4  Chemical Composition

Part 1: Chemical composition of the meteorites (4.1 - 4.7)
We use the chemical composition of meteorites to learn about the formation and evolution of the protoplanetary disk, from which our planetary system formed. The composition of the meteorites is the result of the composition of the meteorite parent body formation region and subsequent processes, which primarily occurred on the meteorite parent bodies. The meteorite composition is one way of meteorite classification, which means, this classification contains information about the meteorite formation regions. Learn about the chemical composition of meteorites and what we can learn from this about their formation and evolution.
Learning Goals
Explain the key importance of the CI-chondrites for cosmo- but also geochemistry. Recall the meaning of ›cosmochemical abundance of the elements‹, and describe this. Name the 10 most abundant elements, their relative abundance and cosmochemical character (-> needs to be known). Generally illustrate the element abundance of the various meteorite groups. Relate the meteorite compositions to their formation. Describe the meteorite volatile depletion and give mechanisms how these formed.

Part 2: Chemical fractionation trends in meteorites, Earth, ISM and the galaxy (4.8 - 4.13)
The chemical composition of Meteorites, the Earth, etc. are the result of their initial composition, their various formation processes, and their subsequent evolution. Their composition is thereby changed on a large scale – the entire Earth, meteorites in general, the ISM, etc. –, but also on small scales, e.g., the composition of individual minerals. Learn about the composition of meteorites, Earth, ISM, and the galaxy on a large scale. Inform yourself about their fractionation trends, and how these were established.
Learning Goals
Explain why Earth and meteorites have different Mg/Si- and Al/Si/ratios. Give reasons for these differences. Explain why highly siderophile elements (HSE) are higher in Earth’s mantle than expected. Argue why the refractory elements are depleted in the ISM-gas. Provide evidence why our solar system has a typical Mg/Si-ratio. Explain why the Mg/Si-ratio of other stars is evidence for fractional condensation (-> needs to be known) in other protoplanetary disks, but likely also the protoplanetary disk from which our planetary system was born. Describe how the chemical composition of our galaxy evolved as a whole, and provide reasons for this.

Bonus 1: Presenting elements and their compositions (3.7 - 3.12)
The element composition of meteorites and their components are presented and determined in various ways. Learn and understand two of the frequently used techniques: normalisation und modal recombination. Learning Goals
Confidently apply normalisation to e.g., present element compositions of bulk meteorites in category plots. Confidently apply modal recombination to e.g., calculate bulk element compositions of chondrules.

Bonus 2: Presenting isotopes and their compositions (4.13 - 3.12)
Our Solar System had an eventful start. More and more extra-solar solar- and planetary systems are discovered, in various stages of their early formation. Using this information and together with the much more and much more detailed information from meteorites, it is then possible to reconstruct the formation and evolutionary phases of our Solar System. Describe the events in the first millions of years when planetary systems in general, and our Solar System in particular form and formed, respectively. Learning Goals
Explain why isotopes are presented as ratios and not absolute values. Give arguments why isotope ratios are presented relative to a standard. Confident usage and explanation of the various stable and radioactive isotope notations. Explain the approximation a-1 = Ln(a).

Bonus 3: Isotope composition of chondrules (3.7 - 3.12)
Chondrules have variable bulk isotope compositions, which can be highly variable for the various elements. This mostly depends on the process that fractionated the isotopes. The bulk isotope compositions of chondrules in a single chondrite often show a distinct distribution. Learn about bulk chondrule isotope compositions and distributions, and what can be learned from these about chondrule formation conditions and regions. Learning Goals
Recall bulk chondrule isotope intervals of important elements. Relate the width of such intervals to the mechanism of isotope fractionation processes. Describe the distribution of bulk chondrule isotope compositions. Use the bulk chondrule isotope compositions as evidence for their formation from a common reservoir.

4.1 The fundamental Element-Correlation Solar Photosphere vs. CI Chondrites

This most fundamental plot illustrates the pivotal importance of meteoritics. Meteorites, and in particular CI chondrites have essentially the same composition than the Sun, except for the He, He and a number of volatile elements. The Sun concentrates >99.9% of the mass of the Solar System and should therefore compositionally be representative of the entire Solar System. As CI chondrites have the same composition as the Sun, so are they compositionally representative of the Solar System. This allows to study the composition of the Solar System in high precision in the laboratory. CI chondrites are therefore also the standard reference for studying geochemical processes.

