. RESULTS AND DISCUSSION 2.1. Buckypaper Microstructure. The BNNT buckypaper studied in this work is shown in Figure 1a. It is comprised of a randomly oriented network of high-aspect nanotubes. As illustrated in the figure, the individual nanotubes are comprised of alternating “B” and “N” atoms arranged in a honeycomb-like structure. The as-received samples (Tekna, Canada) were synthesized by solvent dispersion and vacuum infiltration processes.34 In addition to fibrils, the nontubular particles seen in the SEM micrograph are B and BN impurities.30 The buckypaper has a very high degree of porosity of ∼80% and, consequently, a very low mass density of ∼0.4 g/cm3.34 The microstructure is comprised of a large number of nodes, where multiple nanotubes intersect and make contact with each other (Figure 1b). These nodal points are interconnected by nanotube struts, creating a mesh-like framework. The majority of the nodes are 4° junctions, that is, four nanotube struts emanate from the node (encircled in yellow). However, some of these junctions are characterized by five or more nanotube struts/ bundles emerging from the nodal junctions (encircled in red), suggesting a highly interconnected microstructure. The crosssection of the buckypaper reveals a multilayer architecture, with multiple BNNT meshes stacked over each other (Figure 1c). The multilayer architecture is formed during a vacuum filtration process, during which the BNNT dispersion is filtered through a polycarbonate membrane.34 One-dimensional (1D) tubes tend to stack one over the other during filtration, resulting in multiple layers in the buckypaper. This interconnected nanotube network provides structural stability to the buckypaper, as it can be folded and handled without breaking (shown in Figure 1a). 2.2. In Situ Investigation of Indentation-Induced Local Compression at Elevated Temperatures. The buckypaper was subjected to compression by a flat-ended diamond probe of 10 μm diameter. These tests were performed inside the scanning electron microscope (SEM) for real-time visualization of load transfer in the nanotube assembly (Figure 2a). The indentation tests were performed at room temperature and 250 and 500 °C. Load−displacement curves reveal an elastoplastic deformation behavior. An increase in deformation is noticed as the temperature increases (Figure 2b). The difference between deformation at room temperature and 250 °C is very marginal (around 8 μm), but there is visible enhancement of the total penetration depth at 500 °C (∼12 μm). However, the final residual plastic deformation is not too high even at 500 °C (∼6.2 μm), as compared to room temperature indentation (∼5.7 μm). This suggests activation of additional elastic mechanisms with increasing temperature. Supporting Information Videos S1, S2, and S3 show the real-time SEM videos of the indentation for the tests performed at room temperature and 250 and 500 °C, respectively. The buckypaper exhibits a trampoline-like behavior. The compression loading visibly stretches the buckypaper in the vicinity of the indenter probe up to radial distances > 10 μm (Figure 2a), indicating effective load transfer/ dispersion due to the interconnected nanotube network. During compression, the stress−strain relationship is expressed as a volumetric change:38

• What are the stress-transfer mechanisms in a BNNT assembly, and how do multiple nanotubes interact with each other during mechanical deformation? • How does temperature influence the mechanical properties of BNNT assembly? • How are thermal transport and mechanical properties/ deformation mechanisms of BNNT interrelated? • How does a freestanding BNNT structure respond to cyclic loading cycles, and are there any associated viscoelastic deformation mechanisms? This study presents a comprehensive assessment of hightemperature mechanics of a multilayer assembly of BNNT, henceforth referred to as BNNT-buckypaper, at multiple microstructural length scales. In situ nanoindentation inside an electron microscope is performed to probe the local deformation due to compressive action. Mechanical deformation of the buckypaper is probed for multiple indentation loading−unloading cycles, to evaluate the localized fatigueresistance of BNNT assembly at elevated temperatures. Digital image correlation analysis of real-time videos is performed to visualize the microstructure strain distribution mechanisms. The bulk mechanical properties are investigated by macrotensile tests, coupled with in situ high-speed camera imaging to observe failure mechanisms. The interplay between mechanical stresses and thermal transport in the buckypaper is analyzed by simultaneous thermal imaging during the tensile test. Dynamic mechanical investigations are performed to examine the viscoelasticity of BNNT buckypaper. These mechanical investigations are performed up to 750 °C, which is the temperature around which the major structural transformations/oxidative degradation of BNNT initiate. The application of BNNT buckypaper in a polymer nanocomposite is demonstrated, and the mechanical characteristics are examined. The real-time observation of material deformation at multiple length scales is a powerful approach to probe the stress-transfer characteristics of macroscale assemblies. The findings of this study provide critical high-temperature deformation information about BNNTs. These mechanistic insights are vital for engineering and predicting the mechanical properties of hightemperature materials and architectures based on BNNT.

ΔV = −C bΔP V0

(1)

where V0 is the initial volume, ΔV is the volume change due to compression, ΔP is the pressure change or the stress applied during indentation, and Cb is the compressibility of the 4404


Article

ACS Applied Nano Materials Δt = −C bΔP t0

(2)

where t0 is the initial thickness of the buckypaper, which is 80 μm. Change in thickness (Δt) is equal to the indenter displacement (d), and change in pressure (ΔP) is the indentation stress, which can be computed as F/Ap, where F is indentation force and Ap = πr2, is the area of indenter probe compressing the material. The peak displacements at 5 mN applied force (at room temperature and 250 and 500 °C) are plotted in Figure 2b. Substituting these values into eq 2, the compressibility of BNNT buckypaper can be determined as a function of temperature. The localized compressibility of the buckypaper increases from 1.7 GPa−1 at room temperature to ∼2.4 GPa−1 at 500 °C (Figure 2c). The elastic modulus of the buckypaper as a function of temperature is computed from the compressibility values, based on the following relation: E=

3(1 − 2ν) Cb

(3)

where ν is the Poisson’s ratio, which can be assumed to be ∼0.15 for the buckypaper.39 On the basis of eq 3, the elastic modulus of BNNT buckypaper is computed to be ∼1.2 GPa at room temperature, but drops to 0.8 GPa at 500 °C (Figure 2c). Upon compression, the nanotube pillars that interconnect the multiple layers of buckypaper will undergo buckling. BNNT is highly flexible, and elastic buckling of nanotube happens by kink formation, schematically shown in Figure 2d.40 As the applied load increases, each nanotube can undergo multiple kinking phenomena, shown in a previous report.40 During kink formation, the lattice structure of BNNT is distorted, creating dislocation-like defects and missing atomic planes. Golberg and co-workers reported a spring-like recovery of kinks upon removal of mechanical load.40 Because of their reversible character, they are also referred to as “momentary kinks”. However, for bending angles greater than 115°, the nanotubes tend to retain a residual buckle, responsible for permanent deformation.40 This explains the elastoplastic load−displacement behavior observed in Figure 2b. It is known that BNNTs are often characterized by the presence of vacancies, particularly if the growth process is far from thermodynamic equilibrium.41,42 Thermal vacancy migration is one of the possible mechanisms for reorganization and agglomeration of the vacancy defects to form dislocation lines. The activation barrier for the migration event is reported to be 3−6 eV,41 making the formation of the dislocation line at room temperature thermodynamically infeasible. However, vacancy migration would be favored at elevated temperatures, giving rise to the formation of extended dislocation defects in BNNT. The presence of dislocations is likely to enhance the deformation experienced by the nanotubes due to dislocation motion. This was evident from relatively higher penetration depths noticed during high-temperature nanoindentation in comparison to the room-temperature indentation depths (Figure 2b). In addition to the intrinsic vacancy effects, compression of the buckypaper structure will also induce intertube sliding. Energy is consumed to overcome van der Waals attraction between the tubes for sliding to happen. The intertube mechanisms are likely to play a more important role during bulk-scale and high-load testing and are discussed later in section 2.4 on macroscale tensile testing. It is noteworthy that the buckypaper is highly porous (∼80% porosity) and characterized by very low density of ∼0.4 g/cm3. The loss in elastic modulus at temperature as high as 500 °C is

