SPE Lithology-independent porosity measurement. Continuous producibility/permeability estimates. Fluid characterization capability.

Transcription

1 SPE A Next-Generation Wireline NMR Logging Tool L. DePavia, SPE, N. Heaton, SPE, D. Ayers, R. Freedman, SPE, R. Harris, B. Jorion, J. Kovats, SPE, B. Luong, N. Rajan, R. Taherian, K. Walter, D. Willis, Schlumberger; Jeffrey Scheibal, SPE, Serena Garcia, SPE, Shell Exploration and Production Company Copyright 2003, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the SPE Annual Technical Conference and Exhibition held in Denver, Colorado, U.S.A., 5 8 October This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box , Richardson, TX , U.S.A., fax Abstract This paper discusses design and implementation of a newgeneration nuclear magnetic resonance (NMR) wireline logging tool. The tool is a multifrequency, eccentered gradient-field design with multiple depths of investigation (DOI) spanning a range of several inches. The broad range of DOIs enables radial profiling of saturation distribution and formation damage. Because of the eccentered mode of operation and sensor design, the DOIs are maintained regardless of hole size and temperature. The DOIs range from 1 to 4 in. The tool features multiple sensors, including a large antenna for fluid characterization and complementary smallaperture antennae for high-resolution acquisition modes. The sensors can be operated either separately or simultaneously at logging speeds of up to 3,000 ft/hr. Comparison of the responses is used to provide high-resolution identification of long T1 fluids such as light hydrocarbons. New NMR acquisition sequences such as Diffusion Editing (DE ) are readily implemented on the new tool, which has a highly flexible pulse programmer. Combining the new acquisition methods with the multifrequency capability of the tool and Magnetic Resonance Fluid (MRF * ) characterization processing provides robust fluid saturation and oil viscosity answers. Introduction The development of NMR well-logging technology has been motivated by the unique set of answers that NMR can provide. In particular the benefits include: Mark of Schlumberger Lithology-independent porosity measurement. Continuous producibility/permeability estimates. Fluid characterization capability. Over the course of the last decade, NMR logging has experienced a remarkable evolution. Wireline NMR tools now log faster and provide considerably better signal-to-noise-ratio (SNR) than the early tools. 1,2 The enhancements in data quality provided by changes in hardware design 3,4 were followed by significant improvements in many of the NMR answers. In recent years there has been increasing emphasis on fluid typing applications of NMR tools. It has been recognized for some time that NMR logs provide a useful alternative in situations where conventional log analysis is difficult or unreliable such as in low-resistivity or low-contrast pay environments. One aspect of NMR fluid characterization that has marked its progress is the continual and rapid development of new acquisition methods. For example, new multimeasurement diffusion-based techniques 5 have been successfully deployed recently. Results obtained so far suggest that NMR data can provide quite detailed fluid characterization, going beyond simple hydrocarbon detection and saturation estimation. A new tool, the Magnetic Resonance expert (MRX * ), has been developed to provide: Robust fluid characterization. Multiple well-defined and evenly spaced DOIs for invasion profiling. High vertical resolution. Standard NMR answers at high logging speeds. In order to achieve these objectives a multi-sensor design with a programmable tool architecture was adopted. A high degree of pulse sequence flexibility is essential in order to introduce new measurements as they are conceived. For example, DE 5 sequences were implemented at the initial stages of field testing and have subsequently been optimized based on the results obtained. Constraints on magnet size and radio frequency pulse strength ultimately limit NMR measurements to the nearwellbore region. However, the value of increasing the radial extent of NMR measurements is evident. First, deeper measurements provide a greater likelihood of sensing native fluids in shallow invasion environments. Second, rugosity

