Unconventional gas has become an important resource to help meet our future energy demands. Although plentiful, it is difficult to produce this resource, when locked in a massive sedimentary formation. Among all unconventional gas resources, tight gas sands represent a big fraction and are often characterized by very low porosity and permeability associated with their producing formations, resulting in extremely low production rate. The low flow properties and the recovery factors of these sands make necessary continuous efforts to reduce costs and improve efficiency in all aspects of drilling, completion and production techniques. Many of the recent improvements have been in well completions and hydraulic fracturing. Thus, the main goal of a hydraulic fracture is to create a long, highly conductive fracture to facilitate the gas flow from the reservoir to the wellbore to obtain commercial production rates. Fracture conductivity depends on several factors, such as like the damage created by the gel during the treatment and the gel clean-up after the treatment. This research is focused on predicting more accurately the fracture conductivity, the gel damage created in fractures, and the fracture cleanup after a hydraulic fracture treatment under certain pressure and temperature conditions. Parameters that alter fracture conductivity, such as polymer concentration, breaker concentration and gas flow rate, are also examined in this study. A series of experiments, using a procedure of “dynamical fracture conductivity test”, were carried out. This procedure simulates the proppant/frac fluid slurries flow into the fractures in a low-permeability rock, as it occurs in the field, using different combinations of polymer and breaker concentrations under reservoirs conditions. The result of this study provides the basis to optimize the fracturing fluids and the polymer loading at different reservoir conditions, which may result in a clean and conductive fracture. Success in improving this process will help to decrease capital expenditures and increase the production in unconventional tight gas reservoirs.

This work is about fracture conductivity; how to measure and model it based on experimental data. It is also about how to determine the relative importance of the factors that affect its magnitude and how to predict its magnitude based on these factors. We dynamically placed the slurry hereby simulating the slurry placement procedure in a field-scale fracture. We also used factorial and fractional factorial designs as the basis of our experimental investigation. The analysis and interpretation of experimental results take into account the stochastic nature of the process. We found that the relative importance of the investigated factors is dependent on the presence of outliers and how they are handled. Based on our investigation we concluded that the investigated factors arranged in order of decreasing impact on conductivity are: closure stress, polymer loading, flow back rate, presence of breaker, temperature and proppant concentration. In particular, we find that at high temperatures, fracture conductivity was severely reduced due to the formation of a dense proppant-polymer cake. Also, dehydration of the residual gel in the fracture at high flow back rates appears to cause severe damage to conductivity at higher temperatures. This represents a new way of thinking about the fracture cleanup process; not only as a displacement process, but also as a displacement and evaporative process. In engineering practice, this implies that aggressive flow back schemes are not necessarily beneficial for conductivity development. Also, we find that at low proppant concentrations, there is the increased likelihood of the formation of channels and high porosity fractures resulting in high fracture conductivities. The uniqueness of this work is a focus on the development of a conductivity model using regression analysis and also the illustration of a procedure that can be used to develop a conductivity model using dimensional analysis. We reviewed both methodologies and applied them to the challenge of modeling fracture conductivity from experimental studies.