Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.252227
Title: The effect of pore structure on gas and liquid permeability in crystalline rocks
Author: Phillips, Matthew David
ISNI:       0000 0001 3489 0118
Awarding Body: University of London
Current Institution: University College London (University of London)
Date of Award: 2001
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
An investigation into the effect of pore structure on gas and liquid permeability in crystalline rocks has been performed on thirteen samples in order to give advice on future predictions of liquid permeabilities in low permeability rocks. It was concluded that this could be achieved by performing simple gas permeability experiments and correcting using the empirical formula presented. An empirical gas/liquid correction was determined using a techniques previously employed by Bloomfield and Williams (1995), which took the form: log10 k = 1.75log10 kg+11.03 where gas permeabilities (to Nitrogen), kg were in the range of 2.02 x 10-18 m2 to 1.74 x 1016 m2 and liquid permeabilities (to water), k, were in the range of 5.72 x 10-21 m2 to 4.01 x 10-17m2. After comparison with other empirical correlations (Bloomfield and Williams, 1995; Klinkenberg, 1941 and Lovelock, 1977) it was inferred that a 'crossover' (indicative of a change of dynamics in a system, in this case the flow of gas through porous media) was observed at kg = 1 x 10-16 m2, where values above this crossover were described by the empirical formula determined by Bloomfield and Williams and values below were described by the formula presented above. Upon further analysis, it was shown that the value of the crossover was equivalent to a Knudsen number, 8-1, which has been shown empirically in the literature to correspond to the transition flow regime where a combination of both slip and Knudsen flow occur. In addition, an extended experimental liquid permeability programme, where simple effective pressure was increased systematically, was performed on five samples chosen for their differences in pore structure (determined from MICP experiments and thin section analysis) to determine what effect pore structure had on porosity and permeability. It was found that no general trend was observed and that it varied dramatically for each sample though, on an individual basis, critical diameters that control the reduction in permeability were observed. Analytical modelling was employed to predict the liquid permeabilities of generic types of crystalline rock. Three models were used: a periodically constricted conduit model (based on Dullien, 1975a) extended to elliptical pores and parallel plates; a percolation model (Katz and Thompson, 1987) and the Carman-Kozeny model (1927). It was found that the Carman-Kozeny model was completely inadequate in the prediction of the transport properties of low permeability crystalline rocks. The other models proved that it was possible to predict the liquid permeability of certain samples, but no parameter identifying which choice of model would best predict the permeability of certain individual samples was recognised. The physical assumptions that underpin the Klinkenberg (1941) corrections, were investigated using three different gas types and four samples. It was found that the Klinkenberg correction, and hence the Klinkenberg test, was not adequate in predicting the liquid permeability where k 10-17 m2. In addition, the form of the Klinkenberg pore constant, b, found in the Klinkenberg correction: k = k_g (1+b/p_m) where pm is the mean gas pressure at which the permeability experiment was performed, was compared directly to the equivalent constant derived from kinetic theory (Scott and Dullien, 1962a). It was found that while the two forms were very similar, b, consistently overestimated the value of the average sample pore radius, r, by up to two orders of magnitude and furthermore, exhibited a dependence upon the pore/crack aspect ratio, [xi], for the sample not considered in the constant derived from kinetic theory.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.252227  DOI: Not available
Keywords: Geology
Share: