Earth Mantle

Consequently, Earth's mantle today is a complex assemblage of lithologically and compositionally heterogeneous materials: peridotites (or their high-pressure equivalents), which have undergone an unknown number of melt extraction events in the shallow mantle, and other mafic components, namely recycled crustal materials.

From: Encyclopedia of Geology (Second Edition) , 2021

The mantle

Kent C. Condie , in Earth as an Evolving Planetary System (Fourth Edition), 2022

Introduction

Earth's mantle plays an important role in the evolution of the crust and provides the thermal and mechanical driving forces for plate tectonics. Heat liberated by the core is transferred into the mantle where most of it (>   90%) is convected through the mantle to the base of the lithosphere. The remainder may be transferred upward by mantle plumes generated near the core-mantle boundary layer. In the past decade, the resolution of structures in the mantle from seismology has greatly improved and we can now see some details as small as 100   km. The mantle is also the graveyard for descending lithospheric slabs, and the fate of these slabs is subject of ongoing discussion and controversy. Do they partially collect at the 660-km discontinuity in the upper mantle as they descend and what effect do they have on convection in the mantle? Do mantle plumes exist, and if they exist, how and where are they generated and what role do they play in mantle-crust evolution? Another exciting mantle topic is that of the origin and growth of the lithosphere and if its role in plate tectonics has changed with time. The Archean continental lithosphere, for instance, is considerably thicker than post-Archean continental lithosphere and geochemical data from xenoliths suggest it had quite a different origin. Still another hot topic is the origin of isotopic differences in basalts, which reflect different compositions and ages of mantle sources. How did these sources form and survive for billions of years in a convecting mantle? These are some of the questions we will address in this chapter.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B978012819914500010X

Volume 3

Weiwei Wu , ... Huichao Rui , in Encyclopedia of Geology (Second Edition), 2021

Analogy of Oceanic Lithosphere

Earth's mantle plays an important role in the evolution of the crust and provides the driving forces for plate tectonics, which eventually determines the planet's habitability. As an important component of the Earth's mantle, however, the study of modern oceanic lithosphere is limited by the tools available. Basic information has obtained through geophysical methods, such as acoustic imaging, seismology, magnetism, magneto-tellurics, gravimetry and heat flow measurements (Anderson, 2006; Helffrich, 2006). Nevertheless, as a result of limitations of resolution of the geophysical methods, the shape, relief, compositional structure, segmentation geometry and other internal details of the oceanic lithosphere created along the mid-ocean ridges remain elusive and not fully understand. In other hand, in situ abyssal samples are obtained from ocean floors and transform faults through dredging, drilling programs and sampling by deep sea submarines (Warren, 2016), however, it is difficult and expensive to study. Moreover, the relative position of some dredged abyssal samples in an oceanic lithosphere is not well constrained. Since the realization that ophiolites represent fragments of ancient oceanic lithosphere, studying ophiolites opens the door for geologists to contribute fundamentally to understanding oceanic lithosphere due to the high-fidelity magmatic and stratigraphic records preserved in them.

Partial melting of the Earth's upper mantle beneath the mid-ocean ridges is the main driving force for the chemical differentiation of the silicate Earth (Condie, 2016). Theoretically, people assume that the crust is enriched in incompatible elements and left behind is a residual mantle that is preferentially depleted in those elements, but well homogenized by vigorous convective stirring. However, studies of MORB and abyssal peridotite have also identified the occurrence of compositional heterogeneities in the mantle beneath mid-ocean ridges and diminish the valid ity of this assumption (e.g., Hofmann, 1997; Warren, 2016). For instance, the abyssal peridotites underlain the mid-ocean ridge supposed to undergo limited melting at the present-day ridge axis. However, an ultra-refractory mantle block, based on infertile major, trace, and modal compositions, demonstrates at the 15–16°N region around the Fifteen-Twenty transform fault on the MAR (Paulick et al., 2006; Seyler et al., 2007; Godard et al., 2008), indicating previous episodes of melt extraction. By contrast, some peridotites from the Romanche transform fault on the MAR (Seyler and Bonatti, 1997) show more fertile compositions than depleted mantle (DM) (Workman and Hart, 2005), implying enriched source mantle. In addition, studies on all the long-lived radiogenic isotope systems show ultra-depleted isotopic compositions, such as Nd (Salters and Dick, 2002; Cipriani et al., 2004; Mallick et al., 2014), Sr (Warren et al., 2009), Pb (Warren and Shirey, 2012), Hf (Stracke et al., 2011; Mallick et al., 2015) and Os (Harvey et al., 2006; Liu et al., 2008), providing unequivocal evidence for pre-existing depletions. All the lines of evidence obtained from global abyssal peridotites suggest a modern heterogeneous oceanic mantle. However, the nature of ancient oceanic mantle (older than 180   Ma) is unconstrained due to most of them had been subducted into deeper mantle.

