Gubbins D., Herrero-Bervera E. Encyclopedia of Geomagnetism and Paleomagnetism

Подождите немного. Документ загружается.

minerals siderite and rhodochrosite (38 and 32 K, respectively). It is

thus advised to use SIRM demagnetization only for a reconnaissance

study of mineral phases present. A useful method to distinguish

between the two above cases is to measure a SIRM acquired at room

temperature during a zero-field cycle to a low temperature. Magnetic

phase transitions will be recorded in this experiment, while Néel tem-

peratures will not.

For some materials low-temperature magnetic properties measured on

warming may depend on the mode of preceding cooling, i.e., on whether

a sample has been cooled in a zero (zero-field cooling, ZFC) or in a strong

magnetic field (field cooling, FC). In particular, this is the case for mag-

netically hard minerals, in which thermoremanent magnetization (TRM)

acquired during cooling in a strong field to a given low temperature

is much higher than a SIRM, which can be induced by a single applica-

tion of the same field at the same temperature. Demagnetization curves

of the two remanences will then be strongly discrepant but converge near

room temperature. Another variant of ZFC vs FC measurements, which

must not be confused with the former one, consists in measuring magne-

tization during warming in a relatively weak, usually 2–20 mT, DC field,

after sample being cooled once in a zero and second time in the DC field

equal to the measurement field. This method can be used to characterize

the superparamagnetic behavior. The two magnetization vs temperature

curves would diverge if the blocking of superparamagnetic moments

occurs below room temperature.

Measurements of magnetic hysteresis loops as a function of tem-

perature potentially would yield the most detailed information on the

magnetic mineralogy of the sample. However, this is a rather time-

consuming experiment, and relatively little has been reported so far

on low-temperature hysteresis properties of minerals and rocks.

Variation of AC susceptibility at low temperatures is also a useful

characteristics of the magnetic mineralogy. An important advantage of

susceptibility measurements is that they allow to detect minerals that

show antiferromagnetic ordering below Néel temperatures. On the other

hand, low-temperature susceptibility signal due to ferrimagnetic and

antiferromagnetic phases is progressively masked by a contribution

from paramagnetic minerals which increases proportionally to 1/T.

The remainder of this article presents an overview of low-

temperature magnetic properties of most common iron minerals found

in terrestrial rocks.

Magnetite and maghemite

Magnetite (Fe

), iron ferrite, is the longest known and most ubiqui-

tous natural magnetic material and probably also the most extensively

studied. It occurs, in a varying concentration, in most volcanic, sedi-

mentary, and metamorphic rocks, and forms economically significant

iron ore deposits. Furthermore, magnetite can be formed biogenically

by a wide variety of organisms, from prokaryotes to humans (cf.

Kirschvink et al., 1985).

Low-temperature anomalies in the magnetic susceptibility and mag-

netization of magnetite were discovered in 1910–1920s (Weiss and

Renger, 1914; Weiss and Forrer, 1929). However, only much later it

was proposed (Verwey, 1939; Verwey and Haayman, 1941) that these

anomalies, together with those in electrical properties, heat capacity,

etc., are due to the phase transition, which is now called the Verwey

transition. The transition is accompanied by a slight monoclinic dis-

tortion of the cubic spinel crystalline lattice of magnetite stable at

ambient temperature, which results in a large, by over an order of mag-

nitude, increase of the magnetocrystalline anisotropy. The transition

temperature (T

) is 125 K in stoichiometric crystals with low internal

stress (Walz, 2002). Substitution of iron with vacancies (Aragón et al.,

1985; Aragón, 1992; Özdemir et al., 1993) and foreign metal cations

(Miyahara, 1972; Kozłowski et al., 1996a,b; Brabers et al., 1998)

are known to lower the T

and eventually suppress the transition.

Stresses have the same effect (Samara, 1968; Rozenberg et al.,

1996). The low-temperature behavior of the magnetic properties, and

particularly of a remanent magnetization during cooling from room

temperature in zero magnetic field, is also strongly affected by the iso-

tropic point at about 130 K, the temperature where the first constant of

magnetocrystalline anisotropy K

momentarily turns zero (Bickford,

1950; Syono, 1965). The isotropic point temperature appears much

less sensitive to nonstoichiometry than the T

(Kąkol and Honig,

1989; Kąkol et al., 1994). A review of low-temperature magnetic

properties of magnetite from the point of view of rock magnetism

has been recently given by Muxworthy and McClelland (2000a).

The Verwey transition is easily recognizable in the demagnetization

curves of a SIRM acquired at a low temperature (e.g., 10 K), as a sharp

drop of magnetization (Figure M38). The temperature where the mag-

netization decrease rate is maximal is usually taken as a transition tem-

perature. This has a justification in studies on large nearly perfect

synthetic magnetite crystals (Aragón, 1992), where the magnetization

drop occurs within 1 K. However, in fine grains magnetization

decreases more gradually, particularly when magnetite is slightly oxi-

dized (Figure M39, Özdemir et al., 1993), and such an assignment

of the transition temperature may be somewhat arbitrary. Perhaps for

this reason the exact values of T

in rocks are seldom reported, and

the observation of the transition is taken as only a qualitative test for

the presence of magnetite. Meanwhile, actual values of transition tem-

peratures can be of interest because of their sensitivity to minute

variations of oxygen stoichiometry and metal substitution, thus

providing information inaccessible with other methods. Recent exam-

ples that illustrate use of the low-temperature magnetometry of magne-

tite to solve magnetomineralogical problems include: (i) a method to

detect the presence of biogenic magnetite produced by magnetotactic

bacteria, showing a distinctive difference between the demagnetization

curves of SIRMs given below the Verwey transition after ZFC and FC,

respectively (Moskowitz et al., 1993); (ii) a model explaining the mag-

netic susceptibility signal in Chinese loess by the effect of natural

grass fires (Kletetschka and Banerjee, 1995); and (iii) tracing of che-

mical alteration of goethite (a-FeOOH) during heating (Özdemir and

Dunlop, 2000).

Magnetite and magnetite-bearing rocks can be further characterized

by a variation of a remanence acquired at room temperature during a

zero-field cycle crossing the Verwey temperature. On initial cooling,

Figure M38 Change in the remanent moment obtained by

cooling in a magnetic field of 1 T to 4.5 K during warming in a

zero field, for samples of Fe

3d

with d ¼ 0.000 (squares),

0.003 (triangles), and 0.006 (circles). (After Arago

n (1992)

ã American Physical Society, with the permission of the publisher.)

516 MAGNETIC PROPERTIES, LOW-TEMPERATURE

magnetization is lost between ca. 200 K and the T

; demagnetization is

grain-size dependent, being by far the most effective in large multido-

main grains, where it is essentially complete at the isotropic point

(Dunlop, 2003). Below the transition temperature, magnetization is

nearly constant until on warming the temperature reaches the vicinity

of T

. The behavior on crossing the T

from below depends primarily

on stress: in low-stress samples remanence tends to be further lost

(Muxworthy and McClelland, 2000b), while highly stressed ones

show a rebound (Heider et al., 1992; King and Williams, 2000). In

the paleomagnetic practice, cycling the rock sample through the isotro-

pic point and Verwey temperature is known as the low-temperature

cleaning (LTC, Ozima et al., 1964). LTC should at least partly remove

the unwanted secondary components of NRM carried by large multi-

domain grains of magnetite. However, perhaps because physical

mechanisms of remanence demagnetization during LTC remained elu-

sive for so long (Muxworthy and McClelland, 2000b; Dunlop, 2003),

it has been seldom used in paleomagnetic studies.

