Rare earth element geochemistry of ferromanganese deposits from the Indian Ocean

Marine Chemistry, 38 (1992) 185-208

185

Elsevier Science Publishers B.V., Amsterdam

Rare earth element geochemistry of ferromanganese deposits from the Indian Ocean B. Nagender Nath a'~, V. Balaram b, M. Sudhakarc'l and W.L. PliJgera aAbteilungfur Angewandte Lagerstiittenlehre, Aachen University of Technology, Siisterfeldstrasse 22, D-5100 Aachen, Germany bGeochemistry Division, National Geophysical Research Institute, UppalRoad, Hyderabad-500 007, India CE401, Sea-Use Programme, London School of Economics, Houghton Street, London WC2A 2AE, UK (Received 15 October 1990; revision accepted 31 January 1992)

ABSTRACT Nagender Nath, B., Balaram, V., Sudhakar, M. and Pliiger, W.L., 1992. Rare earth element geochemistry of ferromanganese deposits from the Indian Ocean. Mar. Chem., 38:185-208. Fourteen manganese nodules and three ferromanganese crusts from the Indian Ocean were analysed for major and minor elements and the 14 naturally occurring rare earth elements (REE). The REE were analysed by inductively coupled plasma-mass spectroscopy (ICP-MS). The samples were selected systematically from the Western Indian Ocean and the Central Indian Basin, to represent a seamount top, slope, an abyssal hill, siliceous sediment affected by terrigenous influx, a highly productive siliceous environment, red clay and carbonate sedimentary domains. Although REE zonation is observed in one oriented nodule, with relative enrichment in the top, evidence of top-bottom fractionation appears to have been obliterated as a result of the nodules being turned over. Correlations between Ca, P, Fe and REE in nodules suggest that the REE primarily reside in the iron oxyhydroxide and phosphatic phases. An authigenic origin is attributed to these elements. The nodules and crusts from the Western Indian Ocean and the shallower depths of the Central Indian Basin are 8-MnO2 rich, and are characterized by high concentrations of REE and higher positive cerium anomalies. These two areas are in the realm of cold Antarctic Bottom Water (AABW), which may enhance the oxidative scavenging of Ce by particles and its subsequent incorporation into manganese nodules. All the nodules and crusts show Gd-Tb anomalies. A diagenetic nodule with a palagonitic, smectite-rich nucleus exhibits an unusual heavy REE (HREE) enrichment with no significant Ce anomaly.

INTRODUCTION

Numerous studies of the chemical and mineralogical compositions of ferromanganese nodules have documented large variations both regionally and Correspondence to: B. Nagender Nath, Geological Oceanography Division, National Institute of Oceanography, Dona Paula, Goa - - 403 004, India. ~Permanent address: Geological Oceanography Division, National Institute of Oceanography, Dona Paula, Goa m 403 004, India.

0304-4203/92/$05.00 © 1992 Elsevier Science Publishers B.V. All rights reserved.

186

B. NAGENDER NATH ET AL.

locally (see reviews by Glasby, 1977; Cronan, 1980; Baturin, 1988; Halbach et al., 1988 ). The rare earth elements (REE) in manganese nodules and their associated sediments also display considerable variability on both regional and local scales (Baturin, 1988 ). The major and minor element chemistry of F e - M n deposits from the Central Indian Basin has been reported in many recent papers (e.g. Cronan and Moorby, 1981; Moorby and Cronan, 1981; Ahmed and Hussain, 1987; Mukhopadhyay and Nagendernath, 1988; Sudhakar, 1989 ). These studies have emphasized the role of hydrogenetic contributions, oxic and suboxic diagenesis, the thickness of the acoustically transparent layer (ATL), and the local topography in the geochemistry of manganese nodules. Seawater is believed to be a major contributor of REE to manganese nodules (Goldberg, 1961; Goldberg et al., 1963). In contrast to other trace elements, REE are not believed to be contributed to nodules from porewater diffusive fluxes (Elderfield et al., 1981 a; Piper, 1988 ). Recent studies (Aplin, 1984; Ingri and Ponter, 1987; Calvert et al., 1987; Glasby et al., 1987; Kunzendorf et al., 1987, 1989; Piper, 1988 ) have shown that a clear understanding of the uptake and distribution of REE in manganese nodules and associated phases may allow us to constrain models of REE cycling in the oceanic environment. Negative cerium anomalies in nodules were originally thought to be solely due to submarine hydrothermal activity (Elderfield and Greaves, 1981 ), but later reports showed the presence of such anomalies in hydrogenetic nodules also (Calvert et al., 1987; Glasby et al., 1987). Most studies on Fe-Mn deposits have been more or less confined to the Pacific Ocean. The recent review by Baturin (1988) showed the lack of data on REE in Atlantic and Indian Ocean nodules. We have found analyses of only 22 nodules from the Indian Ocean in the published literature (Pachadjanov et al., 1963; Glasby et al., 1978; Volkov, 1979; Tlig, 1982; all referred to by Baturin ( 1988); and Ben Othman et al., 1989). In this paper, we attempt to address the following points: ( 1 ) top-bottom variations of REE in oriented nodules and the role of diagenesis in the supply of REE to manganese nodules from the Indian Ocean; ( 2 ) distribution and behaviour of all 14 naturally occurring REE and their relation with the major chemical constituents in nodules; (3) variations of the REE in suites of ferromanganese nodules from different environments such as shallow basins, deeper abyssal plains, seamount top, and different sedimentary environments; and (4) the role of bottom-water redox conditions on the REE geochemistry of nodules. GENERAL FEATURES OF THE STUDY AREA

Nodules were collected from both the Central Indian Basin and the Mascarene Basin. The Central Indian Basin is bordered o n t h e west by the Cha-

REE GEOCHEMISTRYOF FERROMANGANESENODULES

187

TABLE 1 Sampling station details and description of samples Sample no.

Type

Latitude Longitude Depth (°S) (°E) (m)

GAR-EN Encrustation

12.500

76.333

649H

Encrustation

12.644

76.282

PSET

Encrustation

14.500

75.000

MAR- 1

Large nodule from shallow basin

14.988

54.983

MAR-3

Small nodule from shallow basin

16.016

55.980

MAR-22 Small nodule 16.000 from an abyssal hill of shallow basin

55.983

198-0

Nodule from flank of seamount

11.000

73.250

142-0

Nodule from top of a seamount

11.000

73.000

Description

5050-5100 Thin brownish oxide layer firmly encrusted over greyish ( < 1 cm) brown consolidated mud underlain by mottled brown hardened mud &MnO2 dominant Black very thin oxide layer ( < 1 cm). 5000 Rough (fine granular) surface texture. Loosely encrusted over brownish palagonitic material ~-MnO2 dominant with todorokite traces 4881-5173 Oxide and substratum are visually distinct, but firmly attached. Oxide black ( 1.5 cm thick). Rough (coarse granular) surface texture. Substratum - light brown hard mud/palagonite with speckles &MnO2 dominant 4300 Irregular elongated slab ( 10-11 cm). Gritty/Knobbley surface. Smooth and rough surface texture. Nucleus - - irregular, elongated, dispersed yellowish brown altered hardened mud, extending almost entire length of the nodule ~-MnO2 dominant, feldspar and quartz present 4300 Rectangular, irregular, small (3-4 cm); smooth texture (granular). Nucleus-elongate, large, dispersed, yellowish brown hardened mud. Definite control of shape by nucleus ~MnO2 with todorokite and quartz 4130 Spheroidal to cubical (smoothed edges); smooth surface texture. Nucleus - - triangular, unconsolidated material. No control of nucleus on overall shape Only ~-MnO2 5031 Discoidal, medium-sized (5.3 cm) with subcircular outline; surface very rough (coarse granular). Nucleus - clay (with quartz and illite) ~-MnO2 dominant with minor amounts of todorokite. 3580 Rhombohedral, small ( 3.8 cm) with smooth (fine granular) surface. Nucleus - - clay (feldspar, phillipsite ) £MnO2 dominant with minor amounts of todorokite

188

B. NAGENDER NATH ET AL.

TABLE 1 (Continued) Sample no.