Volatile and/or inert elements: H, He, N, C, O, noble gases.

They have the almost similar composition as the Sun. As the Sun concentrates more than 99.9% of the mass of the solar system, the Sun’s composition should be representative for the entire Solar System – and then so should the CI chondrites.

✗… it is produced in the Sun and therefore apparently depleted in meteorites.

✗… it is consumed in the Sun and therefore apparently depleted in meteorites.

✗… it is produced in the Sun and therefore apparently over-abundant in meteorites.

✓… it is consumed in the Sun and therefore apparently over-abundant in meteorites.

✓True

✗False

✗The chromosphere

✓The photosphere

✗The protuberances

✗The corona

4.2 CI Chondrite Element Abundances

The most abundant element is O. This is followed by the 3 about similar abundant elements Mg, Si and Fe, which are closely followed by S. About an order of magnitude below are Al, Ca and Ni, which is why CAIs are rare. Only a little less abundant as these are Na, Cr and Mn. All these elements are represented in the abundant silicates, metal, sulphide and the CAI-minerals. The REE are 6 orders of magnitudes below the main elements Mg, Si and Fe, which is why REE element concentrations are given in ppm.

O, Mg, Si, Fe, S. About an order of magnitude below Mg/Si/Fe: Al, Ca, Ni. Then e.g., Na, Cr, N. Hydrogen is commonly not mentioned, but only a little rarer than S.

P, Cl, K, Ti, Mn, Co, Zn, Cu.

✗Because it is normalised to Mg and CV chondrites.

✓Elements with even atomic numbers are more stable.

✗The differences are only because of the log scaling

✗Because only elements with odd neutron numbers are shown.

✓About 6 orders of magnitude

✓About 10(-6)

✗About 10(-6) orders of magnitude

✗About 10(6) orders of magnitude

✗Because these only condense at very high temperatures.

✓Because Ca,Al are about one order of magnitude rarer than Mg/Si/Fe.

✗Because Ca,Al are about two orders of magnitude rarer than Mg/Si/Fe.

✗Because Ca,Al has abut similar abundances than Mg/Si/Fe, bust most of the Ca,Al is concentrated in chondrule mesostases.

4.3 Mean Bulk Element Compositions of the Various Chondrite Groups

An important discriminating characteristic among chondrite groups is their bulk composition. A full set of bulk composition of the various chondrite groups is highly beneficial.

Element concentrations are often presented in normalised form. For this, values need to be divided by each other. It would not be possible to directly compare normalised values to each other, if these were the result from dividing element concentrations with different pseudo-units (i.e. %, ppm, mg/g, etc) by each other. Therefore, the element concentrations are all given in the same pseudo-unit in the table. It is then, actually, possibly to apply any mathematical operation with the element concentrations, e.g., addition (for analysis totals) or multiplication (for modal recombination).

✗In wt%

✓In µg/g

✓In wt-ppm

✗In wt-ppm%

✓True

✗False

4.4 Iron Meteorite Cooling Rates Using Island Widths

Various methods exist to determine the cooling rate of iron meteorites. One method measures the micro-micro-structure of taenite (Ni-rich metal). This taenite decomposes on the µm-scale into various, Ni-rich metals. This produces a cloudy structure with islands set in the centres of a honeycomb structure. The size of these islands correlate negatively with the cooling rate. This means, smaller islands form at high cooling rates, while larger island form at low cooling rates. This method can further be used for other meteorite groups that contain metal.

We can learn about the evolution of the parent body, e.g., its size and cooling time.

0.5 to 500 K/Ma.

✓10-500 nm

✗10-500 µm

✗10-500 mm

✗True

✓False

4.5 Classification and Comparison by Bulk Chondrite Composition

Bulk chondrite compositions are among the criteria to classify chondrite groups. Bulk chondrite compositions further provide insights into characteristics and processes of the reservoirs in which the chondrites formed.

