Skip to main content Accessibility help



  • Access

  • Remote Compositional Analysis
  • Techniques for Understanding Spectroscopy, Mineralogy, and Geochemistry of Planetary Surfaces
  • Online publication date: November 2019
  • pp 3-20



      • Send chapter to Kindle

        To send this chapter to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle.

        Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Available formats

        Send chapter to Dropbox

        To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Dropbox.

        Available formats

        Send chapter to Google Drive

        To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Google Drive.

        Available formats
Export citation


A variety of features in the visible and near-infrared regions that are observed in remote sensing applications are the result of electronic transitions, typically involving cations of transition metals, most commonly Fe and Ti, or the molecular species S. The position and intensity of these features are sensitive not only to the particular cation, but also to its oxidation state, the particular phase in which it occurs, the geometric structure of the site that it occupies, and interactions between and among neighboring cations. Often these features are diagnostic for the host mineral.

1 Electronic Spectra of Minerals in the Visible and Near-Infrared Regions

George R. Rossman and Bethany L. Ehlmann

1.1 Origin of Electronic Spectra of Minerals

Many of the spectral features of minerals in the visible to near-infrared region (VNIR; defined here as ~0.4–2.5 µm) arise from electronic transitions within and between transition elements and the anions chemically bound to them. Thousands of minerals have color or wavelength-variable properties in this portion of the spectrum. Metal ions including vanadium, chromium, manganese, iron, cobalt, nickel, and copper, usually in either the 2+ or 3+ oxidation state, are responsible for the color of many minerals. However, only a few of these elements, typically iron, titanium, and oxygen, are important in most remote sensing applications of rocky bodies. Many features arise from electronic transitions of electrons between the d orbitals of a metal ion, while some spectroscopic features arise from interactions between atoms.

1.2 Units

Wavelengths are commonly expressed in nanometers (nm) or micrometers (μm) and, in older literature, Ångstrom units. Literature on mineral spectroscopy and mineral chemistry often uses nm, while the remote sensing literature typically uses µm. The conversion among them is:

1000 nm = 1 μm = 10,000 Å. (1.1)

The spectrum can also be presented in energy units, usually wavenumbers, which are the reciprocal of the wavelength, and are usually expressed in reciprocal cm. The advantage of wavenumbers is that absorptions are symmetrical in energy coordinates but not in wavelength coordinates. Spectroscopic energies can also be expressed in electron volts, but this is more commonly encountered in the physics literature.

Wavenumbers (cm1)=10000000/nm = 10000 / μm)1000 nm= 1μm = 10,000 cm1;    400 nm = 0.4 μm = 25,000 cm11cm1= 1.23984×104eV;     8065.54 cm1= 1 eV = 1239.8 nm.(1.2)

Spectra are usually displayed in either reflectance units or absorbance units. Reflectance spectra must be taken in comparison to a standard. In a laboratory setting, the standard can be a colorless polytetrafluoroethylene-based plastic such as Spectralon®, aluminum, or, in the near-IR (NIR) region, gold. Spectra are presented as the percent (0–100%) or fraction (0–1) of sample reflectance relative to the standard versus wavelength. In spacecraft applications, the comparison standard is typically the solar flux. Reflected light data collected by spacecraft are typically expressed as I/F [radiance/(solar irradiance/π)]. For a given viewing geometry, these data can be further corrected for angular dependencies in scattering properties (see Hapke, 1981, for definitions of different types of reflectance).

Even though remotely obtained spectra are the composite response of many components in a field of view, many fundamental studies of mineral spectra are conducted with single crystals. In chemistry, such studies are usually presented in absorbance units, where

Absorbance = ‒log10(Transmission). (1.3)

An absorbance of 1 means that 10% of the incident light is passing through the crystal; an absorbance of 2 means that 1% of the light passes through. Beer’s law formulations (I = I0e–αd) are also sometimes used to derive an absorption coefficient (α) where the intensity of initial light (I0) is compared to the intensity after transmission (I) through a given thickness of material (d).

Because most mineral crystals are anisotropic, fundamental studies of single crystals usually measure the spectrum with polarized light vibrating along the fundamental optical directions of the crystal. The refractive indices of a crystal for light traveling in different directions relative to the crystal axes form an optical indicatrix, mathematically, an ellipsoidal surface. Crystals that belong to the orthorhombic, monoclinic, and triclinic crystal systems will have three independent spectra that can display very different absorption properties (biaxial indicatrix). Crystals in the tetragonal and hexagonal systems will have two different spectra (uniaxial indicatrix), while isotropic, cubic crystals will have only one spectrum (spherical indicatrix). In general, spectra can be named either according to the crystal axes in which the vibration occurs (e.g., E\\c or the c-spectrum) or by the symbol for the index of refraction that would be measured in the vibration direction. For biaxial crystals with three independent optical directions we have the α, β, and γ spectra (also called the X, Y, and Z spectra). For uniaxial crystals with two different spectra there are two independent optical orientations: the E⊥c direction, also called the ω-spectrum, and the E\\c direction, which is also called the ε-spectrum.

1.3 Crystal Field Transitions

The spectra of metal ions, particularly those of first-row transition elements, Ti through Cu, are often interpreted with the use of Crystal Field Theory. The d-orbital electrons are the valence (outermost) electrons in the case of these metals. For an isolated transition metal ion, electrons occupy any d orbital with equal probability. However, in a mineral, electrostatic fields produced by the anions (usually oxygen) surrounding the central metal ion separate the metal ion’s d orbitals into different energy levels. This allows the d-orbital electrons to undergo transitions between orbital energy states. Their transitions to different energy levels under the influence of VNIR light give rise to much of the color we see and the spectra we measure of minerals.

This can be understood for the case of a metal ion surrounded by six oxide ions (ligands) arranged in perfect octahedral symmetry. If the metal ion were floating in free space with no oxide or other anions near it, all five orbitals in the 3d level would have the same energy (Figure 1.1a). But when the metal ion is in an octahedral arrangement of oxide ions, the d orbitals split into groups of two different energies (Figure 1.1b) reflecting the different interactions the d-orbital electron clouds have with the surrounding ligands.

