RMR_Poster_Final
- 1. Effect of pH and Ionic Strength (Salinity) on Geochemical Water-Rock Interactions During Hydraulic Fracturing in the Frontier Formation of the Powder River Basin, Wyoming Bratcher, Jordan1, Herz-Thyhsen, Ryan1, Kaszuba, John1,2 1) Department of Geology and Geophysics, University of Wyoming 2) School of Energy Resources, University of Wyoming Experimental Design Experimental Design Completion Report Fluid Additives USGS Data Formation Water Geochemical Model Experimental Frac Fluid Chemical Field Purpose Mass Fraction Frontier Fm. Water Solvent 86.9% Bauxite Proppant 11% HCl Dissolves Rock 0.12% Tetramethylammonium Chloride Clay stabilizer 0.088% Methanol Prevents corrosion and reduces friction 0.07% Sodium Erythorbate Prevents metal precipitation 0.0009% Results Conclusions Future Work Acknowledgements and References 0.01 0.1 1 10 100 1000 10000 -50 0 50 100 150 Conc.(mol/kg) Time (hrs) Neutral pH Experiment 0.01 0.1 1 10 100 1000 10000 -50 0 50 100 150 Time (hrs) Low pH Experiment SiSi Ca Ca Li Li Sr Sr Ab Qtz 50 m Cc Qtz 50 m 0.13 0.135 0.14 0.145 0.15 0.155 0 1 2 3 4 5 6 7 8 -50 50 150 250 350 450 550 650 pH pH of Neutral Experiment pH of low pH Experiment Ionic Strength of Neutral pH Experiment Ionic Strength of Low pH Experiment I(molal) Time (hrs) Experimental pH and Ionic Strength vs. Predictive Model Model 87% Water 2% Additives 11% Proppant Figure 2: Percent composition of frac fluid by volume. Water is by far the most important component. Background Hornbuckle 1-11H Herz-Thyhsen, 2014 (unpublished) Figure 4: Hornbuckle 1-11H well located in the southwest corner of the Powder River Basin. Figure 3: Experimental Apparatus. Pressure vessel consists of a stainless steel bomb housing a flexible gold bag capable of sampling during experimentation while maintaining constant temperature and pressure. Figure 1: Hydraulic fracturing overview. Large volumes of fresh water are needed to fracture a well which needs to be collected for treatment and disposal after production. http://imgbuddy.com/hydraulic-fracturing.asp ii i Quartz 64% Calcite 20% Illite 10% Albite 5% Pyrite 1% i iii iii • USGS data from the surrounding research area were compared to geochemical models to design the Frontier Fm. water chemistry (Figure 7) • Geochemical model allows Frontier Fm. mineralogy (Figure 6) to equilibrate with water. Water chemistry generated by model approximates average USGS Frontier Fm. chemistry well (Figure 7) Figure 6: Simplified Frontier Fm. composition. Five minerals were chosen to represent the Frontier Fm. for use in geochemical models (Herz-Thyhsen, 2014 (unpublished)). Well Data Average Model ii Figure 7: Development of Frontier Fm. water chemistry. Blue triangles represent USGS Frontier Fm. water chemistry taken from surrounding wells. Red circle is the statistical average of blue triangles. Green star represents formation water predicted by geochemical model. Figure 5: Development of experimental frac fluid. Data from several different sources were used to develop the frac fluid composition used in experiments. Figure 8: Experimental frac fluid chemistry. Fluid additives were selected from the Hornbuckle 1-11H (Figure 4) completion report based on functionality, lab safety, and accessibility (Herz- Thyhsen, 2014 (unpublished)). • 4 experiments over a range of pH and ionic strengths take place for a month at a time each (Figure 9) • Geochemical models are used to predict and analyze experimental results • Rock samples are taken from the Frontier Fm. interval of Hornbuckle 1-11H core (Figure 4), reacted with frac fluid, and both liquids and solids are analyzed • Data from 84 wells surrounding the Hornbuckle 1-11H were averaged and compared to geochemical model results (Figure 7) • Experimental ionic strength range designed to represent naturally existing ionic strength range in Frontier Fm. water and extended to include more saline flowback/ produced waters • Make use of produced and flowback waters • Free up freshwater for other uses • Mitigate transport, treatment, and disposal costs • Predict and track downhole changes quickly Research Impacts • Calcite and albite both show strong dissolution features among experiments however calcite is much more sensitive to low pH • Geochemical model predicts pH, ionic strength, and most major aqueous species within reasonable accuracy • This information forms a basis for predicting and tracing downhole chemical changes based on frac fluid chemistry • Understanding downhole chemistry is the first step to replacing freshwater with produced/flowback water in hydraulic fracturing • Compare results of experiments 1, 2, and 3 to better understand the effect of ionic strength on geochemical water-rock interactions Funding/ Support: UW Cretaceous Tight Oil Consortium II, School of Energy Resources Laboratory/Technical Aid: Norbert Swoboda-Colberg, Susan Swapp, Janet Dewey, Virginia Marcon, Alexa Socianu, Quin Miller, Jonathan Pullam, Bianca Majumder, Amber Zandanel References • Heier “Trace Elements in Feldspars-A Review” Norsk. Geol. Tidsskr., v. 42, 415-454 • Paquette and Reeder “Relationship Between Surface Structure, Growth Mechanism, and Trace Element Incorporation in Calcite” Geochimica et Cosmochimica Acta, v. 59, 735-749 Experimental Design 0.01 0.1 1 10 100 1000 10000 100000 1000000 -50 0 50 100 150 200 250 300 350 400 450 500 550 600 650 Conc.(mol/kg) Time (hrs) Neutral pH Experiment SO4 Cl Na K Mg Ca Si Li Sr Al Model Initial pH = 7.3 T = 115 C P = 350 bar I = ~0.15 molal 0.01 0.1 1 10 100 1000 10000 100000 1000000 -50 0 50 100 150 200 250 300 350 400 450 500 550 600 650 Conc.(mol/kg) Time (hrs) SO4 Cl Na K Mg Si Li Sr Al ModelLow pH Experiment Initial pH = 7.3 T = 115 C P = 350 bar I = ~0.15 molal Figure 9: Initial conditions of each experiment. Subsequent figures focus on the effect of pH on geochemical reactions (expt. 2 and 4). • Geochemical model adequately predicts pH • Ionic strength predictions within 10% difference • Geochemical model accurately predicts concentrations of most aqueous species for experiments 2 and 4 Figure 9: SEM images of dissolving albite (left) and calcite (right) in reacted chips. While both minerals show dissolution features, calcite shows a stronger dissolution response to low pH. Fluid Characterization Figure 10: Comparison of pH and ionic strength for experiments 2 and 4 to predictive model. Model predicts pH of both low and neutral experiments well while ionic strength is still within 10% difference. Figure 11: Comparison of aqueous species of experiment 4 to predictive model. Na, Cl, SO4, and Si are predicted well by model while K and Ca are predicted to be lower concentrations. Figure 12: Comparison of aqueous species from experiment 2 to predictive model. Ca and K are still predicted at lower concentrations. Noticeable differences between experiment 2 and 4 are the elevated concentrations of Ca, Sr, and Mg. Figure 13: Comparison of major and minor elements associated with calcite and albite between experiments 2 and 4. Ca and Sr show the strongest response to low pH conditions indicating an increase in dissolution of calcite. Li and Si show little response meaning albite dissolution is relatively unaffected by changing pH on experimental time scales. • Fresh water usage in hydraulic fracturing is critical (Figure 1 and 2) • 2 - 8 million gallons freshwater required per well (USGS) • Freshwater typically hauled to drilling sites • Wastewater is collected during production to be treated and disposed of (Figure 1) • The potential to replace freshwater with produced and flowback water exists • The ability to use geochemical analysis and models to predict and track downhole changes in siliciclastic reservoirs such as the Frontier Formation is appealing Results Robert Kirkwood – background image Seyfried et al., 1987 Gold Bag Sample Port Pressure Vessel