Class: enstatite chondrites; group: EH. R also have quite high Mg/Si ratios. All just mentioned chondrites have about 1.3-1.4 times CI chondritic Mg/Si ratios.

The refractory elements such as Ca and Al are generally enriched in carbonaceous chondrites, between about 1.1 and 1.4 times CI chondritic. The refractories are depleted to about 0.9 times CI chondritic in all other chondrites – i.e., OC, EC, R and K.

✗To CI and Mg

✓To CI and any element

✗To CI

✗To Mg

✗To any element

✓True

✗False

✓CC

✗OC

✗EC

✓K

✗R

4.6 Volatile Element Depletion in Carbonacous Chondrites

Carbonaceous chondrites are to different degrees monotonically depleted in volatile elements. This monoton trend, however, stops at the most volatile elements, forming a plateau. The depletion trend might be explained by incomplete condensation, and the plateau by admixing some CI material

The volatile elements are monotonically depleted. The higher volatile elements have a flat, plateau like pattern, i.e., these are depleted to the same extent relative to CI.

Either by incomplete condensation or evaporation from an initially CI like reservoir.

✗True

✓False

✗True

✓False

✓True

✗False

4.7 The Urey-Craig Diagram

The various chondrite groups are oxidised/reduced to various degrees, representing variable O fugacities in their respective formation regions. Changes along the x-axis represent these variable O fugacities, while changes along the y-axis in fact represent variable metal abundances. Enstatite chondrites are the most reduced, while CM and CI chondrites are the most oxidised and metal-poor. Carbonaceous chondrites are generally more oxidised and metal-poor. Ordinary chondrites have anti-correlated metal abundance and oxidation-degree, i.e., silicates in LL chondrites are richest in Fe, although LL chondrites have the lowest bulk Fe – and vice versa in H chondrites.

4.8 Cosmochemical Fractionation Trend and Terrestrial Array

An intriguing combination of two trends is seen when Al/Si is plotted vs. Mg/Si. Terrestrial rocks mainly fall along a trend with a negative slope. This is explained by fractional crystallisation: melts are enriched in Si and Al and plot at low Mg/Si and high Al/Si, while residual rocks have complementary ratios. This negative correlation is called the ›terrestrial array‹. Bulk chondrites, on the contrary, fall on a straight positive correlation in this plot, i.e., Mg and Al de/increase together. This is called the ›cosmochemical fractionation trend‹. CI chondrites and the Sun plot close together on the cosmochemical fractionation trend. Curiously, the bulk silicate Earth (BSE) plots almost exactly on the intersection of the terrestrial array and the cosmochemical fractionation trend. The BSE composition is, however, deficient in Si when compared to the Sun. One explanation might be a significant fraction of Si (up to several wt%) in the Earth’s core. Alternatively, the Earth was formed from a Si-poor reservoir. A further observation is that carbonaceous chondrites (CC) have similar Mg/Si ratios, but vary in Al/Si ratios, likely due to various Ca,Al-rich inclusion (CAI) abundances. These CAIs were likely added later to the CC. In general, the cosmochemical fraction trend seems to indicate variations in Si among the various planetary materials. The origin of this variation is, however, still unclear.

Always: Mg/Si, Al/Si. Sun: 0.91, 0.09; BSE: 1.09, 0.11; spinel peridotites: 1.1 to 1.4, 0.02 to 0.1 differentiated rocks: 0 to 0.2; 0.2 to 0.45; chondrites: 0.55 to 0.9, 0.45 to 0.1.

The BSE has a higher Mg/Si ratio than the Sun, i.e., the BSE is depleted in Si relative to the Sun. The currently most popular explanation are several wt% Si in the Earth’s core, i.e., the Earth likely started with the same Mg/Si ratio as the Sun, but a significant fraction of Si was incorporated in the Earth’s core during the initial differentiation of the Earth.

