James R. Hein

Comparison of the partitioning behaviours of yttrium, rare earth elements, and titanium between hydrogenetic marine ferromanganese crusts and seawater

In order to evaluate details of the partitioning behaviours of Y, rare earth elements (REEs), and Ti between inorganic metal oxide surfaces and seawater, we studied the distribution of these elements in hydrogenetic marine ferromanganese (Fe-Mn) crusts from the Central Pacific Ocean. Nonphosphatized Fe-Mn crusts display shale-normalized rare earths and yttrium (REYSN) patterns (Y inserted between Dy and Ho) that are depleted in light REEs (LREEs) and which show negative anomalies for Ysn, and positive anomalies for LaSN, EuSN, GdSN, and in most cases, Cesn. They show considerably smaller Y/ Ho ratios than seawater or common igneous and clastic rocks, indicating that Y and Ho are fractionated in the marine environment. Compared to P-poor crusts, REYSN patterns of phosphatized Fe-Mn crusts are similar, but yield pronounced positive Ysn anomalies, stronger positive LaSN anomalies, and enrichment of the HREEs relative to the MREEs. The data suggest modification of REY during phosphatization and indicate that studies requiring primary REY distributions or isotopic ratios should be restricted to non-phosphatized (layers of) Fe-Mn crusts. Apparent bulk coefficients, Kdm, describing trace metal partitioning between nonphosphatized hydrogenetic Fe-Mn crusts and seawater, are similar for Pr to Eu and decrease for Eu to Yb. Exceptionally high values of KDCe, which are similar to those of Ti, result from oxidative scavenging of Ce and support previous suggestions that Ce (IV) is a hydroxide-dominated element in seawater. Yttrium and Gd show lower KD values than their respective neighbours in the REY series. Results of modelling the exchange equilibrium between REY dissolved in seawater and REY sorbed on hydrous Fe-Mn oxides corroborate previous studies that suggested the surface complexation of REY can be approximated by their first hydroxide binding constant. Negative “anomalies” occur for stabilities of bulk surface complexes of Gd, La, and particularly Y. The differences in inorganic surface complex stability between Y and Ho and between Gd and its REE neighbours are similar to those shown by the stabilities of complexes with aminocarboxylic acids and are significantly larger than those shown by stabilities of complexes with carboxylic acids. Hence, sorption of Y and REEs onto hydrous Fe-Mn oxides may contribute significantly to the positive YSN and GdSN anomalies in seawater.

Uptake of elements from seawater by ferromanganese crusts: solid-phase associations and seawater speciation

Marine Fe–Mn oxyhydroxide crusts form by precipitation of dissolved components from seawater. Three hydrogenetic crust samples (one phosphatized) and two hydrothermal Mn-oxide samples were subjected to a sequential-leaching procedure in order to determine the host phases of 40 elements. Those host-phase associations are discussed with respect to element speciation in seawater. The partitioning of elements between the two major phases, Mn oxide and Fe oxyhydroxide, can in a first-order approximation be explained by a simple sorption model related to the inorganic speciation of the elements in seawater, as has been proposed in earlier models. Free and weakly complexed cations, such as alkali and alkaline earth metals, Mn, Co, Ni, Zn, Tl(I), and partly Y, are sorbed preferentially on the negatively charged surface of the MnO2 in hydrogenetic crusts. The driving force is a strong coulombic interaction. All neutral or negatively charged chloro (Cd, Hg, Tl), carbonate (Cu, Y, Pb, and U), and hydroxide (Be, Sc, Ti, Fe, Zr, Nb, In, Sn, Sb, Te, Hf, Ta, Bi, Th, and Tl(III)) complexes and oxyanions (V, Cr, As, Se, Mo, and W) bind to the slightly positively charged surface of the amorphous FeOOH phase. While coulombic interaction can explain the sorption of the negatively charged species, the binding of neutral species is based on specific chemical interaction. Organic complexation of elements in deep-ocean water seems to be at most of minor importance. Surface oxidation can explain some strong metal associations, e.g. of Co and Tl with the MnO2 and Te with the FeOOH. Sorption reactions initially driven by coulombic forces are often followed by the formation of specific bonds between the adsorbate and the atoms of the oxide surface. Differences in the associations of some metals between the non-phosphatized and phosphatized hydrogenetic crusts and between the hydrogenetic and the hydrothermal samples reflect the different physico-chemical environments of formation and speciations in oxic seawater vs. less-oxic fluids, especially for the redox-sensitive metals such as Mo and V. These environmental-related differences indicate that the methodology of chemical speciation used here in combination with spectroscopic methods may allow for the detection of changes in paleoceanographic conditions recorded during the several tens of millions of years of crust growth.

