MAGNETIC
RESULTS
The results of
measurements 1 through 5 (see preceding paragraph) are given in Tables T1
and T2.
Measurements 6 and 7 are discussed separately.
Whole-Core
Pass-Through Measurement: Natural Remanent Magnetization Intensity and Magnetic
Susceptibility
Within the five drilling
sites (1065, 1067, 1068, 1069, and 1070), where rock magnetic screening of
samples was made, there are considerable variations in magnetic properties and
demagnetization behavior among the various lithologies, which will be described
elsewhere. The most common features, however, can be summarized as follows: A
remagnetization imparted by the coring process is commonly encountered as noted
during previous legs (e.g., Gee et al., 1989; Zhao et al., 1994). This
remagnetization is characterized by NRM inclinations that are strongly biased
toward vertical value (+90°) in many cores. This remagnetization most severely
affected the external portions of the cores (presumably because the outside of
the core is physically closer to the magnetized core barrel). As shown in
Figures F2 and F3,
upon demagnetization to 60 mT, a significant decrease in intensity and a shift
of inclination toward negative values were observed, suggesting the presence of
drilling-induced remagnetization. At all Leg 173 sites, variations in magnetic
susceptibility generally parallel the variations in NRM intensity (e.g., see
Fig. F2). The
general trend of the susceptibility data curve was used to characterize the
magnetic material contained within the cored material and for correlating
sedimentary intervals between selected drilled sites. A good example was
presented during shipboard study (Shipboard Scientific Party, 1998, p. 207)
where sections of magnetic susceptibility profiles from Holes 1068A and 1067A
can be correlated.
AF
and Thermal Demagnetization of Discrete Samples
To investigate the nature
of the remanent magnetization of the discrete samples from Leg 173 sites,
selected samples were stepwise AF or thermally demagnetized. The NRM intensity
or direction of the minicore samples did not change significantly (<5%) after
zero-field storage for 2 weeks. As mentioned above, the vertically directed
drilling-induced magnetization is often present in Leg 173 cores. In most cases,
this steeply downward component of magnetization is not very resistant to AF
demagnetization (e.g., see Fig. F4A).
Thermal demagnetization of minicore samples also successfully removed this
drilling-induced magnetization component (Fig. F4B).
As shown in Figure F4B,
the drilling-induced remagnetization component (with inclination >75°) is
removed after 250°C demagnetization and a characteristic component (with
inclination ~50°) can be identified. As mentioned, we used magnetic
susceptibility to monitor the production of new magnetic materials during
thermal demagnetization that might have altered the remanence. Apart from small,
insignificant fluctuations, the susceptibility of minicore samples generally did
not change until after they had been heated to 350°C. Above this temperature,
many samples showed a decrease or increase in susceptibility. In addition,
progressive thermal demagnetization on several minicores of metasediments and
serpentinized peridotite revealed that a component of viscous remagnetization (VRM)
parallel to the present-day magnetic field has been recorded in these rocks. The
measured VRM has been demonstrated to be useful for orienting cores for
structural studies (see the Site 1067-1070 chapters in Whitmarsh, Beslier,
Wallace, et al., 1998). Examples of this VRM application are given in Table T3.
Results from AF and
thermal demagnetization of basement samples from Sites 1067, 1068, and 1070 show
behavior that cannot be interpreted in such a simple fashion. As shown in Figure
F5, some of
these basement samples exhibited signs of drilling-induced magnetization as
evidenced by the steep inclinations in the initial demagnetization measurements
(AF demagnetization to 15 mT effectively removed this overprint; see Fig. F5A).
Others show a "soft" VRM component that was removed at low-temperature
demagnetization steps (NRM = 300°C), followed by a stable component of
magnetization (300° to 566°C demagnetization; see Fig. F5B).
Although not every section was sampled and measured, it seems reasonable to
conclude that drilling has had a lesser effect on the NRM of basement samples as
compared with that of sedimentary cores.
It is interesting to note
that the stable component for serpentinized peridotite samples from Site 1068
has a much shallower inclination compared to the expected inclination at the
drilling site. The magnetically cleaned inclinations (mean = 42.9°; N = 17; 
95
= 3.7°) in the serpentinized peridotites are systematically shallower than the
inclination expected today (59°) at the drilling sites but are statistically
indistinguishable from the Jurassic-Cretaceous inclinations for Iberia (33.4°-
45.2° and 95
~10°; see Van der Voo, 1969; Galdeano et al., 1989). A similar observation was
also made in serpentinized peridotites from Site 1070. Assuming the inclination
represents the primary remanence at the time when these rocks were formed, the
similarity between the observed and the expected inclinations is consistent with
the notion that the drill sites were part of the Iberia plate at the time of
acquisition of the magnetization. Alternatively, the shallow inclination could
indicate that these sections have been tilted after the acquisition of the
magnetization, or a combination of both.
