See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325521361

Rare earth elements and yttrium (REY) variability with water depth in
hydrogenetic ferromanganese crusts

Article  in  Chemical Geology · June 2018

DOI: 10.1016/j.chemgeo.2018.05.045

CITATIONS READS

9 149

5 authors, including:

Keishiro Azami Naoto Hirano

The University of Tokyo Tohoku University
3 PUBLICATIONS   10 CITATIONS    107 PUBLICATIONS   1,224 CITATIONS   

SEE PROFILE SEE PROFILE

Kazutaka Yasukawa
The University of Tokyo
52 PUBLICATIONS   364 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Hawaiian mantle plume View project

All content following this page was uploaded by Keishiro Azami on 27 February 2019.

The user has requested enhancement of the downloaded file.

 Chemical Geology 493 (2018) 224–233

Contents lists available at ScienceDirect

Chemical Geology
journal homepage: www.elsevier.com/locate/chemgeo

Rare earth elements and yttrium (REY) variability with water depth in T
hydrogenetic ferromanganese crusts
K. Azamia,⁎,1, N. Hiranoa,b, S. Machidac,d, K. Yasukawad,c, Y. Katod,c,e
a
Graduate School of Science, Tohoku University, Aramaki-aoba 6-3, Aoba-ku, Sendai 980-8578, Japan
b
Center for Northeast Asian Studies, Tohoku University, Kawauchi 41, Aoba-ku, Sendai 980-8576, Japan
c
Ocean Resources Research Center for Next Generation, Chiba Institute of Technology, 2-17-1 Tsudanuma, Narashino, Chiba 275-0016, Japan
d
School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
e
Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Ntushima-cho 2-15, Yokosuka, Kanagawa 237-0061, Japan

A R T I C LE I N FO A B S T R A C T

Editor: Karen J. The origins of ferromanganese (Fe–Mn) crusts that form in seawater and elemental partitioning into these crusts have
Keywords: previously been studied using samples from water depths shallower than 4 km. Here we report geochemical analyses
Rare earth elements of Fe–Mn crusts sampled from petit-spot volcanoes within the Pacific Ocean at water depths of 5.2–6.0 km. The Fe–Mn
Ferromanganese crusts crusts have low Mn/Fe ratios (0.63–1.5), high contents of Co, Ni, and rare earth elements and Y (REY), and positive Ce
Petit-spot volcanoes anomalies, indicating a hydrogenetic origin. The Fe–Mn crusts are characterized by lower shale-normalized La/Pr
Pacific Ocean ratios (LaSN/PrSN) than typical hydrogenetic Fe–Mn crusts from the Pacific Ocean. Despite negative LaSN/PrSN and
NdSN/SmSN and positive EuSN/GdSN and DySN/HoSN correlations with water depth, the distribution coefficients of the
REY are constant. This indicates that the observed REY variability with water depth in these hydrogenetic Fe–Mn
crusts reflects that of seawater. In particular, a strong correlation between LaSN/PrSN and water depth has potential for
use as a (paleo)ocean hydro-barometer. The variety of REY compositions of Fe–Mn crusts should be interpreted on the
basis of distribution coefficients calculated for each water depth.

1. Introduction complexes, and as mono- and/or di-carbonate complexes, respectively

(Ohta and Kawabe, 2000a). REY partition coefficients between Fe–Mn
Submarine ferromanganese (Fe–Mn) crusts and nodules are con- crusts and seawater result in measured negative La, Gd, and Y
sidered a potential ocean-floor mineral resource due to their high anomalies in Fe–Mn crusts, as these elements have lower stabilities in
contents of minor metals such as Co, Ni, and rare earth elements and Y surface (hydroxide) complexes than do their neighbors in the REY series
(REY) (Hein et al., 2013). Based on their composition, they are classi- (Bau et al., 1996). All hydrogenetic Fe–Mn crusts generally show si-
fied into three genetic types: hydrogenetic, diagenetic, and hydro- milar shale-normalized REY patterns, being enriched in light to middle
thermal. However, Fe–Mn crusts around hydrothermal vents, such as REY, having flat middle to heavy REY (Hein et al., 1999; Prakash et al.,
near submarine active volcanoes, locally show compositions inter- 2012; Bau et al., 2014), and exhibiting a prominent positive Ce
mediate between those of the hydrogenetic and hydrothermal types anomaly.
(Kuhn et al., 1998). Usui and Glasby (1998) described a Fe–Mn crust Many studies have investigated the chemical compositions of Fe–Mn
from a hydrothermal vent as having a hydrothermal interior layer and a crusts and, although most of the analyzed crusts were obtained from
hydrogenetic exterior layer. water depths shallower than 4 km, a relationship between water depth
Experimental studies have shown that REY distribution in hydro- and elemental ratios has been identified. For example, total REY con-
genetic Fe–Mn crusts represents the exchange equilibrium of REY3+ centrations in Fe–Mn crusts increase with water depth (DeCarlo and
between surface complexes on Fe and Mn (oxyhydr)oxides and solution McMurtry, 1992; Nozaki et al., 2016), and those in seawater linearly
complexes in seawater, with the exception of redox-sensitive Ce (Bau, increase with water depth throughout the ocean (Zhang and Nozaki,
1999; Ohta and Kawabe, 2000a, 2000b, 2001; Bau and Koschinsky, 1996; Nozaki and Alibo, 2003a, 2003b). Although Fe–Mn crusts do not
2009; Prakash et al., 2012). Surface complexes on Fe–Mn (oxyhydr) grow on abyssal plains (5–6 km water depth) that are covered by pe-
oxides and solution complexes in seawater generally exist as hydroxide lagic sediments, they do grow on exposed basement (e.g., seamounts,

⁎
Corresponding author.
E-mail address: [email protected] (K. Azami).
1
Present address: School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan

https://doi.org/10.1016/j.chemgeo.2018.05.045
Received 21 November 2017; Received in revised form 28 May 2018; Accepted 30 May 2018
Available online 01 June 2018
0009-2541/ © 2018 Elsevier B.V. All rights reserved.
K. Azami et al. Chemical Geology 493 (2018) 224–233

Fig. 1. Location maps of the sampling sites. (A) Samples 878-3B and D8-18 collected offshore of northeast Japan. (B) Samples 1201-R02 and 1206-R07 collected
offshore of Minamitorishima Island. Petit-spot volcanoes have been previously reported in both regions (Hirano et al., 2006, 2010).

