Application of the Split Hopkinson Resonant Bar Test for Seismic Property Characterization of Hydrate-bearing Sand Undergoing Water Saturation

Abstract

Application of the Split Hopkinson Resonant Bar test for seismic property characterization of hydrate-bearing sand undergoing water saturation Seiji Nakagawa and Timothy J. Kneafsey Lawrence Berkeley National Laboratory, Berkeley, California, USA ABSTRACT: Conventional resonant bar tests allow the measurement of seismic properties of rocks and sediments at low frequencies (several kilohertz). However, the tests require a long, slender sample which is often difficult to obtain from the deep subsurface and weak and fractured formations. We present an alternative low-frequency measurement technique to the conventional resonant bar tests. This technique involves a jacketed core sample placed between a pair of long, metal extension rods with attached seismic source and receiver—the same geometry as the split Hopkinson pressure bar test for large-strain, dynamic impact experiments. Because of the added length and mass to the sample, the resonance frequency of the entire system can be lowered significantly, compared to the sample alone. The proposed “Split Hopkinson Resonant Bar (SHRB)” test is applied in two steps. In the first step, extension and torsion-mode resonance frequencies and attenuation of the system are measured. Then, numerical inversions for the compressional and shear wave velocities and attenuation are performed. We initially applied the SHRB test to synthetic materials (plastics) for testing its accuracy, then used it for measuring the seismic velocities and attenuation of a rock core containing supercritical CO 2 , and a sediment core while methane hydrate formed in the pore space. 1. INTRODUCTION Seismic properties of fluid-filled, porous geological materials can be frequency-dependent, thus laboratory measurements should be performed at field-employed frequencies. Unfortunately, laboratory measurement of low- frequency (<10 kHz) seismic properties is a difficult task with traditional ultrasonic equipment and limited core size. Quasi-static techniques (e.g., [1]) allow measurements of low-frequency seismic properties continuously up to several hundreds of hertz, but such measurements are difficult and not commonly performed. Conventional resonant bar tests allow measuring seismic properties of rocks and sediments at relatively low frequencies (typically hundreds of hertz to several kilohertz) relatively easily. One disadvantage of this technique, however, is that the test requires a long, slender sample, which is often difficult to obtain from deep rock /sediment and weak and fractured formations. If the conventional resonant bar test is applied to 5− 10 cm-long cores typically available from such formations, the resonance frequency can be as high as tens of kilohertz. Alternatively, a composite bar, consisting of a metal extension rod (including a seismic source), jacketed rock or sediment core sample, and another metal rod (including a seismic receiver), can be used to reduce the resonance frequency of the system. The reduction in the resonance frequency increases for longer length and larger mass of the extension rods. (Tittmann [2] proposed the use of additional mass in resonant bar tests to reduce resonance frequencies.) Because this sample geometry is the same as the split Hopkinson pressure bar test used primarily for large-strain, dynamic impact experiments, we call this technique the “Split Hopkinson Resonant Bar (SHRB)” test. In the following, we will first describe the testing methodology of the SHRB test, then our experimental setup and the measurement procedures. The principles of numerical inversion for elastic moduli and seismic velocities and attenuation from the laboratory measurements will also be discussed, along with necessary corrections to the model for removing measurement artifacts. Following the description of the method, we will present applications of the SHRB test in the laboratory. First, we will show the accuracy of the technique when applied to synthetic materials (plastics) having known material properties. Then we will apply the technique to a partially water saturated sediment core injected with methane gas to form methane hydrate within the pore space, and a subsequent water flood of the sample. Finally, we will summarize the strengths and weakness of the SHRB test and make recommendations for its use and further development. 2. METHOD 2.1. General Methodology The SHRB test is applied in two steps. First, laboratory measurements are conducted for resonance frequencies and attenuation of the entire system (extension bars + sample). In the next step, these measurements are used as an input to a numerical inversion code for determining the Young’s modulus (E) and shear modulus (G) of the sample itself. These moduli are combined with separately known or measured material density of the sample to determine seismic velocities and attenuation.