Type

Latitude Longitude Depth (°S) (°E) (m)

Description

219-O

Base of a seamount

12.750

76.250

5350

320A

Nodule from siliceous clay/

12.248

75.501

5269

10.997

74.249

5204

Granular to coarse granular surface texture. Nucleus - - weathered rock fragment (phillipsite, feldspar and pyroxene ) Todorokite dominant, ~MnO2 present Rough textured medium-sized nodule. Nucleus - - fibrous palagonite with phillipsite crystals (?) and smectite Todorokite and &MnO2 in almost equal quantities Todorokite with minor ~-MnO2

17.998

78.000

4904

ooze

284A 75C

134/l 0 O-1T

Diagenetic nodule Hydrogenetic nodule

Nodule from shallow basin Top of an oriented nodule

O- 1B

Bottom of the above nodule

O-2T

Top of second oriented nodule from the above location

O-2B

Bottom of second oriented nodule

Somali-Seychelles Basin 13.087

75.019

5270

Triangular, irregular, small (3.2 cm) nodule. Surface texture smooth. Nucleus - - altered rock material Minor amounts of fi-MnO2. No major peaks observed Smooth textured nodule &MnO2 only Medium-sized ( 5-6 cm ) nodule with uneven top surface and flat bottom (half-hamburger shape). Serpulid growths on surface. Surface texture smooth (fine granular). Nucleus - - brownish yellow altered rock No mineralogy data Discoidal, medium-sized (4-5 cm) nodule with a hump on top. Threefourths of the nodule is stained with sediment. Serpulid growths on top. Surface texture smooth (fine granular) 6-MnO2 with minor todorokite Todorokite and ~-MnO2

Description of samples 198-0, 142-0 and 219-0 is from Mukhopadhyay ( 1988 ).

gos-Laccadive and Mid-Indian Oceanic Ridges, on the south by the Southeastern Indian Ridge, on the east by the Ninety East Ridge, and merges in the north into the Bay of Bengal (Schlich, 1982). The Mascarene Basin lies between Madagascar to the west and the Mascarene Plateau to the east. This Basin corresponds to the northwest extension of the Madagascar Basin. The two basins are separated by the Mauritius Fracture Zone. To the south, the Mascarene Basin abuts the northeastern flank of the Madagascar Ridge. To

189

REE GEOCHEMISTRYOF FERROMANGANESENODULES

50*E

70*

90 °

110"

130 °

20°N

0°

2 oOs

50 °

70 °

90 °

1100

130"

Fig. 1. Sample location map.

the north, the basin extends towards the Farquhar Group, the Amirante Trench and the Seychelles Bank (Schlich, 1982 ). The nodules were recovered from the vicinity of Tromelin Island, midway between Madagascar and Cargados Carajos. The volcanic outcrops of this island may be contributing nuclei for this suite of nodules. The depths observed in this area are above the carbonate critical depth when compared with those of most of the Central Indian Basin (Kolla and Kidd, 1982 ). A brief description of the samples is presented in Table 1 and their locations are shown in Fig. 1. ANALYTICAL METHODS

The major and minor elements were analysed at the Rheinisch-Westf~ilische Technische Hochschule (RWTH, Aachen, Germany by X-ray fluores, cence spectrometry (XRF) using glass beads which were prepared with a Specpure lithium tetraborate flux, the sample-to-flux ratio being 1 : 7. Mineralogy of the manganese nodules and crusts was determined by a Siemens Xray diffractometer using a molybdenum target. For the determination of REE, the samples including two replicates were

190

B. NAGENDER NATH ET AL.

TABLE 2 R E E d e t e r m i n a t i o n s (values in p p m ) o b t a i n e d using I C P - M S c o m p a r e d with results o b t a i n e d by ion-exchange I C P - O E S , SSMS a n d I N A A Nodule standard

La

Ce

Pr

Nd

Sm

Eu

Gd

Tb

Dy

P-I P-I P-I P-1 P-1 A-I A-1 A-I A-I

105 103 107 120 82 115 112 133 130

318 288 296 289 280 656 676 668 >300

27.5 27.0 21.7

114 135 143 113 110 94 105 85.3 94

27.2 33.8 35.3 30.4 28.0 20.4 24.7 20.9 21.0

7.44 7.75 9.04 6.57 6.80 5.81 6.10 4.48 4.80

33.8 27.0 32.5 29.4 24.0 34.3 23.6 26.5 22.0

4.53 -

25.99 4.73 2 9 . 2 0 5.13 2 7 . 0 0 4.93 2 5 . 0 0 5.10 25.80 5.09 22.40 4.70 22.00 5.30

(ICP-MS) (ICP-OES) (ICP-OES) (INAA) (SSMS) (ICP-MS) (ICP-OES) (INAA) (SSMS)

23.0

4.20 4.20 3.80

Ho

Er

Tm

13.3 1.72 15.2 13.4 13.0 15.6 2.19 14.4 15.0 -

Yb

Lu

13.26 12.30 11.50 13.80 13.00 15.40 11.80 16.30 13.50

1.75 1.89 1.66 1.85 2.21 1.92 2.16 -

I C P - M S values are f r o m this study. I C P - O E S values are f r o m Ingri a n d P o n t e r ( 1987 ), I N A A values are f r o m F l a n a g a n a n d G o t t f r i e d ( 1 9 8 0 ) a n d SSMS values are f r o m R a n k i n a n d G l a s b y ( 1979 ).

digested in a mixture of HCI, HC104 and HF in an open digestion system. The resulting dried residue was dissolved in HNO3. The REE were measured by inductively coupled plasma-mass spectrometry (ICP-MS) using a VG Plasmaquad at the National Geophysical Research Laboratory, Hyderabad. The details of the method and instrumental settings have been described in detail by Balaram and Saxena ( 1988 ). In contrast to IDMS (isotope dilution mass spectrometry), which can be used only for polyisotopic elements, and ICP, which generally requires prior chromatographic separation from the matrix, ICP-MS offers very low detection limits for all 14 REE, simple spectra and relatively fast analytical turnaround (Jarvis, 1988 ). Considering its low abundance in geological material, 115in was used as an internal standard and to monitor the signal drift during analysis; 100 ng m l - 1 of indium was added to all the samples and standards. After optimizing the ~15in signal, the system was operated covering a mass range of m / z 113-176. The analytical accuracy of the analysis was checked by preparing duplicates of US Geological Survey nodule standards A-1 and P-1. The results obtained are shown in Table 2, and are compared there with the published values. The precision was better than _+1% for all the major elements except for Mg ( _+3%), and better than _+6% for REE except Ho ( _ 8%). RESULTS

AND

DISCUSSION

The results of the major element and REE analyses are presented in Tables 3 and 4. Top-bottom fractionation or R E E zonation

Elderfield et al. (1981a) and Murphy and Dymond (1984) have found significant variation in the REE concentrations of tops and bottoms of Pacific

REE GEOCHEMISTRYOF FERROMANGANESENODULES

191

TABLE 3 Major element data (concentration in wt.%) of manganese nodules and crusts (analysis performed on samples dried at 110 ° C) Sample no.

Fe

Mn

Ti

Ca

K

P

Si

Al

Mg

GAR-EN a 649H a PSET a MAR-I MAR-3 MAR-22 198-O 142-O 219-O 320A 284A 75C 134/10 O-1 T b O-1 Bb O-2T b 0 - 2 Bb

14.53 17.17 10.14 17.25 18.23 20.32 8.48 13.82 8.99 8.11 7.51 10.94 19.04 14.72 14.51 20.19 15.98

8.70 19.39 19.01 14.85 13.48 16.80 27.99 21.47 28.46 24.58 25.24 6.47 17.09 23.27 22.45 26.97 30.76

0.61 0.70 0.55 0.83 0.95 1.08 0.36 0.47 0.36 0.33 0.34 0.61 0.70 0.47 0.46 0.83 0.95

1.34 2.00 1.49 2.28 2.08 2.17 1.92 1.88 1.85 1.45 1.59 0.65 2.08 1.83 1.69 2.03 2.09

1.74 0.63 1.21 0.93 1.09 0.50 0.66 0.92 0.69 0.81 0.85 2.74 0.64 0.57 0.59 0.36 0.32

0.19 0.22 0.19 0.25 0.28 0.30 0.12 0.20 0.14 0.09 0.09 0.07 0.25 0.21 0.18 0.23 0.18

14.43 6.60 8.57 9.84 10.00 6.17 6.58 8.21 6.36 8.95 9.00 18.84 7.79 5.85 6.49 5.06 5.15

5.12 1.37 3.58 3.11 3.38 1.98 1.78 2.18 1.96 2.99 2.73 6.32 1.75 1.79 2.13 1.42 1.80

1.21 1.30 2.01 1.40 1.28 1.19 1.66 1.54 1.57 2.03 1.98 1.65 1.19 1.53 1.57 1.51 1.80

aFerromanganese crusts. bTops and bottoms of two oriented nodules.