Figure 2. (a) In situ nanoindentation of BNNT buckypaper inside SEM chamber. Indentation compression leads to elastic stretching of buckypaper due to highly interconnected network. (b) Nanoindentation load−displacement response of buckypaper at room temperature and 250 and 500 °C temperatures. (c) Local compressibility and elastic modulus of buckypaper determined during nanoindentation as a function of temperature. (d) Schematic representation of BNNT buckling/kinking mechanism active during indentation compression. Real-time SEM videos are provided as Supporting Information Videos S1, S2, and S3.

buckypaper. Since the localized indentation deformation is largely one-dimensional in nature, the volume change in eq 1 can be expressed as a change in thickness (Δt) during compression:38 4405


Article


Figure 3. (a) Nanoindentation load−displacement response for cyclic loading−unloading (50 cycles) with hysteresis loops at room temperature and 250 and 500 °C. (b) Microstructure strain contours developed in the buckypaper during indentation loading for the first and 50th cycles are shown, attesting to long distance stress-transfer and accumulation of residual strain over the 50 deformation cycles (scale bar, 20 μm), (c) Schematic representation of the detachment−attachment mechanism at nanotube junctions leading to topological transformation due to stretching of the buckypaper (during cyclic indentation). The real-time SEM videos are provided as Supporting Information Videos S4, S5, and S6.

very minimal. All of these factors make it a very promising reinforcement candidate for lightweight metal, ceramic, and

polymer composites which are employed in high-temperature applications. 4406


Article

ACS Applied Nano Materials 2.3. Localized in Situ Cyclic Deformation at Elevated Temperatures. While the freestanding buckypaper displays stable elastic properties at elevated temperatures, the ability to retain structural integrity and mechanical properties over repeated deformation cycles is vital for real-world applications. The buckypaper was subjected to cyclic loading−unloading− reloading quasi-static compression by in situ nanoindentation, to examine the deformation characteristics after prolonged local loading. The load−displacement curves demonstrate viscous effects, evident from the formation of hysteresis loops during cyclic loading (Figure 3a). The cyclic indentation was performed for 50 cycles. Supporting Information Videos S4, S5, and S6 show real-time deformation inside the SEM, for cyclic tests performed at room temperature and 250 and 500 °C, respectively. The videos reveal compression−recovery sequences as the indenter is loaded and unloaded. As noted previously, the buckling of nanotube pillars interconnecting the buckypaper layers is reversible in nature.40 The formation and annihilation of kinks provide this high degree of reversible cyclic deformation, making the freestanding BNNT structure damage-tolerant. Nevertheless, some residual plastic strain is also noticed, due to a certain degree of permanent buckle (when the nanotubes deform excessively, typically >115° angle40). As a result, some of the energy imparted by the indenter is dissipated, resulting in the hysteresis loop seen in Figure 3a. It is noteworthy that van der Waals forces of attraction between the nanotubes hold together the BNNT assembly. There are van der Waals interactions between individual nanotubes, as well as the multiple layers constituting the buckypaper (as shown in Figure 1c). The van der Waals forces acting between multiple layers can be approximated by the equation for interplane interactions:43

FVDW =

A 6πh3

It is seen from Supporting Information Videos S4, S5, and S6 that the indentation loading/unloading not only compresses the buckypaper along the z-axis but also causes elastic stretching and recovery around the indenter. The interconnected and entangled nanotubes in the buckypaper act as a network of springs, with the tendency to revert to the original configuration upon unloading. Digital image correlation (DIC) analysis of the real-time snapshots was performed to obtain microstructure strain contours. In DIC analysis, the real-time snapshots of the microstructure (during deformation) are compared/analyzed with respect to the starting snapshot, to capture the changes in specimen surface incrementally as the force is applied.45 The software tracks and correlates the pixels in the deformed snapshots with respect to the starting image. This relative comparison aids in determining the full field strain map, showing spatial distribution of strains in the microstructure as a function of mechanical loading.46,47 Figure 3b shows the strain maps for room-temperature indentation, clearly showing the long-distance stress-transfer, evident from the development of strains even ∼50 μm away from the local indentation. The ability to disperse the force provides the buckypaper assembly a high degree of trampoline-like elasticity (Videos S1−S6) and prevents local failure/rupture of the BNNT assembly even for local indentation-induced stresses as high as ∼60 MPa. The DIC maps with colored contours indicate a variable stress-state, which is highly dependent on the local disentanglement and reentanglement phenomena. The map clearly shows that the microstructure strains are pronounced at peak loading state as the local depression caused by the indenter probe is transmitted as elastic stretching in the vicinity. During the first loading cycle, the local strains are obtained to be as high as ∼3% when 5 mN load is applied. It is also noted that even after complete unloading, there are residual strains in the buckypaper (|ε| < 1%), which signifies some degree of “unrecovered” permanent deformation. This residual strain progressively builds up after each cycle. The final residual strain at the end of the 50th cycle was observed to be higher than 1% (Figure 3b). The permanent deformation of the buckypaper can be related to topological transformation of the nanotube network during indentationinduced stretching. As observed in Figure 1b, the buckypaper microstructure is comprised of nodes where the nanotubes intersect. During the stretching and release of the buckypaper (due to cyclic loading−unloading), the nanotubes are likely to undergo detachment−attachment phenomena at these nodal junctions.48 Upon unloading, the nanotubes do not necessarily reattach with the same partner; many of the nanotubes tend to pair with new partners to form the nodal junctions. This leads to topological transformation, resulting in some degree of residual permanent deformation observed in Figure 3a,b. This detachment−attachment mechanism active during the stretching of buckypaper is illustrated in Figure 3c. It is noteworthy that the nanoindentation modulus values (0.8−1.2 GPa) are much higher than the “tensile” elastic modulus reported in literature for BNNT buckypaper (∼10−20 MPa),34 primarily because nanotube disentanglement is likely to be more prominent in tension resulting in compromised stiffness. This difference in tension−compression mechanics would be vital when engineering buckypaper-based composites. Hence, the next section explores the tensile properties of the buckypaper as a function of temperature. 2.4. High-Temperature Tensile Deformation at Macrostructural Scale. To examine the bulk structural scale mechanical properties, the buckypaper was subjected to tensile

(4)

where FVDW is the van der Waals force between the plane sheets, A is the Hamaker’s coefficient, and h is the interplanar separation. Since BNNT layers are porous, this equation can be modified to account for the porosity: FVDW =

(1 − p)A 6πh3

(5)

where p is the porosity, which is 0.8 for the buckypaper investigated in this study.34 The energy change during cyclic compression can be modeled as E=

dFVDW A = dh 30πh4

(6)

