42. GEOCHEMISTRY OF RARE EARTHS AND OTHER TRACE ELEMENTS …€¦ · geochemistry of major, minor,...

Pisciotto, K. A., Ingle, J. C, Jr., von Breymann, M. T., Barron, J., et al., 1992Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 127/128, Pt. 1

42. GEOCHEMISTRY OF RARE EARTHS AND OTHER TRACE ELEMENTS IN SEDIMENTS FROMSITES 798 AND 799, JAPAN SEA1

Yoshitaka Minai,2 Ryo Matsumoto,3 Yoshio Watanabe,4 and Takeshi Tominaga2

ABSTRACT

Nineteen trace elements, including seven rare earth elements (REE's), and 10 major and minor elements in 76 sedimentsamples from Sites 798 (Oki Ridge) and 799 (Yamato Trough) were determined by means of instrumental neutron activationanalysis and X-ray fluorescence spectrometry. Most REE patterns (chondrite-normalized) of the sediments from both sites werenearly identical to the patterns of terrigenous materials. The cerium anomaly (slightly positive) frequently appeared in REEpatterns of the sediments (200-750 mbsf) from Site 799. Cerium may be selectively incorporated into the sediments withhydrogenous manganese precipitation. However, the degree of the anomaly was not well correlated with manganese content,suggesting that cerium may behave as a trivalent REE (like the other REE's) during diagenesis while manganese is transportedin the sediment column accompanied by reduction to a lower oxidation state. The Th/Sc ratio of the sediments from Sites 798and 799 tended to decrease with penetration depth. Such a depth profile may indicate a decrease in basic volcanism activitiesfrom the Pliocene (Site 798) and Miocene (Site 799). The La/Yb ratio and degree of europium anomaly also varied with depth,which may imply that two or more components with different REE patterns were supplied throughout sedimentation at sites inthe Japan Sea.

INTRODUCTION

The geology and geophysics of the Japan Sea have been exten-sively studied over the past two decades as a typical backarc basinsystem. However, the geological history of the Japan Sea system isstill controversial. From Sites 798 (Oki Ridge) and 799 (YamatoTrough) of Leg 128 (Fig. 1), 590-m and 1084-m long sediment coreswere recovered. These cores provide information on changes ofsedimentation environments from the Pliocene (Site 798) or earlyMiocene (Site 799) to the Holocene. In this article, we report thegeochemistry of major, minor, and trace elements, including sevenrare earth elements (REE's), in sediments from the Japan Sea, basedon the analysis of 76 sediments from Sites 798 and 799. The geochem-istry of REE and other trace elements has provided useful informationfor estimating the origin and formation processes of marine sediments(e.g., Bender et al., 1971; Matsumoto et al., 1985; Minai and Tomi-naga, 1989; McLennan et al., 1990; Toyoda and Masuda, 1990;Toyoda et al., 1990).

We have studied the geochemistry of REE's and other traceelements in deep sea sediments collected from the Pacific (Minai,1982; Minai and Tominaga, 1989) and from DSDP sites at the NankaiTrough, the Japan Trench, and adjacent regions (Minai et al., 1985).Previously, a model for classification of oceanic sediment was pro-posed (Matsumoto et al., 1985) based on concentrations of the REE'sreported in the literature (including our studies mentioned above). Onthe basis of this review we have proposed that selective depletion ofcerium in the REE pattern (negative cerium anomaly) of oceanicsediments can be regarded as a measure of hydrothermal activity. Asreported previously (Minai et al., 1985), such an anomaly was ob-served in the REE patterns of pelagic sediment from Site 294 (westernPhilippine Basin) and of sedimentary rocks near the basements at

Pisciotto, K. A., Ingle, J. C, Jr., von Breymann, M. T., Barron, J., et al., 1992. Proc.ODP, Sci. Results, 127/128, Pt. 1: College Station, TX (Ocean Drilling Program).

2 Department of Chemistry, Faculty of Science, University of Tokyo, Hongo, Tokyo113, Japan.

3 Geological Institute, Faculty of Science, University of Tokyo, Hongo, Tokyo 113,Japan.

4 Geological Survey of Japan, Tsukuba, Ibaraki 305, Japan.

Sites 296 (the Shikoku Basin) and 436 (the Japan Trench). One of ourobjectives in this article was to investigate traces of hydrothermalactivities in the sediment columns from the Japan Sea, based on thegeochemistry of REE's and other trace elements.

The primary objective of this work is to elucidate the sedimentationhistory of the Japan Sea. It has been claimed that Japan Sea sedimen-tation was mainly controlled by both volcanic activities (accompaniedby hydrothermal activities) and rapid influx of terrigenous materialsderived from the continent and the Japanese Islands (Karig, Ingle, etal., 1975; Tamaki, 1988). In addition, it is probable that the opening ofthe Japan Sea was caused by complicated processes of rifting andspreading of centers at various ages. As shown in our studies onsediments from various regions around the Japanese Islands (Minai etal., 1985), concentrations of REE's and other trace elements and theirrelative correlations are extremely useful for elucidating the origin ofterrigenous materials in the sediments. It can be expected that detailedstudies of REE patterns and their relations to the abundances of otherelements in sediments may provide clues to elucidating the complexhistory of sedimentation processes in the Japan Sea.

EXPERIMENTAL

For each sample, a few grams of sediment were washed with deion-ized water to remove water-soluble species (e.g., NaCl) in the sedimentcolumns before being dried in an oven at 50° C. The air-dried sampleswere finely pulverized for X-ray fluorescence analysis (XRF) and instru-mental neutron activation analysis (INAA).

Five hundred milligrams of the pulverized sample were ignited at600°C to determine loss on ignition (LOI). After the determination ofLOI, the powders were fused with anhydrous lithium borate to preparea bead sample for XRF. The analysis was carried out with a PhillipsIW-1400 XRF spectrometer. Ten major and minor elements (Si, Ti,Al, Fe, Mn, Mg, Ca, Na, K, and P) were determined with correctionof LOI, applying calibration curves prepared by analysis of variousstandard rocks.

One hundred milligrams of the pulverized sample was taken forINAA. The portion was sealed in polyethylene film for neutronirradiation. Ten to twenty samples together with standard rocks (JB-1and JLk-1 issued by the Geological Survey of Japan), whose recom-mended values of the elemental concentrations were presented in

719

Y. MINAI, R. MATSUMOTO, Y. WATANABE, T. TOMINAGA

42 °N

40'

38'

36C

128°E 130° 132° 134'

Figure 1. Locations of Sites 798 and 799, Leg 128.

Ando et al. (1989) and Ebihara et al. (1985), were irradiated for 18-24hr with 5x1011 cm~V1 thermal neutron flux in a TRIGA-II nuclearreactor at Rikkyo University. Iron wires were applied for monitoringneutron flux at the sample positions. For 5-7 days after the end ofirradiation, As, Ba, Br, U, La, Sm, and Lu concentrations weredetermined by means of gamma-ray spectrometry, using an intrinsicgermanium semiconductor detector (Laboratory for Earthquake Chem-istry, University of Tokyo) or a lithium-drifted germanium semiconduc-tor detector (Radioisotope Center, University of Tokyo). Also, for 20-40days after the end of irradiation Co, Cr, Cs, Hf, Sb, Se, Ta, Th, Ce, Eu,Tb, and Yb concentrations were determined by the same method. Ana-lytical precisions estimated from the counting statistics for Sample128-799A-3H-4, 41-^-3 cm, as an example, are as follows: As (11%),Ba (17%), Br (15%), Co (1%), Cr (2%), Cs (9%), Hf (4%), Sb (7%),Se (0.3%), Ta (5%), Th (1%), U (11%), La (1%), Ce (1%), Sm (1%),Eu (3%), Tb (15%), Yb (7%), and Lu (9%).

136° 138° 140° 142°

RESULTS

Site 798

Site 798 (Fig. 1) is located at a small basin on the top of Oki Ridge(Hole 798B: 37°03.848'N, 134°79.961'E). A517.9-m long sedimentcore of late early Pliocene to Holocene sediment was recovered fromHole 798B at this site, where the water depth was 911.1 m. At anaverage sedimentation rate of about 120 m/m.y. the pelagic-hemipe-lagic sediments were deposited above the carbonate compensationdepth (Ingle, Suyehiro, von Breymann, et al., 1990). The sedimentcolumn was divided into three lithological units: Unit I (0-220 mbsf,latest Quaternary to latest Pliocene), Unit II (220-455 mbsf, latePliocene), and Unit III (455-518 mbsf, late to late early Pliocene)(Ingle, Suyehiro, von Breymann, et al., 1990).