TABLE 4 REE concentrations (in ppm) in ferromanganese nodules and crusts Sample no. La GAR-EN 649H PSET MAR-1 MAR-3 MAR-22 198-O 142-O 219-O 320A 284A 75C 134/10 O-1 T O-I B O-2T O-2B

Ce

Pr

Nd

Sm

Eu

Gd

Tb

Dy

Ho

Er

Tm

Yb

126 546 26.70 110 24.14 6.23 31.37 4.40 22.35 4.51 11.32 1.92 11.02 243 860 49.93 201 48.82 12.49 58.24 7.33 40.53 6.90 19.76 2.74 16.76 107 872 22.57 93 20.94 5.29 32.17 3.68 21.21 3.72 10.21 1.48 10.17 160 989 34.21 137 28.64 7.52 38.58 5.20 28.57 5.44 15.30 1.91 13.34 197 1384 41.22 165 35.78 8.95 49.97 6.18 34.36 6.13 17.43 2.40 15.91 228 1517 44.75 176 37.64 9.45 55.54 6.53 36.37 6.74 18.16 2.57 17.17 125 347 31.38 128 30.05 8.02 36.04 4.87 26.42 4.51 12.26 1.72 11.90 180 616 41.65 164 36.50 9.64 44.72 6.35 32.90 5.48 15.59 2.06 13.91 121 446 32.38 133 33.88 8.33 36.56 5.25 27.35 4.63 13.30 1.87 11.76 90 241 27.39 130 55.53 17.02 63.15 8.13 41.35 9.84 30.58 6.57 41.33 87 245 21.54 88 21.24 5.50 25.30 3.t7 17.38 3.21 8.43 1.12 7.68 57 341 12.24 49 10.51 3.19 16.79 2.09 10.43 2.01 5.52 0.68 4.73 168 910 36.75 148 32.25 8.13 44.98 6.00 32.01 5.85 15.95 2.16 14.86 181 715 43.74 177 41.22 10.96 53.35 7.21 37.84 6.50 17.17 2.49 16.21 175 726 41.35 170 38.64 10.49 52.40 6.81 35.90 6.26 17.55 2.39 16.03 216 835 49.61 206 48.43 12.31 63.56 8.10 44.75 7.96 20.53 2.90 19.23 147 549 34.27 139 30.81 8.33 40.29 5.21 29.71 5.27 13.96 1.65 13.45

Sample details are as described in Table 3.

Lu 1.72 2.38 1.41 1.93 2.25 2.32 1.64 1.97 1.73 8.52 0.92 0.62 2.05 2.07 2.24 2.69 1.71

192

B. NAGENDER NATH ET AL 20-

•- - 0-I TOP o o o_I BOTTOM 6 6 0-2 TOp

A

10-

I

L

i

20

"--°

B

219-0

o

198-0

o

L~ \ ~Z

•

75C

6

20

320A 28~A

.~

c 10

1

20-

D

• PSET O GAR-EN ~ 649H

O

10

1

20-

102

i

i

i

~

i

h

~

E A ~

i

i

i

i

~

t

I

•

" HAR I

A

• HAR 22

o--o MAR /

. . . . . . . Lg Ee Pr Nd Sm

2

Eu G

'dtO'y' Tb Ho

' I Er Tm

~

' b lu

REEGEOCHEMISTRYOF FERROMANGANESENODULES

193

manganese nodules. To confirm the prevalence of such a variation in the nodules from the Indian Ocean, tops and bottoms of two oriented nodules were analysed for REE in the present study. The analytical results and the shale-normalized REE patterns of these samples are shown in Tables 3 and 4 and Fig. 2, respectively. The shale values of REE used to normalize are from Piper (1974b) and Sholkovitz ( 1988 ), and represent a mean value of North American, European and Russian shales. The Mn/Fe ratios in these samples vary from 1.3 to 1.9. Of the two nodules studied for top-bottom fractionation, one nodule (O-2) has a compositionally distinct top and bottom zonation similar to that reported initially by Raab (1972) and later by other workers (Calvert et al., 1978; Moore, 1984; Dymond et al., 1984; Friedrich et al., 1988). In contrast, nodule O-1 shows almost equal Mn/Fe ratios (Table 5 ). TABLE5 Important elemental ratios used for genetic interpretations Sample no.

Mn/Fe

Ce/La

Ce/Nd

La/Sm

Yb/Sm

GAR-EN 649H PSET MAR-1 MAR-3 MAR-22 198-O 142-0 219-0 320A 284A 75C 134/10 O-1 T O-1 B 0-2 T 0-2 B

0.6 I. 1 1.8 0.9 0.7 0.8 3.3 1.6 3.2 3.0 3.4 0.6 0.9 1.6 1.5 1.3 1.9

4.32 3.54 8.17 6.17 7.01 6.67 2.77 3.43 3.70 2.69 2.81 6.01 5.41 3.95 4.14 3.87 3.75

4.98 4.28 9.41 7.20 8.41 8.63 2.72 3.76 3.35 !.85 2.76 6.95 6.14 4.05 4.27 4.06 3.96

5.24 4.98 5.10 5.59 5.51 6.05 4.16 4.91 3.56 1.62 4. l0 5.40 5.21 4.39 4.53 4.76 4.42

0.45 0.34 0.49 0.47 0.45 0.46 0.40 0.38 0.35 0.74 0.36 0.45 0.46 0.39 0.42 0.40 0.44

Sample details are as given in Table 3.

Fig. 2. Shale-normalized patterns of the samples studied. (A) Tops and bottoms of the oriented nodules O- l and 0-2. (B) Nodules from the seamount environment. The sample from the seamount top (142-O) shows the maximum REE concentration. (C) Patterns for three varieties of nodules. Sample 75C is a typical hydrogenetic nodule from a red clay region and shows the biggest possible cerium anomaly. Sample 284A is representative of a diagenetic nodule with a low positive Ce anomaly. Smectite-containingnodule 320A is enriched in HREE. (D) Patterns of ferromanganese encrnstations. (Note the strong positive cerium anomalies.) (E) Nodules recovered from the Western Indian Ocean. All of them have identical patterns and higher REE concentrations than the nodules from the Central Indian Basin.

194 1OOO-

B. NAGENDER NATH ET AL. 20~

Lo 750-

r:,8-1

o-1

v~

Ce~

O-2T

r

~oo-

tw W O 250

55-

Sme • B O.

T Eu ~

.........

12

I .........

14

O-2T Eu

• t .........

16

I .........

18

e i .........

20

22

IRON Fig. 3. Diagram showing the relationship between Fe (%), Ce and other REE (ppm) in the oriented nodules. B and T represent bottoms and tops, respectively.

The oxide layers from the top and bottom of nodule 0-2 have differing mineralogy. Whereas there is more 6-MnO2 than todorokite in the top part of the nodule, no significant differences are found between 6-MnO2 and todorokite in the bottom of the nodule. A strong quartz peak and higher concentrations of AI and Si (Table 3 ) indicate the presence of larger amounts of aluminosilicate debris in the bottom of the nodules. Similar to the results found by Elderfield et al. ( 1981 a), a positive co-variation between REE and Fe (Fig. 3 ) is observed. Fe enrichment in the nodule tops and Mn phase enrichment in nodule bottoms (as found in this study, and by Banakar (1990) and Pattan and Mudholkar (1990) for the same area) indicate the role of diagenesis. Apart from Si and A1 (aluminosilicate phase ), and Mn, P is the only major element which shows a consistent zonation. The zonation of P and Fe are similar. The association of REE with Fe and P could be due to two competing phases - - Fe oxyhydroxide and P phase from phosphatic fish debris (Elderfield et al., 1981b; Calvert and Piper, 1984) - - or could be caused by incorporation in a single iron (phosphate) oxyhydroxide phase alone (Glasby and Thijssen, 1982; Glasby et al., 1987 ). The depletion of the REE in the bottoms of nodules may be explained as follows. Enhanced concentrations of REE in porewater compared with seawater and upward diffusion into overlying sedimentary layers has been observed in anoxic, reducing nearshore sediments (Elderfield and Sholkovitz, 1987; German and Elderfield, 1989; Sholkovitz et al., 1989). There is a lack of porewater REE data in the deep-sea pelagic areas to suggest a contribution of REE through early diagenetic processes, but the similarity in vertical and

REE GEOCHEMISTRY OF FERROMANGANESE NODULES

195

interoceanic REE distributions to nutrients in seawater (Elderfield and Greaves, 1982; De Baar et al., 1983, 1985a,b; Klinkhammer et al., 1983 ) may be suggestive of a significant upward diagenetic flux of REE from the sediments (De Baar et al., 1988 ). In spite of these findings, most of the nodule bottoms studied so far (in this study, and by Elderfield et al. (1981a) and M u s h y and Dymond (1984) ) show REE depletion. Similarly, many of the nodule bottoms studied so far (including those studied in this work) have lower iron contents (see, e.g. Banakar (1990) and Pattan and Mudholkar ( 1990 ), who considered the same study area); by implication, fewer iron oxyhydroxide surfaces may be available for adsorptive scavenging, which could lead to a lesser enrichment of REE in bottoms. No REE zonation is observed in the O-1 nodule of this study (Table 4 ) and Nodule 2 of Elderfield et al. (198 l a). The similarities in metal concentrations in tops and bottoms and the mineralogy (~MnO2 and todorokite of roughly equal intensities) seems to be due to turning over of the nodule. Based on studies of 226Ra/23°Th and 23°Th/232Th activity ratios, differences in 23°Th and 231pa, and inconsistent transition metal top/bottom variations, Krishnaswamy and Cochran ( 1978 ) and Moore ( 1984 ) have concluded that nodules turn over on a 1000-10 000 year time-scale in Pacific. Evidence for the turnover of nodules from the Central Indian Basin has also been found by Banakar (1990). At the same time, we do not visualize the possibility of nodule 0-2 turning over, as three-fourths of this nodule was stuck in the sediment (Table 1 ).

