_________________________________________________________

Novel routes of h-BN nanostructure aerogel synthesis via freeze casting, carbon substitution reaction and template-assisted AACVD for applications in energy and telecoms
__________________________________________________________

Abstract

Overwhelming current research has shown that hexagonal Boron Nitride (h-BN) has a unique mix of outstanding thermal conductivity, mechanical and electrical insulating properties, which makes it a promising material for wide range of applications.

Polymer materials, which have excellent dielectrics properties, are currently used in radiofrequency applications as radome materials and laminated glass composites, however their thermal conductivity is poor, and the mechanical properties need improvement. Therefore, combining h-BN with polymer materials to form composites could solve these issues. A low dielectric constant, and low loss tangent are important parameters for radiofrequency applications, as these parameters indicate how the material will interfere and impact the information transmittance through electromagnetic waves.

The dispersion of the filler inside the polymer matrices has a profound impact on the thermal, mechanical and dielectric properties. An aerogel as a full 3D network could help to control the dispersion of the polymers, and thus is a promising route for BN-polymer composites. Furthermore, by forming an aerogel you have more control over the porosity, and other network parameters. Currently there are two challenges in the field, making BN aerogels with control over porosity and crystallinity, and infiltrating the polymer inside the structure.

This review will provide an overview to BN and BN composite materials and discuss the current landscape of methods and aerogel-functionalisation techniques. Developing new processes for efficient polymer infiltration could lead to the next generation of h-BN-polymer composites with the ideal properties for radome applications.

Tables of contents

Abstract

1. Introduction

2. Hexagonal Boron Nitride an introduction

2.1. Theory on the thermal conductivity of materials

2.1.1. Different models in thermal conductive polymer composite materials

2.2. Dielectric properties of materials

2.2.1. Theory on the dielectric properties of composites materials

2.3. Boron nitride composites materials

2.3.1. Study of Boron Nitride Polypropylene composites

2.3.2. Study of Boron nitride/Polyethylene composites

2.3.3. Boron nitride functionalisation

3. Introduction to hexagonal Boron Nitride Aerogel structure composites

3.1. Carbon Aerogel structure (Freeze casting method)

3.2. Carbon substitution reaction

3.3. h-BN coating (Template-assisted AACVD)

3.4. Polymer infiltration in porous structure

4. Conclusions

References

1.    Introduction

Over the coming decade government and industries will focus more funds in energy storage capacities, developing renewable energy and reducing the fuel consumption which are one of the significant incoming issues of the XXI century. For example, one way to reduce fuel consumption in transport is to develop new advanced materials to offer the same mechanical strength properties but in a less dense material. Regarding the energy storage capacities one of the big challenge is to exhaust the heat generated during the charge or the discharge which decrease the performance of the battery. In addition, a lot of research is also carried out to improve the life comfort of people, for example airplane company want to propose to their passenger a high-speed Wi-Fi connection. This could be possible by performing new radome materials exhibiting low dielectric constant and loss tangent at higher frequency.

The nanomaterials showed a great interest in the past 30 years showing unusual and outstanding chemical and physics properties.  A lot of researcher think that nanomaterials could be the solution of current and future issues. The hexagonal Boron Nitride materials (h-BN) is one of them. It has the advantages to present a high thermal conductivity and strong mechanical properties while being an electrical insulator with low dielectric constant and loss tangent. For these reasons h-BN appears to be a promising candidate to solve the issues listed above.

2.    Hexagonal Boron Nitride an introduction

Hexagonal boron nitride (h-BN) is a highly adaptable material with multifarious implications for use in the modern world. Specifically, its lightweight, superb oxidation resistance, chemical stability (passivity to reactions with acids, alkalis and melts), hydrophobicity, high thermal conductivity and thermal stability (with a melting point near 2600 ℃)¹, are strong indications of its utility.  Most importantly, h-BN’s singular atomic and crystal structure allow it change. Hexagonal  Boron nitride is a planar material with a structure similar to graphene, composed of boron and nitrogen atoms. The B-N-B-N bonds are arranged in a hexagonal shape. Noteworthy are each monolayer, boron and nitrogen atoms, which are bound by strong sp2 covalent bonds which act to form a network of (BN)3 rings as shown in Figure 1. The hexagons of neighbouring planes in h-BN are superposed, i.e., B and N atoms are in succession along the c-axis (bonded together by Van Der Walls force), while in graphite, they are shifted by half a hexagon. Moreover, due to the difference in electronegativity of B and N, the B−N bonds in BN materials are partially ionic, in contrast with the purely covalent C−C bonds in graphitic structures. These can lead to the so-called “lip−lip” interactions between neighbouring layers/shells in BN nanostructures (i.e., chemical bonds form as bridges or “spotwelds” between the atoms of adjacent layers/shells). This phenomenon contributes to a metastable energy minimum through decreasing the number of dangling bonds at the edges/tips and reducing the “frustration” effect (i.e., when B−B and N−N bonds form instead of the energetically more favourable B−N bonds).

These weak interplanar bonding allows h-BN layers to slide over each other easily and thus make h-BN an excellent alternatives in solid lubricant applications^2,3.

Figure 1: Chemical bond (left) and crystal structure (right) of Boron Nitride

h-BN is as an exceptional electrical insulator with good dielectric properties (small dielectric constant, low volume resistivity and a low loss tangent)^1,4. The location of the bonding electrons has an important impact on the h-BN dielectric properties. The electronegativity of nitrogen is higher than boron. Thus, electrons in the BN sp2 orbitals establish strong bonding with charge localization closer to nitrogen atoms than the boron atoms. In addition, electrons in π orbitals are more localized on the nitrogen atoms^3,5. On the other hand, in opposition to h-BN, in Graphite the electrons are delocalized, where they  can move through the structure to allow electrical conduction through the graphite network^6,7. Additionally, other crucial aspects include boron nitride’s ability in terms of thermal stock resistance, environmental safety and low toxicity (except in nanometre scale, toxicity no established)^1,6,8.

Notwithstanding the aforementioned benefits for boron nitride, one must consider its limitations, for when it is compared to graphite, we see that h-BN is also anisotropic. As a result, the thermal, mechanical and electrical properties of h-BN will be dependant following a measurement parallel or perpendicular to the c axis of the crystal structure as shown in Table 1. In addition, h-BN has poor sintering properties, resulting from the strong covalence of B-N bond has a rather low self-diffusion coefficient. Consequently, even sintered under high temperature (>2000°C)  with pressure assistance, it is challenging to manufacture dense materials^1,9. The low relative density limits the application of h-BN; as a result, h-BN is usually used as additives to other materials such as ceramics and polymers to enhance their properties. A detailed review⁹ of h-BN ceramic composite materials can be found by Duan et al.

Table 1: Main characteristics of hexagonal boron nitride edited from V. Feng review^{10,4,8,9,11–13}

Density	2.27g/cm3	Band Gap Energy	5.95 eV
Bulk modulus	36.5GPa	Minimum Direct Band Gap	6.47 eV
Young’s modulus (300K)	32 ±GPa	Dielectric Constant	║to c axis 5.06
	3GPa		⊥ to c axis 6.85
Debye Temperature	400K	Loss tangent (9.3GHz)	0.0002
Moh’s Hardness	1.5	Volume resistivity (293K)	1013-1015 Ω.cm
Lattice constant	a0=2.504 Å	Thermal conductivity	400 W/m.K
			║ to c axis ≤ 30W/m.K
	c0=6.661 Å		⊥ to c axis ≤ 600W/m.K
Specific Heat	0.8J/g.K	Thermal expansion	║to c axis 37.7×10-6/K
			⊥to c axis -2.72×10-6/K

Advantageously, we can use h-BN filled polymer composite materials to improve the thermal conductivity of composites. Regarding to the discussion on the h-BN/polymer composite in scholarly literature, some general models for effective thermal conductivity of composite will be explored in the next section of this literature review. These models will help us to predict the thermal conductivity of the composites, and at the same time provide information on the heat transfer mechanisms in these composites.