Analytical XRF and INAA results are listed in Table 1 (for majorand minor elements) and Table 2 (for trace elements). Depth profiles

Table 1. Major and minor element contents in the sediments from Site 798.

Sampleno.

123456789

1011121314151617181920

Core, section,interval (cm)

798A-1H-4, 86-881H-5, 94-965H-2, 86-885H-3, 85-82

11H-2, 97-9911H-3, 90-92

798B-14H-4, 92-9418X-2, 90-9221X-4, 98-10025X-1 ,95-9730X-3, 90-9233X-2, 90-9237X-3, 10-1240X-4, 123-12543X-5, 90-9247X-6, 30-3249X-6, 88-9051X-5, 90-9252X-2, 88-9053X-5 ,42-44

Depth(mbsf)

5.366.94

39.0640.5596.5798.00

127.51164.30195.79230.65281.10308.10347.30378.93409.10447.43468.48486.30491.38505.12

SiO2

(%)

52.9648.5438.8656.1756.6447.1654.4259.7664.2764.8456.8367.8758.6762.28

63.5761.1964.7060.2460.67

TiO2

(%)

0.620.530.420.580.590.520.510.560.640.490.650.430.680.57

0.550.700.600.730.58

A12O3

(%)

14.2813.0011.4814.4214.2513.6312.2013.1013.9013.1815.4010.7116.3614.07

13.2815.9713.5115.6512.79

Fe2O3a

(%)

5.334.974.414.774.835.394.724.724.974.836.853.826.435.12

5.115.934.975.894.93

MnO(%)

0.0530.0520.0380.0460.0360.0570.0380.0410.0400.0420.0440.0260.0360.026

0.0210.0430.0360.0410.025

MgO(%)

2.302.331.811.761.701.961.701.851.701.612.131.121.971.63

1.462.071.822.041.76

CaO(%)

3.4910.6318.692.170.97

11.978.224.950.990.650.980.280.580.38

0.420.620.670.760.56

Na2O(%)

1.421.111.051.531.130.981.181.111.131.291.040.950.840.89

1.041.120.901.180.93

K2O(%)

2.532.331.582.522.342.302.112.312.662.653.022.022.962.36

2.252.962.392.842.40

P2O5

(%)

0.1420.1370.1360.1190.1160.1490.1430.1190.1010.1010.1150.0700.0890.088

0.0720.0830.0910.0820.113

LOI(%)

16.8916.3821.5415.9217.3815.8814.7611.489.62

10.2212.9312.7111.3812.5913.8712.239.31

10.3210.5515.24

' Total iron as Fe->CK.

720

GEOCHEMISTRY OF RARE EARTHS AND OTHER TRACE ELEMENTS

of elemental concentrations are shown in Figure 2 (for major andminor elements) and Figure 3 (for trace elements).

Site 799

Site 799 (Fig. 1) is located in the Kita-Yamato Trough (Hole799A: 39°2.046'N, 133°86.685'E; Hole 799B: 39°22.048'N,133°86.645'E). From Holes 799A and 799B, a 1084.0-m longsediment column (early Miocene to Holocene age) was recoveredbeneath 2073 m of water (Ingle, Suyehiro, von Breymann, et al.,1990). The sediment column was composed of five lithological

units: Unit I (0-170 mbsf, Quaternary to late Pliocene), Unit II(170-457 mbsf, late Pliocene to late Miocene), Unit III (457-800mbsf, late to middle Miocene), Unit IV (800-1020 mbsf, middle toearly Miocene), and Unit V (1020-1084 mbsf, early Miocene). Theaverage sedimentation rate was about 70 m/m.y., although the rateranged from 15 m/m.y. to 175 m/m.y (Ingle, Suyehiro, von Breymann,et al., 1990).

Determined concentrations of the analyzed elements are listed inTable 3 (for major and minor elements) and Table 4 (for trace elements).Depth profiles of elemental concentrations are shown in Figure 4 (formajor and minor elements) and Figure 5 (for trace elements).

2.5

1.5 -

10

1 F

1I

* G

r -r—•-——»-̂ ~ •ΦΦ

0.06100 200 300

Depth (mbsf)400 500 100 200 300

Depth (mbsf)4 0 0 5 0 0

Figure 2. Depth profiles of major and minor element contents (%) in the sediments from Site 798: A. SiO2. B. TiO2. C. A12O3. D. Fe2O3 (totaliron). E. MnO. F. MgO. G. CaO. H. Na2O. I. K2O. J. P2O5. Broken lines indicate boundaries between lithological units.

721


Table 2. Trace element concentrations in the sediments from Site 798.

Sampleno.

1234567S9

1011121314151617181920


798A-1H-4, 86-881H-5, 94-965H-2, 86-885H-3, 85-82

11H-2, 97-9911H-3, 90-92

798B-14H-4, 92-9418X-2, 90-9221X-4, 98-10025X-1,95-9730X-3, 90-9233X-2, 90-9237X-3, 10-1240X-4, 123-12543X-5, 90-9247X-6, 30-3249X-6, 88-9051X-5, 90-9252X-2, 88-9053X-5, 42-44

Depth(mbsf)

5.366.94

39.0640.5596.5798.00

127.51164.30195.79230.65281.10308.10347.30378.93409.10447.43468.48486.30491.38505.12

As(ppm)

6126

111364765

249

11149

17126

1111

Ba(ppm)

963917664805828799705830982825999

108310851252829903957870

23251032

Br(ppm)

90282374762246272025252712189

1976

1719

Co(ppm)

13.212.511.49.8

10.314.511.312.013.25.6

21.710.312.514.717.710.524.512.923.711.5

Cr(ppm)

65575259816160626247806684686049626083

100

Cs(ppm)

8.48.06.97.07.97.86.97.07.75.49.97.8

10.47.98.96.99.07.08.77.7

Hf(ppm)

3.53.12.94.13.13.13.03.23.63.03.34.53.53.13.76.34.23.93.83.2

Sb(ppm)

1.63.21.71.11.11.40.91.31.01.12.01.11.11.51.41.51.82.02.12.4

Se(ppm)

13.612.61111.112.312.512.211.312.88.5

14.412.215.613.214.711.916.813.717.916.0

Ta(ppm)

1.81.61.53.11.51.41.61.62.62.31.94.01.91.92.16.11.82.11.61.2

Th(ppm)

12.411.010.012.311.611.110.611.813.29.9

14.014.614.111.713.015.512.911.011.410.5

U(ppm)

8.57.74.7

13.714.79.27.44.44.35.47.5

25.16.08.57.15.55.54.66.79.1

La(ppm)

32.630.630.636.329.732.128.329.534.625.535.840.236.631.734.457.736.631.533.128.8

DISCUSSION

Major and Minor Element Geochemistry

Site 798

As plotted in Figure 2A, the SiO2 content gradually increases withsub-bottom depth up to 200 mbsf. Beneath about 200 mbsf, the SiO2

content is nearly constant at around 60%. At the upper sections of thecore (above about 200 mbsf), the SiO2 content in the two samples isvery low, because calcium carbonates develop in the samples asshown in the depth profile of CaO content (Fig. 2G). Calcium contentin the samples in the lower sections of the core (below about 200mbsf) is less than 1%. At these sections of the sediment columndilution with calcium carbonates does not lead to decrease in thecontents of other elements. Depth profiles of A12O3 and TiO2 contents(Fig. 2C and B) in the upper sections are identical to the depth profileof SiO2 content (Fig. 2A). Aluminum and titanium are regarded aselements derived from terrigenous detritus, and Si/Al ratio of terrig-enous silicate is around three (Leinen, 1977), which is nearly identicalto the ratios of the sediments from the portions of the sedimentcolumn. Such similarity among depth profiles of SiO2, A12O3, andTiO2 contents in the upper sections of the core suggests that silica willalso be supplied in the form of aluminosilicates derived from terrig-enous materials. Contents of these elements can vary with degree ofdilution by calcium carbonates, whereas the Si/Al ratio is preserved.However, in the lower sections (below about 120 mbsf), the depthprofiles of A12O3 and TiO2 contents are different from that of SiO2 content:SiO2 content (Fig. 2A) inversely changes against A12O3 (Fig. 2C) andTiO2 contents (Fig. 2B). Such a difference in depth profiles suggestsincorporation of biogenic silica into the sediments.