Interelement relations and their implications Correlation matrices have been prepared for three data subsets and are presented in Tables 6 and 7. Table 6 shows the correlation between bulk chemical data and the major elements and elemental ratios of nodules and encrustations. Such elemental ratios were previously used for genetic interpretations (e.g. Addy, 1979; Elderfield et al., 1981a,b; Aplin, 1984; Calvert et al., 1987; Glasby et al., 1987; Kunzendorf et al., 1987 ). The relations among REE are shown in Table 7. It is known that there is an indeterminate tendency for negative correlations between variables of similar compositions (Skala, 1979; Glasby et al., 1985 ). However, our data do show non-trivial positive correlations between Fe and Ti, Fe and P, Ca and P, Si and K, A1 and K, etc. Although the postulated relationships as shown below may exist, they are likely to be weaker than the magnitude of correlations indicated. Although Mn is not related to any other element, other major associations noticed are as follows: a group comprising Fe, Ti, P and the light REE up to Nd (La, Ce, Pr and Nd ), and a group consisting of Si, AI and K. We designate the above associations as Mn-oxide phase, Fe-oxide/oxyhydroxide phase and aluminosilicate phase respectively. Mn has no strong association with REE

Mn

-0.32 0.41 -0.77 -0.17 -0.80 -0.77 0.44 0.76 -0.61 -0.64 -0.59 -0.14 0.27 0.07 ±0.33 0.28 0.33 0.44 0.43 0.32 0.36 0.38 0.31 0.29 0.22 0.27 0.21

-0.26 0.88 0.59 -0.29 0.85 -0.25 -0.26 -0.70 -0.78 0.37 0.45 0.63 -0.16 0.68 0.82 0.81 0.71 0.67 0.30 0.18 0.50 0.46 0.52 0.37 0.27 - 0.00 0.03 -0.12

0.46 -0.12 0.75 -0.09 -0.05 -0.55 -0.72 0.56 0.63 0.67 -0.05 0.44 0.62 0.81 0.45 0.41 0.09 -0.01 0.30 0.20 0.29 0.21 0.12 -0.09 -0.05 -0.16

Ti

-0.83 0.73 -0.79 -0.80 -0.40 -0.03 -0.01 0.04 0.19 -0.24 0.72 0.75 0.58 0.77 0.75 0.49 0.37 0.56 0.57 0.63 0.47 0.40 0.12 0.17 0.02

Ca

-0.41 0.97 0.96 0.03 -0.34 0.35 0.33 0.24 0.14 -0.72 -0.59 -0.24 -0.72 -0.74 -0.68 -0.60 -0.68 -0.69 -0.73 -0.61 -0.55 -0.34 -0.39 -0.25

K

n = 17; values above 0.61 are significant at 99% level.

La Ce Pr Nd Sm ,Eu Gd Tb Dy Ho Er Tm Yb Lia

REE 3+

AI Mg Mn/Fe Ce/La Ce/Nd La/Sm Yb/Sm

Si

Ti Ca K P

Mn

Fe

-0.40 -0.37 -0.72 -0.61 0.49 0.55 0.63 -0.13 0.62 0.78 0.89 0.67 0.61 0.24 0.11 0.43 0.39 0.46 0.30 0.23 -0.03 0.00 -0.14

P

0.96 -0.02 -0.35 0.27 0.27 0.21 0.22 -0.69 -0.59 -0.26 -0.71 -0.73 -0.63 -0.54 -0.64 -0.65 -0.69 -0.53 -0.47 -0.26 -0.31 -0.16

Si

0.09 -0.34 0.39 0.37 0.19 0.29 -0.69 -0.60 -0.19 -0.73 -0.74 -0.63 -0.54 -0.61 -0.64 -0.68 -0.50 -0.45 -0.22 -0.27 -0.13

Al

0.66 -0.25 -0.34 -0.65 0.29 -0.37 -0.58 -0.59 -0.45 -0.38 -0.02 0.09 -0.13 -0.15 -0.17 -0.04 0.02 0.20 0.21 0.29

Mg

Correlation coefficient matrix for complete data vs. major elements and important elemental ratios

TABLE6

-0.61 -0.67 -0.76 0.04 -0.24 -0.45 -0.65 -0.26 -0.21 0.10 0.16 -0.10 -0.06 -0.08 -0.02 0.05 0.18 0.19 0.24

Mn/Fe

0.99 0.67 0.12 -0.13 0.08 0.69 -0.12 -0.18 -0.40 -0.44 -0.19 -0.29 -0.24 -0.26 -0.26 -0.30 -0.29 -0.29

Ce/La

0.76 0.05 -0.08 0.17 0.75 -0.05 -0.12 -0.40 -0.45 -0.17 -0.28 -0.22 -0.26 -0.28 -0.34 -0.33 -0.35

Ce/Nd

-0.45 0.02 0.40 0.66 0.16 0.05 -0.47 -0.58 -0.26 -0.31 -0.24 -0.43 -0.50 -0.67 -0.66 -0.72

La/Sm

-0.08 -0.34 -0.08 -0.31 -0.22 0.24 0.39 0.23 0.19 0.14 0.45 0.53 0.73 0.73 0.81

Yb/Sm

>

m

=:

Z

Z

REE GEOCHEMISTRYOF FERROMANGANESENODULES

197

TABLE 7 Correlation coefficient matrix for REE

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb

La

Ce

Pr

Nd

Sm

Eu

Gd

Tb

Dy

Ho

Er

Tm

Yb

Lu

1

0.74 1

0.96 0.58 1

0.93 0.53 0.99 1

0.60 0.19 0.74 0.82 1

0.45 0.06 0.61 0.70 0.98 1

0.72 0.41 0.82 0.88 0.96 0.92 1

0.70 0.32 0.83 0.89 0.97 0.93 0.98 1

0.76 0.40 0.87 0.92 0.96 0.90 0.99 0.99 1

0.54 0.27 0.66 0.74 0.95 0.96 0.96 0.95 0.94 1

0.44 0.20 0.57 0.65 0.94 0.97 0.92 0.91 0.89 0.99 1

0.14 -0.03 0.27 0.38 0.82 0.90 0.75 0.74 0.70 0.89 0.94 1

0.15 -0.01 0.29 0.39 0.82 0.90 0.76 0.76 0.71 0.90 0.95 0.99 1

-0.02 -0.13 0.11 0.22 0.72 0.83 0.63 0.63 0.57 0.81 0.88 0.98 0.98

n = 17; values above 0.61 are significant at 99% level.

(Table 6 ), except for a statistically insignificant relation with middle REE (Sm and Eu ). Strong negative correlations are displayed between Si, A1 and K and REE (mostly significant at 99% confidence level). The three major groups observed here are in accordance with the Pacific nodules (Calvert and Price, 1977). The first, second and third elemental groups may represent the diagenetic, hydrogenetic and aluminosilicate phase, respectively. Negative correlations between REE and the aluminosilicate phase indicate that the surface transfer of continental detritus or the occlusion of detrital material (Ehrlich, 1968 ) is an unlikely mechanism for the incorporation of REE into manganese nodules in the present study. This gains support from Sr and Nd isotope studies reported initially by Goldstein and O'Nions ( 1981 ) and recently by Ben Othman et al. (1989). The 87Sr/S6Sr ratios for manganese nodules from various oceans studied (Ben Othman et al., 1989) range between 0.70895 and 0.70918 and cluster close to, with a slight deviation from, the seawater value (0.70919). Also, the Crqdvalue is reported to be - 8.9 for Indian Ocean nodules and therefore similar to the seawater value ( - 8 ) for this region (Ben Othman et al., 1989). In addition, higher aluminosilicate contents seem to dilute the REE concentrations, as is evidenced by nodule 75C (Tables 3 and 4). The lack of correlation between the REE and the Mn phase, and their inverse relationship with Mn/Fe ratios (T~tble 6), supports the idea that the nodules with marked diagenetic signature (higher Mn/Fe ratios) show a depletion of REE (Calvert et al., 1987 ), or may be due to the preference of REE for the Fe phase. Sample 284A in this suite of samples is taken as a represent-

