The M-/X-asteroid menagerie: Results of an NIR spectral survey of 45 main-belt asteroids


Corresponding author. E-mail:


Abstract– Diagnostic mineral absorption features for pyroxene(s), olivine, phyllosilicates, and hydroxides have been detected in the near-infrared (NIR: approximately 0.75–2.50 μm) spectra for 60% of the Tholen-classified (Tholen 1984, 1989) M-/X-asteroids observed in this study. Nineteen asteroids (42%) exhibit weak Band I (approximately 0.9 μm) ± Band II (approximately 1.9 μm) absorptions, three asteroids (7%) exhibit a weak Band I (approximately 1.05–1.08 μm) olivine absorption, four asteroids (9%) display multiple absorptions suggesting phyllosilicate ± oxide/hydroxide minerals, one (1) asteroid exhibits an S-asteroid type NIR spectrum, and 18 asteroids (40%) are spectrally featureless in the NIR, but have widely varying slopes. Tholen M-asteroids are defined as asteroids exhibiting featureless visible-wavelength (λ) spectra with moderate albedos (Tholen 1989). Tholen X-asteroids are also defined using the same spectral criterion, but without albedo information. Previous work has suggested spectral and mineralogical diversity in the M-asteroid population (Rivkin et al. 1995, 2000; Busarev 2002; Clark et al. 2004; Hardersen et al. 2005; Birlan et al. 2007; Ockert-Bell et al. 2008, 2010; Shepard et al. 2008, 2010; Fornasier et al. 2010). The pyroxene-bearing asteroids are dominated by orthopyroxene with several likely to include higher-Ca clinopyroxene components. Potential meteorite analogs include mesosiderites, CB/CH chondrites, and silicate-bearing NiFe meteorites. The Eos family, olivine-bearing asteroids are most consistent with a CO chondrite analog. The aqueously altered asteroids display multiple, weak absorptions (0.85, 0.92, 0.97, 1.10, 1.40, and 2.30–2.50 μm) indicative of phyllosilicate ± hydroxide minerals. The spectrally featureless asteroids range from metal-rich to metal-poor with meteorite analogs including NiFe meteorites, enstatite chondrites, and stony-iron meteorites.


The M- and X-asteroid taxonomic designations of Tholen (1984, 1989) are based on visible λ spectra and albedo information. The X-class includes all asteroids in the ECAS data set (Zellner et al. 1985) that have featureless visible-λ spectra as the only classification criterion. The X-class becomes nondegenerate if object albedo information is available as a second criterion. This produces three subclasses, the E-, M-, and P-asteroids that have relatively high, moderate, and low albedos, respectively. For example, the 38 Tholen M/MU/MU: asteroids in this work have an albedo range approximately from 7 to 30% (Tedesco et al. 2002). The MU and MU: variants of the M-class designation indicate “unusual” and “unusual and noisy” data (Tholen 1989). The five X-/XD-class asteroids have a wider albedo range, from 6 to 34% (Tedesco et al. 2002). The one Tholen S-asteroid in this article, 71 Niobe, has an albedo of approximately 31% (Tedesco et al. 2002).

Prior to 2001 and the introduction of the SpeX 0.8–5.5 μm spectrograph at the NASA Infrared Telescope Facility (IRTF; Rayner et al. 2003, 2004), M-asteroid NIR spectra were reported as being spectrally featureless, which led to the suggestion that enstatite chondrites and NiFe meteorites were the most likely analogs for these asteroids (Chapman and Salisbury 1973; Gaffey and McCord 1979). Polarimetric observations by Dollfus et al. (1979) supported the metallic interpretation by suggesting that particulate metallic surfaces with grain diameters of 20–50 μm as being the most consistent with their polarimetric data. Lupishko and Belskaya (1989), however, suggested that the M-asteroids 16 Psyche, 21 Lutetia, 22 Kalliope, 69 Hesperia, and 110 Lydia, have a silicate surface component with enstatite chondrites and stony-iron meteorites as the most consistent options. Jones et al. (1990) reported that 55 Pandora and 92 Undina, both M-asteroids, displayed 3 μm absorptions, whereas 16 Psyche did not. The work by Jones et al. (1990) was the first suggestion of a hydrous mineral component for some M-asteroids, which were previously thought to be anhydrous.

In addition, radar observations suggest that the M-asteroid population displays both a larger average radar albedo and larger overall radar albedo variability (Magri et al. 1999; Shepard et al. 2010). A high radar albedo is the most robust indicator suggesting the presence of metal on the surfaces of atmosphereless bodies (Ostro 1993). Ostro et al. (1985) reported differing radar albedos for 16 Psyche and 97 Klotho, with 16 Psyche as the most likely one to have a significant surface metal component. Mitchell et al. (1995) reported a high radar albedo for 216 Kleopatra (0.44 ± 0.15), which was subsequently revised upward to 0.7 (Ostro et al. 2000) and 0.68 ± 0.28 (Magri et al. 2007). Shepard et al. (2008) reported an all-observation radar albedo average for 216 Kleopatra of 0.60 ± 0.15.

Three-micron and visible λ spectra from Rivkin et al. (1995, 2000) and Busarev (2002) suggested the presence of hydrous mineralogical environments (i.e., phyllosilicates, hydrated salts, etc.) for some M-asteroids, which provided additional hints of mineralogic variability within this taxonomic group. Rivkin et al. (1995, 2000) reported that 21 Lutetia, 22 Kalliope, 55 Pandora, 77 Frigga, 92 Undina, 110 Lydia, 129 Antigone, 135 Hertha, 136 Austria, and 201 Penelope display a 3 μm absorption feature. Mineralogically, the implication of a 3 μm absorption feature in an asteroid’s NIR spectrum is the presence of H2O- or OH-bearing minerals, such as the phyllosilicate group of minerals and various hydroxide minerals (Calvin and King 1997). However, a 3 μm absorption is not diagnostic and may mask otherwise anhydrous objects (Gaffey et al. 2002, and references therein).

Clark et al. (2004) and Hardersen et al. (2005) reported the detection of weak approximately 0.9 μm absorptions in the NIR spectra for some M-/X-asteroids. Clark et al. (2004) reported approximately 0.9 μm features in the NIR spectra of 16 Psyche, 110 Lydia, 216 Kleopatra, and 785 Zwetana, whereas Hardersen et al. (2005) reported similar features for 16 Psyche, 69 Hesperia, 110 Lydia, 125 Liberatrix, 201 Penelope, and 216 Kleopatra.

Since then, several research teams have undertaken both radar and NIR spectral studies of the M-/X-asteroid population (Birlan et al. 2007; Ockert-Bell et al. 2008, 2010; Shepard et al. 2008, 2010; Fornasier et al. 2010; this paper). Birlan et al. (2007) reported featureless NIR spectra of eight M-asteroids (325 Heidelberga, 497 Iva, 558 Carmen, 687 Tinette, 860 Ursina, 909 Ulla, 1280 Baillauda, and 1564 Srbija) that are most consistent with a NiFe meteorite analog via spectral curve matching. One additional asteroid, 766 Moguntia, exhibits a weak olivine absorption feature with a band depth of approximately 7% (Birlan et al. 2007). Spectral curve matching, MGM modeling, and application of the Shkuratov scattering model suggest the CV3 chondrite, Allan Hills (ALH) 84028, and the H6 ordinary chondrite, Ozona, as the best analogs for 766 Moguntia. Nedelcu et al. (2007) reported rotationally resolved NIR spectra of 21 Lutetia. Their spectra exhibit slope variations and they suggest enstatite chondrites (EL6) and carbonaceous chondrites (CI1, CV3) as the best meteorite analogs, depending on the nature of each NIR spectral slope for 21 Lutetia.