✗… removal of forsterite

✗… addition of forsterite

✗… removal of CAIs

✓… addition of CAIs

✓higher

✗approximately similar

✗lower

✗EC – OC – CC

✗CC – EC – OC

✗EC – CC – OC

✗OC – CC – EC

✓CC – OC – EC

✗OC – EC – CC

4.9 Bulk Silicate Earth Element Composition and Depletion

Bulk silicate Earth (BSE) is the composition of the Earth minus its metal core. Lithophile refractory and main elements up to about Mg have CI chondritic normalised concentrations. Elements with lower T50% condensation temperatures than Mg are increasingly depleted in the BSE. Siderophile refractory and main elements abundances are depleted relative to CI (and Mg). The degree of depletion depends on the partition of the according elements between core and mantle, i.e., between metal and silicate. This partition depends on the pressure, O-fugacity and changing valence state of some of these elements, which, of course, also depend of the ambient conditions such as pressure or O-fugacity. All highly siderophile elements (HSE) are solely expected in the metal core, i.e., their normalised abundances should essentially be 0. This is not the case, and although the normalised HSE abundances are still very low in BSE, this cannot be the remnant from Earth’s differentiation into core and mantle. It is therefore commonly suggested that a late and large meteorite impact (called ›late veneer‹) with CI chondritic composition dissolved in the Earth’s mantle, thereby adding and enriching it in HSE. The volatile depletion trend is generally independent of the litho-/sidero-/chalco-/atmophile character of an element. This could be explained by a general volatile depletion of the reservoir from which Earth formed. However, individual elements (e.g., In, Pb) or groups of elements (e.g., S/Se/Te) deviate from the volatile depletion. These deviations are not yet fully understood, but likely as well related to ambient and/or varying element states such as pressure, O-fugacity or valence.

The highly siderophile elements are less depleted in the BSE than expected from their partition coefficients. These predict that the HSE should essentially be completely concentrated in the Earth’s core. It is therefore suggested that a large asteroid hit the Earth after it differentiated into core and mantle. This asteroid contained solar siderophile element abundances, which were dissolved in the Earth’s mantle, thereby increasing its siderophile element budget. This late asteroid collision and its consequences is called the ›late veneer‹ event.

The material from which Earth accreted likely formed in a reservoir with an elevated ambient temperature. In this case, the volatile elements remain to some degree – depending on their specific 50% condensation temperatures – in the gas phase. This means, the gas phase in the region in which the building blocks of Earth formed was likely enriched in volatile elements.

✓True

✗False

✗After the refractory elements

✓About in the middle of the main elements

✗After the main elements

✗Shortly after the volatile elements

✓True

✗False

4.10 Element Abundances in the ISM Gas

The ISM gas phase is enriched in volatile elements and depleted in refractory elements. The refractory elements likely reside in grains that condensed from the gas phase. This process likely explains the refractory rich and volatile depleted patterns of many bulk chondrites, but also chondrules, which represent a high-temperature component of chondrites.

The ISM gas phase is rich in volatile and depleted in refractory elements. This indicates incomplete condensation of the elements. This means, incomplete condensation might be a valid explanation for the volatile depleted patterns seen in bulk chondrites.

There is no difference. In fact, their similarity is a strong indication for incomplete condensation, as otherwise it might be rather explained by fractional condensation of some particular mineral phases.

✗… the same element pattern as the ISM

✓… the complementary element pattern as the ISM

✓… refractory elements

✗… volatile elements

✗True

✓False

4.11 Mg-Si Ratios of Stars in the Vicinity

Mg and Si are two main elements in our solar system. The Mg/Si ratios of stars in the vicinity of our solar system is close to the Mg/Si ratio of the Sun, which is about 1.05 (atom-ratio). Components that formed in the protoplanetary disk, such as chondrules, should therefore only contain olivine and pyroxene. Some addition of Fe to these minerals would further increase the olivine/pyroxene ratio. No silica is expected. The silica observed in chondrules, must therefore represent the result of a fractionation process. Observations of protoplanetary disks have also shown the presence of silica. As their Mg/Si ratio is similar to our solar system, the same argument can be made, i.e., the observed silica must represent the result of a fractionation process. Hence, it appears similar process formed the first solids of various protoplanetary disks, including our own.

Most stars have Mg/Si ratios in the range of about 0.8 and 1.2, quite similar to our Sun.

