INNOVATION March-April 2012
Table 1. Composition of flow-back fluids from shale gas formations.
Ca 2+ (mg/L)
HCO - (mg/L) 3
Cl - (mg/L)
TDS (mg/L)
Specific Gravity
Na + (mg/L)
Composition → Shale
Subsequently, periodic extraction of formation brines can lead to their storage in on-site impound- ments, recycling or shipping for treat- ment, recycling of flow-back fluids and deep-well disposal of the brines. Selected inorganic chemical parameters of the
Fayetteville
15,000 1.01
5,400
260
1,300
8,000
Marcellus
72,000 1.05
24,000 2,900
260
44,000
Barnett
40,000 1.03
12,000 2,200
290
24,000
Source: Aqua-Pure Ventures, Calgary, Alberta
waves through the casing and the cement to the forma- tion, are employed to verify the bonding of the casing to the formation (ie, a cement bond log). Use of geotextiles or engineered clays can prevent seepage through surface impoundments to shallow groundwater or streams. Groundwater Contamination by Methane The presence of methane in some shallow wells and aquifers is not an unusual occurrence in western Canada. Under highly reducing conditions, acetate or CO 2 may be reduced to methane by microbes present in the aquifer producing biogenic methane. Shale gas or thermogenic methane, however, is principally produced in deep for- mations at elevated temperatures and pressures. Biogenic and thermogenic methane can be differentiated through their isotopic signatures involving the measurement of their stable hydrogen and carbon isotope ratios. Figure 2 shows the isotopic domains of the various origins of methane. Some US states, such as Pennsylvania, recommend sampling groundwaters for “pre-drill parameters” before shale gas development. The various sources of hydro- carbon gases however can complicate the identification of their origin when based solely on their measured concentrations. For example, some shallow aquifers in Alberta and Texas produce both methane and ethane by biogenesis. Other shallow aquifers in Ontario, Illinois, and Iowa apparently produce only methane. Sometimes
flow-back fluids are shown in Table 1 for three US shale gas developments: the Fayetteville in Arkansas, Marcel- lus in Pennsylvania and Barnett in Texas. Included in these brines would be synthetic organic chemicals such as polyacrylamide, a gelling agent, surfactants for wet- tability control and biocides to inhibit well fouling and sulphate reduction. In a recent case in which the US Environmental Pro- tection Agency (EPA) installed two deep monitoring wells between the zone of hydraulic fracturing and the zone of groundwater production beneath Pavillion, Wyoming, the EPA wells showed evidence of thermogenic methane, potassium hydroxide, diesel and gasoline range organics and other synthetic organic chemicals that the EPA associ- ated with hydraulic stimulation operations, ie, fracking. The EPA also argued that state records indicated that there was evidence of poor well completions in several produc- ing wells in the area, ie, poor cement bonding of produc- tion or other casing to the adjacent formations. Hydraulic fracturing will produce micro-seismic signals that can be monitored from an adjacent borehole. This information was used by Kevin Fisher, General Manager of Pinnacle-Halliburton, a microseismic test- ing service company, to claim that there is no evidence of vertical fractures propagating “out of zone” from the shales upward into shallow aquifers, which was the con- cern in hearings before the US Congress in 2010. Fisher’s data suggest that the separation between the base of fresh groundwater and the top of the vertical fracture is at least 1,000 m in the Barnett and Marcel- lus Shales. Even greater vertical separations should be expected—2 km or more—in Northeastern BC. However, if fracking occurs at shallow depths, as at Pavillion, Wyoming or (anecdotally) in two cases in Alberta, then contamination of shallow wells by fracturing chemicals or natural gas is obviously possible and avoidable. Rather than out-of-zone fracturing, evidence from the US indicates that shallow groundwater contamina- tion by methane from shale gas development appears to be caused by poor cement completions of gas wells (Bainbridge Township, Ohio; Dimmock, Pennsylvania or Pavillion, Wyoming) and leakage from poorly con- structed surface impoundments (anecdotal). These causes can be readily prevented by good drilling and geoenvi- ronmental engineering practice, respectively. Downhole geophysical logs, measuring the travel time of sound
Figure 2 . Isotopic domains for methane
-150
Biogenic CO 2
-200
Thermogenic
-250
Biogenic
CH4 ‰ VSMOW -300
Thermogenic reservoir gas
δ 2 H
Ambiogenic (crystalline terrains)
Biogenic acetate fermentation
Kettle Point fm, Ontario
-350
Alberta Coal Bed Methane
Alliston buried valley aquifer, Ontario
-400
-110
-90
-70
-50
-30
-10
δ 13 C
‰ VPDB
CH4
2 9
M a r ch/A p r i l 2 012
i n n o va t i o n
Made with FlippingBook - Online catalogs