Investigation of longitudinal and transverse dispersion in stable displacements with a high viscosity and density contrast between the fluids

Experimental and numerical investigations into the effect of heterogeneities on the recovery of heavy oil by VAPour EXtraction (VAPEX)

“…This difference is similar to that seen by Das and Butler as can be seen from fig:6 where we have plotted the observed and predicted oil drainage rates as a function of the square root of permeability [16]. This discrepancy has also been observed by other workers [1][2][3][4][5][6][7][8]12,13,[15][16][17][20][21][22][23][25][26][27][28]50,52] and suggests that there is some problem with the Butler-Mokrys equation.…”

“…This flow configuration with glycerol displacing ethanol was chosen in order to eliminate viscous fingering (see discussion in [3,49]). Transverse dispersion was not measured as previous work by Alkindi et al [5] has shown that its effect in these experiments is negligible.…”

“…Following Alkindi et al [5,6] we used glycerol to represent the oil and ethanol to represent the vapour. These analogue fluids were chosen so that:…”

“…This analysis assumes first contact miscibility between the oil and the vapour. This is clearly much simpler than performing a fine grid simulation to predict VAPEX performance but unfortunately seems to systematically underpredict the drainage rate seen in laboratory experiments [1][2][3][4][5][6][7][8]13,14,[16][17][18][20][21][22][23][24][25][26][27][28] In contrast, the only published field pilot (in Soda Lake, Saskatchewan) obtained oil production rates a factor of 10 lower than predicted. A further uncertainty in the field scale application of VAPEX is the impact of geological heterogeneity.…”

Effective mass diffusion and dispersion in random porous media

“…In the literature there are many expressions for estimations of longitudinal dispersion that have been developed from different studied systems. A brief summary of models can be found in the work of Lugo-M endez et al In a general sense, our numerical predictions are closer to those for Alkindy et al [34] and Lugo M endez et al, [35] mainly for Pe p <10 , which is attributed to similarities in the studied microstructure. For Pe p >10 our results are between results of Alkindy and Lugo-M endez and coworkers.…”