2.1. Theory on the thermal conductivity of materials

 

A well description of the different thermal conductivity models used in the past centuries will be summarized below.

Firstly, the simple definition of the thermal conductivity is translated in the the basic principle called Fourier’s Law, in equation 1 of thermal conductivity which is the ability of a materials to conduct heat¹⁴.

q=-kdTdx

(1)

The constant, k, is referred to as thermal conductivity in W/m.K.

Thermal conductivity, k, depends on position and temperature (Equation 2):

k=k[r⃗,Tr⃗,t]

(2)

However, for most homogenous materials k can be treated as a constant¹⁴, k = k(T), but for anisotropic materials, thermal conductivity varies with orientations. Moreover, in heterogeneous materials e.g. composite materials, thermal conductivity can vary with spatial locations.

2.1.1.     Different models in thermal conductive polymer composite materials

Two basic models were proposed by Agari and al. the parallel and series conduction models, as shown in Figure 2^15,16.

Figure 2: Schematics of two simple conductive models¹⁵

Equation 3 shows the parallel conduction mode, with the composite materials would have the highest effective thermal conductivity^15,16.

kC=1-ϕkm+ϕkf

(3)

Where

ϕ is the volume fraction of filler, k_c, k_m, and k_m are respectively the thermal conductivities of the composite material, the polymer matrix and the filler. Turning to Equation 4, we can see the series conduction model. This will result in the lowest effective thermal conductivity for the composite materials^15,16.

1kc=1-ϕkm+ϕkf

(4)

Moreover,  the two extreme models, which, as we have just seen, were inaccurate. Maxwell and Eucken (Equation 5) as well as Bruggeman^17,18 (Equation 6) designed two models to calculate the effective thermal conductivity of composite materials with random distribution of fillers in the matrix, bridging the previous gap.

kc=kmkf+2km+2ϕ(kf-km)kf+2km-ϕ(kf-km)

(5)

1-ϕ=kf-kCkf-kmkmkC13

(6)

As conclusion find in the cited review¹⁰, both models presume that the fillers do not interact with each other and are uniformed sphere particles that are randomly distributed in a homogenous matrix medium. Yet the Maxwell-Eucken model gives a fairly accurate prediction at low filler concentration (under 25 vol%), whereas the Bruggeman model is relevant for predicting the effective thermal conductivity of composites containing a higher fraction of filler (over 20 vol%)¹⁹.

However, these models still present some problematic parameters. Such as particles are supposed spherical, and that the fillers don’t interact each other¹⁶. Agari thus, trying to rectify these inaccuracies, motivated the parallel and series models and developed a new semi-empirical model (Equation 7)²⁰:

log⁡kC=ϕC2log⁡kf+(1-ϕ)⁡log⁡C1km

(7)

Two new parameters, C1 (fillers) and C2 (host polymer) are proposed, which are determined experimentally^15,20 which take into account the dispersion state of the fillers.

Additionally, Agar presents a model which includes the effect of aspect ratios of filler on the thermal conductivity of composites, which can be seen in the modified Equation 8²¹. The aspect ratios in nanomaterials fields have a significant impact in term of properties. That’s why I thought this model will be pertinent:

log⁡kC=ϕClog⁡LD+Elog⁡kf+1-ϕlog⁡C1km

(8)

Where

Clog⁡LD+E=C2

C and E are constants determined experimentally and L/D is the aspect ratio of the filler (Length/Diameter). In conclusion Agari models are able to predict the thermal conductivity of composites quite accurately compare to other methods^16,17. On the other hand, one of the disadvantage of the Agari models is that delicate experiments need to be completed to calculate the corresponding constant values before precise calculations can be entirely determined¹⁷.

2.2. Dielectric properties of materials

Dielectric materials have been widely used in wireless communication systems as resonators, antennas, filters, substrates, etc. In recent years, the millimetre wavelength range has received much attention for applications in communication systems. It is expected to enable high-capacity and high-speed information transfer. The dielectric materials used for these systems must have a low dielectric constant (ε_r) to reduce the propagation delay time and minimize the capacitive coupling effects. At the same time, ε_r relates to the size of the dielectric device. Therefore, the ability of optimizing the ε_r value is desired. The low dielectric loss (tan (δ) a quantitative measure of the dissipation of energy resulting from the interaction of electromagnetic waves with the material) of the materials is effective in reducing the signal attenuation. In addition to the dielectric properties, the materials must satisfy several other requirements, such as high thermal conductivity, low moisture absorption, and suitable mechanical stiffness. Ceramic filler-filled polymer matrix composites are considered to be promising candidates for satisfying these requirements.

2.2.1.     Theory on the dielectric properties of composites materials

Several theoretical models have been published to anticipate the dielectric constant of composites in function of fillers content. For single filler/polymer two components composite different models were proposed since the 19^th century. We will here focus on three different models. Firstly, the Lorentz-Lorenz effective-medium model proposed in 1880^22,23 a simple model between the fillers and the matrix, shown in Equation 12.

εc-1εc+2=faεa-1εa+2+fbεb-1εb+2    (12)

Where f_a represents the volume fraction of the fillers and f_b represents the volume fraction of the polymer matrix. Respectively ε_c, ε_a and ε_b are the dielectric constant of the composite, the fillers and the polymer. The advantage of this model it’s that it can be generalized to systems containing more than two phases by adding more terms. On the other hand, it doesn’t include intermolecular interaction to more accurately describe dense materials.

Hence, taking these limitations into account in 1904 Maxwell Garnett exposed his effective-medium model²⁴. As any such theory, it aims to approximate a complex electromagnetic medium such as a colloidal solution of gold microparticles in water with a homogeneous effective medium. The Maxwell Garnett mixing formula (Equation 13) gives the permittivity of this effective medium (or, simply, the effective permittivity) in terms of the permittivities and volume fractions of the individual constituents of the complex medium.

εc-εbεc+2εb=faεa-εbεa+2εb     (13)

Where

fa=43πra3∕Vis the volume fraction occupied by the phase a, V is the total volume of the system, ε_c, ε_a and ε_b are respectively the dielectric constant of the composite, the fillers and the polymer.

Equation 12 and 13 have the same general form:

εc-εhεc+2εh=faεa-εhεa+2εh+fbεa-εhεa+2εh    (14)

Where ε_h is the dielectric function of a host medium. Thus, ε_h equals 1 in void and ε_b for the Lorentz-Lorenz and Maxwell Garnett expressions, respectively. If f_a > f_b, then the more appropriate choice for ε_h in the Maxwell Garnett case is ε_a. However, the resulting values of ε_c are different for the two choices. Bruggeman resolved this dilemma by proposing that neither phase should be given reference, but that the inclusions should be considered as being embedded in the effective medium itself¹⁸. In the above formulation this is equivalent to choosing ε_h = ε_c, in which case the left-hand side of Equation 14 vanishes and:

0=faεa-εcεa+2c+fbεa-εcεa+2εc    (15)

This is the Bruggeman effective-medium expression, or in conventional terminology the effective medium approximation (EMA). The Maxwell Garnett and Bruggeman expressions have been derivated many different ways, generally by requiring that the scattering of a wave off an inclusion shall vanish in the forward direction. Recent treatments^25,26 have attempted to understand their differences more fundamentally in terms of the microstructure of the composite material. These investigations showed that the Maxwell Garnette result follows from a coated-sphere configuration, where inclusions of phase a are completely surrounded by material of phase b. The Bruggeman expression follows from an aggregate model, where particles of phase a and phase b are mixed on a random basis. This conclusions ore, of course, completely consistent with the corresponding choice of host material, ε_h, in the derivation of Equations 13 and 15. We comment finally on the importance of microstructure in effective medium theories. The average microstructure has obviously entered equation 14 through the volume fractions f_a and f_b. But the detailed microstructure has also entered implicitly through the assumptions that the inclusions where spherical and no-interacting.