Magnetic properties of the low-temperature monoclinic phase of

magnetite, although of no direct importance in paleomagnetism,

proved to be an interesting subject of study itself, receiving much

recent interest (Schmidbauer and Keller, 1996; Muxworthy, 1999;

Özdemir, 2000; Kosterov, 2001, 2002; Özdemir et al., 2002; Smirnov

and Tarduno, 2002; Kosterov, 2003). In particular, it appears that below

the Verwey transition temperature dependences of the magnetic hyster-

esis parameters are qualitatively different in truly multidomain and in

smaller, pseudosingle-domain grains, respectively: in multidomain

grains both M

and the coercive force are much higher after a ZFC while

a more complex relationship between ZFC and FC hysteresis para-

meters is observed in pseudosingle-domain grains (Kosterov, 2001,

2002). Magnetic properties of the low-temperature phase also strongly

depend on magnetite stoichiometry (Özdemir et al., 1993; Kosterov,

2002, 2003), which is a result of the variation of the magnetocrystalline

anisotropy with the degree of oxidation (Kąkol and Honig, 1989).

Memorizing the value of a small to moderate magnetic field ( <100

mT) applied during cooling through the Verwey transition has been

demonstrated (Smirnov and Tarduno, 2002).

Maghemite (g-Fe

) is a fully oxidized analog of magnetite, in

which 1/9 of iron cations in the spinel lattice are replaced with vacan-

cies. It is a common constituent of soils (Taylor and Schwertmann,

1974) and also has been found in deep-sea sediments (Harrison and

Peterson, 1965). Unlike magnetite, maghemite does not show phase

transitions below room temperature, and remanence given at a given

low temperature decreases monotonously on warming (Morrish

and Watt, 1958; Moskowitz et al., 1993). SIRM given at room tem-

perature changes reversibly on ZFC to a low temperature (de Boer

and Dekkers, 1996).

Titanomagnetites

Titanomagnetites are spinel iron-titanium oxides of the composition

3x

. Stable single-phase oxides exist for x between 0 (magne-

tite) and 0.96 (Kąkol et al., 1991a). The end member with nominal

composition Fe

TiO

is known as mineral ulvöspinel. A titanomagne-

tite with x  0.6 (TM60) is the dominant magnetic mineral in fresh

submarine basalts; in subaerial basalts, unoxidized titanomagnetites

typically have compositions TM60-TM70 (Dunlop and Özdemir,

1997, Section14.4). Ti-rich titanomagnetites have only a local signifi-

cance in sediments, where the eroded material from volcanic rocks is

readily available. Titanomagnetites of intermediate compositions are

found relatively less frequently in nature than those close to magnetite

and TM60, respectively.

Titanomagnetites with compositions richer in Ti than TM04 do not

exhibit the Verwey transition (Kozłowski et al., 1996a). However,

the variation of the magnetocrystalline anisotropy remains similar to

that in magnetite for the wide range of compositions. At room tem-

perature, K

is negative; with the decrease of temperature its absolute

value first increases and then starts to decrease approaching the isotro-

pic point. With the increase of x the isotropic point temperature first

shifts toward lower temperatures, being below 77 K for the material

in the TM18-TM36 range (Syono, 1965; Kąkol et al., 1991b), and

then starts to increase again, so that the material with the nominal com-

position TM41 has the isotropic point of about 125 K, very close to

that of magnetite (Kąkol et al., 1991b). For even more Ti-rich compo-

sitions, K

is positive at all temperatures below 300 K (Syono, 1965;

Kąkol et al., 1991b) and increases sharply to very high values below

200 K.

Low-temperature dependences of SIRM and low-field susceptibility

for millimeter-size single crystals of titanomagnetite in the composi-

tion range from TM0 to TM60 have been reported by Moskowitz

et al. (1998, Figure M40). Titanomagnetites from TM05 to TM28

show a sharp decrease of SIRM between 40 and 80 K. Low-field sus-

ceptibility of these titanomagnetites increases rapidly in this tempera-

ture range. For TM05 it reaches a maximum at about 90 K and then

slightly decreases; for TM19 and TM28 the susceptibility increase just

becomes considerably slower above 110 K. Since the material under

study is multidomain, this behavior is likely controlled by the tempera-

ture variation of magnetocrystalline anisotropy. Compositions around

TM40 show a somewhat more complex behavior: SIRM first drops

rapidly in about the same range as for more Ti-poor compositions,

and then shows further, less pronounced decrease between 120

and 150 K. The latter should be due to the isotropic point at about

125 K (Kąkol et al., 1991b). In even more Ti-rich titanomagnetites

TM55-TM60 SIRM decreases rapidly between 50 and 70 K, and more

Figure M39 Normalized saturation isothermal remanence (SIRM)

curves of submicron magnetites during warming from 5 to

300 K in zero field. Top: reduced (apparently stoichiometric)

magnetites; bottom: surface-oxidized magnetites. The grain size

is indicated on the curves. (After O

zdemir et al. (1993)

ã American Geophysical Union, with the permission of the

author and the publisher.)

MAGNETIC PROPERTIES, LOW-TEMPERATURE 517

gradually above this temperature, bearing some resemblance to the

behavior of ultrafine superparamagnetic grains.

Magnetic hysteresis of intermediate titanomagnetites as a function

of low temperature has been studied only on millimeter-sized single-

crystal samples, for a rather limited number of compositions from

TM40 to TM80, and in the magnetic fields not exceeding 1.5 T, which

is likely to be too low to fully saturate the samples, especially more

Ti-rich ones (Tucker, 1981; Schmidbauer and Readman, 1982). The

most distinctive feature of these compounds is a large increase of the

coercive force with decreasing temperature, which is due to the sharp

rise of the magnetocrystalline anisotropy. For compositions from

TM50 to TM80 the rise of the H

starts at about 150 K and for

TM40 at 70 K (Schmidbauer and Readman, 1982). In addition, both

and the coercive force appear to depend on the mode of preceding

cooling, the latter being much higher and the former considerably

lower after ZFC.

It must be borne in mind that the above experiments were done on

large crystals, necessarily multidomain. To the best of author’s knowl-

edge, no low-temperature magnetic studies have been carried out on

synthetic intermediate titanomagnetites of micron to submicron grain

size. The low-temperature data on rocks, which contain such grains,

are also scarce. Moskowitz et al. (1998) have reported contrasting

low-temperature SIRM(T) curves for basaltic samples from the center

and a chilled margin of an Icelandic dike, respectively. Samples from

the center of the dike yielded curves closely resembling those for syn-

thetic Ti-rich (x > 0.4) titanomagnetites, while samples from the dike

margin showed curves with a break-in-slope around 50 K. It may be

suggested that rapid cooling of the dike margin had produced a titano-

magnetite phase with x  0.2, and the discrepancy between SIRM(T)

curves of these samples and those of synthetic titanomagnetite of simi-

lar composition is due to a difference in grain size. Similar SIRM(T)

curves have been observed in late Pleistocene to Holocene sediments

from Zacapu basin in central Mexico, where the abundant titanomag-

netite of composition ranging from TM20 to TM55 comes from

nearby volcanic activity (Ortega et al., 2002). Further complications

are likely to arise if titanomagnetites are oxidized to form titanomaghe-

mites, as is the case for most submarine basalts older than several mil-

lions of years. In such rocks, Matzka et al. (2003) have observed even

a self-reversal of SIRM given at room temperature during a low-tem-

perature cycle in zero field. Magnetization crossed to negative values

around 70 K during the cooling leg but returned reversibly to initial

values upon the subsequent warming. In summary, much has yet to

be done to fully understand the low-temperature behavior of naturally

occurring intermediate titanomagnetites.

