Soil and groundwater pollution by chlorinated solvents such as tricholorethene (TCE) and tetrachloroethene (PCE) is a frequent problem in the industrialized world. Chlorinated solvents, characterized by a low solubility and a density greater than water, form dense non-aqueous phase liquids (DNAPLs) when released in the subsurface. DNAPLs accumulate along low permeability layers and slowly dissolve in groundwater acting as a long-term source of contamination that can last for decades. Remediation of chlorinated solvent DNAPLs is recognized as one of the most challenging problems in the field of environmental remediation. In situ bioremediation (ISB) is a promising and cost-effective technology for their removal that relies on the activity of specialized microorganisms able to transform chlorinated compounds to ethene (a non-toxic product) via a stepwise anaerobic process called organohalide respiration (OHR). ISB has been applied successfully for the treatment of dissolved phase plumes since the early 1980’s. However, its application for source zones, where contaminants are present as DNAPLs, is relatively recent and has only been developed in the last decade. One of the major issues limiting source zone ISB is the acidification of the groundwater due to the transformation of chlorinated compounds by organohalide-respiring bacteria (OHRB) and the production of organic acids by fermentative microbial populations. OHRB are inactivated when the pH is below 5-6 and therefore pH buffer amendments are required when the soil buffering capacity is insufficient. In field applications, the most common method used for pH adjustment is the injection of soluble buffers such as sodium bicarbonate. However, this method requires frequent injections and constant monitoring as alkalinity is rapidly consumed. Therefore, there is a need to develop more efficient and long-lasting buffering strategies. The objective of this thesis was to develop a novel method for long-term control of groundwater pH that relies on the use of ground silicate minerals. Silicate minerals may act as a long-term source of alkalinity release as i) they dissolve slowly compared to carbonates and ii) their dissolution rate and solubility is pH-dependent and increase with acidic pH. In addition, they are easily available at an affordable cost as a raw material or as a by-product of industrial processes. Silicate minerals are the most common rock forming mineral and constitute a very diverse group with highly variable dissolution rates, solubilities and compositions. Only a restricted numbers of these minerals present appropriate characteristics to act as buffering agents. A screening methodology, based on numerical simulations, thermodynamic and kinetic considerations, was developed to select potential candidates for pH control. A geochemical model including the main microbial processes driving groundwater acidification and silicate mineral dissolution was developed as well. This model provides a useful design tool to estimate the mineral requirement in the perspective of field applications. The results of numerical simulations showed that a dozen silicate minerals have the potential to act as buffering agents. Abiotic batch experiments were conducted with five silicate minerals (nepheline, fay alite, forsterite, diopside and andradite) to validate and improve the geochemical model. Abiotic experiments confirmed the buffering potential of these minerals and revealed the importance of secondary precipitation, a process not included in the original formulation of the model. Precipitation of secondary phases can decrease the reactivity of silicates, reduce the aquifer porosity and precipitate nutrients. Therefore, prediction of secondary precipitations was included in the model in order to predict this type of reaction. The influence of silicate mineral dissolution on OHRB and fermentative bacteria was investigated in batch cultures. As expected, the five silicate minerals (except nepheline) were able to maintain the pH in the tolerance range for the three microbial consortia tested. However, transformation of cA-DCE to ethene was completely inhibited in most of the experiments in the presence of minerals. These results showed that compatibility of silicate minerals with the bacterial community involved in in situ bioremediation has to be carefully evaluated prior to their use for pH control at a specific site. Subsequently, the long-term buffering potential of the most promising buffering agents (diopside, fayalite, forsterite) was tested in continuous-flow column studies simulating chloroethene source zone conditions for six and a half month and a half. In contrast to batch experiments, transformation of cis-DCE to ethene was not inhibited by mineral dissolution in continuous flow systems. Olivine minerals (such as fayalite and forsterite) appeared as suitable pH buffering agents. They successfully maintained the pH in the neutral range (7.5 for forsterite and 6.5 for fayalite) and sustained the activity of OHRB bacteria. In contrast, the buffering potential of diopside rapidly decreased due to the formation of a less-reactive cation-depleted leached layer at the mineral surface. This thesis demonstrated the potential of silicate minerals to act as a long-term source of alkalinity release for groundwater pH control. A global strategy for the selection of appropriate buffering agents based on site characteristics was developed. This methodology was applied to the particular case of chlorinated solvent ISB but can be extended to any groundwater remediation technology requiring close to neutral pH conditions.