These assumptions are hardly ever satisfied in actual heterogeneous materials, which generally have a random and complex form. The shape of the particles determines how effectively they are screened, which affects the microscopic polarizations and fields, which in turn determines the functional relationship between ε_c and the dielectric functions of the constituents.

2.3.  Boron nitride composites materials

Manufacturing a multifunctional material with a combination of high thermal conductivity (k_eff), high electrical resistivity (σ), strong mechanical properties, low dielectric constant and loss tangent (ε, tan (δ)) has recently gained interest due to its potential wide range of applications. As previously mentioned in this literature review, the practical outcomes of this could be extensive. For example, it can be used in batteries to dissipate the heat generated during the charge/discharge, and in radomes materials to increase the mechanical strength and thermal properties while maintaining or improving the radio frequency transparency in the high frequency range. A lot of different polymers such as Epoxy, Polystyrene and polyethylene (PE) are used as electrical insulators materials for their good dielectrics properties, on the other hand they are lacking of suitable thermal and mechanical properties^27,28,29. With the incorporation of thermally conductive, strong, but electrical insulating ceramic fillers into the polymer matrix, the composite materials have the ability to maintain his good dielectric properties and  improving the thermal conductivity and mechanical strength of polymers. To solve this issue different ceramic such as zinc oxide (ZnO, keff=60 W.m∙K^-1, density=5.61 g/cm3), aluminium oxide (Al2O3, keff=30 W.m∙K^-1, density=4.00 g/cm3)³⁰ were used, but on the other hand they are too heavy and not enough strong mechanically. Among all, h-BN has the remarkable potential to be the filler of choice due to its lower density (2.27g/cm3), low dielectric constants (around 5), higher thermal conductivity (600 W.m∙K^-1 for x-y plane and 30 W.m∙K^-1 for z-plane) as well as mechanical strength (Experiment data on Boron nitride nanotubes yield stress: 1.1-1.3 TPa³¹, Youg’s Modulus: 722-1100 GPa^32,33, Tensile strength: 11-32 GPa³⁴). Table 2 below shows the thermal conductivity of some boron nitride filled composite materials reported from literatures.

Table 2: Thermal conductivity of different composite filled with h-BN adapted from V. Feng review¹⁰

Polymer matrix	Thermal conductivity of polymer	Fraction of filler	Thermal conductivity of composite	Dielectrics properties	Mechanical properties	Size of filler	Reference
Epoxy resin	0.202 W/m.K	31 vol% h-BN	2.3 W/m.K	–	–	N. A	Bujard35
Epoxy resin	0.214 W/m.K	69 wt% h-BN	3.6 W/m.K	–	Young’s Modulus: 13.3 GPa	13.8 µm h-BN	Xia and al36
Epoxy Resin	0.214 W/m.K	Mix of 43.6 wt% h-BN and 26.3 wt% Kenaf fibre	6.418 W/m.K	–	Young’s Modulus: 8.2 GPa	13.8 µm h-BN 11.5µm Kenaf fibre	Xia and al36
Epoxy resin	0.2 W/m.K	40 vol% h-BN 40 vol% AlN	8.0 W/m.K	–	–	18µm h-BN 14.4µm AlN	Hong and al37
Epoxy resin	0.20 W/m.K	42.79 vol% BN	3.62 W/m.K	–	Young’s Modulus: 12 GPa	50 µm A-BN d:10 µm l:30 µm W-BN	Kim and al38
Epoxy resin	0.20 W/m.K	26.5 vol% h-BN and c-BN	19.0 W/m.K	–	–	0.4 and 0.2 µm h-BN 1 µm c-BN	Yung and Liem39
Epoxy resin	0.202 W/m.K	60 wt% h-BN	1.052 W/m.K	Dielectric constant:5.6 Loss Tangent:0.072	Flexural Strength: 80 MPa	0.6-1.2 µm h-BN	Gu and al27
Epoxy-ETDS	0.2 W/m.K	70 wt% h-BN	4.11 W/m.K	–	Young’s Modulus: 8.8 GPa	12 µm h-BN	Kim and al40
Epoxy-ETDS	0.202 W/m.K	70 wt% h-BN	3.88 W/m.K	–	Young’s Modulus: 7.8 GPa	12 µm h-BN	Kim and al40
Epoxy resin	0.15 W/m.K	40 wt% h-BN	0.96 W/m.K (vertical)	–	Young’s Modulus: 4.55 GPa	5 µm h-BN	Lin and al41
Epoxy resin	0.236 W/m.K	5 wt% h-BN	0.329 W/m.K	Loss tangent: 0.01	–	0.1-3 µm, 4.8 nm thickness h-BN	Yu and al42
HDPE	0.26 W/m.K	35 vol% h-BN	1.24 W/m.K	Dielectric constant: 4 Volume resistivity: 1015Ω.cm-1	Tensile strength: 30MPa	0.5 µm h-BN	Zhou and al28,43
HDPE	0.45-0.55 W/m.K	50 vol% h-BN	3.66 W/m.K	–	–	N.A	Lee and al44
HDPE	0.26 W/m.K	30 vol% 5:1 BN:Al2O3	1.40 W/m.K	–	–	0.5 µm h-BN	Zhou and al28,43
PP	0.233 W/m.K	55 wt% h-BN	2.438 W/m.K	–	–	10 µm h-BN	Cheewaw-uttipong45
PP	0.22 W/m.K	25 wt% h-BN	0.471 W/m.K	–	–	5-10 µm h-BN	Chen29
PP	0.22 W/m.K	29 vol% h- BN	1.9 W/m.K	–	Storage modulus: 4.2 GPa Loss modulus:880 MPa	7-10 µm h-BN	Cheewaw-uttipong46

2.3.1.     Study of Boron Nitride Polypropylene composites

Polypropylene polymer filled with boron nitride was studied by Cheewawuttipong and al.^45,46 and more precisely the influence of the size of the h-BN particles and the viscosity of the Polypropylene. Two different size of h-BN (BN-s: 1-2 µm, and BN-l: 7-10µm) were mixed in high or low viscosity polypropylene (PP-H, MFR = 7-8 g/10 min, PP-L, Melt flow rate (MFR) = 26-29 g/10 min). A batch kneader at a barrel temperature of 250 ºC was used to mix the fillers and the polymer. They used the compression moulding technique at 200°C and 19.6MPa to create the composite samples.

Regarding the thermal conductivity, we observe that the thermal conductivity increases with the h-Bn content. Also, there is no significant influence of the initial viscosity of the polypropylene on the thermal conductivity values. The main impact come from the size of the h-BN, where the thermal conductivity is more important with the largest particles as shown in Figure 3 below.