Based on eq 6, the energy changes inversely to the fourth power of interlayer separation, h. As noted before, the compressibility of the buckypaper assembly increases with the increase in test temperature (Figure 2c). As a result, the separation “h” between the layers will decrease, resulting in enhanced viscous effects/energy dissipation (based on eq 6). This is evident from the nature of F−h hysteresis loops shown in Figure 3a, with more prominent hysteresis loops at 500 °C. In addition to interlayer van der Waals interactions, intertube sliding will be active during mechanical loading of the buckypaper. The interfacial sliding may involve stick−slip events, which can lead to energy dissipation.44 The ability to dissipate energy is promising for superior impact tolerance. 4407


Article


snapshots show that the crack initiates at the location of stress-concentration, which is a precrack tip in this case (snapshots no. 1 and 2). The crack propagation is characterized by extensive deflections (snapshot no. 3). The crack advances in the buckypaper via disentanglement and detachment of BNNTs held together by van der Waals forces. Since the nanotubes in buckypaper microstructure are clustered, the disentanglement takes place locally in the microstructure, at the sites of least resistance to detachment or sites with poor intertube entanglement. The 3D network of nanotubes delays crack propagation due to local disruptions in the crack pathway. This was evident from the zigzag crack pathway observed in Figure 5. The high-speed-camera video (Supporting Information Video S7) shows the crack propagation takes place in multiple steps, with intervening pauses or halts. This is attributed to extensive crack-bridging by BNNTs. The pulled-out nanotubes acting as crack-bridges are encircled in snapshot no. 4. These bridges were found to be as long as 90−100 μm and enhance the resistance to failure. For crack to propagate, the applied mechanical energy is absorbed for breaking the nanotube bridges. This results in delayed crack advancement and failure, as observed in Supporting Information Video S7, with multiple halts during crack propagation. The final snapshot (snapshot no. 6) after failure shows a zigzag tearing of the buckypaper. The deformation mechanisms at different temperatures were examined by postfailure electron microscopy of buckypaper specimens after the samples were cooled to room temperature (Figure 6). The failed cross-section reveals delamination/ separation of layers constituting the buckypaper (Figure 6a). The extent of detachment is more prominent at elevated temperatures. For tensile failure at 750 °C, it can be seen that the nanotube pillars are pulled out. Thermal transformation of BNNT at elevated temperatures16 is responsible for poor interlayer cohesion. SEM micrographs in Figure 6b reveal transformation and coalescence of BNNTs to form platelet-like structures at 500 and 750 °C. It is reported in our previous work that BNNT undergoes “selective”, limited unzipping at elevated temperatures under oxidative conditions.17 High-temperature exposure results in diffusive bonding of the unzipped nanoribbons. This will result in poor interconnectivity of the nanotube network, leading to deteriorated failure-resistance. It is hypothesized that defect sites undergo oxidative transformations due to lower activation barrier for oxidation. This results in certain regions in the microstructure having a discontinuous/ disrupted nanotube network, responsible for poor load-bearing ability at 750 °C (Figure 4). During tensile deformation, the buckypaper microstructure undergoes reorganization, such that the tubes are aligned along the loading direction. This aids in enhanced load-bearing ability as nanotubes display superior mechanical properties along their axial length. The realignment phenomenon is shown in the SEM micrographs in Figure 6c. The images show nanotube bundles pulled out due to tension. The realignment of nanotubes leads to bridge formation, observed in real-time high-speed-camera video (Supporting Information Video S7). The realignment behavior was noticed for each individual layer of the buckypaper assembly. However, the pull-out is not as prominent at 750 °C. The thermal transformation results in deterioration of the nanotube network. This was observed upon close examination of the failed crosssection (Figure 6d). Coalescence of BNNTs at high temperatures results in arrested ability for nanotube reorientation. During tensile loading of BNNT, Stone−Wales (SW) defects are created after a critical mechanical strain.50 SW defects sites

loading and the stress−strain characteristics were obtained over a range of temperature, i.e., at room temperature and 250, 500, and 750 °C (Figure 4). The buckypaper was characterized by an

Figure 4. Tensile stress−strain curves for BNNT buckypaper deformed at room temperature and 250, 500, and 750 °C.

ultimate tensile strength (UTS) of ∼2.8 MPa at room temperature. However, it should be noted that, with 80% porosity, the “true” cross-section area that bears tensile load is much lower: Atrue ∼ (1 − p)Aapparent, where p is porosity and Aappparent is the apparent cross-section area determined by physically measuring the width and thickness of the tensile specimens. When normalized by taking porosity into consideration, the “true” tensile strength at room temperature is computed to be ∼13.8 MPa. A progressive drop in strength was observed with an increasing temperature. The true strength was found to be ∼9.8, 7.9, and 3.8 MPa at 250, 500, and 750 °C, respectively. The strain at failure is ∼8.4% at room temperature, but it drops to ∼2.4% at 750 °C. Despite this drop, the mechanical strength of buckypaper still surpasses/rivals the reported strengths for most polymers and elastomers, which are rarely stable above 300 °C.49 Therefore, BNNT buckypaper is a promising low-density material candidate for load-bearing structures at much higher service temperatures where other conventional low-density materials, such as polymer foams and elastomers do not even survive. To develop insight into failure mechanisms in real time, the tensile deformation of buckypaper at room temperature was captured by high-speed camera. The sequence of events during tensile loading is summarized in Figure 5. The real-time

Figure 5. Sequence of deformation events recorded by high-speed camera at 5,000 frames/s during tensile testing of BNNT buckypaper at room temperature. 4408


Article


defect goes down, and therefore, the critical strain for plastic yielding also goes down. This explains the observed drop in failure strain for BNNT buckypaper as the deformation temperature increases (Figure 4). Mechanically induced defects can be created at lower tensile strains, thereby accentuating the plastic deformation. At elevated temperatures, the buckypaper experiences complex thermomechanical conditions: tensile stresses induce realignment of the BNNT clusters, elevated temperatures create Stone−Wales defect sites in the nanotubes, and selective morphological transformations alter the microstructure. In order to examine the influence of high-temperature exposure in real time, the tensile buckypaper specimen was heated to 500 °C and the temperature gradient was probed by thermal imaging. In Figure 7a, the tensile setup with a contact heater is

Figure 6. Scanning electron micrographs, revealing tensile failure mechanisms at different temperatures: (a) delamination of layers in the assembly, (b) morphological transformations at elevated temperatures, (c) alignment of nanotube bundles due to tensile loading, and (d) comparison of interlayer load-bearing mechanism at room temperature and 750 °C.

Figure 7. (a) Thermomechanical deformation of BNNT buckypaper exposed to a contact heating element during high-temperature tensile testing. (b) Infrared thermal image/map of the buckypaper experiencing tensile loading. (c) Corresponding temperature profile along the buckypaper length.

are characterized by B−B and N−N homonuclear bonds.51 These defect sites are susceptible to structural transformations due to bond rotation, creating dislocations in the nanotube. Therefore, creation of defect during tensile deformation marks the onset of plastic yielding. The energy associated with SW defect (ESW) is related to tensile strain (ε):50 ESW = 5.6 − 10.8ε − 37.5ε sin(2χ + 37.3°)

shown. The buckypaper rests on the heater during the tensile test, such that the heater makes contact with the central region of the buckypaper strip. Simultaneous thermal imaging was performed during the tensile test at 500 °C. An instantaneous temperature map during the test is shown in Figure 7b. It is seen that the temperature of the specimen is high over the heating element (∼540 °C), but it drops near the grips or away from the heating element (points A and B marked in the image). The corresponding temperature profile is plotted in Figure 7c. It is noteworthy that the buckypaper is porous (very high porosity of 80%); therefore the actual area/volume that transfers the heat is limited. Additionally, heat loss/dissipation to the atmosphere is also responsible for the drop in the specimen temperature in the regions not in direct contact with the heater. As a result,