The depth profile of Fe2O3 content (Fig. 2D) below about 200 mbsfis also similar to the depth profiles of A12O3 and TiO2 contents, suggestingterrigenous supply of iron into the sediment column. However, in theupper sections variation of Fe2O3 content is much smaller than those ofSiO2, A12O3, and TiO2 contents (Fig. 2A, C, and B, respectively), whichimplies that some of the iron in the upper sections is contained in theiron-bearing minerals formed through hydrogenous precipitation anddiagenetic processes. Contents of Na2O, K2O, P2O5, and MgO (Fig. 2H,I, J, and F, respectively) in the lower column also correlate with A12O3

content. However, the depth profiles of contents of these elements in theupper sections differ from the depth profile of A12O3 content, suggestingdiagenesis and the contribution of other source materials. Content ofMnO, ranging from 0.02% to 0.06%, correlates poorly with A12O3 andTiO2 contents throughout the sediment column. Manganese seems to be

supplied in the form of hydrogenous precipitates, and may migrate tothe sediment column after burial.

Site 799

The content of SiO2 in the sediments from Site 799 (Fig. 4A) rangesfrom 40% to 80%, with several exceptions where calcium or manganesecarbonates are dominant, as seen in the depth profiles of CaO (Fig. 4G)or MnO (Fig. 4E) content. The depth profile of A12O3 content (Fig. 4C)agrees well with that of the TiO2 content (Fig. 4B) throughout the column.As has been mentioned, such agreement between the depth profiles wasalso seen in the sediment column from Site 798. Similar depth profilesof A12O3 and TiO2 contents imply that both elements were supplied asterrigenous detritus with a similar Al/Ti ratio. The depth profile of SiO2

content is different from those of A12O3 and TiO2 contents, especially inseveral samples diluted by carbonates. This suggests that siliceous bio-genics (e.g., diatoms) precipitated through sedimentation at a rate differ-ent from that for terrigenous detritus, and that diagenetic processes affectthe depth profile of SiO2 content. The depth profile of Fe2O3 content (Fig.4D) at the sections beyond 540 mbsf agrees well with depth profiles ofA12O3 and TiO2 contents (Fig. 4C and B, respectively). At the sectionsbelow 540 mbsf, the depth profile of Fe2O3 content does not correlatewith the profiles of A12O3 and TiO2 contents, suggesting the contributionof other materials bearing iron in the older sedimentation. In particular,the Fe2O3 content is substantially low at around 750-950 mbsf, whererhyolitic elastics are abundant, as shown in lithological descriptions ofthe sediment column (Ingle, Suyehiro, von Breymann, et al., 1990). Therelationship among A12O3, TiO2, and Fe2O3 contents restricts the chemicalcompositions of the materials supplied by rhyolitic volcanism. Perhapsthe Al/Ti ratios of the volcanic materials were identical to the ratios of theterrigenous detritus supplied throughout the sedimentation of the column.However, less iron was contained in the volcanic materials.

It is also remarkable in depth profiles of major and minor elementsthat the MnO content is extraordinarily high in a part of the column,as plotted in Fig. 4E. The manganese content is above 0.1 % in morethan half of the samples analyzed. Besides, the MnO content exceeds1% (up to 18%) in several samples. Such accumulation of manganeseat the particular positions of the sediment column is extensivelydiscussed by Matsumoto (this volume). Such high MnO content isused as a measure of diagenetic formation of manganese-rich carbon-ate concretions which were abundant in the sediment column (Ingle,Suyehiro, von Breymann, et al., 1990; Matsumoto, this volume). Noother elements show such a depth profile in the sediment column. Man-ganese geochemistry of modern sediments in the Japan Sea is influ-

722


Table 2 (continued).

Sampleno.

1234567S91011121314151617IS1920

Ce(ppm)

71645975657058647957799178708112488758574

Sm(ppm)

5.55.74.95.84.84.74.64.75.23.66.65.86.05.56.57.67.05.65.96.4

Eu(ppm)

1.31.21.11.11.11.21.01.11.41.41.41.21.31.21.31.11.51.21.41.4

Tb(ppm)

0.700.760.680.790.690.660.620.810.620.520.970.740.690.800.821.121.121.060.860.83

Yb(ppm)

2.22.21.82.31.92.11.91.91.91.62.52.02.22.02.32.82.62.32.62.8

Lu(ppm)

0.530.410.310.580.580.550.540.370.400.320.620.420.430.550.360.550.440.500.540.40

enced by oxidation-reduction conditions and water depth (Masuzawa,1987). The large differences in MnO content between Sites 798 and799 may be related to differences in water depth at sedimentation.

Iron was also partly contained in carbonate minerals (Matsumoto,this volume; Minai et al., this volume). However, the depth profileof Fe2O3 content agrees with those of A12O3 and TiO2 contents, atleast over 540 mbsf. This means that most of the iron is not containedin carbonates, in contrast to manganese. The correlation betweenFe2O3 and A12O3 contents is shown in Figure 6. Except for samples26, 41, 44, 53, and 57, Al/Fe ratio is almost constant throughout thecolumn. Samples 26 and 57, whose Fe2O3 contents are excessive incomparison with their A12O3 contents, contain large amounts ofcarbonate minerals (Matsumoto, this volume): rhodochrosite andmagnesite (sample 26), and dolomite (sample 57), suggesting thatsubstantial amounts of iron may be contained in carbonate mineralsin these samples. In contrast, Fe2O3 content in samples 41, 44, and53 is substantially less than what would be expected based on theirA12O3 content. Because A12O3 content correlates with TiO2 content,it is not likely that aluminum is selectively enriched in these samples.Perhaps terrigenous detritus or volcanic material with lower Fe2O3

content was supplied during the sedimentation of samples 41, 44,and 53.

Depth profiles of Na2O and K2O contents also agree with those ofA12O3 and TiO2 contents. Similar depth profiles of A12O3, TiO2, Na2O,and K2O contents suggest that most of the sodium and potassium in thesediments was also supplied from terrigenous sources throughout thesedimentation. Figure 7 shows the correlation between Na2O and K2Ocontents. In most of the samples, K2O/Na2O ratios are around 2, whereasthe ratios in samples 41,42,44, 50, and 53 are quite different from thosein the others. This may imply that another source material with a differentK/Na ratio contributed to sedimentation of these samples. However, wemust be very careful in utilizing the ratio to characterize the sourcematerials in the sediments because the K20/Na20 ratio can largelychange with diagenetic alteration of the volcanic materials. As has beenreported in the studies on low-temperature alteration of abyssal tholeiites(Hart, 1969; Hart and Staudigel, 1982), the ratio increases with degree ofalteration with accumulation of potassium into the secondary mineralsformed in low-temperature alteration. The Japan Sea sediments analyzedin this work contain relatively large amounts of volcanic materials. Suchdiagenetic processes may affect the variations in the K2O/Na2O ratios ofthe sediments. Depth profiles of P2O5 (Fig. 4J) and MgO contents (Fig.4F) are totally different from the depth profiles of the elements fromterrigenous detritus (e.g., aluminum). Magnesium content is particularlyhigh in some samples where carbonates are abundant.

Trace Element Geochemistry

The geochemistry of trace elements, except for REE, in thesediments from Site 798 and 799 is discussed in this section. Thedetermined trace elements (except for REE) are As, Ba, Br, Co, Cr,Cs, Hf, Sb, Se, Ta, Th, and U. The geochemistry of REE's will bediscussed in the next section in comparison with that of major, minor,and other trace elements.

Site 798

Figure 3 presents depth profiles of the elements discussed in thissection. It is sometimes difficult to interpret geochemical behaviorsof trace elements in sediment columns from such depth profilesbecause the depth profile tends to be affected by occasional dilution(e.g., with carbonates). Usually, the elemental ratio (or correlation) oftwo elements can provide more information for interpreting thegeochemistry of trace elements in sediments because such occasionalincorporation of materials does not influence the depth profile of theelemental ratio.