knolls, ridges, and escarpments) (Hein et al., 1999), which are generally Japan from 6 km depth, and two samples (1201-R02 and 1206-R07)
shallower in depth. Petit-spot volcanoes are a recently identified type of from offshore of Minamitorishima Island from 5.2 km depth (Fig. 2;
volcanic activity, first reported from the northwest Pacific Ocean, and Table 1) for detailed study.
are found at water depths of 5–6 km. These are monogenetic volcanoes The substrate of Fe–Mn crust samples D8-18 and 1206-R07 is highly
situated on subducting and old oceanic plates at sites of plate flexure vesicular basaltic rock, whereas that of sample 878-3B is peperite that
(Hirano et al., 2006). As such, petit-spot volcanic regions are favorable includes fragments of highly vesicular lava. Petit-spot volcanic rocks
for Fe–Mn crust growth in the deep submarine environment. show high vesicularity, despite the great water depth (Hirano et al.,
This study reports new geochemical data for Fe–Mn crusts from 2006; Okumura and Hirano, 2013). The substrate of sample 1201-R02,
petit-spot volcanoes and surrounding regions at depths of 5.2–6.0 km, collected ~20 km from a petit-spot volcano, is a pelagic sedimentary
which is much deeper than typical examples on the slopes of sea- rock. The basaltic substrates of D8-18 and 1206-R07 have palagonite
mounts. These geochemical data reveal a systematic variation in REY rims that are an alteration product of the quenched glass rind on lava
composition as compared with Fe–Mn crusts developed at shallower (Table 1). As the rate of palagonitization of basaltic glass in seawater is
depths. We describe the relationships between the elemental ratios of ~0.003 to 0.02 μm/yr (Moore et al., 1985), the eruption ages of the
the REY series and water depth, and assess the implications for REY substrates of D8-18 (3.0 mm thick palagonite rim) and 1206-R07
distribution and (paleo)oceanographic studies. (2.0 mm thick palagonite rim) are estimated to be 0.15–1.00 Ma and
0.10–0.67 Ma, respectively. These ages are considerably younger than
2. Samples the oceanic plate and are a characteristic feature of petit-spot volcanism
(Hirano et al., 2006).
The Fe–Mn crusts were sampled in the western North Pacific, off- The crust samples were each cut into three or four slabs (Table 1;
shore of northeast Japan, and in the western Pacific near Fig. 2) to examine the chemical variations from interior to exterior. The
Minamitorishima (Marcus) Island (Fig. 1) (Hirano et al., 2006; Hirano slabs were sequentially numbered from the surface to the substrate with
et al., 2010; Hirano, 2011). The samples were collected by research the exception of D8-18-a, 878-3B-a, 878-3B-b, and 1206-R07-a. For
cruises of the Japan Agency for Marine-Earth Science and Technology example, sample 1201-R02 has three layers, from surface to substrate: a
(JAMSTEC): dredge No. D08 during cruise KR04-08 by R/V Kairei dark brown layer (1201-R02-1 and 1201-R02-2), a dense black layer
(offshore northeast Japan), and dives 878, 1201, and 1206 by the (1201-R02-3), and a dark brown layer including pelagic sediments
submersible Shinkai 6500 from R/V Yokosuka (offshore Minamitor- (1201-R02-4). D8-18-a is a section that is thought to have grown on the
ishima Island). The slopes of petit-spot volcanoes offshore of northeast base of the basaltic gravel substrate, as it is coated by soft sediment
Japan are composed of hyaloclastites, pillow lavas, peperites, and (Fig. 2C). Sample 878-3B-a is a relatively dark layer found locally be-
volcanic bombs. Lava flows (including pillow tubes) and tabular bodies tween the substrate and 878-3B-2 (Fig. 2B). Sample 878-3B-b is a
of sedimentary rock were sampled offshore of Minamitorishima Island. granular surface section on the side of the substrate. Sample 1206-R07-
We selected two samples (D18-8 and 878-3B) from offshore of northeast a is a granular surface on substrate inside of a lava tube.

225
K. Azami et al. Chemical Geology 493 (2018) 224–233

Fig. 2. Photographs of the studied samples. (A) Sample D8-18-a is coated with soft sediments and is considered to have grown with being covered by seafloor
sediment. (B) Sample 878-3B is hyaloclastite from a petit-spot volcano. (C) Sample 1201-R02 was taken from tabular sedimentary rocks near pillow lavas. (D) Sample
1206-R07 was collected from a tubular lava of a petit-spot volcano.

Table 1
Information on the samples analyzed in this study.
Sample Depth (m) Latitude (N) Longitude (E) Average thickness (mm) Note

D8-18 On bottom 6 Substrate is vesicular basalt with a 3 mm palagonite rim

5911 37°29.74′ 149°44.66′
Off bottom
6003 37°30.44′ 149°44.23′
D8-18-1 2 Upper 2 mm of crust
D8-18-2 3 Lower 3 mm of crust
D8-18-a 1.3 Partly beneath the substrate and in contact with mud
878-3B 5979 37°37.602′ 149°30.498′ 10 Substrate is peperite that includes vesicular basalt
878-3B-1 3.4 Upper 3.4 mm of crust
878-3B-2 4.9 Lower 4.9 mm of crust
878-3B-a 2–4 Darker layer locally developed between the substrate and 878-3B-2
878-3B-b 3.4 Granular surface section on the side of the substrate
1201-R02 5363 23°32.454′ 154°40.142′ 33 Substrate is silty sedimentary rock
1201-R02-1 3.8 Upper 3.8 mm of crust (dark brown layer)
1201-R02-2 4.8 Next 4.8 mm of crust (dark brown layer)
1201-R02-3 15.4 Next 15.4 mm of crust (dense black layer)
1201-R02-4 12 Lower 12.0 mm of crust (dark brown layer including pelagic sediments)
1206-R07 5201 23°28.242′ 154°34.464′ 8 Substrate is vesicular and tubular basalt with a 2 mm palagonite rim
1206-R07-1 3.2 Upper 3.2 mm of crust
1206-R07-2 4.8 Lower 4.8 mm of crust
1206-R07-a 2 On inside of tubular lava

3. Methods analyzed by inductively coupled plasma–mass spectrometry (ICP–MS;

Agilent 7500c, Agilent, Santa Clara, California) following the proce-
All samples were analyzed for major elements using X-ray fluores- dures described by Kato et al. (2005, 2011). Analytical accuracy for
cence spectrometry (XRF; ZSX Primus II, Rigaku, Matsumoto, Japan). SRM JMS-2 (determined by Takaya et al., 2014) is ≤5%, and the re-
Analytical accuracy, as obtained by analysis of standard reference lative standard deviation (RSD; n = 2) is ≤3%. Both XRF and ICP–MS
material (SRM) JB-2, is ≤2%, as determined by the Geological Survey analyses were carried out at the Department of Systems Innovation, the
of Japan (GSJ). Trace elements (Co, Ni, Cu, and Zn) and REY were University of Tokyo, Japan.

226
K. Azami et al. Chemical Geology 493 (2018) 224–233

Table 2
Results of major and trace element analyses of Fe–Mn crusts.
Sample Mn Fe Mn/Fe Co Ni Cu Zn
(%) (%) (μg/g) (μg/g) (μg/g) (μg/g)

D8-18-1 15.7 22.4 0.7 829 1239 537 351

D8-18-2 15.6 23.3 0.67 996 1154 655 363
D8-18-a 24.1 14.7 1.64 943 5872 3065 691
878-3B-1 14.7 23.4 0.63 748 965 607 363
878-3B-2 17.5 24.6 0.71 945 1466 1009 400
878-3B-a 23.5 20 1.18 433 6356 3385 570
878-3B-b 20.7 21.4 0.97 2120 3314 704 388
1201R02-1 16.3 13.1 1.24 2387 1246 848 397
1201R02-2 22.9 26.3 0.87 3090 2024 1259 450
1201R02-3 24.8 24 1.04 4010 2011 1261 405
1201R02-4 16 24.5 0.65 4662 3063 1782 448
1206R07-1 17.1 22 0.78 2338 1255 843 358
1206R07-2 17.5 22.4 0.78 2513 1609 1390 410
1206R07-3 31.9 20.8 1.54 10,443 4190 652 427