(7)

where χ is the chirality (χ ϵ [0°,180°]). On the basis of eq 7, the formation of SW defects becomes energetically favorable at higher strains. Due to the nature of nanotube assembly in buckypaper, intertube slipping is expected to be prominent during tensile loading and the strains experienced by individual nanotube might not be high enough for defect formation. However, at elevated temperatures, the activation barrier for SW 4409


Article


Figure 8. (a) Loss tangent as a function of dynamic loading frequency for BNNT buckypaper recorded at room temperature and 250, 500, and 750 °C. (b) Schematic illustration of the unzipping−zipping mechanism responsible for viscoelasticity in BNNT buckypaper assembly.

flash diffusivity measurement, and was found to be only 0.17 W/ (m·K). This is anticipated because BNNT primarily conducts heat along its axial direction. During tensile deformation, the nanotubes align in the plane of the buckypaper, along the loading axis. Very few nanotubes are oriented perpendicular to the tensile loading plane, resulting in 3 orders of magnitude lower conductivity. The difference in thermal transport in and out of the buckypaper plane has implications for the mechanical deformation behavior. As noted before, plastic yielding of buckypaper is induced by Stone−Wales defects. Superior transport along the buckypaper will result in accentuated thermal migration of defects to form dislocations, activating plastic deformation mechanisms. Contrary to this, limited outof-plane thermal transport results in arrested plasticity. This was evident from the SEM micrographs: Primarily nonplastic mechanisms such as delamination or detachment of nanotube pillars were prominent for out-of-plane deformation (Figure 6a), as opposed to plastic mechanisms such as nanotube reorientation, stretching, and pull-out observed in the buckypaper plane (Figure 6c). Therefore, there is an interplay of multiscale thermomechanical mechanisms during tensile deformation of BNNT buckypaper, which manifests as multistage yielding and pop-ins in the stress−strain curves (Figure 4). 2.5. Viscoelasticity at Elevated Temperatures. The unique microstructure of buckypaper, with a network of nanotube junctions and struts, induces time-dependent

simultaneous thermal transport will take place during the tensile test, with heat flow from the hot central region to the relatively low temperature outer regions of the buckypaper strip. The heat energy transfer is computed by Fourier’s law: q ̇ = −λA(ΔT /Δx)

(8)

where q̇ is the heat energy transfer per unit time, λ is the thermal conductivity, A is the cross-section area (5 mm × 80 μm), ΔT is the temperature difference across two points, and Δx is the distance between the two points. Considering the points A (near the grip) and P (over the heating element), the temperature gradient (ΔT) is ∼430 °C and the heat transfer distance (Δx) is 4 mm. The thermal conductivity of porous materials (λ) is determined by Loeb’s relation:52

λ = λs(1 − p)

(9)

where λs is the thermal conductivity of solid material and p is the porosity. The reported peak thermal conductivity of “aligned” BNNTs is 2400 W/(m·K).11 As observed in Figure 6c, the nanotubes tend to align along the tensile axis during the deformation; therefore, this value of conductivity for aligned BNNT can be substituted to eq 9 for λs. The in-plane conductivity of the porous buckypaper is then computed to be 480 W/(m·K) from eq 9. Therefore, the heat flow rate is obtained to be 20.4 W (based on eq 8). The out-of-plane thermal conductivity of the buckypaper was also determined by thermal 4410


Article

ACS Applied Nano Materials W = σcluster Δx Ar

mechanical characteristics. The buckypaper was subjected to dynamic tensile loading, with the mean axial displacement of 20 μm and a displacement amplitude of ±10 μm. The buckypaper was subjected to low deformations to avoid plastic yielding or failure of the assembly. Dynamic deformation results in a phase lag (δ) between the applied strain and the resultant stresses developed in the material.53 This phase lag results in a complex deformation state: a component of the applied energy is stored as potential energy (elastic component), whereas some of the energy is dissipated during the loading cycle (loss component).54 The materials with significant loss component are viscoelastic in nature; the storage (E′) and loss (E″) moduli of such materials are expressed in terms of the phase lag angle, δ:55 σ E′ = cos δ (10) ε E″ =

σ sin δ ε

This work of unzipping is computed to be ∼100 J per unit area, per loading cycle. This explains the remarkable energydissipation ability observed in Figure 8a. At the high frequency of dynamic mechanical tests, the loading−unloading becomes much faster. The increase in zipping−unzipping phenomena results in enhanced energy dissipation, which manifests as a generally rising trend for tan δ with increasing frequencies (Figure 8a). A marginal drop in the value of loss tangent for frequencies above 80 Hz suggests the simultaneous activation of other deformation mechanisms, such as reorientation/alignment along the loading direction, which start dominating over the viscous mechanism (zipping−unzipping). During dynamic mechanical loading, frictional effects (atomic scale stick−slip instabilities) between sliding nanotubes can also play a role in energy dissipation. As the temperature increases, thermal excitations reduce stick−slip jumps.44 Additionally, the meshlike nanotube network is altered at elevated temperatures due to coalescence of nanotubes shown in Figure 6b,d. As a result, there are fewer intertube junctions where zipping−unzipping phenomena take place during dynamic loading. This leads to suppressed loss tangent at temperatures exceeding 500 °C. Nevertheless, the retention of viscoelasticity even up to 750 °C opens up windows for multifarious applications. 2.6. Application of BNNT Buckypaper in Composites. Intrigued by the excellent load-bearing capability of BNNTs, we prepared polymer composites by introducing the buckypaper in PDMS resin. The composites were fabricated by casting technique, such that the buckypaper was sandwiched between the polymer (Figure 9a). Tensile behavior of these specimens was measured and compared with pure PDMS. To assess the effect of introducing varying weight fractions of buckypaper in the resin, two compositions were examined: 0.1 and 0.4 wt % BNNT in PDMS. The stress−strain plots show remarkable enhancement in the load-bearing capacity, the stresses are higher for the composite specimens as compared to pure PDMS (Figure 9b). PDMS is a highly stretchable elastomer, with remarkably high failure strains. Addition of BNNT leads to relatively lower failure strains compared to pure PDMS; nevertheless, the strains are reasonably enough (in the range of ∼30−60%) for load-bearing applications. It is seen that 0.1 wt % BNNT-reinforced composite exhibits a tensile strength of ∼49 kPa, and strength is ∼105 kPa for 0.4 wt % BNNT addition. At such low concentration of 0.4 wt %, the stresses are ∼3 times higher than that of neat PDMS. The elastic modulus computed from the slope of the curve (initial linear region) goes up from ∼110 kPa for pure PDMS to ∼1.4 MPa on adding 0.4 wt % BNNT, which is an ∼1170% enhancement. This is clear evidence for the potential of buckypaper as filler for lightweight polymers that are typically characterized by lower stiffness, limiting their application in load-bearing structures. Moreover, buckypaper allows for developing classical laminated, sandwichstructured composites, overcoming the time, resources, and challenges often associated with dispersing the nanofillers in the matrix.60−62 It is noteworthy that for applications requiring stretchability and superior failure strains (than the values demonstrated here), the weight fraction of buckypaper should be further reduced and the microstructure needs to be engineered so as to produce an intimate polymer/nanotube interface.