Figure 8 shows the depth profile of the Th/Sc ratio. The ratio showsmixing proportions between terrigenous detritus from continentalcrust (Th/Sc = 1, Taylor and McLennan, 1985) and andesitic-to-basicvolcanics (e.g., Th/Sc = 0.16 for andesite crust and Th/Sc < 0.01 foroceanic crust, Taylor and McLennan, 1985). As reported, the ratio hasbeen successively applied in turbidite sedimentation near continents(McLennan et al., 1990). As is seen in Figure 8, the ratio is almost 1through the column, with slight depletion in the sections below 400mbsf, implying that terrigenous material of Th/Sc = 1 was suppliedthroughout sedimentation, but volcanic material related to island-arcor basin-type volcanism was more active in the Pliocene than in theQuaternary. The ratio of surface sediments at the Japan Sea (Minai,1982) is about 1, indicating dominant contribution of continental crustin sedimentation of the modern Japan Sea. The constant Th/Sc ratiosuggests that similar sedimentation processes have occurred through-out the Quaternary and late Pliocene in the Japan Sea.

In Figure 9, correlations between Cr (ppm) and A12O3 (%) areshown with open circles. Although two samples (12 and 20) indicateenrichment of chromium relative to aluminum, correlation betweenthe two elements is fairly good, suggesting that chromium was alsosupplied from terrigenous sources to the sediment column.

Site 799

The depth profile of the Th/Sc ratio is plotted in Figure 10. The ratiogradually decreased with depth, from about 1 (Quaternary) to about0.6-0.7 (Miocene). Since the Th/Sc ratio reflects incorporation of andes-itic and basaltic materials, the depth profile of the Th/Sc ratio suggests agreater contribution of andesitic and basaltic materials in sedimentationat Kita-Yamato Trough from the Miocene to the Pliocene. However,several samples (samples 4, 15, 17, 41, 50, and 51) show much higherTh/Sc ratios than the ratio estimated from interpolation in the depthprofile. A high concentration of thorium relative to scandium concentra-tion may reflect rhyolitic volcanic activity in the sedimentation of thesamples. Some of the samples differ in chemical compositions from theother sediments: carbonate concretion is abundant in sample 4, and thechemical compositions of samples 41, 50, and 51 (Tables 3 and 4) arequite different from the adjacent sediments, as shown in the depth profiles(Figs. 4 and 5). However, no remarkable differences in chemical compo-sitions are seen for samples 15 and 17. Analysis of more samples fromthese sections may provide further information on variation of sedimen-tary history in those periods.

In Figure 9 correlations between Cr concentration (ppm) andA12O3 content (%) are shown with dots. A good correlation betweenconcentrations of the two elements suggests that terrigenous detritusagain is the major source of chromium in sedimentation for Site 799.The aluminum content is also correlated with Cs, Hf, Ta, and Th

723


10

3000

„ 2000

Im 1000

80

1m 40

100 200 300

Depth (mbsf)

400 500 100 200 300

Depth (mbsf)

400 500

Figure 3. Depth profiles of trace element concentrations (ppm) in the sediments from Site 798: A. As. B. Ba. C. Br. D. Co. E. Cr. F. Cs. G. Hf.H. Sb. I. Se. J. Ta. K. Th. L. U. M. La. N. Ce. O. Sm. P. Eu. Q. Tb. R. Yb. S. Lu. Broken lines indicate boundaries between lithological units.

concentrations. Such correlation also suggests association of theseelements with aluminosilicates in terrigenous detritus. However, insamples 10, 20, 41, 44, 50, and 53, chromium concentration relativeto A12O3 content is very different from the concentration in the othersamples (Fig. 9). Concentrations of cesium, hafnium, tantalum, andthorium in several samples also occasionally differ from those in allthe other samples, which implies that these elements were sometimesincorporated into the sediment without association with aluminosili-cates. It is well known that hafnium is present in zircon because ofsubstitution for the zirconium ion. Deposition of such accessoryminerals may lead to extraordinary abundances of these elements, asseen in their depth profiles (Fig. 5).

Concentrations of As, Ba, Br, Co, Sb, and U do not correlate wellwith A12O3 content. Concentration of barium increases with the depth(70-200 mbsf), and then decreases (200-580 mbsf) (Fig. 5B). Con-centrations of Br and Sb (Fig. 5C and H, respectively) are also high inthe range from 100 to 600 mbsf. Such variations of elemental concen-trations (especially barium concentration) with depth are interpretedas changes in sedimentation environment relating to change in waterdepth (von Breymann et al., 1990, and this volume). The depth profileof cobalt concentration (Fig. 5D) is essentially identical to those of Ba,Br, and Sb concentrations (Fig. 5B, C, and H, respectively). However,the Co concentration is also high in the 600-750 mbsf range, whereasBa, Br, and Sb concentrations are not substantially high. Perhaps cobalt

724


15

• . i

/ v

I . 1

V1.1

0.9

0.7

05

3

1 2

Q

A-7R

ü 80 -

200 300Depth (mbsf)

500

Figure 3 (continued).

was supplied to the sediments at that range by processes different fromthose supplying barium, bromine, and antimony.

REE Geochemistry

Figure 11 shows typical REE patterns (chondrite-normalized) ofthe Quaternary (Fig. 11 A) and Pliocene (Fig. HB) sediments fromSite 799. The chondritic values used are those determined by Masudaet al., (1973). The REE patterns indicate LREE enrichment relativeto HREE, implying that terrigenous detritus can be regarded as majorsources supplying REE's to the sediments.

Depth profiles of La/Yb, Ce/Ce*, and Eu/Eu* ratios are shown inFigures 12 (Site 798) and 13 (Site 799); Ce* and Eu* are cerium and

200 300

Depth (mbsf)

500

europium concentrations interpolated in a chondrite-normalizedREE pattern assuming no cerium and europium anomalies. Valuesof Ce* and Eu* are calculated from chondrite-normalized concen-trations of La, Sm, and Tb (LaN, SmN, TbN) and concentrations ofCe and Eu in chondrite (Ce(chondrite) and Eu(chondrite)) using thefollowing equations:

Ce* = Ce(chondrite) × (4 × LaN - SmN)/5,

Eu* = Eu(chondrite) × (2 × SmN - TbN)/3.

The La/Yb ratios in Figures 1 2 and 13 indicate degree of LREEfractionation from HREE since lanthanum and ytterbium represent

725


100

80

g 60

g 40

20

0

20

O 10

Ff

J11

>

200 400 600

Depth (mbsf)

800 1000 200 400 600

Depth (mbsf)

800 1000

Figure 4. Depth profiles of major and minor element contents (%) in the sediments from Site 799: A. SiO2. B. TiO2. C. A12O3. D. Fe2O3 (totaliron). E. MnO. F. MgO. G. CaO. H. Na2O. I. K2O. J. P2O5. Broken lines indicate boundaries between lithological units.

LREE and HREE, respectively. The Ce/Ce* and Eu/Eu* ratios (alsoplotted in Figures 12 and 13) indicate selective enrichment or deple-tion of cerium and europium from the other REE's. We can use theCe/Ce* and Eu/Eu* ratios to represent degrees of anomaly of ceriumand europium concentrations relative to concentrations of theother REE's.

Most Ce/Ce* ratios of the sediments from Site 798 (Fig. 12B) areless than 1.15, while they range from 1.25 to 1.7 in most of thesediments in the 200-750 mbsf range from Site 799 (Fig. 13B).Cerium anomalies have been reported in marine-derived substances:mud, carbonaceous ooze, hydrothermal metalliferous precipitate,seawater, fish debris, chert, and manganese nodule (e.g., Goldbergand Arrhenius, 1958; Piper, 1974; Murray et al., 1990; Shimizu and

Masuda, 1977; Toyoda et al., 1990; Matsumoto et al., 1988; Minai etal., 1985; Minai and Tominaga, 1989). In contrast, for continentalcrust and volcanic elastics on land, cerium anomalies have scarcelybeen reported (Henderson, 1984). Cerium tends to be oxidized totetravalent state from trivalent state, which is the ordinary oxida-tion state of REE's, in marine environments. Oxidation of cerium inmarine environments leads to behaviors of cerium that differ fromthose of the other trivalent REE's. However, europium, which is alsounique for the stability of its divalent state, is trivalent, like the otherREE's in normal marine environments (Henderson, 1984). It isnotable that manganese nodules (e.g., Piper, 1974) and settling parti-cles (Masuzawa and Koyama, 1989) tend to accumulate more ceriumthan the other trivalent REE's, suggesting association of cerium with

726


Table 3. Major and minor element contents in the sediments from Site 799.

Sampleno.

