Liz Chapman, PhD, Geochemist ECHELON Applied Geosciences · Marcellus produced waters • Four Pennsylvania counties Bradford Westmoreland Washington Greene • Different sample types:

Liz Chapman, PhD, Geochemist ECHELON Applied Geosciences

1 © Copyright 2014 EchelonAGC

For a variety of fluids (fresh water, brines, AMD-impacted and co-produced waters) And many geologic rock types and materials Coal, shale, permeable limestone and sandstone aquifers Deep and shallow units Cements/grouts/combustion byproducts (coal fly ash)

Must be able to identify contaminant source as well as provide ongoing monitoring

Introduced tracers Major and trace element geochemical signatures

Natural isotopic signatures © Copyright 2014 EchelonAGC

Four naturally-occurring stable isotopes, including 87Sr and 86Sr

87Sr is supplemented by the slow decay of 87Rb (‘radiogenic isotope’) Half-life: 48.8 Ga

87Sr/86Sr increases with time

© Copyright 2014 EchelonAGC Capo et al., 1998

Rocks with different compositions and geological histories develop distinct 87Sr/86Sr ratios Reflect sources of Sr available during

formation

Waters which interact with these units can inherit their 87Sr/86Sr

87Sr/86Sr in geologic materials is an indicator of both age and geochemical origin

© Copyright 2014 EchelonAGC


0.712 –

0.703 –

0.709 –

0.706 –

Millions of Years 0 200 400 600

– 40

– -80

– 0

– -40

87Sr/86Sr

(RIVERS)

εSrSW =104

87Sr/86Srsample87Sr/86Srseawater

−1


Ubiquitous and abundant One of the most abundant

trace elements in crustal rocks

Systematics understood Proxy for calcium Sr substitutes for Ca in

feldspars, carbonates, sulfate minerals


Does not fractionate appreciably during physical, chemical, or biological processes (e.g. evaporation, mineral dissolution/precipitation, oxidation, sorption, plant/animal uptake)

Isotopic composition not affected by dilution or mixing


http://www.seenobjects.org/2004-12-09-sparkler�

~2 µg (10-6 g) Sr Minimum amount of

sample to process: Carbonate: < 10 mg Shale: 20-50 mg Groundwater: 0.5 – 1L Produced water: <500 mL AMD: <500 mL Brines: <500 mL


Can be diagnostic of specific environmental reservoirs, both natural and anthropogenic

Allows for sensitive quantification of source inputs, especially if source 87Sr/86Sr are very different

Quantifies amount of mixing

Can be used to understand processes along migration pathways


To use Sr isotopes as a tracer, you must have: The isotopic composition of all

sources of Sr to the system The extent of isotopic variation

within individual sources


Since some samples might have very low Sr concentrations, it is important to have very clean sample bottles for collection HDPE bottles (Nalgene) Acid-washed (ex. p. 22-27,

http://water.usgs.gov/owq/FieldManual/chapter3/final508Chap3book.pdf)

Samples for Sr isotope analysis should not be acidified in the field unless ultrapure acids are available

When possible, separate aliquots should be taken for major/trace element concentrations, and Sr isotopes


Solid samples (cores, well cuttings, ash, soil, rocks) must be pulverized and leached to release Sr Simulate fluids that may

be interacting with the rock


Sr concentrations must be accurately measured before processing for Sr isotopes Sr separation performed in Class 100 clean lab


Rapid Sr separation (5x faster than traditional methods) Vacuum-assisted Disposable columns

Isotope measurement by multicollector ICP-MS (4-5 samples/hr) Typical measurement

uncertainty: 0.002% See Wall et al., 2013


Marcellus produced waters • Four Pennsylvania counties Bradford Westmoreland Washington Greene

• Different sample types:

Individual well single samples

Impoundment samples

Produced water time series

© Copyright 2014 EchelonAGC Chapman et al., 2012; Kolesar et al., 2013; Capo et al., 2014

0

4

8

12

16

2

6

14

10

10 20 30 40 50

Sr SW ε

Bradford Co. Bradford Co.

Washington Co.

Greene Co.

