SUMMARY Compressional deformation in fluids and rocks is influenced by similar viscoelastic effects, as in shear case. In this paper we introduce the importance of bulk viscosity and modulus in frequency-dependent response of elastic velocities. We conducted experiments to measure bulk modulus and attenuation of two heavy-oil saturated rock samples by confining pressure cycling method under varying oscillation frequencies (within teleseismic frequency band 0.001 - 1 Hz), and compared these measurements to a more conventional axial stress-strain technique. We plan to extend the frequency range of the pressure cycling apparatus as well as to modify the setup in order to measure frequency-dependent bulk viscosities of viscoelastic fluids. INTRODUCTION The bulk modulus of any pore fluid must be used to interpret the seismic response and perform a fluid substitution for Direct Hydrocarbon Indicator analysis.
We demonstrate that the oscillating pore pressure method is suitable for measuring basic set of rock’s poroelastic properties, such as drained bulk modulus, Biot-Willis and Skempton’s coefficients. The oscillating pore pressure method has been initially developed as a method to determine hydraulic properties of rocks (Kranz et al., 1990; Fischer, 1992). We equip our samples with strain gages and subsequently measure deformation, caused by a harmonic pore pressure pulse, propagating through sample’s porous space. By doing so we simultaneously determine poroelastic and hydraulic properties. One of the main goals of the poroelastic measurement is to improve the estimation of the storage capacity from purely hydraulic pore pressure oscillating method. We show that the storage capacity values from the oscillating pore pressure method alone are overestimated by the order of magnitude; measured poroelastic properties for the tested samples are in agreement with literature data. Hence, oscillating pore pressure method is capable of accurately measuring poroelastic properties.
The oscillating pore pressure method for measurements of permeability and storage capacity has been previously presented in details (Kranz et al., 1990; Fischer, 1992; Song and Renner, 2007). It is based on inducing time-harmonic diffusive fluid flow through the porous sample and subsequent measurement of the upstream and downstream pressure signals. The propagated through the specimen pressure sinusoid is always phaselagged and attenuated on the downstream side. Once the amplitude ratio and phase has been calculated, we can solve the diffusivity equation with oscillatory boundary conditions and estimate the permeability and storage capacity of the sample.
Although it allows accurate determination of permeability, the estimation of the storage capacity yields large errors, sometimes on the order of several magnitudes (Bernab´e et al., 2006). As mentioned by Bernab´e et al. (2006) the values of the storage capacity are particularly uncertain if the experiment employs large downstream reservoirs. This is usually done in case of high permeability samples (100 mD and higher) in order to achieve measurable phase and amplitude ratio of the two pressure signals.
The storage capacity of the sample is a poroelastic constant, responsible for fluid storage in the sample and defined as (Wang, 2000; K¨umpel, 1991):
Naturally occurring gas hydrates contain significant amounts of natural gas which might be produced in the foreseeable future. Thus, it is necessary to understand the pore-space characteristics of hydrate reservoirs, especially the pore-scale distribution of hydrates and their interaction with the sediment. The goal of this study is to determine how the hydrates are distributed in the pore space and the implications of this porescale distribution for hydrate saturation estimates from seismic and acoustic velocities.
Laboratory measurements were conducted to obtain information about the distribution of hydrates in the pore space of synthetic sediments (glass beads). Tetrahydrofuran (THF) was used as a guest molecule since THF hydrate is a proxy for naturally occuring hydrate. We performed micro X-ray computed tomography (MXCT) on laboratory formed glass-bead samples. MXCT images indicate that THF hydrates form in the pore space with little to no contact to the grain surfaces. We observed salt precipitation at grain contacts and in small pore space. These hydrate-bearing sediments appear to follow a pore-filling model but contained salt cement. Based on this knowledge, it may be possible to calibrate seismic and well logging data to calculate the amount of natural gas stored in a hydrate reservoir. This information will help to make decisions regarding the producibility of methane hydrates and to develop safe production schemes.
Gas hydrates are clathrate structures of natural gases. They require low temperatures and high pressures for stability. These conditions are met in shallow sediments in Arctic permafrost regions and beneath the seafloor along continental slopes. The estimated amount of natural gas, mainly methane, stored in hydrate reservoirs exceeds the amount of natural gas stored in conventional resources by at least one order of magnitude (Meyer, 1981; Dobrynin et al., 1981; Collett et al., 2009). The widespread occurence of gas hydrates in permafrost and shallow marine sediments is well established (Collett et al., 2009). Anderson et al. (2008) and Dallimore et al. (2008) demonstrated that gas-hydrate production can be developed with existing oil and gas production technology. For successful production of methane gas from hydrate reservoirs we need to obtain knowledge about physical properties of gas-hydrate bearing sediments.
The most common geophysical methods used to characterize and quantify gas hydrates in nature are seismic surveying and well logging. In order to calibrate and interpret these field measurements, laboratory studies are necessary to determine the bulk physical properties of hydrate-bearing sediment. Currently, it is possible to predict the existence of gas hydrates from geophysical measurements. However, the techniques used to estimate hydrate saturation based on either seismic data or well logs require further development (Collett and Lee, 2012). Thus, the amount of hydrates stored in a reservoir remains uncertain in some cases. Information about the distribution of gas hydrates in the rock is necessary to determine hydrate saturation.