198

B. NAGENDER NATH ET AL.

ative of diagenetic nodules (with todorokite dominant). This sample shows the lowest concentrations of REE (except for sample 75C). As designated above, the Fe-bearing phase may represent a hydrogenetic component. There are two schools of thought regarding the correlation between Fe, P and REE in the manganese nodules. Based on elemental correlations and leaching experiments, Elderfield et al. ( 198 lb) suggested that two phases govern the REE distribution in the manganese nodules: (a) a phosphatic phase from fish debris and (b) an Fe-oxyhydroxide phase. In contrast, Calvert and Price ( 1970, 1977 ) suggested that Fe and P are present as a ferriphosphate phase. Later Glasby and Thijssen ( 1982 ) and Glasby et al. ( 1987 ) similarly attributed this association to a single mineral phase - - iron phosphate. However, none of the above works considered the calcium data of nodules in their REE interpretations. As mentioned above, the REE in this study are associated with Fe and P, either as an iron phosphate phase or as two separate phases. However, it may be noted that the trivalent REE co-vary well with Ca (Table 6). In ferromanganese nodules, the REE are bound initially by two phases: a phosphatic phase and a surface layer of phosphate which had previously been adsorbed onto hydrous iron oxides. Upon diagenesis, the REE may be incorporated into recrystallized biogenic apatite (Jonasson et al., 1985 ). The association of REE with Ca may thus be due to their presence in minor biogenic apatite (calcium phosphate). Liu et al. (1988) have attributed the striking peak for Sm enrichment in a manganese nodule observed by Goldberg et al. (1963) to apatite co-precipitation with the Mn-Fe oxyhydroxides and replacement of Ca 2÷ by the middle REE, particularly Sm 3+ whose ionic radius is similar to that of Ca 2+. The proportion of REE associated with biogenic apatite could not be detected, as no apatite peak could be observed from X-ray diffractograms. As calcium and phosphate contents added together form less than 5% (Table 3 ), standard X-ray diffraction techniques may not reveal any apatite peaks. Following Calvert et al. ( 1987 ), La/Sm and Yb/Sm ratios have been used (Tables 5 and 6) to study the REE fractionations. La/Sm ratios for this suite of samples indicate that the nodules from the Western Indian Ocean (Mascarene and Seychelles Basins ) are LREE enriched compared with the nodules from the Central Indian Basin. The La/Sm ratio varies from 5.0 to 5.2 in crusts and from 1.6 to 5.4 in manganese nodules of the Central Indian Basin (Table 5 ). Among the nodules from the Central Indian Basin, only those from the top of the seamount (Sample 142-0 ) and from the pelagic clay (Sample 75C ) have high La/Sm ratios (Table 5 ). On the other hand, the three nodules from the Western Indian Ocean (Mascarene Basin) all show a higher La/Sm ratio than the shale ratio. The nodule from the shallowest depth (Table 1 ) shows markedly the highest La/Sm ratio (Table 5 ). Therefore, only the nodules and encrustations from shallower depths (Table 1 ) show LREE enrichment in concurrence with the findings by Calvert et al. ( 1987 ).

REE GEOCHEMISTRY OF FERROMANGANESE NODULES

199

The HREE behave as a separate entity in this work. HREE are correlated well among themselves (Table 7); they relate only to middle REE (mainly Sm and Eu) and do not exhibit any relationship with major elements (Table 6 ). Such an observation suggests a possible fractionation of REE in this group of samples. The shale-normalized patterns (Fig. 2 ) depict a convex structure with enriched middle REE compared with LREE or HREE (excluding Ce and excepting Sample 320A, which is discussed below). HREE fractionation is further supported by Yb/Sm ratios. None of the samples have a higher Yb/ Sm ratio (except Sample 320A--Table 5 ) than that of shale (0.47; Sholkovitz, 1988 ). This indicates an HREE depletion in these nodules. This is in contrast to the HREE enrichment in seawater (Goldberg et al., 1963; Hogdahl et al., 1968; Elderfield and Greaves, 1982; De Baar et al., 1983, 1985a,b; Elderfield, 1988; Piepgras and Jacobsen, 1988 ) compared with the shale REE pattern (Gromet et al., 1984). HREE enrichment in seawater has often been attributed to their greater complexing capacities (Turner et al., 1981; Cantrell and Byrne, 1987 ) and to the greater stability of HREE complexes in seawater (Goldberg et al., 1963; Elderfield and Greaves, 1982). Hence, LREE compared with HREE are more susceptible to removal from seawater and to subsequent incorporation into manganese nodules. This may explain why we find LREE enrichment and HREE depletion in these nodules.

Cerium anomaly variations Large positive Ce anomalies have been often reported in manganese nodules from various locations (see Fleet (1984) for an earlier review and Brookins (1989) for a recent review). They are attributed to the oxidation of Ce(III) to CeO2(s) (Goldberg, 1961 ), although negative Ce anomalies are also reported for a few cases (Courtois and Clauer, 1980; Elderfield and Greaves, 1981; Calvert et al., 1987; Glasby et al., 1987 ). Many workers (e.g. Glasby, 1973; Piper, 1974a; Addy, 1979; Elderfield et al., 1981 a,b; Shaw and Wasserburg, 1985, and others) concur with the idea of Goldberg ( 1961 ) that Ce is preferentially taken up by nodules. All the samples studied here except Sample 320A show positive Ce anomalies (Figs. 2 and 4) between 0.12 (Sample 198-O) and 0.62 (crust PSET) (Fig. 4). The Central Indian Basin nodules, except the hydrogenetic nodule recovered from the red clay sedimentary domain, show low positive Ce anomalies (Fig. 4). All the nodules from the Western Indian Ocean (Mascarene and Somali-Seychelles Basins) show markedly high Ce anomalies. Ce anomalies show a negative relation with Mn/Fe (Fig. 5 ), implying that they are lowest in the nodules with the maximum diagenetic signature. Strong correlations between Ce and Ti, and between P and Fe (Table 6 ) indicate that Ce may be hosted by the Fe oxyhydroxide phase, and is probably removed from seawater in its higher valence state together with Co and Ti. A weak correla-

200

MAR

B. NAGENDER NATH ET AL.

1

MAR 3 MAR 22 134/10

l

GAR-EN

1

649H

I

PSET 198-0 142-0 219-0

~i!i.

~ []

WlO n o d u l e s

[]

ClB n o d u l e s

crusts

320A

284A 75C

0-1T

ii!iiii :jiiiiiiiiiiiiiii!i!iiiill ii! !i iiiiii/ iiiii[

0-1B 0-2T 0-2B I

0 0.1

I

I

0.3

I

I

0.5

1

I

1

0.7

Fig. 4. Magnitude of cerium anomalies. Ce anomalies are calculated following Elderfield and Greaves (1981), using the formula log (3 Ce,/2 (La,+Ndn)), where subscript n represents shale-normalized values. They are highest in all the samples from the Western Indian Ocean, but in the Central Indian Ocean are high only in the encrustations and the hydrogenetic nodule.

tion between Ce and Ca in this study and the commonly observed negative Ce anomalies of the phosphatic phases (Fleet ( 1984 ) and references therein) favours the partitioning of most of the Ce into the Fe phase. The samples with 6-MnO2 (vernadite) as the major ferromanganese mineral have the highest Ce anomalies and all these 6-MnO2-bearing nodules are recovered from the shallower depths (Table 1). 6-MnO2 is the main phase in the hydrogenetic substance forming in the well-oxygenated deep-sea environment and is intimately intergrown with FeOOH.nH20 and aluminosilicates (Halbach, 1986). The association of an enhanced Ce anomaly with 6-MnO2 could be due to its presence in the Mn phase (Koeppenkastrop and De Carlo, 1988, 1990) or may be due to a probable catalysing effect of the MnO2 surface on the oxidation of C e 3+. The variation of Ce anomalies, 6-MnO: contents and LREE/HREE fractionations with geographic locations has been correlated with the bottom-water

REEGEOCHEMISTRYOF FERROMANGANESENODULES I , I J J i l

"'~'

I ....

. . . . . . . . .

....

201

I . . . . . . . . .

mm m

m

[]

m m

E

m

0

m

m

mm i

m

0 0

IB fll

u

I

. . . . . . . . .

I . . . . . . . . .

1

I . . . . . . .

2

I . . . . . . . . .

