A Study of the Effects of Diffusion on Pulsed Neutron Capture Logs(includes associated papers 7670 and 7901 )

Abstract

The response of the Neutron Lifetime Log TM is studied in viewof the diffusion of thermal neutrons in a wellbore geometry. Qualitative theoretical arguments are given for the diffusioneffects of a log-inject-log (LIL) procedure. Experimentalverification that diffusion exactly cancels out the LILprocedure is presented. Introduction One application of pulsed neutron logging has been todistinguish qualitatively among gas-, oil- and saltwaterbearing formations. The parameter ordinarily measuredis the macroscopic, thermal neutron-absorptioncrosssection, which is sensitive to the chlorinecontent of the formation and can sense the saltwatersaturation of reservoir rock. A particularly usefuladvantage of pulsed neutron logs is that they can measurecased wells quantitatively. One presumed theoreticaldeficiency has been much discussed and contended. Neutron diffusion, a well known physical phenomenon, has been recognized or at least proposed - as a possiblesource of error when measuring cased wells. Certainlythis phenomenon sullies the otherwise simple relationship between the inherent nuclear properties of a formationand, measured by commercial instruments. This expectedinfluence of thermal neutron diffusion on the NeutronLifetime Log TM (NLL) for a particular case was investigated experimentally and is discussed in the first part of thispaper. Another application of pulsed neutron logs hasreceived considerable attention in recent years.Various people have tried to use pulsed neutron capture logs in alog-inject-log (LIL) procedure to ascertain residual oil saturation quantitatively in formations that have beendepleted by secondary recovery operations. It has beenargued that the effect of neutron diffusion on themeasured must be accounted for to attain the limitingaccuracy of the method. On the other hand, it has beenrecognized that the LIL procedure involves the comparisonof two separate measurements, each of which is influenced equally by diffusion effects (if any exist). This beingthe case, the net result of neutron diffusion is zero.Demonstration of this cancellation is verifiedexperimentally and discussed in the last pan of our paper. Theory of Measurement Thermal neutrons are not detected directly in the systemthat produces the NLL. Instead, gamma rays with energygreater than 2.2 million electron volts (MeV) aremeasured. This procedure effectively samples the thermal neutron population in the formations surrounding theborehole. The NLL instrument emits cyclical bursts of14-MeV neutrons, which suffer subsequent elastic andinelastic collisions and slow down to thermal velocities. Since protons are the best moderators of fast neutrons, any fluid-filled porosity results in a rapid slowdown.Typical sand formations saturated with water haveslowdown times of 10 to 20 seconds for 14-MeV neutrons. Thus, waiting an appropriate time after the fast neutronburst assures that observed gamma rays are caused bythermal neutron reactions. The detection of gamma rays generally is conceded tobe superior to neutron detection. This advantage isprimarily because most gamma rays can penetrate fartherto reach the detector than thermal neutrons. Furthermore, gamma rays travel with the speed of light, while neutronsdiffuse slowly, especially through the highly absorbingborehole fluids that come between the formation and thedetector. JPT P. 1788^