(11)

The ratio of loss modulus and storage modulus (E″/E′), tan δ, is called the loss tangent and is a measure of the energydissipation ability.56 The dynamic mechanical analysis (DMA) of the buckypaper was performed from room temperature up to 750 °C. The loading/unloading frequency-dependent loss behavior of the buckypaper is shown in Figure 8a for different temperatures. The buckypaper is characterized by excellent energy-dissipation ability, with tan δ value as high as ∼0.5 at room temperature. The loss tangent drops as the temperature increases. Nevertheless, the material retains the viscoelastic nature even up to 750 °C (with tan δ recorded in the range of ∼0.05−0.1). During dynamic loading−unloading cycle, the nanotube network experiences microstructure transformation by unzipping/zipping of intertube junctions or nodes.57 During the loading cycle, the nanotubes at the node separate (unzipping), resulting in a reduction in nanotube entanglements.48 The unzipping process requires energy to overcome van der Waals attraction forces between BNNTs, resulting in energy dissipation. During the unloading cycle, the separated nanotubes come together at the nodal points (zipping of nodes). The zipping event is associated with re-formation of intertube entanglements. This dynamic loss mechanism is elucidated in Figure 8b. The buckypaper can be considered to be a dense cluster of nanotubes. The strength of a nanoparticle cluster (with spherical nanoparticles) is modeled as58 σcluster =

1 − pF pd 2

(13)

(12)

where p is the porosity, F is the interparticle binding force, and d is the particle size (diameter in the case of spherical particles). Equation 12 can be modified for 1D nanoparticle clusters by replacing the d2 term in the denominator with the product of tube diameter and length, d × l. Tang and co-workers determined the van der Waals attraction/binding force (F) between two overlapping BNNTs to be ∼96 nN by in situ TEM investigation.59 This binding force value corresponds to the nanotube diameter (d) ∼ 27.5 nm and overlapping length (l) ∼ 87 nm.59 Substituting these values to the equation for σcluster, the strength of BNNT clusters is found to be ∼10 MPa, which must be overcome for unzipping of nanotube junctions. During dynamic tensile loading, the buckypaper is subjected to a cyclic displacement (Δx) of 10 μm. Therefore, the work done per unit area in a single loading cycle is computed by 4411


Article


The microstructure of buckypaper-based composites is examined by the SEM analysis of the fractured surface. The incorporated buckypaper within the matrix is seen in Figure 9c. The fractured surface shows the multiple layers of the buckypaper. The signs of pull-out are observed, indicating stress-transfer occurred from the polymer matrix to the BNNT sheets. Since buckypaper is porous, there is some degree of resin infiltration as seen in Figure 9d. During tensile loading, the polymer coated nanotubes are pulled and they resist fracture. The signature of extensive crack-bridging can be noticed in the SEM micrograph (Figure 9d). These load-bearing mechanisms, such as nanotube pull-out, delamination of buckypaper layers, and crack-bridging manifest as jumps/drops in the tensile stress−strain curves for the composites (Figure 9b). Good interface wetting between the polymer and the filler material is vital for improving the mechanical performance of the composite. The SEM micrograph in Figure 9e shows that well-coated nanotubes aid in bearing the loads. However, it should be noted that the microstructure also consists of regions with relatively inferior wetting and infiltration of the polymer within the buckypaper. Large, unfilled interstices are observed in Figure 9e. Improving the polymer infiltration within the buckypaper can significantly enhance the failure strains observed for the composites. Tailoring the viscosity of the resin, processing temperature/pressure, and buckypaper porosity are some of the strategies that can be used to design nanocomposite microstructures with superior wetting and infiltration. These findings clearly demonstrate that the freestanding BNNT buckypaper architecture has practical application for developing advanced composite materials. Although this study is demonstrating the application in polymers, this filler can very well be employed for metal and ceramic matrix composites because of the excellent high-temperature stability of BNNT.

3. CONCLUSIONS This work seeks to unravel the mechanics of stress-transfer in a BNNT macroassembly at multiple length scales, viz., the role of defects in individual nanotubes, stress-transfer between multiple nanotubes, and mechanical interactions between different layers of buckypaper (summarized in Figure 10). The high-temperature mechanical properties of a freestanding BNNT buckypaper are probed for the first time. In situ characterization techniques are adopted for real-time examination of deformation mechanisms. Digital image correlation analysis of the realtime snapshots is performed for visualizing microstructure deformation. (1) Localized compression during nanoindentation revealed a trampoline-like effect, with excellent load transfer through the nanotube network. The buckypaper displays excellent flexibility and resistance to failure for local compressive stresses as high as 60 MPa even at 500 °C. (2) The cyclic loading−unloading−reloading indentation of the buckypaper demonstrated excellent damage-tolerance over a long period of time as well as at elevated temperatures (250 and 500 °C). Deformation for 50 cycles and local stresses up to 60 MPa resulted in a minimal residual strain ∼ 1%. (3) The freestanding buckypaper is capable of bearing tensile stresses up to 750 °C, although the failure strength is found to drop from ∼14 MPa at room temperature to ∼4 MPa at 750 °C. Nevertheless, a wide range of service temperatures in which BNNT retains its structural integrity and load-bearing ability is exciting for high-temperature applications.

Figure 9. (a) Cast PDMS specimens with BNNT buckypaper embedded inside to make sandwich-structured composites. (b) Tensile stress−strain plot for PDMS- and buckypaper-reinforced composites. SEM micrographs of fractured surface of the composite showing: (c) buckypaper pull-out/delamination (inset, low-magnification picture of the failed surface), (d) polymer infusion between buckypaper layers and crack-bridging by nanotubes, and (e) regions of excellent and inferior wetting/infiltration by polymer. 4412


Article


Figure 10. Schematic representation of mechanical interactions/stress-transfer phenomena in a freestanding macroscopic BNNT assembly at multiple hierarchical levels. temperature mechanical stage equipped with sample and probe heaters was used. The upper limit of the temperature range was chosen to be 500 °C, since the sample heaters used for in situ indentation testing were designed and capable for sample heating up to this temperature. The real-time high-resolution SEM videos corresponding to nanoindentation loading/unloading were also recorded. Bulk mechanical properties were examined by performing tensile tests of the buckypaper using a microtensile stage equipped with a 450 N capacity load cell (MTI Instruments, Albany, NY, USA). A IL5 Fastec high-speed camera (Fastec Imaging, San Diego, CA, USA) was used to record the crack initiation and propagation phenomena during tensile loading at 5,000 frames/s. For high-temperature tensile test (up to 750 °C), a heating element was fitted in the tensile stage that heats the specimen during the entire duration of the test by direct contact. The tensile tests were performed in air. Viscoelastic behavior was examined by tensile dynamic mechanical analysis (DMA) using ElectroForce 3100 (TA Instruments, New Castle, DE, USA) at room temperature and 250, 500, and 750 °C. The DMA studies were performed in displacement control mode, with programmed mean displacement and dynamic displacement amplitudes of 20 and 10 μm, respectively. A low-load cell with 250 g capacity was used. The dynamic response was obtained as loss tangent over a frequency range of 10−100 Hz. 4.3. Digital Image Correlation Analysis. DIC analysis of realtime in situ SEM videos was performed to compute microstructure strains as a function of applied indentation load. VIC-2D DIC software (Correlated Solutions, Irmo, SC, USA) was used to determine the strain maps. The software computes local strains in the microstructure by comparing real-time snapshots with respect to the starting image (when no load is applied). 4.4. Thermal Imaging. To ascertain the effect of thermal transport on high-temperature mechanical response, an infrared thermal imaging camera (FLIR T450SC, Wilsonville, OR, USA) was used to record the tensile deformation at 500 °C in real time. The recorded video was used to determine temperature gradient in the specimen. 4.5. Thermal Conductivity. A light flash apparatus (NETZSCH LFA 467 HT HyperFlash, Germany) was used to determine the thermal conductivity of the buckypaper along its thickness. The measurements