3

4

Mn/Fe

Fig. 5. Scatter plot between cerium anomalies and Mn/Fe ratios showing a negative relationship, which seems to be driven by the dominant mineral present in these nodules. Nodules with higher Mn/Fe ratios, i.e. todorokite-bearing nodules, have lower Ce anomalies.

temperatures. Bottom-water potential temperatures in the Indian Ocean are governed to a large extent by the movement of Antarctic Bottom Water (AABW). The lowest bottom-water potential temperature is - 0.6 ° C ( Kolla et al., 1976 ) in the Atlantic-Indian Basin (in the west) and the South Indian Basin (in the east ), and the highest temperature is 1.2 ° C in the Arabian Basin (in the west) and the Central Indian Basin (in the east). Reported temperature values for the Central Indian Basin (0.96-1.2 °C ) are much higher (Kolla et al., 1976 ) than those found in the Mascarene Basin (0.6-0.9 °C). Kolla et al. (1976) have suggested that although the AABW enters the Central Indian Basin through sills in the Ninety East Ridge and Southeast Indian Ridge, the influx of AABW is less in the Central Indian Basin than in other basins of the Indian Ocean. On the other hand, the low bottom-water temperatures suggest definite AABW effects in the Mascarene Basin. On the whole, it has been observed (by Kolla et al. ( 1976, 1980) and Kolla and Kidd (1982) and references therein) that the AABW effects are more marked on the seafloor in the southern and western regions of the Indian Ocean. In addition, the bottom-water dissolved oxygen levels of the Central Indian Basin are lower than in the Mascarene Basin (Wyrtki, 1971 ). The Central Indian Basin nodules with higher bottom-water temperatures and low dissolved oxygen contents show lesser REE enrichments, low degree of fractionation and lower Ce anomalies. On the other hand, the nodules from the Mascarene and Somali-Seychelles Basins show enriched REE contents relative to those of the Central Indian Basin (enrichment represented by Western Indian Ocean to Central Indian Basin ratios is in the order Ce

202

B. NAGENDER NATH ET AL.

(2.24)>La (1.31)>Lu (l.24)>Er (l.21)>Tm, Yb ( 1 . 2 ) > H o ( 1.19 ) > Gd, Pr ( 1.15 ) > Nd ( 1.13 ) > Sm ( 1.04) ), and Ce anomalies. The Ce anomaly of the Somali-Seychelles nodule is lower (0.43) than that of Mascarene Basin nodules (0.49-0.55), reflecting the flow of AABW from south to north. These findings lead us to assume that in addition to the mineralogical control, hydrographic features and redox conditions also play a role in the REE concentrations.

Gadolinium-terbium anomalies When compared with the neighbouring elements, the shale-normalized Gd value shows a positive departure, with a corresponding negative Tb anomaly, in all the nodules and crusts studied here (Fig. 2 ). Similar distinct Gd positive anomalies and depressed Tb values have been reported for seawater by De Baar et al. (1985b). They have also been previously reported in Fe-Mn oxide crusts from the Pacific Ocean by Hein et al. (1988) and De Carlo (1990). Gd anomalies in seawater have been attributed to a higher degree of complexation of Gd, resulting in reduced scavenging of dissolved Gd relative to its neighbours Eu and Tb (De Baar et al., 1985b). It may therefore be expected that marine authigenic deposits would, on average, exhibit a complementary negative Gd anomaly (De Baar et al., 1985b), whereas the REE patterns observed in the present study are similar to the seawater pattern. Such an observation is of interest, as the REE patterns of nodules are generally a mirror image of the seawater pattern. For example, there is a gradual increase in the shale-normalized HREE pattern of seawater but a decrease in the pattern for the nodules studied here. If the aqueous complexing capacity of Gd is as high as reported by De Baar et al. (1985b), the nodules should have shown a depleted behaviour similar to that of HREE. Additional data from various locations are needed to explain the behaviour of Gd in the oceanic environment. The mechanism leading to this similarity in the shale-normalized Gd patterns of seawater and manganese nodules may be analogous to the process of retention of negative cerium anomalies from seawater by manganese nodules at certain locations (Piper, 1974a; Courtois and Clauer, 1980; Elderfield and Greaves, 1981; Calvert et al., 1987; Glasby et al., 1987).

HREE enrichment in Sample 320A Sample 320A shows an REE pattern (Fig. 2 ) markedly different from those of other nodules, with a significant HREE enrichment and no Ce anomaly. The interior of this nodule, including the type of nucleus, was found to be different from that of other nodules. This sample is composed of brownish, fibrous material with disseminated shining phillipsite (?) crystals. X-ray dif-

REE GEOCHEMISTRY OF FERROMANGANESE NODULES

203

fraction studies revealed quartz, illite, smectite, plagioclase and kaolinite. Smectite was confirmed by the usual method of glycolation (Warshaw and Roy, 1961). Furthermore, this material was identified as palagonite. Palagonite is a mixture of altered, hydrated, oxidized glass, and composed of a dioctahedral smectite with significant Mg in the octahedral sheet (Eggleton and Keller, 1982 ). Palagonites are the common alteration products of submarine basalts and glasses (Honnorez, 1981, and references therein; Miihe and Frenzel, 1989; Zhou and Fyfe, 1989 ). However, there is little agreement on the overall magnitude or the potential for fractionation among REE during the low-temperature alteration of basalts (McLennan, 1989, and references therein; Grauch, 1989, and references therein). The halmyrolytic minerals formed from basalts, such as smectite and phillipsite, display REE patterns similar to those of seawater (Piper, 1974a,b; Bernat, 1975; Courtois and Hoffert, 1977; Desprairies and Bonnot-Courtois, 1980). The REE pattern of Sample 320A (Fig. 2), with no Ce anomaly and HREE enrichment, is comparable with that for the smectites from the South Pacific which are formed from halmyrolitic reactions involving volcanic lithogenous material and seawater (Piper, 1974a; Courtois and Hoffert, 1977). These variations underline the importance of REE in nucleus material and the potential effects of the nucleus material on the bulk REE contents. The importance of the nucleus may be more significant in the nodules with low oxide/nucleus ratios. This further suggests that one should be cautious in comparing the results of separated oxide and bulk analyses of whole samples. SUMMARY

The REE distributions in the nodules and crusts from the Western Indian Ocean exhibit notable enrichments (of 10-30%) relative to the nodules from the more diagenetic environment of the Central Indian Basin. For Ce, the enrichments are more pronounced (up to 220%). The preferential enrichment of Ce and other REE in the nodules from the Western Indian Ocean is consistent with the oxygenated conditions created by the AABW flow. Superimposed on these oceanographic conditions are redox conditions within the sedimentary environment. The restriction of higher REE contents and higher Ce anomaly values to &MnO2 nodules and crusts, and of the opposite pattern to todorokite nodules, and the inverse relation between some of the REE (especially Ce) and Mn/Fe suggest that nodule REE chemistry may also be linked to sediment diagenesis. Interelement correlations between Ca, Fe, P and trivalent REE support the idea of Elderfield et al. (198 lb) that REE probably reside in two phases, an iron oxyhydroxide phase with phosphate chemisorbed onto it, and a minor apatitic phase. On the other hand, Ce seems to be associated with vernadite

204

B.NAGENDERNATHETAL.

that is richer in iron oxyhydroxide than todorokite, with MnO2 catalysing the oxidation of Ce to an insoluble tetravalent form. ACKNOWLEDGEMENTS The sampling work was carried out with the financial support of the Department of Ocean Development, G o v e r n m e n t of India, as part of their programme 'Polymetallic Nodules Survey'. B.N. thanks the G e r m a n Academic Exchange Service ( D A A D ) for support while writing this paper. We thank the Directors of the National Institute of Oceanography, Goa, the National Geophysical Research Institute, Hyderabad, and the Institut f'tir Mineralogie u n d Lagerst~ttenlehre, Aachen, for providing facilities. B.N. and M.S. thank R.R. Nair, Project Leader, for encouragement to undertake this work, Dr. R. M u k h o p a d h y a y for providing the three samples from seamount environment and unpublished data, and M. Shyam Prasad for crust samples and for his help during microscopic work on Sample 320A. Jai Sankar is thanked for word-processing work. We wish to express our sincere thanks to Drs. H. Kunzendorf, G.P. Glasby, P.M. Herzig, A.C. Aplin, S.E. Calvert and E.D. Sholkovitz for helpful suggestions and c o m m e n t s on an earlier version of the manuscript; and Professor P.J. Wangersky and two a n o n y m o u s reviewers for their work on a later version.

REFERENCES Addy, S.K., 1979. Rare earth element patterns in manganese nodules and micronodules from northwest Atlantic. Geochim. Cosmochim.Acta, 43:1105-1115. Ahmed, S.M. and Hussain, A., 1987. Geochemistry of ferromanganese nodules from the Central Indian Basin. Mar. Geol., 77:165- t70. Aplin, A.C., 1984. Rare earth element geochemistryof Central Pacific ferromanganeseencrustations. Earth Planet. Sci. Lett., 71: 13-22. Balaram, V. and Saxena, V.K., 1988. Determination of rare earth elements (REE) in standard geologicalreference samples by InductivelyCoupled Plasma-Mass Spectrometry (ICP-MS) using indium as internal standard. First Intemational Conference on Plasma Source Mass Spectrometry, Durham University, UK, 12-16 September 1988, p. 44. (Abstr.) Banakar, V.K., 1990. Uranium-thorium isotopes and transition metal fluxes in two oriented nodules from the Central Indian Basin: implications for nodule turnover. Mar. Geol., 95: 71-76. Baturin, G.N., 1988. The Geochemistry of Manganese and ManganeseNodules in the Ocean. D. Riedel, Dordrecht, 342 pp. Ben Othman, D., White, W.M. and Patchett, P.J., 1989. The geochemistryof marine sediments, island arc magma genesis, and crust-mantle recycling.Earth Planet. Sci. Lett., 94:1-2 I. Bernat, M., 1975. Les isotopes de ruranium et du thorium et les terres rares dans renvironment marin. Cah. ORSTOM S6r. Geol., 7: 65-83. Brookins, D.G., 1989. Aqueous geochemistryof rare earth elements. In: B.R. Lipin and G.A.

