Polymer composites exhibit many properties not observed in miscible polymer blends, such as anisotropic stiffness. In composites, one thermoplastic polymer can exist in another in the form of spheres, ellipsoids, fibrils, and lamella as a dispersed phase.¹ This arrangement is affected by viscoelasticity, blend ratio, and processing conditions. Nanofiber formation in fibrillar blends (those which possess a fibrillar dispersed phase in the immiscible blend) is relevant to producing polymer composites with improved mechanical properties,² but is not understood for all industrially relevant composites.

One such composite consists of polybutylene terephthalate (PBT) nanofibers in cellulose acetate butyrate (CAB) matrices. Industrially relevant nanoscale fiber assembly will require high-throughput production protocols.³ The final shape and size of the dispersed phase depends on processing parameters and polymer properties,⁴ but a relevant systematic study has not been reported. We have investigated PBT nanofiber formation in CAB matrices, with an emphasis on measuring polymer physical properties and nanofiber morphology development.

We formed PBT nanofibers in a twin-screw extruder. This is a process of continuous deformation, elongation, orientation, and coalescence of PBT micelles to nanofibers in the CAB matrix. Plate and He have indicated that good fibrillation is achieved when the viscosity ratio is within the 0.1<η_d/η_m<10 range,^{5, 6} where η_d and η_m are the dispersed phase viscosity and matrix viscosity, respectively. We fixed CAB with three types of PBT in our study. At shear rates of 30, 50, and 80s⁻¹ (see Figure 1), the viscosity ratio was in the 1<η_d/η_m<10 range, and the fibrillar dispersed phase remained after removing the matrix phase.

Figure 2(a) and (b) presents the initial morphological development of three types of PBT fibers. The sheeting mechanism⁷ may explain the quick morphological changes in the initial stage of blending. This mechanism leads to sheet or ribbon formation when a large piece of dispersed phase is dragged across a hot surface. Due to shear flow and interfacial tension, the sheets were unstable and began to break up when holes formed in them. When the holes attained sufficient size and concentration, they coalesced and formed network structures.

Figure 2(c) shows the metaphase development of PBT dispersed phases. The average diameters of PBT1^#, PBT2^#, and PBT3^# fibers were 160, 165, and 176nm, respectively (see Figure 3). Massive fibers formed by network structure breakup at SP−2^#, and an enormous reduction in phase size occurred from SP−1^# to SP−2^#.

Figure 2(d) displays the later morphological development of the PBT dispersed phase. The average diameters of three PBT nanofibers were 142, 139, and 125nm, respectively (see Figure 3). Smaller-diameter fibers formed in the later development. The diameter distribution decline from SP−2^# to SP−3^# indicates that the later morphological development leads to further fiber-diameter reduction. We schematically outline the entire nanofiber formation process in Figure 4.

In summary, we prepared well-defined nanofibers from PBT/CAB immiscible blends. We found that the holistic developmental trends of the PBT dispersed phase were nearly the same. Fibers began to form even under the shear flow of the twin-screw extruder. The dispersed-phase morphology-development mechanism involved the formation of sheets, holes, and network structures, followed by size reduction and nanofiber formation. Viscosity ratio, blend ratio, and shear rate affected nanofiber diameter distribution. In the future, we will produce more types of thermoplastic nanofibers and use them in practical applications, such as filtration biomaterials.