As a survivor of oceanic lithosphere suffered plate subduction, ophiolite can provide "older" information. In fact, geochemical (e.g., REE, HFSE) and geochronological data (e.g., Re-Os isotopic) from many ophiolites have shown strong evidence for compositional heterogeneity in their mantle units (Aldanmaz et al., 2009; O'Driscoll et al., 2012; Piccardo et al., 2014). The crustal and mantle sequences in some ophiolites may not represent a simple melt-residual relationship as generally believed (Snow et al., 2000; Tsuru et al., 2000; Walker et al., 2002; Alard et al., 2005; Gervilla et al., 2005; Frei et al., 2006; Shi et al., 2007, 2012; Marchesi et al., 2011; O'Driscoll et al., 2012; González-Jimenez et al., 2012, 2013). For example, Os-rich alloys from the Neo-Tethyan Dongqiao ophiolites, China, show smooth PGE patterns and 187Os/188Os ranges from 0.12003 to 0.12194 (yielding Re-depletion ages ≥   1.1   Ga), which may represent residue of a sub-continental lithospheric mantle (Shi et al., 2007). No consensus has been reached about the compositional heterogeneity observed in ophiolitic mantle. It is generally suggested that these "older" peridotite mantle may be fragments of the subcontinental lithospheric mantle (SCLM) (Hassler and Shimizu, 1998; Rampone et al., 2005; Shi et al., 2007; O'Reilly et al., 2009; González-Jimenez et al., 2013) or long-term preservation of refractory domains in the incompletely homogenized asthenospheric mantle (Liu et al., 2008). Irrespective to the origin of the compositional heterogeneity, the way in which these "older" peridotites isolated from the vigorous convective mantle is still a matter of active debate. Anyway, based on the studies comparison ophiolite (ancient oceanic lithosphere) and modern oceanic lithosphere, it indicates that the Earth's upper mantle is heterogeneous, at least from the oldest ophiolite to the present day.

The ophiolite analogy to the ancient oceanic crust has its disadvantages. We assume that if present-day oceanic lithosphere is representative of ancient oceanic lithosphere (ophiolite) then comparisons of their respective petrologic, geologic, and physical properties should reveal strong similarities. The basic flaw in this inference is that it is assumed that the present-day processes that give rise to new oceanic crust beneath mid-ocean ridge are the same as those that produced ancient oceanic crust (ophiolite) in the past and can be directly compared. However, the physical and chemical state of the Earth's mantle has been changing with the evolution of the planet since it formed ~   4.5   Ga ago, as well as the changing global tectonic regimes (Condie, 2016). As for instance during the Hadean, radiogenic heat production in Earth was three to five times greater than that at present resulting in hotter mantle in ancient time (about 100–300   °C hotter in early Archean than at present) (Galer, 1991; Condie, 2016).

Another most serious disadvantage is the great volume difference between present-day oceanic spreading centers and on-land ophiolites that have formed in ancient oceans now vanished. Considering the volume of ophiolites tectonically emplaced during the Phanerozoic time compared to the amount of oceanic crust formed during this same period of time, less than 0.001% of the oceanic crust has survived from subduction (Coleman, 1977). It therefore seems unlikely that this small proportion of ancient oceanic lithosphere (ophiolite) can reveal all of the secrets of the oceanic lithosphere evolution.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780081029084000746

The Mantle

Kent C. Condie , in Earth as an Evolving Planetary System, 2005

Publisher Summary

The Earth's mantle plays an important role in the evolution of the crust and provides the thermal and mechanical driving forces for plate tectonics. The mantle is also the graveyard for descending lithospheric slabs, and the fate of these slabs in the mantle is a subject of ongoing discussion and controversy. Because subducted plates are relatively cool, they decrease the temperature of nearby mantle, leaving relatively warm mantle in the regions between two subduction zones. These broad, warm regions of mantle, known as "mantle upwellings" are relatively buoyant and rise, providing the return flow. Mantle upwellings elevate the Earth's surface a few hundred meters, producing superswells. Because they elevate the temperature of the uppermost mantle, widespread small degrees of melting occur near the tops of upwellings, giving rise to volcanism and to mafic underplating of the crust. Another characteristic of upwellings is that they contain most of the modem hotspots and, hence, most of the modem mantle plumes.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780120883929500049