Shepard et al. (2008, 2010) and Ockert-Bell et al. (2008, 2010) have conducted coordinated radar and NIR spectral observations of members of the M-/X-asteroid population in an effort to better constrain the mineralogic and meteoritic diversity of these asteroids. Shepard et al. (2008, 2010), as well as previous workers, have obtained radar observations for 19 M-/X-asteroids. These include 16 Psyche, 21 Lutetia, 22 Kalliope, 83 Beatrix, 97 Klotho, 110 Lydia, 129 Antigone, 135 Hertha, 216 Kleopatra, 224 Oceana, 325 Heidelberga, 347 Pariana, 497 Iva, 678 Fredegundis, 758 Mancunia, 771 Libera, 779 Nina, 785 Zwetana, and 796 Sarita (Shepard et al. 2010). As a group, their radar albedos vary from 0.07 to 0.60 with 13 of the asteroids having radar albedos <0.30 (Shepard et al. 2010). Due to the wide range of variability in radar albedo seen for some of their asteroids and the presence of silicate minerals (Hardersen et al. 2005; Ockert-Bell et al. 2008, 2010; Fornasier et al. 2010), Shepard et al. (2010) suggested that most M-asteroids are collisional mixes, mostly analogous to stony-iron meteorites and high-Fe carbonaceous chondrites. However, individual analogs for specific asteroids suggested by Shepard et al. (2010) span a wide range that includes NiFe meteorites, CH/CB chondrites, enstatite chondrites, and stony irons.

Ockert-Bell et al. (2008, 2010) and Fornasier et al. (2010) presented M-/X-asteroid NIR spectra for 22 and 30 asteroids, respectively. While Ockert-Bell et al. (2008) focused on analyzing asteroid NIR continuum slopes, they reported weak, approximately 0.9 μm band depths for 16 Psyche and 129 Antigone of approximately 1.2–1.5%. Examination of fig. 2 in Ockert-Bell et al. (2008) suggests the presence of weak, approximately 0.9 μm features for 55 Pandora, 110 Lydia, 216 Kleopatra, and 872 Holda. Ockert-Bell et al. (2010) reported weak, approximately 0.9 μm features for 16 Psyche, 22 Kalliope, 77 Frigga, 129 Antigone, 135 Hertha, 136 Austria, 250 Bettina, 441 Bathilde, 497 Iva, 678 Fredegundis, 771 Libera, and 872 Holda. They report approximately 0.9 ± 1.9 μm features for 55 Pandora, 110 Lydia, 216 Kleopatra, 347 Pariana, 758 Mancunia, 779 Nina, and 785 Zwetana (Ockert-Bell et al. 2010). Ockert-Bell et al. (2010) note that the highest radar albedo asteroids tend to also have weak approximately 0.9 and 1.9 μm absorption features.

Fornasier et al. (2010) report weak visible-λ and/or NIR absorptions for 17 M-asteroids that include 16 Psyche, 22 Kalliope, 55 Pandora, 69 Hesperia, 110 Lydia, 129 Antigone, 132 Aethra, 498 Tokio, 516 Amherstia, 755 Quintilla, and 872 Holda. Visible-λ absorptions are either at approximately 0.43 μm or 0.50 μm; NIR absorptions span a wider range, but dominate in the Band I region from approximately 0.87 to 1.16 μm. NIR absorptions at longer wavelengths are reported for 755 Quintilla and 516 Amherstia (range: approximately 1.37–1.95 μm) (Fornasier et al. 2010). The collection of absorption features identified by Fornasier et al. (2010) are attributed to minerals that include chlorites, serpentinites, antigorite, pigeonite, augites, orthopyroxene, Fe-bearing pyroxene, olivine, oldhamite/troilite, and unknown phases.

Taken as a cumulative body of research, the available evidence suggests that the mineralogic and meteoritic diversity of the M-/X-asteroid taxonomic classes continues to expand. The intent of this article is to add to the existing knowledge of the M-/X-asteroid taxons and to search for additional evidence of mineralogic, meteoritic, and spectral diversity.

Observations and Data Reduction

All observations were conducted at the NASA Infrared Telescope Facility, Mauna Kea, Hawai’i, using the SpeX spectrograph in the prism mode (i.e., low resolution, R approximately 95), 0.75 to 2.50 μm, 0.8″ × 15″ slit (Rayner et al. 2003, 2004). The full observational protocols are described in Hardersen et al. (2004, 2005, 2006a), but the standard procedure involves obtaining NIR spectra of the target asteroids, standard (i.e., extinction) stars (as close to G2V as possible) for modeling the first-order extinction coefficients above Mauna Kea, and solar analog stars to correct for slope differences when using non-G2V extinction stars. To minimize spectral data dispersion and to securely detect weak NIR spectral absorption features, a signal-to-noise ratio (SNR) ≥100 is desired for each nightly average asteroid spectrum. Oftentimes, the nightly asteroid spectrum is an average of only those spectra with the best telluric corrections (i.e., the least data scatter in the telluric water vapor regions at approximately 1.4 and 1.9 μm). See Table 1 for a compilation of the observing circumstances for the asteroids reported in this article.

