THERMAL DECAY TIME / PULSED NEUTRON LOG BASICS
This page describes the standard TDT or PNL tools which have been the mainstay of cased hole logging since the early 1970s. They are still in use and many older logs are in your well files – these may be useful as baseline logs to assist in reservoir moniroring projects. More modern tools are called Induced Gamma Ray Spectroscopy or Advanced Pulsed Neutton Logs.

The thermal decay time log is a record of the rate of capture of thermal neutrons in a formation after it is bombarded with a burst of 14 MeV neutrons. An electronic neutron generator in the tool produces pulses of neutrons which spread into the borehole and formation.  Known as pulsed neutron capture or neutron lifetime logs, various service companies use the following mnemonics:  TDT, PNL, NLL, PDK, etc.

On older logs, the primary derived value from the pulsed neutron device is the neutron decay time (TAU) for Schlumberger logs, and the Neutron Half Life (LIFE) for Dresser logs.

TAU is the time for the neutron population to fall to 1/e (37%) of its original value and is related to the formation capture cross section (SIGMA), by the following equation:

      1: SIGMA = 4.45 / TAU for the Schlumberger tool

For the Dresser tool, LIFE represents Neutron half-life, the time required for the neutron cloud to decay to one half its original concentration.  Therefore,

      2: SIGMA = 3.15 / LIFE for the Dresser tool

On modern logs, and many older ones, the SIGMA curve is displayed and the above calculation is not needed.

Where:

  SIGMA = capture cross section (capture units)

  TAU = neutron decay time (msec)

  LIFE = neutron half life (msec)

Older TDT tools had a single pulsed neutron source; the modern version emits a dual neutron burst.  Logs from either might be encountered so an understanding of their different designs and processing algorithms will help the analyst to understand the outputs and limitations of the tool.

References:

1.  Thermal Neutron Decay Time Logging using Dual Detection

       J.T. Dewan, C.W. Johnstone, L. A. Jacobson, W. B. Wall, R.P. Alger, 
       SPWLA, 1973

2.  Dual-Burst Thermal Decay Time Logging Principles
     D.K. Stelman, R.A. Adolph, M. Mahdavi, W.E. Preeg,
     SPE Formation Evaluation, 1988

NEUTRON PHYSICS BASICS
The capture cross section SIGMA is defined as the relative ability of a material to "capture" or absorb free thermal neutrons. Chlorine has a high capture cross section and hydrogen has a low capture cross section.

The neutrons are quickly slowed down to thermal energies by successive collisions with atomic nuclei of elements in the surrounding media. The thermalized neutrons are gradually captured by elements within the neutron cloud, and, with each capture, gamma rays are emitted. The rate at which these neutrons are captured depends on the nuclear capture cross sections, which are characteristic of the elements making up the formation and occupying its pore volume. The gamma rays of capture which are emitted are counted at one or more detectors in the logging tool during different time gates following the burst, and from these counts the rate of neutron decay is automatically computed. One of the results displayed is the thermal decay time, TAU, which is related to the macroscopic capture cross section of the formation, SIGMA, which is also displayed.

Because chlorine is by far the strongest neutron absorber of the common earth elements, the response of the tool is determined primarily by the chlorine present (as sodium chloride) in the formation water. Like the resistivity log, therefore, the measured response is sensitive to the salinity and amount of formation water present in the pore volume. Consequently, when formation water salinity permits, thermal decay time logging provides a means to recognize the presence of hydrocarbons in formations which have been cased, and to detect changes in water saturation during the production life of the well. The TDT log is useful for the evaluation of oil wells, for diagnosing production problems, and for monitoring reservoir performance.

TDT-K: regions of formation and borehole decay of capture gamma ray curve.

MODELING THERMAL DECAY DATA
Modeling the neutron decay time to arrive at an accurate formation SIGMA is a complex task.  Not only do neutrons undergo two major interactions downhole:  one with the borehole and one with the formation, but neutron decay time is affected by both neutron capture and neutron diffusion. Diffusion contributes to the decrease in neutron density during the measuring period, causing the decay time to be too slow, resulting in an overestimation of SIGMA. 