Titanohematites

The composition of minerals belonging to the titanohematite series can

be described by the chemical formula Fe

2y

, the end members

being minerals hematite (a-Fe

) and ilmenite (FeTiO

), respectively.

These minerals crystallize in the rhombohedral system; the space group

describing crystal symmetry is R

3C for y < 0.5 and R3 for y > 0.5.

Compositions close to both hematite and ilmenite are quite common

in igneous rocks, while truly single-phase titanohematites of an inter-

mediate composition (0.2 < y < 0.8) can be obtained only in the labo-

ratory by quenching from high temperatures. Natural intermediate

titanohematites of nominal composition 0.5  y  0.7, found in rapidly

chilled dacitic pyroclastic rocks, received much interest in rock magnet-

ism due to the ability to acquire a self-reversed TRM (Uyeda, 1958;

Hoffman, 1992). Magnetic properties of titanohematites are deter-

mined by their composition: (i) for 0  y < 0.5 an overall ordering is

antiferromagnetic with a weak parasitic ferromagnetism due to a spin

canting (Dzyaloshinsky, 1958; Moriya, 1960); (ii) compositions with

0.5 < y < 0.7 are ferrimagnetic; (iii) titanohematites with 0.7 < y 

0.93 show rather complex magnetic structures, which will be discussed

in more detail below; (iv) compositions with y greater than about 0.93

show antiferromagnetism. Curie temperatures of titanohematites

decrease approximately linearly with increasing ilmenite content up

to y  0.93, reaching the values below room temperature at y  0.75

(Nagata, 1961).

Hematite is the most stable iron oxide in oxidizing conditions

and, as such, is found in many sedimentary and in some highly

oxidized volcanic rocks. Its small net saturation moment of about

0.4 A m

1

(Morrish, 1994) results primarily from a spin canting.

Another mechanism that can produce a magnetic moment is a presence

Figure M40 Normalized saturation isothermal remanence (SIRM)

curves of synthetic titanomagnetites during zero-field warming

from 5 to 300 K: (a) TM0-TM41 and (b) TM55-TM61. SIRM was

given in a 2.5 T field at 5 K. Single-crystal samples are denoted

by solid symbols and polycrystalline samples are denoted by

open symbols. (c) Normalized in-phase susceptibility curves

of synthetic titanomagnetites during warming from 15 to 300 K.

Single-crystal samples are TM0, TM05, TM19, TM28, TM41,

TM55, and polycrystalline samples are TM40 and TM60. The

in-phase susceptibility was measured at 1000 Hz. (After

Moskowitz et al. (1998), with the permission of the author and the

publisher, Elsevier.)

518 MAGNETIC PROPERTIES, LOW-TEMPERATURE

of defects or impurity atoms in the hematite crystalline lattice (Bucur,

1978). Since hematite is much less strongly magnetic than magnetite

or maghemite, the two latter minerals dominate the NRM in the case

when concentrations of these minerals are comparable. However, in

an important class of sedimentary rocks called red beds, NRM is

almost entirely carried by hematite.

Below room temperature, bulk hematite exhibits the so-called Morin

transition (Morin, 1950). At the Morin transition spin directi ons

change from the basal plane to the rhombohedral c-axis, and the spin

canting, and the magnetic moment associated with it, seems to vanish.

The defect moment, however, remains largely intact. This explains, at

least qualitatively, the peculiar behavior of room temperature rema-

nence being cycled in a zero field through the Morin transition (Haigh,

1957; Gallon, 1968; Bucur, 1978): magnetization lost on cooling is lar-

gely, but not completely, restored on warming ( Figure M41). Still

unresolved is an existence of the transition temperature hysteresis dur-

ing magnetization cycling in a zero field. Transition temperatures

determined from cooling curves are often lower than those determined

from warming curves, sometimes by as much as 20



Alike the Verwey transition in magnetite, the Morin transition is

sensitive to impurities. Titanium is known to suppress the transition

in concentrations as low as less than 1% (Kaye, 1962), while the

transition exists until 10% of Al substitution (da Costa et al., 2002).

The transition becomes more gradual for highly aluminous hematites,

in the sense that low- and high-temperature phases appear to coexist

over a considerable range of temperatures, and the effective transition

temperature is lowered to about 180 K for hematite with 10% Al.

The transition temperature is lowered and the transition itself appears

considerably suppressed in fine grains (Bando et al., 1965; Schroeer

and Nininger, 1967; Nininger and Schroeer, 1978; Amin and Arajs,

1987). This may be the reason why the Morin transition is rarely found

in hematite-bearing rocks (Dekkers and Linssen, 1989).

Titanohematites with y up to about 0.7 do not show any distinctive

low-temperature magnetic properties (Ishikawa and Akimoto, 1957).

On the contrary, more Ti-rich compositions show fairly complex beha-

vior (Figure M42). At very low temperatures, a spin glass structure

seems to arise (Arai and Ishikawa, 1985; Arai et al., 1985a,b; Ishikawa

et al., 1985), the spin glass transition temperatures T

being consider-

ably lower than the respective Curie temperatures. Between these two

transition points compositions with y  0.8– 0.9 show superparamag-

netic behavior (Ishikawa et al., 1985). The spin glass phase is capable

to carry a relatively strong remanence, which disappears at the T

. The

coercive force also increases sharply below this temperature reaching

300 –600 mT at 4.2 K. It must be noted, however, that the above

studies have been carried out using the synthetic materials. Whether

their natural analogs would show similar magnetic properties remains

largely unknown.

The second end member of the titanohematite series, ilmenite,

orders antiferromagnetical ly below its Néel temperature of 57  2K

(Senftle et al., 1975). Antiferromagnetism in the low-temperature

phase persists up to about 6.6% of Fe

3þ

substitution, while Néel

temperatures shift to below 50 K (Thorpe et al., 1977). Ilmenite there-

fore contributes only to induced magnetization. A susceptibility

peak at 45 K, which may be due to ilmenite, has been reported for a

basalt sample from the Hawaiian deep drill hole SOH-1 (Moskowitz

et al., 1998).

Iron hydroxides

Hydrous iron oxides are produced commonly by weathering of

iron-bearing rocks in ambient conditions. They are also ubiquitous in

Figure M41 Variation of IRM (0.6 T), acquired at room temperature

in a dispersed hematite powder, during a zero-field cycle to 193 K

(–80



C). (Modified from Haigh (1957), with the permission of the

publisher, Taylor & Francis Ltd., http://www.tandf.co.uk/journals.)

Figure M42 Low-temperature magnetic-phase diagram of

ilmenite-rich part of the FeTiO

-Fe

system. Down triangles

indicate the temperature where TRM disappears, up triangles the

temperature of the susceptibility peak. Closed symbols refer to

single crystals, open symbols to polycrystalline samples. Note that

x  1–y. (After Ishikawa et al. (1985), with the permission of the

publisher, the Physical Society of Japan.)

MAGNETIC PROPERTIES, LOW-TEMPERATURE 519

marine sediments (Murray, 1979). Of various iron hydroxides, the

orthorhombic form, goethite (a-FeOOH), is the most stable. Goethite

is an antiferromagnet but, like hematite, often carries a small mag-

netic moment believed to be of defect origin (Strangway et al.,

1968; Hedley, 1971). Below its Néel temperature of 120



C (in well-

crystalline samples) goethite is extremely magnetically hard, requiring

fields in excess of 10 T to approach the magnetic saturation. An

intense TRM, nearly equal to the spontaneous magnetization J

, can

be acquired by cooling from T

in a strong field. Upon warming in

a zero field it is demagnetized almost linearly with temperature until

fully vanishing somewhat below T

, but on the consequent cooling

in a zero field a considerable part of TRM is restored (Figure M43a,

Rochette and Fillion, 1989). TRM is far greater than an IRM that could

be acquired by application of the same field at a low temperature.