Table 1.   Summary of the observational circumstances for the 45 asteroids observed at the NASA Infrared Telescope Facility (IRTF). Individual asteroid and stellar spectra were limited to individual exposure times of 120 s. Stellar spectral classifications obtained from the Simbad Astronomical Database ( Asteroid apparent magnitudes obtained from the JPL Horizons ephemeris during time of observation ( UT start/stop times represent either (1) the time interval between the beginning of the first and last set of observations, or (2) the time interval for continuous set(s) of observations.
AsteroidDate(s) of observation (UT)Apparent magnitudeNo. of spectra obtainedNo. of spectra used to produce spectrumStart time (UT) Stop time (UT)Total integration time (s)Airmass range of observationsExtinction star usedSolar analog star usedWeather notes
16 Psyche3/22/200210.7120120705113518001.167–1.268HD 95868 (G2V)SAO 120107 (G5 III)Clear
21 Lutetia8/16/200812.168681318142827201.245–1.574HD 285660 (G0V)HD 28099 (G2V)Initial summit clouds in early evening; otherwise, clear
22 Kalliope5/9/200411.4205120512298001.406–1.417HD 155415 (G2V)SAO 120107 (G5 III)Clear
22 Kalliope2/27/200811.27060123215217001.020–1.201SAO 119805 (G1V)SAO 120107 (G5 III)Clear
55 Pandora1/22/200712.61641001010144586101.087–1.380HD 94794 (F8V)SAO 120107 (G5 III)Clear
69 Hesperia5/3/200112.254541126145421501.304–1.364SAO 141608 (G2V)SAO 120107 (G5 III), SAO 31899 (G3V)Clear
69 Hesperia5/4/200112.21616132914258801.197–1.288SAO 141608 (G2V)SAO 31899 (G3V)Clear
71 Niobe1/6/200912.9403052065848001.092–1.405SAO 74132 (G4V)HD 28099 (G2V)Mostly clear, but some cirrus clouds are present
77 Frigga1/23/200713.230651763136001.077–1.286HD 9986 (G5V)SAO 120107 (G5 III)Clear
97 Klotho5/9/200412.54013842125430001.246–1.321HD 141308 (G2V)SAO 120107 (G5 III)Clear
97 Klotho5/10/200412.55412826134454501.369–1.488HD 141308 (G2V)SAO 120107 (G5 III)Light cirrus present across the sky
110 Lydia5/2/200112.6303062492021001.066–1.144SAO 119191 (G0V)SAO 120107 (G5 III)Clear
110 Lydia5/3/200112.6381854595935401.127–1.284SAO 119191 (G0V)SAO 120107 (G5 III), SAO 31899 (G3V)Clear
110 Lydia5/4/200112.61616915103017601.153–1.467SAO 119191 (G0V)SAO 31899 (G3V)Clear
125 Liberatrix3/23/200213.3808053591548001.036–1.396SAO 118176 (G1V)SAO 120107 (G5 III)Clear
129 Antigone8/19/200511.410106306471501.412–1.503HD 141247 (F9V)None.Mostly clear skies. Overhead cirrus at beginning of night
129 Antigone8/20/200511.430306056376901.322–1.465HD 141247 (F9V)None.Mostly clear skies. Cirrus near horizons only
129 Antigone8/21/200511.44066046318001.331–1.450HD 141247 (F9V)None.Cirrus overhead beginning of night. Otherwise clear remainder of night
132 Aethra5/21/200812.820201158125024001.402–1.410SAO 185949 (G5V)SAO 27996 (G5V)Variable abundance of cirrus clouds present around the sky
132 Aethra8/18/200814.3101053960712001.200–1.215HD 154067 (G2/G3V)HD 28099 (G2V)Clear
135 Hertha10/3/200410.21101584194411901.037–1.130HD 6302 (G8V)HD 1835 (G3V)Clear
136 Austria3/22/200212.85050825140535001.100–1.514SAO 139102 (G8V)SAO 120107 (G5 III)Clear
184 Dejopeja8/20/200513.12071139132620001.183–1.518HD 211839 (G3V)None.Mostly clear skies. Cirrus near horizons only
201 Penelope5/2/200112.240401038123727001.140–1.293SAO 140198 (G2V)SAO 120107 (G5 III)Clear
201 Penelope5/3/200112.25042900123229801.138–1.272SAO 140198 (G2V)SAO 120107 (G5 III), SAO 31899 (G3V)Clear
201 Penelope7/29/200610.970701000104113201.197–1.203HD 194428 (G2/G3V)SAO 31899 (G3V)Early evening fog. Otherwise, clear
216 Kleopatra4/29/200112.6393960092844701.090–1.434SAO 138119 (G8V)SAO 120107 (G5 III)Variable cirrus clouds present during the observations
216 Kleopatra5/4/200112.7242461984824801.086–1.332SAO 138119 (G8V)SAO 31899 (G3V)Clear
224 Oceana10/2/200412.1741570391755801.072–1.620HD 6302 (G8V)HD 1835 (G3V)Clear
250 Bettina4/19/200512.75020550102427001.001–1.629HD 88371 (G2V)SAO 120107 (G5 III), SAO 31899 (G3V)Clear skies, low relative humidity
325 Heidelberga4/29/200114.2447357494801.155–1.167SAO 138119 (G8V)SAO 120107 (G5 III)Variable cirrus clouds present during the observations
325 Heidelberga1/23/200715.03491259140740801.485–1.977HD 121867 (G2V)SAO 120107 (G5 III)Clear
338 Budrosa7/30/200613.180801018132396001.059–1.352SAO 146477 (F8V)SAO 31899 (G3V)Early fog and high relative humidity, followed by clear skies
347 Pariana5/9/200411.95619821122931001.143–1.537HD 141308 (G2V)SAO 120107 (G5 III)Clear
369 Aeria4/19/200513.260271153150372001.215–1.463HD 157842 (G2V)SAO 120107 (G5 III)Clear skies, low relative humidity
382 Dodona4/21/200512.34020929102424001.307–1.338HD 115106 (G2V)SAO 120107 (G5 III)Clear
413 Edburga4/20/200515.810014526932120001.007–1.307HD 104076 (G0V)SAO 120107 (G5III)Clear skies, low relative humidity
417 Suevia8/20/200514.05050847135960001.060–1.385HD 217786 (F8V)None.Mostly clear skies. Cirrus near horizons only
417 Suevia8/21/200514.03016937130636001.061–1.184HD 217786 (F8V)None.Cirrus overhead beginning of night. Otherwise clear remainder of night
418 Alemannia1/23/200714.546141034122855201.148–1.555HD 98281 (G8V)SAO 120107 (G5 III)Clear
418 Alemannia1/24/200714.570701027133284001.106–1.575HD 98281 (G8V)SAO 120107 (G5 III)Clear
441 Bathilde8/19/200512.83030940114713501.112–1.378HD 198273 (G2V)None.Mostly clear skies. Overhead cirrus at beginning of night
497 Iva4/21/200516.0802051281396001.000–1.188SAO 98655 (F8V)SAO 120107 (G5 III)Clear
498 Tokio10/9/200111.566130613154501.487–1.539SAO 53622 (G2V)SAO 31899 (G3V)Clear
504 Cora1/22/200713.9262050574831201.011–1.131HIP 15904 (G0V)SAO 120107 (G5 III)Clear
504 Cora1/23/200713.9201071475124001.071–1.147HIP 15904 (G0V)SAO 120107 (G5 III)Clear
516 Amherstia10/4/200413.126783190631201.066–1.136HD 7352 (G0V)HD 1835 (G3V)Patchy cirrus present
558 Carmen1/22/200412.96850811112761201.004–1.281HD 75528 (G1V)HD 28099 (G2V)Clear
739 Mandeville8/21/200512.840880685348001.296–1.401HD 203619 (G3V)None.Cirrus overhead beginning of night. Otherwise clear remainder of night
758 Mancunia8/19/200512.64040919121224001.219–1.389HD 214414 (G3V)None.Mostly clear skies. Overhead cirrus at beginning of night
766 Moguntia1/20/200414.58019618102396001.053–1.195HD 41330 (G0V)HD 28099 (G2V)Clear
785 Zwetana4/19/200511.73030110812347501.040–1.162BD+06 2840B (G0V)SAO 120107 (G5 III)Clear skies, low relative humidity
796 Sarita1/24/200715.310101416144312001.442–1.586HD 129485 (G6V)SAO 120107 (G5 III)Clear
798 Ruth8/18/200815.32051308140124001.183–1.407SAO 76384 (G4V)HD 28099 (G2V)Clear
857 Glasenappia8/17/200815.92091334142824001.266–1.571HD 28099 (G2V)HD 28099 (G2V)Clear
857 Glasenappia8/18/200815.988141714449601.200–1.303HD 28099 (G2V)HD 28099 (G2V)Clear
860 Ursina8/16/200813.45040752112560001.117–1.460HD 209712 (F6V)HD 28099 (G2V)Initial summit clouds in early evening; otherwise, clear
860 Ursina1/7/200915.610444651512001.396–1.583HD 215763 (F9V)HD 28099 (G2V)Clear
872 Holda10/3/200414.0341163375240801.143–1.431HD 223498 (G7V)HD 1835 (G3V)Clear
931 Whittemora4/19/200514.6302964390136001.006–1.205SAO 100038 (G2V)SAO 120107 (G5 III)Clear skies, low relative humidity
1210 Morosovia1/22/200715.4363654880543201.011–1.195HIP 15904 (G0V)SAO 120107 (G5 III)Clear
1210 Morosovila1/24/200715.5302050955736001.011–1.022HIP 15904 (G0V)SAO 120107 (G5 III)Clear
1461 Jean-Jacques7/29/200615.536563974843201.206–1.314HD 146720 (G3V)SAO 31899 (G3V)Early evening fog. Otherwise, clear

Observations for the August 2005 IRTF observing run did not utilize solar analog (i.e., G2V) stars. Within this article, this applies to the NIR spectra of 129 Antigone, 184 Dejopeja, 417 Suevia, 441 Bathilde, 739 Mandeville, and 758 Mancunia. Extinction corrections for these asteroids utilized either F9V or G3V stars. The overall NIR spectral slope for the asteroids using the F9V standard stars (129 Antigone, 417 Suevia, 441 Bathilde) will be somewhat decreased, mostly at the shorter λ. The variations will be relatively minor as the spectral classes only span three subclasses from G2V stars; the overall slope change for the asteroids using G3V standard stars (184 Dejopeja, 739 Mandeville, 758 Mancunia) will be trivial.

