A novel catalyst-free methodology has been developed to prepare few-layer hexagonal boron nitride nanosheets using a bottom-up process. Scanning electron microscopy and transmission electron microscopy (both high and low resolution) exhibit evidence of less than ten layers of nanosheets with uniform dimension. X-ray diffraction pattern and other additional characterization techniques prove crystallinity and purity of the product.

1. Introduction

Due to structural similarities to graphene, hexagonal boron nitride (h-BN) nanosheets have become an extremely desirable material over the last few years. While many synthetic efforts have been made to combine these two materials for several electronic applications [15], recently few-layer h-BN nanosheets have been successfully developed as substrates for graphene [68]. Furthermore, h-BN nanosheets are useful in a number of versatile applications due to unique inherent physical properties [911]. h-BN exhibits excellent chemical and mechanical stability and it is thermally conductive and is electrically insulating material with a wide band gap (5-6 eV) as well. While h-BN is being used as an electrical insulator in thermally conductive materials [12], composites of h-BN have been the preferred species in aircrafts for their radiation shielding properties [13].

Although the developments of graphene and graphene composites are continuously advancing, few-layer h-BN nanostructures are comparatively less investigated due to synthetic difficulties [14]. Preparation of h-BN nanostructures involves one of the two common approaches, top-down or bottom-up. The top-down process mainly includes mechanical or chemical exfoliation of h-BN nanosheets from bulk h-BN [1517]. Although it is one of the most common techniques currently used to produce nanomaterials on a large scale, the major disadvantage of this method is the imperfection of surface structure imparted during the process [18]. On the contrary, the bottom-up approach yields nanostructures with minimal defects and superior chemical homogeneity. Typically, chemical vapor deposition (CVD) is used as a bottom-up approach. However, the requirement of substrate and extreme reaction conditions makes this synthetic route less desirable. Involvement of catalysts in a CVD process not only restricts industrial scaling-up of the method, but also yields products with metal impurities [6, 19]. It is interesting to note a recent report in which defect-free h-BN nanocrystals were produced at 1000°C in a tube furnace [20]. Consequently, a controlled cost-effective synthetic methodology to prepare h-BN nanosheets on a large scale is warranted. To the best of our knowledge, there have been no reports to date on a bottom-up methodology that avoids catalysts altogether in producing pristine few-layer h-BN nanosheets at a significantly lower temperature.

Herein we report a bottom-up synthesis for few-layer h-BN nanosheets by an autoclave pyrolysis technique that can potentially be scaled up for industrial manufacturing purposes.

2. Materials and Methods

All the chemicals used were from Sigma-Aldrich, KBH4 (99.998%), NH4Cl (99.998%), and NaN3 (99.5%). The HCl used for purification was from Fisher Scientific. In a typical pyrolysis experiment, the reagents were assembled in an Ar glovebox containing <2.00 ppm O2. The B and N precursors, KBH4 (82.31 mmol), NH4Cl (82.45 mmol), and NaN3 (82.60 mmol), were thoroughly mixed together and transferred to a 100 mL autoclave that was sealed and then heated with varying times (12 h for BNS12, 24 h for BNS24, and 48 h for BNS48) at 800°C in a vertical furnace. After the reaction was completed, the crude product from the autoclave was washed three times with 3 M HCl with sonication between each wash. The next step was to wash with deionized water, until the pH of the decant liquid became neutral. Sonication was carried out between each water wash. The drying process included washing few times with acetone followed by evaporation of the solvent using a rotary evaporator. Finally, the purified product was dried under high vacuum at room temperature overnight.

3. Results and Discussion

The synthetic methodology involved mixing of NH4Cl, NaN3, and KBH4 in equimolar proportions, in an inert atmosphere. The resulting mixture was heated at 800°C, inside a tightly closed stainless steel autoclave to produce h-BN nanosheets (Scheme 1). In order to optimize the reaction conditions and to analyze the time dependency on the morphology of the product, the pyrolysis was carried out at varied time interval of 12 to 48 h. Detailed reaction protocols are included in the supplementary information.