Therefore, FC and ZFC warming curves of goethite are very strongly

different for both bulk and nanocrystalline goethite (Figure M43b,

Guyodo et al., 2003). Both features can be used as a discriminatory

criterion for this mineral.

Other iron hydroxides often found in sediments and soils include

lepidocrocite (g-FeOOH) and ferrihydrite (5Fe

9H

O). These

minerals are paramagnetic at room temperature and do not contri-

bute to NRM; however, both can transform to ferrimagnetic phases

(maghemite) on moderate heating. Recent studies of the low-

temperature magnetic properties (Zergenyi et al., 2000; Hirt et al.,

2002) have revealed that below their ordering temperatures (100 K

for ferrihydrite and 50–60 K for lepidocrocite) these minerals can

acquire a considerable, on the order of several tenths of A m

1

remanence which decays nearly exponentially on heating in zero

field, mimicking the superparamagnetic behavior of ultrafine parti-

cles of strongly magnetic minerals, e.g., magnetite or maghemite

(Figure M44).

Iron sulfides

Of a variety of iron sulfides occurring in nature, monoclinic 4C pyr-

rhotite (Fe

) is the most extensively studied. The magnetic transition

at 30–34 K, initially discovered in early 1960s (Besnus and Meyer,

1964), has been since found in a number of natural and synthetic

pyrrhotites (Rochette, 1988; Dekkers, 1989; Dekkers et al., 1989;

Rochette et al., 1990). Phenomenologically, the behavior of the SIRM

and the coercive force on crossing the transition has much in common

with the much more extensively studied Verwey transition in magne-

tite. Part of a SIRM given at some temperature below the transition is

demagnetized at the latter. A SIRM acquired above the transition,

when cycled through the transition in zero magnetic field is also partly

demagnetized, and a rebound of magnetization at the transition is

observed during warming (Figure M45). The latter feature is grain-size

dependent, becoming progressively less pronounced in finer grains.

Pyrrhotite grains several tens of microns in size show an irreversible

loss of remanence even during a zero-field cycle to 77 K (Dekkers,

1989). The M

ratio and the coercive force increase sharply below

the transition, indicating that the low-temperature phase is in the

single-domain magnetic state. These properties can serve as discrimi-

natory criteria for monoclinic pyrrhotite in the case when heating the

sample above room temperature is impossible or undesirable.

Thiospinel of iron, mineral greigite (Fe

) once thought to be rare,

is being identified in growing number of environments, contributing to

NRM of sediments and soils (see e.g., Snowball and Torii, 1999 and

references therein). Its low-temperature magnetic properties have been

the subject of several recent studies (Roberts, 1995; Snowball and

Torii, 1999; Dekkers et al., 2000); however, no distinctive features

between 4.2 K and room temperature have been observed. Generally,

SIRM given at a low temperature decreases monotonously on warming

in zero-field albeit at different rates (Roberts, 1995), while room-

temperature SIRM hardly show any changes on cooling to 4 K

(Dekkers et al., 2000, Figure M46). The latter study also reported on

the magnetic hysteresis measurements on synthetic greigite. A peak

in M

was found at 10 K, and both the coercive force and coercivity

of remanence increased significantly below at about 30 K. Reasons

for this behavior are not clear.

Pyrite (FeS

) is often found in sediments affected by diagenesis.

Theoretical calculations show that pyrite and its polymorph marcasite

are low-spin paramagnets whose magnetic susceptibility is temperature

independent (Hobbs and Hafner, 1999). However, in practice an

increase of susceptibility is often observed in both compounds below

Figure M43 (a) Saturation magnetization J

(crosses) in 10

1

and high-field susceptibility w

(open circles) in 10

9

1

vs temperature for two goethite samples. J

and saturation

remanence J

are equal for the G70 sample. The upper curve of J

(triangles) is obtained by heating peak TRM at 4.2 K and the lower

one (dots) by cooling peak TRM from 290 K. (b) Field-cooled (FC)

and zero-field-cooled (ZFC) remanence curves for 350-nm (left

vertical axis) and 3.5-nm goethite (right vertical axis) samples.

(After Rochette and Fillion (1989) and Guyodo et al. (2003)

ã American Geophysical Union, with the permission of the

authors and the publisher.)

520 MAGNETIC PROPERTIES, LOW-TEMPERATURE

10 K (Burgardt and Seehra, 1977; Seehra and Jagadeesh, 1979). Most

likely it is due to otherwise undetectable impurities showing usual

Curie-Weiss paramagnetism.

Pyrrhotites richer in iron than Fe

are antiferromagnetic at room

temperature, but transform to a ferrimagnetically ordered phase at

200



C, and eventually to monoclinic pyrrhotite on further heating.

Other Fe-S minerals (mackinawite, smythite) have been often consid-

ered metastable in ambient conditions. These minerals received so

far a limited interest in rock magnetism, and their magnetic properties

at low temperatures are largely unknown.

Siderite an d rhodochrosite

Siderite (iron carbonate, FeCO

) and rhodochrosite (manganese carbo-

nate, MnCO

) usually precipitate jointly in anoxic sedimentary envir-

onments when iron and manganese are present in soluble form. The

two compounds may form solid solutions spanning the entire range

of intermediate compositions. Although paramagnetic at room tem-

perature, siderite has some importance in paleomagnetism, since sec-

ondary maghemite is produced by weathering siderite-bearing rocks,

and also because siderite easily converts to magnetite at heating to

400



C, thereby putting an effective limit to thermal demagnetization

(Ellwood et al., 1986, 1989). Siderite orders antiferromagnetically

below the Néel temperature of 37–40 K (natural samples of varying

purity, cf. Jacobs, 1963; Robie et al., 1984). MnCO

is a canted anti-

ferromagnet with slightly lower Néel temperature of 32 K (synthetic

crystal, Borovik-Romanov et al., 1981). Both compounds can acquire

remanence below T

, which is lost quite sharply on warming in zero

field. These minerals could be thus confused with pyrrhotite, which

shows the magnetic transition in the same temperature range. How-

ever, siderite and rhodochrosite can still be distinguished from pyrrho-

tite since both minerals show a large (over one order of magnitude

for siderite) difference between SIRMs acquired after FC and ZFC,

respectively (Housen et al., 1996; Frederichs et al., 2003), not

observed for pyrrhotite (Figure M47). It might be expected that

cycling of room-temperature SIRM in siderite- and pyrrhotite-bearing

samples will also distinguish between the two minerals.

Vivianite

Vivianite, iron hydrophosphate (Fe

(PO

)

8H

O), usually precipitates

in moderately to highly productive, iron-rich sedimentary environ-

ments. At low temperatures, vivianite shows antiferromagnetic order-

ing in each of two sublattices formed by Fe

2þ

ions occupying

nonequivalent positions (van der Lugt and Poulis, 1961). The two sub-

lattices are canted by approximately 42



(Forsyth et al., 1970). How-

ever, conflicting results have been reported on changes of the

antiferromagnetic order with temperature. Forstat et al. (1965) have

Figure M45 Examples of the 34-K transition in samples of

pyrrhotite with contrasting grain size. Crosses correspond to

cooling and circles to rewarming. (Modified from Dekkers et al.