NIR spectra were reduced using either IRAF/SpecPR (Gaffey 2003) or Spextool (Cushing et al. 2004). Both programs perform the same primary functions, which include flux summation and extraction, extinction corrections, channel shifting, and spectral averaging. SpecPR performs additional analysis functions such as absorption feature continuum removal, polynomial fitting, and band center/band area calculations. Spextool, an IDL routine, does not include analysis functions. However, analysis functions equivalent to those for SpecPR are accomplished using MATLAB subroutines.

Infrared Telescope Facility/SpeX LXD observations of 22 Kalliope were also conducted for this study. Both prism mode and cross-dispersed spectra (approximately 1.9–4.1 μm) were obtained and used to compare with similar observations by Rivkin et al. (2000). The SpeX LXD data are reduced using standard NIR spectral reflection techniques, which include dark and flat field corrections, and application of a bad pixel mask. The raw cross-dispersed data is different from prism data, as it is vertically offset across six spectral orders on the CCD chip. The data were extracted as a 1-D flux array, clipped at the ends to exclude low SNR data, and only the highest SNR images were used in the final average. Error bars are the standard deviation of the mean of each channel. Wavelength calibration in the K-band uses argon lines obtained by SpeX and in the L-band using atmospheric absorption lines. The data are normalized at 2.2 μm.

Individual NIR absorption bands are isolated using a linear continuum that meets the local maximum on either side of an absorption feature. A continuum-removed feature is calculated as the ratio of the reflectance of the absorption feature to the linear continuum. A polynomial is then fit to the isolated feature to estimate the band center. Reported band centers are averages of multiple polynomial fitting attempts. Band error estimates are dominated by the point-to-point data scatter of the feature, but are typically smaller for approximately 0.9–1.0 μm (Band I) features than the approximately 1.9–2.0 μm (Band II) features.

Continuum-removed features were corrected for temperature effects, when necessary, because NIR mineral absorptions become narrower and/or experience a wavelength shift compared with equivalent laboratory spectra obtained at room temperature (approximately 293 K) (Singer and Roush 1985; Moroz et al. 2000; Hinrichs and Lucey 2002). Asteroid surface temperatures at the time of observation were calculated using the Thermflx program of M. J. Gaffey using the Standard Thermal Model (STM; Lebofsky and Spencer 1989) with inputs that include the semimajor axis, phase angle, albedo, emissivity (0.856), and beaming factor (0.75) across an arbitrary wavelength interval (i.e., 0.4–20.0 μm). Using the techniques in Burbine et al. (2009), all Band I temperature shifts are smaller than the errors in the reported band centers and usually amounts to a shift of only 0.001 μm. Asteroids with measurable Band II features, however, experienced substantially larger temperature-induced shifts and were corrected by the methods of Burbine et al. (2009).

Pyroxene chemistry estimates are determined from the methods of Gaffey et al. (2002) and Burbine et al. (2009), where the former is based on the work of Adams (1974) and Cloutis and Gaffey (1991). The pyroxene chemistry calculations are most precise when orthopyroxene ± Type B clinopyroxene are the sole mafic silicates present. Pyroxene chemistry estimates can be skewed by the presence of olivine and Type A clinopyroxene that can shift the Band I centers (Gaffey et al. 2002). The presence of Type A and/or Type B high-Ca clinopyroxene can be inferred based on the position of the band parameters in the pyroxene band-band plot (Gaffey et al. 2002), and Type A clinopyroxene can be directly detected by the presence of a uniquely invariant double Band I minimum in the approximately 1 μm region (Schade et al. 2004; Hardersen et al. 2006a). The Burbine et al. (2009) calibration is based on the howardite, eucrite, diogenite (HED) meteorites, where pyroxene is the dominant mafic silicate mineral present. Both techniques are used in this article for comparison.

Mineral Mixtures

Laboratory VNIR spectra of several mineral, metal, and meteorite samples were obtained to allow comparison of the laboratory spectra to the asteroid spectra. All new spectra were obtained at the University of Winnipeg using an Analytical Spectral Devices (ASD) FieldSpec Pro HR spectrometer (Cloutis et al. 2006). Absolute reflectances from 0.35 to 2.50 μm was measured for all samples, which were uniformly ground to <45 μm particle size fractions. Spectra were measured at a viewing geometry of i = 30°, e = 0°. Olivine (OLV003), orthopyroxene (PYX042), metal (Meteorite Hills [MET] 101), and CH chondrite (Pecora Escarpment [PCA] 91467) samples were used in this work. PYX042 has a chemistry of Wo0.4Fs12.8 and Al2O3 abundance of 0.09 wt%. OLV003 has a chemistry of Fo90.4. The pyroxene within PCA 91467 has a chemistry of Fs1–5 that ranges as high as Fs16 (Score and Lindstrom 1993). This sample has a weathering grade of B/C, which indicates moderate to severe weathering (Score and Lindstrom 1993). The reflectance spectra measured at the NASA-supported RELAB facility were measured at i = 30°, e = 0° in bidirectional reflectance mode relative to halon for the approximately 0.3–2.5 μm region at 5 nm intervals, and corrected for minor irregularities in halon’s absolute reflectance in the 2.0–2.5 μm region. Details of the RELAB facility are available at the RELAB website (

VNIR spectra were obtained for olivine:metal and orthopyroxene:metal mixtures to study the effects of changing relative amounts of powdered metal on the overall NIR spectral slope and the reduction of mineral absorption band depths. Mixtures were sampled at 10 wt% intervals for olivine:metal and orthopyroxene:metal mixtures. The laboratory VNIR spectra were also compared with asteroid spectra to attempt to constrain the potential abundance of metal on asteroid surfaces.


Physical and Dynamic Characteristics

As a group, the asteroids in this article do not possess any distinct or uniquely identifiable dynamical or physical characteristics that suggest they are genetically related. Of the Tholen M-asteroids in this article, 30 of 38 have an IRAS albedo within the range from 0.14 to 0.25 (Tedesco et al. 2002). Seven M-asteroids have IRAS albedos <0.14, and one (55 Pandora) has an IRAS albedo >0.30 (Tedesco et al. 2002; Shepard et al. 2010). Two of the Tholen X-/XD-asteroids have an IRAS albedo <0.14, three have an IRAS albedo in the range from 0.14 to 0.25, and one (504 Cora) has an IRAS albedo >0.30 (Tedesco et al. 2002).

Dynamically, these asteroids range in semimajor axis from approximately 2.15–3.20 AU. There is no large-scale preference of dynamical groupings of these asteroids when plotting eccentricity versus semimajor axis or sine of the orbital inclination versus semimajor axis. However, several asteroids are members of a few different asteroid families. All the olivine-bearing asteroids in this article—766 Moguntia, 798 Ruth, and 1210 Morosovia (see discussion below)—are members of the Eos dynamical family (Zappala et al. 1990). Five asteroids: 77 Frigga, 110 Lydia, 125 Liberatrix, 201 Penelope, and 224 Oceana, are members of the Kozai 30 family (Kozai 1979). Fifty-five Pandora, 441 Bathilde, and 872 Holda are members of the Kozai 36 family (Kozai 1979). Both 755 Quintilla and 758 Mancunia are members of the Kozai 60 family (Kozai 1979). A total of 325 Heidelberga and 382 Dodona are members of the Kozai 63 family (Kozai 1979). 338 Budrosa and 558 Carmen are members of the Williams 124 family (Williams 1979, 1989, 1992). It is important to note that common family membership is not a guarantee of a direct genetic relationship due to the presence of interlopers.

