Miscible oil-based-mud (OBM) filtrate contamination poses a major challenge to the acquisition of representative fluid samples using wireline formation testers (WFTs). A sound understanding of the OBM-filtrate cleanup process and identification of first-order impact parameters are of paramount importance for the design of next-generation WFT probes that can operate in OBM-filtrate environments with enhanced efficiency.

We have constructed a numerical model for OBM-filtrate cleanup using an equation-of-state (EOS) compositional fluid-flow simulator. The numerical cleanup model honors the physics of multicomponent-fluid flow and the thermodynamics of phase behavior. Simulation results exhibit close agreement with analytical predictions and with field data for the time dependence of contamination during sampling. First-order impact parameters were identified through a sensitivity study using the numerical model. It has been found that the clean-up function is predominantly governed by permeability anisotropy, porosity, cleanup flowrate, viscosity ratio, depth of invasion, distance between the WFT probe and a sealing boundary, formation thickness, and wellbore radius. A response-surface-based contamination model (RSCM) was developed using the above-described simulation investment with additional runs. RSCM constitutes a rapid approximate model and can serve as a prejob-planning or real-time-analysis tool. Our simulation and rapid-modeling results compare well with empirical observations made in the field. In particular, the rate of change of miscible contamination with time has been found to vary between t0.3 and t0.6, with t0.45 representing a good average value. For the first time, modeling has been shown to give essentially the same results as empirical observations.


In the development of deepwater prospects and other capital-intensive exploration and production projects, understanding the nature of hydrocarbon fluids in terms of chemical and physical properties, phase behavior, spatial distribution, and hydraulic and thermodynamic communication is of critical importance. Fit-for-purpose design of completions and production facilities and optimal planning of reservoir production strategies strongly depend on the adequate characterization of the physical and chemical properties of the fluids. In many deepwater and other high-cost wells, WFT fluid samples may be one of the few sources of fluid properties reliable enough for economic screening. Therefore, it is imperative that representative high-quality WFT samples be collected early in any exploration or appraisal campaign.

Sampling in wells drilled with OBM presents special challenges. Unlike water-based-mud (WBM) filtrate, OBM filtrate is miscible with the in-situ hydrocarbons and cannot be physically separated from the crude oil subsequent to sampling. Mathematical decontamination is used instead. Nevertheless, there exists a maximum allowable level of contamination for such methods to work. This level depends on the type of involved hydrocarbons and the OBM. For instance, it is challenging to back out the original viscosity and saturation pressure of black-oil samples contaminated at levels greater than 10% by weight in the live fluid (Elshahawi et al. 2008). A reliable extrapolation to original crude-oil properties requires accurate determination of the level of contamination. This is easier to obtain for black oils than for condensates or biodegraded fluids. Yet, the process still can be hampered by the highly variable quality of laboratory analyses (Mullins and Schroer 2000). Our preferred approach is to ensure that contamination is reduced to acceptably low levels during sampling and acquire samples that require little parameter extrapolation. As for any other optimization problem, however, maximizing sample quality is not absolute. Instead, it is subject to a number of constraints. There are practical limits on how long a sampling station may last. For instance, rig time for some offshore and deep wells can be very costly, and the probability of tool sticking generally increases with longer pumping times. As such, if a sample is collected much later than necessary, valuable rig time may be wasted. On the other hand, under certain circumstances, contamination levels may remain high even after extended pumping because of the miscible nature of OBM-filtrate invasion. In some cases, no realistic clean-out time will result in contaminations that are acceptably low. To be effective, a decision to continue or discontinue such a sampling station must be decided early on during the sampling process. So when is the right time to sample? There can be no unique answer to this question even on a well-by-well basis. The optimal length of time to adequately cleanup a sample is dependent on a number of fluid and rock properties that will undoubtedly vary from one station to another. These include the rate of cleanup, the extent of filtrate invasion, and a large number of rock, fluid, and geometrical parameters. The optimization must, therefore, take place on the spot, in real time. Elshahawi et al. (2007) suggests that there is no replacement for real-time monitoring and control. Our objective in this work is to study the physics of the sampling process, understand the main controls of sample cleanup, and use this understanding to enhance our ability to implement real-time monitoring and control.

This content is only available via PDF.
You can access this article if you purchase or spend a download.