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758


799A-3H-4, 41-436H-2, 44-469H-2, 36-389H-2, 133-135

12H-6,43^515H-2, 39-4118H-5, 40-4222X-1,91-9324X-2, 40-4225X-2, 4 1 ^ 327X-3, 126-12830X-2, 40-4233X4, 40-4236X-2, 4 1 ^ 336X-5,41^338X4, 38-4038X-6, 39-4042X-4, 40-4245X4, 40-4249X-1,40-42

799B-6R-1,40-427R4, 32-34

12R-1, 125-12712R-3, 40-4215R-1,40-4217R-1,93-9518R-2, 40-4120R-7, 30-3224R-3,114-11627R-1,40-4229R-3, 50-5229R4, 50-5229R-7, 50-5230R-1,49-5234R-2, 40-42

b34R-2, 44-50c34R-2, 44-5035R-1, 1-335R-1, 107-113

d35R-l, 107-11341R-1, 40-4246R-3, 50-5247R-3, 52-5449R-2, 19-2151R-1, 39-41

e53R-1,46-48f53R-1,46-4856R-1, 36-3856R-3, 49-5157R-1,40-4259R-1,38-4060R-3, 45-4761R-2, 49-5261R-5, 100-102

g61R-6, 52-55h61R-6, 52-5562R-3, 15-1763R-3, 18-20

Depth(mbsf)

15.6141.4470.1671.13

105.63127.99161.6194.71215.00224.71245.66262.40292.80318.81323.31341.08344.09379.90408.80443.00495.00504.52549.15551.30577.40597.23607.80634.50667.94692.90715.30716.80721.30721.89761.70761.74761.74769.51770.57770.57827.50878.80888.52905.89923.69943.06943.06971.86974.99981.50

1000.881013.651021.791026.801027.821027.821032.651042.28

SiO2

(%)

55.6854.9141.9621.2848.2267.3767.1651.4259.4959.3548.5263.5364.3369.9573.1271.7463.9851.1664.2654.2656.3756.5959.1162.1659.7013.07

56.5565.6356.3766.3067.1767.52

75.5679.78

82.1914.2368.1579.7273.8167.13

75.577.61

58.6465.71

69.0870.7172.9571.3369.11

9.9063.06

TiO2

(%)

0.720.630.440.100.480.440.480.510.530.580.210.420.290.150.250.310.390.430.430.650.480.430.390.450.430.04

0.310.420.560.480.410.41

0.240.21

0.200.010.450.240.350.36

0.290.28

0.540.23

0.470.360.350.320.120.050.43

A12O3

(%)

15.2614.6110.276.02

12.3211.1210.6212.1812.7013.536.939.998.386.318.118.50

11.1310.1010.5914.7412.3711.0310.2211.3611.803.78

9.4911.1913.6511.5910.7810.28

7.557.23

4.943.45

12.247.689.25

12.73

8.468.22

12.4211.84

10.4012.429.178.686.481.61

10.59

Fe2O3a

(%)

6.706.434.912.515.394.224.555.025.405.772.244.513.051.832.883.384.934.604.446.225.514.574.914.835.536.20

5.454.275.514.373.813.61

2.621.96

2.150.861.581.962.852.90

2.662.33

6.243.87

4.142.802.873.031.502.634.33

MnO(%)

0.080.080.040.071.240.660.225.801.590.40

13.190.953.172.200.340.130.246.170.661.880.172.191.300.150.30

18.03

6.980.741.680.620.950.26

1.040.16

0.102.410.040.040.240.11

0.730.26

0.291.33

0.070.160.331.174.124.461.30

MgO(%)

2.902.671.990.062.843.112.633.503.453.692.653.562.441.271.512.032.924.343.253.082.562.692.892.402.004.63

2.633.503.932.792.632.15

0.960.77

0.7816.103.291.011.722.64

1.651.34

2.241.61

1.681.401.481.511.05

15.842.13

CaO(%)

3.104.37

17.4635.73

9.110.580.433.381.250.635.640.551.282.110.370.220.313.070.571.221.102.501.200.400.52

10.44

3.020.611.100.610.700.39

0.630.29

0.2522.72

0.310.240.360.69

0.840.43

0.800.90

0.511.040.631.255.93

25.962.22

Na2O(%)

1.501.811.190.761.181.241.391.301.361.070.651.251.160.901.230.741.521.091.131.181.091.040.830.840.880.10

0.770.980.970.900.910.89

0.590.63

0.840.081.360.335.471.73

0.740.68

1.480.73

1.082.321.000.860.500.181.10

K2O(%)

3.153.212.080.962.532.232.412.602.892.971.092.161.590.901.771.682.562.222.233.142.432.182.022.262.170.19

1.852.512.982.452.152.02

1.050.96

0.960.060.241.371.951.84

1.491.46

3.633.48

1.902.611.521.430.620.311.81

P2O5

(%)

0.1260.1170.1460.0770.2660.0750.0740.1430.1010.0890.1480.0720.0780.0680.0540.0530.0540.1260.0720.1230.0920.1610.0840.0960.1020.226

0.1440.0820.2000.1020.1010.098

0.0800.074

0.0750.1010.0610.0820.1190.136

0.0930.092

0.1770.100

0.1810.1310.1170.1800.1380.0700.207

LOI(%)

10.8011.1719.2031.9116.448.96

10.0314.1411.2411.9318.7413.0214.2314.3310.3911.2111.9616.6912.3613.5017.8316.6117.0615.0316.5643.30

12.8110.0813.039.81

10.4012.38

9.687.95

7.539.9712.277.358.809.71

7.507.31

13.5410.20

10.516.059.57

10.2210.4538.9812.82

aTotalironasFe2O3.Darker portion.

c Lighter portion.d Dolomite concretion.e Top of the interval.f Bottom of the interval.g Darker portion.h Lighter portion.

hydrogenous manganese precipitation. Besides, in a seawater columnit has been reported (Elderfield and Sholkovitz, 1987; German andElderfield, 1989,1990) that the profile of cerium is similar to that ofmanganese because the oxidation-reduction potential of manganese(IV) is identical to the potential of cerium (IV) oxides (Leinen, 1977).It has been recognized that manganese in marine sediments is mainlysupplied by hydrogenous precipitation (e.g., Matsumoto et al., 1985).

The association of cerium with manganese in marine environmentsand the similarity of these elements in oxidation-reduction chemistrysuggest that the positive cerium anomaly observed in the REE pat-terns of the samples at Site 799 indicates a contribution of hydroge-nous manganese precipitation to the sediment. However, MnOcontent in the sediments does not correlate well with positive ceriumanomaly, as shown in Figure 14. It is notable that manganese

727


Table 4. Trace element concentrations in the sediments from Site 799.

Sampleno.

1234567891011121314151617181920212223242526272S293031323334353637383940414243444546474H49505152535455565758


799A-3H-4,41-«6H-2, 44-169H-2, 36-389H-2, 133-13512H-6, 43^515H-2, 39-4118H-5,40-4222X-1, 91-9324X-2, 40^225X-2, 41-4327X-3, 126-12830X-2, 40-4233X-4, 40-4236X-2,41^*336X-5, 41-4-338X-4, 38-1038X-6, 39^042X-4, 40-4245X-4, 40-4249X-1,40-42799B-6R-1,40-42

7R-4, 32-3412R-1, 125-12712R-3, 40-4215R-1,40-4217R-1,93-9518R-2, 40-4120R-7, 30-3224R-3,114-11627R-1,40-4229R-3, 50-5229R-4, 50-5229R-7, 50-5230R-1,49-5234R-2, 40-42a34R-2,44-50b34R_2, 44-5035R-1, 1-335R-1, 107-113C35R-1, 107-11341R-1,40-4246R-3, 50-5247R-3, 52-5449R-2, 19-2151R-1, 39-41d53R-1,46-48e53R-l, 46-4856R-1,36-3856R-3,49-5157R-1,40-4259R-1,38-4060R-3, 45-4761R-2, 49-5261R-5, 100-102f61R-6, 52-55861R-6, 52-5562R-3, 15-1763R-3,18-20

Depth(mbsf)

15.6141.4470.1671.13105.63127.99161.60194.71215.00224.71245.66262.40292.80318.81323.31341.08344.09379.90408.80443.00495.00504.52549.15551.30577.40597.23607.80634.50667.94692.90715.30716.80721.30721.89761.70761.74761.74769.51770.57770.57827.50878.80888.52905.89923.69943.06943.06971.86974.99981.501000.881013.651021.791026.801027.821027.821032.651042.28