2 2 SPE effects, which can degrade shallow NMR measurements, may be alleviated by increasing the DOI. The new tool acquires data independently at a range of well-defined DOIs up to 4 in. Field test results have clearly confirmed the value of the deeper measurements both for sensing native fluids and for reducing rugosity effects. In addition, it has been observed that valuable information concerning formation fluids can be derived by comparing NMR measurements acquired at different DOIs. Combining the radial profiling capability of the new tool with the new diffusion techniques described above provides saturation profiles of the near-wellbore region. Unlike most other logging tools, NMR tools have extremely well-defined sensitive volumes determined by the dimensions of the antenna and its relation to the static magnet. This property of the NMR measurement is useful for radial profiling, as discussed above. It is also helpful for highvertical resolution measurements, 6,7 provided that the antenna aperture can be made sufficiently small. Another objective of the new design was to maintain a high-resolution measurement for providing standard rock quality and producibility answers. This has been achieved by incorporating independent high-resolution antennae in addition to the main antenna, which is used for fluid characterization and radial profiling. This paper builds on an earlier publication that presented initial results of an experimental version of the new NMR logging tool. 8 Here we provide detailed specifications of the tool and present examples that demonstrate the value of the unique design features. Tool Design and Specifications The tool has a main antenna, designed for fluid typing applications, and two high-resolution antennae. A schematic picture of the tool is shown in Fig. 1. The main antenna operates at multiple frequencies and is intended primarily for fluid characterization applications. There are eight different frequencies of operation corresponding to independent measurement volumes (shells) with evenly spaced DOIs. The shell volumes form concentric arcs in front of the antenna, as illustrated in Fig. 1. Fig. 1 Schematic representation of the new NMR tool. The volumes and DOIs are independent of temperature. The high-resolution antennae operate at a single frequency, corresponding to a slightly shallower DOI than the main antenna. These antennae provide rock quality and producibility answers. A flexible pulse sequence programmer allows multiple frequencies (main antenna) and the different antennae to be addressed in a single acquisition mode. Data acquisition is designed so that idle time is minimized and data density is maximized. The tool is run eccentered, using bowsprings to press the antenna against the borehole wall. The eccentered design enables the tool to be conveyed on pipe and operated in large holes and in deviated wells. It also ensures that the measurement volumes and depths of investigation are independent of hole size. The main specifications of the new generation tool currently being deployed are summarized in Table 1. Examples presented later in this paper were obtained with an experimental (EXP) tool, for which the specifications differ slightly from those given below. Table 1 NMR Tool Specifications Tool length (ft) 32.7 Sonde diameter (in.) 5.0 Tool weight (lbm) 1,200 Max temperature ( F) 300 Max pressure (kpsi) 20 Min echo spacing (ms) 0.45 Frequency (khz) 1, Field gradient (G/cm) Depth of investigation (in.) 1.25, 1.5, 1.9, 2.3, 2.7, 3.1, 3.4, 3.7, 4 Vertical resolution (in.) 4, 18 Min hole size (in.) 5 7/8 Logging speeds (ft/hr) 3, Notes 1. Frequency, field gradient and DOI are provided separately for the high-resolution antenna (first entry). Ranges of values are given for the main antenna. 2. Vertical resolution specifications are defined by antenna aperture in axial direction. Values provided separately for high-resolution antenna and main antenna. 3. Logging speed depends on acquisition mode. The range of DOI accessed by the tool falls in what is often regarded as the invaded zone, where movable native fluids have been largely replaced by mud filtrate. Nonetheless, experience has shown that there remain many environments, most notably in wells drilled with oil-based muds (OBM), where invasion is shallow and NMR tool measurement volumes include significant proportions of native fluids. However, if only one measurement volume is available, or the span of measurement volumes is small, it may not be possible to determine relative fractions of native fluid and filtrate. Results obtained during field testing have shown that invasion profiles can often be observed by comparing results from different shells and antennae. Because the NMR signal amplitude is proportional to the average hydrogen index of formation fluids, variation in NMR porosity or free-fluid volume with DOI may be interpreted in terms of varying fractions of filtrate and native fluids with different hydrogen indices.

