Over the past decade, comprehensive investigations have been conducted to develop graphene based two-dimensional (2D) materials to harness their excellent and unprecedented properties such as high electrical conductivity, optical transparency, mechanical strength, and flexibility. To fully utilize these functionalities, a number of methods including chemical vapor deposition, liquid shear exfoliation, sonication, ball milling, electrochemical exfoliation and the recently conceived compressible flow exfoliation (CFE) have been successfully studied. However, these methods can introduce significant amount of defects during exfoliation into the graphene crystal structures that have a strong influence on their properties. This study is designed to compare the defect and flake quality between CFE and bath or probe sonication process for producing high-performance conductive ink from exfoliated graphene.

In our CFE process, graphite is rapidly jettisoned through a small orifice using high-pressure gases without the need for any time-based treatment, unlike other shear-based liquid processes. Shear-induced exfoliation occurs due to the high velocities that expanding and accelerating gases can achieve in small orifices coupled with viscous friction effects resulting in a high shear rate (γ>105 s-1) experienced by the graphite particles. In contrast, in the sonication methods, an ultrasonic transducer is used to induce unstable cavitation bubbles in a liquid medium, which upon their inevitable collapse emanate a shock wave. The energy of this shockwave is sufficient to fragment nearby bulk graphite powders into smaller lengths as well as thickness along the weak, secondary c-axis. But when the bulk particle is fragmented into smaller flakes, a good number of edge and basal plane defects are introduced into the flakes. The defect population increases as the time for sonication rises. The occurrence of a disorder-activated D peak at 1330 cm-1 in the Raman spectra of graphene is indicative of defects, in particular, those which disrupt the sp2 hybridization. Such defects can be interpreted to be the creation of new edges, vacancies or substitutions, with the ratio between the peaks intensities of the D to G peak (ID/IG) providing a qualitative indication of their population. In our study, we found the ratio of D peak to G peak is significantly less in CFE than that of bath sonicated graphene-ID/IG=0.66 for CFE graphene and ID/IG=1.1 for bath sonicated graphene. In contrast, the bulk graphite powder had ID/IG ratio of 0.62. The increased quantity of defects in bath sonication may be attributed to the prolonged sonication time which is well known to be responsible for reducing the flake length and hence, introducing more edges. The flake quality of exfoliated graphene was also verified using atomic force microscope (AFM) and transmission electron microscope (TEM).