As(ppm)

14510322144544224614162379753734777438716

45424545135439446317

Ba(ppm)

750840871540129019403100325040003728313033003310122011841721299023402140237024301960120013801040910116069013109501076285011201420930939969194

1072

6006805906505106206405206441664470393550270353412268447

Br(ppm)

2440321940265259414126634439172831473448683624493510211357111088899412

186443449417356427

Co(ppm)

2221268294439506424346049332422414236416039644488194555374122262824192510314

361911924172223103129212013539

Cr(ppm)

6256491156634451512722402818243441444544765245746584824395054403635273128527

4283722312726346017114826393718851

Cs(ppm)

7.83.55.31.56.68.14.86.36.83.42.54.93.61.93.847.95.45.65.36.35.55.25.95.30.85.13.95.575.54.85.24.222.21.80.31.8

1.32.73.62.93.62.92.63.15.12.92.63.92.53.331.20.74

Hf(ppm)

4.45.53.72.42.53.92.73.23.21.71.42.51.91.16.21.84.62.62.42.23.12.22.22.72.60.62.52.23.23.232.72.62.41.41.71.50.21.6

5.41.42.15.31.61.91.72.332.74.62.842.32.210.52.9

Sb(ppm)

2.52.44.01.53.54.85.08.98.73.23.79.94.33.62.83.94.28.63.95.19.16.59.37.212.91.55.61.83.15.22.42.82.32.22.12.31.80.31.9

0.923.12.13.13.63.33.36.93.50.821.42.32.71.50.63.5

Se(ppm)

1514104131512141366129578121212121511111313513714171414131488716

1071011976913691081086212

Ta(ppm)

1.52.41.91.11.01.40.81.00.80.80.60.80.60.44.20.95.70.90.90.90.80.90.61.21.30.21.11.11.21.410.810.90.50.40.50.10.5

1.10.60.710.60.50.70.711.12.20.90.80.70.60.20.10.8

Th(ppm)

1414108121410121255964126209991110991021071013999754414

154676656911257576317

U(ppm)

4.73.86.25.62.31.72.33.15.35.46.64.72.84.45.94.26.84.62.15.04.53.05.57.07.94.45.43.73.85.35.47.46.923.2

9.08.45.21.75.2

5.93.86.54.93.94.23.53.34.24.96.47.04.95.43.84.30.83.5

La(ppm)

39433528323426363536202420133418272925263625293634112722293726232328182412418

34141920171514282427502217182012529

a Darker portion.b Lighter portion.c Dolomite concretion.

Top of the interval.e Bottom of the intervaf Darker portion.g Lighter portion.

carbonates occurred at the ODP sites in the Japan Sea (Matsumoto, thisvolume). Manganese oxide precipitation and diagenetic transforma-tion in the sediment column are regarded as important processes in theformation of manganese carbonate concretions (Matsumoto, this vol-ume). Manganese content is primarily dependent on the amounts ofmanganese carbonates in the sediments from Site 799. In the processof formation of manganese carbonates in the sediment column, man-

ganese (IV) is reduced to manganese (II), which is more mobile thanmanganese (IV) because of its higher solubility into pore water.Tetravalent cerium can be reduced to the lower oxidation state, trivalentcerium, during reduction of manganese (IV) in the diagenetic proc-esses. Differences in the oxidation state after reduction (manganese (II)and cerium (III)) can lead to fractionation of cerium from manganeseafter burial and the formation of manganese-bearing carbonates with-

728


Table 4 (continued).

Sampleno.

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

Ce(ppm)

8195666671838190913748735740724375757380627580821083697659112171746885496628741

7827464741393746545511447394946271064

Sm(ppm)

7.27.56.14.86.66.64.98.07.26.44.85.75.12.96.03.65.25.95.65.58.15.16.57.07.82.86.54.86.38.95.45.85.75.64.86.92.60.94.3

6.92.74.64.63.83.53.24.04.46.510.04.73.84.34.02.60.95.6

Eu(ppm)

1.41.21.10.81.41.31.01.71.40.61.01.21.00.60.60.70.91.21.11.11.61.11.31.31.50.61.90.91.61.71.01.11.01.41.41.40.50.20.8

1.10.50.91.31.00.60.61.20.90.70.80.91.00.80.80.50.21.1

Tb(ppm)

0.950.780.820.801.000.930.691.051.260.431.120.800.850.650.900.500.960.880.620.590.870.791.101.041.020.651.120.890.991.220.800.980.600.940.761.090.370.130.61

1.110.330.740.870.530.700.440.760.571.041.980.700.630.870.570.370.150.70

Yb(ppm)

3.23.02.62.33.23.51.94.03.81.14.12.53.72.23.11.13.94.02.92.34.02.93.92.02.72.43.22.42.73.72.32.42.03.91.72.71.00.51.7

2.81.22.02.91.91.31.31.92.43.55.32.41.82.11.91.40.42.6

Lu(ppm)

0.550.410.380.340.420.500.280.570.540.390.710.430.550.300.600.270.540.500.400.310.500.340.570.590.500.470.520.450.450.490.430.460.420.710.380.740.240.140.42

0.440.310.400.450.310.270.260.310.670.950.800.530.270.340.390.260.130.60

out cerium enrichment. The possible reduction of cerium to the triva-lent state implies that cerium behaves as an ordinary trivalent REE inthe sediment column after reduction in diagenesis. Hence, it can beexpected that the degree of the cerium anomaly did not change afterburial, although the MnO content changed in diagenetic processes afterreduction. Also, it is noteworthy that the degree of positive ceriumanomaly could be a measure for estimating the amount of hydrogenousprecipitation correlating to manganese influx at sedimentation even ifthe MnO content changed in diagenesis after burial.

The depth profile of the Ce/Ce* ratio in sediments from Site 799(Fig. 13B) shows that degree of positive cerium anomaly is relativelyhigh in the 200-750 mbsf range. We can expect that the rate ofhydrogenous precipitation of manganese is high in that range. Dur-ing the sedimentation, terrigenous detritus (without cerium anom-

aly) and hydrogenous precipitates of manganese (associated withcerium-enriched REE's) would be deposited on the ocean bottom.However, it is noteworthy that some of the samples (e.g., samples10 and 21) show a negative cerium anomaly (Ce/Ce* ratio below 1).Negative cerium anomaly has been reported for seawater (e.g.,Goldberg and Arrhenius, 1958; Elderfield and Sholkovitz, 1987:German and Elderfield, 1989, 1990) and for mud, carbonaceousooze, chert, and fish debris in oceanic environments (e.g., Piper,1974; Shimizu and Masuda, 1977; Toyoda et al., 1990; Matsumotoet al., 1988; Minai and Tominaga, 1989). Perhaps REE's in seawaterwith a negative cerium anomaly were transferred to such sedimentsas they were formed, through adsorption on biogenic materials orhydrogenous precipitation (except for manganese hydroxides carry-ing REE's with positive cerium anomaly).

It has been claimed that the negative cerium anomaly is a measureof hydrothermal activity, based on comparison of manganese con-tents, sedimentation rates, and REE patterns of marine sediments(Matsumoto and Minai, 1985; Matsumoto et al., 1985; Minai et al.,1985). At Site 799, it has been reported that the MnO content wasstrongly changed by diagenetic processes after burial (Matsumoto,this volume). Since we cannot compare the degree of cerium anomalywith MnO content in the sediment before diagenesis, the negativecerium anomaly observed in some samples cannot be a decisive clueto hydrothermal activity in the Japan Sea.

From the REE patterns and the Ce/Ce* ratio, we can assume atleast three sources supplying REE's in the sediment column. Terrig-enous detritus can be regarded as the major source of REE's in thesediments because the REE patterns of all samples are similar to thoseof continental crust (showing relative enrichment of LREE to HREE).Hydrogenous precipitates of manganese with positive cerium anom-aly can also be regarded as one of the sources because of the positivecerium anomaly frequently observed in the REE patterns of thesediments. Biogenic or hydrogenous precipitates scavenging REE'sin seawater with negative cerium anomaly may be the third source ofREE's in some sediments at Site 799. It seems that the cerium anomalycould decipher mixing ratios among the materials derived from thesesources into the sediments.