REE GEOCHEMISTRY OF FERROMANGANESE NODULES

205

McKay (Editors), Geochemistry and Mineralogy of Rare Earth Elements. Mineral. Soc. Am. Rev. Mineral., 21: 201-225. Calvert, S.E. and Piper, D.Z., 1984. Geochemistry of ferromanganese nodules from DOMES Site A, northern equatorial Pacific: Multiple diagenetic metal sources in the deep sea. Geochim. Cosmochim. Acta, 48: 1913-1928. Calvert, S.E. and Price, N.B., 1970. Composition of manganese nodules and manganese carbonates from Loch Fyne, Scotland. Contrib. Mineral. Petrol., 29:215-233. Calvert, S.E. and Price, N.B., 1977. Geochemical variation in ferromanganese nodules and associated sediments from the Pacific Ocean. Mar. Chem., 5: 43-74. Calvert, S.E., Price, N.B., Heath, G.R. and Moore, T.C., 1978. Relationship between ferromanganese nodules' composition and sedimentation in a small survey area of the equatorial Pacific. J. Mar. Res., 36: 161-183. Calvert, S.E., Piper, D.Z. and Baedecker, P.A., 1987. Geochemistry of the rare earth elements in ferromanganese nodules from DOMES Site A, northern equatorial Pacific. Geochim. Cosmochim. Acta, 51: 2331-2338. Cantrell, K.J. and Byrne, R.H., 1987. Rare earth element complexation by carbonate and oxalate ions. Geochim. Cosmochim. Acta, 51: 597-605. Courtois, C. and Clauer, N., 1980. Rare earth elements and strontium isotopes of polymetallic nodules from southeastern Pacific Ocean. Sedimentology, 27: 687-695. Courtois, C. and Hoffert, M., 1977. Distribution des terres rares dans les srdiments superficiels du Pacifique sud-est. Bull. Soc. Grol., 7: 65-83. Cronan, D.S., 1980. Underwater Minerals. Academic Press, London, 362 pp. Cronan, D.S. and Moorby, S.A., 1981. Manganese nodules and ferromanganese oxide deposits from the Indian Ocean. J. Geol. Soc. London, 138: 527-539. De Baar, H.J.W., Bacon, M.P. and Brewer, P.G., 1983. Rare earth distribution with a positive Ce anomaly in the Western North Atlantic Ocean. Nature, 301: 324-327. De Baar, H.J.W., Bacon, M.P., Brewer, P.G. and Bruland, K.W., 1985a. Rare earth elements in the Pacific and Atlantic Oceans. Geochim. Cosmochim. Acta, 49:1943-1959. De Baar, H.J.W., Brewer, P.G. and Bacon, M.P., 1985b. Anomalies in rare earth distributions in seawater: Gd and Tb. Geochim. Cosmochim. Acta, 49:1961-1969. De Baar, H.J.W., German, C.R., Elderfield, H. and Gaans, P.V., 1988. Rare earth element distributions in anoxic waters of the Cariaco Trench. Geochim. Cosmochim. Acta, 52: 12031219. De Carlo, E.H., 1990. Separation of lanthanide series elements in marine Fe-Mn crusts by ionexchange chromatography and determination by ICP/AES. Sep. Sci. Tech., 25 (6): 781-798. Desprairies, A. and Bonnot-Courtois, C., 1980. Relation entre la composition des smectites d'altrration sous-marine et leur cortege de terres rares. Earth Planet. Sci. Lett., 48:124-130. Dymond, J., Lyle, M., Finney, B., Piper, D.Z., Murphy, K., Conard, R. and Pisias, N., 1984. Ferromanganese nodules from MANOP Sites H, S and R--Control of mineralogical and chemical composition by multiple accretionary processes. Geochim. Cosmochim. Acta, 48: 931-949. Eggleton, R.A. and Keller, J., 1982. The palagonitization of limburgite glass--A TEM study. Neues Jahrb. Mineral. Monatsh., 7: 321-336. Ehrlich, A.M., 1968. Rare earth abundances in manganese nodules. Ph.D. Dissertation, MIT, Cambridge, MA, 222 pp. Elderfield, H., 1988. The oceanic chemistry of the rare earth elements. Philos. Trans. R. Soc. London, Ser. A, 325: 105-126. Elderfield, H. and Greaves, M.J., 1981. Negative cerium anomalies in the rare earth element patterns of oceanic ferromanganese nodules. Earth Planet. Sci. Lett., 55:163-170. Elderfield, H. and Greaves, M.J., 1982. The rare earth elements in seawater. Nature, 296: 214217.

206

B. NAGENDER NATH ET AL.

Elderfield, H. and Sholkovitz, E.R., 1987. Rare earth elements in the porewaters of reducing nearshore sediments. Earth Planet. Sci. Lett., 55:163-170. Elderfieid, H., Hawkesworth, C.J., Greaves, M.J. and Calvert, S.E., 1981 a. Rare earth element zonation in Pacific ferromanganese nodules. Geochim. Cosmochim. Acta, 45:123 l - 1234. Eiderfield, H., Hawkesworth, C.J., Greaves, M.J. and Calvert, S.E., 198lb. Rare earth element geochemistry of oceanic ferromanganese nodules and associated sediments. Geochim. Cosmochim. Acta, 45:513-528. Flanagan, F.J. and Gottfried, D., 1980. USGS rock standards. III. Manganese nodule reference samples USGS-Nod A-1 and USGS-Nod P-1. U.S. Geol. Surv. Prof. Paper, 1155, 36-39 pp. Fleet, A.J., 1984. Aqueous and sedimentary geochemistry of the rare earths. In" P. Henderson (Editor), Rare Earth Element Geochemistry. Elsevier, Amsterdam, pp. 343-373. Friedrich, G., Pltiger, W.L. and Kunzendorf, H., 1988. Chemical composition of manganese nodules. In: P. Halbach, G. Friedrich and U. von Stackelberg (Editors), The Manganese Nodule Belt of the Pacific Ocean. Ferdinand Enke, Stuttgart, pp. 37-50. German, C.R. and Elderfield, H., 1989. Rare earth elements in Saanich Inlet, British Columbia, a seasonally anoxic basin. Geochim. Cosmochim. Acta, 53:2561-2572. Glasby, G.P., 1973. Mechanisms of enrichment of the rarer elements in marine manganese nodules. Mar. Chem., l: 105-125. Glasby, G.P., 1977. Marine Manganese Deposits. Elsevier, Amsterdam, 523 pp. Glasby, G.P. and Thijssen, T., 1982. Control of the mineralogy and composition of marine manganese nodules by the supply of divalent transitional metal ions. Neues Jahrb. Mineral., 145: 291-307. Giasby, G.P., Hunt, J.L. and Rennet, R.M., 1985. Trace element analysis of marine sediments from the southwest Pacific. N.Z. Soil Bur. Sci. Rep., 53, 62 pp. Glasby, G.P., Gwozdz, R., Kunzendorf, H., Friedrich, G. and Thijssen, T., 1987. The distribution of rare earth and minor elements in manganese nodules and sediments from the equatorial and SW Pacific. Lithos, 20:97-113. Goldberg, E.D., 196 I. Chemistry in the Oceans. In: M. Sears (Editor), Oceanography, Vol. 67. Am. Assoc. Adv. Sci., Washington, D.C., pp. 583-587. Goldberg, E.D., Koide, M., Schmitt, R.A. and Smith, R.H., 1963. Rare earth distributions in the marine environment. J. Geophys. Res., 68:4209-4217. Goldstein, S.L. and O'Nions, R.K., 1981. Nd and Sr isotopic relationships in pelagic clays and ferromanganese deposits. Nature, 292: 324-327. Grauch, R.J., 1989. Rare earth elements in metamorphic rocks. In: B.R. Lipin and G.A. McKay (Editors), Geochemistry and Mineralogy of Rare Earth Elements. Mineral. Soc. Am. Rev. Mineral., 21: 147-167. Gromet, L.P., Dymek, R.F., Haskin, L.A. and Korotev, R.L., 1984. The "North American Shale Composite". Its compilation, major and trace element characteristics. Geochim. Cosmochim. Acta, 48: 2469-2482. Halbach, P., 1986. Processes controlling the heavy metal distribution in Pacific ferromanganese nodules and crusts. Geol. Rundsch., 75 ( 1 ): 235-247. Halbach, P., Friedrich, G. and von Stackelberg, U., 1988. The Manganese Nodule Belt of the Pacific Ocean. Ferdinand Enke, Stuttgart, 255 pp. Hein, J.R., Schwab, W.C. and Davies, A.S., 1988. Cobalt and platinum rich ferromanganese crusts and associated substrate rocks from the Marshall Islands. Mar. Geol., 78: 255-283. Hogdahl, O.T., Melson, S. and Bowen, V.T., 1968. Neutron activation analysis of lanthanide elements in seawater. In: R.A. Baker (Editor), Trace Inorganics in Water. Adv. Chem. Ser., 73: 308-325. Honnorez, J., 1981. The aging of the oceanic crust at low temperatures. In: C. Emiliani (Editor), The Sea, Vol. 7. The Oceanic Lithosphere. Wiley, New York, pp. 525-587.