Fig. 3. Mn–Fe–(Co + Ni + Cu) × 10 ternary diagram showing data of the

All samples were split and pulverized, and then dried at 105 °C for Fe–Mn crusts. Hydrothermal and hydrogenetic fields are from Bonatti et al.
~12 h prior to heating at 950 °C for 6 h for XRF analysis and loss-on- (1972).
ignition (LOI) determinations. Glass beads for XRF analyses were made
by mixing 0.4 g of sample powder and 4.0 g of lithium tetraborate flux hydrogenetic compositions (Fig. 3) (Bonatti et al., 1972). Notably,
with additional drops of 2% lithium bromide (aq) as a release agent. samples D8-18-a and 878-3B-a have lower Co and higher Ni and Cu
This mixture was heated at 1190 °C by a bead sampler (TK-4100; Tokyo contents than samples from the same region, indicating a influence of
Kagaku, Tokyo, Japan). The sample solutions for ICP–MS were made by diagenetic origin. In contrast, samples 878-3B-b and 1206-R07-a have
digesting 0.05 g of dried sample powders in 800 μL of HClO4, 4 mL of approximately double the Co contents of the other samples from each
HF, and 2 mL of HNO3 in a 12 mL Teflon PFA vial at 130 °C for ~2 h. region.
The dissolved sample was progressively evaporated at 110 °C for 12 h,
160 °C for 6 h, and 190 °C to dryness. The residue was then added to
2 mL of aqua regia and dissolved at 90 °C for 12 h. This solution was 4.2. Rare earth elements and yttrium
evaporated at 110 °C, and the residue was dissolved in 4 mL of HNO3,
1 mL of HCl, and 5 mL of ultrahigh-purity water (> 18 MΩ cm) at 90 °C. Rare earth elements and yttrium (REY) are important elements for
Finally, the solution was gravimetrically diluted to 1:4000 with ultra- understanding the origin and development of Fe–Mn deposits, parti-
high-purity water. Spectral interferences from oxides and hydroxides cularly when major and trace elements do not show clear differences
were corrected following Aries et al. (2000). XRF and ICP–MS data were amongst potential origins (i.e., hydrothermal and diagenetic) (Bau
normalized on an anhydrous basis (Tables 2 and 3). et al., 2014). The REY contents and shale-normalized REY (REYSN)
patterns of the samples are given in Table 3 and shown in Fig. 4. We
used PAAS (Post-Archean Average Australian Shale; Taylor and
4. Results McLennan, 1985) for data normalization. All samples show hydro-
genetic characteristics, generally having high overall REY contents
4.1. Major elements (excluding 878-3B-a), positive Ce anomalies, and YSN/HoSN < 1 (Bau
et al., 2014). However, the LaSN/PrSN (0.67–0.86) and YSN/HoSN ratios
The major element compositions of the Fe–Mn crusts are given in (0.53–0.60) of all the samples are lower than those of typical hydro-
Table 2. The Mn/Fe ratios of all samples are between 0.63 and 1.64. genetic Fe–Mn crusts from the Pacific Ocean (e.g., Hein et al., 2012;
Although data for most of the samples (D8-18-1, D8-18-2, 878-3B-1, Nozaki et al., 2016).
878-3B-2, 1201-R02-2, 1201-R02-4, 1206-R07-1, and 1206-R07-2) plot Samples 878-3B-b and 1206-R07-a have high Co contents and the
between the hydrogenetic and hydrothermal fields on a highest CeSN/CeSN* ratios from their respective regions, indicating very
Mn–Fe–(Co + Ni + Cu) × 10 ternary diagram, some show slow growth rates for these samples (Kuhn et al., 1998). It has been

Table 3
Rare earth element and Y data for the Fe–Mn crusts, along with shale-normalized (PAAS) anomalies. All REY data are reported in μg/g. CeSN* = 0.5LaSN + 0.5PrSN.
Sample La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu CeSN/CeSN* LaSN/PrSN YSN/HoSN

D8-18-1 129.1 1033 44.3 175.1 42.8 10.4 41 6.4 36.7 94.4 6.6 17.8 2.5 16.2 2.3 3.08 0.68 0.53
D8-18-2 154.4 973.2 51.6 205.1 49.1 12.1 47.1 7.3 41.8 113.6 7.5 20.4 2.9 18.7 2.8 2.47 0.7 0.56
D8-18-a 87.4 678.9 30.4 123 31.1 7.6 30.4 4.8 27.7 75.5 5 13.8 2 12.9 1.9 2.97 0.67 0.56
878-3B-1 150.6 916.5 51.9 205.6 49.6 11.9 47.1 7.3 41.4 110 7.3 19.6 2.7 17.5 2.6 2.33 0.68 0.56
878-3B-2 177 1022 58.2 232.8 55 13.5 52.6 8.1 46.1 121.4 8.1 21.7 3.1 19.5 2.9 2.28 0.71 0.55
878-3B-a 50.4 186.1 17.1 69.5 16.9 4.4 16.6 2.6 14.9 40.6 2.7 7.3 1.1 6.8 1 1.43 0.69 0.56
878-3B-b 158.3 1636 49.3 192.4 45.5 11 44 6.8 39 102.8 7 19 2.7 17.6 2.7 4.21 0.75 0.54
1201R02-1 164.1 1798 50.7 203.3 48.8 11.8 48.8 7.6 44.4 120.5 8.2 22.7 3.3 21.3 3.2 4.49 0.76 0.54
1201R02-2 226.9 1471 66.3 266.6 61.1 15 62.4 9.7 58 164.9 10.9 30.2 4.3 27.9 4.2 2.74 0.8 0.56
1201R02-3 223 1732 61.1 242.1 53.4 13 55.3 8.6 51.6 152.2 9.8 27.6 4 25.6 3.9 3.4 0.85 0.58
1201R02-4 190.7 1769 51.6 202.6 45.1 10.8 45.6 7.1 43 135 8.3 23.5 3.5 22.7 3.5 4.09 0.87 0.6
1206R07-1 141.9 1701 44.8 177.6 42.1 10.3 43.4 6.7 39.1 103.6 7.3 20.3 3 19.2 2.8 4.85 0.74 0.53
1206R07-2 180.4 1401 56.2 224.3 53.3 13.2 54.4 8.5 51 146 9.5 26.7 3.9 25.5 3.9 3.17 0.75 0.56
1206R07-3 183.3 2066 52 203.1 46 11.2 46.8 7.5 44.6 124.1 8.5 24.2 3.6 23 3.5 4.84 0.83 0.54