Cross-plot showing exponential model results versus intrinsic SIGMA: near detector measures higher than far response, due to more diffusion effects.

With older tools, the decay time curve was approximated using a sum of exponentials, from which the borehole decay contribution needed to be removed.  The residual (non-exponential) diffusion was subtracted  from formation sigma, using chartbook solutions based on laboratory calibrations.  It was also necessary to make appropriate environmental corrections.  The exponential method created a dependence of formation sigma on the nature of the borehole fluid.  The modern dual burst system accounts explicitly for thermal neutron diffusion, using a closer solution to the neutron diffusion equation. This model precludes the use of chartbook and environmental corrections.  The improved results are obvious in the cross-plot comparing measured formation SIGMA (both models) versus inherent SIGMA.  

Comparison of measured formation SIGMA versus SIGMA intrinsic, exponential and diffusion models

EVOLUTION OF THERMAL DECAY TOOLS
Schlumberger tools have evolved through three design stages.  The 1 ^11/16” OD allows the tool to be run through tubing.

The Schlumberger TDT-K system utilizes two detectors and two variable time gates (plus a background gate) to sample the capture gamma radiation decay following the neutron burst. The width and positions of the time gates, as well as the neutron burst width and burst repetition rate, are varied in response to signals that are related to SIGMA (or more precisely, related to the formation decay rate, TAU). A hardware delay following the burst allowed the borehole signal to decay, meaning the updated TAU was generated mainly from the formation signal.  The main outputs are SIGMA and a ratio curve; the values were entered into a chart to determine apparent porosity and apparent water salinity.

 

DUAL BURST TDT LOGGING
The biggest change to TDT logging came with the TDT-P tool, developed to provide the most complete and accurate pulsed neutron capture measurement, to account for diffusion effects.  The tool features a dual burst neutron generator and is a kit upgrade to the TDT-M  tool.  A low short burst generates a short (borehole) decay component, and a lengthy, high burst characterizes the long formation decay component. Additionally, the exponential decay model and chartbook solutions of prior tools is replaced by the nuclear diffusion equation model which accounts for non-exponential diffusion and capture in both the wellbore and the formation.  The output is a correct formation sigma, independent of borehole fluid, diffusion coefficients and porosity.  User inputs are minimal (borehole size, casing size and weight).  Other inputs necessary for the diffusion model, are measured by the tool.

Timing gate configuration versus counts, dual burst TDT tools

The TDT-M system utilizes sixteen time gates and one of four possible neutron burst widths and burst repetition rates. Counts from the sixteen gates are combined to form two "sum" gates (plus a background gate) from which SIGMA is computed. As in the TDT-K system, the combination of gates used to form the "sum" gates, as well as the burst width and repetition rate, are selected according to SIGMA (or TAU) of the formation.  Because borehole diffusion is mainly a constituent of the near detector counts, this configuration allows SIGMA to be calculated preferentially using the far detector, giving a more accurate formation SIGMA.

Two processing methods are possible:  a real-time, single component wellsite method, and an offsite VAX model, in which 2 components are fitted independently; the offsite result is deemed accurate when the dual component results are in agreement.  The wellsite model provides the initial SIGMA formation values to the offsite model.  In general, the wellsite model is sufficient for monitoring water movement but the offsite model is recommended in heterogeneous or unknown lithologies, (e.g. having shales, clays or anhydrite), or in complex borehole geometries.  The improved precision of the offsite model makes it a better choice for critical operations such as Log-Inject-Log. 

Other service companies offer similar tool designs.

TDT / PNL DATA ANALYSIS BASICS
The main use of the TDT log is to determine water saturation from the contrast between SIGMA for hydrocarbons and saline water.  Other uses are for porosity, apparent salinity and gas detection. 

The water saturation is based on the sum of the capture cross sections, in a mathematical treatment similar to the sonic, density and neutron logs.