Koenigsberger
Ratio
The Koenigsberger ratio, Q,
is defined as the ratio in a rock of remanent magnetization to the induced
magnetization in the Earth's field. In general, the Koenigsberger ratio is used
as a measure of the stability to indicate a rock's capability of maintaining a
stable remanence. The International Geomagnetic Reference Field value at the Leg
173 site (45,000 nT = 35.83A/m) was used for calculating Q, where
- Q
= NRM (A/m)/(k
[SI] × H
[A/m])
and H
is the local geomagnetic field. The variation of the Koenigsberger ratios in
Table T1 in
general resembles that of the NRM. For example, the serpentinized peridotite
samples from Site 1070 have a higher intensity of remanence than the overlying
sedimentary rocks, and consequently the Koenigsberger ratio of the former is
higher than that of the latter. Similar examples of this correlation are also
seen in samples from other Leg 173 sites. The Koenigsberger ratios for many
claystone and nannofossil chalk samples from Leg 173 sites are <1.0,
indicating that induced magnetization would be more than or comparable to
remanent magnetization. The low value of the Koenigsberger ratio for these
samples also indicates the presence of low-coercivity magnetic minerals that
carry an unstable remanence and are more susceptible to an external magnetic
field. The other parameters measured are the median-demagnetizing field or the
unblocking temperature (Tb),
both of which represent the stability of remanence.
Low-Field
Magnetic Susceptibility
Volume magnetic
susceptibility of natural materials in a weak magnetic field depends on the
abundance and grain size of ferromagnetic minerals. As listed in Table T1,
discrete samples of sedimentary rocks show low magnetic susceptibility values,
indicating a very low concentration of ferro(i)magnetic minerals in these rocks.
The susceptibility of serpentinized peridotite samples from Sites 1067, 1068,
and 1070, on the other hand, is much higher than that of the overlying
sediments. This is expected because they are known to have very different
magnetic mineralogies. In this study, initial susceptibility was often routinely
measured after each measurement step during thermal demagnetization in order to
monitor changes in magnetic mineralogy during heating. For example, maghemite to
hematite transition at ~300°C produces a decrease in susceptibility, which is a
useful means of identifying the presence of this mineral (Opdyke and Channell,
1996).
Hysteresis
Loop Parameters
Saturation magnetization
(Js), saturation remanence (Jr), coercive force (Hc),
and remanent coercivity (Hcr, from back-field experiments) are
parameters that can be determined from a hysteresis loop. Hysteresis loop
parameters are useful in characterizing the intrinsic magnetic behavior of
rocks. Thus, they are helpful in studying the origin of remanence. In this
study, hysteresis loops and the associated parameters Jr/Js,
Hc, and Hcr were obtained using an alternating gradient
magnetometer (Princeton Measurements Corporation) capable of resolving magnetic
moments as small as 5 × 10-8 emu (Flanders, 1988). Hysteresis
parameters determined from 12 representative samples from Leg 173 sites are
presented in Table T2.
Saturation magnetization (Js) is a measure of the total amount of
magnetic mineral in the sample. The coercivity, Hc, is a measure of
magnetic stability. The two ratios, Jr/Js and Hcr/Hc,
are commonly used as indicators of domain states and, indirectly, grain size.
For magnetite, high values of Jr/Js (>0.5) indicate
small (<0.1 µm or so) single-domain (SD) grains, and low values (<0.1)
are characteristic of large (>15-20 µm) multidomain (MD) grains. The
intermediate regions are usually referred to as pseudo-single domain (PSD). Hcr/Hc
is a much less reliable parameter, but conventionally SD grains have a value
close to 1.1, and MD grains should have values >3-4 (Day et al., 1977).
Figure F6
displays the ratios of the hysteresis parameters for Leg 173 samples containing
mainly titanomagnetite as magnetic minerals plotted on a Day et al. (1977)-type
diagram. Such a representation provides qualitative information on the magnetic
grain sizes from SD to PSD to large MD. The samples analyzed in this study
indicate that the magnetic grain sizes of the Site 1069 samples fall near the
boundary between SD and PSD, whereas the samples from Sites 1070 and 1068 are in
the PSD range. The Fe-Ni metal Sample 173-1068A-26R-4W, 18-20 cm (awaruite), is
an exception that falls in the MD region. Hysteresis behavior of three yellow
chalk samples from Site 1069 and two amphibolite breccia samples from Site 1067
were also studied, but the results are complex and are not readily explained at
present.