Mantle Dynamics

D. Bercovici , ... Y. Ricard , in Treatise on Geophysics (Second Edition), 2015

7.07.4.4 Slab Heat Flux and Plate Velocities

The Earth's mantle is thought to be about 70–80% heated internally by primordial heat and radiogenic sources, and the rest from other sources, primarily from cooling of the core (see Chapter 7.11 ). As plumes are thought to carry the heat from the core (see next section), heat flow through the oceanic lithosphere, or equivalently cooling by slabs, accounts for most of the Earth's heat flow. Indeed, one can use these heat flow arguments to predict slab and plate velocities and check for self-consistency with the boundary layer force balance arguments of Section 7.07.4.1.1 , which thus verifies that slabs are an integral part of convective heat transport (Bercovici, 2003).

At a few 100   km depth beneath the lithosphere, mantle heat loss is primarily due to the downward injection of cold material by subducting slabs (analogous to dropping ice cubes in hot water). The energy-flux balance would thus require that fQ  = wAρc p ΔT where Q  ≈   30–38   TW is the net heat output through the top of the mantle (Schubert et al., 2001) f  =   0.8–0.9 is the fraction of Q accounted for by slab cooling (since the remaining heat transport by mantle plumes probably accounts for of order 10–20% of the net heat flux; Davies, 1988a; Jaupart et al., 2007; Lay et al., 2008; Pollack et al., 1993; Sleep, 1990), ρ  ≈   3000   kg   m−   3 is slab density, cp   =   1000   J   kg   K−   1 is heat capacity, ΔT  ≈   700   K is the average slab thermal anomaly as in Section 7.07.4.1.1 , and w is a typical vertical velocity of a slab. Lastly, A  ≈   2πRδ is the total horizontal cross-sectional area of all slabs crossing this particular depth where δ  ≈   100   km is a typical slab thickness, and the horizontal length of all slabs is estimated by the circumference of the Earth since most slabs occur in a nearly large circle around the Pacific basin and thus R  ≈   6000   km. Using these numbers, we can solve for slab velocity w  = fQ/(ρc p ΔTδ2πR)   ≈   10   cm year−   1, which is precisely the typical velocity for fast, active plates and predicted by slab-pull arguments ( Section 7.07.4.1.1 ). In the end, that 'active' plates cool, thicken, subside, and eventually sink as subducting slabs at plate velocities, while also cooling the mantle is tantamount to saying slabs and plates are convective currents.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780444538024001354