227
K. Azami et al. Chemical Geology 493 (2018) 224–233

we need to examine the possibility that the Fe–Mn crusts were con-
taminated by this volcanism. In general, Mg concentrations of Fe–Mn
crusts are low compared with petit-spot volcanic rocks (i.e., the sub-
strate of the samples). The mean Mg concentration of the Fe–Mn crusts
obtained in this study (1.53 wt%; n = 14) is similar to those of crusts on
old seamounts in the Central Pacific (1.42 wt%; n = 7; Wang et al.,
2015), Western Pacific (1.65 wt%; n = 11; Hein et al., 2012), and the
entire Northwest Pacific (1.33 wt%; n = 1478; Hein et al., 1999). The
average Mg content of petit-spot volcanic rocks is 5.75 wt% (n = 153)
(Machida et al., 2015). As such, the Mg concentrations of our samples
indicate that contamination by petit-spot volcanism is negligible.
It is necessary to discern the origins of Fe–Mn crusts because the
REY distribution depends on the origin (e.g., exchange equilibrium with
the REY host phases) (Prakash et al., 2012). Data for all samples ob-
tained in this study plot outside the hydrothermal field in Fig. 3, due to
high REY contents, positive Ce anomalies, and negative YSN/HoSN ratios
(Figs. 4 and 5). These results are consistent with a hydrogenetic origin,
with the exception of samples D8-18-a and 878-3B-a. This suggests that
hydrothermal activity on the petit-spot volcanoes was too small in scale
and/or of too short a duration to develop hydrothermal components
within the thin Fe–Mn crusts. Samples D8-18-a and 878-3B-a have both
hydrogenetic and diagenetic components, due to their high Cu and Ni
contents relative to Co, small positive Ce anomalies (878-3B-a), and low
REY contents as compared with typical hydrogenetic crusts (Figs. 4 and
5). The location of D8-18-a, which was covered by sediment and lo-
cated beneath the substrate, is consistent with it being a diagenetic
nodule rather than a hydrogenetic crust. Sample 878-3B-a could record
an unusual ocean environment comparable with that on the Shatsky
Fig. 4. (A) Shale-normalized REY spider diagram for the Fe–Mn crusts. Rise, where hydrogenetic Fe–Mn crusts are enriched in Ni and Cu (Hein
Hydrogenetic (Hein et al., 2012), diagenetic (Bau et al., 2014), and hydro- et al., 2012). Given that diagenetic Fe–Mn deposits generally scavenge
thermal (Bau et al., 2014) REY data fields are also plotted. (B) Chondrite-nor- REY from sedimentary pore water rather than from the overlying sea-
malized (McDonough and Sun, 1995) REY spider diagram for the Fe–Mn crusts. water (Bau et al., 2014), we focused on the REY compositions of the
hydrogenetic Fe–Mn crusts to investigate the REY distribution between
reported that the sides of Fe–Mn crusts commonly have more granular the Fe–Mn crusts and seawater (i.e., excluding samples D8-18-a and
878-3B-a).
surfaces and grow at lower rates than the tops of crusts (Hein et al.,
1999). Thus, the slower growth rates of these two samples are attrib-
uted to their position on the sides of the Fe–Mn crusts. The genesis of 5.2. REY distribution in hydrogenetic Fe–Mn crusts
Fe–Mn deposits is generally identified using a Ce anomaly vs. Nd dis-
crimination diagram (Fig. 5). Data for samples 878-3B-a and D8-18-a REY concentrations in hydrogenetic Fe–Mn crusts can be diluted by
plot in the region between hydrogenetic and diagenetic origins, as detrital materials that contain very low REY concentrations. Thus, a
shown by their high Ni and Cu, and relatively low Co contents. direct comparison of absolute REY concentrations, which depend on the
proportion of contaminating aluminosilicate debris, is unsuitable for
examining the REY distribution in Fe–Mn oxides. Instead, the ratios
5. Discussion between two adjacent elements in the REY series (excluding redox-
sensitive Ce) provide useful information about the REY distribution
5.1. Origin of Fe–Mn crusts without influence from debris. Fig. 6 shows the relationships between
water depth and (Z)REESN/(Z+1)REESN (Z = atomic number of REE,
Given that there are a number of young volcanoes in the study area, excluding Ce and Pm.), YSN/HoSN, and Mn/Fe ratios, as compared with

Fig. 5. Ce anomaly vs. Nd content discrimination diagram

showing the data from offshore of northeast Japan and
Minamitorishima Island from this study. Hydrogenetic data
are from DeCarlo and McMurtry (1992), Bau et al. (1996),
Baturin and Dubinchuk (2009), and Wang et al. (2015). Di-
agenetic, hydrothermal, and mixed hydrogenetic–diagenetic
data are from Bau et al. (2014).

228
K. Azami et al. Chemical Geology 493 (2018) 224–233

(caption on next page)

229
K. Azami et al. Chemical Geology 493 (2018) 224–233

Fig. 6. Water depth vs. shale-normalized REY ((Z)REESN/(Z+1)REESN; YSN/HoSN) and Mn/Fe ratios in Fe–Mn crusts. Correlation coefficients (r) were calculated from
the data from offshore of Minamitorishima Island, using only surface layer samples. LaSN/PrSN ratios in Fe–Mn crusts show a clear negative correlation with water
depth (correlation coefficient: R2 = 0.73). Data for hydrogenetic Fe–Mn crusts offshore of Minamitorishima are from Nozaki et al. (2016) and Machida et al. (2016)
(including the surface layers on the top sides of two hydrogenetic nodules). Data for hydrogenetic Fe–Mn crusts from other regions in the Pacific Ocean are from
DeCarlo and McMurtry (1992), Bau et al. (1996), Hein et al. (2012), and Wang et al. (2015).