REE GEOCHEMISTRY OF FERROMANGANESE NODULES

207

Ingri, I. and Ponter, C., 1987. Rare earth abundance patterns in ferromanganese concretions from the Gulf of Bothnia and the Barents Sea. Geochim. Cosmochim. Acta, 51: 155-161. Jarvis, K., 1988. Low level determination of the rare earth elements in rocks and minerals by inductively coupled plasma mass spectrometry. (Abstr.) Chem. Geol., 70 ( 1/2): 175. Jonasson, R.G., Bancroft, G.M. and Nesbitt, H.W., 1985. Solubilities of some hydrous REE phosphates with implications for diagenesis and seawater concentrations. Geochim. Cosmochim. Acta, 49:2133-2139. Klinkhammer, G., Elderfield, H. and Hudson, A., 1983. Rare earth elements in seawater near hydrothermal vents. Nature, 305:185-188. Koeppenkastrop, D. and de Carlo, E.H., 1988. Adsorption of rare earth elements from seawater onto synthetic mineral phases. EOS, 69(44): 1254. Koeppenkastrop, D. and de Carlo, E.H., 1990. Distribution of rare earth elements between seawater and synthetic mineral phases. EOS, 71 (43): 1417. Kolla, V. and Kidd, R., 1982. Sedimentation and sedimentary processes in the Indian Ocean. In: A.E.M. Nairn and F.G. Stehli (Editors), The Ocean Basins and Margins. Vol. 6. The Indian Ocean. Plenum, New York, pp. 1-45. Kolla, V., Sullivan, L., Streeter, S.S. and Langseth, M.G., 1976. Spreading of Antarctic Bottom Water and its effects on the floor of the Indian Ocean inferred from bottom-water potential temperature, turbidity and sea-floor photography. Mar. Geol., 21: 171-189. Kolla, V., Eittreim, S., Sullivan, S., Kostecki, J.A. and Burckle, L.H., 1980. Current-controlled abyssal microtopography and sedimentation in Mozambique Basin, Southwest Indian Ocean. Mar. Geol., 34: 171-206. Krishnaswami, S. and Cochran, J.K., 1978. Uranium and thorium series nuclides in oriented ferromanganese nodules: growth rates, turnover times, and nuclide behaviour. Earth Planet. Sci. Lett., 40: 45-62. Kunzendorf, H., Gwozdz, R., Friedrich, G. and Glasby, G.P., 1987. Minor and trace elements in manganese nodules and sediments from the Pacific. In: R.W. Hurst and T.E. Davies (Editors), The Practical Applications of Trace Elements and Isotopes in Environmental Biogeochemistry and Mineral Resource Evaluation. Theophrastus, Athens, pp. 181-196. Kunzendorf, H., Gwozdz, R., Glasby, G.P., Stoffers, P. and Renner, R., 1989. The distribution of rare earth elements in manganese micronodules and sediments from the equatorial and southwest Pacific. Appl. Geochem., 4 (2): 183-194. Liu, Y.G., Miah, M.R.U. and Schmitt, R.A., 1988. Cerium: A chemical tracer for paleo-oceanic redox conditions. Geochim. Cosmochim. Acta, 52:1361-1371. McLennan, S.M., 1989. Rare earth elements in sedimentary rocks: Influence of provenance and sedimentary processes. In: B.R. Lipin and G.A. McKay (Editors), Geochemistry and Mineralogy of Rare Earth Elements. Mineral. Soc. Am. Rev. Mineral., 21:169-199. Moorby, S.A. and Cronan, D.S., 1981. The distribution of elements between coexisting phases in some marine ferromanganese oxide deposits. Geochim. Cosmochim. Acta, 45:1855-1877. Moore, W.S., 1984. Thorium and radium isotopic relationships in manganese nodules and sediments at MANOP Site S. Geochim. Cosmochim. Acta, 48: 987-992. Miihe, R. and Frenzel, G., 1989. On the submarine palagonitization of volcanic glasses. In: K.S. Balasubramaniam, G. Faure, J. Goni, P.L.C. Grubb, D. Evangelou, P.A. Hill, C. LahodneySarc, E. Mendelovici, A.J. Melfi, A.P. Nikitana, W.F. Pickering and S.S. Augustithis (Editors), Weathering; Its Products and Deposits, Vol. I. Processes. Theophrastus, Athens, pp. 383-419. Mukhopadhyay, R., 1988. Morphological and geochemical studies of the polymetallic nodules from four sectors in the Central Indian Ocean Basin between 11 degrees south and 16 degrees south latitudes. Ph.D. Thesis, University of Calcutta, 93 pp. Mukhopadhyay, R. and Nagendernath, B., 1988. Influence of seamount topography on the local

208

a. N A G E N D E R N A T H ET AL.

facies variation in ferromanganese deposits in the Indian Ocean. Deep-Sea Res., 35:14311436. Murphy, K. and Dymond, J., 1984. Rare earth element fluxes and geochemical budget in the eastern equatorial Pacific. Nature, 307: 444-447. Pattan, J.N. and Mudholkar, A.V., 1990. The oxidation state of Mn in ferromanganese nodules and deep-sea sediments from the Central Indian Ocean. Chem. Geol., 85:171-181. Piepgras, D.J. and Jacobsen, S.B., 1988. The isotopic composition of neodymium in the N. Pacific. Geochim. Cosmochim. Acta, 52: 1373-1381. Piper, D.Z., 1974a. Rare earth elements in ferromanganese nodules and other marine phases. Geochim. Cosmochim. Acta, 38: 1007-1022. Piper, D.Z., 1974b. Rare earth elements in the sedimentary cycle: a summary. Chem. Geol., 14: 285-304. Piper, D.Z., 1988. The metal oxide fraction of pelagic sediment in the equatorial North Pacific Ocean: A source of metals in ferromanganese nodules. Geochim. Cosmochim. Acta, 52:21272145. Raab, W., 1972. Physical and chemical features of Pacific deep-sea manganese nodules and their implications to the genesis of nodules. In: D.R. Horn (Editor), Ferromanganese Deposits on the Ocean Floor. National Science Foundation, Washington, DC, pp. 31-49. Rankin, P.C. and Glasby, G.P., 1979. Regional distribution of rare earth and minor elements in manganese nodules and associated sediments in the southwest Pacific and other localities. In: J.L. Bischoffand D.Z. Piper (Editors), Marine Geology and Oceanography of the Pacific Manganese Nodule Province. Plenum, New York, pp. 681-697. Schlich, R., 1982. The Indian Ocean: Aseismic Ridges, Spreading Centres and Oceanic Basins. In: A.E.M. Nairn and F.G. Stehli (Editors), The Ocean Basins and Margins. Vol. 6. The Indian Ocean. Plenum, New York, pp. 51-148. Shaw, H.F. and Wasserburg, G.J., 1985. S m - N d in marine carbonates and phosphates: Implications for Nd isotopes in seawater and crustal ages. Geochim. Cosmochim. Acta, 49: 503518. Sholkovitz, E.R., 1988. Rare earth elements in the sediments of the N. Atlantic Ocean, Amazon Delta, and East China Sea: Reinterpretation ofterrigenous input patterns to the oceans. Am. J. Sci., 288: 236-281. Sholkovitz, E.R., Piepgras, D.J. and Jacobsen, S.B., 1989. The porewater chemistry of rare earth elements in Buzzards Bay sediments. Geochim. Cosmochim. Acta, 53 ( 11 ): 2847-2856. Skala, W., 1979. Some effects of the constant-sum problem in geochemistry. Chem. Geol., 27: 1-9. Sudhakar, M., 1989. Ore grade manganese nodules from the Central Indian Basin: An evaluation. Mar. Mining, 8(2): 201-214. Turner, D.R., Whitfield, M. and Dickson, A.G., 1981. The equilibrium speciation of dissolved components in freshwater and seawater at 25 degrees centigrade and 1 atm. pressure. Geoclaim. Cosmochim. Acta, 45:855-881. Warshaw, C.M. and Roy, R., 1961. Classification and a scheme for the identification of layer silicates. Geol. Soc. Am., Bull., 72: 1453-1492. Wyrtki, K., 1971. Oceanographic Atlas of the International Indian Ocean Expedition. National Science Foundation, Washington, DC, 531 pp. Zhou, Z. and Fyfe, W.S., 1989. Palagonitization of basaltic glass from DSDP Site 335, Leg 37: Textures, chemical composition and mechanism of formation. Am. Mineral., 74:1045-1053.