Westmoreland Co. n

0.710 0.711 0.712 87Sr/86Sr

© Copyright 2014 EchelonAGC Chapman et al., 2012



15 20 25 30 35 40 45 50 55 60 0.0

0.1

0.2

0.3

0.4

Sr SW ε

Sr/C

a

Bradford Co. PW Westmoreland Co. PW Greene Co. PW Washington Co. PW Pittsburgh Coal AMD Stream water

to Venango brine (0.015, +111)

Combination of εSr and Sr/Ca likely to distinguish between sources in nearly all cases Even small amounts (<0.1%) of Marcellus produced waters

can significantly shift stream εSr


87Sr/86Sr of Marcellus produced waters allow for extremely sensitive tracking

Potential applications: Verification of safe water disposal Determination of origin of dissolved

constituents in surface and ground waters affected by multiple sources Quantification of mixing

Produced water reflects: Original composition of injection water Mobilized constituents from shale Formation waters liberated by fracing

Sr measured in flowback up to 27 months post-frac



Strontium concentrations plateau within the first year of flowback


εSr continue to increase even after 2 years

Greene A

Greene B

Greene C


Linear correlations indicate the increase in εSr is due to mixing of two endmembers

Produced water

Leachates


Stewart et al., in press © Copyright 2014 EchelonAGC

Higher- εSr endmember most likely pore/formation water rather than soluble salts


Sr isotopes show a change over time from frac fluid + formation water primarily formation water Mixing models combined with leachate geochemistry

suggest that dissolved salts in produced water originate in formation water rather than as soluble salts within the shale

Greene County site with: Six Marcellus laterals One vertical Marcellus well Four Upper Devonian (UD) gas wells One shallow groundwater spring

Sr measured before and after hydraulic fracturing of laterals

© Copyright 2014 EchelonAGC Kolesar Kohl et al., 2014



Most UD wells show no change after fracturing (p values >0.05)

For isotopic shifts to be considered significant enough to suggest Marcellus fluid incursion, εSr would need to decrease by 1-3 units

Kolesar Kohl et al., 2014


Sr isotope values fall between Marcellus and Upper Devonian values

Values shift on a semiannual basis (±0.8 from the mean)

Spring water contains very little Sr

Very sensitive to any potential mixing with produced water



The only well in the study that showed a significant change in Sr isotope values after horizontal wells were fractured (from +33.8 to +35.9)

Sr concentration also increased by ~200 mg/L

New pathways within the Marcellus were opened up by fracturing



Calculated mixing models between produced waters and spring water

Most sensitive elements: Ba, Br, Cl, Sr

Elemental ratios (Sr/Ca, Br/Cl) less sensitive than absolute concentrations



Greater sensitivity than elemental conc., especially in waters with natural seasonal variation

Unlike elemental concentration, Sr isotopes can distinguish between UD and Marcellus produced waters



Subsequent to hydraulic fracturing, no significant migration of Marcellus-derived fluids was observed in Upper Devonian or shallow groundwater units Shift in Sr isotopes of vertical Marcellus well suggests

fracturing opened new flowpaths within the unit Sr isotopes show greater sensitivity to potential brine

migration than elemental concentrations or ratios

History of fossil fuel activities: Oil and gas extraction since mid-1800’s, Upper Devonian sands (Bradford,

Venango) Hilltop strip mines (Clarion, Brookville coals)


Unplugged wells Artesian flow (>20 gpm) High total dissolved solids (TDS) Sulfate-dominated >100 mg/L iron


Gas well discharges: High sulfate, iron Low sodium, chloride

AMD: High sulfate Low iron

Oil and gas brines: High sodium High chloride


Siderite nodules (FeCO3) Coal overburden

Siderite cement Shallow sandstone drinking

water aquifers


Chapman et al., 2013 © Copyright 2014 EchelonAGC

1-3: Gas well discharges

4-7: AMD

8: Unaffected aquifer water

9-10: Venango oil and gas brines

11-13: Siderite nodules

14-16: Siderite cement

Chapman et al., 2013 © Copyright 2014 EchelonAGC

87Sr/ 86Sr Sr SW ε

- 0.710

- 0.712

- 0.714

- 0.716

- 0.718 Venango brines

Siderite nodules

Siderite cement

Gas well discharges AMD

Unaffected aquifer


87Sr/ 86Sr

Siderite cement

- 0.719

- 0.718

- 0.717

- 0.716

- 0.715


Supported by saturation indices generated by PHREEQc modelling software


Variety of sample types are often necessary to understand subsurface processes

Strontium isotopes able to: Differentiate between coal- and oil/gas-related

inputs Distinguish mineralogically identical but

genetically different sources


Inactive coal mine partially filled by injection of grout to mitigate AMD 98% coal utilization byproducts (CUB) 2% Portland cement