Figure 3: Thermal conductivity as a function of BN content for each PP/BN composite

Dynamic analyses were performed to measure the storage and loss modulus of each samples. As shown in Figure 4 we can see that the storage and loss modulus increase with the filler content. However, we can’t conclude about the correlativity between the modulus and the viscosity of the polypropylene. This tendency would agree with the thermal conductivity at high BN concentration. This implies that the network structure of h-BN in PP/BN composite have an influence on their solid mechanical properties.

Figure 4: Storage modulus and loss modulus as a function of BN content for various PP/BN composites at temperature of 25 °C

2.3.2.     Study of Boron nitride/Polyethylene composites

Polyethylene polymer has an interest in telecommunications fields due to his low dielectric constant (ε_r=2.0)²⁸ and loss tangent (tan (δ)=0.0012)⁴⁷. However, the thermal conductivity and mechanical properties could be drastically improved by the addition of h-BN fillers.

Zhou et al. studied boron nitride reinforced polyethylene composites. And regarding the dielectric constant the authors recorded that adding 30 vol% h-BN filler inside the HDPE matrix will increase the dielectric constant from 2 to 4 and decrease the volume resistivity from 10¹⁷ to 10¹⁵ Ω.cm²⁸ as shown in the figure 5.

Figure 5: Dielectrical constant and volume resistivity of the composites as a function of filler content.

In relation to the mechanical properties Zhou also investigated the elongation and the stress at break as described in figure 6 and 7.

Figure 6 and 7: Elongation at break of the composites as a function of volume fraction of filler (left). Dielectrical constant and volume resistivity of the composites as a function of filler content (right)

 

It was found that with increasing filler concentration the stress at break increased compared to that of pure HDPE. However, at very low filler content the stress at break decreased; the stress at break of pure HDPE decreased from 26MPa to 10.1MPa, when filled with 3.0% of BN by volume. Because the filler can be distributed only in the amorphous phase of HDPE due to its high crystalline degree, thus, the local concentration of the filler in amorphous phase in HDPE is high. In this case, the filler served as defects in fact⁴⁸.

Mixing raw h-BN fillers in different polymers matrices increase the properties of the pure polymer in term of thermal and mechanical properties and maintain the dielectrics constants in a good range. Researchers tried to improve the potential of these composites by enhancing the network arrangement of the fillers inside the matrix. The functionalisation of the surface of the h-BN could offer new possibilities of rearrangement and interaction with the polymer.

2.3.3.     Boron nitride functionalisation

It is expected that many novel properties can emerge from a material through its smart functionalization, either physical or chemical. This is also proven to be applicable in case of h-BN. Its properties can be tailored, and many brand-new features and applications can be created directly via such functionalization. However, the high chemical stability/inertness of h-BN hinders its modifications. This makes the corresponding attempts to functionalize BN structures, both physically and chemically, a challenging research topic.

The surface modification of boron nitride is an effective method to improve the dispersion of h-BN in the polymer matrices²⁷. On the other hand, the parallel planes of boron nitride have no surface functional groups for chemical modification. In contracts, the vertical planes contain reactive groups that provide possibility for chemical bonding and dispersion in solution⁴⁰.

The authors compare two different study of chemical functionalisation and their impact on the thermal conductivity of the composites. The first one from Gu and al’s work reported using γ-aminopropyltriethoxysilane to modify h-BN, which allowed a slight enhancement of thermal conductivity of 60 wt% h-BN/epoxy composite from 0.997 to 1.052 W/m∙K²⁷. The second one inspired by Gu et al.’s work, Kim et al. proposed three different chemical functionalisation of h-BN fillers with sodium hydroxide followed with two different silane based surface curing agents: 3-glycidoxypropyltrimethoxysilane (KBM-403) and 3-choropropyltrimethoxysilane (KBM-703)⁴⁰. They believed these large particles tend to form fewer mismatch or damaged surface layer between filler and matrix thus minimized the thermally resistance in the composite. This point goes in contradiction with the conclusion observed by Gu and al²⁷, where they mentioned that decreasing the size of the fillers would allow more filler particles to surround the polymer  and allow the formation of a thermal conductive network in the final composites²⁷.

Other studies^41,42,49 have reported other chemical reactions to improve the thermal conductivity and mechanical properties of the final composites as indicated in Table 2, Table 3^{55,56, 57–64,65–67,}  and Figure 7⁶³.

Figure 7: Summary of chemical functionalization strategies of h-BN bulk-/nanomaterials. A charge is denoted when the compensating functional group is unknown⁶³.

Table 3: Mechanical and thermal conductivity enhancement of polymers using functional BN nanomaterials

 

3.    Introduction to hexagonal Boron Nitride Aerogel structure composites

Regarding the previous thermal and mechanical properties achieved by the different composites materials studied above we can see that they reached a limit in term of improvement. The h-BN functionalisation brought new possibilities to break this limit by favouring the creation of a network inside the polymer matrix to improve the thermal conductivity and mechanical strength while maintaining low dielectric constant. Based on this conclusion I wanted to focus my DPhil project on the extent to which a functionalised h-BN 3D aerogel filled with the proper polymer could break this plateau. The first step involves synthesising a carbon foam made of graphene oxide sheets mixed with carbon nanotubes, then chemically and thermally reducing to obtain graphene sheets and carbon nanotube mixture foam. The second step will be to coat this foam with h-BN and anneal the structure under oxygen to conserve only the h-BN coating, or use carbon substitution synthesis to replace the C-C bonds with B-N bonds. Then the functionalization of the obtained h-BN porous structure will be investigated to offer the best chance for his polymer infiltration. The details of this protocol are explained in the last part of my literature review.

3.1. Carbon Aerogel structure (Freeze casting method)

Since the first report of aerogels by Kistler in 1930s⁶⁴ several ultralight cellular materials such as silica aerogels⁶⁵, CNT aerogels⁶⁶, have been prepared to exploit their wide range of applications. Haiyan Sun and al synthesised an ultralight carbon aerogel made of carbon nanotubes and graphene sheets. They used a suspension of commercial carbon nanotubes (CNTs) and graphene oxide sheets (GO) stirred in water and then freeze dried for few days. They chemically reduced the graphene oxide with hydrazine vapour at 160°C for 24hrs.

This is the delicate part of this study, due to the fact that hydrazine it’s a really instable and toxic compound. We can find in the literature different techniques to reduce the graphene oxide sheets in graphene sheets.

Songfeng Pei and al⁶⁷ made an interesting review on the different possibilities to reduce graphene oxide, such as thermal reduction, chemical reduction, photocatalyst reduction or  solvothermal reduction. The advantage of this review is that the authors control and compare the yield of the reduction mechanism with electrical conductivity measurements and the evaluation of C/O ratio by X-ray photoelectron  spectroscopy (XPS) as shown in Table 4. The authors found that the thermal reduction of graphene is one of the most effective and safe technique. As a result, annealing reduction is usually carried out in vacuum⁶⁸, or an inert⁶⁹ or reducing atmosphere^69,70,71,72. Becerril et al.⁶⁸ have reduced GO films by thermal annealing at 1000 °C and found that a quality vacuum (<10^-5 Torr) is key for the recovery of GO, otherwise the films can be quickly lost through reaction with residual oxygen in the system. The same condition should also be considered in inert atmospheres.

