The Geochemical Logging Tool (GLT) uses three separate modes of gamma-ray spectroscopy to obtain measurements of most of the major oxides which make up sedimentary and igneous rocks. Initial measurements provide estimates of Si, Al, Fe, Ca, K, U and Th (together with H and Cl). Estimates of Ti, S and Gd are obtained later with further processing. The GLT provides gross geochemical information about the formation which is particularly useful when combined with other logs. The data can be used directly for the characterization of geological sequences and phenomena, and are excellent for geotechnical zoning. However, due to its relatively low measurement precision (see "Data Quality and Limitations" below), the GLT is best employed in environments where there is a marked variation in the geochemistry of the rocks.

The GLT consists of four components. At the top is a natural gamma ray tool. Beneath this is a compensated neutron porosity tool, which in the GLT is used solely as a carrier for a Californium (252Cf) source. Californium is used instead of the conventional AmBe source (see CNT1) because its spectrum has a lower energy (2 MeV instead of 4.5 MeV), thus reducing the number of fast neutron reactions which would interfere with measurements taken by the tools below. Next is the aluminum activation clay tool, which is essentially a natural gamma ray tool with a modified spectrometer (7 windows instead of 5) to allow a more detailed analysis of the spectrum. Finally, a gamma ray spectrometry tool is located at the bottom of the string. A boron exclusion sleeve surrounds the gamma ray spectrometry tool and increases the signal-to-noise ratio by shielding the path of fast neutrons from borehole fluid and reducing the number of capture reactions in the borehole itself, thus counteracting the effects of chlorine and water present in the borehole. The sleeve also reduces the interference of iron from the tool housing.

Measurements and data processing

The natural gamma ray tool measures the abundance of K, U and Th from the natural gamma radiation given out by these elements (see also NGT). A sodium iodide detector is used for the measurement and this also provides a spectrum of the background radiation which is required for subsequent processing. Data are collected as the toolstring is pulled up the borehole so that natural gamma-ray measurements are made before the formation is activated by the neutron and gamma spectroscopy tools.

The next two tools in the string allow the measurement of the Al concentration. The 252Cf source in the compensated neutron porosity tool causes the neutron activation of Al, in which the natural isotope 27Al absorbs thermal neutrons and produces the isotope 28Al, which decays with a half life of 2.24 minutes and emits 1.78 MeV gamma rays. The aluminum activation clay tool measures the gamma spectrum of the activated formation and the Al component is determined by subtracting the input from the natural gamma ray tool spectrum. There is some spectral interference in the aluminum measurement from silicon which is corrected during the land-based processing.

The gamma ray spectrometry tool can operate in two timing modes: inelastic, which mainly measures the neutron reactions in the high energy range; and capture-tau mode, which employs prompt neutron capture reactions to measure elemental concentrations. This report describes how the gamma ray spectrometry tool functions in capture-tau mode, which is how it is normally used in the ODP. For an example of its use in inelastic mode the reader should refer to ODP Leg 164.

The gamma ray spectrometry tool uses a ‘minitron’ tritium source to bombard the formation with pulsed 14 MeV neutrons. Through scattering reactions with the atoms in the formation, the neutrons progressively lose energy until they reach a thermal energy at which they can be captured by elemental nuclei in the rock. When this occurs the nucleus emits a gamma ray at a unique energy, characteristic for each element. The emitted gamma rays are measured by a spectrometer consisting of a sodium iodide detector and a 256-channel analyzer. During logging, the gamma ray spectrometry tool provides estimates of Si, Fe, Ca, Cl and H. In ODP boreholes the Cl and H relate virtually entirely to the sea water in the borehole. Later land-based processing permits the removal of Cl and H from the spectra, and the additional extraction of estimates for Ti, S and Gd.

Following data acquisition, the elemental concentrations measured by the GLT are expressed as decimal fractions and the elements are normalized to unity. Further processing, sometimes referred to as the 'oxide closure procedure', converts the major elements (Si, Al, Ca, Fe, S, Ti, K, Cl and H) to weight percent oxides. The trace elements (U, Th and Gd) are expressed in parts per million. Post-cruise processing also allows the expected errors on the GLT data to be calculated.

The elements measured by the GLT account for the bulk of the chemistry of most common rocks; the only significant elements not measured are Na, Mg and possibly Mn. Under favorable circumstances an estimate of these missing elements may be obtained by comparing a calculation of the photoelectric factor (Pe) from the elements measured above, with the direct measurement of Pe made by the lithodensity tool (HLDS). The difference in these Pe values is, within limits of error, due to the unmeasured elements, and may be recast as either Na or Mg, or some combination, where a fixed ratio of the elements has to be assumed.