(1989) ã American Geophysical Union, with the permission of

the author and the publisher.)

Figure M44 (a) Thermal demagnetization of IRM acquired at 5 K

in a 2.5 T field during warming to 300 K after cooling the sample

in zero-field cooling (ZFC) or in a 2.5 T field (FC) for two

lepidocrocite samples with different crystallite size. (b) Thermal

demagnetization of IRM acquired at 10 K in a 2.5 T field during

warming to 300 K for a ferrihydrite sample. (After Hirt et al.

(2002) and Zergenyi et al. (2000) ã American Geophysical

Union, with the permission of the authors and the publisher.)

MAGNETIC PROPERTIES, LOW-TEMPERATURE 521

found two sharp anomalies in the specific heat at 9.60 and 12.40 K,

respectively, which were attributed to two distinct antiferromagnetic-

paramagnetic transitions. On the other hand, in a neutron diffraction

experiment Forsyth et al. (1970) found only one antiferromagnetic-

paramagnetic transition at 8.84  0.05 K. A recent rock magnetic

study (Frederichs et al., 2003) has confirmed previous data on the

magnetic susceptibility, and found that below 10 K vivianite can

acquire only a small remanence on the order of 10

3

1

Iron silicates

Silicates are the most abundant minerals in the Earth’s crust. They

often contain iron in either ferrous (Fe

2þ

) or ferric (Fe

3þ

) form. At

low temperatures, iron-rich silicates typically order antiferromagneti-

cally, and the observed Néel temperatures range from <10 K to

about 100 K (Coey et al., 1982; Coey and Ghose, 1988). Examples

are fayalite (Fe

SiO

) and ferrosilite (FeSiO

), iron-rich end members

of olivine and pyroxene series, respectively. For the former Néel

temperature is 65 K (Santoro et al., 1966), and for the latter 43 K

(Sawaoka et al., 1968). However, iron-rich silicates are rare in nature

and in most cases iron cations are diluted with nonmagnetic cations

such as Mg

2þ

and Al

3þ

. With increasing nonmagnetic dilution, Néel

temperatures are progressively lowered and the magnetic ordering

eventually suppressed. In the olivine series (Fe

1–x

)

SiO

the anti-

ferromagnetic transition disappears for x  0.3 (Hoye and O’Reilly,

1972). Some sheet silicates, while having the antiferromagnetic ground

state, can exhibit the magnetic hysteresis and acquire remanent magne-

tization at low temperatures. An example is mineral minnesotaite, or

ferrous talc, of chemical formula Fe

(OH)

, which acquires an

IRM(2 T) of about 40 A m

1

at 4.2 K (Ballet et al., 1985). The

underlying mechanism of this behavior is that in minnesotaite the

metamagnetic transition from an antiferromagnetic to a ferromagnetic

state occurs in an anomalously low field of about 0.5 T. After once

being subjected to fields higher than the metamagnetic threshold, the

sample follow hysteresis loop never to return to the antiferromagnetic

state (Coey and Ghose, 1988). In summary, silicate minerals may have

certain potential to contribute to the low-temperature magnetic signal

of rocks, particularly to the magnetic susceptibility; whether and how

often this is really the case remains to be seen.

Andrei Kosterov

Bibliography

Amin, N., and Arajs, S., 1987. Morin temperature of annealed submi-

cronic a-Fe

particles. Physical Review B, 35: 4810 –4811.

Aragón, R., 1992. Magnetization and exchange in nonstoichiometric

magnetite. Physical Review B, 46: 5328–5333.

Aragón, R., Buttrey, D.J., Shepherd, J.P., and Honig, J.M., 1985. Influ-

ence of nonstoichiometry on the Verwey transition. Physical

Review B, 31: 430–436.

Arai, M., and Ishikawa, Y., 1985. A new oxide spin glass system of

(1–x)FeTiO

-xFe

. III. Neutron scattering studies of magnetiza-

tion processes in a cluster type spin glass of 90FeTiO

-10Fe

Journal of the Physical Society of Japan, 54: 795–802.

Arai, M., Ishikawa, Y., Saito, N., and Takei, H., 1985a. A new oxide

spin glass system of (1–x)FeTiO

-xFe

. II. Neutron scattering

Figure M47 Warming curves of IRM acquired at 2 K in a 5 T

magnetic field after cooling in zero field (open symbols) and in a

5 T field (solid symbols) for well-crystalline siderite (a) and

rhodochrosite (b). (After Frederichs et al. (2003), with the

permission of the author and the publisher, Elsevier.)

Figure M46 (a) Thermal demagnetization of IRM acquired at 5 K

in a 2.5 T field for four greigite samples. (b) Behavior of IRM

(1.2 T) acquired at room temperature for synthetic greigite

during zero-field cooling to 4 K. Sample EXP31 was stored for

22 months before the measurements, whereas sample G932

was a “fresh” sample. (After Roberts (1995) and Dekkers et al.

(2000), with the permission of the authors.)

522 MAGNETIC PROPERTIES, LOW-TEMPERATURE

studies of a cluster type spin glass of 90FeTiO

-10Fe

. Journal

of the Physical Society of Japan, 54: 781–794.

Arai, M., Ishikawa, Y., and Takei, H., 1985b. A new oxide spin glass

system of (1–x)FeTiO

-xFe

. IV. Neutron scattering studies on a

reentrant spin glass of 79FeTiO

-21Fe

single crystal. Journal of

the Physical Society of Japan, 54: 2279–2286.

Ballet, O., Coey, J.M.D., Mangin, P., and Townsend, M.G., 1985. Fer-

rous talk—a planar antiferromagnet. Solid State Communications,

55: 787–790.

Bando, Y., Kiyama, M., Yamamoto, N., Takada, T., Shinjo, T., and

Takaki, H., 1965. The magnetic properties of a -Fe

fine parti-

cles. Journal of the Physical Society of Japan, 20: 2086.

Besnus, M.J., and Meyer, A.J.P., 1964. Nouvelles données expérimen-

tals sur le magnétisme de la pyrrhotine naturelle.InProceedings

of the International Conference on Magnetism. Nottingham: Insti-

tute of Physics, pp. 507–511.

Bickford, L.R. Jr., 1950. Ferromagnetic resonance absorption in mag-

netite single crystals. Physical Review, 78: 449–457.

Borovik-Romanov, A.S., Egorov, V.M., Ikornikova, N.Y., Kreines,

N.M., Ozhogin, V.I., and Prozorova, L.A., 1981. Antiferromagnetic

carbonates of the 3d elements. Soviet Physics—Crystallography,

26: 623–630.

Brabers, V.A., Walz, M.F., and Kronmüller, H., 1998. Impurity effects

upon the Verwey transition in magnetite. Physical Review B, 58:

14163–14166.

Bucur, I., 1978. Experimental study of the origin and properties of the

defect moment in single domain haematite. Geophysical Journal of

the Royal Astronomical Society, 55: 589–604.

Burgardt, P., and Seehra, M.S., 1977. Magnetic susceptibility of iron

pyrite (FeS

) between 4.2 and 620 K. Solid State Communications,

22: 153–156.

Coey, J.M.D., and Ghose, S., 1988. Magnetic phase transitions in sili-

cate minerals. In Ghose, S., Coey, J.M.D., and Salje, E. (eds.),

Structural and Magnetic Phase Transitions in Minerals. New

York: Springer-Verlag, pp. 162–184.