MANTLE PLUMES AND HOT SPOTS

D. Suetsugu , ... T. Kogiso , in Encyclopedia of Geology, 2005

Dynamics of Mantle Plumes

In Earth's mantle, the Rayleigh number, a measure of the vigour of convection, is estimated to be 5 × 106 to 5 × 107. For this range, time-dependent convection, with instabilities originating at thermal boundary layers, is expected. Plume formation may, however, be suppressed by large-scale flow related to plate motions, which, if sufficiently fast, advects growing instabilities towards large-scale upwellings before individual plumes can form. For somewhat slower large-scale flow, individual plumes may exist, but are advected towards, and hence cluster in, regions of large-scale upwellings. A thermal boundary layer exists above the core–mantle boundary, which is therefore a likely plume source depth. Another thermal boundary and possible source depth is at the 660. Plumes may also originate at chemical boundaries within the mantle. No conclusive evidence exists for those, but anticorrelation of bulk sound and shear wave anomalies in the lowermost mantle indicates that the D″ layer at the base of the mantle may be chemically distinct from the overlying mantle. Topography of the chemical boundary could be up to a few 100 km, and plumes may rise from the high points (cusps) on the boundary. Plumes may entrain material from the underlying layer; the amount of entrained material depends on the density difference between the two layers, due to chemical and temperature differences. Both laboratory and numerical experiments show that a plume that is less viscous than the surrounding mantle tends to have a large, roughly spherical or mushroom-shaped head, followed by a narrow tail (conduit) connecting the head and the source region (Figure 5). Scaled to Earth dimensions, a head diameter of several 100 to 1000 km is expected, if it rises from the lowermost mantle. Such a size is also required to explain the volume of many large igneous provinces, thus adding support to the notion that these originate from upwellings from the lowermost mantle, whereas hotspots not associated with flood basalts may be caused by plumes from shallower depth. Conduits should be only ≈100 km wide. Thinner conduits may occur for strongly temperature-dependent viscosity. If mantle viscosity increases strongly with depth, conduit diameter could be up to a few hundred kilometres in the lower mantle. Ascent times of plume heads through the whole mantle are estimated to be ≈10 My to several tens of millions of years. For nonlinear mantle rheology, decrease of effective viscosity due to larger stresses around plume heads may cause rise times shorter than for Newtonian viscosity. As a plume head rises, the surrounding mantle heats up and becomes buoyant and less viscous. Mantle material is entrained into the rising plume, which therefore contains a mixture of materials from the source region and ambient mantle. Formation of a flood basalt is thought to occur when a plume head reaches the base of the lithosphere. Subsequently, the conduit may remain in existence as long as hot material is flowing in at its base – for 100 My or longer. The conduit consists of a narrow core, where most of the material transport occurs, and a thermal halo. Due to thermal diffusion, heat is lost from plume conduits as they traverse the mantle. Though strong plumes, such as the Hawaii plume, are not significantly affected, heat loss significantly reduces the temperature anomaly expected for weaker plumes. For weak plumes from the core–mantle boundary, with anomalous mass flux of ≲500–1000 m3 s−1, the sublithospheric temperature anomaly is low and no melting is expected. Weak hotspots may therefore have shallower origins. On the other hand, the temperature anomaly of strong plumes, inferred from observations, is much less than expected from the temperature drop across a thermal boundary layer between core and mantle. This may indicate a chemically distinct layer at the base of the mantle, with plumes rising from its top. Mantle plumes may coexist with superplumes, and conduits are expected to be tilted and distorted in large-scale mantle flow. The rising of a tilted conduit may cause further entrainment of ambient mantle material. If the tilt exceeds ≈60°, the conduit may break into separate diapirs, which may lead to extinction of the plume. As a consequence of conduit distortion, overlying hotspots are expected to move. Hence, mantle plumes probably do not provide a fixed reference frame. However, if plumes arise from a high-viscosity lower mantle, hotspots should move much more slowly than lithospheric plates move. Conduits are likely to be time variable, with disturbances traveling along them; these may be wave-like or may take the shape of secondary plume heads. Waves are associated with increased conduit flux, which may explain flux variations in mantle plumes. Ascending plumes interact with mantle phase transitions. The 660 somewhat inhibits flow across but is unlikely to block penetration of plumes. Experiments involving high pressure suggest phase relations of a pyrolite mantle such that, at the high temperatures of mantle plumes, this phase boundary does not hinder flow across. Beneath the lithosphere, buoyant plume material flows out of the conduit, spreads horizontally in a low-viscosity asthenosphere, and is dragged along with moving plates. Plume material buoyantly lifts up the lithosphere and causes a hotspot swell. Partial melt extraction at the hotspot may leave behind a buoyant residue that also contributes to swell formation. Plume material does not necessarily erupt directly above the conduit. It may also flow upward along the sloping base of the lithosphere, and enhanced melting may occur at steep gradients. A sloping base exists near spreading ridges. If ridge and hotspot are less than a few hundred kilometres apart, eruption of volcanics may occur at the ridge rather than, or in addition to, directly above the plume. Also, in other cases, such as in Africa, the spatial distribution of plume-related melting and magmatism may be controlled by the lithosphere rather than by the plume position. Formation of vertical fractures and ascent of magma through the lithosphere preferably occur for tensile lithospheric stresses. Loading of the lithosphere by hotspot islands causes stresses that may influence formation of fractures and therefore determine the spacing of hotspot islands along tracks. Feeding of plume material to a nearby ridge may put the lithosphere above the plume under compression and shut off eruption directly above the plume. If a hotspot (e.g., Iceland) is located close to the ridge, the viscosity contrast between plume and ambient mantle may become a factor 1000 or more beneath thin lithosphere. Such large viscosity variations facilitate ridge-parallel flow of plume material and help to explain geochemical anomalies south of Iceland. Propagation of pulses in plume flux explains the V-shaped topography and gravity anomalies at the Reykjanes Ridge. Probably not all intraplate volcanism is caused by plumes as described. In many cases, the origin of intraplate volcanism may be shallow, due to cracks in the lithosphere caused by tensional stresses, or due to edge-driven convection at locations where lithospheric thickness varies laterally.