3 SPE Log Examples Radial Profiling. The radial profiling acquisition mode utilizes four different shells corresponding to DOIs of 1.5, 1.9, 2.3, and 2.7 in. The shells are accessed sequentially such that the tool moves forward by about one antenna length during the course of one complete measurement cycle. This strategy ensures that the first measurement on each shell benefits from the long polarization time provided by the prepolarizing magnet in front of the antenna. Fig. 2 illustrates the acquisition timing. Fig. 2 Schematic representation of the radial profiling acquisition. The bars represent the positions of the measurement volumes as the tool (shown on the left, with antenna at lower right) moves up the borehole. Shells 1, 2, 3, and 4 are represented by black, green, red, and blue bars respectively. On each visit to a shell, an enhanced precision mode 4 (EPM) suite is acquired. The EPM parameters (WT = wait time, TE = echo spacing, NECHO = number of echoes, REPT = number of repetitions) used for the logs presented below are given in Table 2. The radial profile log is typically run at 850 ft/hr with a sample interval of 18 in. Note that these parameters refer to the EXP tool that was used to acquire the examples presented here. Current tools use slightly modified acquisition parameters. In particular, standard echo spacings (TE) in the current tool are 0.45 ms instead of the 0.6 ms for all the shells used for this acquisition mode. Fig. 3 Radial profile logs acquired in a Texas water well. Gamma ray and induction logs are shown in Tracks 1 and 2. Track 3 compares density porosity with porosity from the new tool at 2.7 in. Track 4 presents all four porosities from the new tool. Brown and blue shading indicates fines invasion and washout respectively. T2 distributions are shown in Tracks 5 through 8. Gamma ray and induction logs are shown in Tracks 1 and 2. Track 4 compares the four porosities. Brown shading shows the porosity deficiency in Shell 1 (1.5 in.) relative to Shell 4 (2.7 in.) and indicates whole mud invasion. The radial profiling results show that significant quantities of mud solids, accounting for 10 to 15 p.u. in porosity, invade about 1.5 in. into the formation. There is relatively little fines invasion beyond 2 in. Blue shading at approximately 320 ft and 500 ft indicates porosity excess in Shell 1 relative to Shell 4. The excess porosity in the shallower measurements is due to mud signal that appears in washouts. Although shallow measurements may be affected by mud signal or formation damage, the deepest measurement still provides a robust porosity. By comparing the well-separated measurement volumes, radial profiling provides a valuable hole quality indicator. Table 2 Radial Profile Acquisition Parameters Measurement WT (ms) TE (ms) NECHO REPT The first radial profile example was acquired in a water well in Texas. Logs were acquired in a well drilled through a shallow, high-porosity sandstone formation. Logs acquired in previous wells in the same area had indicated that the formation was invaded by whole mud, which affected porosity and other log measurements. One of the objectives was to use the log from the new tool to identify and, if possible, characterize the extent of the formation damage. Fig. 3 shows the radial profile logs acquired in the water well using the new tool.

4 4 SPE Fig. 4 shows a radial profiling log acquired in a deepwater Gulf of Mexico well. This well was drilled with OBM and traversed gas-bearing sands. neutron and density logs (Track 3) are well separated, indicating a high shale content. Fig. 4 Radial profile logs across well-defined, gas-bearing sands in a Gulf of Mexico well. Gamma ray and deep resistivity logs are shown in Tracks 1 and 2. Track 3 shows neutron and density porosity logs. Porosity logs (MRP) for 1.5, 1.9, 2.3, and 2.7 in. DOI are plotted in Track 4. Also shown in Track 4 and shaded in red is a minimum gas volume log computed from free-fluid volume logs. T2 distributions are shown in Tracks 5-8. In this interval, the gas-bearing sands are easily identified by the conventional logs shown in Tracks 1 to 3. In particular, there is a well-defined density neutron crossover (red shading in Track 3) in five of the six sand packets. At approximately x630 ft there is another short section that appears to be sand from the neutron density logs; however, there is no crossover indicating gas. The porosity logs are presented in Track 4. In each of the gas-bearing sand packets, there is a clear reduction in apparent porosity for the deeper