Coey, J.M.D., Moukarika, A., and Ballet, O., 1982. Magnetic order in

silicate minerals. Journal of Applied Physics, 53: 8320–8325.

da Costa, G.M., Van San, E., De Grave, E., Vandenberghe, R.E.,

Barrón, V., and Datas, L., 2002. Al hematites prepared by homo-

geneous precipitation of oxinates: material characterization and

determination of the Morin transition. Physics and Chemistry of

Minerals, 29: 122–131.

de Boer, C.B., and Dekkers, M.J., 1996. Grain-size dependence of the

rock magnetic properties for a natural maghemite. Geophysical

Research Letters, 23: 2815–2818.

Dekkers, M.J., 1989. Magnetic properties of natural pyrrhotite. II.

High- and low-temperature behaviour of J

and TRM as a function

of grain size. Physics of the Earth and Planetary Interiors, 57:

266–283.

Dekkers, M.J., and Linssen, J.H., 1989. Rock magnetic properties of

fine-grained natural low-temperature haematite with reference to

remanence acquisition mechanisms in red beds. Geophysical Jour-

nal International, 99, 1–18.

Dekkers, M.J., Mattéi, J.L., Fillion, G., and Rochette, P., 1989. Grain-

size dependence of the magnetic behavior of pyrrhotite during its

low-temperature transition at 34 K. Geophysics Research Letters,

16: 855–858.

Dekkers, M.J., Passier, H.F., and Schoonen, M.A.A., 2000. Magnetic

properties of hydrothermally synthesized greigite (Fe

). II. High-

and low-temperature characteristics. Geophysical Journal Interna-

tional, 141: 809–819.

Dunlop, D.J., 2003. Stepwise and continuous low-temperature demag-

netization. Geophysical Research Letters, 30: 1582, doi: 10.1029/

2003GL017268.

Dunlop, D.J., and Özdemir, Ö., 1997. Rock Magnetism: Fundamentals

and Frontiers. Cambridge: Cambridge University Press.

Dzyaloshinsky, I., 1958. A thermodynamic theory of “weak” ferro-

magnetism of antiferromagnetics. Journal of Physics and Chemis-

try of Solids, 4: 241–255.

Ellwood, B.B., Balsam, W., Burkart, B., Long,. G.J., and Buhl, M.L.,

1986. Anomalous magnetic properties in rocks containing the

mineral siderite: paleomagnetic implications. Journal of Geophysi-

cal Research, 91: 12779–12790.

Ellwood, B.B., Burkart, B., Rajeshwar, K., Darwin, R.L., Neeley, R.A.,

McCall, A.B., Long, G.J., Buhl, M.L., and Hickcox, C.W., 1989.

Are the iron carbonate minerals, ankerite and ferroan dolomite, like

siderite, important in paleomagnetism? Journal of Geophysical

Research, 94: 7321–7331.

Forstat, H., Love, N.D., and McElearney, J., 1965. Specific heat of

(PO

)

8H

O. Physical Review, 139: A1246–1248.

Forsyth, J.B., Johnson, C.E., and Wilkinson, C., 1970. The magnetic

structure of vivianite, Fe

(PO

)

8H

O. Journal of Physics C: Solid

State Physics, 3: 1127–1139.

Frederichs, T., von Dobeneck, T., Bleil, U., and Dekkers, M.J., 2003.

Towards the identification of siderite, rhodochrosite, and vivianite

in sediments by their low-temperature magnetic properties. Physics

and Chemistry of Earth, 28: 669–679.

Gallon, T.E., 1968. The remanent magnetization of haematite single crys-

tals. Proceedings of the Royal Society of London A, 303:511–524.

Guyodo, Y., Mostrom, A., Lee Penn, R., and Banerjee, S.K., 2003.

From nanodots to nanorods: Oriented aggregation and magnetic

evolution of nanocrystalline goethite. Geophysical Research Let-

ters, 30: 1512, doi: 10.1029/2003GL017021.

Haigh, G., 1957. Observations on the magnetic transition in hematite

at 15



C. Philosophical Magazine, 2: 877–890.

Harrison, C.G.A., and Peterson, M.N.A., 1965. A magnetic mineral

from the Indian Ocean. American Mineralogist, 50: 704–712.

Hedley, I.G., 1971. The weal ferromagnetism of goethite (a-FeOOH).

Zeitschrift für Geophysik, 37: 409–420.

Heider, F., Dunlop, D.J., and Soffel, H.C., 1992. Low-temperature and

alternating field demagnetization of saturation remanence and ther-

moremanence in magnetite grains (0.037 to 5 mm). Journal of

Geophysical Research, 97: 9371–9381.

Hirt, A.M., Lanci, L., Dobson, J., Weidler, P., and Gehring, A.U.,

2002. Low-temperature magnetic properties of lepidocrocite. Jour-

nal of Geophysical Research, 107: 10.1029/2001JB000242.

Hobbs, D., and Hafner, J., 1999. Magnetism and magneto-structural

effects in transition-metal sulphides. Journal of Physics: Con-

densed Matter, 11: 8197–8222.

Hoffman, K.A., 1992. Self-reversal of thermoremanent magnetization in

the ilmenite-hematite system: Order-disorder, symmetry, and spin

alignment. Journal of Geophysical Research, 97:10883–10895.

Housen, B., Banerjee, S.K., and Moskowitz, B.M., 1996. Low-

temperature magnetic properties of siderite and magnetite in

marine sediments. Geophysical Research Letters, 23: 2843–2846.

Hoye, G.S., and O’Reilly, W., 1972. A magnetic study of the ferro-

magnesian olivines (Fe

1–x

)

SiO

,0< x < 1. Journal of Phy-

sics and Chemistry of Solids, 33: 1827–1834.

Ishikawa, Y., and Akimoto, S., 1957. Magnetic properties of the

FeTiO

-Fe

solid solution series. Journal of the Physical Society

of Japan, 12: 1083–1098.

Ishikawa, Y., Saito, N., Arai, M., Watanabe, Y., and Takei, H., 1985. A

new oxide spin glass system of (1–x)FeTiO

-xFe

. I. Magnetic

properties. Journal of the Physical Society of Japan, 54: 312–325.

Jacobs, I.S., 1963. Metamagnetism of siderite (FeCO

). Journal of

Applied Physics, 34: 1106–1107.

Kąkol, Z., and Honig, J.M., 1989. Influence of deviation from

ideal stoichiometry on the anisotropy parameters of magnetite

3(1–d)

. Physical Review B, 40: 9090–9097.

Kąkol, Z., Sabol, J., and Honig, J.M., 1991a. Cation distribution and

magnetic properties of titanomagnetites Fe

3–x

(0  x < 1).

Physical Review B, 43: 649–654.

MAGNETIC PROPERTIES, LOW-TEMPERATURE 523

Kąkol, Z., Sabol, J., and Honig, J.M., 1991b. Magnetic anisotropy of

titanomagnetites Fe

3–x

,0 x  0.55. Physical Review B, 44:

2198–2204.

Kąkol, Z., Sabol, J., Stickler, J., Kozłowski, A., and Honig, J.M.,

1994. Influence of titanium doping on the magnetocrystalline ani-

sotropy of magnetite. Physical Review B, 49: 12767–12772.

Kaye, G., 1962. The effect of titanium on the low temperature transi-

tion in natural crystals of haematite. Proceedings of the Physical

Society, 80: 238–243.

King, J.G., and Williams, W., 2000. Low-temperature magnetic prop-

erties of magnetite. Journal of Geophysical Research, 105:

16427–16436.

Kirschvink, J.L., Jones, D.S., and Mac Fadden, B.J., (eds.), 1985.

Magnetite Biomineralization and Magnetoreception in Organisms:

A New Biomagnetism. New York: Plenum Press.

