Here we present preliminary experiments for synthesis of carbide materials from bacterial cellulose that has been infiltrated with tungsten trioxide nanoparticles. Bacterial cellulose decomposes to carbon upon heat treatment around 900 °C. Upon further heat treatment in the presence of the metal oxide, this carbon reacts with the metal oxide to form a metal carbide. Bacterial cellulose is easily obtained using a microbial fermentation process¹. The product is a random network of nanometer-scaled cellulose fibers featuring high crystallinity. In contrast to cellulose obtained from plants, bacterial cellulose is pure cellulose. Hence, the carbides that could be obtained from this carbonized material are expected to exhibit superior properties than other bio-synthesized carbides. Furthermore, we aim at fabricating engineered cellulose scaffolds by manipulating bacterial movement in desired trajectories. The functionalization of these scaffolds would then lead to engineered carbides.

Bacterial cellulose (BC) is widely used as the base of the desert nata de coco, and is gaining recognition as a performance material in wound dressing. In this work we use commercially available nata de coco (purchased as Jubes Nata de Coco) as the bacterial cellulose and tungsten trioxide (WO₃) as the metal oxide nanoparticle, as we aim to synthesis tungsten carbide (WC). Cubes of nata de coco were washed repeatedly in ultra-pure water to eliminate the flavored syrup in which they were originally suspended. Once washed, the BC cubes were immersed in an alcohol-based WO₃dispersion and shook gently for 30 minutes. After 30 minutes, the cubes were taken out of the dispersion and directly introduced in a tube furnace. Heat treatment was performed with a heating rate of 5 °C/min and constant nitrogen gas flow. Based on previous heating protocols, a dwell of 30 minutes at 300 °C was first used to eliminate oxygen from the chamber, followed by a dwell of 3 hours at 1300 °C. XRD and SEM/EDS analysis were performed to characterize the carbonaceous materials obtained after the heat treatment.

Bacterial cellulose can hold up to 90% water of its total volume ². Hence significant amount of shrinkage occurs during the heat treatment. XRD characterization was performed after heat treatment and the results shown in Figure 1. Tungsten carbide (WC) has been formed along with its weaker form W₂C. A significant amount of metallic tungsten (W) is still present in the material. Heat treatment at higher temperature and longer dwell time may eliminate these W and W₂C to obtain only WC. From the SEM image in Figure 2a, it seems like carbide has been formed only on the surface and distribution of the carbide material is non-uniform on the surface. Figure 2b shows the SEM image of the tungsten carbide particles sitting on carbon background. The synthesized WC features particle size ranging from tens of nanometers up to 1.5 μm. The EDS results of the particles (Table 1) also confirm that the particles are of tungsten carbide. Our current hypotheses are that most of the WO₃nanoparticles could infiltrate only couple of microns inside from the surface the BC cube. Alternatively, the particles could have been pushed out from the matrix during water evaporation.

Ongoing work is on characterizing the extent of nanoparticle infiltration in the BC matrix before and after water evaporation during heating; using critical point drying to obtain an aerogel before heat treatment; and optimizing the heat treatment process to obtain entirely WC. 

References:

1. Z. Y. Wu, C. Li, H. W. Liang, J. F. Chen, and S. H. Yu, Angew. Chemie - Int. Ed., 52, 2925–2929 (2013).

2. a. Svensson et al., Biomaterials, 26, 419–431 (2005) http://linkinghub.elsevier.com/retrieve/pii/S014296120400198X. 

Table 1: Composition of the grains obtained from EDS of Figure 2b