Figures are not included in the pdf file.
High-amplitude pressure pulsing or mechanical excitation of a saturated porousmedium under a pressure gradient increases the flow rate of the liquid alongthe direction of the flow gradient. Experiments show that this occurs forsinglephase liquid systems and two-phase liquid systems (e.g. water- wet,paraffin oil mobile phase) under various conditions and system parameters. Thepresence of free gas in the system leads to a delay of the effect because theexcitation energy is dissipated in compressing the gas.
Experiments have been performed in a wide range of configurations (cylindricalcells and flat plate simulators), grain sizes (30 to 2000 microns), viscosities(I to -10,600 cP), and flow factors (e.g. with and without sand flow). Mobilephase flow rate increases of from 30% to over a thousand percent (in the caseof the most viscous oils) were measured. These beneficial flow rate effectstook place in sand packs without saturation changes, without fabric changes,and under conditions of constant external head They cannot be rationalizedwithin Darcy theory as this theory contains no inertial terms. A new theory hasbeen developed.
The flow enhancement effect requires large strains; seismic amplitude strainsare inadequate. The nature of the excitation is important: high-amplitudenon-seismic pulses dominated by low frequencies are best. This is most easilyachieved in the laboratory and in the field by pressure pulsing, as opposed toother excitation methods.
One of the most interesting series of experiments was related to physical(visual and quantitative) demonstrations that viscous fingering instabilitiescan be suppressed by pulsing applied to the less viscous invading phase. Theseresults have important implications in the execution of water floods in thefield and can lead to methods of converting old water floods tostabilized
BASIC FLOW ENHANCEMENT PRINCIPLES
Before our experimental activity is defined and results described, it is bestto explain why results do not fall within the "conventional" view of porousmedia mechanics. What is being done during system pulsing is not radical, butcurrently accepted poromechanics models cannot correctly account for suchdynamic effects. A better theory was developed well before the experiments, andit helped guide the testing program. This theory will now be qualitativelydescribed.
Scientists and engineers working in fluid flow have been taught that thequasi-static Darcy flow paradigm (q ? ?p/ ?l), where gradient is amacroscopically defined quantity ( ?p/ ?l) =(Pl-P2)/l), is asufficient theory for porous media flow over a wide range of conditions.Perhaps some inability to correctly predict flow rates or dispersion behaviorin clays, shales or fractured media is admitted, but otherwise Darcy theory isaccepted uncritically.
Similarly, geophysicists working with porous media wave mechanics have beentaught that Biot-Gassmann theory is sufficient to describe porous media wavepropagation, given a wavelength much greater than the particle size. Neither ofthese "fundamental" theories is complete, although each may be sufficient forpractical purposes under certain restrictive conditions.
Darcy theory is a quasi-static theory, and contains no inertial terms.
Bentonite organo-clay/anthracite mixture in the granular form (EC-1 00) wasused in filtration (column) studies in treating four representativeoil-in-water emulsions. The oilin- water emulsions used were as follows:standard mineral oil (SMO); Kutwell45 (KUTJ and Valcool (VAL), two cuttingoils; and refinery efluent (RE) from the Co-operative Oil Refinery, Regina,Saskatchewan The concentrations of oil in oily waters varied from 8.3 to 69.3mg/L. Eight-hour column studies were conducted in a 19 mm ID, 450 mm/l200 mmlong cast acrylic pipe with organo-clay/anthracite depth of 300 mm/l000 mm. TheSMO, KUT, and VAL oil-in-water emulsions were pumped into the column at fourflow rates of 3, 6, 9, and 12 mL/min (0.3, 0.5, 0.8, and I.0gpm/ft2, respectively). Column breakthrough studies were conductedin a 19 mm ID, 1200 mm long cast acrylic pipe using the organo-clay/anthracitemixture of IO00 mm depth. The study was conducted for SIUO, KUT, VAL and REoil-in-water emulsions with a flow rate of 12 mUmim (I gpm/ft2). Theeight-hour column tests with 300 mm bed depth and all oilin- water emulsionsindicated that generally, the oil removal efficiencies decreased with anincrease in flow rate. The percentage redaction in oil removal efficiency was29 and 37 for SMO, 51 and 59 for Km, and 9 and 57 for VAL when the flow ratewas increased from 3 mUmin to 6 and 9 mLJmin, respectively. The results of theeight-hour experiments with 1000 mm depth of organo-clay/anthracite bed andwith a flow rate of I2 mUmin showed that oil removal eficiency for SMO, KUT,and VAL varied between 65 and 70 percent. In the case of RE which is a treatedand highly stable emulsion, the oil removal efficient was found to be 99.5percent. The results from the breakthrough studies clearly indicated that theThomas' equation provides a reasonable fit of the data. The oil-sorptioncapacities (x/m) based on a mass balance analysis were found to be 0.0036,0.0019, 0.0015, and 0.0018 for SMO, KUT VXL, and RE, respectively. The analysisof breakthrough data using Thomas model resulted in similar values of x/m. Theresults also showed that uptake of oil by organo-clay/anthracite mixture canwell be described by a simple equation involving time such as Weber and Mowismodel.