TY - JOUR
T1 - Chemical diffusion between a fracture and the surrounding matrix
T2 - Measurement by computed tomography and modeling
AU - Polak, Amir
AU - Grader, Abraham S.
AU - Wallach, Rony
AU - Nativ, Ronit
PY - 2003/4
Y1 - 2003/4
N2 - A medical-based X-ray CT scanner was used to monitor the diffusion of NaI into the matrix of a 20-cm-long, 5-cm diameter fractured chalk core. The core was retrieved from a core hole at a depth of 18.3 m and was artificially fractured along its axis using a Brazilian-like test. The NaI solution flowed continuously along the vertically oriented fracture and the transient lateral concentration distribution within the matrix at different cross sections along the core was monitored by two-dimensional 2-mm-thick slices through the sample and an in-plan pixel resolution of about 0.25 mm. The lateral concentration distribution within the matrix was characterized by a sharp decrease at a thin matrix layer adjacent to the fracture/matrix interface followed by diffusion-type concentration distribution elsewhere. This concentration variation suggests that a thin transition layer exists along the fracture/ matrix interface where the diffusion coefficient is higher than that of the bulk matrix. The higher diffusion coefficient of the transition layer is possibly related to minifissures that develop when fractures are formed. After 6 days of tracer injection into the fracture inlet, distilled water was injected for 11 days, forming a reverse concentration gradient and back diffusion. A mathematical model that assumes diffusion within the matrix and a linear concentration variation through the transition layer from its value in the fracture to its time-dependent value at the transition layer/matrix, interface was developed. Very good agreement was obtained between the predicted and measured concentrations during both the diffusion and back diffusion phases. Application of the model to a field site in the Negev desert, Israel, suggested that the rock matrix that had been subjected to 20 year of contaminant diffusion would require more than 200 years before it would stop releasing contaminants into the intersecting fractures (a parabolic process). According to these calculations, remediation efforts based on clean water injection into the fractures are not feasible.
AB - A medical-based X-ray CT scanner was used to monitor the diffusion of NaI into the matrix of a 20-cm-long, 5-cm diameter fractured chalk core. The core was retrieved from a core hole at a depth of 18.3 m and was artificially fractured along its axis using a Brazilian-like test. The NaI solution flowed continuously along the vertically oriented fracture and the transient lateral concentration distribution within the matrix at different cross sections along the core was monitored by two-dimensional 2-mm-thick slices through the sample and an in-plan pixel resolution of about 0.25 mm. The lateral concentration distribution within the matrix was characterized by a sharp decrease at a thin matrix layer adjacent to the fracture/matrix interface followed by diffusion-type concentration distribution elsewhere. This concentration variation suggests that a thin transition layer exists along the fracture/ matrix interface where the diffusion coefficient is higher than that of the bulk matrix. The higher diffusion coefficient of the transition layer is possibly related to minifissures that develop when fractures are formed. After 6 days of tracer injection into the fracture inlet, distilled water was injected for 11 days, forming a reverse concentration gradient and back diffusion. A mathematical model that assumes diffusion within the matrix and a linear concentration variation through the transition layer from its value in the fracture to its time-dependent value at the transition layer/matrix, interface was developed. Very good agreement was obtained between the predicted and measured concentrations during both the diffusion and back diffusion phases. Application of the model to a field site in the Negev desert, Israel, suggested that the rock matrix that had been subjected to 20 year of contaminant diffusion would require more than 200 years before it would stop releasing contaminants into the intersecting fractures (a parabolic process). According to these calculations, remediation efforts based on clean water injection into the fractures are not feasible.
KW - Chalk
KW - Computerized tomography
KW - Contaminant transport
KW - Diffusion
KW - Fractured rock
UR - http://www.scopus.com/inward/record.url?scp=0242486356&partnerID=8YFLogxK
U2 - 10.1029/2001WR000813
DO - 10.1029/2001WR000813
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:0242486356
SN - 0043-1397
VL - 39
SP - SBH101-SBH1014
JO - Water Resources Research
JF - Water Resources Research
IS - 4
ER -