Figure 1.1 Energy diagram for 3d orbitals and their electron probability clouds. (a) Orbitals in free space. (b) Orbitals in an octahedron of oxide ions. (c) The electron clouds of the d orbitals in relationship to the oxide ions in an octahedral arrangement.

Iron in the 2+ oxidation state has six electrons, the valence electrons, in the 3d orbital. In an octahedral coordination environment, these go into the 3d orbitals as pictured in Figure 1.2a because the electrons are energetically more stable when pairing of electrons is minimized. An electronic transition will occur when light of an appropriate energy interacts with the Fe2+ ion and promotes an electron from a lower energy orbital to a higher energy orbital (Figure 1.2b). Each configuration of the electrons is an electronic state of the system. If the total number of unpaired electrons is not changed during the transition from the electronic ground state to a higher energy state, this is called a spin-allowed transition. If the total number of unpaired electrons changes, this transition is called a spin-forbidden transition because such a transition is about 100 times less likely to occur than a spin-allowed one.

Figure 1.2 Electron configurations for Fe2+. (a) The ground state in an octahedral coordination environment. (b) The spin-allowed excited state that gives rise to the primary NIR absorption bands. Here, the total number of unpaired electrons has not changed in the excited state. (c) A spin-forbidden state in which the total number of unpaired electrons has changed in the electronic excitation. A comparison of the relative splitting of ground state d orbitals for Fe2+ ion in (d) regular octahedral, (e) regular tetrahedral, and (f) a representative distorted coordination.

The intensities of electronic bands relate to their spin-allowed or spin-forbidden properties. Spin-forbidden transitions produce absorption bands that are commonly much weaker than the spin-allowed bands. However, interactions between cations, as explained in a following section, can dramatically increase the intensity of formally spin-forbidden bands and produce other features of high intensity when cations in different oxidation states interact. In addition, electronic transitions that involve transfer of charge from anions to cations (Sections 1.4 and 1.5) can also be of much higher intensity, but are usually centered in the ultraviolet portion of the spectrum.

Qualitative predictions of the spectrum of metal ion complexes can be obtained from Tanabe–Sugano diagrams. These diagrams usually present energy states for complexes in ideal, octahedral coordination that can be used to interpret the number of spin-allowed and spin-forbidden absorption bands and their widths, and, with suitable experimental parameters, can provide predictions of where bands will occur. Most ions in minerals are not in ideal octahedral coordination, so these diagrams often do not accurately interpret mineral spectra, but they do indicate which absorption bands will split into multiple components for metal ions in crystal sites of low symmetry. These diagrams, along with other concepts previously discussed, are reviewed in more detail in several books and articles about mineral spectroscopy (Karr, 1975; Burns, 1993; Rossman, 2014).

Another important factor in determining the number and wavelengths of absorption bands from a metal ion is the symmetry and distances of the ions surrounding the central metal ion. The number and energies of absorption bands strongly depend on the symmetry (Figure 1.2d–f). In a perfectly regular octahedron, Fe2+ will have one possible transition from the lower to the higher set of orbitals (Figure 1.2b). In a perfectly regular tetrahedral coordination environment, the energy difference between the orbitals will be smaller; consequently, the absorption will occur at longer wavelengths, but still with only one absorption band. However, coordination environments of ideal symmetry are almost never encountered. In nearly all minerals, the metal ion is in a coordination environment distorted from ideal symmetry. In such cases, the energies of the orbitals will split and multiple absorption bands will be possible. This fact is crucial for understanding the relatively broad nature of absorption bands in spectra of common rock-forming minerals. For example, in olivine the broad Fe-related electronic absorption observed is, in reality, a set of overlapping absorptions, caused by the existence of numerous 6-coordinated sites of different dimensions and symmetries that occur as the atoms around the iron vibrate due to thermal energy. In pyroxene, the wavelength of the Fe-related electronic absorption in the distorted 6-coordinated M(2) site shifts systematically with Ca, Fe, and Mg substitution that changes the dimensions of the octahedral site.

Group theory provides symbolic names for each of the electronic states of the system. These names convey the spin state, degeneracy, and symmetry of the electron cloud (for further reading see Harris & Bertolucci, 1989; Cotton, 1990). For example, in a perfectly octahedral coordination environment, the Fe2+ ground state has a 5T2g symmetry designation, where the T indicates that the state is triply degenerate, the 5 is the number of unpaired electrons +1, and the 2g relates to the symmetry of the electron cloud. The excited state has a designation 5Eg, where the E symbolizes a doubly degenerate state. A single absorption band occurs when the electron is promoted from the 5T2g state to the 5Eg state. In coordination environments of lower symmetry, the T state can split into three different electronic states and the E state can split into two, each with a different energy. In orthopyroxene, the electronic ground state of Fe2+ in the low symmetry M(2) site splits into three different states labeled 5A1, 5A2, and 5B2, and the excited state splits into a 5B1 and a 5A1 state, each of which is no longer degenerate (Goldman & Rossman, 1977).

Electronic absorption bands can be temperature sensitive. They typically broaden at higher temperatures and sharpen at lower temperatures. Fundamental studies of minerals and chemicals are often conducted at liquid nitrogen or even liquid helium temperatures to sharpen absorptions and allow determination of band centers at high spectroscopic resolution. Particularly for targets below ~150 K, consideration of shifts may be relevant in interpretation of remotely collected spectra.

Absorptions can also shift position or change intensity as mineral sites are distorted and metal–oxygen bond distances change at elevated temperatures (e.g., Aronson et al., 1970; Sung et al., 1977). High-temperature spectra are important in planetary science for interpreting the composition of bodies that are several hundreds of degrees warmer than Earth such as Mercury, Venus, and lavas on Jupiter’s moon, Io.

