|Kinetic Parameters for the Exchange of Hydrate Formers
||Last Reviewed 6/3/2016
This project will investigate the kinetics associated with producing natural gas from hydrate bearing geologic media via unconventional technologies. The principal production technology of concern for this research will be that of exchanging carbon dioxide (CO2) and nitrogen (N2) with clathrated methane (CH4). The so called guest-molecule exchange technology is an attractive technology from the perspective of its potential for maintaining the geomechanical integrity of the reservoir formation, in addition to its carbon neutral potential. As with other unconventional technologies, its ability to produce natural gas will depend on our understanding of the processes and our ability to exploit this understanding. The approach taken during this project will be to understand the kinetic mechanisms that control the exchange of hydrate formers using numerical simulation to interpret field-scale trials, laboratory experiments to determine kinetic parameters, and code comparison to verify mathematical models and solution schemes.
Pacific Northwest National Laboratory (PNNL), Richland, WA
The numerical simulation component of the project involves the development and application of the computer code STOMP-HYDT-KE to investigate the production of geologic accumulations of natural gas hydrates via depressurization, thermal stimulation, inhibitor injection, or guest-molecule swapping. The STOMP-HYDT-KE computer code solves conservation equations for energy; water mass; mobile N2, CO2, and CH4; clathrated N2, CO2, and CH4; and salt or inhibitor mass. Solving conservation equations for both mobile and clathrated hydrate formers (i.e., N2, CO2, and CH4) allows the code to mode swapping process as being kinetic. Whereas systems of interest have been generally limited to those involving aqueous-gas-hydrate phases, STOMP-HYDT-KE additionally considers nonaqueous liquid, ice, and precipitated salt phases. The STOMP-HYDT-KE also has capabilities for solving coupled geochemistry problems via its ECKEChem module, giving the code thermal-hydrological-geochemical (THC) capabilities. An element of the project will be to extend these capabilities to include integrated geomechanical capabilities, allowing the simulator to model changes in the geomechanical state of the hydrate bearing reservoir throughout the production process.
Important issues regarding reservoir stimulation techniques, safety, and cost must be addressed before large-scale commercial recovery of natural gas from hydrates can be attempted. Reservoir modeling is an important tool for addressing these issues; however, applying this modeling tool requires access to reliable thermodynamic, kinetic, and physical property data for gas hydrates and physicochemical properties of the hydrate-bearing sediments themselves. Laboratory studies to characterize gas chemistries of synthetic gas hydrate sands have been conducted using a residual gas analyzer. The proposed experiments will leverage the results of previous experiments where the gas composition within a synthetically rich methane hydrate core was successfully monitored, allowing the use of acquired hydrate dissociation kinetics. Furthermore, proposed measurements using the pressurized x-ray diffraction technique are unique and will be some of the first reported. This technique was successfully used to track mineral dissolution, carbonation reactions, and mineral volume changes. The types of structural information gained from this technique are believed to improve fundamental understanding of mechanisms that occur during the gas swapping process.
Budget Period 1 - During the first budget period, the project team will investigate the kinetics of exchanging CO2 and N2 with clathrated CH4 in hydrate-bearing geologic media. The project comprises two distinct components: (1) numerical investigation of the 2012 Iġnik Sikumi gas hydrate field trial, and (2) experimental investigation of kinetic exchange processes in laboratory-scale hydrate-bearing unconsolidated sands. The principal objective of the numerical component will be to provide an interpretation of the data gathered at Iġnik Sikumi Well #1. The experimental component of this project is designed to provide kinetic exchange parameters needed for the numerical simulation. The principal objective of the two experiments is to provide an order of magnitude value to the kinetic exchange parameters for the field-scale simulations of the Iġnik Sikumi gas hydrate field trial.
Budget Period 2 - The project team will expand the investigations of the first budget period during the second budget period. The team will continue to run numerical simulations of the 2012 Iġnik Sikumi gas hydrate field trial in order to resolve disagreements between simulation results and field trial observations and provide a more thorough interpretation of the field results. Laboratory experiments designed to provide kinetic parameters under controlled conditions will continue, and a code comparison study focused on expanding the International Hydrate Code Comparison Study to problems involving gas hydrates of mixtures of CH4, CO2, and N2 hydrate formers were proposed. West Virginia University is preparing a suite of problems for the code comparison study, which involves hydrates of pure components and component mixtures of CH4, CO2, and N2. The laboratory experiments and code comparison study are currently unfunded.