Depth profiles of La/Yb and Eu/Eu* ratios in Figures 12 and 13may indicate complicated sedimentation processes at Sites 798 and799. As shown in Fig. 15, the La/Yb (Fig. 15A) and Eu/Eu* (Fig.15B) ratios are poorly correlated with the Ce/Ce* ratio, which isregarded as an indicator of the mixing proportion of hydrogenousprecipitates in the sedimentation. The poor correlation of the La/Yband Eu/Eu* ratios with the Ce/Ce* ratio suggests that these ratioscannot be regarded as indicators of hydrogenous precipitation. Itimplies that variations of the La/Yb and Eu/Eu* ratios in the sedi-ment columns are related to changes of diagenetic processes in thecolumn or mixing ratios among terrigenous sources with differentLa/Yb and Eu/Eu* ratios.

In Figure 16, La/Yb ratio is plotted against MnO/Al2O3 ratio. Inthis plot, La/Yb ratio tends to decrease with the increase of theMnO/Al2O3 ratio. Usually, MnO/Al2O3 ratios are regarded as indica-tors of hydrogenous contribution to sedimentation since manganeseand aluminum are supplied as hydrogenous precipitate and terrig-enous detritus, respectively. However, in the sediments from Site799, manganese was accumulated after burial at particular layersforming manganese carbonates (Matsumoto, in this volume),whereas aluminum was immobile in the sediment column. There-fore the MnO/Al2O3 ratios in the sediments also indicate manga-nese accumulation by diagenesis.

In Figure 17 dependencies of La and Yb concentrations on A12O3

content are plotted. Lanthanum and ytterbium represent LREE andHREE, respectively. Concentration of La (Fig. 17A) correlates withA12O3 content, as do TiO2, Na2O, K2O, Cr, Cs, Hf, Ta, and Th contents.However, as seen in Figure 17B, the Yb concentrations, particularlyin the sediments from Site 799, are poorly correlated with the A12O3

content. As mentioned earlier, terrigenous materials were regarded as

729


F

5000

4000

3000

2000

1000

0

80

60

40

20

0

B

C v\ I t

r

-?- 10 •

, G

H

Ai

100

80

60

40

20

0

D

I

{

20

15

I 10o

5

0

I

If r

200 400 600

Depth (mbsf)

800 1000

J 1IL

200 400 600

Depth (mbsf)

800 1000

Figure 5. Depth profiles of trace element concentrations (ppm) in the sediments from Site 799: A. As. B. Ba. C. Br. D. Co. E. Cr. F. Cs. G. Hf.H. Sb. I. Se. J. Ta. K. Th. L. U. M. La. N. Ce. O. Sm. P. Eu. Q. Tb. R. Yb. S. Lu. Broken lines indicate boundaries between lithological units.

a major source of REE in the sediments. Such differences in correla-tion plot against A12O3 content between La and Yb concentrationsmay suggest that source materials with higher HREE concentrationwere also supplied to the sediment column. Otherwise, differencesduring diagenesis in geochemical behaviors of LREE and HREE maylead to a discrepancy in the correlation plots against the A12O3 content.

It is well recognized that complexation of REE with organicand inorganic ligands depends on the ion radius of REE (Cantrelland Byrne, 1987; Byrne and Kim, 1990). Although dependence ofthe binding strength of naturally-occurring organic materials onthe ion radii is not well understood, HREE with its smaller ionradii tends to bind more strongly with simple organic monocar-

boxylates as analogues of the natural organic substances thanLREE with its larger ionic radius (Byrne and Kim, 1990). Poten-tially, complexation with organic substances in the sediments isimportant in assisting in the dissolution and migration of REE in thesediment column. The stronger binding of HREE than that of LREEwith organic materials may lead to higher mobility of HREE accom-panied by diagenetic accumulation of manganese, as observed inthe sediments from Site 799.

The behavior of REE in diagenesis has been extensively examinedin the sediments from Leg 127 (Murray et al., this volume). Althoughsedimentation environments differed among the sites from Legs 127and 128, work on REE geochemistry in the sediments from the ODP

730


30

2 0 •

~ 20 -

10 "

20 -

K

L

M

Λ

TV r

ii

ft

120

80

40

0

12

N

J1 . 1

4 -

o

V•Vfk•A200 400 600

Depth (mbsf)

800 1000

1.0

0.8

0.6

0.4

0.2

0.0

s

I200 400 600

Depth (mbsf)

800 1000

Figure 5 (continued).

sites suggests that diagenetic processes are extremely important inREE geochemistry of marine sediments. Figures 12C and 13C showthe depth profiles of the Eu/Eu* ratios in the sediments from Sites 798and 799, respectively. It has been reported that Eu/Eu* ratios incrust-derived materials range from 0.6 to 1 depending on formationage of the crust (e.g., McLennan et al., 1990). Most of the Eu/Eu*ratios in the sediments from Sites 798 and 799 are in the range of theEu/Eu* ratio in crust-derived materials. Terrigenous detritus derivedfrom the continental crust on the Eurasia Plate and the younger crustof the island-arc (partly having a continental-type crust) have beendeposited to form the sediments of the Japan Sea. Variation of theEu/Eu* ratio may be interpreted as variation of mixing ratios amongvarious types of terrigenous detritus around the Japan Sea through thegeological periods. It should be mentioned that the Eu/Eu* ratio insome samples is lower than 0.6. The Eu/Eu* ratio in sample 51,

containing rhyolitic volcanic ash, is the lowest among the samplesanalyzed. Plagioclase fractionation in the formation of rhyoliticmagma can lead to selective depletion of europium (negativeeuropium anomaly). Extreme depletion of europium in some sedi-ments may indicate sedimentation of rhyolitic volcanic ash. Perhapsthe depth profile of the Eu/Eu* ratio traces changes in rhyoliticvolcanic activity and mixing proportions among multiple terrigenoussources in the sediment column.

CONCLUSIONS

We conclude the following. (1) A positive cerium anomaly maybe a measure of hydrogenous precipitation, although it does notcorrelate with MnO content, which is sometimes regarded as a meas-ure of hydrogenous precipitation. (2) Correlation between La/Yb and

731


oo

Figure 6. Correlation between Fe2O3 (total iron) and A12O3 contents (%) in thesediments from Sites 798 (circles) and 799 (dots).

<M

Figure 7. Correlation between Na2O and K2O contents (%) in the sedimentsfrom Sites 798 (circles) and 799 (dots).

MnO/Al2O3 ratios, together with the correlations of La and Ybconcentrations to A12O3 content, may suggest migration of HREEthrough the sediment column. It also shows that diagenesis is impor-tant in the REE geochemistry of marine sediments. (3) Depth profilingof the europium anomaly may indicate variations of rhyolitic volcanicactivities and terrigenous sources throughout the sedimentation. (4)The depth profile of the Th/Sc ratio is an indicator of variation in basicvolcanic activities in the sedimentation. (5) REE pattern and correla-tion of REE's with other elements suggest that terrigenous detritus isa major source of REE in the sediments. (6) Several major and traceelements (e.g., chromium) are concluded to mainly derive from terrig-enous materials.

ACKNOWLEDGMENTS

The authors thank Prof. H. Wakita of the University of Tokyo forthe use of his laboratory for INAA. The authors also thank Ms. I.Kusudo of Science University of Tokyo for her devoted contributionto sample preparation. This research was partly supported by theInter-University Joint Research Program for Use of the Rikkyo Re-search Reactor.

REFERENCES

Ando, A., Kamioka, H., Terashima, S., and Itoh, S., 1989.1988 values for GSJrock reference samples, "Igneous rock series." Geochem. J., 23:143-148.

Bender, M., Broecker, W., Gornitz, V., Middel, U., Kay, R., Sun, S.-S., andBiscaye, P., 1971. Geochemistry of three cores from the East Pacific Rise.Earth Planet. Sci. Lett., 12:425^33.

Byrne, R. H., and Kim, K.-H., 1990. Rare earth element scavenging inseawater. Geochim. Cosmochim. Acta, 54:2645-2656.

Cantrell, K. J., and Byrne, R. H., 1987. Rare earth element complexation bycarbonate and oxalate ions. Geochim. Cosmochim. Acta, 51:597-605.

Ebihara, M., Minai, Y, Kubo, M. K., Tominaga, T., Aoki, N., Nikko, T.,Sakamoto, K., and Ando, A., 1985. Revaluation of lanthanide abundancesin Japanese standard rock samples, JG-1 and JB-1. Anal. Sci., 1:209-213.