measurements relative to the shallowest measurement. The variation in MRX porosity with DOI reflects the invasion front of the OBM filtrate. The shallower shells sense more of the filtrate and less gas relative to the deeper shells. Because gas has a much lower hydrogen index than the filtrate, the deep measurements read lower porosity. Note that the porosities from all four shells overlay in the sand at approximately x630 ft, where the neutron density logs provide no crossover. Inspection of the T2 distributions (Tracks 5 to 8) further confirms that the free-fluid volume decreases for the deeper shells over the gas-bearing sand packets. Bound-fluid volumes are similar for all four shells, as expected. The curve bounding the red area in Track 4 is a minimum gas volume log generated from the difference in free-fluid volumes between Shell 1 (DOI = 1.5 in.) and Shell 4 (DOI = 2.7 in.). Observe the remarkable similarity between this new log and the neutron density crossover. A second radial profiling log was acquired over another section of the same well that passed through a laminated sand. The results are presented in Fig. 5. In this case, the conventional logs do not provide clear evidence of hydrocarbon. The gamma ray (Track 1) is quite featureless over the entire interval. The deep resistivity log (Track 2) ranges from 1 to 2 ohm-m, but is also rather characterless. The Fig. 5 Radial profile logs across laminated sands in a Gulf of Mexico well. Gamma ray and deep resistivity logs are shown in Tracks 1 and 2. Track 3 shows neutron and density porosity logs. Porosity logs (MRP) for 1.5, 1.9, 2.3, and 2.7 in. DOI are plotted in Track 4. Also shown in Track 4 and shaded in red is a minimum gas volume log computed from free-fluid volume logs. T2 distributions are shown in Tracks 5-8. The porosity logs from the new tool are shown in Track 4. Also shown in Track 4 is the light-hydrocarbon indicator, which bounds the red shaded area. Although the indicator is less pronounced than in the other gas-bearing sands of the previous example, there is nonetheless a clear free-fluid deficit in the deeper shells confirming the presence of gas. High-Resolution Logging. The new tool s high-resolution acquisition mode combines the high-resolution antennae with one of the shells of the main antenna. The high-resolution mode timing is illustrated in Fig. 6. Fig. 6 Schematic representation of the high-resolution acquisition. The bars represent the positions of the measurement volumes as the tool (left) moves down the borehole. The highresolution antenna and measurements are shown in orange. The light and dark gray bars indicate successive overlapping measurements on the main antenna. The sequence is designed such that the high-resolution antenna acquires non-overlapping measurements with long effective polarization times provided by the pre-polarizing magnet. Measurements on the main antenna are overlapping, and the effective polarization time is just the main antenna idle time, which is about 1.3 sec. As with the radial profiling

5 SPE scheme, EPM sequences are acquired each time the antennae are accessed in the measurement cycle. High-resolution acquisition parameters (WT = wait time, TE = echo spacing, NECHO = number of echoes, REPT = number of repetitions) used in the example shown below are provided in Table 3. The long wait time (measurement 1) for the main antenna is shown in parentheses. Table 3 High-Resolution Acquisition Parameters Measurement WT (ms) TE (ms) NECHO REPT 1 10, (1,300) The high-resolution mode may be logged either up or down. Typical logging speed for this mode is 650 ft/hr and the sampling interval is 5 in. A high-resolution-mode log was acquired over the same laminated sand shown in the previous example (radial profiling, Fig. 5). In this case the sequence combined the highresolution antenna (DOI = 1.25 in.) with Shell 1 from the main antenna (DOI = 1.5 in.). Fig. 7 compares the MRX highresolution logs with CMR-Plus logs acquired over the same interval. measurement precision. In effect, the two antennae provide a simultaneous main and repeat pass. In addition, the correlation with corresponding CMR-Plus curves is also remarkable. There are, however, some differences. In particular, the CMR- Plus free-fluid volume reads higher in some of the sand laminations, while in other sands it agrees well with the new tool s free-fluid volumes. Inspection of the T2 distributions in Tracks 5 7 shows that the free-fluid peak is well separated from the bound fluid, so these differences cannot be due to