Edmunds, Neil Roger (Laricina Energy Ltd)
Solvent-additive processes (SAP) are a promising, but challenging technology. Perhaps the biggest challenge from an engineering point of view, is that simulators probably work some of the time, but not all of the time; and there is no information about where the line between occurs, or what the correct answer should be, after the line is crossed. Other serious problems are the many degrees of freedom in SAP process design, and the non-linear relationships between process inputs and economic results. There are too many possible designs to try randomly for even a single reservoir, and there is limited theory to interpolate or scale available experimental data.
This paper attempts to assemble some known pieces of the puzzle, and to explore how they may fit together to explain and predict SAP performance characteristics
First, some familiar PVT relationships are presented, with examples using temperature as the independent variable. This helps to clarify the choice of solvent, as a function of reservoir pressure, and also to understand the effect of the increasing solvent "dose??. It is shown that SAP will create a double front, one where the water is condensed, and a second where the solvent is absorbed by, and drains with, the oil. A vapor blanket separates the two fronts.
Secondly, simple estimates are given for the temperature distribution in the vapor blanket (i.e. solvent-active zone). Together with PVT data for the same pressure, these allow the thickness of a vapor blanket to be estimated.
Finally, SAP mass transport limits are considered, by observing that the second front essentially constitutes VAPEX. The Butler-Mokrys theory is discussed, in view of its failure to predict certain experimental results; it is argued that this results from neglect of capillary pressure effects, which in fact are dominant at the front. A purely empirical correlation by Nenniger is introduced, which can be rearranged to predict the maximum solvent speed, also as a function of temperature.
With the decrease in conventional oil and gas reserves throughout the world and an ever-increasing demand for fossil-fuel based energy and resulting high oil prices, focus has been shifting to unconventional and heavy oil and bitumen. Grosmont carbonates in Northern Alberta have been estimated to contain at least 300 billion barrels of heavy oil or bitumen. However, recovering this oil is extremely difficult because of the complexity associated with carbonate reservoirs in general, e.g., the Grosmont unit is known to possess a triple porosity system-matrix, fractures and vugs, based on core studies. The second problem is the fluid itself, which is highly viscous bitumen, immobile at reservoir conditions. To extract this bitumen from very heterogeneous carbonate rock, both heat and dilution using solvents may be needed. This paper reports the results and analysis of hot solvent experiments conducted on original Grosmont carbonate cores.
The Vapor Extraction (Vapex) process and its many hybrid variants have attracted a great deal of attention as potentially less energy intensive alternatives for exploiting heavy oil and bitumen resources. However, despite much work over the past two decades, uncertainty remains about the basic mechanisms, the scaling aspects and the most appropriate methods of numerically simulating these processes. This paper offers some insights into several of these outstanding questions. The questions are examined in the context of an extensive and well-documented set of Vapex experiments carried out by Maini and his colleagues over the past 10 years.
We have experimented with different methods of simulating these experiments using a physics-based reservoir simulator. Despite the high permeability (greater than 200 Darcys in all of the experiments), we find that capillary pressure plays a significant role in the drainage. The simulations suggest that most of the drainage takes place in the capillary transition zone along the edge of the vapor chamber, rather than in the single-phase zone ahead of it which has not yet been contacted by vapor.
It has been emphasized in the literature that the near-linear scaling of oil rate with height observed in the experiments is dramatically different from the square root of height dependence predicted by the original analytic model of Vapex. However, the experiments also show an increasing solvent fraction in the produced oil phase as height increases. When this "solvent mixing?? effect is separated out of the rates, the remaining height dependence is less than linear, though still greater than square root of height.
The relative roles of molecular diffusion and mechanical dispersion in Vapex have been widely discussed in the literature. Generally, mechanical dispersion is expected to play a larger role in these high permeability experiments (vis-à-vis the field), due to larger fluid velocities. We present a method of inferring the diffusion/dispersion present in the simulations, despite a hidden component of numerical dispersion caused by the numerical method itself. We find that the experiments are well matched with values of diffusion and dispersion in line with literature correlations, and that the contribution of mechanical dispersion is perhaps not as large relative to that of molecular diffusion as might be expected.