Figure 5. Numerical model of an axisymmetric mantle plume with strongly temperature-dependent viscosity. Colours indicate temperature (red = hottest, blue = ambient mantle); black lines indicate marker chains. Plots are at 14-My time intervals. Reproduced with permission from van Keken P (1997) Evolution of starting mantle plumes: a comparison between numerical and laboratory models. Earth and Planetary Science Letters 148: 1–11.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B0123693969001313

The Mantle and Core

D.G. Pearson , N. Wittig , in Treatise on Geochemistry (Second Edition), 2014

Abstract

The fragments of Earth's mantle (xenoliths) that erupted in deeply derived volcanic rocks provide unique windows into the chemistry of the deep Earth and the formation of the earliest continents. They also provide one of the most direct ways of estimating the primitive composition of the Earth's mantle. The geochemical measurements of mantle xenoliths beneath the oldest parts of the continents' 'cratons' can be used to constrain the age, depth of formation, and hence tectonic environment that led to the formation of deep, stable lithospheric keels extending to over 200   km. The processes that obscure the primary geochemical signatures in mantle xenoliths in the context of new stable isotopic variations are reviewed. New compositional versus depth data are compiled that are used along with recent parameterizations of polybaric mantle melting to evaluate models for the formation of deep cratonic keels.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780080959757002059

Mantle Dynamics

C. Jaupart , ... J.-C. Mareschal , in Treatise on Geophysics (Second Edition), 2015

7.06.7.3 The Peculiarities of Mantle Convection: Observations

Convection in the Earth's mantle proceeds in peculiar ways. One distinctive feature is the triangular age distribution of the seafloor ( Figure 7 ; Becker et al., 2009; Labrosse and Jaupart, 2007; Parsons, 1982), which is at odds with other convecting systems as well as with the parameterized schemes discussed earlier. We show in Appendix H the age distribution at the upper boundary of several convective layers. None of them resemble that of Earth, which illustrates current limitations in reproducing mantle convection processes.

A few other peculiar features of mantle convection are worth mentioning. Heat loss is unevenly distributed at the surface. The Pacific Ocean alone accounts for almost 50% of the oceanic total and 34% of the global heat loss of the planet. This is due in part to the large area of this ocean and in part to its high spreading rate. Oceanic plates are transient, such that changes of oceanic heat loss may occur when a new ridge appears or when one gets subducted. For example, the heat flux out of the Atlantic Ocean is about 6   TW, 17% of the oceanic total (Sclater et al., 1980). This ocean has almost no subduction and started opening only at 180   My. At that time, the generation of a new mid-ocean ridge led to an increase of the area of young seafloor at the expense of old seafloor from the other oceans and hence to enhanced heat loss. From the standpoint of dynamics, the most challenging features of mantle convection are perhaps the large variations of plate speeds and dimensions that exist. With the small number of plates present, averaging values of spreading velocity and plate size may well be meaningless.

One key feature of Earth is the presence of large continents at the surface, which play an important role. They do not allow large heat fluxes through them and generate boundaries with complicated shapes that constrain mantle flow. They exert a strong control on secular cooling and may well be responsible for the triangular distribution of ocean floor ages (Coltice et al., 2012; Grigné and Labrosse, 2001; Labrosse and Jaupart, 2007; Lenardic et al., 2005).

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780444538024001263

Earthquake Thermodynamics and Phase Transformations in the Earth's Interior

William F. McDonough , in International Geophysics, 2001

1.3 Early Evolution of the Earth

The formation of the Earth's core and mantle at about 4.6–4.5 Ga via planetary accretion and core separation suggests a very hot early Earth that continues to convect vigorously in order to remove heat from the planet's interior. This thermal evolution has had a marked influence on the chemical evolution of the planet (Davies, 1990). The mantle and its properties regulate the removal of heat from the Earth's core. Likewise, the atmosphere controls the Earth's capacity to radiate heat to outer space.