The atomic Mg/Si ratio of forsterite is 2, of enstatite is 1, and of the solar system about 1.05. During equilibrium condensation, all Si should be consumed by enstatite and a little forsterite. No Si is left to form silica. Silica can, however, form during fractional condensation. In this case, the condensing forsterite produces a SiO-rich gas. As then not all forsterite reacts with this SiO to enstatite, some SiO is left over to produce silica. The mineralogical zonation in some chondrules with the sequence from core to rim with olivine-pyroxene-silca, and in many chondrules with the sequence olivine-pyroxene supports this scenario. As stars in the vicinity of the Sun have similar Mg/Si ratios as our Sun, and because silica has been observed in protoplanetary disks, it can be concluded that fractional condensation also occurs in these protoplanetary disks. This suggests that the process of protoplanetary disk evolution is similar among protoplanetary disks – including the one from which our Solar System formed.

✗… an indicator for its size.

✓… an indicator for its age.

✗… an indicator for its brightness.

✗… a measure of the star’s log(Mg/H) ratios to the Sun’s log(Mg/H) ratio

✓… a measure of the star’s log(Fe/H) ratios to the Sun’s log(Fe/H) ratio

✗… a measure of the star’s log(Si/H) ratios to the Sun’s log(Si/H) ratio

✗1.5

✓1.05

✗0.5

✗2

✗0.05

✗… equilibrium condensation

✓… fractional condensation

✗… thermal condensation

4.12 Normalisation

Normalising data is essentially calculating a ratio of the analytical data, divided by standard data, e.g., CI bulk data. This is a single normalisation. In a double normalisation two ratios are calculated: the first step is the same as in a single normalisation. Then, the result is further divided by e.g., one element, i.e., the normalised result from the first step of this element. Typical double normalisations are: ›normalised to CI and Mg‹, but also ›Ca/Al ratios normalised to Mg‹, i.e.. ›Ca/Al/Mg‹. It is mandatory to add what unit the data had that were used for the normalisation, e.g.,wt% or atom%. As the normalisation is always a ratio, the designation would be e.g., mass-ratio or atom-ratio.

It is the simple division of the element concentration(s) in a sample by the element concentration(s) in a standard (e.g., CI chondrites).

These are two simple divisions. First, the element concentration(s) in a sample is/are divided by the element concentration(s) in a standard (e.g., CI chondrites). Second, the concentrations of all elements in the sample, but als in the standard are divided by the element concentration of one element (e.g., Mg) in the sample and standard, respectively. The first and second step can be inverted.

✓sample normalised to CI and Mg

✓sample/CI/Mg

✗CI/sample/Mg

✗True

✓False

✗True

✓False

4.14 Isotope Notation

Isotopes are presented as ratios, so these can be compared among samples. The most common notation is: delta=((Rsmp/Rstd)-1)*1000, with Rsmp being an isotope ratio in the sample and Rstd being the isotope ratio in a standard.

Isotopes are not measured as ratios. Each isotope is measured individually. The isotope ratio is then calculated. Isotopes are presented as ratios, as absolute concentrations are highly impractical and not directly comparable. The absolute isotope concentration in a sample not only depends on the relative concentrations of the various isotopes of an element. The absolute isotope concentration in a sample also depends on the concentration of the corresponding element. Hence, for a sensible comparison of absolute isotope concentrations, it would be required to also know the concentration of the corresponding element. And even if this were the case, this would introduce the additional error of the element concentration measurement. The isotope ratio of a sample is, in contrast, independent of the concentration of the corresponding element. Hence, it is sensible to present isotope compositions as isotope ratios.

When the sample and standard have identical compositions and are divided by each other, the result is 1. Intuitively, if we compare two things and there is no difference, we might think something like ›there is zero difference between what I am comparing‹. To mimic this thought, 1 is subtracted, which means, if the sample and standard are identical, the isotope composition of the sample is 0.

Capital Delta essentially designates the y-axis intercept of a fractionation line parallel to a reference fractionation line (e.g., the TFL).

The heavier isotope. For example, in the oxygen 3-isotope plot deltaO18/16 is on the x-axis and deltaO17/16 on the y-axis.