(4) BNNT buckypaper demonstrates viscoelastic response upon dynamic tensile loading. The loss tangent (tan δ) is recorded to be as high as 0.5, suggesting excellent energydissipation/impact-resistance ability. The viscoelastic characteristics are observed at elevated temperatures too (up to 750 °C). (5) The application of BNNT buckypaper in a polymer composite is demonstrated. The composite exhibits enhanced resistance to deformation and is characterized by remarkably higher stiffness (∼1170% improvement). The ease of introducing buckypaper into the matrix without employing complex processing techniques makes it a promising reinforcement candidate. Superior thermal stability and mechanical performance of BNNT provide a colossal opportunity to develop the next generation of advanced materials. The findings in this work provide critical insights for engineering advanced nanocomposites based on BNNT assemblies which have a wide range of applications in space exploration, heat shields, aircraft bodies, nuclear structures, sensor and actuator systems, and thermal management systems.

4. EXPERIMENTAL SECTION 4.1. Material and Microstructure. The freestanding BNNT buckypaper used in this study was obtained from Tekna (Quebec, Canada). The buckypaper sheet has a thickness of ∼80 μm, with individual nanotube diameters in the range of ∼5−10 nm. The buckypaper microstructure and failure mechanisms were examined using JEOL JSM-6330F field emission SEM (Tokyo, Japan). 4.2. Multiscale Mechanical Characterization. Nanoscale “localized” mechanical response of the buckypaper was captured by in situ nanoindentation (Picoindenter, Hysitron PI 87, Bruker, Billerica, MA, USA). The nanoindenter stage was installed inside a dual beam JEOL JIB-4500 focused ion beam/SEM system (Tokyo, Japan) for in situ imaging. A flat-ended diamond probe with 10 μm diameter was used. For high-temperature indentation testing (up to 500 °C), a high4413


Article

ACS Applied Nano Materials were performed by exposing one of the sample surfaces to short energy light pulse, which recorded the sample temperature at the other surface. Buckypaper was coated with graphite prior to the measurement, to avoid light transmission. 4.6. Composite Fabrication. A two component PDMS (SilGel 612, Wacker Chemie AG, Munich, Germany) was used. The buckypaper strips were introduced in the resin prior to curing, such that the strips are suspended between the polymer. The resin, with buckypaper suspended inside, was cured in a Petri dish by heating at 100 °C for 30 min. Dogbone shaped samples were then punched out of the cured film to perform tensile tests.

■

(3) Golberg, D.; Bando, Y.; Huang, Y.; Terao, T.; Mitome, M.; Tang, C.; Zhi, C. Boron Nitride Nanotubes and Nanosheets. ACS Nano 2010, 4, 2979−2993. (4) Song, W.-L.; Wang, P.; Cao, L.; Anderson, A.; Meziani, M. J.; Farr, A. J.; Sun, Y.-P. Polymer/Boron Nitride Nanocomposite Materials for Superior Thermal Transport Performance. Angew. Chem., Int. Ed. 2012, 51, 6498−6501. (5) Kim, K. S.; Couillard, M.; Shin, H.; Plunkett, M.; Ruth, D.; Kingston, C. T.; Simard, B. Role of Hydrogen in High-Yield Growth of Boron Nitride Nanotubes at Atmospheric Pressure by Induction Thermal Plasma. ACS Nano 2018, 12, 884−893. (6) Wei, X.; Wang, M.-S.; Bando, Y.; Golberg, D. Tensile Tests on Individual Multi-Walled Boron Nitride Nanotubes. Adv. Mater. 2010, 22, 4895−4899. (7) Arenal, R.; Wang, M.-S.; Xu, Z.; Loiseau, A.; Golberg, D. Young Modulus, Mechanical and Electrical Properties of Isolated Individual and Bundled Single-Walled Boron Nitride Nanotubes. Nanotechnology 2011, 22, 265704. (8) Zhao, Y.; Chen, X.; Park, C.; Fay, C. C.; Stupkiewicz, S.; Ke, C. Mechanical Deformations of Boron Nitride Nanotubes in Crossed Junctions. J. Appl. Phys. 2014, 115, 164305. (9) Chang, C. W.; Fennimore, A. M.; Afanasiev, A.; Okawa, D.; Ikuno, T.; Garcia, H.; Li, D.; Majumdar, A.; Zettl, A. Isotope Effect on the Thermal Conductivity of Boron Nitride Nanotubes. Phys. Rev. Lett. 2006, 97, No. 085901. (10) Chang, C. W.; Han, W.-Q.; Zettl, A. Thermal Conductivity of BC-N and BN Nanotubes. J. Vac. Sci. Technol., B: Microelectron. Process. Phenom. 2005, 23, 1883−1886. (11) Belkerk, E. B.; Achour, A.; Zhang, D.; Sahli, S.; Djouadi, M.-A.; Yap, Y. K. Thermal Conductivity of Vertically Aligned Boron Nitride Nanotubes. Appl. Phys. Express 2016, 9, No. 075002. (12) Bai, X.; Golberg, D.; Bando, Y.; Zhi, C.; Tang, C.; Mitome, M.; Kurashima, K. Deformation-driven electrical transport of individual boron nitride nanotubes. Nano Lett. 2007, 7, 632−637. (13) Ciofani, G.; Danti, S.; D’Alessandro, D.; Ricotti, L.; Moscato, S.; Bertoni, G.; Falqui, A.; Berrettini, S.; Petrini, M.; Mattoli, V.; Menciassi, A. Enhancement of Neurite Outgrowth in Neuronal-Like Cells following Boron Nitride Nanotube-Mediated Stimulation. ACS Nano 2010, 4, 6267−6277. (14) Kang, J. H.; Sauti, G.; Park, C.; Yamakov, V. I.; Wise, K. E.; Lowther, S. E.; Fay, C. C.; Thibeault, S. A.; Bryant, R. G. Multifunctional electroactive nanocomposites based on piezoelectric boron nitride nanotubes. ACS Nano 2015, 9, 11942−11950. (15) Thibeault, S. A.; Kang, J. H.; Sauti, G.; Park, C.; Fay, C. C.; King, G. C. Nanomaterials for Radiation Shielding. MRS Bull. 2015, 40, 836− 841. (16) Golberg, D.; Bando, Y.; Kurashima, K.; Sato, T. Synthesis and Characterization of Ropes made of BN Multiwalled Nanotubes. Scr. Mater. 2001, 44, 1561−1565. (17) Nautiyal, P.; Loganathan, A.; Agrawal, R.; Boesl, B.; Wang, C.; Agarwal, A. Oxidative Unzipping and Transformation of High Aspect Ratio Boron Nitrie Nanotubes into “White Graphene Oxide” Platelets. Sci. Rep. 2016, 6, 29498. (18) Chen, Y.; Zou, J.; Campbell, S. J.; Le Caer, G. Boron Nitride Nanotubes: Pronounced Resistance to Oxidation. Appl. Phys. Lett. 2004, 84, 2430−2432. (19) Nautiyal, P.; Zhang, C.; Boesl, B.; Agarwal, A. Non-Equilibrium Wetting and Capture of Boron Nitride Nanotubes in Molten Aluminum during Plasma Spray. Scr. Mater. 2018, 151, 71−75. (20) Nautiyal, P.; Gupta, A.; Seal, S.; Boesl, B.; Agarwal, A. Reactive Wetting and Filling of Boron Nitride Nanotubes by Molten Aluminum during Equilibrium Solidification. Acta Mater. 2017, 126, 124−131. (21) Lahiri, D.; Singh, V.; Li, L. H.; Xing, T.; Seal, S.; Chen, Y.; Agarwal, A. Insight into Reactions and Interface between Boron Nitride Nanotube and Aluminum. J. Mater. Res. 2012, 27, 2760−2770. (22) Agarwal, A.; Boesl, B.; Nautiyal, P.; Rudolf, C.; Zhang, C. Aluminum-Boron Nitride Nanotube Composites and Method for Making the Same. U.S. Patent App. 15/182,042, 2017.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsanm.9b00817. Video S1, in situ indentation of BNNT buckypaper at room temperature, inside SEM (MP4) Video S2, in situ indentation of BNNT buckypaper at 250 °C, inside SEM (MP4) Video S3, in situ indentation of BNNT buckypaper at 500 °C, inside SEM (MP4) Video S4, in situ cyclic indentation loading−unloading− reloading (50 cycles) of BNNT buckypaper at room temperature, inside SEM (MP4) Video S5, in situ cyclic indentation loading−unloading− reloading (50 cycles) of BNNT buckypaper at 250 °C, inside SEM (MP4) Video S6, in situ cyclic indentation loading−unloading− reloading (50 cycles) of BNNT buckypaper at 500 °C, inside SEM (MP4) Video S7, high-speed camera video of tensile failure of BNNT buckypaper at room temperature (captured at 5000 frames per second) (MP4)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: agarwala@fiu.edu. ORCID