Here is the log response equation for the SIGMA measurement with only hydrocarbon and water in the porosity:

3: SIGMA= PHIe * Sw * SIGw (water term)

                    + PHIe * (1 - Sw) * SIGh (hydrocarbon term)

                    + Vsh * SIGsh (shale term)

                    + (1 - Vsh - PHIe) * Sum (Vi * SIGi) (matrix term)

Where:

  SIGh = log reading in 100% hydrocarbon

  SIGi = log reading in 100% of the ith component of matrix rock

  SIGMA = log reading

  SIGsh = log reading in 100% shale

  SIGw = log reading in 100% water

  PHIe = effective porosity (fractional)

  Sw = water saturation in reservoir (fractional)

  Vi = volume of ith component of matrix rock

  Vsh = volume of shale (fractional)

This equation is solved for Sw by assuming all other variables are known or previously calculated:

      4:  SIGw = 22.0 + 0.000404 * WS(ppm)

      5: SIGm = Sum (Vi * SIGi)

      6: PHIe = TPHI from log if available and no gas OR from open hole logs

      7: SWtdt = ((SIGMA - SIGm) - PHIe * (SIGh – SIGm) - Vsh * (SIGsh - SIGm))

                    / (PHIe * (SIGw - SIGh))

*Note:  the Sigma value of water varies with salinity, temperature and pressure (see SLB chart Gen-12, Capture Cross Section of NaCl Water Solutions)

The ratio of counts in the near to far spaced detector is recorded and used as an estimate of formation porosity, in a fashion similar to the CNL neutron log*. Earlier TDT logs had only one detector, so no ratio porosity was available.  In the dual detector tool, porosity is derived from a ratio of formation gate counts and sigma, and is corrected for borehole salinity.

*CAUTION: The early dual detector TDT logs do not give a good value for porosity in dolomite reservoirs. Always compare PHItdt to core or open hole log analysis whenever possible.

On most TDT tool, a visual comparison of the count rates from the near and far detectors are used to detect gas in clean, high porosity formations, with the far detector curve moving to the left and the near moving slightly to the right.  However, this was not definitive; a change in porosity could mimic a gas response. The newer TDT-P tool provides a porosity independent method of gas detection, via the inelastic count rate far detector curve (INFD).

TDT / PNL LOG CURVE NAMES

Older TDT Logs - Single Burst

Curves                                        Units                       Abbreviations

thermal decay time                      sec                      TAU

capture cross section                  c.u.                         SIGMA

casing collar                                mv                          CCL

* gamma ray                                         api                          GR

Modern TDT Logs - Dual Burst

Curves                                       Units                       Abbreviations

thermal decay time                     sec                     TAU

capture cross section                  c.u.                         SIGMA

corrected borehole sigma            c.u.                         SIBH

borehole corr. formation sigma     c.u.                         SIGM

inelastic counts far detector         cps                         INFD

casing collar                               mv                          CCL

near count rate                           cps                         NCPS, N1, TSCN (total selected counts for near detector)

far count rate                              cps                         FCPS, F1, TSCF (total selected counts for far detector)

* gamma ray                                         api                          GR

* ratio                                                   unitless                   RATIO

* TDT Porosity                            fractional                 TPHI

EXAMPLES OF TDT / PNL LOGS
 

TDT-K type log with gamma ray GR, count rate ratio, sigma, near  and far count rate overlay N1 and F1. On many logs, TDT porosity TPHI replaced the ratio curve. Curve name abbreviations vary widely depending on era and service supplier.  The relative behaviour of near and far count rate curves F1, N1 in track 3 were used to detect gas.

Gas detection example TDT-P: density neutron crossover from open hole logs in Track 1, near and far count rate crossover in Track 3. INFD, the inelastic count rate far detector curve (Track 3), increases with far and near count curves, confirming gas response.

Pulsed neutron (PDK-100) reservoir monitoring example. Original openhole analysis (Tracks 5 and 6), PDK saturation 10 years later (Track 4), and 13 years later (Track 3 )

TDT-P display of wellsite processing, showing changing oil water contact over four passes (track 1 Borehole Sigma), with porosity (track 2) and Formation Sigma (track 3).

Overlay of Wellsite processing (field) with Offsite processing results