Kletetschka, G., and Banerjee, S.K., 1995. Magnetic stratigraphy of

Chinese loess as a record of natural fires. Geophysical Research

Letters, 22: 1341–1343.

Kosterov, A.A., 2001. Magnetic hysteresis of pseudo-single-domain

and multidomain magnetite below the Verwey transition. Earth

and Planetary Science Letters, 186: 245–253.

Kosterov, A.A., 2002. Low-temperature magnetic hysteresis properties

of partially oxidized magnetite. Geophysical Journal International,

149: 796–804.

Kosterov, A.A., 2003. Low-temperature magnetization and AC sus-

ceptibility of magnetite: effect of thermomagnetic history. Geophy-

sical Journal International, 154:58–71.

Kozłowski, A., Kąkol, Z., Kim, D., Zaleski, R., and Honig, J.M.,

1996a. Heat capacity of Fe

3–a

(M¼Zn, Ti, 0  a  0.04).

Physical Review B, 54: 12093–12098.

Kozłowski, A., Metcalf, P., Kąkol, Z., and Honig, J.M., 1996b. Electri-

cal and magnetic properties of Fe

3–z

(z < 0.06). Physical

Review B, 53: 15113–15118.

Matzka, J., Krása, D., Kunzmann, T., Schult, A., and Petersen, N.,

2003. Magnetic state of 10–40 Ma old ocean basalts and its impli-

cations for natural remanent magnetization. Earth and Planetary

Science Letters, 206: 541–553.

Miyahara, Y., 1972. Impurity effects on the transition temperature of

magnetite. Journal of the Physical Society of Japan, 32: 629–634.

Morin, F.J., 1950. Magnetic susceptibility of aFe

and aFe

with

added titanium. Physical Review, 78: 819 –820.

Moriya, T., 1960. Anisotropic superexchange interaction and weak

ferromagnetism. Physical Review, 120:91–98.

Morrish, A.H., 1994. Canted Antiferromagnetism: Hematite. Singa-

pore; River Edge, NJ: World Scientific.

Morrish, A.H., and Watt, L.A.K., 1958. Coercive force of iron oxide

micropowders at low temperatures. Journal of Applied Physics,

29: 1029–1033.

Moskowitz, B.M., Frankel, R.B., and Bazylinski, D.A., 1993. Rock

magnetic criteria for the detection of biogenic magnetite. Earth

and Planetary Science Letters, 120: 283–300.

Moskowitz, B.M., Jackson, M., and Kissel, C., 1998. Low-temperature

magnetic behavior of titanomagnetites. Earth and Planetary

Science Letters, 157: 141–149.

Murray, J.W., 1979. Iron oxides. In Burns, R.G., (ed.), Marine

Minerals. Washington, DC: Mineralogical Society of America,

pp. 47–98.

Muxworthy, A.R., 1999. Low-temperature susceptibility and hysteresis

of magnetite. Earth and Planetary Science Letters, 169:51–58.

Muxworthy, A.R., and McClelland, E., 2000a. Review of the low-

temperature magnetic properties of magnetite from a rock magnetic

perspective. Geophysical Journal International, 140: 101–114.

Muxworthy, A.R., and McClelland, E., 2000b. The causes of low-

temperature demagnetization of remanence in multidomain

magnetite. Geophysical Journal International, 140:115–131.

Nagata, T., 1961. Rock Magnetism. Tokyo: Maruzen Co.

Nininger, R.C. Jr., and Schroeer, D., 1978. Mössbauer studies of the

Morin transition in bulk and microcrystalline a-Fe

. Journal of

Physics and Chemistry of Solids, 39: 137–144.

Ortega, B., Caballero, C., Lozano, S., Israde, I., and Vilaclara, G.,

2002. 52000 years of environmental history in Zacapu basin,

Michoacan, Mexico: the magnetic record. Earth and Planetary

Science Letters, 202: 663–675.

Özdemir, Ö., 2000. Coercive force of single crystals of magnetite

at low temperatures. Geophysical Journal International, 141:

351–356.

Özdemir, Ö., and Dunlop, D.J., 2000. Intermediate magnetite forma-

tion during dehydration of goethite. Earth and Planetary Science

Letters, 177:59–67.

Özdemir, Ö., Dunlop, D.J., and Moskowitz, B.M., 1993. The effect of

oxidation on the Verwey transition in magnetite. Geophysical

Research Letters, 20: 1671–1674.

Özdemir, Ö., Dunlop, D.J., and Moskowitz, B.M., 2002. Changes in

remanence, coercivity and domain state at low temperature in mag-

netite. Earth and Planetary Science Letters, 194: 343–358.

Ozima, M., Ozima, M., and Nagata, T., 1964. Low temperature treat-

ment as an effective means of “magnetic cleaning” of natural rema-

nent magnetization. Journal of Geomagnetism and Geoelectricity,

16:37–40.

Roberts, A.P., 1995. Magnetic properties of sedimentary greigite

(Fe

). Earth and Planetary Science Letters, 134: 227–236.

Robie, R.A., Haselton, H.T. Jr., and Hemingway, B.S., 1984. Heat

capacities and entropies of rhodochrosite (MnCO

) and siderite

(FeCO

) between 5 and 600 K. American Mineralogist, 69:

349–357.

Rochette, P.,1988. La Susceptibilité Anisotrope des Roches Faiblement

Magnétiques: Origines et Applications, Thèse d’Etat, Université de

Grenoble.

Rochette, P., and Fillion, G., 1989. Field and temperature behavior of

remanence in synthetic goethite: paleomagnetic implications. Geo-

physical Research Letters, 16: 851–854.

Rochette, P., Fillion, G., Mattéi, J.L., and Dekkers, M.J., 1990.

Magnetic transition at 30–34 Kelvin in pyrrhotite: insight into a

widespread occurrence of this mineral in rocks. Earth and Plane-

tary Science Letters, 98: 319–328.

Rozenberg, G.Kh., Hearne, G.R., Pasternak, M.P., Metcalf, P.A., and

Honig, J.M., 1996. Nature of the Verwey transition in magnetite

(Fe

) to pressures of 16 GPa. Physical Review B, 53: 6482–

6487.

Samara, G.A., 1968. Effect of pressure on the metal-nonmetal transition

and conductivity of Fe

. Physical Review Letters, 21: 795–797.

Santoro, R.P., Newham, R.E., and Nomura, S., 1966. Magnetic proper-

ties of Mn

SiO

and Fe

SiO

. Journal of Physics and Chemistry of

Solids, 27: 655–666.

Sawaoka, A., Miyahara, S., and Akimoto, S., 1968. Magnetic

properties of several metasilicates and metagermanates with

pyroxene structure. Journal of the Physical Society of Japan, 25:

1253–1257.

Schmidbauer, E., and Keller, R., 1996. Magnetic properties and

rotational hysteresis of Fe

and g-Fe

particles 250 nm in

diameter. Journal of Magnetism and Magnetic Materials, 152:

99–108.

Schmidbauer, E., and Readman, P.W., 1982. Low temperature mag-

netic properties of Ti-rich Fe-Ti spinels. Journal of Magnetism

and Magnetic Materials, 27:114–118.

Schroeer, D., and Nininger, R.C. Jr., 1967. Morin transition in a-Fe

microcrystals. Physical Review Letters, 19: 632–634.

Seehra, M.S., and Jagadeesh, M.S., 1979. Temperature-dependent

magnetic susceptibility of marcasite (FeS

). Physical Review B,

20: 3897–3899.