Pyroxene-Bearing Asteroids

Seventeen Tholen M/MU-asteroids, two Tholen X-asteroids, and one Tholen XD-asteroid exhibit weak approximately 0.9 μm (Band I) ± approximately 1.9 μm (Band II) absorptions with continuum-removed band depths that range from approximately 1 to 14%. Table 2 lists these asteroids with their band parameters and complementary data, such as radar albedo and 3 μm absorptions. The band parameters for the six asteroids with weak pyroxene features from Hardersen et al. (2005) have been remeasured and included in Table 2. Table 3 includes the band parameters for the five asteroids with Band I and Band II absorptions, which include temperature-corrected band centers and the derived pyroxene chemistries using the Gaffey et al. (2002) and Burbine et al. (2009) calibrations. Figures 1 and 2 show the average NIR spectrum for each pyroxene-bearing asteroid, except for 857 Glasenappia. Figure 3 shows the continuum-removed Band I absorptions for 250 Bettina, 347 Pariana, 497 Iva, and 516 Amherstia, which are typical Band I absorptions for these asteroids.

Table 2.   Physical, orbital, and taxonomic characteristics of the pyroxene-bearing M-/X-asteroids.
NameTholen classIRAS diameter (km)IRAS albedoa (AU) Band I (μm) Band I depth (%)Band II (μm)Radar albedo3 μm feature?IRAS albedo
  1. Taxonomic classifications from Tholen (1984, 1989).

  2. IRAS albedo/diameter from Tedesco et al. (2002).

  3. 3 μm data from Rivkin et al. (2000).

  4. a129 Antigone albedo from Shepard et al. (2008) and references therein.

  5. bData from Shepard et al. (2010) and references therein. Morrison (1977) reports a radiometric albedo = 0.18 ± 0.04. Radar data from Shepard et al. (2010). Band II centers listed in Table 2 are the nontemperature-corrected values.

  6. cData from Band II averages from Hardersen et al. (2005).

16 PsycheM253.160.12032.9210.932 ± 0.008∼1–20.42 ± 0.10No0.1203
69 HesperiaM138.130.14022.9890.923 ± 0.011; 0.909 ± 0.009∼21.78–1.83c0.1402
110 LydiaM86.090.18082.7320.914 ± 0.006; 0.914 ± 0.004; 0.903 ± 0.005∼21.75–1.90c0.20 ± 0.12Yes0.1808
125 LiberatrixM43.580.22532.7430.920 ± 0.006∼2–3No0.2253
129 AntigoneM0.18a2.8670.918 ± 0.003; 0.930 ± 0.008∼30.36 ± 0.09Yes0.1800
184 DejopejaX66.470.18973.1790.931 ± 0.004∼2No0.1897
201 PenelopeM68.390.16042.6780.932 ± 0.005; 0.945 ± 0.008; 0.917 ± 0.010∼21.82–1.83cYes0.1604
216 KleopatraM135.070.11642.7910.923 ± 0.003∼21.79–1.80c0.60 ± 0.15No0.1164
250 BettinaM79.750.25813.1480.914 ± 0.007∼20.2581
338 BudrosaM63.110.17662.9110.939 ± 0.008∼1–21.933 ± 0.0180.1766
347 ParianaM51.360.18452.6120.919 ± 0.008∼2–30.36 ± 0.090.1845
369 AeriaM60.000.19192.6490.920 ± 0.003∼21.865 ± 0.023No(?)0.1919
382 DodonaM58.370.16103.1190.934 ± 0.007∼1–20.1610
417 SueviaX40.690.19602.807See text for discussion∼20.1960
418 AlemanniaM34.100.18782.5930.922 ± 0.010; 0.910 ± 0.005∼2–30.1878
497 IvaM31 ± 60.085 ± 0.010b2.8500.941 ± 0.006∼1–21.918 ± 0.0190.24 ± 0.08No0.0850
516 AmherstiaM73.100.16272.6760.927 ± 0.003∼4–51.911 ± 0.0390.1627
558 CarmenM59.310.11612.9070.940 ± 0.006∼2–30.1161
796 SaritaXD44.960.19662.6350.914 ± 0.008∼21.846 ± 0.0270.25 ± 0.10No0.1966
857 GlasenappiaMU15.030.23182.190Flat feature rom ∼0.94 to 1.08 μm∼14%1.958 ± 0.028No(?)0.1572
Table 3.   Band parameters and pyroxene chemistry estimates for 338 Budrosa, 369 Aeria, 497 Iva, 516 Amherstia, and 796 Sarita. Band centers have been temperature-corrected via the methods in Burbine et al. (2009).
AsteroidBand I (μm)Band II (μm)Gaffey et al. (2002)
Pyroxene chemistry
Burbine et al. (2009)
Pyroxene chemistry
338 Budrosa0.9391.919Wo12Fs38Wo8Fs39
369 Aeria0.9201.886Wo3Fs22Wo3Fs26
497 Iva0.9391.937Wo11Fs39Wo9Fs41
516 Amhertia0.9281.898Wo8Fs25Wo5Fs31
796 Sarita0.9141.846Wo0Fs11Wo0Fs18
Figure 1.

 Average NIR reflectance spectra of 16 Psyche, 69 Hesperia, 110 Lydia, 125 Liberatrix, 129 Antigone, 184 Dejopeja, 201 Penelope, 216 Kleopatra, 250 Bettina, and 338 Budrosa.

Figure 2.

 Average NIR reflectance spectra of 347 Pariana, 369 Aeria, 382 Dodona, 417 Suevia, 418 Alemannia, 497 Iva, 516 Amherstia, 558 Carmen, and 796 Sarita.

Figure 3.

 Continuum-removed Band I absorption features for 250 Bettina, 347 Pariana, 497 Iva, and 516 Amherstia. Band depths for the pyroxene-bearing asteroids in this article range from approximately 1 to 5%.

For the asteroids in Table 2 that only exhibit a Band I absorption, the Band I centers for all but two asteroids (417 Suevia and 857 Glasenappia) range from approximately 0.89 to 0.95 μm. Without a measurable Band II feature, it is not possible to use the existing calibrations to estimate the average surface pyroxene chemistry for each asteroid. However, it is instructive to note that all but three of the orthopyroxene laboratory samples measured for the pyroxene band–band plot from Gaffey et al. (2002) have Band I centers ≤0.936 μm. In Table 1, nine asteroids (69 Hesperia, 110 Lydia, 125 Liberatrix, 129 Antigone, 216 Kleopatra, 250 Bettina, 347 Pariana, 418 Alemannia, 796 Sarita) have Band I centers <0.930 μm. While a few higher-Ca Type B clinopyroxene samples in the band–band plot have band centers <0.930 μm, the vast majority have band centers that range from approximately 0.95 to 1.07 μm (Gaffey et al. 2002). This suggests that most of these asteroids have a surface mineralogy dominated by orthopyroxene or an orthopyroxene/Type B clinopyroxene mixture.

Sixteen Psyche, 201 Penelope, 382 Dodona, and 558 Carmen have Band I centers that range from 0.930 to 0.945 μm. Compared with the asteroids showing the shorter-λ Band I centers, these asteroids have a higher proportion of Type B clinopyroxene on their surfaces. 417 Suevia’s Band I feature has a relatively flat bottom that extends from approximately 0.82 to 1.17 μm, which prevents a determination of a Band I center. The breadth of this feature suggests a multipyroxene ± olivine surface mineralogy that accounts for the breadth of this feature and the presence of multiple overlapping weak absorptions (Cloutis and Gaffey 1991; Gaffey et al. 2002; Schade et al. 2004).

