400
11 Manganese: Predominant Role of Nodules and Crusts
study of marine manganese nodules and crusts where
the minerals making up these deposits are fine grained
and the powder diffraction patterns for many of these
minerals are similar (Post 1999). This makes the
unambiguous identification of these minerals difficult
(Manceau et al. 1992a). For this reason, other techniques have been used to study the mineralogy of
these deposits at the micron (µm) level such as electron
diffraction (Varentsov et al. 1991) and X-ray absorption
spectroscopy (XAS). XAS includes both X-ray
absorption near-edge structure (XANES) and extended
X-ray absorption fine structure (EXAFS) (Manceau et
al. 1992a,b, 2002; Drits et al. 1997; Silvester et al. 1997).
Studies using these techniques were initiated by F.V.
Chukhrov, Director of the Institute of Geology of Ore
Deposits of the Russian Academy of Sciences in
Moscow, in the 1980s (e.g. Chukhrov et al. 1989) and
achieved world recognition following joint FrancoRussian work on this topic.
Using these methods, it was established that the
10 Å phase in manganese nodules actually belongs to
one of at least seven minerals each with different structural characteristics: asbolane (a Co-Ni-bearing manganese oxide mineral), buserite-I, buserite-II, todorokite,
mixed-layered phases: asbolane-buserite-I, buserite-Ibuserite-II and buserite-I-“defective lithiophorite“
(Manceau et al. 1987, 1992b; Drits et al. 1997). In the
case of the manganese crust from Krylov Seamount
located in the eastern North Atlantic on the Cape Verde
plate, Varentsov et al. (1991a) showed that the lower
layers of the crust consisted dominantly of Fevernadite (δMnO 2 ) and goethite with admixtures of
mixed layered asbolane-buserite whereas the upper
layers consisted of Fe-vernadite and feroxyhite
(δFeOOH) with subordinate amounts of birnessite,
mixed-layered asbolane-buserite and goethite (cf.
Varentsov et al. 1991b). This sequence was thought to
reflect a significant hydrothermal input into the crust
when it first formed near the crest of the Mid-Atlantic
Ridge which was followed by a dominantly hydrogenous origin. In addition, asbolane has been shown to
occur as flakes and microflakes in manganese crusts
and nodules from the Atlantis Fracture Zone and
Krylov Seamount in the Atlantic Ocean, the Kurchatov
Fracture Zone and the Chile Plate in the Pacific Ocean
and in manganese micronodules in nannofossil ooze
from the Indian Ocean (Chukhrov et al. 1982, 1983).
An important question is the anomalous position
of Co in the geochemistry of manganese nodules.
Glasby and Thijssen (1982) have interpreted this in
terms of the crystal field characteristics of Co. On this
basis, Co
3+
(d
6
) is stable in nodules in the low spin
state with octahedral coordination and has an ionic
radius of 0.53 Å which is almost identical to that of
Fe
3+
and Mn
4+
. As such, Co may substitute in either
MnO 2 or FeOOH as the trivalent ion but not in the
interlayer spacing of 10 Å manganate as the divalent
ion as is the case for Ni
2+
, Cu
2+
, Zn
2+
and Mn
2+
. The
ability of Co to substitute in both Mn and Fe oxyhydroxides leads to its correlation with either Mn or Fe
in nodules or crusts depending on the environment of
formation (Burns and Burns 1977, 1980; Halbach et al.
1983; Giovanoli and Arrhenius 1988). Experimental
evidence has confirmed that Co
3+
is the dominant form
of Co in Mn nodules (Dillard et al. 1984; Hem et al.
1985; Manceau et al. 1997).
11.4.9 Dating
Deep-sea Mn nodules and crusts are amongst the
slowest growing minerals on Earth (Manheim
1986). They have a minimum growth rate of 0.8
mm Ma
-1
(Puteanus and Halbach 1988) which is
equivalent to the formation of about one unit cell
every year. Shallow-water concretions grow about
10
4
times faster. Several methods have been used
to date deep-sea manganese nodules and crusts
(Ku 1977; Mangini 1988).
One of the earliest methods used to date nodules
involved K-Ar dating of the volcanic nucleus of the
nodule. This method is of limited value because
submarine weathering of the core often invalidates
the results. In addition, the method makes no
allowance for any time gap between the formation of
the nucleus and subsequent Mn accretion or for
hiatuses in growth of the Mn deposit. Only a minimum
average growth rate is therefore obtained. Although
this method was used by early workers (e.g. Barnes
and Dymond 1967), it is too crude to be of much value
now.
In addition, Moore and Clague (2004) have recently
demonstrated the progressive thickening of ferromanganese crusts along the Hawaiian Ridge with
increasing distance from the Hawaiian hotspot (Loihi
Seamount). From a comparision of the maximum
thicknesses of the submarine crusts along the ridge
with radiometric ages of adjacent subaerial features,
these authors estimated the growth rates of the crusts
to be about 2.5 mm Ma
-1
. This growth rate was then
used to deduce the ages of landslide deposits and
volcanic fields occurring away from the axis of the
ridge from the estimated ages of associated crusts.
This may be considered to be an example of indirect
dating of ferromanganese crusts.