Low-Temperature
Properties
Transitions in the
magnetic properties of magnetite, pyrrhotite, and hematite occur at low
temperatures and they provide a potential means of magnetic mineral
identification. Magnetite exhibits a crystallographic phase transition from
cubic to monoclinic at 110-120 K. Associated with this transition, the
anisotropy constant goes through zero as the easy axis of magnetization changes
from [100] to [111] (Nagata et al., 1964). Low-temperature measurements were
made on 16 representative samples to help characterize the magnetic minerals and
understand their rock magnetic properties. These measurements were designed to
determine the Néel temperature and other critical temperatures of a magnetic
substance and were made from 10 K to room temperature on 100-300 mg subsamples
in a Quantum Design Magnetic Property Measurement System (MPMS) at the
University of Minnesota. By definition, a ferrimagnetic mineral grain is
superparamagnetic if its volume is smaller than the critical value 25 kT/K
(where k = Boltzmann's constant, T
= 300 K in this study, and K = the magnetic anisotropy constant per unit
volume), so that its net remanence over 100 s is zero (Cullity, 1972). However,
as T is
decreased toward 0 K, the thermal energy kT
decreases and K increases (both serving to aid magnetic stability) so that all
grains that were superparamagnetic at 300 K will be able to retain thermally
stable remanent magnetization near 0 K (Banerjee et al., 1992). The so-called
Verwey transition in magnetite can be observed by a decrease in intensity of an
IRM at low temperature as it is allowed to warm through or cool through the
transition temperature. In this study, these experiments included (1) cooling
the sample from room temperature (300 K) down to 10 K (in some cases to 5 K) in
a steady magnetic field of 2.5 T and measuring the remanence at 5 K intervals;
(2) measuring the saturation isothermal remanence magnetization (SIRM) at 5 and
10 K (SIRM5 and SIRM10, respectively) and then warming it to 300 K in zero field
while measuring the remanence value every 5 K.
As shown in Figure F7,
the low-temperature curves of SIRM both in zero-field warming and in a 2.5 T
field cooling display a variety of features. These include an unblocking
temperature in the vicinity of 40-60 K, most likely caused by ferrimagnetic
pyrrhotite or greigite (Dekkers et al., 1989; Rochette et al., 1990) and a
decrease in remanence in the 100-120 K range, most likely caused by the
magnetocrystalline anisotropy constant, k1, of magnetite going to
zero in this temperature, known as the Verwey transition (Verwey et al., 1947).
The low-temperature data are one of the major lines of evidence for the presence
of Fe-Ni metals in Site 1068 peridotite, (titano)magnetites in the
"fresher" peridotites of Site 1070, and maghemites in the
"altered" part of the peridotite section (Zhao, 2000).
Curie
Temperature
Curie temperature is the
temperature below which a magnetic mineral is magnetically ordered. Because this
value is a sensitive indicator of composition, it is useful in understanding the
magnetic mineralogy. In this study, Curie temperature was determined by
measurement of magnetic moment vs. temperature (using the Princeton MicroMag
Vibrating Sample Magnetometer at the University of Minnesota), because the
magnetic moment drops to zero about the Curie temperature. We conducted
thermomagnetic analyses in an inert atmosphere on 16 samples chosen to be
representative of the Leg 173 cores.
Figure F8
shows high-temperature magnetic moment runs of a representative
"altered" and a "fresher" peridotite sample from Site 1070.
The heating and cooling curves for the "altered" sample display a
significant drop of magnetic moment ~420°C (Fig. F8A).
This drop may be indicative of a fraction of maghemite, which could be
responsible for the observed remanent magnetization. For the "fresher"
peridotite sample, the results show Curie temperatures between 550° and 580°C,
indicative of the presence of titanomagnetite (Fig. F8B).
Low-Temperature
AC Susceptibility Measurements
In a nonlinear and
sinusoidally varying applied field, the magnetic response is determined by
factors including field dependence and time or frequency dependence (i.e.,
viscosity) (Jackson et al., 1998). Previous work by Worm et al. (1993) and
Jackson et al. (1998) has suggested that ferrimagnetic pyrrhotite and
titanomagnetites exhibit strong field dependence of AC susceptibility in a large
applied field. To investigate the field- and frequency-dependent susceptibility
of the Leg 173 cores, AC susceptibility measurements were made on selected
samples with a LakeShore Model 7130 AC susceptometer at the University of
Minnesota. The temperature dependence of the in-phase (
)
susceptibility between 15-300 K was measured at two frequencies (95 and 1000
Hz), with AC field amplitudes between 100 and 1000 A/m.
The main observations in
this study can be outlined as follows:
- The field- and frequency-dependent susceptibility curves for the
"fresher" peridotites from Site 1070 are similar to those of
synthetic titanomagnetite 55 (TM55) (Jackson et al., 1998), with some field
dependence and no frequency dependence.
- For the "altered" peridotite, the field- and
frequency-dependent relationships are directly opposite to those in the
"fresher" zone, with some frequency dependence but no field
dependence.
- There is no detectable frequency dependence of susceptibility or
its field dependence for these awaruite-bearing samples from Site 1068,
probably because their susceptibility is a superposition of paramagnetic,
antiferromagnetic, and ferrimagnetic susceptibilities from different
Fe-bearing phases.
Figure F9
shows an example of the temperature dependence of magnetic susceptibility
between 15 and 300 K for an "altered" peridotite sample. More examples
and detailed descriptions can be found in Zhao (2000).
