- Title
- Application of supercritical carbon dioxide in engineered geothermal system
- Creator
- Remoroza, Alvin I.
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2013
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- The present thesis is concerned with geothermal energy and specifically focuses on engineered geothermal systems (EGS), which are among a portfolio of technology options for power generation from geothermal resources. In this cyclic approach (also known as hot-dry-rocks or enhanced geothermal systems), high pressure water (i.e. "geofluid") is first pumped down a borehole (known as injection well) into a bed of hot fractured rock and forced to travel through the bed, capturing the heat content of the rocks. The hot water is then extracted from a second borehole (known as production well) and sent into a binary power plant, where its thermal energy is converted to electricity. The cooled water exiting the power plant is then injected back into the ground to resume the cycle. The aim of this thesis is to advance the understanding of CO₂ based EGS power generation process and verify the merits of using CO₂ rather than water for heat extraction from fractured hot dry rocks. The work has been largely driven by the suitable thermodynamic and transport properties of supercritical CO₂ (scCO₂), which makes it a desirable candidate for harnessing geothermal energy from hot dry rocks, particularly in regions where water resources are scarce. However, only a limited number of studies were carried out in the past to assess the viability of the CO₂ based EGS concept. Most of these studies were theoretical examinations of the heat extraction and exergy analysis under a limited range of operational parameters. In addition, research work on the fluid-rock interactions relevant to CO₂ based EGS is also limited and needs further investigation. The present thesis addresses the above knowledge gaps through a combined experimental and theoretical study, resulting to an accurate description of the entire CO₂ based EGS power generation process encompassing the reservoir, wellbore and power plant cycle as well as the fluid-rock geochemical interaction. The specific objectives of the project underlying this thesis were: (1) model development and simulation of the entire CO₂ and H₂O based EGS and the associated power plant cycles, (2) optimisation studies and sensitivity analysis of operating and design parameters affecting CO₂ and H₂O based EGS performance, (3) performance comparison of CO₂ based EGS and H₂O based EGS under the same operating and reservoir conditions, (4) examination of the effect of reservoir parameters on both CO₂ and H₂O based EGS concepts through detailed 3D reservoir simulations, (5) design and fabrication of a fluid-rock interaction apparatus capable of simulating EGS conditions, and (6) experimental investigation of the fluid-rock interactions at reservoir conditions and its likely impact on the performance characteristics of CO₂ and H₂O based EGS. One dimensional (1D), 2D, and 3D models of integrated reservoir-wellbore-power plant cycle were developed to provide an overall description of fluid flow in fractured reservoir (channel flow) and in radial fluid flow in homogeneous porous media. It was also created to investigate "3D effects" as well as transient changes during the power generation process. The thermosiphon power generation process was used in CO₂ based EGS model simulations while the Organic Rankine Cycle binary plant with isopentane as the working fluid was employed in the H₂O based EGS simulations. Mass and energy balance equations associated with the integrated 1D and 2D reservoir-wellbore-power plant cycle model simulations were solved using the Engineering Equation Solver (EES). In the integrated 3D reservoir-wellbore-power plant cycle modelling, the transient geofluid mass and heat flow rates in the reservoir were simulated using TOUGH2/ECO2N software packages while the wellbore flow and power plant cycle calculations were carried out using EES. The use of TOUGH2/ECO2N was validated and calibrated by replicating the results of prior studies done by Pruess (2008) where TOUGH2/EOSM simulator was used. A fluid-rock interaction apparatus with titanium made wetted components was designed and fabricated to conduct batch and flow-through experimental studies of rock samples with CO₂ and H₂O at pressures up to 50 MPa and temperatures up to 400°C. Surface granite from Moonbi near New England Highway, NSW and drill core samples from Mossgiel 1 and Nambucurra 1 boreholes at Murray-Darling Basin, NSW were collected and used as representatives of hot-dry-rock (HDR) EGS reservoir rock formations. The granite samples were pulverised and analysed for particle size distribution as well aselement (fused-bead XRF) and mineral (Rietveld quantitative XRD) compositions prior to any experiments. Fluid-rock interaction experiments were conducted for up to 15 days at different simulated reservoir pressures (20 and 35 MPa) and temperatures(200 and 250°C). Fluid effluents were analysed using ICP-OES, and the reacted pulverised granite samples were subjected to further