1.4 Oxygen-to-Metal Charge Transfer

Another common feature in the spectra of many minerals is the oxygen-to-metal charge transfer transition. This feature arises from absorption of photons with enough energy to transfer charge density from an oxygen ligand to the central metal ion. Oxygen-to-iron charge transfer is most commonly encountered in common rock-forming minerals where the band is usually centered in the ultraviolet region. The higher the charge state of the central metal ion, the lower the energy of the absorption band will be. Oxygen-to-Fe3+ charge transfer bands sometimes tail into the visible portion of the spectrum, where they absorb in violet and blue and often produce a rusty orange-red color. Oxygen-to-metal charge transfer absorptions are normally much more intense than those arising from transitions within the d orbitals of metal ions.

1.5 Intervalence Charge Transfer

Intervalence Charge Transfer (IVCT) refers to a process in which two metal ions in close proximity to each other in a structure transfer an electron between them, thereby temporarily changing the oxidation state of both cations. Absorption bands in the optical spectrum from IVCT can be comparatively intense, and only a little IVCT produces spectroscopic features and color in the visible spectral region. In the geological world, only two such interactions are commonly encountered: Fe2+–Fe3+ and Fe2+–Ti4+. A third, Ti3+–Ti4+, is occasionally found in meteorites.

For these interactions to occur, cations need to be adjacent to each other in the mineral structure, often sharing a common edge or face of the coordination polyhedron. Both Fe2+–Fe3+ and Fe2+–Ti4+ IVCT are particularly common in terrestrial minerals such as micas, pyroxenes, amphiboles, and tourmalines and are the origin of the dark color of many minerals including magnetite and ilmenite. Fe2+–Fe3+ IVCT in sites of near-octahedral coordination is found in the 630–820 nm region. Fe2+–Ti4+ IVCT (Figure 1.3a) is typically found in the 425–460 nm region for 6-coordinated near-octahedral cations such as in pyroxenes (Mao et al., 1977; Mattson & Rossman, 1988). The Ti3+–Ti4+ IVCT, observed in pyroxenes and hibonite from meteorites (Dowty & Clark, 1973; Burns & Vaughn, 1975), occurs near 690 nm in meteoritic hibonite from Murchison (Rossman, 2019).

Figure 1.3 Transmission spectra. (a) Clinopyroxene from the Angra dos Reis meteorite showing Fe2+–Ti4+ IVCT near 480 nm and the Fe2+ features near 1000 and 1200 nm discussed in Section 1.6. (Modified from Mao et al., 1977.) (b) A 200 μm thick augite crystal showing the absorption bands from Fe2+ in the geometrically distorted M(2) site near 1000 and 2400 nm, and the weaker bands from Fe2+ in the nearly octahedral M(1) site near 970 and 1200 nm. Weak absorption from Cr3+ appears near 450 and 650 nm. (c) A 200 μm thick diopside crystal showing comparatively weak absorption bands from Fe2+ in the geometrically distorted M(2) site near 1000 and 2400 nm, and the stronger bands from Fe2+ in the nearly octahedral M(1) site near 1000 and 1200 nm. Absorption near 800 nm arises from Fe2+–Fe3+ IVCT.

In a number of terrestrial minerals adjacent sites may have different coordination polyhedra including edge-shared octahedra and tetrahedra in cordierite or edge-shared octahedra and distorted cubes in garnets. In these cases, the wavelengths of the IVCT bands will differ from those of the edge-shared octahedra. A number of different mineral examples are reviewed in Burns (1981).

1.6 Spectra of Key Minerals

There are currently more than 5400 known mineral phases, but only a small number of them contribute electronic absorptions routinely associated with remotely sensed spectra in the VNIR region. These phases include pyroxenes, olivines, feldspars, iron-bearing layered silicate minerals, and iron oxides. A number of other phases such as iron carbonates, iron sulfates, and other sulfur species are occasionally encountered. While they are only components contributing to the whole spectroscopic signature of an object, these minerals and their electronic absorptions carry important information for revealing the geological history of an object. In this section, we review the spectra of select, important phases. Many examples of the spectra of mineral single crystals with other cations are presented in Rossman (2014).

Iron is the element most commonly causing absorptions in the VNIR spectral region and is responsible for the color of common rock-forming minerals. In the primary igneous minerals, iron is usually found in the 2+ oxidation state, often either in sites that are somewhat distorted from ideal octahedral 6-coordination or in irregular sites of higher coordination number. Frequently, iron occurs in more than one distinct site in the crystal structure of the host mineral. Sulfur species of mixed oxidation state are important on some outer Solar System bodies (e.g., Io: Nash et al., 1980; Carlson et al., 1997). Other metal cations such as V, Cr, Mn, Ni, and Cu are important contributors to the spectra of terrestrial minerals and are responsible for the spectacular colors of many museum-quality minerals. To date, they have not played a significant role in remotely sensed spectra of other planetary bodies.

1.6.1 Pyroxenes

Pyroxenes, (Ca, Mg, Fe)2(Si, Al)2O6, are important minerals in many planetary bodies and are an excellent example of how structural distortion affects spectral properties. The two components of the pyroxene absorption bands of Fe2+ become increasingly separated as the sites become more distorted from octahedral geometry due to cation substitutions. In the case of the pyroxene M(2) site, the two components can be separated by about 1000 nm (Figure 1.3b).

The spectrum of augite in Figure 1.3b, a terrestrial clinopyroxene, shows prominent absorptions at about 1000 nm in the beta polarization and near 2300 nm in the alpha direction. These two bands arise from Fe2+ in the M(2) site of pyroxene, which is highly distorted from an octahedral geometry. Two weaker bands near 970 and 1200 nm are due to Fe2+ in the less distorted M(1) site. Small, sharp spin-forbidden transitions are observed at wavelengths less than 1.0 µm. In contrast, the spectrum of diopside in Figure 1.3c has comparatively little contribution from the M(2) site and is primarily dominated by Fe2+ absorption from the M(1) site.