Iron and manganese oxide mineralization in the Pacific

Abstract Iron, manganese, and iron-manganese deposits occur in nearly all geomorphologic and tectonic environments in the ocean basins and form by one or more of four processes: (1) hydrogenetic precipitation from cold ambient seawater, (2) precipitation from hydrothermal fluids, (3) precipitation from sediment pore waters that have been modified from bottom water compositions by diagenetic reactions in the sediment column and (4) replacement of rocks and sediment. Iron and manganese deposits occur in five forms: nodules, crusts, cements, mounds and sediment-hosted stratabound layers. Seafloor oxides show a wide range of compositions from nearly pure iron to nearly pure manganese end members. Fe/Mn ratios vary from about 24 000 (up to 58% elemental Fe) for hydrothermal seamount ironstones to about 0.001 (up to 52% Mn) for hydrothermal stratabound manganese oxides from active volcanic arcs. Hydrogenetic Fe-Mn crusts that occur on most seamounts in the ocean basins have a mean Fe/Mn ratio of 0.7 for open-ocean seamount crusts and 1.2 for continental margin seamount crusts. Fe-Mn nodules of potential economic interest from the Clarion-Clipperton Zone have a mean Fe/Mn ratio of 0.3, whereas the mean ratio for nodules from elsewhere in the Pacific is about 0.7. Crusts are enriched in Co, Ni and Pt and nodules in Cu and Ni, and both have significant concentrations of Pb, Zn, Ba, Mo, V and other elements. In contrast, hydrothermal deposits commonly contain only minor trace metal contents, although there are many exceptions, for example, with Ni contents up to 0.66%, Cr to 1.2%, and Zn to 1.4%. Chondrite-normalized REE patterns generally show a positive Ce anomaly and abundant ΣREEs for hydrogenetic and mixed hydrogenetic-diagenetic deposits, whereas the Ce anomaly is negative for hydrothermal deposits and ΣREE contents are low. However, the Ce anomaly in crusts may vary from strongly positive in East Pacific crusts to slightly negative in West Pacific crusts, which may reflect the redox conditions of seawater. The concentration of elements in hydrogenetic Fe-Mn crusts depends on a wide variety of water column and crust surface characteristics, whereas concentration of elements in hydrothermal oxide deposits depends of the intensity of leaching, rock types leached, and precipitation of sulphides at depth in the hydrothermal system.

North Atlantic Deep Water export to the Southern Ocean over the past 14 Myr: Evidence from Nd and Pb isotopes in ferromanganese crusts

The intensity of North Atlantic Deep Water (NADW) production has been one of the most important parameters controlling the global thermohaline ocean circulation system and climate. Here we present a new approach to reconstruct the overall strength of NADW export from the North Atlantic to the Southern Ocean over the past 14 Myr applying the deep water Nd and Pb isotope composition as recorded by ferromanganese crusts and nodules. We present the first long-term Nd and Pb isotope time series for deep Southern Ocean water masses, which are compared with previously published time series for NADW from the NW Atlantic Ocean. These data suggest a continuous and strong export of NADW, or a precursor of it, into the Southern Ocean between 14 and 3 Ma. An increasing difference in Nd and Pb isotope compositions between the NW Atlantic and the Southern Ocean over the past 3 Myr gives evidence for a progressive overall reduction of NADW export since the onset of Northern Hemisphere glaciation (NHG). The Nd isotope data allow us to assess at least semiquantitatively that the amount of this reduction has been in the range between 14 and 37% depending on location.

Long-term eruptive activity at a submarine arc volcano

Three-quarters of the Earths volcanic activity is submarine, located mostly along the mid-ocean ridges, with the remainder along intraoceanic arcs and hotspots at depths varying from greater than 4,000 m to near the sea surface. Most observations and sampling of submarine eruptions have been indirect, made from surface vessels or made after the fact. We describe here direct observations and sampling of an eruption at a submarine arc volcano named NW Rota-1, located 60 km northwest of the island of Rota (Commonwealth of the Northern Mariana Islands). We observed a pulsating plume permeated with droplets of molten sulphur disgorging volcanic ash and lapilli from a 15-m diameter pit in March 2004 and again in October 2005 near the summit of the volcano at a water depth of 555 m (depth in 2004). A turbid layer found on the flanks of the volcano (in 2004) at depths from 700 m to more than 1,400 m was probably formed by mass-wasting events related to the eruption. Long-term eruptive activity has produced an unusual chemical environment and a very unstable benthic habitat exploited by only a few mobile decapod species. Such conditions are perhaps distinctive of active arc and hotspot volcanoes.