Therefore, a reducing gas such as H₂ is added to consume the residual oxygen in the atmosphere. Furthermore, because of the high reducing ability of hydrogen at elevated temperatures, the reduction of GO can be realized at a relatively low temperature in a H₂ atmosphere.Wu et al. reported that GO can be well reduced at 450 °C for 2 h in an Ar/H₂ (1:1) mixture with a resulting C/O ratio of 14.9 and conductivity of ~1×10³ S/cm.

Table 4: Comparison of the reducing effect of GO by different methods⁶⁷

After the obtention of this full ultralight carbon foam the next step will be to change the carbon bonds in Boron and Nitrogen bonds while maintaining the same 3D structure. Two methods will be presented below the carbon substitution reaction and the h-BN CVD coating technique.

3.2. Carbon substitution reaction

The synthesis of boron nitride nanotubes (BNNTs) is really complex to control, and the yield of such materials are really low in lab scale production. Some laboratories  investigate the plasma synthesis method⁷³ and reach 30g/hr but the complexity of the reactor chamber and the high temperature needed (8000 K) leading to the limited accessibility to researchers producing large scale of BNNTs. That is why I thought it will be useful to use the advance up-scaling research in CNTs and then try to use their remarkable shape as a 3D substrate structure for carbon substitution reaction in Boron and Nitrogen bonds.

Han and al.⁷⁴ detailed a method involving carbon nanotubes substitution reaction to develop the synthesis of mass quantities of boron nitride nanotubes. Boron oxide vapor was reacted with nitrogen gas in the presence of carbon nanotubes to form boron nitride nanotubes, whose diameters and lengths are similar to those of the starting carbon nanotubes. It is proposed that carbon atoms of carbon nanotubes can be fully substituted by boron and nitrogen atoms through a general chemical reaction.

For confirming the speculated CNTs substituted reaction, BNNTs were chosen to be synthesized because BNNTs have the same layered structure and also close lattice constants compared with CNTs. The designed reaction can be expressed as:

B2O3+3Cnanotubes+N2→2BNnanotubes+3CO

It was expected that boron oxide gas generated from the B2O3 powder would flow up towards the region containing the carbon nanotubes and react with the nanotubes and the nitrogen (N2) gas. The reaction was carried out in an induction-heating system with susceptors made of graphite. B2O3 powder was placed in an open sintered graphite crucible and then covered with CNTs. Relatively pure multishell CNTs with typical diameters of approximately 10 nm used here were prepared by a metal-catalyzed chemical vapor deposition (CVD) method. The crucible was held in a flowing nitrogen atmosphere at 1500°C for 30min.

After this they characterised the materials with XRD, transmission electron microscopy (TEM), energy dispersive x-ray spectroscopy (EDS, and electron energy loss spectroscopy (EELS)⁷⁴. The  XRD confirmed the presence of hexagonal BN (two layered repeating units) and rhombohedral BN (r-BN: three-layered repeating units) (Figure 8). The TEM showed nice BNNTs with the same shape that the initial CNTs and with an interplanar spacing equals to 0.335nm which is consistent with the interplanar distance 0.333nm in both bulk h-BN and r-BN (Figure 9). The EELS spectra showed that individual nanotubes correspond to B-N sp² bonds and reveal the absence of K-edge absorption for carbon which indicate that all the carbon nanotubes reacted with the Nitrogen and B₂O₃ (Figure 10). Nevertheless, the EDS revealed the presence of iron nanoparticles encapsulated inside the BNNTs which was used as a catalyst for the synthesis of CNTs.

              

Figure 8 and 9 : On the left XRD spectrum and on the right TEM image of BNNTs

Figure 10: EELS spectra of BNNTs

This paper showed that carbon substitution reaction works with CNTs at high temperature (1500°C) and with the presence of Nitrogen and  B₂O₃. On the other hand, they didn’t mention the impact of the initial shape of the CNTS, aspect ratio, if they are Multi Walled (MWCNTs) or single walled carbon nanotubes (SWCNTs). And also, they only used spectroscopy characterization techniques and didn’t characterise the mechanical, electrical or thermal resistance properties of their BNNTS. Because maybe it’s just a nice h-BN “mould” that they obtained but with worth chemical and physical properties.

Another study from Golberg and al.⁷⁵ based on the same synthesis method than above, proved by characterised the BNNTs with High Resolution TEM (HRTEM) and EELS spectrum that the BNNTs produced from MWCNTs by a substitution reaction exhibit the zig-zag arrangement of all graphitic tubular sheets. Also noticed that two phenomena appeared, the first one is that the number of shells decrease and second one is that the alignment of the nanotubular shells is along the tube axis which lead to a flat tip-end termination of the BN nanotubes as shown in Figure 11.

Figure 11: representative HRTEM image of flat tip-end termination of a zig-zag nanotube.

The last study from D. Goldberg, Y. Bando and al was focused on carbon substitution reaction from only SWCNTs compare to the MWCNTs used in the previous study. They obtained by playing with different temperature B_x-C_1-x, B_x-C_1-x-y-N_y and BN single walled nanotubes. At 1523K they obtained only B_x-C_1-x SWNTs, at 1623K they obtained a mixture of B_x-C_1-x, B_x-C_1-x-y-N_y and fraction of BN SWNTs, finally at 1803K only BN SWNTs, with h-BN and r-BN crystal structure, were observed using HRTEM and EELS analysis. This study presents the advantage to use another source of carbon nanotubes, and also showed that playing with the temperature parameter we can obtain different materials as Boron doped and Boron Nitrogen doped SWCNTs.

Therefore, regarding the synthesis of my BN aerogel from my MWCNTs and graphene sheets we will have to pay attention to the decrease in number of shells of the MWCNTs and control that the temperature is uniform inside the furnace to avoid any Carbon by products. Also, these articles were useful regarding the characterisation techniques used to analyse this type of materials. The CVD techniques used for h-BN deposition on different type of substrate could offer more advantages such as using lower temperature, avoid the shells decrease phenomenon on MWCNTs and the flat tip-end termination.

3.3. h-BN coating (Template-assisted AACVD)

Template assisted method allow the BN materials to copy the structure of the templates with then a step to remove the template. Synthesis temperature for this method are usually lower than of carbon substitution reaction (over 1500°C) because the substrate doesn’t take part in the reaction and are easy to remove after the growth of BN by a simple annealing step under air. In addition, the morphology, size and properties of the template have an impact on the final BN products. For example, BN ultra-light (1.6mg/cm³) structure were obtained with Ni foam used as a catalyst and template at the same time. However, this route cannot be applied for the synthesis of BN based porous materials without a catalyst, the Ni template restricts the size and morphology of the final BN foam.

