Controls of low injectivity caused by interaction of reservoir and clogging processes in a sedimentary geothermal aquifer (Mezőberény, Hungary)

Low injectivity is often experienced in geothermal doublets installed in sandstone reservoirs. This even led to a shut-down of the Mezőberény (Hungary) geothermal site. An on-site campaign was carried out in January 2021 to prepare a stimulation aiming to enhance the transmissivity of the sedimentary reservoir and the near well-bore zone of this site. Previous studies have concluded that insufficient injectivity may be linked to a high skin effect in the near well-bore zone and pore clogging in combination with the low net sandstone content of the fluvio-deltaic reservoir. A chemical soft stimulation based on the injection of hydrochloric acid (HCl) was successfully used to unclog and recover the well injectivity. Despite of such empirical evidence, the geochemical mechanisms leading to both, detrimental formation of clogging and the HCl-driven transmissivity restoration, have not been elucidated yet. This work presents the results of a novel analysis aiming at (a) predicting the dominant type of clogging forming in the near well bore zone, (b) quantifying the drop in hydraulic conductivity as clogging occurs and (c) support the optimization of the HCl dosage during the chemical soft stimulation. The study is supported by new experimental datasets never presented before from the Mezőberény site and a geochemical model simulating the main mechanisms involved in the clogging and unclogging processes. It is concluded that the biofilm formation was the dominant, while the precipitation of calcite and amorphous ferrihydrite – later reduced to magnetite by microbes – was the secondary clogging mechanism: In the long-term (yearly scale) simulating the hydraulic conductivity showed a decline with forming scales, therefore biofilm was presumably responsible for the experienced rapid (one month) clogging. When modelling the chemical stimulation, the estimated amount of precipitated minerals (Case 1) was dissolved already with 2.5 moles of HCl per liter water (~ 10 m/m%). Therefore, the chosen 20 m/m% of HCl might had a beneficial effect dissolving the potentially higher amount of scaling (Case 2) and/or the carbonate minerals of the matrix near the well-bore. Overall, it is concluded that the chemical and the microbial analysis together with the geochemical model was critical to tailor the remediation attempts and to propose further development or reconstruction of the surface system before going into operation to prevent recurrent impairments.