COBALT- AND PLATINUM-RICH FERROMANGANESE CRUSTS AND ASSOCIATED SUBSTRATE ROCKS FROM THE MARSHALL ISLANDS

Abstract Ferromanganese crusts cover most hard substrates on seafloor edifices in the central Pacific basin. Crust samples and their associated substrates from seven volcanic edifices of Cretaceous age along the Ratak chain of the Marshall Islands are discussed. The two most abundant substrate lithologies recovered were limestone, dominantly fore-reef slope deposits, and volcanic breccia composed primarily of differentiated alkalic basalt and hawaiite clasts in a phosphatized carbonate matrix. The degree of mass wasting on the slopes of these seamounts is inversely correlated with the thickness of crusts. Crusts are generally thin on limestone substrate. Away from areas of active mass-wasting processes, and large atolls, crusts may be as thick as 10 cm maximum. The dominant crystalline phase in the Marshall Islands crusts is δ-MnO 2 (vernadite). High concentrations of cobalt, platinum and rhodium strongly suggest that the Marshall Islands crusts are a viable source for these important metals. Many metals and the rare earth elements vary significantly on a fine scale through most crusts, thus reflecting the abundances of different host mineral phases in the crusts and changes in seawater composition with time. High concentrations of cobalt, nickel, titanium, zinc, lead, cerium and platinum result from a combination of their substitution in the iron and manganese phases and their oxidation potential.

Global occurrence of tellurium-rich ferromanganese crusts and a model for the enrichment of tellurium

Abstract Hydrogenetic ferromanganese oxyhydroxide crusts (Fe-Mn crusts) precipitate out of cold ambient ocean water onto hard-rock surfaces (seamounts, plateaus, ridges) at water depths of about 400 to 4000 m throughout the ocean basins. The slow-growing (mm/Ma) Fe-Mn crusts concentrate most elements above their mean concentration in the Earth’s crust. Tellurium is enriched more than any other element (up to about 50,000 times) relative to its Earth’s crustal mean of about 1 ppb, compared with 250 times for the next most enriched element. We analyzed the Te contents for a suite of 105 bulk hydrogenetic crusts and 140 individual crust layers from the global ocean. For comparison, we analyzed 10 hydrothermal stratabound Mn-oxide samples collected from a variety of tectonic environments in the Pacific. In the Fe-Mn crust samples, Te varies from 3 to 205 ppm, with mean contents for Pacific and Atlantic samples of about 50 ppm and a mean of 39 ppm for Indian crust samples. Hydrothermal Mn samples have Te contents that range from 0.06 to 1 ppm. Continental margin Fe-Mn crusts have lower Te contents than open-ocean crusts, which is the result of dilution by detrital phases and differences in growth rates of the hydrogenetic phases. Correlation coefficient matrices show that for hydrothermal deposits, Te has positive correlations with elements characteristic of detrital minerals. In contrast, Te in open-ocean Fe-Mn crusts usually correlates with elements characteristic of the MnO2, carbonate fluorapatite, and residual biogenic phases. In continental margin crusts, Te also correlates with FeOOH associated elements. In addition, Te is negatively correlated with water depth of occurrence and positively correlated with crust thickness. Q-mode factor analyses support these relationships. However, sequential leaching results show that most of the Te is associated with FeOOH in Fe-Mn crusts and ≤10% is leached with the MnO2. Thermodynamic calculations indicate that Te occurs predominantly as H5TeO6− in ocean water. The speciation of Te in ocean water and charge balance considerations indicate that Te should be scavenged by FeOOH, which is in agreement with our leaching results. The thermodynamically more stable Te(IV) is less abundant by factors of 2 to 3.5 than Te(VI) in ocean water. This can be explained by preferential (not exclusive) scavenging of Te(IV) by FeOOH at the Fe-Mn crust surface and by Fe-Mn colloids in the water column. We propose a model in which the extreme enrichment of Te in Fe-Mn crusts is likely the result of an oxidation reaction on the surface of FeOOH. A similar oxidation process has been confirmed for Co, Ce, and Tl at the surface of MnO2 in crusts, but has not been suggested previously to occur in association with FeOOH in Fe-Mn crusts. Mass-balance considerations indicate that ocean floor Fe-Mn deposits are the major sink for Te in the oceans. The concentration and redox chemistry of Te in the global ocean are likely controlled by scavenging on Fe-Mn colloids in the water column and Fe-Mn deposits on the ocean floor, as is also the case for Ce.