Several attempts have been made to date Mn
nodules and crusts by paleontological methods. Initial
11 Manganese: Predominant Role of Nodules and Crusts
study of marine manganese nodules and crusts where
the minerals making up these deposits are fine grained
and the powder diffraction patterns for many of these
minerals are similar (Post 1999). This makes the
unambiguous identification of these minerals difficult
(Manceau et al. 1992a). For this reason, other techniques have been used to study the mineralogy of
these deposits at the micron (µm) level such as electron
diffraction (Varentsov et al. 1991) and X-ray absorption
spectroscopy (XAS). XAS includes both X-ray
absorption near-edge structure (XANES) and extended
X-ray absorption fine structure (EXAFS) (Manceau et
al. 1992a,b, 2002; Drits et al. 1997; Silvester et al. 1997).
Studies using these techniques were initiated by F.V.
Chukhrov, Director of the Institute of Geology of Ore
Deposits of the Russian Academy of Sciences in
Moscow, in the 1980s (e.g. Chukhrov et al. 1989) and
achieved world recognition following joint FrancoRussian work on this topic.
Using these methods, it was established that the
10 Å phase in manganese nodules actually belongs to
one of at least seven minerals each with different structural characteristics: asbolane (a Co-Ni-bearing manganese oxide mineral), buserite-I, buserite-II, todorokite,
mixed-layered phases: asbolane-buserite-I, buserite-Ibuserite-II and buserite-I-“defective lithiophorite“
(Manceau et al. 1987, 1992b; Drits et al. 1997). In the
case of the manganese crust from Krylov Seamount
located in the eastern North Atlantic on the Cape Verde
plate, Varentsov et al. (1991a) showed that the lower
layers of the crust consisted dominantly of Fevernadite (δMnO 2 ) and goethite with admixtures of
mixed layered asbolane-buserite whereas the upper
layers consisted of Fe-vernadite and feroxyhite
(δFeOOH) with subordinate amounts of birnessite,
mixed-layered asbolane-buserite and goethite (cf.
Varentsov et al. 1991b). This sequence was thought to
reflect a significant hydrothermal input into the crust
when it first formed near the crest of the Mid-Atlantic
Ridge which was followed by a dominantly hydrogenous origin. In addition, asbolane has been shown to
occur as flakes and microflakes in manganese crusts
and nodules from the Atlantis Fracture Zone and
Krylov Seamount in the Atlantic Ocean, the Kurchatov
Fracture Zone and the Chile Plate in the Pacific Ocean
and in manganese micronodules in nannofossil ooze
from the Indian Ocean (Chukhrov et al. 1982, 1983).
An important question is the anomalous position
of Co in the geochemistry of manganese nodules.
Glasby and Thijssen (1982) have interpreted this in
terms of the crystal field characteristics of Co. On this
basis, Co
3+
(d
6
) is stable in nodules in the low spin
state with octahedral coordination and has an ionic
radius of 0.53 Å which is almost identical to that of
Fe
3+
and Mn
4+
. As such, Co may substitute in either
MnO 2 or FeOOH as the trivalent ion but not in the
interlayer spacing of 10 Å manganate as the divalent
ion as is the case for Ni
2+
, Cu
2+
, Zn
2+
and Mn
2+
. The
ability of Co to substitute in both Mn and Fe oxyhydroxides leads to its correlation with either Mn or Fe
in nodules or crusts depending on the environment of
formation (Burns and Burns 1977, 1980; Halbach et al.
1983; Giovanoli and Arrhenius 1988). Experimental
evidence has confirmed that Co
3+
is the dominant form
of Co in Mn nodules (Dillard et al. 1984; Hem et al.
1985; Manceau et al. 1997).
11.4.9 Dating
Deep-sea Mn nodules and crusts are amongst the
slowest growing minerals on Earth (Manheim
1986). They have a minimum growth rate of 0.8
mm Ma
-1
(Puteanus and Halbach 1988) which is
equivalent to the formation of about one unit cell
every year. Shallow-water concretions grow about
10
4
times faster. Several methods have been used
to date deep-sea manganese nodules and crusts
(Ku 1977; Mangini 1988).
One of the earliest methods used to date nodules
involved K-Ar dating of the volcanic nucleus of the
nodule. This method is of limited value because
submarine weathering of the core often invalidates
the results. In addition, the method makes no
allowance for any time gap between the formation of
the nucleus and subsequent Mn accretion or for
hiatuses in growth of the Mn deposit. Only a minimum
average growth rate is therefore obtained. Although
this method was used by early workers (e.g. Barnes
and Dymond 1967), it is too crude to be of much value
now.
In addition, Moore and Clague (2004) have recently
demonstrated the progressive thickening of ferromanganese crusts along the Hawaiian Ridge with
increasing distance from the Hawaiian hotspot (Loihi
Seamount). From a comparision of the maximum
thicknesses of the submarine crusts along the ridge
with radiometric ages of adjacent subaerial features,
these authors estimated the growth rates of the crusts
to be about 2.5 mm Ma
-1
. This growth rate was then
used to deduce the ages of landslide deposits and
volcanic fields occurring away from the axis of the
ridge from the estimated ages of associated crusts.
This may be considered to be an example of indirect
dating of ferromanganese crusts.
Several attempts have been made to date Mn
nodules and crusts by paleontological methods. Initial