data for hydrogenetic Fe–Mn crusts reported in previous studies depth in the Fe–Mn crusts (Fig. 6).
(DeCarlo and McMurtry, 1992; Bau et al., 1996; Hein et al., 2012; Wang If (Z)REESN/(Z+1)REESN in the Fe–Mn crusts depends on the Mn/Fe
et al., 2015; Machida et al., 2016; Nozaki et al., 2016). The data for the ratio of the crust, then the distribution coefficients with seawater must
surface layers (3–5 mm thick) of Fe–Mn crusts (Nozaki et al., 2016; this change with Mn/Fe for a given water depth. Fig. 7 shows the re-
study) and hydrogenetic Fe–Mn nodules (top side) (Machida et al., lationship between water depth and the KD (distribution coefficient) of
(Z)
2016) from offshore of Minamitorishima Island are characterized as REE/(Z+1)REE and Y/Ho (calculated using Eq. (1) below) in order to
modern Fe–Mn oxides near our study area. confirm whether variations in (Z)REESN/(Z+1)REESN are controlled by
These Fe–Mn crusts show positive correlations between water depth seawater composition. The composition of seawater has been reported
and EuSN/GdSN and DySN/HoSN ratios, and negative correlations be- at site LM9 (40°25.66′N, 144°29.74′E) in the Northwest Pacific, where
tween water depth and LaSN/PrSN, NdSN/SmSN, and Fe/Mn ratios the data were obtained at depth intervals of ~500 m (Zhang, 1995). We
(Fig. 6). Although YSN/HoSN ratios generally decrease with increasing calculated KD by using the seawater composition from the closest water
water depth, YSN/HoSN ratios show considerable scatter at depths depth at this site, as follows:
of < 3 km, which might be due to the high YSN/HoSN ratios of carbonate (Z )
REE crusts /(Z + 1)REE crusts
debris, which reflect the seawater value (Tanaka et al., 2003, 2004). KD ((Z )REE /(Z + 1)REE ) = (Z )
Although TmSN/YbSN ratios have a relatively large correlation coeffi- REE seawater /(Z + 1)REE seawater (1)
cient with water depth (r = −0.68), the variation in TmSN/YbSN is where Z is the atomic number of the REE, excluding Ce and Pm. Fig. 7
small (< 0.06). In Fig. 6, Hadal Fe–Mn crusts from the Northwest Pa- demonstrates that the KD of (Z)REE/(Z+1)REE in the surface layer of the
cific (black triangles in Fig. 6) lie along the same linear trends as the Fe–Mn crusts from offshore of Minamitorishima Island are generally
surface layers of Fe–Mn crusts from offshore Minamitorishima Island constant, despite the relatively clear negative correlation between
(black flame symbols in Fig. 6). Similar trends can be recognized in water depth and the Mn/Fe values of the crusts (Fig. 6). Moreover, a
other Pacific Fe–Mn crusts, although the data exhibit much wider leaching experiment on Fe–Mn crusts from the Indian Ocean (Prakash
scatter than in the trends defined by the Northwest Pacific samples. This et al., 2012) revealed that the proportion of La in the Mn oxide phase
scatter could be attributed to the wide geographic region of origin and (70%–75%) versus the Fe oxyhydroxide phase (20%–25%) is constant,
variable ages of the Fe–Mn crusts in the compiled data. irrespective of water depth (1.5–5.3 km), Mn/Fe ratios (0.38–5.86), and
LaSN/PrSN ratios show a strong negative correlation with water LaSN/PrSN ratios (1.19–0.77). Therefore, the co-variations of (Z)RE-
depth (correlation coefficient: r = −0.85). We note that the LaSN/PrSN ESN/(Z+1)REESN in hydrogenetic Fe–Mn crusts with water depth are not
ratios of the deepest samples offshore of northeast Japan (6 km water caused by variable Fe/Mn ratios, but simply reflect the co-variations of
depth) have the lowest values (< 56% lower than the other samples). It (Z)
REESN/(Z+1)REESN in seawater.
has been generally considered that all hydrogenetic Fe–Mn crusts show To further explore this trend, the degree of the tetrad effect on REE
similar REYSN patterns, with LaSN/PrSN > 1 regardless of their forma- distribution patterns of the Fe–Mn crust was assessed using Pr/Pr*, Tb/
tion environment (Bau et al., 1996, 2014). However, LaSN/PrSN ratios of Tb*, Dy/Dy*, Tm/Tm*, and Yb/Yb* values, as proposed by Irber (1999)
crusts from water depths > 5 km are ≪1 (Fig. 6). Given that the mean as follows:
LaSN/PrSN value of petit-spot volcanic rocks is 1.05 (n = 92; Machida
et al., 2015) and the LaSN/PrSN values of the Fe–Mn crusts are < 1, this Pr/Pr∗ = KD (Pr)/(KD (La)1/3 × KD (Nd)2/3)
cannot be due to contamination by petit-spot volcanic rock fragments.
PrSN/NdSN ratios do not show a strong correlation with water depth Tb/Tb∗ = KD (Tb)/(KD (Gd)2/3 × KD (Ho)1/3)
and, as such, the negative correlation between LaSN/PrSN ratios and
Dy/Dy∗ = KD (Dy)/(KD (Gd)1/3 × KD (Ho)2/3)
water depth must be due to a relative decrease in La in the Fe–Mn
crusts. Tm/Tm∗ = KD (Tm)/(KD (Er)2/3 × KD (Lu)1/3)
To explain the co-variations of REY ratios in Fe–Mn crusts with
water depth (Fig. 6), we need to consider factors such as complexation Yb/Yb ∗ = KD (Yb)/(KD (Er)1/3 × KD (Lu)2/3)
of REY in solutions and the chemical composition of Fe–Mn crusts.
Fig. 8 clearly shows that the degree of the tetrad effect is also
Dissolved REY in seawater exists dominantly as mono-carbonate com-
constant with water depth and, as such, the clear REE ratio variations in
plexes, REY(CO3)+, or di-carbonate complexes, REY(CO3)2− (Ohta and
the Fe–Mn crusts with water depth cannot be attributed to the tetrad
Kawabe, 2000a). The partitioning of REY into Mn oxide and Fe oxy-
effect, but is instead attributed to REE ratio variations in seawater.
hydroxide phases depends on the ratio between mono- and di-carbonate
complexes of dissolved REY in seawater (related to the “tetrad effect”),
because these have different surface charges and thus favor either the 5.3. Implications for marine geology and geochemistry
strongly negative Mn oxide or slightly positive Fe oxyhydroxide
(Langmuir, 1997). In Fe–Mn crusts, La, Gd, and Y are more easily in- REY concentrations and their ratios in seawater vary with water
corporated into the Mn oxide phase, whereas Sm, Dy, and Yb are more depth during processes by which REY are adsorbed by settling particles
compatible in the Fe oxyhydroxide phase as compared with neigh- (debris coated by organic material or oxides) in surface seawater and
boring REY (Bau and Koschinsky, 2009; Prakash et al., 2012). In par- released into the deep sea (Piepgras and Jacobsen, 1992; Alibo and
ticular, the difference in coordination numbers of the absorbed La be- Nozaki, 1999; Nozaki and Alibo, 2003a, 2003b; Haley et al., 2004;
tween Fe oxyhydroxide and Mn oxide phases results in La being Akagi, 2013). In seawater, La and Gd have high stability in the REE
preferentially incorporated into the Mn oxide phase (Ohta et al., 2009a, series, which is associated with the tetrad effect caused by the absence
2009b). In addition, the Mn/Fe ratios of Fe–Mn crusts decrease with of the 4f electron in trivalent La and the half-filled 4f electron orbital of
increasing water depth. These processes might explain the co-variation trivalent Gd (Alibo and Nozaki, 1999; Nozaki and Alibo, 2003b).
of LaSN/PrSN, NdSN/SmSN, EuSN/GdSN, and DySN/HoSN ratios with water Therefore, La and Gd are preferentially released from precipitating
particles. Given that the La content of particles (leached by acetic acid)

230
K. Azami et al. Chemical Geology 493 (2018) 224–233

Fig. 7. Water depth vs. distribution coefficients between Fe–Mn crusts and seawater for ratios of neighboring elements in the REY series. Data for hydrogenetic
Fe–Mn crusts are the same as in Fig. 6. The REYSN pattern of seawater is from Zhang (1995). Correlation coefficients (r) were only calculated from the surface layer
sample data from offshore of Minamitorishima Island.

decreases with increasing water depth (Sholkovitz et al., 1994; Lerche seawater increase at a different rate from those of Sm (Zhang et al.,
and Nozaki, 1998), the La/Pr ratio in seawater also decreases with in- 1994). Given that Nd and Sm are neither neighboring elements in the
creasing water depth globally (e.g., Piepgras and Jacobsen, 1992; Alibo REE series nor in the same tetrad section (Nd is in the La–Nd section,
and Nozaki, 1999). The variations in Eu/Gd ratios might be caused by a whereas Sm is in the Pm–Gd section), their behaviors in seawater might
similar mechanism. Despite their similar ionic radius and valency, Y differ from each other and could thus cause the negative correlation
shows a different complexation behavior from Ho (and heavy REE) in between NdSN/SmSN and seawater depth. The variations in DySN/HoSN
seawater, due to the absence of 4f electrons in Y (Byrne and Lee, 1993; with water depth cannot be explained by their electron configuration
Bau, 1996; Nozaki et al., 1997; Ohta and Kawabe, 2000b). Conse- that results in the tetrad effect. Instead, the difference in their mean
quently, the YSN/HoSN ratio of seawater decreases with water depth residence times (Dy = 740 yr and Ho = 1820 yr; Nozaki, 2001) and the
(Nozaki and Alibo, 2003b). Vertical profiles of Nd concentrations in variability in REY carriers (e.g., siliceous particles such as diatoms,