The study of Y. Song and al.⁷⁶ used use carbon aerogel templates to create BN aerogels via a template-assisted and catalyst free route. This is a good choice to avoid using catalyst because they are often metallic nano-particles and will have a negative impact on the dielectric properties of the final composites materials and his radio frequency transparency properties. The authors used a similar method described above to obtain a carbon aerogel⁷⁷. Then they used the chemical vapour deposition (CVD) which is a promising and controllable technique for the growth of thin films^78–80. They placed the carbon foam inside the furnace and used a 1:1 ratio BN precursor known as borazine (B₃N₃H₆) slowly introduced through a bubbler with 5sccm of Ar flow, in low pressure (100Pa) CVD furnace and a quite low temperature of 900°C for 30min with an Ar/H₂ (5:1) flow of 270sccm. The last step is an annealing process of 30min at 600°C in oxygen to burn the carbon template. Hence, the BN aerogel was synthesised with the composite morphology of graphene like-BN nanosheets and CNTS-like BNNTs. Raman and FT-IR were performed on this BN aerogels and confirm the presence of h-BN structure. SEM and TEM images showed in Figure 12 that the h-BN grown at the surface of the carbone nanotubes because we found that the internal diameter of the BNNTs is egal to the outside diameter of the initial CNTs (ID_BNNTs= 12.2nm compare to OD_CNTs=12.5nm)

Figure 12: SEM and TEM images of carbon aerogel template (left) and BN aerogel (right)

However, the BNNTs show a discontinuous and imperfect layered structure indicating that it has a lower crystallinity than the initial CNTs. Certainly, due to the catalyst free method and low temperature CVD growth environment in the study. Could be interesting to study the crystallinity behaviour of the BNNTs obtained with the same method but at higher growth temperature. The growth time as also its importance because below 30min the BN units are too week to support the 3D structure of the BN aerogel. And above 2hrs the density is too high and will certainly limits the polymer infiltration. Template assisted CVD techniques seems to be a promising method for growing BN aerogel with a catalyst free method. Different other BN precursor can be used such as boric acid inside an alumina boat in the furnace with an ammonia flow  at 900°C for 30-120min⁷⁸ or trymethylborate introduced through an aerosol unit with argon followed by an ammonia gas flow inside the furnace at 700°C for 30-120min⁷⁸.

3.4. Polymer infiltration in porous structure

The final step after obtaining the BN aerogel structure will be to fill this porous structure with a dedicated polymer suitable for the desired applications. Because the aerogel cannot be used as it is our interesting applications in radome materials. This 3D network will need a polymer matrix to support the mechanical stress of his environment, while the Bn network will help the polymer to improve his thermal conductivity and mechanical strength. The previous ultralight aerogel was used as it is for oil absorption properties⁸¹. It could absorb 160 times⁷⁶ his weight in oil while repealing water due to his high hydrophobicity properties. It can then easily be burned to decompose the oil and be used again as water depollution solution.

The main issue with polymer infiltration in a really high porous and non-reactive materials is the non-affinity of the polymer to make interaction with the 3D network and fill all the porosity. Also, the fragility of this BN aerogel need to be take in consideration regarding the techniques to use for the infiltration. High pressure methods have to be excluded, we will privilege vacuum infiltration method assisted by BN functionalisation.

X. Zeng and al.⁸² realized a 3D boron nitride nanosheets networks from Ice-templated assembly strategy. They used a combination of liquid exfoliation and non-covalent functionalisation methods to be able to disperse the Boron Nitride naosheets (BNNs) in deionized water. They used a simple technique for the infiltration of the epoxy polymer inside their 3D BNNs structure, they just completely immerged the samples inside the epoxy resin for 2hr following by a vacuum step at 70°C. Few factors could improve the infiltration of the epoxy by immersion. Firstly, the viscosity of the epoxy need to be lowest as possible (could increase the temperature to increase the viscosity but not too high because then the epoxy will start to cure, as it’s a exothermic reaction, and become more and more viscous), secondly, the functionalisation of the BNNs need to be investigated to be the most suitable function added to the BNNs to increase the affinity with the epoxy polymer. They reported an enhancement of the thermal conductivity of the final composite (with only 9.29vol% of BNNs) up to 3.5 W.m^-1.K^-1.

One solution that I suggest would be to use the high-pressure infiltration method on this structure will be first to use the immersion technique it will filled the majority of the porosity, let the polymer cured, then the composites will have sufficient strength to resist to the high-pressure infiltration method which will allowed to fill the remain porosity with the same polymer to improve the thermal and mechanical properties.

The general lack of scientific review on polymer infiltration method in porous ceramic structure is unusual. This means that research need to be done in this area. In addition, with the absence of mechanical characterisation of 3D porous filled polymer composites. The functionalisation of h-BN materials, the infiltration of polymer inside an ultra-porous 3D network and the mechanical characterisation of these composites seems that it would be one of the most challenging experimental part of my project but at least the most interesting.

4.    Conclusions

Hexagonal Boron Nitride is an outstanding material with a great future in the field of composite materials exhibiting extraordinary thermal, mechanical and dielectrics properties. Different theoretical models can predict the behaviour of a complex composite materials involving single or hybrid fillers inside the polymer matrices regarding their thermal or dielectrics properties. Some Thermal models fit well at low fillers concentration like Maxwell-Eucken under 25 vol% or Bruggeman model above 20 vol%. The same behavior is observed following the Maxwell Garnett or Bruggeman model for the dielectric constant prediction. This allowed us to predict the maximum or minimum values of the processed composites or to understand, try to identify the reasons when the values are out of the expected range. The behaviour of the fillers inside the matrix is complex and governed by internal interactions which are dependent on different factors such as the size, shape and end chemical group.

Current research in h-BN composites materials has showed huge improvement in thermal conductivity, up to +2000%, mechanical strength, up to +120% while maintaining a low dielectric constant (2.5-5) and loss tangent (10^-4). However, this research also underlined some limits like the ability to have a synergy between the fillers and the polymer matrix to allow them to build an efficient network for the conduction of the phonon (high thermal conductivity), the reduction of the presence of failure growth location in the matrix (mechanical strength) and a less interfering structure to the electromagnetic field (radio frequency transparency). That’s why the development of novel h-BN aerogel structure composites and their functionalisation need to be intensified because this method is the way to produce a 3D network inside the polymer matrix with the combination of a hybrid structure with another remarkable nanomaterials such as Boron Nitride Nanotubes (BNNTs). The advances in Chemical Vapour Deposition techniques or carbon substitution reaction coupled with freeze casting method allow the synthesis of high porosity h-BN aerogel, but the next challenge will be to proceed to a full polymer infiltration inside this nano-porous structure. The potential of these materials is expected to be on another range a magnitude compared to the current h-BN composite materials and would lead to the next generation of materials for a variety of applications in the fields of energy and telecommunications.

References

1. Lipp, A., Schwetz, K. A. & Hunold, K. Hexagonal boron nitride: Fabrication, properties and applications. J. Eur. Ceram. Soc. 5, 3–9 (1989).

2. Watanabe, S., Miyake, S. & Murakawa, M. Tribological properties of cubic, amorphous and hexagonal boron nitride films. Surf. Coatings Technol. 49, 406–410 (1991).

3. Ooi, N., Rairkar, A., Lindsley, L. & Adams, J. B. Electronic structure and bonding in hexagonal boron nitride. J. Phys. Condens. Matter 18, 97–115 (2006).

4. Steinborn, C. et al. Correlation between microstructure and electrical resistivity of hexagonal boron nitride ceramics. J. Eur. Ceram. Soc. 33, 1225–1235 (2013).

5. Muramatsu, Y., Kaneyoshi, T., Gullikson, E. M. & Perera, R. C. C. Angle-resolved soft X-ray emission and absorption spectroscopy of hexagonal boron nitride. Spectrochim. Acta – Part A Mol. Biomol. Spectrosc. 59, 1951–1957 (2003).

6. Eichler, J. & Lesniak, C. Boron nitride (BN) and BN composites for high-temperature applications. J. Eur. Ceram. Soc. 28, 1105–1109 (2008).

7. Search, H., Journals, C., Contact, A., Iopscience, M. & Address, I. P. Graphite Interplanar Bonding : Electronic Delocalization. Europhys. Lett. 403, (1994).

8. Paszkowicz, W., Pelka, J. B., Knapp, M., Szyszko, T. & Podsiadlo, S. Lattice parameters and anisotropic thermal expansion of hexagonal boron nitride in the 10-297.5 K temperature range. Appl. Phys. A Mater. Sci. Process. 75, 431–435 (2002).

