Premium
Deep well injection of brine from Paradox Valley, Colorado: Potential major precipitation problems remediated by nanofiltration
Author(s) -
Kharaka Yousif K.,
Ambats Gil,
Thordsen James J.,
Davis Roy A.
Publication year - 1997
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/97wr00573
Subject(s) - brine , nanofiltration , anhydrite , aquifer , geology , groundwater , gypsum , salinity , precipitation , geochemistry , reverse osmosis , geochemical modeling , hydrology (agriculture) , calcite , environmental science , mineralogy , membrane , chemistry , geotechnical engineering , paleontology , biochemistry , oceanography , physics , organic chemistry , meteorology
Groundwater brine seepage into the Dolores River in Paradox Valley, Colorado, increases the dissolved solids load of the Colorado River annually by ∼2.0 × 10 8 kg. To abate this natural contamination, the Bureau of Reclamation plans to pump ∼3540 m 3 /d of brine from 12 shallow wells located along the Dolores River. The brine, with a salinity of 250,000 mg/L, will be piped to the deepest (4.9 km) disposal well in the world and injected mainly into the Mississippian Leadville Limestone. Geochemical modeling indicates, and water‐rock experiments confirm, that a huge mass of anhydrite (∼1.0 × 10 4 kg/d) likely will precipitate from the injected brine at downhole conditions of 120°C and 500 bars. Anhydrite precipitation could increase by up to 3 times if the injected brine is allowed to mix with the highly incompatible formation water of the Leadville Limestone and if the Mg in this brine dolomitizes the calcite of the aquifer. Laboratory experiments demonstrate that nanofiltration membranes, which are selective to divalent anions, provide a new technology that remediates the precipitation problem by removing ∼98% of dissolved SO 4 from the hypersaline brine. The fluid pressure used (50 bars) is much lower than would be required for traditional reverse osmosis membranes because nanofiltration membranes have a low rejection efficiency (5–10%) for monovalent anions. Our results indicate that the proportion of treatable brine increases from ∼60% to >85% with the addition of trace concentrations of a precipitation inhibitor and by blending the raw brine with the effluent stream.