Elderfield, H., and Sholkovitz, E. R., 1987. Rare earth elements in the pore watersof reducing nearshore sediments. Earth Planet. Sci. Lett., 82:280-288.

German, C. R., and Elderfield, H., 1989. Rare earth elements in Saanich Inlet,British Columbia, a seasonally anoxic basin. Geochim. Cosmochim. Acta,53:2561-2571.

, 1990. Rare earth elements in the NW Indian Ocean. Geochim.Cosmochim. Acta, 54:1929-1940.

Goldberg, E. D., and Arrhenius, G.O.S., 1958. Geochemistry of Pacific pelagicsediments. Geochim. Cosmochim. Acta, 13:153-212.

Hart, S. R., 1969. K, Rb, Cs contents and K/Rb, K/Cs ratios of fresh and alteredsubmarine basalts. Earth Planet. Sci. Lett., 6:295-303.

Hart, S. R., and Staudigel, H., 1982. The control of alkalies and uranium inseawater by ocean crust alteration. Earth Planet. Sci. Lett., 58:202-212.

Henderson, H. (Ed.), 1984. Rare Earth Element Geochemistry: Amster-dam (Elsevier).

Ingle, J. C , Jr., Suyehiro, K., von Breymann, M. T, et al., 1990. Proc. ODP,Init. Repts., 128: College Station, TX (Ocean Drilling Program).

Karig, D. E., Ingle, J. C , Jr., et al., 1975. Init. Repts. DSDP, 31: Washington(U.S. Govt. Printing Office).

Leinen, M., 1977. A normative calculation technique for determining opal indeep-sea sediments. Geochim. Cosmochim. Acta, 41:671-676.

Masuda, A., Nakamura, N., and Tanaka, T, 1973. Fine structures of mutuallynormalized rare-earth patterns of chondrites. Geochim. Cosmochim. Acta,37:239-248.

Masuzawa, T., 1987. Early diagenesis in deep-sea sediments of the Japan Sea:type, controlling factor, and diffusion flux. /. Earth Sci., Nagoya Univ.,35:249-267.

Masuzawa, T, and Koyama, M., 1989. Settling particles with positive Ceanomalies from the Japan Sea. Geophys. Res. Lett., 16:503-506.

Matsumoto, R., and Minai, Y, 1985. Genetical classification of deep seasediments on the basis of Mn content and REE pattern. J. Sediment. Soc.Jpn., 21:1-9.

Matsumoto, R., Minai, Y, and Iijima, A., 1985. Manganese content, ceriumanomaly, and rate of sedimentation as aids in the characterization andclassification of deep-sea sediments. In Nasu, N., Kobayashi, K., Uyeda,S., Kushiro, I., and Kagami, H. (Eds.) Formation of Active Ocean Margins:Tokyo (Terra Scientific), 913-939.

Matsumoto, R., Minai, Y, and Okamura, M., 1988. Geochemistry anddeposition environments of bedded chert of the Cretaceous Shimantogroup, Shikoku, southwest Japan. Mod. Geol., 12:197-224.

McLennan, S. M., Taylor, S. R., McCulloch, M. T, and Maynard, J. B.,1990. Geochemical and Nd-Sr isotopic composition of deep-sea tur-bidites: crustal evolution and plate tectonic associations. Geochim.Cosmochim. Acta, 54:2015-2050.

Minai, Y, 1982. Geochemical studies of the ocean floor rocks and sediments[Doctoral dissert.]. Univ. of Tokyo.

Minai, Y, Matsumoto, R., and Tominaga, T, 1985. Geochemistry of thedeep sea sediments from the Nankai Trough, the Japan Trench, andadjacent regions. In Kagami, H., Karig, D. E., and Coulbourn, W. C , etal., Init. Repts. DSDP, 87: Washington (U.S. Govt. Printing Office),643-657.

Minai, Y, and Tominaga, T, 1989. Neutron activation analysis of rare earthelements in deep sea sediments from the Pacific Ocean and the Japan Sea.J. Radioanal. Nucl. Chem., 137:351-363.

Murray, R. W, Buchholtz ten Brink, M. R., Jones, D. L., Gerlach, D. C , andRuss, G. P., Ill, 1990. Rare earth elements as indicators of different marinedepositional environments in chert and shale. Geology, 18:268-271.

Piper, D., 1974. Rare earth elements in the sedimentary cycle: a summary.Chem. Geol., 19:285-304.

732


Shimizu, H., and Masuda, A., 1977. Cerium in chert as an indicator of marineenvironment of its formation. Nature, 266:346-348.

Tamaki, K., 1988. Geological structure of the Japan Sea and its implications.Chishitsu Chosasho Geppo, 39:269-365.

Taylor, S. R., and McLennan, S. M., 1985. The Continental Crust: Its Compositionand Evolution: Oxford (Blackwell).

Toyoda, K., and Masuda, A., 1990. Sedimentary environments and chemicalcomposition of Pacific pelagic sediments. Chem. Geol, 88:127-141.

Toyoda, K., Nakamura, Y., and Masuda, A., 1990. Rare earth elements ofPacific pelagic sediments. Geochim. Cosmochim. Acta, 52:1557-1569.

von Breymann, M. T, Emeis, K.-C, andCamerlenghi, A., 1990. Geochemistryof sediments from the Peru upwelling area: results from Sites 680, 682,685, and 688. In Suess, E., von Huene, R., et al., Proc. ODP, Sci. Results,112: College Station, TX (Ocean Drilling Program), 491-503.

Date of initial receipt: 18 March 1991Date of acceptance: 11 September 1991Ms 127/128B-180

2 -oen

100 200 300

Depth (mbsf)

400 500

Figure 8. Depth profile of Th/Sc ratio of the sediments from Site 798. Broken lines indicateboundaries between lithological units.

120

o

Figure 9. Correlation between Cr concentration (ppm) and A12O3 content (%)in the sediments from Sites 798 (circles) and 799 (dots).

733


üCO

200 400 600

Depth (mbsf)

800 1000

Figure 10. Depth profile of Th/Sc ratio in the sediments from Site 799. Broken lines indicateboundaries between lithological units.

La Ce SmEu 7b YbLu

Figure 11. Typical REE patterns (chondrite-normalized) of the sediments fromSite 799: A. Quaternary sediment samples 1 (dot), 2 (circle), and 5 (invertedtriangle and solid square). B. Pliocene sediment samples 13 (dot), 14 (circle),and 16 (solid square).

734


S

100 200 300

Depth (mbsf)

400 500

Figure 12. Depth profiles of La/Yb (A), Ce/Ce* (B), and Eu/Eu* (C) ratios of the sediments fromSite 798. Broken lines indicate boundaries between lithological units.

735


40

30

JQ

| 20

10 F

0

2.0

<Dü

ü

LU

LU

1.5 -

1.0 -

0.5

1.0

0.8 -

0.6 ^

0.4 -

0.2

. A J

B

t

. i ,

. c

I i

Λr200 400 600

Depth (mbsf)

800 1000

Figure 13. Depth profiles of La/Yb (A), Ce/Ce* (B), and Eu/Eu* (C) ratios of the sediments from

Site 799. Broken lines indicate boundaries between lithological units.

736


oO

0.01 0.1 1 10 100MnO (%)

Figure 14. Dependence of Ce/Ce* ratio on MnO content (%) in the sedimentsfrom Sites 798 (circles) and 799 (dots).

0.001 0.01Mn0/AloC>

C. O

Figure 16. Dependence of La/Yb ratio on MnO/Al2O3 ratio in the sedimentsfrom Sites 798 (circles) and 799 (dots).

30

20

10

n

»

-

-

- % ***%*i

# * '•• . i .

A

•

2.0

3LU

LU

Figure 15. Dependence of La/Yb (A) and Eu/Eu* (B) ratios on Ce/Ce* ratioof the sediments from Sites 798 (circles) and 799 (dots).

Figure 17. Dependence of La (A) and Yb (B) concentrations (ppm) on A12O3

content (%) in the sediments from Sites 798 (circles) and 799 (dots).

737

42. GEOCHEMISTRY OF RARE EARTHS AND OTHER TRACE ELEMENTS …€¦ · geochemistry of major, minor,...

Documents

Transcript of 42. GEOCHEMISTRY OF RARE EARTHS AND OTHER TRACE ELEMENTS …€¦ · geochemistry of major, minor,...