9. Duan, X. et al. Review on the properties of hexagonal boron nitride matrix composite ceramics. J. Eur. Ceram. Soc. 36, 3725–3737 (2016).

10. FENG, V. A review on Hexagonal Boron Nitride Filled Composites, Route Towards 3D Printable Multifunctional. Review (2017).

11. Arnaud, B., Lebègue, S., Rabiller, P. & Alouani, M. Huge excitonic effects in layered hexagonal boron nitride. Phys. Rev. Lett. 96, 1–4 (2006).

12. Green, J. F., Bolland, T. K. & Bolland, J. W. Theoretical elastic behavior for hexagonal boron nitride. J. Chem. Phys. 64, 656–662 (1976).

13. Sichel, E. K., Miller, R. E., Abrahams, M. S. & Buiocchi, C. J. Heat capacity and thermal conductivity of hexagonal pyrolytic boron nitride. Phys. Rev. B 13, 4607–4611 (1976).

14. Lienhard IV, J. H. & Lienhard V, J. H. A HEAT TRANSFER TEXTBOOK, fourth edition. Phlogist. Press 766 (2012). doi:978-04864793161

15. Agari, Y. & Uno, T. Estimation on thermal conductivities of filled polymers. J. Appl. Polym. Sci. 32, 5705–5712 (1986).

16. Kumlutas, D. & Tavman, I. H. A numerical and experimental study on thermal conductivity of particle filled polymer composites. J. Thermoplast. Compos. Mater. 19, 441–455 (2006).

17. Lin, F., Bhatia, G. S. & Ford, J. D. Thermal conductivities of powder‐filled epoxy resins. J. Appl. Polym. Sci. 49, 1901–1908 (1993).

18. Bruggeman, D. A. G. Berechnung verschiedener physikalischer Konstanten von heterogenen Substanzen. III. Die elastischen Konstanten der quasiisotropen Mischk??rper aus isotropen Substanzen. Ann. Phys. 421, 160–178 (1937).

19. Pietrak, K. & Winiewski, T. S. A review of models for effective thermal conductivity of composite materials. Open Access J. J. Power Technol. 95, 14–24 (2015).

20. Agari, Y., Ueda, A. & Nagai, S. Thermal conductivities of composites in several types of dispersion systems. J. Appl. Polym. Sci. 42, 1665–1669 (1991).

21. Agari, Y., Ueda, A. & Nagai, S. Thermal conductivity of a polyethylene filled with disoriented short‐cut carbon fibers. J. Appl. Polym. Sci. 43, 1117–1124 (1991).

22. Lorentz, H. A. The theory of electrons and its applications to the phenomena of light and radiant heat. (1916).

23. Lorenz, L. Ueber die Refractionsconstante. Ann. Phys. 247, 70–103 (1880).

24. J. C. Maxwell Garnett. Colours in Metal glasses and in Metallic Films. Philos. Trans. R. Soc. London 204, 121–137 (1904).

25. Niklasson, G. A., Granqvist, C. G. & Hunderi, O. Effective medium models for the optical properties of inhomogeneous materials. Appl. Opt. 20, 26 (1981).

26. Hogg, S. W. et al. Dielectric constants for mixed media. J. Phys. D. Appl. Phys. (1977).

27. Gu, J., Zhang, Q., Dang, J. & Xie, C. Thermal conductivity epoxy resin composites filled with boron nitride. Polym. Adv. Technol. 23, 1025–1028 (2012).

28. Zhou, W., Qi, S., Li, H. & Shao, S. Study on insulating thermal conductive BN/HDPE composites. Thermochim. Acta 452, 36–42 (2007).

29. Chen, L. et al. Thermal Conductivity Performance of Polypropylene Composites Filled with Polydopamine-Functionalized Hexagonal Boron Nitride. PLoS One 1–16 (2017). doi:10.1371/journal.pone.0170523

30. Sim, L. C., Ramanan, S. R., Ismail, H., Seetharamu, K. N. & Goh, T. J. Thermal characterization of Al2O3and ZnO reinforced silicone rubber as thermal pads for heat dissipation purposes. Thermochim. Acta 430, 155–165 (2005).

31. Golberg, B. D., Bando, Y., Tang, C. & Zhi, C. Boron Nitride Nanotubes **. Adv. Mater. 2413–2432 (2007). doi:10.1002/adma.200700179

32. Verma, V., Jindal, V. K. & Dharamvir, K. Elastic moduli of a boron nitride nanotube. Nanotechnology 435711, (2007).

33. P, A. et al. Elastic modulus and resonance behavior of boron nitride nanotubes. Appl. Phys. Lett. 2527, 1–4 (2012).

34. Wei, B. X., Wang, M., Bando, Y. & Golberg, D. Tensile Tests on Individual Multi-Walled Boron Nitride Nanotubes. Adv. Mater. 0044, 4895–4899 (2010).

35. Bujard, P. Thermal Conductivity of Boron Nitride Filled Epoxy Resin: Temperature Dependence and Influence of Sample Preparation. Intersoc. Conf. Therm. Phenom. Fabr. Oper. Electron. Components I-THERM ’88 41–49 (1988). doi:10.1109/ITHERM.1988.28676

36. Xia, C. et al. Hybrid boron nitride-natural fiber composites for enhanced thermal conductivity. Sci. Rep. 6, 1–8 (2016).

37. Hong, J. P. et al. High thermal conductivity epoxy composites with bimodal distribution of aluminum nitride and boron nitride fillers. Thermochim. Acta 537, 70–75 (2012).

38. Kim, K., Kim, M. & Kim, J. Thermal and mechanical properties of epoxy composites with a binary particle filler system consisting of aggregated and whisker type boron nitride particles. Compos. Sci. Technol. 103, 72–77 (2014).

39. K. C. Yung, H. L. Enhanced Thermal Conductivity of Boron Nitride Epoxy-Matrix Composite Through Multi-Modal Particle Size Mixing. J. Appl. Polym. Sci. 106, (2007).

40. Kim, K., Kim, M., Hwang, Y. & Kim, J. Chemically modified boron nitride-epoxy terminated dimethylsiloxane composite for improving the thermal conductivity. Ceram. Int. 40, 2047–2056 (2014).

41. Lin, Z., Liu, Y., Moon, K. S. & Wong, C. P. Enhanced thermal transport of hexagonal boron nitride filled polymer composite by magnetic field-assisted alignment. Proc. – Electron. Components Technol. Conf. 1692–1696 (2013). doi:10.1109/ECTC.2013.6575801

42. Yu, J. et al. Interfacial modification of boron nitride nanoplatelets for epoxy composites with improved thermal properties. Polymer (Guildf). 53, 471–480 (2012).

43. Zhou, W., Qi, S., An, Q., Zhao, H. & Liu, N. Thermal conductivity of boron nitride reinforced polyethylene composites. Mater. Res. Bull. 42, 1863–1873 (2007).

44. Lee, G. W., Park, M., Kim, J., Lee, J. I. & Yoon, H. G. Enhanced thermal conductivity of polymer composites filled with hybrid filler. Compos. Part A Appl. Sci. Manuf. 37, 727–734 (2006).

45. Cheewawuttipong, W., Tanoue, S., Uematsu, H. & Iemoto, Y. Thermal Conductivity of Polypropylene Composites With Hybrid Fillers of Boron Nitride and Vapor-Grown Carbon Fiber. (2016). doi:10.1002/pc

46. Cheewawuttipong, W., Fuoka, D., Tanoue, S., Uematsu, H. & Iemoto, Y. Thermal and Mechanical Properties of Polypropylene / Boron Nitride Composites. Energy Procedia 34, 808–817 (2013).