The accretion of the inner planets was likely to have occurred in a relatively high temperature environment. The collapse of the protosolar cloud to form the sun produced a hot central region in the solar system and a considerable temperature gradient across it (Boss, 1998; Lewis, 1974). The inner solar system (e.g., perhaps out to 3 AU or more) experienced high-temperature processing (Boss, 1998), including the melting of grains, inclusions, and chondrules (high-temperature components of meteorites). The earliest formed materials have crystallization ages on the order of < 3 Ma (or less) after T 0 (Allegre et al., 1995a). It is these high-temperature condensates, along with lesser amount of lower-temperature condensates that coalesced to form grains and larger sized fragments, that further accreted to form larger bodies, including planetesimals and ultimately planets. A considerable amount of thermal energy is evolved in this process, resulting in substantial internal heat in the latter stages of planet building that must be dissipated from the planet's interior. In addition, the giant impact event hypothesized to form the Moon would have significantly heated the Earth (Melosh, 1990).

If the Earth accreted from a mixture of silicates and metal particles followed by metal-silicate differentiation, separation of the Earth's core would heat the mantle (Birch, 1965; Elsasser, 1963; Flasar and Birch, 1973). It was calculated that the gravitational energy released by core formation would be converted into thermal energy, which would be enough to heat the mantle by about 1000–2000°C—thus driving mantle convection (Davies, 1990).

How fast this heat was dissipated to space depends on the nature of the early atmosphere. If the Earth had a significant gaseous envelope surrounding it throughout most of its accretion, it would have enhanced the chances of the upper portion of the mantle being wholly molten through thermal blanketing and greenhouse heating of the surface. Alternatively, if there was no atmosphere, the planet's heat is rapidly lost to space by radiation and little to no extensive melting of the mantle would have occurred.

The preceding considerations lead to the suggestion that the Earth's mantle experienced large-scale melting during accretion and core formation. Collectively these processes start the convective engine for the mantle. Given the likely event of the outer portion of the mantle experiencing significant global melting, one would expect that the mantle would have also experienced some degree of differentiation (crystal–liquid separation). However, there is no geochemical and/or isotopic evidence, based on a wide spectrum of crustal and mantle rocks (including peridotites and komatiites), in support of this global differentiation process (see the review by Carlson, 1994). Thus, if differentiation of the mantle occurred in the Hadean, its effects have been completely erased by the processes of rapid and vigorous convection.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/S0074614201800772

Planets, Asteriods, Comets and The Solar System

A.N. Halliday , in Treatise on Geochemistry (Second Edition), 2014

2.8.11.2 Earth's Hadean Environment

The nature of the Earth's mantle, crust, atmosphere, and hydrosphere after the Earth cooled following the cessation of the main stages of accretion has been the subject of a fair amount of research, particularly given the lack of data upon which to base firm conclusions. A particularly topical and important question is to what extent it may have been possible for life to develop during this period (Mojzsis et al., 1996; Sleep et al., 2001). A great deal has been written on this, and because it is covered elsewhere in the Treatise on Geochemistry, only cursory background is provided here. A schematic of the Hadean as best it can be currently given is shown in Figure 32 .

Figure 32. Schematic showing the timescales for various events through the 'Dark Ages' of the Hadean.

Following the giant impact, the Earth would have cooled rapidly by convective transport from the deep mantle to the surface. In simple terms, the hotter the Earth, the faster it convects, so any magma ocean following the giant impact would have been short-lived. However, the Moon would have been much closer to Earth than it is today; it would have occupied about a third of the night sky. This close proximity would have led to tidal effects that have been discussed by Zahnle et al. (2007). The deep mantle, in particular, would have been subject to heating and convective overturn for the first 100   Ma after the giant impact. It has been proposed by some that Earth's mantle may house hidden deep early formed reservoirs (Boyet and Carlson, 2005; Murphy et al., 2010; Tolstikhin and Hofmann, 2005; Tolstikhin et al., 2006). This is hard to reconcile with the likely degree of mantle overturning in the early Earth. Indeed, the very fact that asthenospheric mantle heterogeneity is so young (<   2   Ga) provides powerful evidence that through the Archean there was efficient mixing.

Even after this period, Earth's interior was a few 100   K hotter because of secular cooling from accretion and far greater radiogenic heat production. The Earth's heat flow was two to three times higher. Convection should have been much more vigorous (Chase and Patchett, 1988; Galer and Goldstein, 1991). Therefore, one can assume that more heat was escaping via mantle melting and production of oceanic crust, so plates would have been smaller. This, in turn, means more mantle-derived volatiles such as CO2 were being released throughout the Hadean. It also means that there was more hydrothermal alteration of the ocean floor. Therefore, CO2 was converted to carbonate in altered basalt and returned to the mantle at subduction zones (if those really existed already).