Osmium isotope variations in the oceans recorded by Fe-Mn crusts

This study presents osmium (Os) isotope data for recent growth surfaces of hydrogenetic ferromanganese (Fe‐Mn) crusts from the Pacific, Atlantic and Indian Oceans. In general, these data indicate a relatively uniform Os isotopic composition for modern seawater, but suggest that North Atlantic seawater is slightly more radiogenic than that of the Pacific and Indian Oceans. The systematic difference in the Os isotopic composition between the major oceans probably reflects a greater input of old continental material with a high Re =Os ratio in the North Atlantic Ocean, consistent with the distribution of Nd and Pb isotopes. This spatial variation in the Os isotope composition in seawater is consistent with a residence time for Os of between 2 and 60 kyr. Indian Ocean samples show no evidence of a local source of radiogenic Os, which suggests that the present-day riverine input from the Himalaya‐Tibet region is not a major source for Os. Recently formed Fe‐Mn crusts from the TAG hydrothermal field in the North Atlantic yield an Os isotopic composition close to that of modern seawater, which indicates that, in this area, the input of unradiogenic Os from the hydrothermal alteration of oceanic crust is small. However, some samples from the deep Pacific (4 km) possess a remarkably unradiogenic Os isotope composition ( 187 Os= 186 Os ratios as low as 4.3). The compositional control of Os incorporation into the crusts and mixing relationships suggest that this unradiogenic composition is most likely due to the direct incorporation of micrometeoritic or abyssal peridotite particles, rather than indicating the presence of an unradiogenic deep-water mass. Moreover, this unradiogenic signal appears to be temporary, and local, and has had little apparent effect on the overall evolution of seawater. These results confirm that input of continental material through erosion is the dominant source of Os in seawater, but it is not clear whether global Os variations are due to the input of mantle or meteoritic material, or simply indicate that the continental source itself is not uniform.

Two Major Cenozoic Episodes of Phosphogenesis Recorded in Equatorial Pacific Seamount Deposits

Seamount phosphorites have been recognized since the 1950s, but this is the first study to provide an in depth exploration of the origin and history of these widespread deposits. Representative samples from equatorial Pacific Cretaceous seamounts were analyzed for chemical, mineralogical, and stable isotope compositions. The phosphorites occur in a wide variety of forms, but most commonly carbonate fluorapatite (CFA) replaced middle Eocene and older carbonate sediment in a deep water environment (>1000 m). Element ratios distinguish seamount phosphorites from continental margin, plateau, and insular phosphorites. Uranium and thorium contents are low and total rare earth element (REE) contents are generally high. REE ratios and shale-normalized patterns demonstrate that the REEs and host CFA were derived from seawater. Strontium isotopic compositions compared with inferred Cenozoic seawater curves define two major episodes of Cenozoic phosphatization: Late Eocene/early Oligocene (39–34 Ma) and late Oligocene/early Miocene (27–21 Ma); three minor events are also indicated. The major episodes occurred at times of climate transition, the first from a nonglacial to glacial earth and the second from a predominantly glacial to warm earth. The paleoceanographic conditions that existed at those times initiated and sustained development of phosphorite by accumulation of dissolved phosphorus in the deep sea during relatively stable climatic conditions when oceanic circulation was sluggish. Fluctuations in climate, sealevel, and upwelling that accompanied the climate transitions may have driven cycles of enrichment and depletion of the deep-sea phosphorus reservoir. As temperature gradients in the oceans increased, Antarctic glaciation expanded and oceanic circulation and upwelling intensified. Expansion and intensification of the oxygen minimum zone may have increased the capacity for midwater storage of phosphorus supplied by dynamic upwelling around seamounts; however, the bottom waters never became anoxic during the phosphogenic episodes. Fluctuations in the CCD and lysocline, CO2 fluxes, and changes in bottom water circulation and temperatures may have bathed the seamount carbonates in more corrosive waters which, coupled with increased supplies of dissolved phosphorus, promoted replacement processes. The late Eocene/early Oligocene phosphogenic episode recorded in seamount deposits is not matched by large phosphorite deposits in the geologic record, whereas the late Oligocene/early Miocene episode and middle Miocene event are matched by large deposits distributed globally. The seamount phosphorites are exposed at the surface of the seamounts and have been for most of the Neogene and Oligocene. The phosphorites do not show signs of etching that would indicate substantial undersaturation of seawater phosphate with respect to CFA. Mass balance calculations indicate that about 5.4–19 × 1012 g of P2O5 are locked up in equatorial Pacific seamount phosphorites. That amount is equivalent to about 2-7 years of the present annual input from rivers.

Actual timing of neodymium isotopic variations recorded by FeMn crusts in the western North Atlantic

Hydrogenetic ferromanganese (FeMn) crusts from the western North Atlantic record variations in the Nd and Pb isotopic composition of Cenozoic deep water preserved during their growth. The timing and cause of the most striking change have been the subject of debate. Some have proposed that the shift took place after 4 Ma in response to the closure of the Panama gateway. Others have argued that the major change in isotope composition occurred as early as 8 Ma. This study presents high-resolution Nd isotope records for crusts previously dated using 10Be/9Be chronology. These data confirm that the shifts in Nd occurred after 4 Ma, consistent with a likely relationship with the closure of the Central American Isthmus and intensification of Northern Hemisphere Glaciation, and in accordance with changes seen in other physical and chemical records. These results illustrate the need for both a robust chronological framework and high-resolution records before a reliable paleoceanographic interpretation can be made of the variations recorded by FeMn crusts.