This innovative technique produced few-layer h-BN nanosheets with high yield in a stainless steel autoclave. However, the exact mechanism of this process is unknown. We believe the high pressure generated during the pyrolysis facilitated the formation of few-layer nanosheets of h-BN. The advantages of this methodology over the CVD methods are manifold. While almost all of the CVD techniques require temperatures higher than 1000°C, a comparatively lower temperature, used in our synthesis, is noteworthy. In addition, the catalyst-free synthetic approach involving inexpensive starting material without the continuous stream of gas made this process unique, cost-effective, and ideal for the large scale production of this material. Nonetheless, the method is simple enough in that the rigorous purification techniques can be avoided and the products can be washed with acid and deionized water to produce pristine nanosheets. The resulting h-BN nanosheets, identified as BNS12, BNS24, and BNS48, were isolated as products from 12 h, 24 h, and 48 h reactions, respectively. The yield of the products, though in slight deviation (less than 4%), was calculated for each reaction (Table 1).

Role of the precursors in the synthesis of h-BN was also investigated systematically. The absence of NaN3 in the equimolar mixture of KBH4 and NH4Cl reduced the yield of h-BN to <40% when reacted for 48 h at 800°C. On the other hand, a complete elimination of NH4Cl from the reactant mixture failed to produce even a trace of h-BN under identical reaction conditions indicating that all three reagents are essential components in producing h-BN nanosheets. Although the exact mechanism for the formation of h-BN in this study is not known, a plausible sequence of reaction steps, shown in Scheme 1, can be deduced based on the formation of hydrogen and nitrogen gases along with the formation KCl and water-sensitive NaNH2 that were removed during the washings using 3 M HCl and deionized water. The in situ formation of ammonia borane (NH3·BH3) in the first step can undergo a fast reaction with NaN3 to produce the desired h-BN with the formation of solid NaNH2 along with the formation of gaseous H2 and N2 as by-products. Since NaNH2 is known to form an infinite polymeric tetrahedral chain, it is possible that the initial formation of the nanosheets of h-BN could occur on the surface of the NaNH2 polymer.

The products were vacuum dried overnight for further characterization. Scanning electron microscopy (SEM), transmission and high resolution transmission electron microscopy (TEM and HRTEM), energy-dispersive X-ray spectroscopy (EDX), electron energy loss spectroscopy (EELS), Fourier transform infrared (FT-IR) spectroscopy, Raman spectroscopy, X-ray powder diffraction (XRD), and UV-Vis spectroscopy are the techniques used for characterizing the product.

The morphology of the BNS samples was investigated using SEM (TeScan Vega II SBH). The products were lightly coated with gold for the ease of imaging purposes. The SEM image of BNS12 exhibits uniform product dimensions (Figure 1) and the average diameter of the nanosheets was found to be approximately between 600 and 800 nm. A narrow size distribution of the products was observed for all of the samples at different time periods. Thus, we can conclude that the size of the product is independent of the reaction times.

TEM (Hitachi H-600) characterization is consistent with the findings from SEM imaging for all of the samples with product dimension falling under similar data ranges (Figures 2(a)2(c)). On the other hand, the HRTEM (JEOL JEM-2100F) determined the number of layers of the h-BN nanosheets. While BNS12 exhibited nanosheets with an average of 6–8 layers (Figure 3(a)), BNS48 showed varied number of layers between 20 and 22 (Figure 3(b)). The spacing between the layers was estimated by measuring the thickness of the sheet and dividing it by the number of layers. The resulting value of 0.33 nm matches perfectly the previously published data [21, 22] as well as those collected via XRD studies. Nonetheless, the HRTEM images evidently predicted that the reaction time has a pronounced effect on the formation of h-BN layers. A controlled reaction time can reduce the number of layers without significantly changing the reaction yields.