The paper also provides some thoughts on questions we believe are still unanswered, including mechanisms behind the height dependent mixing phenomenon and the scaling of the experimental results to the much greater heights and lower permeabilities characteristic of the field.
Thermal and miscible methods are commonly used for in-situ recovery of heavy oil and bitumen. Both techniques have their own limitations and associated shortcomings, often times yielding an inefficient process. The most common thermal method is steam injection, which is highly energy intensive. Steam generation costs and water production affect the economics of the thermal technique adversely. On the other hand, miscible methods are energy effective but their economics depends on the solvent retrieval. Various combinations of these two techniques such as co- or alternate injection of steam and solvent have been proposed as a solution, but no optimum method has yet been developed.
Thermal and miscible methods can be combined by co-injecting solvent with steam or injecting solvent into a pre-heated reservoir. Current work was undertaken to study the performance of solvents at higher temperatures for heavy oil/bitumen recovery. Glass bead packs and Berea sandstone cores were used in the experiments to represent different types of pore structures, porosity and permeability. After saturating with heavy oil, the samples were exposed to the vapor of paraffinic solvents (propane and butane) at a temperature above the boiling point of the solvent, and a constant pressure of 1500 kPa. A mechanical convection oven was used to maintain constant temperature across the setup. The setup was designed in such a way that a reasonably long sample (up to 30 cm) can be tested to analyze the gravity effect. The oil recovered from each of these experiments was collected using a specifically designed collection system and analyzed for composition, viscosity and asphaltene content.
The amount of oil recovered in each case was also analyzed and the quantity and nature of asphaltene precipitated with each of the tested solvents under the prevailing temperature and pressure of the experiment was reported. Optimal conditions for each solvent type were identified for the highest ultimate recovery. It was observed that recovery decreased with increasing temperature and pressure of the system. It was also noticed that butane diluted the oil more than propane which resulted in lower asphaltene content and viscosity of oil produced with butane as a solvent.
Numerical simulations are widely used to predict and history match performance of reservoirs operating under various recovery processes. Simulators can offer quick answers however the answer should be validated against experimental/field data to verify its reliability. Many studies have been conducted to validate SAGD simulation results, but applying thermal simulators to solvent processes with confidence requires more studies and comparisons.
This paper will discuss the following topics on simulation of solvent process:
1. Comparing simulation results against the Nenniger and Dunn correlation on warm solvent drainage process
2. Identifying the key input parameters(s) in solvent simulation
3. Effects of grid block sizes on the results of solvent simulation
Nenniger and Dunn correlation (2008) provides good estimates of laboratory scaled simulation. Nenniger and Dunn recently mentioned the importance of rate limiting step and thickness of the solvent contact layer in the solvent drainage process. In order to understand the sensitivities of solvent simulation to the rate limiting step and thickness of the solvent contact layer, input parameters and grid block sizes were modified to observe their effects on the process.
This paper describes the results of a simulation and optimization study on the application of a solvent-additive steam-assisted gravity drainage (SAGD) process to reservoirs with associated basal water. A review of related studies is provided, together with a discussion of the pros and cons of potential alkane solvents in basal water reservoirs. Economics and the impact of dynamic and ultimate retention are discussed.
A general conclusion drawn from literature is that optimal solvent application to SAGD in reservoirs will likely involve time variations in both rate and composition of the solvent. This is also the case for reservoirs with associated basal water. This results in an optimization problem that has a large number of dimensions, and is very nonlinear. Genetic algorithms, which mimic biological evolution, have been found to be extremely effective in addressing such problems.
A key product of this effort, optimized for a simple clastic reservoir with associated basal water, is presented. The study produced an operable process, which could be described as a new combination of pre-existing concepts. The process offers material improvements in thermal bitumen supply costs, as well as recovery factor. Major reductions in the physical steam/oil ratio (SOR), (and therefore) capital intensity, water use and carbon emissions, are indicated.