231
K. Azami et al. Chemical Geology 493 (2018) 224–233

Fig. 8. Depth vs. degree of the tetrad effect in the REE dis-
tribution patterns of Fe–Mn crusts. See the text for explana-
tions of Pr/Pr*, Tb/Tb*, Dy/Dy*, Tm/Tm*, and Yb/Yb*.
Data for the surface layers of Fe–Mn crusts from offshore of
Minamitorishima Island are from Nozaki et al. (2016) and
Machida et al. (2016). REE data for seawater are from Zhang
(1995). r is the correlation coefficient.

which are enriched in heavy REE; Akagi et al., 2011; Akagi, 2013) may distribution coefficients between Fe–Mn crusts and seawater should
cause the variations in DySN/HoSN. Our results suggest that Fe–Mn consider water depth as well as other factors. The distribution coeffi-
crusts record the signatures of these physicochemical processes con- cients of (Z)REE/(Z+1)REE in the surface layers of several Fe–Mn crusts
trolling the REY ratios of seawater. from offshore of Minamitorishima Island are generally constant, re-
In previous studies that have considered regional variations in the gardless of water depth. The degree of the tetrad effect is also constant
compositions of Fe–Mn crusts and seawater, calculations of REY dis- with depth. These observations indicate that the systematic variations
tribution coefficients have used samples collected from the same re- in REY composition with water depth recorded by the hydrogenetic
gions (e.g., Bau et al., 1996; Ohta and Kawabe, 2000b). In contrast, the Fe–Mn crusts simply reflect depth-dependent changes in seawater REY
compositions of Fe–Mn crusts and/or seawater from a wide range of composition. REE solution complexation and the coordination number
water depths have generally been averaged and used in calculations. of REE in Fe–Mn oxides may be invariable over water depths of 1–6 km.
For example, Takahashi et al. (2007) used the average composition of Further research is required to clarify if this geochemical feature is
deep seawater from depths of 1–6 km, as calculated by Piepgras and specific to the Northwest Pacific or occurs throughout the world's
Jacobsen (1992). However, our results indicate that the REY composi- oceans.
tions of hydrogenetic Fe–Mn crusts are variable and depend on water
depth. When we calculate the REY distribution coefficients of the Acknowledgments
Fe–Mn crusts from depths of 5–6 km (this study) using North Pacific
Deep Water (NPDW; 2.5 km depth) (Nozaki et al., 1999), the KD of The crew and scientific party of the R/V Kairei KR04-08 cruise, R/V
La/Pr is 20% lower than the previously published value. Therefore, Yokosuka YK05-06 and YK10-05 cruises, and the submersible SHINKAI
variations in water depth should be carefully considered when calcu- 6500 are thanked for collecting samples. We thank Dr. K. Nakamura,
lating the distribution coefficients of REY. the University of Tokyo, for helping with our analysis. This study was
The presence of the tetrad effect in the REE distribution patterns funded in part by the Toray Science and Technology Grant (11-5208),
between Fe–Mn crusts and seawater reflects the dissolved species of the Japan Society for the Promotion of Science (JP17K5715), and the
REE in seawater (e.g., Ohta and Kawabe, 2000b). The constant degree Next-generation for ocean resoures exploration of Cross-ministerial
of the tetrad effect (Fig. 8) implies that the solution complexation and Strategic Innovation Promotion Program (SIP).
coordination number of REE on the Fe–Mn oxides are unchanged over
water depths of 1 to 6 km. This is because the degree of the tetrad effect References
reflects a change in the chemical bonding conditions when dissolved
REE are absorbed by Fe–Mn crusts. Akagi, T., 2013. Rare earth element (REE)–silicic acid complexes in seawater to explain
the incorporation of REEs in opal and the “leftover” REEs in surface water: new in-
The correlation between the REY composition of seawater (and
terpretation of dissolved REE distribution profiles. Geochim. Cosmochim. Acta 113,
Fe–Mn crusts, which directly reflect the seawater REY composition) and 174–192. http://dx.doi.org/10.1016/j.gca.2013.03.014.
water depth should be similar irrespective of age, provided the REY are Akagi, T., Fu, F.F., Hongo, Y., Takahashi, K., 2011. Composition of rare earth elements in
settling particles collected in the highly productive North Pacific Ocean and Bering
carried by particles settling to the deep seafloor. This suggests that Sea: implications for siliceous-matter dissolution kinetics and formation of two REE-
LaSN/PrSN ratios of hydrogenetic Fe–Mn crusts have the potential to be enriched phases. Geochim. Cosmochim. Acta 75 (17), 4857–4876. http://dx.doi.org/
used as a (paleo)ocean hydro-barometer. 10.1016/j.gca.2011.06.001.
Alibo, D.S., Nozaki, Y., 1999. Rare earth elements in seawater: particle association, shale-
normalization, and Ce oxidation. Geochim. Cosmochim. Acta 63 (3), 363–372.
http://dx.doi.org/10.1016/S0016-7037(98)00279-8.
6. Conclusions Aries, S., Valladon, M., Polvé, M., Dupré, B., 2000. A routine method for oxide and hy-
droxide interference corrections in ICP-MS chemical analysis of environmental and
geological samples. Geostand. Newslett. 24 (1), 19–31. http://dx.doi.org/10.1111/j.
We report for the first time the REY geochemistry of ultra-deep
1751-908X.2000.tb00583.x.
(5–6 km water depth) hydrogenetic Fe–Mn crusts from petit-spot vol- Baturin, G.N., Dubinchuk, V.T., 2009. Composition of ferromanganese nodules from Riga
canoes. The low LaSN/PrSN and YSN/HoSN ratios of all the samples are a Bay (Baltic Sea). Oceanology 49 (1), 111–120. http://dx.doi.org/10.1134/
S000143700.
feature common to deep submarine Fe–Mn crusts. LaSN/PrSN ratios in Bau, M., 1996. Controls on the fractionation of isovalent trace elements in magmatic and
hydrogenetic Fe–Mn crusts show a strong negative correlation with aqueous systems: evidence from Y/Ho, Zr/Hf, and lanthanide tetrad effect. Contrib.
water depth, which could be utilized as a new (paleo)ocean hydro- Mineral. Petrol. 123 (3), 323–333. http://dx.doi.org/10.1007/s004100050317.
Bau, M., 1999. Scavenging of dissolved yttrium and rare earths by precipitating iron
barometer. Given the systematic variations of REY composition in hy- oxyhydroxide: experimental evidence for Ce oxidation, Y-Ho fractionation, and lan-
drogenetic Fe–Mn crusts with water depth, the estimation of REY thanide tetrad effect. Geochim. Cosmochim. Acta 63 (1), 67–77. http://dx.doi.org/