Data Quality and Limitations

During data acquisition the signal to noise ratio of the gamma-ray measurement can be affected by the following:

• Logging speed. This is normally between 400 and 600 ft/hr, with measurements being made every six inches. The slower the logging speed the greater the measurement precision.
• Borehole fluids and porosity. Due to the large capture cross-section of chlorine and hydrogen more than half of the gamma ray spectrometry tool signal may come from the borehole fluid (normally seawater). This can adversely affect the measurement precision of the other element yields. High porosity rocks can have a similar affect on precision, and it is recommended that the GLT only be used in rocks with less than 40% porosity.
• Hole size. This is of particular importance as oversized holes cause an increase in the signal derived from the borehole fluids, and a decrease in the signal from the formation. The interpretation of geochemical logs should, therefore, always be undertaken in conjunction with caliper logs. Because the aluminum activation clay tool has a low activation energy (2 MeV), aluminum is measured in a much smaller volume of rock than those elements measured by the gamma spectroscopy tool. As a result, with increasing hole size, the aluminum signal decreases rapidly and may reach background levels, whereas the gamma spectroscopy tool elements can still be measured. This problem is compounded by the oxide closure procedure, which forces the major oxides to a constant sum (usually 100%).
• Temperature. Temperature effects can significantly reduce the efficiency of the NaI detector in the gamma ray spectrometry tool, with higher temperatures resulting in a poor signal-to-noise ratio and decreased resolution. Poor resolution will result in gamma-ray peaks appearing in the wrong window and lead to incorrect identification of the element represented. It is recommended that the GLT not be used in temperatures greater than 150°C.

The quality of the data can also be reduced during processing. This can occur due to errors in the spectral inversion of the raw data, inaccuracies in the oxide closure model caused by the presence of unmeasured elements and incorrect oxide factor assumptions. In ODP, shipboard data (particularly petrographic, chemical and diffraction) can often be used to minimize these errors.

One limitation of the GLT is its relatively low spatial resolution. The volume sampled by the GLT approximates to a sphere, with a radius varying from around 0.3-1.0 m, depending on lithology, porosity, composition of the pore fluids and the elemental spectra being determined. At each measurement point (every 15 cm) a number of these spherical samples are averaged. The raw data from the GLT have, therefore, already undergone a certain amount of smoothing. This accentuates the shoulder effect on the logs, which tends to smooth the log responses over sharp lithological boundaries.

Comparisons between GLT derived oxide estimates and similar data obtained from conventional geochemical analyses (e.g. XRF) on core samples should be treated with extreme caution. The two techniques measure substantially different volumes of rock. Furthermore, it is always difficult to precisely match the depths of the core samples with those of the log values, especially when core recovery is low.

Applications

Lithology.

In basement, variations in elemental concentrations will help delineate flow boundaries and characterize alteration vein-filling. In sedimentary environments, where there is a reasonable chemical variation in the rocks, GLT data can be used as an effective indicator of changes in the lithostratigraphy.

Cyclically interbedded lithologies can be identified and analyzed using geochemical logging, and changes in the provenance of sediments can be shown. For example, the FeO, SiO₂ and CaCO₃ results from ODP Hole 950A on the Madeira abyssal plain show distinct downhole alternations (see figure). Horizons which are generally rich in FeO, rich in SiO₂ and poor in CaCO3 show the position of clay-rich organic and volcanic distal turbidites, sourced from volcanic islands and the African margin, to the east of the drill-site. Horizons generally poor in FeO, poor in SiO₂, and rich in CaCO₃ show the position of calc-turbidites, sourced from a chain of seamounts to the west of the plain.

The ratio of certain elemental yields can also be used to emphasize fluctuations or distinct marker horizons in the stratigraphy. For example, elemental yield ratios were used to analyze data from ODP Hole 999B, drilled beneath the Caribbean Sea. The lithology (Si/(Si + Ca)), iron (Fe/(Si + Ca)) and porosity (H/(Si + Ca)) indicator ratios all help to highlight the position of tephra horizons within the formation (see figure).

 

Geochemistry.

Downhole fluctuations in the elemental yields reflect gross variations in geochemistry, which can be used to help categorize the formation. The GLT results from Hole 735B, logged during Leg 118, show a good example of this. This hole penetrated basement of the Southwest Indian Ridge, which between 50-400 mbsf can be subdivided into four distinct units (see figure). The geochemical data clearly delineate Unit 4, which is a Fe-Ti oxide-rich gabbro. Generally low and uniform FeO and TiO₂ values occur in Unit 5, which is a relatively uniform olivine gabbro.

 

Mineralogy.

The oxide data, in combination with data from other logs if appropriate, can be inverted to estimate the proportions of the main minerals in the rock. This information, which can be displayed as mineralogical logs, can often be used to derive other physical properties of the formation, such as magnetic susceptibility or cation exchange capacity (CEC).

Limitations

Environmental Effects:

In holes with severe washouts and in high porosity sediments the signal can be dominated by Al (which has a very high capture cross-section) and H, with very little contributions from the formation. When this occurs, the resulting yields and derived elements are less accurate due to poorer statistics. Likewise, in through pipe or cased holes, the signal is dominated by elemental readings of iron from the pipe and very little of the overall signal comes from the formation.

Log Presentation

The standard presentation of the unprocessed geochemical data includes the elemental yields (CCA, CSI, CSUL, CFE, CHY, CCHL) displayed in decimal fraction along with the formation capture cross section (CSIG) displayed in capture units, and the aluminum curve (ALUM), as wet weight %, from the Aluminum Activation Clay tool .

Tool Specifications

Temperature Rating	300°F (149° C)
Pressure Rating	20 kpsi (13.8 kPa)
Tool Diameter	3 7/8 in. (10 cm)
Tool Length	9.25 ft (2.82 m)
Sampling Interval	6 in (15.24 cm)
Max. Logging Speed	600 ft/hr