Olivine-Bearing Asteroids

The average NIR spectral and physical characteristics of the three Tholen MU-/MU:-olivine-bearing asteroids (766 Moguntia, 798 Ruth, 1210 Morosovia) are shown in Table 4 and Fig. 4. Figure 4 also shows the continuum-removed Band I feature for 766 Moguntia. Band I centers range from 1.047 to 1.068 μm and band depths range from approximately 4 to 10%. The three asteroids are members of the Eos dynamical family (Zappala et al. 1990) and show similar NIR olivine features in the work of Birlan et al. (2007) and Mothé-Diniz et al. (2008).

Table 4.   Physical, orbital, and taxonomic characteristics of the olivine-bearing M-asteroids.
NameTholen classIRAS D (km)IRAS albedoa (AU)Band I (μm)Band I depth (%)FamilyFo %
  1. Family classifications from Zappala et al. (1990).

  2. Forsterite % calculated from Reddy et al. (2011).

  3. IRAS diameters and albedos from Tedesco et al. (2002).

766 MoguntiaMU31.280.15723.0221.068 ± 0.004∼10%Eos60
798 RuthMU43.190.15873.0151.056 ± 0.004∼4–5%Eos84
1210 MorosoviaMU33.650.16953.0121.047 ± 0.009∼8–9%Eos
Figure 4.

 Average NIR reflectance spectra for 766 Moguntia, 798 Ruth, and 1210 Morosovia, and the Band I continuum-removed absorption for 766 Moguntia.

Olivine calibrations using derived Band I centers are poorly constrained due to the small change in band center with increasing Fe2+ content in olivine (King and Ridley 1987; Burns 1993; Reddy et al. 2011). However, applying the recent calibration from Reddy et al. (2011) yields olivine chemistry values of Fo60±7 for 766 Moguntia and Fo84±7 for 798 Ruth showing that these asteroids have rather Mg-rich olivine chemistries. The Band I center for 1210 Morosovia is at a shorter-λ due to a pyroxene component, and olivine calibrations cannot be used (Reddy et al. 2011).

Mothé-Diniz et al. (2008) report Fo∼85–95 for 766 Moguntia and Fo∼40–70 for 798 Ruth, which are somewhat more Mg-rich and Mg-poor, respectively, compared with the results in Table 3. Mothé-Diniz et al. (2008) reported a larger range of Fo values for Eos family members and suggest analogs that include R chondrites, CK chondrites, and partial melts. Birlan et al. (2007) note the weak Band I absorption feature for 766 Moguntia and attribute the relatively shallow band depth to opaque minerals on Moguntia’s surface that were suppressing the olivine absorption.

Hydrated Mineral-Bearing Asteroids

Three Tholen M-asteroids: 22 Kalliope, 55 Pandora, and 132 Aethra, and an S3OS2-Tholen X-asteroid (Lazzaro et al. 2004), 504 Cora, display NIR spectral slopes and absorption features consistent with the presence of phyllosilicate ± hydroxide minerals. Table 5 and Fig. 5 display the dynamical and physical characteristics, and average NIR spectra of these asteroids, respectively.

Table 5.   Physical, orbital, and taxonomic characteristics of the phyllosilicate/hydroxide-bearing M-/X-asteroids.
NameTholen classIRAS D (km)IRAS albedoa (AU)Band minima (μm)3 μm feature?
  1. 3 μm data from Rivkin et al. (2000).

22 KalliopeM181.000.14192.9090.94–(?), 2.30–2.50+Yes
55 PandoraM66.700.30132.7600.85–(?), 0.92–(?), 0.97–(?) 1.10Yes
132 AethraM42.660.17182.6100.91–, 1.10–, 1.40–, 2.28–2.50+
504 Cora30.020.34072.7210.85–, 0.91–, 1.07–
Figure 5.

 Average NIR reflectance spectra for 22 Kalliope, 55 Pandora, 132 Aethra, and 504 Cora. The 22 Kalliope spectra includes L-band data that increases the spectral coverage to approximately 4.1 μm.

132 Aethra was observed in May and August 2008, respectively, and a comparison of their average spectra is shown in Fig. 6. The May 2008 spectrum is an average of 20 spectra, and the August 2008 spectrum is an average of 10 spectra.

Figure 6.

 Overlay of the average spectrum for 132 Aethra from May 2008 (20 spectra) and August 2008 (10 spectra). The steep red slope and weak features in the May 2008 spectrum are absent from the August 2008 spectrum. Unpublished spectra from J. Emery (December 2010–February 2011) are also featureless and more similar to the August 2008 spectrum.

The May 2008 spectrum has a significantly steeper slope than the August 2008 spectrum and exhibits weak absorptions at approximately 0.90, 1.10, and 1.40 μm, along with a more prominent and broader absorption ranging from approximately 2.28 to 2.50+ μm. The August 2008 spectrum has a shallower, but significantly reddish slope; however, it is largely absent the features in the May 2008 spectrum. The features in the May 2008 NIR spectrum at approximately 0.90 and 1.10 μm can be assigned to Fe2+ crystal field transitions and the approximately 1.40 and 2.28–2.50 μm absorptions can be assigned to Mg-OH, H2O, or OH vibrational overtones (Burns 1993; Calvin and King 1997). The spectral differences from the two observing runs can possibly be attributed to variable abundances of surface phyllosilicates with rotation, variable abundances of opaque surface minerals with rotation, the presence of an eclipsing satellite with a different surface mineralogy than the primary, or unidentified observational and instrumental artifacts. A rigorous review of the data does not indicate an obvious observational or instrumental reason(s) to explain the spectral differences.

Comparison of the May 2008 spectrum with multiple RELAB samples of chamosite show the best slope and absorption feature match with Fe-rich sample CHM102. The spectral match is very close from approximately 1.2 to 2.5 μm. The ground chamosite sample has a purity of 98%; no other minerals were identified via X-ray diffraction analysis. The NIR spectra of Fe-poor chamosites are significantly less red (i.e., shallower slopes) than CHM102 (by a factor of 4+), while still displaying multiple weak absorption features attributable to Fe2+, Fe3+, OH, and H2O.

CHM102 displays weak absorptions at approximately 0.70 (Fe2+-Fe3+ IVCT), 0.90 (Fe2+ crystal field transition) and 1.10 μm (Fe2+ crystal field transition), and stronger absorptions at approximately 1.92 μm (H2O), 2.34 μm (metal-OH), and 2.51 μm (metal-OH and/or H2O). Mg-bearing phyllosilicates are characterized by a broad Fe2+ absorption from 0.6 to 1.8 μm that is superimposed on a red NIR slope and includes a weak Fe2+- Fe3+ IVCT absorption at approximately 0.70 μm and weak Fe2+ crystal field transitions at approximately 0.90 and 1.10 μm (Calvin and King 1997). Chamosite is an intermediate Fe-/Mg-bearing phyllosilicate (Fe2+, Mg, Fe3+)5Al(Si3Al)O10(OH,O)8 from the chlorite group and exhibits greater spectral reflectance with increasing Al content (Calvin and King 1997). This relative brightness is consistent with 132 Aethra’s IRAS albedo of approximately 19% (Tedesco et al. 2002) and would more easily allow the appearance of weak absorption features in the NIR spectrum.

Visible-λ spectra from Bus and Binzel (2002) and Fornasier et al. (2010) show an approximately 0.49 μm feature possibly attributable to oldhamite (Burbine et al. 2002); this feature is also seen in the spectra of some E-type asteroids (Clark et al. 2004; Fornasier et al. 2007, 2008). The coexistence of sulfides and phyllosilicates does occur terrestrially in some ore deposits (King and Clark 1989). Some RELAB samples of glauconite, nontronite, and saponite display a similar absorption feature, which is consistent with a hydrated mineral environment (Stewart et al. 2006).