232
K. Azami et al. Chemical Geology 493 (2018) 224–233

10.1016/S0016-7037(99)00014-9. southwest of Australia: unique patterns compared to the South Atlantic data.
Bau, M., Koschinsky, A., 2009. Oxidative scavenging of cerium on hydrous Fe oxide: Geochem. J. 37 (1), 47–62. http://dx.doi.org/10.2343/geochemj.37.47.
evidence from the distribution of rare earth elements and yttrium between Fe oxides Nozaki, Y., Alibo, D.S., Amakawa, H., Gamo, T., Hasumoto, H., 1999. Dissolved rare earth
and Mn oxides in hydrogenetic ferromanganese crusts. Geochem. J. 43 (1), 37–47. elements and hydrography in the Sulu Sea. Geochim. Cosmochim. Acta 63 (15),
http://dx.doi.org/10.2343/geochemj.1.0005. 2171–2181. http://dx.doi.org/10.1016/S0016-7037(99)00142-8.
Bau, M., Koschinsky, A., Dulski, P., Hein, J.R., 1996. Comparison of the partitioning Nozaki, N., Tokumaru, A., Takaya, Y., Kato, Y., Suzuki, K., Urabe, T., 2016. Major and
behaviours of yttrium, rare earth elements, and titanium between hydrogenetic trace element compositions and resource potential of ferromanganese crust at Takuyo
marine ferromanganese crusts and seawater. Geochim. Cosmochim. Acta 60 (10), Daigo Seamount, northwestern Pacific Ocean. Geochem. J. 50. http://dx.doi.org/10.
1709–1725. http://dx.doi.org/10.2343/geochemj.1.0005. 2343/geochemj.2.0430.
Bau, M., Schmidt, K., Koschinsky, A., Hein, J., Kuhn, T., Usui, A., 2014. Discriminating Nozaki, Y., Zhang, J., Amakawa, H., 1997. The fractionation between Y and Ho in the
between different genetic types of marine ferro-manganese crusts and nodules based marine environment. Earth Planet. Sci. Lett. 148 (1-2), 329–340. http://dx.doi.org/
on rare earth elements and yttrium. Chem. Geol. 381, 1–9. http://dx.doi.org/10. 10.1016/S0012-821X(97)00034-4.
1016/j.chemgeo.2014.05.004. Ohta, A., Kawabe, I., 2000a. Rare earth element partitioning between Fe oxyhydroxide
Bonatti, E., Kraemer, T., Rydell, H., 1972. Classification and genesis of submarine iron- precipitates and aqueous NaCl solutions doped with NaHCO3: determinations of rare
manganese deposits. In: Ferromanganese Deposits on the Ocean Floor, pp. 149–166. earth element complexation constants with carbonate ions. Geochem. J. 34 (6),
Byrne, R.H., Lee, J.H., 1993. Comparative yttrium and rare earth element chemistries in 439–454. http://dx.doi.org/10.2343/geochemj.34.439.
seawater. Mar. Chem. 44 (2–4), 121–130. http://dx.doi.org/10.1016/0304-4203(93) Ohta, A., Kawabe, I., 2000b. Theoretical study of tetrad effects observed in REE dis-
90197-V. tribution coefficients between marine Fe-Mn deposit and deep seawater, and in REE
DeCarlo, E.H., McMurtry, G.M., 1992. Rare-earth element geochemistry of ferromanga- (III)-carbonate complexation constants. Geochem. J. 34 (6), 455–473. http://dx.doi.
nese crusts from the Hawaiian Archipelago, central Pacific. Chem. Geol. 95 (3–4), org/10.2343/geochemj.34.455.
235–250. http://dx.doi.org/10.1016/0009-2541(92)90014-V. Ohta, A., Kawabe, I., 2001. REE (III) adsorption onto Mn dioxide (δ-MnO2) and Fe oxy-
Haley, B.A., Klinkhammer, G.P., McManus, J., 2004. Rare earth elements in pore waters hydroxide: Ce (III) oxidation by δ-MnO2. Geochim. Cosmochim. Acta 65 (5),
of marine sediments. Geochim. Cosmochim. Acta 68 (6), 1265–1279. http://dx.doi. 695–703. http://dx.doi.org/10.1016/S0016-7037(00)00578-0.
org/10.1016/j.gca.2003.09.012. Ohta, A., Kagi, H., Nomura, M., Tsuno, H., Kawabe, I., 2009a. Coordination study of rare
Hein, J.R., Koschinsky, A., Bau, M., Manheim, F.T., Kang, J.K., Roberts, L., 1999. Cobalt- earth elements on Fe oxyhydroxide and Mn dioxides: part I. Influence of a multi-
rich ferromanganese crusts in the Pacific. In: Handbook of Marine Mineral Deposits, electron excitation on EXAFS analyses of La, Pr, Nd, and Sm. Am. Mineral. 94 (4),
pp. 239–281 (Chapter 9). 467–475. http://dx.doi.org/10.2138/am.2009.2986.
Hein, J.R., Conrad, T.A., Frank, M., Christl, M., Sager, W.W., 2012. Copper-nickel-rich, Ohta, A., Kagi, H., Nomura, M., Tsuno, H., Kawabe, I., 2009b. Coordination study of rare
amalgamated ferromanganese crust-nodule deposits from Shatsky Rise, NW Pacific. earth elements on Fe oxyhydroxide and Mn dioxides: part II. Correspondence of
Geochem. Geophys. Geosyst. 13 (10). http://dx.doi.org/10.1029/2012GC004286. structural change to irregular variations of partitioning coefficients and tetrad effect
Hein, J.R., Mizell, K., Koschinsky, A., Conrad, T.A., 2013. Deep-ocean mineral deposits as variations appearing in interatomic distances. Am. Mineral. 94 (4), 476–486. http://
a source of critical metals for high-and green-technology applications: comparison dx.doi.org/10.2138/am.2009.2987.
with land-based resources. Ore Geol. Rev. 51, 1–14. http://dx.doi.org/10.1016/j. Okumura, S., Hirano, N., 2013. Carbon dioxide emission to Earth’s surface by deep-sea
oregeorev.2012.12.001. volcanism. Geology 41 (11), 1167–1170. http://dx.doi.org/10.1130/G34620.1.
Hirano, N., 2011. Petit-spot volcanism: a new type of volcanic zone discovered near a Piepgras, D.J., Jacobsen, S.B., 1992. The behavior of rare earth elements in seawater:
trench. Geochem. J. 45 (2), 157–167. http://dx.doi.org/10.2343/geochemj.1.0111. precise determination of variations in the North Pacific water column. Geochim.
Hirano, N., Machida, S., Morishita, T., Ishii, T., Shimizu, K., Ijuin, Y., Machida, S., Hosoi, Cosmochim. Acta 56 (5), 1851–1862. http://dx.doi.org/10.1016/0016-7037(92)
J., Suzuki, T., Kato, J., Imada, S., Hosoya, S., 2010. YOKOSUKA Cruise Report YK10- 90315-A.
05. JAMSTEC. http://www.godac.jamstec.go.jp/darwin/cruise/yokosuka/yk10- Prakash, L.S., Ray, D., Paropkari, A.L., Mudholkar, A.V., Satyanarayanan, M., Sreenivas,
05/e. B., Chandrasekharam, D., Raju, K.A.K., Kaisary, S., Balaram, V., Gurav, T., 2012.