Budget Period 3 – The project team will change the focus of the investigations from the kinetics of exchanging CO2 and N2 with clathrated CH4 in hydrate-bearing geologic media to geomechanical processes associated with producing natural gas hydrates. The results of the Nankai Trough experiment indicate the critical importance of understanding the geomechanical processes in producing natural gas hydrates from suboceanic deposits. During the third budget period, the work will be focused on developing fully coupled capabilities for simulating the deformation of the reservoir and overlying strata with changes in effective stress from changes in pressure and temperature. The geomechanical modeling capabilities will be limited to linear elasticity, but will include the ability to predict mechanical failure through a Mohr-Coulomb criterion. Other operational modes of the STOMP simulator have realized geomechanical capabilities via coupling with the Abaqus simulator, but this approach creates two immediate limitations: 1) the geomechanical component can not be directly altered by the STOMP development team, and 2) the coupled codes can not be converted to a full parallel implementation for execution on distributed memory computers.
The conventional technologies for producing natural gas hydrates from geologic repositories—especially those with pore-filling type hydrates—are reasonably well understood, and numerical simulations have been compared against field trials. In contrast, the guest-molecule-exchange approach for natural gas hydrate production is emerging unconventional technology. Laboratory-scale experiments by ConocoPhillips and researchers at the University of Bergen, Norway, have demonstrated the exchange of CO2 with clathrated CH4, but there have only been a limited number of numerical simulation investigations of the technology. This project provides an opportunity for a recently developed numerical simulator, STOMP-HYDT-KE, to be used to aid in the interpretation of data collected from the 2012 Ignik Sikumi gas hydrate field trial. The ultimate objective of this field of research is to develop numerical simulation tools capable of predicting the performance of the guest-molecule-exchange technology at the reservoir scale, including the geomechanical stability of the process. The work represents a first step in validating a numerical simulator capable of modeling the kinetic exchange of hydrate guest molecules. A credible interpretation of the Ignik Sikumi gas hydrate field trial, realized through numerical simulation, will greatly increase understanding of the fundamental exchange processes for hydrate formers.
Support from U.S. DOE/NETL and KIGAM have yielded simulation capabilities in the STOMP-HYDT-KE simulator that allow for the modeling of fully coupled multifluid hydrologic, heat transfer, hydrate thermodynamics, and geochemistry. Moreover, the simulator is formulated to model the exchange of hydrate formers, hydrate dissociation, and hydrate formation as kinetic processes for a ternary hydrate former system N2, CO2, and CH4. The missing element in this suite of capabilities is the coupling with geomechanics; where, changes in pore pressure and temperature yield changes in effective stress, resulting in rock deformation or failure. These deformations or changes in stresses in turn yield changes in porosity and intrinsic permeability, which directly impact the hydrologic system. The proposed work will allow for the coupling to be integrated into a single simulator with capabilities for execution on sequential, shared-memory parallel, and distributed-memory parallel computers. Kinetic hydrate simulations are computationally expensive and coupling geomechanics adds to that expense, which makes parallel computing a necessity to realize problem solutions to real-world problems at reservoir scales.
The goal of this experimental work is to conduct measurements of methane hydrate dissociation and structural stability in hydrate-bearing sediments using a high-pressure cell and state-of-the-art analytical techniques. The kinetic exchange rates obtained on the ternary gas system will be utilized to validate numerical codes, and the structural data will further support the concept of continuous stability of gas hydrate structures during gas swapping.
The STOMP-HYDT-KE computer code is a reservoir-scale numerical simulator with both sequential and parallel implementations. Parallelism is via OpenMP, using either the PETSc or Lis linear system solvers. STOMP-HYDT-KE has been applied against the 2012 Ignik Sikumi #1 gas hydrate field trial (White and Lee, 2014; Anderson et al., 2014), and against a series of flow-through CH4 replacement experiments (Yonkofski et al., 2016). In addition to this later publication, Catherine Yonkofski’s poster, entitled “Methane Hydrate Production via Carbon Dioxide Storage,” was selected for presentation at the C3E Women in Clean Energy Symposium (http://c3eawards.org/symposium), to be held at Stanford University on May 31, 2016, with a theme of “The Role of Women Internationally in Decarbonizing our Energy Future.” During the flow-through experiments, kinetic rate parameters were determined via inverse modeling against the experimental observations.
A series of guest-molecule exchange experiments have been conducted involving the replacement of mixtures of N2 and CO2 with clathrated CH4 under different temperature and pressure conditions. The target pressure and temperature conditions varied between being within and outside the stability zone for the N2 and CO2 mixture, but always within the stability zone for pure CH4 hydrate. The first study targeted under this task was the development of a standardized procedure to perform in situ monitoring of pore gas chemistry during the replacement of methane in a CH4 hydrate bearing porous sand with CO2 through the titration of a gaseous mixtures consisting of different ratios of N2/CO2. The continuous monitoring of pore gas chemistry would provide clear evidence of the rates associated with the exchange of CH4 with CO2. The experimental procedure involves three main stages: 1) the formation of CH4 hydrate in a porous sandstone, 2) replacement of the core gas with a N2/CO2 gas mixture, and 3) the monitoring of the core gas chemistry over time during the exchange process. The goal of these scoping experiments is to develop kinetic exchange rates and parameters for use in the simulations conducted under the Iġnik Sikumi History Match task.