Pyroxenes are among the most widespread rock-forming minerals in the Solar System. The absorptions caused by Fe2+ in distorted M(1) and M(2) sites can be detected in remote sensing reflectance spectra and related to pyroxene crystal chemistry (Figure 1.4a–c), which in turn can be related to magmatic processes occurring on Solar System bodies. For example, 1 µm and 2 µm absorptions in dark lunar mare terrains were used to establish their volcanic origin and map distinct lava flows (e.g., Pieters, 1978; Staid et al., 2011; Whitten & Head, 2015). Strong pyroxene absorption bands observed for the asteroid Vesta and its family were used to identify it as the parent body for the HED meteorite suite and later mapped with spacecraft data (e.g., McCord et al., 1970; DeSanctis et al., 2012). On Mars, an observed transition from older lavas with low-Ca to younger lavas with high-Ca pyroxenes is inferred to result from thermal evolution of the martian mantle (Mustard et al., 2005; Baratoux et al., 2013). For more reading on pyroxene spectroscopy, see Klima et al. (2011).

Figure 1.4 Reflectance spectra of pyroxenes. (a) Spectra of a variety of pyroxenes from Klima et al. (2011). (b) Pyroxene absorption band positions systematically shift with the crystal chemistry, here represented on a pyroxene quadrilateral. Apices are diopside (CaMgSi2O6), hedenbergite (CaFeSi2O6), enstatite (Mg2Si2O6), and ferrosilite (Fe2Si2O6). (c) These changes have been essential for identifying distinct geologic units with low-Ca and high-Ca pyroxene (LCP and HCP) on Mars (spectra from Mustard et al., 2005), pyroxenes on Vesta (spectra from DeSanctis et al., 2012), and high-Ca pyroxenes in lunar lavas.

(spectra from Pieters, 1986)

1.6.2 The Olivine Series

The spectrum of forsterite provides another example of the role of Fe2+ in two distinct sites in the crystal of (Mg, Fe)2SiO4. Each of the 6-coordinated sites for the metal cations, known as the M(1) and M(2) sites, is significantly distorted from purely octahedral symmetry. Consequently, each site produces a pair of Fe2+ NIR absorption bands, which correspond to the crystal field splitting between the lower energy orbitals and the excited states (Figure 1.5). Because olivine is orthorhombic, the 3 spectra in Figure 1.5a represent polarizations along the a-, b-, and c-axes of the crystal which correspond to the γ, α, and β spectra, respectively. The absorption in the 700–1600-nm region of the forsterite spectrum represents spin-allowed bands of Fe2+. There are absorption bands centered near 830 nm, 1060 nm, 1100 nm, and 1310 nm. The two most intense bands displayed in the γ-spectrum are from iron in the M(2) site. Weak features at less than 800 nm are either spin-forbidden bands of Fe2+ or features from other minor components. The diffuse reflectance spectrum convolves the three spectra, as shown in Figure 1.6. The band positions shift with increasing Fe/(Mg + Fe) ratios. Figure 1.6a compares the spectra of an Mg-poor and an Mg-rich olivine. Olivine spectroscopy is further discussed in Sunshine et al. (1998), Isaacson et al. (2014), and Chapters 4 and 18.

Figure 1.5 Absorption spectra. (a) Olivine (forsterite) from San Carlos, Arizona. (b) Fe-bearing plagioclase feldspar from Lake County, Oregon.

Figure 1.6 Reflectance spectra of iron-containing phases. (a) Spectra of Fe2+ in olivine of two compositions (from Sunshine & Pieters, 1998), plagioclase feldspar (from the NASA/Keck RELAB database at Brown University, spectra by C.M. Pieters), and a spinel (from Cloutis et al., 2004). OMCT = oxygen to metal charge transfer; IVCT = intervalence charge transfer. See text for absorption attributions. (b) Fe(II) and Fe(III) create prominent absorptions in Fe-bearing glasses.

(from Cannon et al., 2017)

1.6.3 Feldspars

Plagioclase feldspars (e.g., CaAl2Si2O8) can have iron substitution and thus an Fe2+ absorption in the NIR region (Figure 1.5b). The dominant absorption centered near 1300 nm arises from Fe2+ in the Ca site, which is significantly distorted from any standard coordination geometry. Features in the 300–500 nm region are from Fe3+ in the Al sites, and features near 3000 nm are from the OH content of the feldspar. The Fe3+ bands are absent in the spectrum of lunar plagioclase returned by the Luna 20 mission (Bell & Mao, 1973; Chapter 18). Plagioclase spectroscopy is further discussed in Cheek (2014).

1.6.4 Spinels

The spinel group minerals of the general formula (XY2O4) are phases that commonly contain Fe2+ in a tetrahedral environment, substituting for Mg. Because there is less electrostatic repulsion from the four oxygen atoms surrounding the iron compared to six oxygen atoms in octahedrally coordinated iron, the lowest energy (longest wavelength) Fe2+ absorption bands occur at lower energies (longer wavelengths) than those from 6-coordinated Fe2+ (Figure 1.7a).

Figure 1.7 Absorption spectra of iron-containing phases. (a) The spectrum of Fe2+ in slightly distorted tetrahedral geometry in a spinel from Tanzania occurs in the 2000–3000 nm NIR region. (b) Spectrum of microparticles of hematite in red jasper. (c) Goethite single crystal spectrum showing two of the three polarization directions. (d) Jarosite spectrum showing strong anisotropy in the Fe3+ absorption bands.

1.6.5 Ferric Oxides

Ferric oxides contain VNIR absorption features due to both crystal field splitting and charge transfer. Hematite (Fe2O3) has a prominent absorption near 860 nm and a shoulder at 630 nm due to Fe3+ in a site of near-octahedral symmetry (Figure 1.7b). Starting at 530 nm, the visible light wing of a strong UV-visible charge transfer of oxygen–Fe3+ dominates the spectrum, making the reflectance very low and obscuring remaining crystal field absorptions. These properties change for nanocrystalline hematite, which has particle sizes <10 nm and is superparamagnetic. When hosted within a neutral matrix, the spectrum of nanocrystalline hematite has a small positive slope in the visible, within the charge transfer band, and lacks a deep 860-nm absorption, presumably because of the broadening of crystal field transitions (Morris et al., 1985). Goethite (Figure 1.7c) and lepidocrocite (FeOOH) are other commonly occurring iron oxides with octahedrally coordinated iron, generating absorptions in different positions. Magnetite (Fe3O4) has a VNIR absorption coefficient more than an order of magnitude higher than other iron oxides due to intervalence charge transfer between ferric and ferrous iron, making it “opaque” in reflectance spectroscopy. It is a very efficient light absorber with low reflectance at all VNIR wavelengths.