different T2 resolution for the two tools. The differences can be explained by the presence of gas or other light hydrocarbon, which is largely expelled from the sensitive region of the CMR-Plus measurement but is sensed by the deeper measurements of the new tool. This result provides further confirmation of the radial profiling log over the same interval, which also indicated the presence of gas. Note that the measurements used in the high-resolution log are in fact the shallowest of the shells. Had Shell 3 (DOI = 2.3 in.) or 4 (DOI = 2.7 in.) been used rather than Shell 1, it is likely that the effect would have been more prominent. There is a significant shale signal with T2 components of less than 1 ms throughout this interval. The 0.2-ms echo spacing of the CMR-Plus tool is more sensitive to these components and therefore provides slightly higher porosity and bound fluid volumes than the EXP tool run with a 0.6 ms echo spacing. Note that current tools use a 0.45 ms echo spacing, which provide greater sensitivity to short T2 components. Fig. 7 High-resolution logs across laminated sands in a Gulf of Mexico well. Gamma ray is shown in the depth track. The deep resistivity log is plotted in Track 1. Tracks 2 and 3 show the NMR free-fluid and porosity logs respectively. Orange curves (HR1) refer to the high-resolution antenna. Black curves (HR2) refer to Shell 1 of the new tool, and green curves indicate CMR-Plus data. Density and neutron measurements are shown in Track 4. The new tool s T2 distributions for the high-resolution and main antenna are shown in Tracks 5 6 respectively. Track 7 shows the CMR-Plus T2 distribution. The excellent agreement between the two sets of the new tool s high-resolution, free-fluid, and porosity logs (Tracks 2 and 3) is very satisfying and lends confidence to the Mark of Schlumberger Saturation Profiling and Fluid Characterization. Current NMR methods to analyze reservoir fluids can be divided into three categories. The first category involves acquisition of a small number of measurements (usually two) in which either the polarization time or the echo spacing is varied. The results are analyzed by either computing difference spectra or comparing T2 distributions from each measurement. The observed differences are then interpreted in terms of oil, water, or gas, based on assumed simplified NMR responses. While these methods are relatively straightforward to implement, the reduced set of measurements do not take advantage of the full potential of the NMR technique and can lead to erroneous interpretation. Greater insight into the molecular physics and NMR properties of formation fluids led to the second category of NMR fluid typing, which uses model-based inversions of multi-measurement data. Two examples of the forward modeling approach are the MACNMR 9 and MRF characterization methods. 10 Not surprisingly, model-based analysis is generally successful in environments where the fluid models provide a good representation of the measured NMR responses. However, in non-ideal situations where fluid NMR properties deviate from the model behavior (e.g., as a result of internal gradients, oil-wet rocks, restricted diffusion), these techniques can lead to inaccurate interpretation. Fortunately, these limitations can largely be overcome by the third and most recent category of NMR fluid characterization. In this new approach, multi-measurement NMR data is analyzed independently of any fluid model. 5,11 The model-independent

6 6 SPE inversion provides two-dimensional maps (D-T2 maps 12 ) correlating T2 relaxation times with molecular diffusion rates. The D-T2 maps constitute a powerful new tool for identifying fluids, calibrating their NMR responses, and optimizing MRF analysis. In parallel with the advances in analysis methods, new NMR acquisition sequences have been developed that are specifically designed to enhance fluid characterization capabilities. In particular, the DE technique 5,8 has been demonstrated to provide substantial improvements in the quality of important fluid diffusion information. This sequence differs from the standard Carr-Purcell-Meiboom-Gill (CPMG) sequence in that the time between successive refocusing pulses is not constant. In a typical DE measurement, the first two echo spacings, t e,l, are long and are followed by a long train of identical echoes with shorter inter-echo spacing, t e,s. The DE sequence independently measures diffusion rates and T2 relaxation times while providing substantially higher echo density and improved answer precision