Natural gas production in the United States has increased dramatically in recent years due to hydraulic fracturing. The cost to harvest natural gas from shale formations has greatly decreased thanks to this process. The developments of this technology have led to cheaper energy for many people but we do not yet know the full scope of the environmental impacts of the process, including the effects that these large scale operations have on air quality. A poorly characterized source of volatile organic compounds (VOCs) and other emissions from hydraulic fracturing are atmospherically vented storage tanks for hydrocarbons and flowback water. Pollutants such as VOCs can vent from these tanks during flowback period as well as the time of normal well operation (the water is then considered “produced” rather than flowback).
In this work 12 samples of hydraulic fracturing flowback fluid are analyzed for their potential to affect atmospheric chemistry. The primary emissions of concern are VOCs, which affect the production of both ozone and particulate matter. Two indicators are used to assess the air quality implications of these samples of flowback fluid: total volatile organic carbon (TVOC) and volatile carbon emission rate.
The total volatile organic carbon is measured by allowing a volume of 200-800 µL of a sample to evaporate into a 60 L Teflon® bag filled with clean air and then sampling the total gas-phase carbon with a flame ionization detector. The results are shown in Figure 1. A wide range of concentrations is observed (0-114 mgC/L), highlighting the range of potential concentrations that can occur at different times and locations in the flowback process. The total volatile organic carbon for many of these samples is quite high considering that the total organic carbon (volatile + non-volatile) of normal surface waters do not typically exceed 10 mgC/L (World Health Organization).
Figure 1. Total volatile organic carbon (TVOC) for 12 samples of flowback fluid.
The emission rate of organic carbon is measured for each sample by slowly blowing clean air over the top and sampling with a flame ionization detector. 10 mL of each sample was used, with ~5 cm2 of the sample exposed to a stream of 1.5 LPM clean air. Figure 2 shows results on volatile carbon emission rates (mgC/min) per the amount of fluid and the surface area of the exposed surface. This measurement could be used to estimate a range of emissions for flowback fluid storage tanks, as long as fluid volume and exposed surface area are known. Again, it should be noted that there is a wide range of measured rates, which indicates that there can be a lot of variation in the composition of flowback fluid.
Figure 2. Evaporation rates for the 12 flowback fluid samples.
We conduct additional experiments and quantify the potential effects of flowback fluid evaporation on concentrations of ozone and particulate matter. Ozone and particulate matter are both EPA criteria pollutants which have threshold levels that are costly to meet for certain areas, and they adversely affect human health. A single hydraulically fractured well uses 2-5 million gallons of water (www.epa.gov) which means that for a region with large hydraulic fracturing activity the implications of these emissions on air quality are important to understand.
See more of this Group/Topical: Environmental Division