1. P. Moschopoulos, Syrakos

Dynamics of viscoplastic filament stretching Journal Article

In: Journal of Non-Newtonian Fluid Mechanics, 284 (104371), 2020.

Abstract | Links | BibTeX | Tags: Filament stretching, Penalized Augmented Langrangian method, Pinching dynamics, Slender filament-2D comparison, viscoplastic materials

@article{Moschopoulos2020b,

title = {Dynamics of viscoplastic filament stretching},

author = {Moschopoulos, P., Syrakos, A., Dimakopoulos, Y., Tsamopoulos, J.},

doi = {10.1016/j.jnnfm.2020.104371},

year = {2020},

date = {2020-08-11},

journal = {Journal of Non-Newtonian Fluid Mechanics},

volume = {284},

number = {104371},

abstract = {We use numerical simulations to study the stretching and pinching of a viscoplastic material that is confined by two coaxial disks when they are pulled apart. The material initially forms a cylindrical bridge between the disks, follows the Heschel-Bulkley model and yields according to the von Mises criterion. We solve the governing equations numerically in their 2D, axisymmetric form using the PAL algorithm (Dimakopoulos et al. (2018)) and the slender filament approximation using the Bercovier and Engelman (1980) regularization. This approximation is valid when the disk radius to the initial length of the bridge, is small, but it is often used even when this aspect ratio is O(1). The results from the 2D solution agree with the experiments of Balmforth et al. (2010) for both a Carbopol gel and a Kaolin suspension, which can be classified as materials of medium and high viscoplasticity, respectively. Depending on the values of the aspect ratio and the yield stress of the material, we show that stretching can or cannot be accurately described by the slender filament approximation. By varying the yield stress and the pulling velocity, we find that discrepancies arise when the viscoplastic character of the flow increases, even though . The inaccuracy of the slender filament approximation stems from the omission of the shear component of the rate of strain tensor from the calculation of its magnitude in the von Mises criterion. This causes a significant overestimation of the effective viscosity, since this component can be comparable to or greater than the normal components outside the filament neck. Thus, larger unyielded regions are predicted by this approximation and the remaining yielded ones carry the burden of the deformation leading to much shorter pinching times. Our results for the pinching process of viscoplastic filaments show the minor role of the yield stress, in agreement with Balmforth et al. (2010). In particular, the process is almost the same as that of power-law fluids with the same value of the shear-thinning exponent n. Our simulations include cases with n < 0.54, where the slender filament approximation is known to fail, and also a case where inertia is accounted for. Still, we confirm that the behavior is similar to that of power-law fluids.},

keywords = {Filament stretching, Penalized Augmented Langrangian method, Pinching dynamics, Slender filament-2D comparison, viscoplastic materials},

pubstate = {published},

tppubtype = {article}

}

We use numerical simulations to study the stretching and pinching of a viscoplastic material that is confined by two coaxial disks when they are pulled apart. The material initially forms a cylindrical bridge between the disks, follows the Heschel-Bulkley model and yields according to the von Mises criterion. We solve the governing equations numerically in their 2D, axisymmetric form using the PAL algorithm (Dimakopoulos et al. (2018)) and the slender filament approximation using the Bercovier and Engelman (1980) regularization. This approximation is valid when the disk radius to the initial length of the bridge, is small, but it is often used even when this aspect ratio is O(1). The results from the 2D solution agree with the experiments of Balmforth et al. (2010) for both a Carbopol gel and a Kaolin suspension, which can be classified as materials of medium and high viscoplasticity, respectively. Depending on the values of the aspect ratio and the yield stress of the material, we show that stretching can or cannot be accurately described by the slender filament approximation. By varying the yield stress and the pulling velocity, we find that discrepancies arise when the viscoplastic character of the flow increases, even though . The inaccuracy of the slender filament approximation stems from the omission of the shear component of the rate of strain tensor from the calculation of its magnitude in the von Mises criterion. This causes a significant overestimation of the effective viscosity, since this component can be comparable to or greater than the normal components outside the filament neck. Thus, larger unyielded regions are predicted by this approximation and the remaining yielded ones carry the burden of the deformation leading to much shorter pinching times. Our results for the pinching process of viscoplastic filaments show the minor role of the yield stress, in agreement with Balmforth et al. (2010). In particular, the process is almost the same as that of power-law fluids with the same value of the shear-thinning exponent n. Our simulations include cases with n < 0.54, where the slender filament approximation is known to fail, and also a case where inertia is accounted for. Still, we confirm that the behavior is similar to that of power-law fluids.