Hematite and some other iron oxides/hydrous oxides have anomalously high absorption intensity compared to many other minerals and chemical compounds. Fe3+ has five d-orbital electrons that populate each of the five d orbitals with one electron each. As such, any electronic transition involves a flip in the spin direction and would require two electrons to pair, an energetically unfavorable process, making Fe3+ electronic transitions formally spin forbidden. However, when two adjacent Fe3+ cations share common oxygen ions, antiferromagnetic–magnetic interactions occur between and among the Fe3+ cations that can cause a large-intensity increase of the iron’s electronic transition (Rossman, 1996). Effectively, one cation undergoes an electronic transition with a spin flip while an adjacent ion undergoes just a spin flip. The combined system of two iron ions would not undergo a total spin change in this process resulting, effectively, in a spin-allowed transition. See also reviews in Morris et al. (1985, 2000) and Chapter 4.

1.6.6 Iron Phyllosilicates

Iron-bearing phyllosilicates are a key indicator of water–rock interaction throughout the Solar System and often form via dissolution and reprecipitation reactions or transformation reactions from the iron-bearing silicates described earlier. Examples include iron–magnesium smectites, the most prevalent phyllosilicate on Mars, and smectites, serpentines, and chlorites found on Mars, asteroids, and meteorites. The spectra of these layered silicates have vibrational absorptions in the infrared longward of 1000 nm (see Chapter 4), as well as electronic absorptions (Figure 1.8a). Octahedrally coordinated Fe2+ and Fe3+ generate absorptions at 900 nm and 650 nm, as well as absorptions centered shortward of 500 nm. Mixed valence samples have a broad Fe2+–Fe3+ charge transfer absorption centered near 700 nm. Also, Fe3+ in tetrahedral coordination can generate an absorption near 430 nm (see Sherman & Vergo, 1988). Collectively, these electronic absorptions allow phyllosilicates to be distinguished. Because of the complications of overlapping broad electronic absorptions with those of mafic silicates in samples, in remote sensing data, the sharp vibrational absorptions are often more diagnostic. Nevertheless, the electronic absorptions can be important, e.g., the 700 nm absorption in asteroid spectra inferred to result from iron phyllosilicates (Vilas et al., 1994) and crystalline hematite in select martian sedimentary deposits (e.g., Fraeman et al., 2013).

Figure 1.8 Reflectance spectra of phyllosilicates and sulfates. (a) In serpentine and chlorite, even small amounts of Fe(II) and Fe(III) produce VNIR absorptions (Spectra compiled from the authors’ libraries; Clark et al., 2007; Chemtob et al., 2015.) (b) Fe(II) and Fe(III) create prominent absorptions with centers shortward of 1000 nm in sulfate minerals. Spectra of natrojarosite, NaFe3(SO4)2(OH)6, and natroalunite, NaAl3(SO4)2(OH)6, with 10% substitution of Fe for Al) are from McCollom et al. (2014). Szomolnokite and copiapite spectra are from Clark et al. (2007).

1.6.7 Carbonates

Fe and Mn substitution in the cation site in carbonates creates broad absorptions centered between 0.9 and 1.5 µm (as described in Gaffey, 1985). The absorption positions of Mn2+ and Fe2+ can be distinguished from each other, differ even for the same cation depending on the dimensions of the site (e.g., Fe within an Mg carbonate vs. Fe within a Ca carbonate), and, in the case of iron, exhibit a deep absorption with less than 1% of the transition metal (e.g., Bishop et al., 2013).

1.6.8 Iron Sulfates

To date, ferric and ferrous iron sulfates are found on Mars, Earth, and (rarely) meteorites. Many ferric iron sulfate crystals display large differences in the intensity of light absorption, with light polarized in different directions in the crystal. In particular, when the polarization direction is aligned in the direction of the Fe–OH–Fe bonds, the intensity of the lowest energy band (6A1g4T1g) and sharper (6A1g4A1g,4Eg) band is increased. This is another example of intensity enhancement in antiferromagnetic systems, as discussed earlier for hematite. In the case of most ferric iron sulfates, and unlike hematite, the magnetic interaction occurs only in specific directions.

A practical consequence of this interaction is that because of the intensity enhancement, certain iron sulfates will be more readily detected than phases that do not have such interactions. Examples relevant to Mars include jarosite and magnesiocopiapite. For example, jarosite family minerals, (K, Na, H3O+)Fe3(SO4)2(OH)6, have four measured absorptions related to Fe electronic absorptions: a broad absorption centered at 930 nm (6A1g4T1g), a shoulder near 600 nm (6A1g4T2g), a strong, broad absorption near 500 nm that is the wing of the charge transfer band that is centered in the UV, and a sharp, narrow feature near 433 nm (6A1g4A1g, 4Eg) (Figure 1.7d) (Rossman, 1976; Sherman & Waite, 1985). As is commonly the case for iron in lower-symmetry sites, these absorptions are strongly anisotropic.

Ferrous and ferric sulfates can be distinguished based on a combination of their electronic absorptions in the VNIR as well as vibrational absorptions in the infrared related to OH, H2O, and SO4 (Figure 1.8b; Chapter 4). The spectra of iron sulfate minerals are reviewed in Bishop and Murad (1996), Bishop et al. (2004), Cloutis et al. (2006), Pitman et al. (2014), Lane et al. (2015), Sklute et al. (2015), and Ling et al. (2016).