The average NIR spectrum of 504 Cora is shown in Fig. 5, and a magnified view of the Band I region is shown in Fig. 7. Reynolds (2007) and Reynolds et al. (2007) initially reported 504 Cora as a low-Fe pyroxene-bearing asteroid, but reanalysis reveals what is seen in Fig. 7, which is a set of weak, narrow absorptions at approximately 0.85 and 0.92 μm that are superimposed on a broader absorption, and another weak feature at approximately 1.08 μm. The weak approximately 1.6 μm absorption in Fig. 5 is telluric.

Figure 7.

 The 0.6–1.2 μm region of the SMASS II visible-λ (open squares; Bus and Binzel 2002) and average IRTF/SpeX NIR spectrum (black diamonds) for 504 Cora. The broader feature in the figure occurs at the overlap of the SMASS II and SpeX data sets. Note the distinct, weak absorptions at 0.85, 0.92, and 1.08 μm in the NIR spectrum. The absorptions are attributed to Fe2+ and Fe3+ crystal field absorptions found in hydroxide and phyllosilicate minerals.

The relatively high reported IRAS albedo for 504 Cora, approximately 30%, is inconsistent with a metal and pyroxene mixture (Cloutis et al. 2010). The approximately 0.85 μm feature is consistent with Fe3+ spin-forbidden bands in hydroxide/oxide minerals such as ferrihydrite and hematite. The same feature has also been reported in the laboratory spectrum of CM carbonaceous chondrites (Gaffey and McCord 1979). The approximately 0.92 and 1.08 μm weak absorptions could be attributed to Fe2+-bearing phyllosilicate (Calvin and King 1997).

55 Pandora has spectral and albedo characteristics that are very similar to those of 504 Cora. The NIR spectrum is broadly similar to that for 504 Cora, most notably in the approximately 0.7–1.2 μm spectral region. While the Band I region for 55 Pandora is somewhat noisier than 504 Cora’s data, there are suggestions of weak absorptions at approximately 0.85, 0.92, 0.97, and 1.10 μm. The 0.97 μm feature is seen in the spectrum of some CM chondrites (Cloutis et al. 2011), whereas the other features are consistent with phyllosilicate phases, as discussed previously.

55 Pandora and 504 Cora have similar IRAS albedos within the range from 0.30 to 0.35. A 3 μm feature has also been detected for 55 Pandora (Jones et al. 1990; Rivkin et al. 2000). The combination of weak absorptions that are indicative of hydroxide/oxide and phyllosilicate minerals suggest a surface mineral environment similar to 504 Cora’s—a heavily hydrated, relatively bright surface that argues for extensive aqueous alteration of the parent asteroid.

An average spectrum for 22 Kalliope is shown in Fig. 5 and includes a weak, yet broad absorption in the approximately 2.30–2.50 μm region that can be attributed to Mg-OH vibrational overtones seen in Mg-serpentines and chamosite (Calvin and King 1997). There is a possible weak feature at approximately 0.94 μm, but the continuum-removed feature seems too narrow for attribution to pyroxene. There is no indication of the weak Fe2+ and Fe3+ absorptions that have been identified in the May 2008 NIR spectrum of 132 Aethra. It is also notable that the spectral slope for 22 Kalliope is not extremely red, as is the case for 132 Aethra and the CHM102 sample.

L-band data from approximately 1.9–4.1 μm were also obtained on the same night. While the L-band spectrum shows a weak absorption in the 3 μm region, the weak feature in the SpeX data from 2.3 to 2.5 μm is absent. The difference in the time of observation of 22 Kalliope for the NIR and L-band data was <1 h, and Kalliope’s surface would have rotated <90° based on its rotation period of 4.148 h (Warner 2007). A separate NIR spectrum of 22 Kalliope obtained on May 9, 2004 UT did not show any indication of a 2.30–2.50 μm absorption (Hardersen et al. 2006b). Rivkin et al. (2000) reported a 3 μm absorption for 22 Kalliope and the possibility of band depth variations.

22 Kalliope has a known satellite companion (Margot and Brown 2003). An initial bulk density of 2.37 ± 0.4 g cm−3 was reported (Margot and Brown 2003), but was later revised upward to 3.35 ± 0.33 g cm−3 (Descamps et al. 2008). The former density is consistent with the bulk densities of some CM chondrites, but is higher than measured for CI chondrites (Britt and Consolmagno 2003). The higher value is inconsistent with both CI and CM chondrites (Britt and Consolmagno 2003). 22 Kalliope’s IRAS albedo (approximately 14%) is also inconsistent with CI and CM chondrite albedos (Gaffey 1976). Chlorite- and serpentine-group minerals have densities ranging from 2.5 g cm−3 to 3.3 g cm−3 (Klein and Hurlbut 1993).

Spectrally Featureless Asteroids

Fifteen Tholen M-asteroids, one Tholen S-asteroid, and two Tholen X-asteroids in this article are spectrally featureless in the NIR. Average NIR spectra are shown in Figs. 8 and 9 and related characteristics in Table 6. As seen in Figs. 8 and 9, these asteroids exhibit a wide variety of spectral slopes. Most of the asteroids in Table 5 exhibit a generally red NIR slope across most or all of the wavelength range from approximately 0.8 to 2.5 μm. However, the spectra of 97 Klotho and 441 Bathilde are flat, and the spectrum of 498 Tokio is distinctly bluish with a pronounced negative slope.

Figure 8.

 Average NIR reflectance spectra of 21 Lutetia, 71 Niobe, 77 Frigga, 97 Klotho, 135 Hertha, 136 Austria, 224 Oceana, 325 Heidelberga, and 413 Edburga.

Figure 9.

 Average NIR reflectance spectra of 441 Bathilde, 498 Tokio, 739 Mandeville, 758 Mancunia, 785 Zwetana, 860 Ursina, 872 Holda, 931 Whittemora, and 1461 Jean-Jacques.

Table 6.   Physical, orbital, and taxonomic characteristics of the near-infrared spectrally featureless M-/X-asteroids.
NameTholen classIRAS D (km)IRAS albedoa (AU)1.8/0.8 μm ratioRadar albedo3 μm feature?
  1. Radar data from Shepard et al. (2008, 2010).

  2. 3 μm data from Rivkin et al. (2000).

21 LutetiaM95.760.22122.4351.1090.24 ± 0.07Yes
71 NiobeS83.420.30522.7561.079/1.1000.19 ± 0.05
77 FriggaMU69.250.14402.6711.381Yes
97 KlothoM82.830.22852.6660.981/1.0040.26 ± 0.05
135 HerthaM79.240.14362.4291.2770.18 ± 0.05Yes
136 AustriaM40.140.14592.2871.137Yes
224 OceanaM61.820.16942.6451.2670.25 ± 0.10
325 HeidelbergaM75.720.10683.2011.4140.17 ± 0.08
413 EdburgaM31.950.14662.5831.289
441 BathildeM70.320.14102.809
498 TokioM81.830.06942.6510.754
739 MandevilleX107.530.06082.742
758 MancuniaX85.480.13173.1900.55 ± 0.14No
785 ZwetanaM48.540.12452.5731.1700.33 ± 0.08No
860 UrsinaM29.320.16182.7961.215/1.292
872 HoldaM30.040.21272.7311.261
931 WhittemoraM45.270.17043.1781.322
1461 Jean-JacquesM32.940.16133.1281.452

Various reflectance ratios and slopes have been measured and analyzed for featureless M-/X-asteroid NIR reflectance spectra (Ockert-Bell et al. 2008, 2010; this work) and laboratory NIR spectra for plausible metallic surfaces (powdered and slab) for M-asteroids (Cloutis et al. 2010; and references therein). The asteroids in Table 5 have 1.8/0.8 μm reflectance ratios that range from 0.754 to 1.452 with 10 of the asteroids having ratios > 1.20. Reflectance ratios for various metallic powders and slabs range from approximately 1.0–1.7 (Cloutis et al. 2010). The blue NIR spectral slope for 498 Tokio and reflectance ratio = 0.754 is inconsistent with the metal NIR spectra from Cloutis et al. (2010) and the NIR spectrum in Fornasier et al. (2010). The remaining asteroids in Table 6 are consistent with both works.