The first-order constraints that exist on the nature of Earth's early exosphere are discussed in Sleep et al. (1989), Sleep and Zahnle (2001), and Zahnle et al. (2007). The Sun was fainter and cooler than today because of the natural start-up of fusion reactions that set it on the main sequence (Kasting and Grinspoon, 1991; Pavlov et al., 2000; Sagan and Chyba, 1997). Therefore, the level of insolation will have been less. There may have been far less marine carbonate. We can infer this from the geologic record for the Archean. It appears that atmospheric CO2 levels were low: most of the CO2 was being recycled to the mantle. Because atmospheric CO2 exerts a profound effect on temperature as a greenhouse gas, from the low levels, one can infer that atmospheric temperatures were cold, unless another greenhouse gas such as methane (CH4) was very abundant (Pavlov et al., 2000). However, clear geochemical evidence for a strong role of methane in the Archean currently is lacking. Impacts, depending on their number and magnitude (Hartmann et al., 2000; Ryder et al., 2000), may have had a devastating effect on the early biosphere. Impact ejecta will react with atmospheric and oceanic CO2 and thereby lower atmospheric CO2 levels reducing atmospheric temperatures still further.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780080959757001236

The Mantle and Core

K. Righter , ... K. Domanik , in Treatise on Geochemistry (Second Edition), 2014

3.12.5.4 Mantle fO2

The oxygen fugacity of Earth's mantle is central to metal/silicate equilibrium, yet it is poorly characterized in experiments. The use of a relative fO2 estimate using run product metal and silicate compositions is common, but ignores the pressure dependence of oxygen fugacity which is sizable. To illustrate the drawbacks of a simplified relative fO2 approach, the example of the relative volumetric properties of FeO and Fe2O3 in silicate melts even at relatively low pressures (Kress and Carmichael, 1991) is considered. There are two problems: first, the pressure dependence of FeO–Fe2O3 equilibria is different than the pressure dependence of the nickel–nickel oxide (NNO) buffer such that use of NNO as a normalization for relative fO2 can be misleading. The volume change of the FMQ (fayalite–magnetite–quartz) buffer (0.17 log fO2 units per GPa) is much closer to the redox state of a silicate melt than is that of the nickel–nickel oxide buffer (0.51 log fO2 units per GPa; Kress and Carmichael, 1991). Second, the compressibilities of FeO and Fe2O3 are much different, so that even pressures of 1–2   GPa will have a large effect on their partial molar volumes. Any calculation of relative fO2 at the high PT conditions of the mantle should include an assessment of the volumetric properties of both the buffer phases, and the phases involved in the redox reaction.

Using this previous work on magmas as an example of the effect of pressure, applying this approach to metal/silicate systems clearly has similar drawbacks. The relative fO2 calculated in the absence of information about, for example, the compressibility of Fe in a core-forming liquid, or FeO in a peridotite melt, is likely to be inaccurate, especially if extrapolated across tens of GPa. For example, if the relative oxygen fugacity of an experimental run of the same silicate and metallic melt composition (ΔIW value) at 20   GPa, 2000   °C is different from that at 5   GPa, 1700   °C, then there is a problem. The best approach to the calculation of relative fO2 for metal/silicate systems involves knowing not only the volumetric properties of Fe and FeO for the IW buffer, but also for Fe in a core-forming liquid, and FeO in a peridotite melt. Such calculations are currently limited to a subset of metallic liquid and silicate melt compositions, due to incomplete data for the diverse composition space that is characteristic of the Earth's mantle and core. However, calculations on these limited compositions (Fe–Ni–S metal and peridotite melt) by Righter and Ghiorso (2012a,b) show for various experiments from the literature that calculated absolute fO2 can be as much as 2 log fO2 units higher than the relative fO2 calculated using the ratio of XFe and XFeO. In addition, and perhaps more importantly, Righter and Ghiorso (2012a,b) show that the fO2 becomes much lower than the IW buffer along a peridotite liquidus to 50   GPa, and can be as reduced as IW-6. These very reducing conditions are not normally considered in early Earth settings, but would have a strong effect on other redox equilibria and also exert controls on the composition of the atmosphere. This is an area where more work will provide much needed constraints on the metal/silicate equilibria.