Arvind Agarwal: 0000-0002-7052-653X Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the Department of Navy Grant N00014-17-12563 and program officer Mr. Anthony C. Smith, Sr. P.N. thanks Florida International University (FIU) Graduate School for the Presidential Fellowship award. B.B. acknowledges the Office of Naval Research DURIP Grant (N00014-16-1-2604) for establishing the in situ nanoindenter facility at FIU. We also recognize the Advanced Materials Engineering Research Institute (AMERI) at FIU for the research facilities used in this study. The authors thank Ms. Noemie Denis (FIU) for providing assistance with schematic preparation for this article.

■

REFERENCES

(1) Golberg, D.; Bando, Y.; Tang, C.; Zhi, C. Boron Nitride Nanotubes. Adv. Mater. 2007, 19, 2413−2432. (2) Fathalizadeh, A.; Pham, T.; Mickelson, W.; Zettl, A. Scaled Synthesis of Boron Nitride Nanotubes, Nanoribbons, and Nanococoons Using Direct Feedstock Injection into an Extended-Pressure, Inductively-Coupled Thermal Plasma. Nano Lett. 2014, 14, 4881− 4886. 4414


Article

ACS Applied Nano Materials (23) Antillon, M.; Nautiyal, P.; Loganathan, A.; Boesl, B.; Agarwal, A. Strengthening in Boron Nitride Nanotube Reinforced Aluminum Composites Prepared by Roll Bonding. Adv. Eng. Mater. 2018, 20 (8), 1800122. (24) Nautiyal, P.; Rudolf, C.; Loganathan, A.; Zhang, C.; Boesl, B.; Agarwal, A. Directionally Aligned Ultra-Long Boron Nitride Nanotube Induced Strengthening of Aluminum-Based Sandwich Composite. Adv. Eng. Mater. 2016, 18, 1747−1754. (25) Bhuiyan, M. M. H.; Wang, J.; Li, L. H.; Hodgson, P.; Agarwal, A.; Qian, M.; Chen, Y. Boron Nitride Nanotube Reinforced Titanium Metal Matrix Composites with Excellent High-Temperature Performance. J. Mater. Res. 2017, 32, 3744−3752. (26) Yamaguchi, M.; Meng, F.; Firestein, K.; Tsuchiya, K.; Golberg, D. Powder Metallurgy Routes toward Aluminum Boron Nitride Composites, their Morphologies, Structures and Mechanical Properties. Mater. Sci. Eng., A 2014, 604, 9−17. (27) Lahiri, D.; Hadjikhani, A.; Zhang, C.; Xing, T.; Li, L. H.; Chen, Y.; Agarwal, A. Boron Nitride Nanotubes Reinforced Aluminum Composites Prepared by Spark Plasma Sintering: Microstructure, Mechanical Properties and Deformation Behavior. Mater. Sci. Eng., A 2013, 574, 149−156. (28) Yamaguchi, M.; Pakdel, A.; Zhi, C.; Bando, Y.; Tang, D.-M.; Faerstein, K.; Shtansky, D.; Golberg, D. Utilization of Multiwalled Boron Nitride Nanotubes for the Reinforcement of Lightweight Aluminum Ribbons. Nanoscale Res. Lett. 2013, 8, 3. (29) Yamaguchi, M.; Tang, D.-M.; Zhi, C.; Bando, Y.; Shtansky, D.; Golberg, D. Synthesis, Structural Analysis, and In Situ Transmission Electron Microscopy Mechanical Tests on Individual Aluminum Matrix/Boron Nitride Nanotube Nanohybrids. Acta Mater. 2012, 60, 6213−6222. (30) Kim, K. S.; Kingston, C. T.; Hrdina, A.; Jakubinek, M. B.; Guan, J.; Plunkett, M.; Simard, B. Hydrogen-Catalyzed, Pilot-Scale Production of Small-Diameter Boron Nitride Nanotubes and Their Macroscopic Assemblies. ACS Nano 2014, 8, 6211−6220. (31) Kim, K. S.; Kim, M. J.; Park, C.; Fay, C. C.; Chu, S.-H.; Kingston, C. T.; Simard, B. Scalable Manufacturing of Boron Nitride Nanotubes and their Assemblies: A Review. Semicond. Sci. Technol. 2017, 32, No. 013003. (32) Xue, Y.; Dai, P.; Zhou, M.; Wang, X.; Pakdel, A.; Zhang, C.; Weng, Q.; Takei, T.; Fu, X.; Popov, Z. I.; Sorokin, P. B.; Tang, C.; Shimamura, K.; Bando, Y.; Golberg, D. Multifunctional Superelastic Foam-Like Boron Nitride Nanotubular Cellular-Network Architectures. ACS Nano 2017, 11, 558−568. (33) Adnan, M.; Marincel, D. M.; Kleinerman, O.; Chu, S.-H.; Park, C.; Hocker, S. J. A.; Fay, C.; Arepalli, S.; Talmon, Y.; Pasquali, M. Extraction of Boron Nitride Nanotubes and Fabrication of Macroscopic Articles Using Chlorosulfonic Acid. Nano Lett. 2018, 18, 1615−1619. (34) Kim, K. S.; Jakubinek, M. B.; Martinez-Rubi, Y.; Ashrafi, B.; Guan, J.; O’Neill, K.; Plunkett, M.; Hrdina, A.; Lin, S.; Denommee, S.; Kingston, C.; Simard, B. Polymer Nanocomposites from Free-Standing, Macroscopic Boron Nitride Nanotube Assemblies. RSC Adv. 2015, 5, 41186−41192. (35) Li, L. H.; Chen, Y. Superhydrophobic Properties of Nonaligned Boron Nitride Nanotube Films. Langmuir 2010, 26, 5135−5140. (36) Smith, M. W.; Jordan, K. C.; Park, C.; Kim, J.-W.; Lillehei, P. T.; Crooks, R.; Harrison, J. S. Very Long Single- and Few-Walled Boron Nitride Nanotubes via the Pressurized Vapor/Condenser Method. Nanotechnology 2009, 20, 505604. (37) Chen, X.; Dmuchowski, C. M.; Park, C.; Fay, C. C.; Ke, C. Quantitative Characterization of Structural and Mechanical Properties of Boron Nitride Nanotubes in High Temperature Environments. Sci. Rep. 2017, 7, 11388. (38) Chen, B.; Cheng, A. H.-D.; Chou, T.-W. A Nonlinear Compaction Model for Fibrous Preforms. Composites, Part A 2001, 32, 701−707. (39) Arif, M. F.; Kumar, S.; Shah, T. Tunable Morphology and its Influence on Electrical, Thermal and Mechanical Properties of Carbon Nanostructure-Buckypaper. Mater. Des. 2016, 101, 236−244.