The EDX spectra of BNS samples were generated over a large area of the products using an INCAx-act Analytical Standard EDS Detector and Figures 4(a)4(c) depict the EDX spectra of BNS12, BNS24, and BNS48, respectively. The EDX data ascertains the elemental composition and purity of h-BN nanosheets. The atomic ratio of B : N was found to be 1 : 1 (atomic weight percentage of B : N = 50.06 : 49.94) by EDX analysis. EELS data of BNS12 was obtained via an electron energy loss detector attached to the JEOL JEM-2100F microscope and Figure 5 represents the EEL spectrum. The peak positions and the splitting patterns in EEL spectrum confirm the sp2 hybridization of BN in h-BN nanosheets. The energy loss around 190 eV and 400 eV corresponded to the K-shell ionization energy of the B and N atoms, respectively [23].

Bonding patterns of the B and N atoms in nanosheets and purity of the products were further verified by FT-IR (ATI Mattson Genesis Series (Figures 6(a)6(c))) and Raman spectroscopy (Figures 7(a)7(c)). The FT-IR peak at 1388 cm−1 is due to the stretching vibration of B-N bonds. A comparatively weaker peak at 815 cm−1 can be attributed to the bending vibration of B-N-B bonds of h-BN [24]. Raman spectroscopy of the BNS samples was performed using a Renishaw inVia Raman microscope with a 532 nm laser source. An intense sharp peak at 1364 cm−1 acts as the definite confirmation of the presence of h-BN in the sample [25].

The UV-visible spectroscopy (Lambda XLS+) was used to determine the UV absorption properties and the corresponding band gaps of the BNS samples (Figures 8(a)8(c)). Very dilute and well dispersed ethanolic solutions of BNS samples were scanned between 200 and 700 nm in a quartz cell. A sharp absorption band at 207 nm (Figure 8(a)) for BNS12 reflects a band gap of 5.98 eV which is consistent with the value reported by Geick et al. for a single crystal of pure h-BN [24] A slight blue shift was noticed for BNS24 and BNS48 (Figures 8(b) and 8(c)) compared to BNS12 that can be attributed to the increasing number of layers in the nanosheets.

The XRD (Rigaku MiniFlex, Cu, 30 kV, 15 mA X-ray) patterns of the h-BN nanosheets, synthesized at 800°C, exhibited interplanar -spacing and intensities that are indicative of h-BN crystallinity [26]. Figures 9(a)9(c) show the spectrum of BNS12, BNS24, and BNS48, respectively, where the indexed peaks are in good agreement with the theoretical values for h-BN (JCPDS 34-0421). The lattice constants of and were calculated using EdPCR component of FullProf Suite software. Nonetheless, all of the XRD patterns suggest formation of highly ordered and pure h-BN nanosheets.

4. Conclusions

In conclusion, we have developed a novel methodology to prepare few-layer h-BN nanosheets in high yields with acceptable purity. A 12 h reaction involving inexpensive starting reagents at moderately low temperature produced nanosheets of uniform dimension and few layers. While the morphology and crystallinity of the product were thoroughly characterized, elemental composition of the products was also determined. A detailed investigation on the variation of reaction temperatures and times along with the application of few-layer h-BN nanosheets for polymer composites is currently being investigated in our laboratories.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.


The authors gratefully acknowledge the financial support from NSF (CHE-0906179). HRTEM and EELS data were obtained from the NUANCE Facilities of Northwestern University and the authors would like to thank Dr. Shu-You Li for his help. Ms. Lori Bross is acknowledged for her timely help with low magnification TEM. Ms. Heather M. Barkholtz and Mr. Karel Waska helped with SEM imaging. Raman spectroscopy of the samples was performed by Mr. Jim Hill at Reinshaw Inc. They acknowledge the help of Dr. K. Vijayaraghavan in preparing the final draft of this paper.