Current Status (June 2016)
In 2012, U.S. DOE/NETL; ConocoPhillips Company; and Japan Oil, Gas, and Metals National Corporation jointly sponsored the first field trial of injecting a mixture of N2/CO2 into a CH4 hydrate-bearing formation beneath the permafrost on the Alaska North Slope. Known as the Ignik Sikumi #1 Gas Hydrate Field Trial, this experiment involved three stages: 1) the injection of a N2/CO2 mixture into a targeted hydrate-bearing layer, 2) a 4-day pressurized soaking period, and 3) a sustained depressurization and fluid production period. Data collected during the three stages of the field trial were made available after a thorough quality check. The Ignik Sikumi #1 data set is extensive, but contains no direct evidence of the guest-molecule exchange process. This study uses numerical simulation to provide an interpretation of the CH4/CO2/N2 guest molecule exchange process that occurred at Ignik Sikumi #1. Simulations were further informed by experimental observations. The goal of the scoping experiments was to understand kinetic exchange rates and develop parameters for use in Iġnik Sikumi history match simulations. The experimental procedure involves two main stages: 1) the formation of CH4 hydrate in a consolidated sand column at 750 psi and 2°C and 2) flow-through of a 77.5/22.5 N2/CO2 molar ratio gas mixture across the column. Experiments were run both above and below the hydrate stability zone in order to observe exchange behavior across varying conditions. The numerical simulator, STOMP-HYDT-KE, was then used to match experimental results, specifically fitting kinetic behavior. Once this behavior is understood, it can be applied to field scale models based on Ignik Sikumi #1. A poster documenting this work was presented at the 2015 AGU Fall Meeting, Ruprecht, C.M., J.A. Horner, and M.D. White. 2015. “Experimental and Numerical Investigation of Guest Molecule Exchange Kinetics based on the 2012 Ignik Sikumi Gas Hydrate Field Trial." Abstract submitted to AGU Fall Meeting, San Francisco, CA. PNNL-SA-112587.
A flow-through CH4 replacement experiment was performed in a sand column at conditions similar to those at Iġnik Sikumi #1 to collect and validate hydrate formation and exchange parameters (slb, Kf, Ke, fi) used in the development of the STOMP-HYDT-KE model. The experiment was performed in three phases: 1) hydrate formation for 40 hours, 2) guest molecule exchange with 77.5/22.5 N2/CO2 molar ratio at 0.38 ml/min titration for 100 hours, and 3) rapid gas vacuum and thermal dissociation for 18 hours. Data were collected through a residual gas analyzer (RGA), an inlet flow syringe pump, an outlet pressure transducer, and eight temperature probes across the column.
The experiment was conducted using a custom built cylindrical, semi-transparent HYDEX column (1.9 cm inner diameter; 20.4 cm length) equipped with a custom built 8-point thermocouple (OMEGA Engineering Inc.) inserted in the center of the column. The thermocouple was designed with a single probe hosting eight temperature sensors set with 2.5 cm spacing across the length of the column. Column pressures were monitored using a 1,500 psi pressure transducer (PX409-USBH Series, OMEGA Engineering Inc.) attached to the downstream end of the outlet port. The assembled column was sealed, wrapped with copper tubing, and placed horizontally in a custom built foam cooler. The column was then pressurized at room temperature to 1,215 psi with CH4 and allowed to equilibrate for a sufficient time for the methane to fully saturate the aqueous phase. The cooler was then packed with ice and the column temperature was cooled to ~2°C by circulating chilled fluid though the copper tubing.
Project Start: June 1, 2013
Project End: September 30, 2018
Project Cost Information:
All DOE Funding
FY13 DOE Share: $90,000
FY14 DOE Share: $80,000
FY15 – DOE Share - $50,000
Total Funding to Date: $220,000
NETL – John Terneus (John.Terneus@netl.doe.gov or 304-285-4254)
PNNL – Mark White (email@example.com or 509-372-6070)
Quarterly Research Progress Report [PDF-315KB] October - December, 2014
Quarterly Research Progress Report [PDF-186KB] July - September, 2014
Quarterly Research Progress Report [PDF-156KB] January - March, 2014
Quarterly Research Progress Report [PDF-418KB] October - December, 2013
Quarterly Research Progress Report [PDF-210KB] July - September, 2013