There also has been a disconnect between the early reduced fO2 required for core formation models (IW-2 to IW-3.6), compared to the Earth's Archean and current upper mantle fO2 which is near FMQ (Arculus, 1985; Delano, 2001; Trail et al., 2011). If the young Earth allowed metallic liquid to pass through its mantle to the core, yet the upper mantle is not reduced enough for iron metal stability, how did Earth's mantle become oxidized? Several ideas have been proposed but they all have drawbacks and a satisfying explanation for this conundrum has remained elusive. One possibility may simply be that the upper mantle has become oxidized over time due to the effects of recycling and plate tectonics. However, no studies have yet revealed a secular trend of oxygen fugacity (e.g., Canil, 2002; Delano, 2001; Eggler and Lorand, 1995). Another idea is that the systematic breakdown of Mg-perovskite into Fe metal and Fe3   +‐bearing silicates has led to natural oxidation of the upper mantle (Frost et al., 2004a; Wade and Wood, 2005). Although this is an intriguing idea, and one that would occur early enough in Earth history to meet the requirements of current models, it is not without problems or questions. For example, the mantle of Mars is just as oxidized as the Earth's (near FMQ) but there is not an Mg-perovskite reservoir in Mars that can produce the oxidation (Righter et al., 2008b). In addition, it is not clear if Fe2O3 is added to the upper mantle by Mg-perovskite dissolution, it may dissociate into FeO and Fe2O3 in response to the low ambient fO2 set by core formation (see also Section 3.12.5.1.5 ). A third and more likely and/or promising possibility is that the mantle was oxidized somewhat by the partitioning of H and C between the core, mantle, magma ocean, and atmosphere (Abe et al., 2000; Kuramoto and Matsui, 1996; Okuchi and Takahashi, 1998). These authors show that C prefers the core while most H prefers the silicate melt, and estimate that the amount of H2O partitioned to silicate melt is large enough to explain the amount of H2O in the hydrosphere and mantle, provides enough oxygen to partially oxidize ferrous iron to ferric iron in the mantle, and perhaps even be the oxidant for metals which may fail to segregate to the core such as late‐accreted highly siderophile elements.

The oxidation state of the early mantle and late accretion events are linked. Many have argued for late (post core formation) accretion of chondritic material to the Earth's upper mantle to explain the near chondritic and elevated HSE abundances. Others have argued that evidence for a late chondritic veneer is recorded in the lunar regolith (Bottke et al., 2010; Puchtel et al., 2008), in the mantle of Mars (Bottke et al., 2010; Brandon et al., 2011), in the mantle of asteroid 4 Vesta (Day et al., 2012), and in the mantle of the angrite parent body (Riches et al., 2012). As noted in Section 3.12.5.1.5 , there is a dearth of high PT partitioning data to evaluate whether such ideas are required for all of these bodies. However, there are additional problems that linger and must be addressed in any scenario, which calls on the addition of chondritic material to a reduced mantle. Addition of chondritic materials (<1 mass %) to a reduced post core formation mantle will result in the reduction of those materials to a mixture of metal and silicate, and the HSE will be partitioned into the metal and then proceed to travel into the core. HSE will not be mixed efficiently into the mantle because the mantle is too reduced. This is a problem for all three bodies – Earth comes closest to a resolution, but as discussed above the oxidation mechanisms for Earth are few and perhaps not robust enough to cause the required oxidation. Scenarios involving volatile species such as H2O and CO2 in the mantle offer the most promise, but quantitative models calibrated over a wide PT range are lacking. Even if the mantle can somehow be oxidized enough that it dissolves HSE, O-bearing sulfide melt will be a host phase for the HSE and would be very mobile such that HSE would be mobilized out of the upper mantle anyway; the addition of oxidized chondritic material during the later stages of the accretion of the Earth could easily have facilitated the segregation of core-forming material by porous flow if temperatures were in excess of the sulfide solidus (Gaetani and Grove, 1999).

One possible solution lies in the oxidizing capacity of a thick atmosphere. Magnesioferrite spinels form during impacts of meteoritic material with the Earth – they form out of the oxidized vapor plume that is created during the impact (Ebel and Grossman, 2005). Such spinels are also capable of hosting significant amounts of many HSE (Ru, Rh, Re, Ir, Os; Righter and Downs, 2001). If the early Earth had an oxidized atmosphere (e.g. Trail et al., 2011) and some HSE were oxidized and condensed in magnesioferrite in an impact event, the HSE could be delivered to a oxidized mantle in an oxidized form (Righter, 2005). They would then have to be mixed efficiently into the primitive mantle. Such a mechanism may work for a subset of the HSE and for the Earth, but it is not a likely mechanism for the Moon or Mars.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780080959757002102