Mine sealing unsuccessful – AMD still leaking 3 years later Major/trace element chemistry not sufficient to

distinguish between discharges which interacted with grout and those that did not


Hamel et al., 2010

46

© Copyright 2014 EchelonAGC Hamel et al., 2010

Eight sites sampled over a period of 2 years

47

© Copyright 2014 EchelonAGC Hamel et al., 2010

Sr isotopes show that discharges received 30-40% of their total Sr from grout material Grout is chemically eroding at a rate of approx.

0.04% (3 x 104 kg) per year


Hamel et al., 2010

Sr isotopes able to distinguish waters interacting with grout material vs waters with no interaction Major/trace element chemistry unable to differentiate

Sr isotopes used to calculate rate of grout dissolution


Hamel et al., 2010

Capo, R.C., Stewart, B.W., Rowan, E., Kolesar, C., Wall, A.J., Chapman, E.C., Hammack, R.W., and Schroeder, K.T., 2014. The strontium isotopic evolution of Marcellus Formation produced waters, southwestern Pennsylvania. International Journal of Coal Geology, available online 28 Dec 2013. Capo, R.C., Stewart, B.W., and Chadwick, O.A., 1998. Strontium isotopes as tracers of

ecosystem processes: theory and methods. Geoderma, v. 82, p. 197-225. Chapman, E.C., Capo, R.C., Stewart, B.W., Hedin, R.S., Weaver, T.J., and Edenborn,

H.M., 2013. Strontium isotope quantification of siderite, brine and acid mine drainage contributions to abandoned gas well discharges in the Appalachian Pleateau. Applied Geochemistry, v. 31, p. 109-118. Chapman, E.C., Capo, R.C., Stewart, B.W., Kirby, C.S., Hammack, R.W., Schroeder, K.T.,

and Edenborn, H.M., 2012. Geochemical and strontium isotope characterization of produced waters from Marcellus Shale natural gas extraction. Environmental Science & Technology, v. 46, p. 3545-3553. Chapman E.C., Capo, R.C., Stewart, B.W., Johnson, J.D., Graney, J.R., Hammack, R.W.,

2011. Geochemical and strontium isotope study of sequentially extracted metals from Marcellus Shale drill core. Geol. Soc. Am. Abstr. Prog., #236-3.

50


Hamel, B.L., Stewart, B.W., Kim, A.G., 2010. Tracing the interaction of acid mine drainage with coal utilization byproducts in a grouted mine: Strontium isotope study of the inactive Omega Coal Mine, West Virginia (USA). Applied Geochemistry, v. 25, p. 212-223. Kolesar Kohl, C.A., Capo, R.C., Stewart, B.W., Wall, A.J., Schroeder, K.T.,

Hammack, R.W., and Guthrie, G.D., 2014. Strontium isotopes test long-term zonal isolation of injected and Marcellus Formation water after hydraulic fracturing. Environmental Science & Technology, v. 48, p. 9867-9873. Kolesar, C.A., Capo R.C., Wall, A.J., Stewart, B.W., Schroeder, K.T., Hammack, R.W.,

2013. Using strontium isotopes to test stratigraphic isolation of injected and formation waters during hydraulic fracturing, AAPG Search and Discovery Article #90163. Stewart, B.W., Chapman, E.C., Capo, R.C., Johnson, J., Graney, J.R., Kirby, C.S., and

Schroeder, K.T., in press. Origin of brines, salts and carbonate from shales of the Marcellus Formation: Geochemical and Sr isotope study of sequentially extracted fluids, Applied Geochemistry.

51


University of Pittsburgh: Rosemary Capo, Brian Stewart, Courtney Kohl, Andrew Wall, James Gardiner Department of Energy: Karl Schoeder, Rick Hammack, Hank

Edenborn Hedin Environmental: Bob Hedin, Ted Weaver

52


Liz Chapman, PhD, Geochemist ECHELON Applied Geosciences · Marcellus produced waters • Four Pennsylvania counties Bradford Westmoreland Washington Greene • Different sample types:

Documents