XRF, XRD, and SEM (scanning electron microscopy) analysis. The following are the key findings of the integrated 1D/2D reservoir-wellbore-power simulations: • The mass flow rate of CO₂ has an inverse relation with the injection temperature in a CO₂ based EGS while there is a direct relationship between the mass flow rate of water and injection temperature in CO₂ based EGS. These contrasting behaviours can be assigned to the fact that an increase in the injection temperature lowers the CO₂ density and hence increases its dynamic viscosity whereas in the case of water, an increase in the injection temperature lowers H₂O kinematic viscosity and thereby decreases the dynamic viscosity. • Reservoir pressure loss is generally higher for H₂O than for CO₂ because of the higher H₂O kinematic viscosity. • CO₂ overall mass flow rate is higher than that of H₂O due to lower average CO₂ kinematic viscosity at reservoir conditions. • Wellbore frictional loss of CO₂ is greater than that of H₂O due to the lower average CO₂ density along the length of the wellbore. • Heat extraction rates of H₂O based EGS is generally higher than those of CO₂ based EGS due to higher specific heat capacity of water. • The thermal siphoning is not practical for H₂O based EGS because the production pressure is usually lower than the injection pressure. • Power generation output of H₂O based EGS are higher than that of CO₂ based EGS and increases almost linearly as injection pressure increases while CO₂ based EGS power output shows a parabolic trend. These were found to be due to the dependency of CO₂ thermodynamic properties on pressure while H₂O thermodynamic properties are almost independent of pressure. • Reservoir temperature does not influence the overall CO₂ mass flow rate, but CO₂ heat extraction rate increases as reservoir temperature increases due to increase in the specific enthalpy change. • The maximum power generation of CO₂ based EGS decreases as reservoir pressure decreases due to lower CO₂ production pressure in the wellhead. • For both CO₂ and H₂O based EGS, shorter injection to production well distance gives higher fluid mass flow rate due to the increase in pressure gradient (pressure drop/distance) between the injection and production wells. • As the ratio of production to injection well increases, the CO₂ based EGS power generation output increases due to diminishing CO₂ frictional loss in individual production well as the number of production well increases. • CO₂ based EGS generally performs better in low permeability reservoirs (typically one order of magnitude decrease in reservoir permeability decreases CO₂ mass flow rates by 27% while H₂O mass flow rates decreases by 67%). • The overall thermal efficiency at any specified injection and reservoir conditions is constant regardless of CO₂ mass flow rate. Batch and flow-through CO₂-rock interaction experiments show that Ca, Fe, Mg, Al, and Si dissolve in scCO₂, which was found to be partly due to the presence of H₂O in the CO₂ stream leaked from the piston accumulator. Geochemical model simulations show that aqueous Si concentration is in equilibrium with the rock minerals after 1 day exposure in the batch experiment. The log of (Na/K) ratios shows the preferential dissolution of albite over k-feldspar. The SEM image analysis of the treated granites shows signs of erosion (i.e. rounded edges and pebble-like surfaces), which is considered to be due to the formation of carbonates in the surface and its subsequent erosion and dissolution (particularly Na₂CO₃ and K₂CO₃) to the fluid. The XFR analysis of the untreated and treated pulverised granites shows very small changes to SiO₂, Al₂O₃, CaO, MgO, Fe₂O₃, Na₂O, and K₂O major oxide compositions consistent with the ICP-OES analytical results. The Na-K-Mg ternary diagram of the data collected from the fluid-rock experiments shows that the aqueous fluid is far from the equilibrium. The presence and/or absence of minerals (hornblende or chlorite) in the starting material influences the log of (Na/K) ratios. Hornblende alters or converts to chlorite in the CO₂-H₂O mixture. Moreover, the concentrations of Ca, Mg, and Fe decreases with time, which is considered to be due to the formation of a passive layer of insoluble carbonate minerals in the surface, thus preventing further fluid-rock interaction. The pressure loss data collected during the flow-through fluid-rock interaction experiments validate the theory that CO₂ has lower reservoir loss than H₂O. The data also validate the correlation between particle size and intrinsic permeability, which predicts that at the same fluid mass flow rate, a medium with a larger particle size has a lower pressure loss.
- Subject
- thermosiphon; engineered geothermal system; hot dry rock; geosequestration
- Identifier
- http://hdl.handle.net/1959.13/1036104
- Identifier
- uon:13209
- Rights
- Copyright 2013 Alvin I. Remoroza
- Language
- eng
- Full Text
- Hits: 609
- Visitors: 921
- Downloads: 408
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Abstract | 356 KB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Thesis | 11 MB | Adobe Acrobat PDF | View Details Download |