47. Anjana, P. S. et al. Dielectric , Thermal , and Mechanical Properties of CeO 2 -Filled HDPE Composites for Microwave Substrate Applications. J. Polym. Sci. Part B Polym. Plast. 48, 998–1008 (2010).

48. Krupa, I., Novák, I. & Chodák, I. Electrically and thermally conductive polyethylene / graphite composites and their mechanical properties. Synth. Met. 145, 245–252 (2004).

49. Gu, J., Liang, C., Dang, J., Dong, W. & Zhang, Q. Ideal dielectric thermally conductive bismaleimide nanocomposites filled with polyhedral oligomeric silsesquioxane functionalized nanosized boron nitride. RSC Adv. 6, 35809–35814 (2016).

50. Zhi, B. C., Bando, Y., Tang, C., Kuwahara, H. & Golberg, D. Large-Scale Fabrication of Boron Nitride Nanosheets and Their Utilization in Polymeric Composites with Improved Thermal and Mechanical Properties. Adv. Mater. 2889–2893 (2009). doi:10.1002/adma.200900323

51. Wang, X. et al. “ Chemical Blowing ” of Thin-Walled Bubbles : High-Throughput Fabrication of Large-Area , Few-Layered BN and C x -BN Nanosheets. Adv. Mater. 4072–4076 (2011). doi:10.1002/adma.201101788

52. Tajaddod, N., Song, K., Green, E. C., Zhang, Y. & Minus, M. L. Exfoliation of Boron Nitride Platelets by Enhanced Interfacial Interaction with Polyethylene. Macromol. J. 315–327 (2016).

53. Wang, X.-B. et al. Biomass-Directed Synthesis of 20 g High-Quality Boron Nitride Nanosheets for Thermoconductive. ACS Nano 9081–9088 (2014).

54. Zhu, H. et al. Highly thermally conductive papers with percolative layered boron nitride nanosheets. ACS Nano 3606–3613 (2014).

55. Coleman, J. N. et al. Two-Dimensional Nanosheets Produced by Liquid Exfoliation of Layered Materials. Science (80-. ). 331, (2011).

56. Zhi, C. Y. et al. Chemically Activated Boron Nitride Nanotubes. Chem. Asian J. 1536–1540 (2009). doi:10.1002/asia.200900158

57. Sainsbury, T. et al. Oxygen Radical Functionalization of Boron Nitride Nanosheets. JACS (2012).

58. Lee, D. et al. Scalable Exfoliation Process for Highly Soluble Boron Nitride Nanoplatelets by Hydroxide-Assisted Ball Milling. ACS (2015). doi:10.1021/nl504397h

59. Xiao, F. et al. Edge-Hydroxylated Boron Nitride Nanosheets as an Effective Additive to Improve the Thermal Response of Hydrogels. Adv. Mater. 7196–7203 (2015). doi:10.1002/adma.201502803

60. Huang, X., Zhi, C., Jiang, P., Golberg, D. & Bando, Y. Polyhedral Oligosilsesquioxane-Modifi ed Boron Nitride Nanotube Based Epoxy Nanocomposites : An Ideal Dielectric Material with High Thermal Conductivity. Adv. Funct. Mater. 1824–1831 (2013). doi:10.1002/adfm.201201824

61. Zhi, C. et al. Boron nitride nanotubes / polystyrene composites. Cambridge J. (2006). doi:10.1557/JMR.2006.0340

62. Zhi, B. C. et al. Towards Thermoconductive , Electrically Insulating Polymeric Composites with Boron Nitride Nanotubes as Fillers. Adv. Funct. Mater. 1857–1862 (2009). doi:10.1002/adfm.200801435

63. Weng Qunhong, Wang Xuebin , Wang Xi, Bando Yoshio, G. D. Functionalized hexagonal boron nitride nanomaterials : emerging properties and applications. Chem. Soc. Rev. 45, 3989–4012 (2016).

64. KISTLER, S. S. Coherent expanded aerogels. Nature 318, (1932).

65. Elaloui, E., Achard, P., Chevalier, J. & Durant, M. Physical properties of silica gels and aerogels prepared with new polymeric precursors OiR. J. Non. Cryst. Solids 186, 1–8 (1995).

66. Zou, J. et al. Ultralight Multiwalled Carbon Nanotube Aerogel. ACS Nano 4, 7293–7302 (2010).

67. Pei, S. & Cheng, H. The reduction of graphene oxide. Carbon N. Y. 50, 3210–3228 (2011).

68. Becerril, H. A. et al. Evaluation of Solution-Processed Reduced Graphene Oxide Films as Transparent Conductors. ACS Nano 2, 463–470 (2008).

69. Wang, X., Zhi, L. & Mu, K. Transparent , Conductive Graphene Electrodes for Dye-Sensitized Solar Cells. nano Lett. (2008).

70. Wu, Z. et al. Synthesis of high-quality graphene with a pre-determined number of layers. Carbon N. Y. 47, 493–499 (2008).

71. Li, X., Wang, H., Robinson, J. T., Sanchez, H. & Diankov, G. Simultaneous Nitrogen Doping and Reduction of Graphene Oxide. JACS 15939–15944 (2009).

72. Wu, Z. et al. Synthesis of Graphene Sheets with High Electrical Conductivity and Good Thermal Stability by Hydrogen Arc Discharge Exfoliation. ACS Nano 3, 411–417 (2009).

73. Kim, K. S., Kingston, C. T., Plunkett, M. & Simard, B. High-yield growth of small-diameter boron nitride nanotubes in radio-frequency induction thermal plasma : role of hydrogen and quenching. (2015).

74. Han, W. et al. Synthesis of boron nitride nanotubes from carbon nanotubes by a substitution reaction. Appl. Phys. Lett. 3085, 21–24 (1998).

75. Golberg, D. et al. Fine structure of boron nitride nanotubes produced from carbon nanotubes by a substitution reaction. Appl. Phys. Lett. 2364, 0–3 (1999).

76. Song, Y. et al. Ultralight boron nitride aerogels via template-assisted chemical vapor deposition. Sci. Rep. 5, 1–9 (2015).

77. Sun, H., Xu, Z. & Gao, C. Multifunctional , Ultra-Flyweight , Synergistically Assembled Carbon Aerogels. Adv. Mater. 2554–2560 (2013). doi:10.1002/adma.201204576

78. Sutter, P., Lahiri, J., Albrecht, P. & Sutter, E. Chemical Vapor Deposition and Etching of High-Quality Monolayer Hexagonal Boron Nitride Films. ACS Nano 7303–7309 (2011).

79. Guo, N. et al. Controllable growth of triangular hexagonal boron nitride domains on copper foils by an improved low-pressure chemical vapor deposition method. Nanotechnology 23, (2012).

80. Shi, Y. et al. Synthesis of few-layer hexagonal boron nitride thin film by chemical vapor deposition. Nano Lett. 10, 4134–4139 (2010).

81. Lei, W., Portehault, D., Liu, D., Qin, S. & Chen, Y. Porous boron nitride nanosheets for effective water cleaning. Nat. Commun. 4, 1777 (2013).

82. Zeng, X. et al. Ice-Templated Assembly Strategy to Construct 3D Boron Nitride Nanosheet Networks in Polymer Composites for Thermal Conductivity Improvement. Small 11, 6205–6213 (2015).