than standard CPMG measurements. One of the principal objectives for the new generation logging tool is to provide robust and comprehensive fluid typing. This is achieved by using DE acquisition sequences together with model-independent inversion and MRF analysis of the data. Combining the multiple-doi measurement capability with fluid characterization provides a saturation profile of the near-wellbore region. Typical saturation profile measurement sequence parameters used in the EXP tool are given in Table 4. The sequence comprises six separate measurements acquired independently on each of four shells. The first two measurements on each shell are standard CPMG trains (t e,l = t e,s ), acquired with different wait times. The remaining four measurements are DE sequences with the same wait time (3 sec) and varying t e,l values. The t e,l times are listed for each shell (S1 to S4) in Table 4. Saturation profiling measurements can be acquired as depth logs (logging speed approximately 200 to 300 ft/hr) or station logs. Table 4 Saturation Profiling Acquisition Parameters WT (ms) t e,l (ms) S1 - S4 t e, s (ms) NECHO 1 10, , 0.6, 0.6, , , , 0.6, 0.6, , , , 3.0, 4.0, , , , 4.0, 5.0, , , , 5.0, 7.0, , , , 7.0, 10, ,000 Fig. 8 shows a set of four D-T2 maps derived from a saturation profiling station log. This log was acquired in a light-oil-bearing sand in a Gulf of Mexico well drilled with OBM. Overlaid on each map are default fluid response lines for dead oil (yellow diagonal line) and water (horizontal light blue line). Each of the four D-T2 maps shows two welldefined main peaks above 0.1 sec on the T2 axis. There is also a weak bound fluid signal with T2 below 0.01 sec. The freefluid peak at T2 0.2 sec, D ~ cm 2 s -1 corresponds to OBM filtrate. The long T2 peak with the higher diffusion rate derives from native hydrocarbon. The fact that the two fluids can be distinguished so clearly in the D-T2 map is interesting and encouraging. Fig. 8 D-T2 maps derived from saturation profiling station log acquired in a hydrocarbon-bearing sand in a Gulf of Mexico well. The DOI of the measurement is given in the top left corner of each map. Overlaid on each map are default fluid response lines for dead oil (yellow diagonal line) and water (horizontal light blue line). These results indicate that the filtrate and native oil remain largely separate in this formation. However, the filtrate peak is shifted to significantly shorter T2 times and higher diffusion rates relative to nominal values for unaltered filtrate at 194 F. The shift in filtrate properties is verified by comparing the D- T2 maps for the hydrocarbon-bearing sand with results obtained in a sand with high-water saturation. Fig. 9 shows a D-T2 map acquired in a depleted sand in the same well. In this case, the measurements were acquired with the high-resolution antenna with a DOI of 1.25 in. Fig. 9 D-T2 map derived from saturation profiling station log acquired in a depleted sand in a Gulf of Mexico well. The DOI of the measurement is 1.25 in. Overlaid on each map are default fluid response lines for dead oil (yellow diagonal line) and water (horizontal light blue line). In this case the filtrate peak appears as expected at a T2 of 1 s and D of cm 2 s -1. Note that the native hydrocarbon peak that appears clearly in the previous example has now practically disappeared, consistent with the highly depleted nature of this sand. The shift in filtrate properties observed in the hydrocarbon sand is due to dissolved gas in the filtrate. Introduction of gas into a liquid hydrocarbon is known to cause an increase in diffusion rate of the liquid and a reduction in the relaxation time. 13 In the examples described above, dissolved gas causes the filtrate diffusion rate, D, to increase

7 SPE by a factor of between 2 and 3. This is accompanied by a reduction in T2 of almost a factor of 5. Conclusions A new wireline NMR logging tool has been developed to address a range of formation evaluation problems. Field testing of the experimental tool confirmed performance and verified new NMR interpretation methodologies. The new tool has a main antenna intended primarily for fluid characterization applications and high-resolution antennae for standard rock quality and producibility answers. The main antenna operates at multiple frequencies, which correspond to different shells, or DOIs. Different shells and different antennae can be addressed in a single acquisition mode. This capability is utilized in the radial profiling application, which detects the radial variation of