1.6.9 Other Sulfur Species

Metal sulfides occur in small amounts in silicate rocks and often act as darkening agents. Their presence is difficult to detect in VNIR remote sensing. SO2 frost, native S, and complex S-bearing organics are constituents that occur in sufficient abundances on outer Solar System bodies to be observed in telescopic and spacecraft data. The prominent charge transfer absorptions extend into the UV and have been used to trace volcanic processes on Io and the products of radiolysis and space weathering on icy bodies. Spectra are reviewed in Nash et al. (1980). Unlike the transition metals discussed earlier that involve d orbitals, molecular sulfur spectra involve s and p orbitals (Meyer et al., 1972; Eclert & Steudel, 2003).

1.6.10 Iron-Containing Glasses

The effects of electronic processes on spectra are observed even in non- or poorly crystalline materials because the metal cations are still within an environment in which the electron orbitals are split energetically due to the presence of other ions. For example, iron-bearing glasses exhibit tetrahedrally and octahedrally coordinated Fe(II), intervalence charge transfer, and oxygen-to-iron charge transfer (Figure 1.6b; Burns, 1993; Cannon et al., 2017). The relative strengths of these absorptions depend on the amount of iron, its coordination (affected by other cations), and the Fe(II)/Fe(III) ratio (Figure 1.6b). The Fe(II)/Fe(III) ratio is indicative of the oxygen fugacity under which the glass formed. Other trace metals, such as Ti, also can generate absorptions in glasses.

1.7 Techniques for Mapping on Planetary Surfaces

The location of band centers in the VNIR is a crucial indicator that a particular mineral species is present. Thus, in the analyses of calibrated spectra, a first step is the identification of band minima. Their locations, shapes, and associations with other minima can be used to uniquely identify mineral species based on the results of the foregoing laboratory data and then spatially mapped. For example, broad absorptions centered at 1300 nm can be uniquely attributed to Fe–plagioclase feldspar if no ~2000 nm feature is apparent that might indicate glass. Olivine’s broad absorption centered near 1000 nm (really a set of superimposed absorptions, as we know from Section 1.6.2) is unique in breadth and thus diagnostic (Figure 1.6a).

For olivines and pyroxenes, which occur in solid solution and whose crystal chemistry is uniquely diagnostic of magmatic processes, additional techniques have been devised to discriminate their solid solution chemistries. The most common implementation is the Modified Gaussian Model (e.g., Sunshine & Pieters, 1993, 1998) whereby the systematic shifts in position, width, breadth, and shape of absorptions in pyroxenes or olivines have been quantified as a function of solid solution chemistry. Information about pyroxene and olivine chemistry is obtained by fitting these parameters to remote sensing data. For more information, see Chapters 4, 11, and 14.

1.8 Geological Significance of Electronic Processes on Solar System Bodies

Because of a confluence of detector technology (abundant and relatively inexpensive charge coupled device (CCD) and complementary metal oxide semiconductor (CMOS) detectors sensitive over the ~400–1000 nm range and requiring no cooling), the relative transparency of the terrestrial atmosphere in the VNIR (for telescopic observation), and the presence of the fingerprints of electronic processes, VNIR spectroscopy of electronic processes has been fruitful in expanding our understanding of the Solar System. Within the asteroid belt, discrete families of pyroxene-rich, olivine-rich, and Fe–phyllosilicate-rich asteroids have been identified and linked to the early evolution and differentiation of planetary bodies (e.g., DeMeo et al., 2009; Chapters 19 and 20). On Mars, changes in the composition of lavas with time and location give insight into temporal and spatial variation in magmatic processes, thus providing key data on the interior evolution of our smaller neighboring planet (e.g., Mustard et al., 2005; Baratoux et al., 2013). Also on Mars, a rich array of secondary Fe-bearing minerals has been identified, which point to ubiquitous oxidation to form ferric oxides and discrete, geologic formations where water–rock interactions formed Fe-rich phyllosilicates, Fe carbonates, or Fe sulfates (e.g., Ehlmann & Edwards, 2014; Chapter 23). On the Moon, differences in pyroxene composition are found in the ancient highlands and later lavas, massive deposits of ferroan anorthosite support models of an early lunar magma ocean, and olivine-enriched deposits and spinel-anorthosite deposits point to excavation of lower crust materials (Chapter 18). The composition of the Venus surface is largely obscured because of atmospheric opacity at key wavelengths, though landed observations hint at the presence of oxidized Fe oxides, specifically hematite (Pieters et al., 1986). Mercury, intriguingly, has no evidence for electronic absorptions, indicating a uniquely iron-poor crust (e.g., Izenberg et al., 2014; Chapter 17). In the outer Solar System, vibrational absorptions of ices and electronic absorptions of sulfur species, rather than transition metal–bearing materials, dominate the surfaces and have been used to trace volcanism, upwelling of materials from subsurface oceans, and radiation-induced weathering of surfaces (e.g., Nash et al., 1980; Carlson et al., 1997).