(40) Golberg, D.; Costa, P. M. F. J.; Lourie, O.; Mitome, M.; Bai, X.; Kurashima, K.; Zhi, C.; Tang, C.; Bando, Y. Direct Force Measurements and Kinking under Elastic Deformation of Individual Multiwalled Boron Nitride Nanotubes. Nano Lett. 2007, 7, 2146−2151. (41) Zobelli, A.; Ewels, C. P.; Gloter, A.; Seifert, G.; Stephan, O.; Csillag, S.; Colliex, C. Defective Structure of BN Nanotubes: From Single Vacancies to Dislocation Lines. Nano Lett. 2006, 6, 1955−1960. (42) Schmidt, T. M.; Baierle, R. J.; Piquini, P.; Fazzio, A. Theoretical Study of Native Defects in BN Nanotubes. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 67, 113407. (43) Nautiyal, P.; Boesl, B.; Agarwal, A. The Mechanics of Energy Dissipation in a Three-Dimensional Graphene Foam with Macroporous Architecture. Carbon 2018, 132, 59−64. (44) Vanossi, A.; Manini, N.; Urbakh, M.; Zapperi, F.; Tosatti, E. Colloquium: Modeling friction: From nanoscale to mesoscale. Rev. Mod. Phys. 2013, 85, 529−552. (45) Nautiyal, P.; Mujawar, M.; Boesl, B.; Agarwal, A. In-situ Mechanics of 3D Graphene Foam Based Ultra-stiff and Flexible Metallic Metamaterial. Carbon 2018, 137, 502−510. (46) Embrey, L.; Nautiyal, P.; Loganathan, A.; Idowu, A.; Boesl, B.; Agarwal, A. 3D Graphene Foam Induces Multifunctionality in Epoxy Nanocomposites by Simultaneous Improvement in Mechanical, Thermal and Electrical Properties. ACS Appl. Mater. Interfaces 2017, 9, 39717−39727. (47) Nautiyal, P.; Zhang, C.; Champagne, V. K.; Boesl, B.; Agarwal, A. In-situ Mechanical Investigation of the Deformation of Splat Interfaces in Cold-Sprayed Aluminum Alloy. Mater. Sci. Eng., A 2018, 737, 297− 309. (48) Li, Y.; Kroger, M. Viscoelasticity of Carbon Nanotube Buckypaper: Zipping-Unzipping Mechanism and Entanglement Effects. Soft Matter 2012, 8, 7822−7830. (49) Ashby, M. Material and Process Selection Charts. The CES EduPack Resource Booklet 2; Cambridge University: Cambridge, U.K., 2010. (50) Dumitrica, T.; Yakobson, B. I. Rate Theory of Yield in Boron Nitride Nanotubes. Phys. Rev. B: Condens. Matter Mater. Phys. 2005, 72, No. 035418. (51) Miyamoto, Y.; Rubio, A.; Berber, S.; Yoon, M.; Tomanek, D. Spectroscopic Characterization of Stone-Wales Defects in Nanotubes. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 69, 121413. (52) Smith, D. S.; Bourret, A. A. J.; Nait-Ali, B.; Pennec, F.; TessierDoyen, N.; Alzina, A.; Otsu, K.; Matsubara, H.; Elser, P.; Gonzenbach, U. T. Thermal Conductivity of Porous Materials. J. Mater. Res. 2013, 28, 2260−2272. (53) Lahiri, D.; Das, S.; Choi, W.; Agarwal, A. Unfolding the Damping Behavior of Multilayer Graphene Membrane in the Low-Frequency Regime. ACS Nano 2012, 6, 3992−4000. (54) Nautiyal, P.; Boesl, B.; Agarwal, A. Harnessing Three Dimensional Anatomy of Graphene Foam to Induce Superior Damping in Hierarchical Polyimide Nanostructures. Small 2017, 13, 1603473. (55) Agrawal, R.; Nieto, A.; Chen, H.; Mora, M.; Agarwal, A. Nanoscale Damping Characteristics of Boron Nitride Nanotubes and Carbon Nanotubes Reinforced Polymer Composites. ACS Appl. Mater. Interfaces 2013, 5, 12052−12057. (56) Nautiyal, P.; Embrey, L.; Boesl, B.; Agarwal, A. Multi-Scale Mechanics and Electrical Transport in a Free-Standing 3D Architecture of Graphene and Carbon Nanotubes Fabricated by Pressure Assisted Welding. Carbon 2017, 122, 298−306. (57) Xu, M.; Futaba, D. N.; Yamada, T.; Yumura, M.; Hata, K. Carbon Nanotubes with Temperature-Invariant Viscoelasticity from −196° to 1000°C. Science 2010, 330, 1364−1368. (58) Eskin, G. I.; Eskin, D. G. Ultrasonic Treatment of Light Alloy Metals; CRC Press: Boca Raton, FL, USA, 2015. (59) Tang, D.-M.; Ren, C.-L.; Wei, X.; Wang, M.-S.; Liu, C.; Bando, Y.; Golberg, D. Mechanical Properties of Bamboo-like Boron Nitride Nanotubes by In Situ TEM and MD Simulations: Strengthening Effect of Interlocked Joint Interfaces. ACS Nano 2011, 5, 7362−7368. 4415


Article

ACS Applied Nano Materials (60) Liao, Y.; Chen, Z.; Connell, J. W.; Fay, C. C.; Park, C.; Kim, J.-W.; Lin, Y. Chemical Sharpening, Shortening, and Unzipping of Boron Nitride Nanotubes. Adv. Funct. Mater. 2014, 24, 4497−4506. (61) Gao, Z.; Zhi, C.; Bando, Y.; Golberg, D.; Serizawa, T. Noncovalent Functionalization of Disentangled Boron Nitride Nanotubes with Flavin Mononucleotides for Strong and Stable Visible-Light Emission in Aqueous Solution. ACS Appl. Mater. Interfaces 2011, 3, 627−632. (62) Zhi, C. Y.; Bando, Y.; Tang, C. C.; Huang, Q.; Golberg, D. Boron Nitride Nanotubes: Functionalization and Composites. J. Mater. Chem. 2008, 18, 3900−3908.

4416


High-Temperature Mechanics of Boron Nitride Nanotube âBuckypaper

Recommend Documents

High-Temperature Mechanics of Boron Nitride Nanotube âBuckypaper