The Thomsen anisotropy parameters are used in seismic processing, in-situ stress estimation, and to describe materials in general. Here we provide a basic overview of potential risks for laboratory measurements on transversely isotropic materials. Then we apply an inversion scheme on data from the literature to determine the most representative sets of Thomsen anisotropy parameters. We interpret the results in terms of the Thomsen parameters e and d for typical Thomsen parameter relationships.
An oscillatory pore pressure method for simultaneous measurements of rock transport properties, such as intrinsic permeability and specific storage capacity, and summarize a laboratory setup, being developed for these purposes. The pore pressure pulsing method has been described before by many researchers, however we attempt to examine the relationship between a rock’s transport properties and oscillation parameters, such as amplitude and frequency.
Liberatore, Matthew W. (Colorado School of Mines ) | Herring, Andrew M. (Colorado School of Mines) | Li, Kejing (Colorado School of Mines) | Bazyleva, Ala (Colorado School of Mines) | Akeredolu, Babajide (Colorado School of Mines) | Prasad, Manika (Colorado School of Mines) | Batzle, Michael (Colorado School of Mines)
The goal of this project was to improve recovery of Alaskan North Slope (ANS) heavy oil resources in the Ugnu formation by improving our understanding of the formation's vertical and lateral heterogeneities via fluid and rock characterization. Although the reserves of heavy oil on the North Slope of Alaska are enormous (estimates are up to 10 billion barrels in place), difficult technical and economic hurdles must be overcome to produce them. The Ugnu formation contains the most viscous, biodegraded oils and standard production methods are ineffective.
Heavy oils are viscoelastic fluids. Thus, it is critical that we understand the properties of the heavy oils we are trying to produce before the geophysical model and modeling plan can be completed. The Ugnu oils (including more than 18 oil, oil/sand, oil/water, and oil/sand/water mixtures) exhibited non-Newtonian characteristics, including shear thinning and a non-zero shear modulus. The complex viscosity of the dead oils has been found to be as high as 7,000 Pa?s and a shear modulus at -10oC above 10,000 Pa (and frequency dependent). A complete set of "live?? oil rheology experiments were completed. A large range of temperatures (-10 to 60oC) and pressures (15 to 2000 psi) were controlled and viscosity measured in novel high-pressure rheology setup.
Saturate-Aromatic-Resin-Asphaltene (SARA) fractions have been measured on site and by an outside laboratory. The SARA technique has large experimental variation when used to measure heavy oils. Asphaltene content varied from 3 to 9% in the same sample measured by CSM and an outside laboratory. A large number of experiments have been completed, including molecular beam mass spectroscopy (MBMS), optical and scanning electron microscopy, and other techniques not reported here. Chemistry signatures from the MBMS and SARA have been correlated with the viscosity of the heavy oils.
Whether hydrate-bearing sediments are formed in nature or in the laboratory, the manner of gas hydrate generation will influence the hydrate habit and distribution, and consequently impact the bulk physical properties. In the last couple years we have begun to better understand the mechanical interactions between hydrate and sediment and how they are controlled by the hydrate formation method. We have developed a conceptual picture of hydrate formation in sediment and how it affects the elastic properties of the hydrated sample. The theory was based on a series of experiments conducted on laboratory-formed gas hydrate-bearing sediments. Ultrasonic velocities were measured in conjunction with MRI in hydrate-bearing Ottawa Sand F110 during hydrate formation and dissociation. P- and s-wave velocities were determined as a function of gas hydrate saturation. Hydrates were formed out of solution using Tetrahydrofuran (THF) and through methane injection into partially water-saturated samples. For the latter, samples with low and high initial water saturation (Swi) were tested. The recorded velocities exhibited a noticeable dependence on Swi. At low saturations (~20%) the hydrate in the sediment acted cementing and increased the ultrasonic velocities dramatically. The final velocities, however, decreased for increasing initial water saturations. Even small changes in the initial water saturation resulted in significant changes in velocities. At high initial water saturations (~80%), the velocity increased linearly with increasing hydrate content even at very low saturations. This behavior differed from the one observed for hydrate formed from out of solution. Ultrasonic velocities recorded during THF hydrate dissociation in sediment did not increase until a critical saturation of 30-35 percent was exceeded. Comparison with model data calculated using the effective medium theory indicated that hydrates formed from a free gas phase and low and high Swi act cementing and load-bearing, respectively. On the other hand, hydrates formed out of solution are pore-filling below and load-bearing above a critical hydrate saturation of about 35-40 percent. This was corroborated with micro X-ray computed tomography.
Adam, Ludmila (Boise State University) | Otheim, Thomas (Boise State University) | van Wijk, Kasper (Boise State University) | Batzle, Michael (Colorado School of Mines) | McLing, Travis L. (Idaho National Laboratory) | Podgorney, Robert K. (Idaho National Laboratory)
Lamb, Andrew P. (Boise State University) | van Wijk, Kasper (Boise State University) | Liberty, Lee (Boise State University) | Revil, Andre (Colorado School of Mines, Dept of Geophysics) | Richards, Kyle (Colorado School of Mines, Dept of Geophysics) | Batzle, Michael (Colorado School of Mines, Dept of Geophysics)