Amthauer G. & Rossman G.R. (1984) Mixed valence of iron in minerals with cation clusters. Physics and Chemistry of Minerals, 11, 3751.
Aronson J.R., Bellotti L.H., Eckroad S.W., Emslie A.G., McConnell R.K., & Thüna P.C. (1970) Infrared spectra and radiative thermal conductivity of minerals at high temperature. Journal of Geophysical Research, 75(17), 34433456.
Baratoux D., Toplis M.J., Monnereau M., & Sautter V. (2013) The petrological expression of early Mars volcanism. Journal of Geophysical Research, 118, 5964.
Bell P.M. & Mao H.K. (1973) Optical and chemical analysis of iron in Luna 20 plagioclase. Geochimica et Cosmochimica Acta, 37, 755759.
Berg B.L., Cloutis E.A., Beck P., et al. (2016) Reflectance spectroscopy (0.35–25 µm) of ammonium-bearing minerals and comparison to Ceres family asteroids. Icarus, 265, 218237.
Bishop J.L. & Murad E. (1996) Schwertmannite on Mars? Spectroscopic analyses of schwertmannite, its relationship to other ferric minerals, and its possible presence in the surface material on Mars. In: Mineral spectroscopy: A tribute to Roger G. Burns. Special publication (Geochemical Society). No. 5. Geochemical Society, Houston, TX, 337358.
Bishop J.L., Dyar M.D., Lane M.D., & Banfield J.F. (2004) Spectral identification of hydrated sulfates on Mars and comparison with acidic environments on Earth. International Journal of Astrobiology, 3, 275285.
Bishop J.L., Perry K.A., Dyar M.D., et al. (2013) Coordinated spectral and XRD analyses of magnesite-nontronite-forsterite mixtures and implications for carbonates on Mars. Journal of Geophysical Research, 118, 635650.
Burns R.G. (1981) Intervalence transitions in mixed valence minerals of iron and titanium. Annual Review of Earth and Planetary Sciences, 9, 345383.
Burns R.G. (1993) Mineralogical applications of crystal field theory. Cambridge University Press, Cambridge.
Burns R.G. & Vaughan D.J. (1975) 2 – Polarized Electronic Spectra. In: Infrared and Raman spectroscopy of lunar and terrestrial minerals (C. Karr, ed.). Academic Press, New York, 3972.
Cannon K.M., Mustard J.F., Parman S.W., Sklute E.C., Dyar M.D., & Cooper R.F. (2017) Spectral properties of martian and other planetary glasses and their detection in remotely sensed data. Journal of Geophysical Research, 122, 249268.
Carlson R.W., Smythe W.D., Lopes-Gautier R.M.C., et al. (1997) The distribution of sulfur dioxide and other infrared absorbers on the surface of Io. Geophysical Research Letters, 24, 24792482.
Cheek L.C. (2014) Foundations of lunar highland crustal mineralogy derived from remote sensing and laboratory spectroscopy of plagioclase-dominated Materials. Brown University Earth, Environmental and Planetary Sciences Theses and Dissertations.
Chemtob S.M., Nickerson R.D., Morris R.V., Agresti D.G., & Catalano J.G. (2015) Synthesis and structural characterization of ferrous trioctahedral smectites: Implications for clay mineral genesis and detectability on Mars. Journal of Geophysical Research, 120, 11191140.
Cloutis E.A., Sunshine J.M., & Morris R.V. (2004) Spectral reflectance-compositional properties of spinels and chromites: Implications for planetary remote sensing and geothermometry. Meteoritics and Planetary Science, 39, 545565.
Cloutis E.A., Hawthorne F.C., Mertzman S.A., et al. (2006) Detection and discrimination of sulfate minerals using reflectance spectroscopy. Icarus, 184, 121157.
Cotton F.A. (1990) Chemical applications of group theory, 3rd edn. Wiley-Interscience, New York.
DeMeo F.E., Binzel R.P., Slivan S.M., & Bus S.J. (2009) An extension of the Bus asteroid taxonomy into the near-infrared. Icarus, 202, 160180.
De Sanctis M.C., Ammannito E., Capria M.T., et al. (2012) Spectroscopic characterization of mineralogy and its diversity across Vesta. Science, 336, 697.
Dowty E.C. & Clark J.R. (1973) Crystal structure refinement and visible-region absorption spectra of a Ti3+ fassaite from the Allende meteorite. American Mineralogist, 58, 230242.
Eckert B. & Steudel R. (2003) Molecular spectra of sulfur molecules and solid sulfur allotropes. In: Elemental sulfur and sulfur-rich compounds II (R. Steudel, ed.). Springer, Berlin, Heidelberg, 3198.
Ehlmann B.L. & Edwards C.S. (2014) Mineralogy of the martian surface. Annual Review of Earth and Planetary Sciences, 42, 291315.
Fraeman A.A., Arvidson R.E., Catalano J.G., et al. (2013) A hematite-bearing layer in Gale crater, Mars: Mapping and implications for past aqueous conditions. Geology, 41, 11031106.
Gaffey S.J. (1985) Reflectance spectroscopy in the visible and near-infrared (0.35–2.55 µm): Applications in carbonate petrology. Geology, 13, 270273.
Goldman D.S. & Rossman G.R. (1977) The spectra of iron in orthopyroxene revisited: The splitting of the ground state. American Mineralogist, 62, 151157.
Hapke B. (1981) Bidirectional reflectance spectroscopy, 1. Theory. Journal of Geophysical Research, 86, 30393054.
Harris D.C. & Bertolucci M.D. (1989) Symmetry and spectroscopy: An introduction to vibrational and electronic spectroscopy. Dover Publications, Mineola, NY.
Horgan B.H.N., Cloutis E.A., Mann P., & Bell J.F. (2014) Near-infrared spectra of ferrous mineral mixtures and methods for their identification in planetary surface spectra. Icarus, 234, 132154.
Isaacson P.J., Klima R.L., Sunshine J.M., et al. (2014) Visible to near-infrared optical properties of pure synthetic olivine across the olivine solid solution. American Mineralogist, 99, 467478.
Izenberg N.R., Klima R.L., Murchie S.L., et al. (2014) The low-iron, reduced surface of Mercury as seen in spectral reflectance by MESSENGER. Icarus, 228, 364374.
Karr C. (1975) Infrared and Raman spectroscopy of lunar and terrestrial materials. Academic Press, New York.