✓+1.34 ‰

✗2.21 ‰

✓+0.47

✓-3.76 ‰

✓-1.34 ppm

✗5.19 ‰

✓True

✗False

✗delta56Fe

✓delta88/86Sr

✗epsilon54Cr

✗True

✓False

✓positive

✗negative

4.15 The approximation a-1 = Ln(a)

This approximation is true for values close to either 0 or 1, depending on how the approximation is formulated. This approximation is used in the context of e.g., isotope variations or fractionation factors, which have values close to either 0 or 1.

4.16 Isotope Notation – Capital Delta

Capital Delta designates the distance of a sample from a reference fractionation line in a 3-isotope plot, e.g., from the terrestrial fractionation line. Multiple samples with the same capital delta then plot along a line parallel to a reference fractionation line – which, of course, then has the distance capital delta from the reference fractionation line.

Capital Delta essentially designates the y-axis intercept of a fractionation line parallel to a reference fractionation line (e.g., the TFL).

The heavier isotope. For example, in the oxygen 3-isotope plot deltaO18/16 is on the x-axis and deltaO17/16 on the y-axis.

✗True

✓False

✓positive

✗negative

✗capitalDelta = deltaO18/16 + 0.522 deltaO17/16

✗capitalDelta = deltaO17/16 - 0.522 deltaO17/16

✓capitalDelta = deltaO17/16 - 0.522 deltaO18/16

4.17 Isotope Notation of Radioactive Elements

Isotope compositions are given relative to a reference isotope composition. Such reference isotopes are often the composition of the Earth or CI chondrites. When the considered isotope is a radioactive system, the reference isotope composition is not fixed, but changes over time. The composition of a sample is then always calculated relative to the current composition of the reference. This means, the reference value is not a fixed value, but the value at a certain time. This is unproblematic for present day samples, which are also referenced to the today’s composition of the reference isotope. But this change is critical to consider when modelling the evolution of a reservoir. In this case the composition of the modelled sample is always referenced to the reference isotope composition at the specific time. This isotope notation of radioactive elements is also the reason why the value of the reference is always 0.

The studied radiogenic isotope ratio in a sample consists of the radiogenic isotope (isoRadio) divided by a stable isotope (isoStab). Both isotopes belong to the same element. This studied isotope ratio is divided by a standard also defined by the same isoRadio and isoStab. This means, this standard isotope ratio also changes over time. To calculate the isotope notation, the isotope ratio of the sample is divided by the isotope ratio in the standard at the time as the sample formed. This means, it is necessary to know the age of the sample, and this age is then used to calculate the isotope ratio of the standard. This calculation is done by using the solar system initial ratio of the isotope ratio, and then calculating how much radiogenic material is added up to the age of the studied sample.

Simply because the isotope notation of a standard is always 0, as therefore standard is divided by the standard. In a radioactive system the radiogenic standard isotope ratio increases over time, but still at each time the increased standard isotope ratio is divided by exactly that increased standard isotope ratio, meaning it always will result in 0 for the isotope notation.

✗parent

✓daughter

✓True

✗False

✗True

✓False

4.18 Bulk Chondrule Stable Isotope Compositions

Chondrule populations of individual chondrites vary in their isotopic compositions. Reported relative deviations of chondrules from standards for the light isotopes up to 18O – and including 41K – range between about -25 to +20‰, and between -134 to +164‰ in the case of 15N. Reported values for all other stable isotope systems are significantly lower, typically ranging between -2 and +1‰, with two chondrules reaching almost -4‰ (88Sr and 114Cd). Isotope data plotting on mass fractionation trends are most likely the result of mass dependent processes. In contrast, isotopes that plot off their mass fractionation trends are in most cases (except for e.g., O isotopes) the result of mixing nucleosynthetic component to the chondrules (-> ›mass independent fractionation‹).

Mostly only a few ±permill. Lighter elements might be in the range of ±percent to – quite rarely – tens of ±percent.

Mass-dependent and mass-independent processes.

✓kinetic

✗mixing

✓equilibrium

✓mass

✗size

✗addition of presolar grains

✗mass

✗size

✗volatility

✗llithophile/siderophile

✗kinetic

✓mixing

✗equilibrium

✗mass

✗size

✓addition of presolar grains

✓small mass

✗small size

✓volatile

✗llithophile