Hirano, N., Takahashi, E., Yamamoto, J., Abe, N., Ingle, S.P., Kaneoka, I., Hirata, T., Distribution of REEs and yttrium among major geochemical phases of marine Fe–Mn-
Kimura, J., Ishii, T., Ogawa, Y., Machida, S., Suyehiro, K., 2006. Volcanism in re- oxides: comparative study between hydrogenous and hydrothermal deposits. Chem.
sponse to plate flexure. Science 313 (5792), 1426–1428. http://dx.doi.org/10.1126/ Geol. 312, 127–137. http://dx.doi.org/10.1016/j.chemgeo.2012.03.024.
science.1128235. Sholkovitz, E.R., Landing, W.M., Lewis, B.L., 1994. Ocean particle chemistry: the frac-
Irber, W., 1999. The lanthanide tetrad effect and its correlation with K/Rb, Eu/Eu∗, Sr/ tionation of rare earth elements between suspended particles and seawater. Geochim.
Eu, Y/Ho, and Zr/Hf of evolving peraluminous granite suites. Geochim. Cosmochim. Cosmochim. Acta 58 (6), 1567–1579. http://dx.doi.org/10.1016/0016-7037(94)
Acta 63 (3–4), 489–508. http://dx.doi.org/10.1016/S0016-7037(99)00027-7. 90559-2.
Kato, Y., Fujinaga, K., Suzuki, K., 2005. Major and trace element geochemistry and Os Takahashi, Y., Manceau, A., Geoffroy, N., Marcus, M.A., Usui, A., 2007. Chemical and
isotopic composition of metalliferous umbers from the Late Cretaceous Japanese structural control of the partitioning of Co, Ce, and Pb in marine ferromanganese
accretionary complex. Geochem. Geophys. Geosyst. 6 (7). http://dx.doi.org/10. oxides. Geochim. Cosmochim. Acta 71 (4), 984–1008. http://dx.doi.org/10.1016/j.
1029/2005GC000920. gca.2006.11.016.
Kato, Y., Fujinaga, K., Suzuki, K., 2011. Marine Os isotopic fluctuations in the early Takaya, Y., Hiraide, T., Fujinaga, K., Nakamura, K., Kato, Y., 2014. A study on the re-
Eocene greenhouse interval as recorded by metalliferous umbers from a Tertiary covery method of rare-earth elements from REY-rich mud toward the development
ophiolite in Japan. Gondwana Res. 20 (2), 594–607. http://dx.doi.org/10.1016/j.gr. and the utilization of REY-rich mud. J. MMIJ 130, 104–114. (in Japanese with
2010.12.007. English Abstract). https://doi.org/10.2473/journalofmmij.130.104.
Kuhn, T., Bau, M., Blum, N., Halbach, P., 1998. Origin of negative Ce anomalies in mixed Tanaka, K., Miura, N., Asahara, Y., Kawabe, I., 2003. Rare earth element and strontium
hydrothermal–hydrogenetic Fe–Mn crusts from the Central Indian Ridge. Earth isotopic study of seamount-type limestones in Mesozoic accretionary complex of
Planet. Sci. Lett. 163 (1), 207–220. http://dx.doi.org/10.1016/S0012-821X(98) Southern Chichibu Terrane, central Japan: implication for incorporation process of
00188-5. seawater REE into limestones. Geochem. J. 37 (2), 163–180. http://dx.doi.org/10.
Langmuir, D., 1997. Aqueous Environmental Geochemistry. Prentice Hall. 2343/geochemj.37.163.
Lerche, D., Nozaki, Y., 1998. Rare earth elements of sinking particulate matter in the Tanaka, K., Ohta, A., Kawabe, I., 2004. Experimental REE partitioning between calcite
Japan Trench. Earth Planet. Sci. Lett. 159 (1–2), 71–86. http://dx.doi.org/10.1016/ and aqueous solution at 25 °C and 1 atm: constraints on the incorporation of seawater
S0012-821X(98)00066-1. REE into seamount-type limestones. Geochem. J. 38 (1), 19–32. http://dx.doi.org/
Machida, S., Hirano, N., Sumino, H., Hirata, T., Yoneda, S., Kato, Y., 2015. Petit-spot 10.2343/geochemj.38.19.
geology reveals melts in upper-most asthenosphere dragged by lithosphere. Earth Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: Its Composition and
Planet. Sci. Lett. 426, 267–279. http://dx.doi.org/10.1016/j.epsl.2015.06.018. Evolution. Blackwell.
Machida, S., Fujinaga, K., Ishii, T., Nakamura, K., Hirano, N., Kato, N., 2016. Geology and Usui, A., Glasby, G.P., 1998. Submarine hydrothermal manganese deposits in the
geochemistry of ferromanganese nodules in the Japanese Exclusive Economic Zone Izu–Bonin–Mariana Arc: an overview. Island Arc 7 (3), 422–431. http://dx.doi.org/
around Minamitorishima Island. Geochem. J. 50. http://dx.doi.org/10.2343/ 10.1111/j.1440-1738.1998.00200.x.
geochemj.2.0419. Wang, G., Jansa, L., Chu, F., Zou, C., Sun, G., 2015. Composition and origin of ferro-
McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. Chem. Geol. 120 (3–4), manganese crusts from equatorial western Pacific seamounts. J. Ocean Univ. China
223–253. http://dx.doi.org/10.1016/0009-2541(94)00140-4. 14 (2), 217–227. http://dx.doi.org/10.1007/s11802-015-2391-9.
Moore, J.G., Fornari, D.J., Clague, D.A., 1985. Basalts from the 1877 submarine eruption Zhang, J., 1995. The Geochemistry of Yttrium and Lanthanides in Seawater (Doctoral
of Mauna Loa, Hawaii; new data on the variation of palagonitization rate with dissertation, Ph.D. thesis). Univ. Tokyo.
temperature. U.S. Geol. Surv. Bull. 1663. Zhang, J., Nozaki, Y., 1996. Rare earth elements and yttrium in seawater: ICP-MS de-
Nozaki, Y., 2001. Rare earth elements and their isotopes. In: Encyclopedia of Ocean terminations in the East Caroline, Coral Sea, and South Fiji basins of the western
Sciences. 4. pp. 2354–2366. http://dx.doi.org/10.1006/rwos.2001.0284. South Pacific Ocean. Geochim. Cosmochim. Acta 60 (23), 4631–4644. http://dx.doi.
Nozaki, Y., Alibo, D.S., 2003a. Importance of vertical geochemical processes in control- org/10.1016/S0016-7037(96)00276-1.
ling the oceanic profiles of dissolved rare earth elements in the northeastern Indian Zhang, J., Amakawa, H., Nozaki, Y., 1994. The comparative behaviors of yttrium and
Ocean. Earth Planet. Sci. Lett. 205 (3), 155–172. http://dx.doi.org/10.1016/S0012- lanthanides in the seawater of the North Pacific. Geophys. Res. Lett. 21 (24),
821X(02)01027-0. 2677–2680. http://dx.doi.org/10.1029/94GL02404.
Nozaki, Y., Alibo, D.S., 2003b. Dissolved rare earth elements in the Southern Ocean,

233

View publication stats