Dense Nonaqueous Phase Liquid Architecture and Dissolution in Discretely Fractured Sandstone Blocks

• Published in: Environmental science & technology 2009-03, Vol.43 (6), p.1877-1883
• Main Authors: Schaefer, Charles E, Callaghan, Amy V, King, Jared D, McCray, John E
• Format: Article
• Language: English
• Subjects: Applied sciences; Aqueous solutions; Correlation analysis; Dissolution; Environmental Processes; Exact sciences and technology; Fractures; Geological Phenomena; Models, Chemical; Pollution; Reynolds number; Rock deformation; Solubility; Studies; Tetrachloroethylene - chemistry; Water Pollutants, Chemical - chemistry
• ISSN: 0013-936X; 1520-5851
• DOI: 10.1021/es8011172

Laboratory experiments were performed in discretely fractured sandstone blocks to evaluate residual dense nonaqueous phase liquid (DNAPL) architecture and dissolution. Tetrachloroethene (PCE) DNAPL residual saturations (DNAPL volume/fracture volume) ranged between 0.18 and 0.52 for the rocks studied. DNAPL−water specific interfacial areas ranged between 19 and 57 cm2/cm3. No measurable correlation was observed between DNAPL−water interfacial area and aperture, aperture ratio, or residual saturation. DNAPL−water interfacial areas were comparable to those reported in sands with grain diameters similar to the rock apertures. However, the DNAPL residual saturation in the fractures were 2−4 times greater than in the sands, suggesting that PCE dissolution rates in rock fractures may be substantially less than in unconsolidated media, as the effective interfacial area per volume of DNAPL in rock fractures was less than in sands. Comparison of dissolution mass transfer coefficients in the bedrock fractures to corresponding mass transfer coefficients measured in sands indicated that dissolution rates in bedrock fractures were substantially less than dissolution rates measured in sands, even after normalization to DNAPL−water interfacial area. The presence of preferential water and DNAPL flow paths within the discrete fractures was shown to have a significant impact on observed DNAPL dissolution rates. DNAPL dissolution was reasonably described by a Reynolds number correlation that incorporated flow characteristics and the DNAPL−water interfacial area.