Senftle, F.E., Thorpe, A.N., Briggs, C., Alexander, C., Minkin, J., and

Griscom, D.L., 1975. The Néel transition and magnetic properties

524 MAGNETIC PROPERTIES, LOW-TEMPERATURE

of terrestrial, synthetic, and lunar ilmenites. Earth and Planetary

Science Letters, 26: 377–386.

Smirnov, A.V., and Tarduno, J.A., 2002. Magnetic field control of the

low-temperature magnetic properties of stoichiometric and cation-

deficient magnetite. Earth and Planetary Science Letters, 194:

359–368.

Snowball, I., and Torii, M., 1999. Incidence and significance of magnetic

iron sulphides in Quaternary sediments and soils. In Maher, B.A., and

Thompson, R. (eds.), Quaternary Climates, Environments and

Magnetism. Cambridge: Cambridge University Press, pp. 199–230.

Strangway, D.W., Honea, R.M., McMahon, B.E., and Larson, E.E.,

1968. The magnetic properties of naturally occurring goethite.

Geophysical Journal of the Royal Astronomical Society, 15:

345–359.

Syono, Y., 1965. Magnetocrystalline anisotropy and magnetostriction

of Fe

-Fe

TiO

series—with special application to rocks mag-

netism. Japanese Journal of Geophysics, 4:71–143.

Taylor, R.M., and Schwertmann, U., 1974. Maghemite in soils and its

origin. I. Properties and observations on soil maghemites. Clay

Minerals, 10: 289–298.

Thorpe, A.N., Minkin, J.A., Senftle, F.E., Alexander, C., Briggs, C.,

Evans, H.T. Jr., and Nord, G.L. Jr., 1977. Cell dimensions and

antiferromagnetism of lunar and terrestrial ilmenite single crystals.

Journal of Physics and Chemistry of Solids, 38:115–123.

Tucker, P., 1981. Low-temperature magnetic hyster esis properties of

multidomain single-crystal titanomagnetite. Earth and Planetary

Science Letters, 54: 167–172.

Uyeda, S., 1958. Thermo-remanent magnetism as a medium of palaeo-

magnetism, with special reference to reverse thermo-remanent

magnetism. Japanese Journal of Geophysics, 2:1–123.

van der Lugt, W., and Poulis, N.J., 1961. The splitting of the nuclear

magnetic resonance lines in vivianite. Physica, 27: 733–750.

Verwey, E.J.W., 1939. Electronic conduction of magnetite (Fe

) and

its transition point at low temperatures. Nature, 44: 327–328.

Verwey, E.J.W., and Haayman, P.W., 1941. Electronic conductivity

and transition point of magnetite (“Fe

”). Physica, 8: 979–987.

Walz, F., 2002. The Verwey transition—a topical review. Journal of

Physics: Condensed Matter, 14: R285–R340.

Weiss, P., and Forrer, R., 1929. La saturation absolue des ferromagné-

tiques et les lois d’approche en fonction du champ et de la tempéra-

ture. Annales de Physique,10

série, 12: 279–374.

Weiss, P., and Renger, K., 1914. Die anfängliche Permeabilität von

Eisen und Magnetit in Funktion der Temperatur und die Abhängig-

keit der Umwandlungspunkte von der Feldstärke. Archiv für

Electrotechnik, 2: 406–418.

Zergenyi, R.S., Hirt, A.M., Zimmermann, S., Dobson, J.B., and

Lowrie, W., 2000. Low-temperature magnetic behavior of ferrihy-

drite. Journal Geophysical Research, 105 : 8297–8303.

Cross-references

Iron Sulfides

Magnetic Mineralogy, Changes due to Heating

Magnetic Proxy Parameters

Rock Magnetism

Rock Magnetism, Hysteresis Measurements

MAGNETIC PROXY PARAMETERS

Introduction

As a rule, a natural system is complex and the result of the interplay of

numerous constituting parameters. To fully describe such systems, ide-

ally each (independent) parameter must be known. However, their

mere number would imply a substantial analytical operation and in

practice one should look for parameters that describe the essential parts

of the system. Because these parameters can only describe the system

in an approximate fashion they are termed “proxy parameters” or

“proxies.” There exists a myriad of proxies including the chemical

analysis of inorganic or organic compounds, the number and types

of living and fossil biota, and the physical properties of a sample, such

as grain-size distribution, porosity, magnetic properties, etc. Basically,

each physical, chemical, or biological parameter that is measurable

reasonably quickly at reasonable cost may serve as proxy to describe

a complete system. One should realize that up to over several thou-

sands of data points are rapidly gathered in a case study. Combinations

of two or more of those parameters may yield valuable proxies as well.

Another important condition for a parameter to serve as proxy is that

its meaning must be understood. This pertains to the analytical point

of view as well as to its functioning in the natural system. In the fol-

lowing, we outline the merit and potential of the so-called mineral-

magnetic proxies.

From the analytical viewpoint, the assets of mineral-magnetic or

rock-magnetic proxy parameters are manifold. They include sensitivity,

rapidity, ease of sample preparation, comparatively modestly priced

instrumentation for data acquisition, nondestructiveness, and in parti-

cular the ability to sense grain-size variation in the ultrafine grain-size

range. Magnetic properties refer to bulk sample properties in contrast

to (electron) microscopical techniques where often only a small frac-

tion of a sample can be analyzed. Therefore, mineral-magnetic para-

meters are well suited as proxy parameters. As drawbacks should be

mentioned that some of the parameters are nonunique, and that their

scales are most often utilized in a relative sense. Often ratios of several

parameters are used to reduce nonuniqueness. The meaning of a ratio

may change depending on the absolute values of numerator and deno-

minator; hence, may vary among case studies. Proper interpretation of

a mineral-magnetic data set therefore requires expert training and a

feel for potential variability in the data cloud.

Magnetism

Mineral-magnetic parameters refer to the concentration and properties

of iron oxides (and some iron sulfides), minerals that occur in trace

amounts in any rock, soil, or even organic tissue. In quantum

mechanics, electrons are assigned orbital and spin magnetic moments.

For magnetism, the spin moment is most important: each electron can

be regarded as a microscopic magnet. In most compounds electrons act

independently and consequently a very low-magnetic moment is mea-

sured in applied magnetic fields and no permanent or remanent

magnetic moment can exist. This is the case in diamagnetic and

paramagnetic substances. Diamagnetism—occurring in compounds

with no unpaired electron spins—is characterized by a small but nega-

tive magnetic moment in applied fields. It is a property of all matter

because all paired electrons yield a diamagnetic moment. Paramagnet-

ism—characteristic of compounds with unpaired electron spins—

yields a positive magnetic moment that outweighs diamagnetic

moments very quickly.

By virtue of exchange interaction, a property described also by

quantum mechanics, the outermost electrons of transition elements

may act collectively in some compounds, provided that their orbitals

sufficiently overlap. For the first row transition elements, this situation

occurs in metal oxides. In nature, iron is the most important element

and the collective electron behavior is termed ferromagnetism (sensu

lato). Several classes of ferromagnetism are distinguished: ferromag-

netism, ferrimagnetism, and antiferromagnetism (Figure M48).

Ferromagnetic and ferrimagnetic materials have very large magnetic

moments in applied fields, several orders of magnitude larger than a

typical paramagnet. Antiferromagnetic materials ideally would have

a net magnetic moment of zero but in reality usually a weak moment

persists. Collective spin behavior gives rise to the existence of perma-

nent or remanent magnetization, i.e., magnetic moments that remain

after removal of an external magnetic field. The magnetic properties

MAGNETIC PROXY PARAMETERS 525