Klima R.L., Dyar M.D., & Pieters C.M. (2011) Near-infrared spectra of clinopyroxenes: Effects of calcium content and crystal structure. Meteoritics and Planetary Science, 46, 379395.
Lane M.D., Bishop J.L., Dyar M.D., et al. (2015) Mid-infrared emission spectroscopy and visible/near-infrared reflectance spectroscopy of Fe-sulfate minerals. American Mineralogist, 100, 6682.
Ling Z., Cao F., Ni Y., Wu Z., Zhang J., & Li B. (2016) Correlated analysis of chemical variations with spectroscopic features of the K–Na jarosite solid solutions relevant to Mars. Icarus, 271, 1929.
Mao H.K., Bell P.M., & Virgo D. (1977) Crystal-field spectra of fassaite from the Angra dos Reis meteorite. Earth and Planetary Science Letters, 35, 352356.
Mattson S.M. & Rossman G.R. (1988) Fe2+-Ti4+ charge transfer in stoichiometric Fe2+,Ti4+-minerals. Physics and Chemistry of Minerals, 16, 7882.
McCollom T.M., Ehlmann B.L., Wang A., Hynek B., Moskowitz B., & Berquó T.S. (2014) Detection of iron substitution in natroalunite-natrojarosite solid solutions and potential implications for Mars. American Mineralogist, 99, 948964.
McCord T.B., Adams J.B., & Johnson T.V. (1970) Asteroid vesta – Spectral reflectivity and compositional implications. Science, 168, 14451447.
Meyer B., Gouterman M., Jensen D., Oommen T.V., Spitzer K., & Stroyer-Hansen T. (1972) The spectrum of sulfur and its allotropes. Advances in Chemistry, 110, 5372.
Morris R.V., Lauer H.V. Jr., Lawson C.A., Gibson E.K. Jr., Nace G.A., & Stewart C. (1985) Spectral and other physicochemical properties of submicron powders of hematite (a-Fe2O3), maghemite (g-Fe2O3), magnetite (Fe3O4), goethite (a-FeOOH), and lepidocrocite (g-FeOOH). Journal of Geophysical Research, 90, 31263144.
Morris R.V., Golden D.C., Bell J.F. III, et al. (2000) Mineralogy, composition, and alteration of Mars Pathfinder rocks and soils: Evidence from multispectral, elemental, and magnetic data on terrestrial analogue, SNC meteorite, and Pathfinder samples. Journal of Geophysical Research, 105, 17571817.
Mustard J.F., Poulet F., Gendrin A., et al. (2005) Olivine and pyroxene diversity in the crust of Mars. Science, 307, 15941597.
Nash D.B., Fanale F.P., & Nelson R.M. (1980) SO2 Frost: UV‐visible reflectivity and Io surface coverage. Geophysical Research Letters, 7, 665668.
Pieters C.M. (1978) Mare basalt types on the front side of the moon – A summary of spectral reflectance data. Proc. 9th Lunar Planet. Sci. Conf., 3, 2825–2849.
Pieters C.M. (1986) Composition of the lunar highland crust from near-infrared spectroscopy. Reviews of Geophysics, 24, 557578.
Pieters C.M., Head J.W. III, Patterson W., et al. (1986) The color of Venus. Science, 234, 13791383.
Pitman K.M., Dobrea E.Z.N., Jamieson C.S., Dalton J.B., Abbey W.J., & Joseph E.C.S. (2014) Reflectance spectroscopy and optical functions for hydrated Fe-sulfates. American Mineralogist, 99, 15931603.
Rossman G.R. (1975) Spectroscopic and magnetic studies of ferric iron hydroxy sulfates: Intensification of color in ferric iron clusters bridged by a single hydroxide ion. American Mineralogist, 60, 698704.
Rossman G.R. (1976) Spectroscopic and magnetic studies of ferric iron hydroxy sulfates: The series Fe(OH)SO4•nH2O and the jarosites. American Mineralogist, 61, 398404.
Rossman G.R. (1988) Optical spectroscopy. In: Spectroscopic methods in mineralogy and geology (F.C. Hawthorne, ed.). Mineralogical Society of America, Washington, DC, 207–254.
Rossman G.R. (1996) Why hematite is red: Correlation of optical absorption intensities and magnetic moments of Fe3+ minerals. In: Mineral spectroscopy: A tribute to Roger G. Burns. Special publication (Geochemical Society). No. 5. Geochemical Society, Houston, TX, 2327.
Rossman G.R. (2014) Optical spectroscopy. Reviews in Mineralogy and Geochemistry, 78, 371398.
Sherman D.M. & Waite T.D. (1985) Electronic spectra of Fe3+ oxides and oxide hydroxides in the near IR to near UV. American Mineralogist, 70, 12621269.
Sherman D.M. & Vergo N. (1988) Optical (diffuse reflectance) and Mössbauer spectroscopic study of nontronite and related Fe-bearing smectites. American Mineralogist, 73, 13461354.
Sklute E.C., Jensen H.B., Rogers A.D., & Reeder R.J. (2015) Morphological, structural, and spectral characteristics of amorphous iron sulfates. Journal of Geophysical Research, 120, 809830.
Staid M.I., Pieters C.M., Besse S., et al. (2011) The mineralogy of late stage lunar volcanism as observed by the Moon Mineralogy Mapper on Chandrayaan‐1. Journal of Geophysical Research, 116, E00G10, DOI:10.1029/2010JE003735.
Sung C.-M., Singer R.B., Parkin K.M., & Burns R.G. (1977) Temperature dependence of Fe2+ crystal field spectra: Implications to mineralogical mapping of planetary surfaces. Proc. 8th Lunar Sci. Conf, 1063–1079.
Sunshine J.M. & Pieters C.M. (1993) Estimating modal abundances from the spectra of natural and laboratory pyroxene mixtures using the Modified Gaussian Model. Journal of Geophysical Research, 98, 90759087.
Sunshine J.M. & Pieters C.M. (1998) Determining the composition of olivine from reflectance spectroscopy. Journal of Geophysical Research, 103, 13,675–13,688.
Vilas F., Jarvis K.S., & Gaffey M.J. (1994) Iron alteration minerals in the visible and near-infrared spectra of low-albedo asteroids. Icarus, 109, 274283.
Whitten J. & Head J.W. (2015) Lunar cryptomaria: Mineralogy and composition of ancient volcanic deposits. Planetary and Space Science, 106, 6781.
Wildner M., Andrut M., & Rudowicz C.Z. (2004) Optical absorption spectroscopy in geosciences: Part I: Basic concepts of crystal field theory; Part 2: Quantitative aspects of crystal fields. In: Spectroscopic methods in mineralogy (A. Beran & E. Libowitzky, eds.). Mineralogical Society of Great Britain and Ireland.