formation fluid hydrogen index in the near wellbore region resulting from filtrate invasion. Another new technique, saturation profiling, involves the acquisition of diffusion editing measurements at multiple DOIs. Model-independent analysis of the data then yields D-T2 maps at each DOI. The maps afford detailed information concerning fluid properties downhole that cannot be obtained by any other logging method. High-resolution acquisition modes combine measurements on the main and high-resolution antennae. Comparison of logs from the two antennae serves as a useful quality control. Differences in apparent porosity and free-fluid volumes for the two antennae can be interpreted in terms of fluid properties or variations in fluid saturations with distance from the borehole. Acknowledgments We are extremely grateful to the operating companies that provided access to their wells during the course of field testing and have provided permission to publish results. Many people other than the listed authors have contributed to the development of the new tool and deserve special mention. In particular we wish to acknowledge Dylan Davies, Martin Poitzsch, Jan Smits, Apo Sezginer, Peter Speier, Boqin Sun, Jaideva Goswami, Larry Lloyd, Dick Peloquin, Pete Cabrera, Bill Endress, Allen Wong, Bill Vandermeer, John Hobbins, Ali Toufaily, Nate Bachman, Agadagba Obehiri, Dan Ward, Kevin McEnaney and Wes Wofford. We would also like to thank Justin Freeman and Mario Winkler of Shell Research and members of the NMR group at Schlumberger-Doll Research, especially Martin Hurlimann and Charlie Flaum, who have collaborated throughout. Nomenclature D = molecular diffusion coefficient, cm 2 s -1 NECHO = number of echoes in NMR measurement REPT = number of measurement repetitions S1 = Shell 1 (1.5-in. depth of investigation) S2 = Shell 2 (1.9-in. depth of investigation) S3 = Shell 3 (2.3-in. depth of investigation) S4 = Shell 4 (2.7-in. depth of investigation) T 1 = longitudinal relaxation time, sec T2 = transverse relaxation time, sec TE = echo spacing in CPMG measurement, c t e,l = long echo spacing in DE measurement, ms t e,s = short echo spacing in DE measurement, ms WT = effective recovery time, ms References 1. Morriss, C.E. et al.: Field Test of an Experimental Pulsed Nuclear Magnetism Tool, Transactions of the SPWLA 34 th Annual Logging Symposium, Calgary, Alberta, Canada, June 13-16, Miller, M.N. et al.: Spin Echo Magnetic Resonance Logging : Porosity and Free Fluid Index Determination, paper SPE presented at the 1990 SPE Annual Technical Conference and Exhibition, New Orleans, U.S.A., September. 3. Prammer, M. et al.: A New Multiband Generation of NMR Logging Tools, paper SPE presented at the 1998 SPE Annual Technical Conference and Exhibition, New Orleans, U.S.A., September McKeon, D. et al.: An Improved NMR Tool Design for Faster Logging, Transactions of the SPWLA 40th Annual Logging Symposium, Oslo, Norway, May 30 - June 3, Hurlimann, M. et al.: Diffusion Editing : New NMR Measurement of Saturation and Pore Geometry, Transactions of the SPWLA 43rd Annual Logging Symposium, Oiso, Japan, 2 5 June, Sezginer, A. et al.: A NMR High Resolution Permeability Indicator, Transactions of the SPWLA 40th Annual Logging Symposium, Oslo, Norway, May 30-June 3, Heaton, N. et al.: High Resolution Bound Fluid, Free Fluid and Total Porosity with Fast NMR Logging, Transactions of the SPWLA 41st Annual Logging Symposium, Dallas, U.S.A., June 4-7, Heaton, N. et al.: Applications of a New Generation NMR Wireline Logging Tool, paper SPE presented at the 2002 SPE Annual Technical Conference and Exhibition, San Antonio, U.S.A., 29- September 2 October 9. Slijkerman, W. F. J. et al.: Processing of Multi-Acquisition NMR Data, SPE presented at the SPE annual conference 3-6 Oct. Houston, Texas, Freedman, R. et al.: A New NMR Method of Fluid Characterization in Reservoir Rocks: Experimental Confirmation and Simulation Results, SPE published in the SPE Journal, Cao Minh, C. et al.: Planning and Interpreting Fluid Characterization Logs, paper SPE presented at the 2003 SPE Annual Technical Conference and Exhibition, Denver, U.S.A., 5 8 October 12. Venktamaranan, L. et al.: Nuclear Magnetic Resonance Measurements and Methods of Analysing Nuclear Magnetic Resonance Data, U.S. Patent No. 6,462,542, (2002). 13. Lo, S. et al.: Correlations of NMR Relaxation Times with Viscosity, Diffusivity, and Gas/Oil Ratio of Methane/Hydrocarbon Mixtures, paper SPE presented at the 2000 SPE Annual Technical Conference and Exhibition, Dallas, U.S.A., 1-4 October.