The liquid carbon dioxide (CO2) fracturing process was studied recently during the completion of a 19-well shallow gas project for Crestar Energy Inc. in southeastern Alberta. The project involved mini fracture treatments that used special tools to record bottomhole pressure, temperature, and density conditions at the perforations. The analysis of the data gathered during this study has shown some surprising results, and has led to the development of a new technique that may enhance the characteristics of the liquid CO2fracturing process, and lower the overall costs associated with this type of treatment. This paper discusses the uses of low-viscosity, low-temperature fracturing fluids, and the ways in which they differ from conventional fracturing fluids. Case studies from the Crestar Energy shallow gas project are used to demonstrate the cost savings realized by combining liquid CO2 with nitrogen (N2) during the fracturing process. Introduction The term low-viscosity, low-temperature fracture fluid refers to the unconventional use of products like liquid CO2 and other compressed gases, without the use of other fluids or additives, to create hydraulic fractures in formations. (The term gas refers to the "normal" state of a substance at atmospheric conditions. When a gas is compressed and cooled, it can become a liquid.) When these products are used in fracturing treatments, their viscosity is less than 1 cP at temperatures below O °C, which is significantly lower than conventional fracturing fluids. Conventional fracturing fluids include water, crude and refined oils, solvents, alcohols, and combinations of these fluids. The variety of treatment fluids reflects the diversity of formations, as well as the need to choose a fluid that is compatible with the reservoir. Conventional fracture theory requires that a fracture fluid have sufficient viscosity to (a) create a pressure greater than the fracture pressure of the formation, and (b) carry proppants into the fracture. This is accomplished by adding viscosity-enhancing chemicals to conventional fluids. These chemicals create a filter cake when the fluid is filtered against the fracture walls. This effectively lowers leak-off values; however, the chemicals and the fluids themselves can leave residual damage that results in less than- optimal productivity(l). Research and development to improve the quality, cleanliness, and performance of fracturing fluids are ongoing, with the ultimate goal being to reduce or eliminate the negative effects of treatment fluids on the reservoirs being treated. Previous papers(2,3) show that low-permeability gas formations can suffer production problems as a result of treatment fluid invasion. Fluids pumped into formations can change formation saturation properties, especially in dry gas wells. The resulting changes in relative permeabilities, capillary pressures, and possible aqueous-phase trapping can all lead to lower production rates. The use of gases like nitrogen and carbon dioxide is not new to the oil and gas industry, and is well documented(4,5,6). These gases (or liquefied gases) add new dimensions to fracturing fluids that are highlighted by the properties of the gases themselves. A property common to all gases is the ability to be compressed.