Meteorite analogs for spectrally featureless asteroids are more difficult to constrain, but enstatite chondrites and NiFe meteorites—assuming moderate albedos from approximately 10 to 20%—remain viable options. Particulate metallic surfaces require red spectral slopes and albedos of approximately 10–15% at 0.56 μm (Cloutis et al. 2010). If IRAS albedos (Tedesco et al. 2002) are equivalent to 0.56 μm albedos, then 21 Lutetia, 71 Niobe, 97 Klotho, 498 Tokio, 739 Mandeville, 872 Holda—and possibly 931 Whittemora and 1461 Jean-Jacques—are not likely to have metallic surfaces.

The recent Rosetta flyby of 21 Lutetia revealed an asteroid with a regolith of varying thickness and an albedo of approximately 19% (consistent with Mueller et al. 2006; inconsistent with Morrison and Zellner 1979), suggesting that the surface is not metallic (Barucci et al. 2010). Shepard et al. (2008, 2010) derived a radar albedo of 0.24 ± 0.07 for 21 Lutetia, which is consistent with enstatite chondrites, CH chondrites, and silicate-bearing iron meteorites. CH chondrites display a weak approximately 0.9 μm pyroxene absorption (i.e., PCA 91467: this work); silicate-bearing iron meteorites may display a pyroxene absorption depending on the surface pyroxene abundance. Orthopyroxene/metal laboratory mixtures produce a weak approximately 0.9 μm absorption feature with a 10/90 wt% mixture of orthopyroxene and powdered metal. The available data currently suggests that enstatite chondrites are the most viable meteorite analog for 21 Lutetia with silicate-bearing iron meteorites (with <10 wt% opx/metal mixture) as another possibility.

Radar albedos have also been obtained for 71 Niobe, 97 Klotho, 135 Hertha, 224 Oceana, 325 Heidelberga, 758 Mancunia, and 785 Zwetana (Shepard et al. 2008, 2010), among which 758 Mancunia and 785 Zwetana have relatively high, and variable, radar albedos, which makes the iron meteorite analog applicable to these asteroids (Shepard et al. 2008, 2010). Ockert-Bell et al. (2008) found a correlation between their “NIR2” continuum slope (1.70–2.45 μm) and radar albedo, which may assist in identifying metal-rich asteroids via NIR spectra.

135 Hertha, 224 Oceana, and 325 Heidelberga have radar and NIR spectral properties consistent with enstatite chondrites, but CH chondrites can be ruled out, as discussed above (Shepard et al. 2008, 2010). 71 Niobe and 97 Klotho are mostly consistent with enstatite chondrites, but their IRAS albedos are higher than typically measured for this meteorite type (Gaffey 1976).

857 Glasenappia

This asteroid is unique in this study, because it is the only asteroid with two well-defined and relatively deep absorptions in the approximately 0.9 and 1.9 μm regions, akin to the NIR spectral characteristics of S-asteroids (Gaffey et al. 1993); see Fig. 10. The continuum-removed Band I absorption is quite broad and extends from approximately 0.80 to 1.60 μm. The bottom of the continuum-removed feature is relatively flat and extends from approximately 0.92 to 1.08 μm with an inflection at approximately 1.20 μm. A Band I center could not be determined due to the flatness of the feature, but the breadth of the feature suggests the presence of overlapping absorptions from at least two distinct mafic silicate minerals. The Band II center = 1.958 ± 0.028 μm. The calculated band area ratio (BAR: Band II/Band I) = 0.64.

Figure 10.

 Average NIR reflectance spectrum of 857 Glasenappia.

From Cloutis et al. (1986), the BAR suggests an approximately 68% relative olivine abundance in an olivine/orthopyroxene mixture on the asteroid’s surface. The rounded Band I feature is also consistent with the S-II asteroids in Gaffey et al. (1993), who note that the shape of this feature is mostly unique to this S-asteroid subtype. The position of the Band II center is consistent with an average surface pyroxene that is either mostly orthopyroxene or an orthopyroxene/Type B clinopyroxene mixture (Gaffey et al. 2002).

Possible meteorite analogs for 857 Glasenappia include L and LL chondrites, which is based on the BAR and Band II center (Gaffey 1976; Gaffey et al. 2002) similarities between the derived olivine abundance and the normative abundances from McSween et al. (1991). Other potential analogs include clinopyroxene-bearing ureilites, brachinites, and olivine/clinopyroxene cumulates (Gaffey et al. 1993).

Analysis and Interpretations

Pyroxene-Bearing Asteroids

For the pyroxene-bearing asteroids that exhibit a measurable Band I feature, Hardersen et al. (2005) offered four possible interpretations: (1) low-Fe pyroxene mantling metallic cores, (2) CB/CH chondrites, (3) smelting products, and (4) collisional debris. The presence of weak pyroxene absorptions rules out the enstatite chondrite and iron meteorite interpretations for these asteroids (Hardersen et al. 2005, and references therein).

The M-/X-asteroids with shorter-λ Band I centers (<0.93 μm) would be consistent with the low-Fe pyroxene mantle/core and CB/CH-chondrite interpretations due to their inferred low Fe2+ content (Brearley and Jones 1998). A continuum-removed absorption for CH-chondrite, PCA 91467, has a band center of approximately 0.92 μm and a band depth of approximately 3%, which is consistent with many of the asteroid band centers in Table 2.

Conversely, those asteroids with Band I centers >0.94 μm (16 Psyche, 201 Penelope, 417 Suevia, 558 Carmen) would be inconsistent with CB/CH chondrites and represent a more Fe-rich pyroxene environment. The asteroids in Table 2 are also inconsistent with CB/CH chondrites as they are more Fe-rich than typically seen in CB/CH chondrites (Brearley and Jones 1998).

The pyroxene-bearing asteroids with Band I and II centers, shown in Table 3, are plotted on the pyroxene band–band plot in Fig. 11. The band centers for 369 Aeria and 516 Amherstia plot within the orthopyroxene region (Gaffey et al. 2002). The derived pyroxene chemistries for 369 Aeria are Wo3Fs22 (Gaffey et al. 2002) and Wo3Fs26 (Burbine et al. 2009). The derived pyroxene chemistries for 516 Amherstia are Wo8Fs25 (Gaffey et al. 2002) and Wo5Fs31 (Burbine et al. 2009). These estimates suggest that orthopyroxene is the dominant mafic silicate mineral on these asteroids’ surfaces, although minor/accessory amounts of Type B clinopyroxene and olivine may also be present (Cloutis et al. 1986). The 338 Budrosa and 497 Iva plot slightly above the pyroxene trend in Fig. 11. This suggests the presence of either an olivine or high-Ca pyroxene phase in addition to orthopyroxene ± Type B clinopyroxene (Gaffey et al. 2002).

Figure 11.

 Pyroxene Band I versus Band II plot with band center data for 338 Budrosa, 369 Aeria, 497 Iva, 516 Amherstia, and 796 Sarita.

369 Aeria displays an NIR spectral slope that is similar to that seen in some mesosiderite NIR spectra (Gaffey 1976). The asteroid’s average pyroxene chemistry is also consistent with the pyroxene chemistry range for some mesosiderites (Weigand 1975; Hewins 1979) and diogenites (Rubin 1997; Mittlefehldt et al. 1998). Weak Band I and Band II features would be expected for an asteroid with a surface composition of orthopyroxene and metal (Vernazza et al. 2009). Laboratory mixtures of orthopyroxene and metal (10/90 wt%) produce a VNIR spectrum with weak pyroxene features, but the slope of the laboratory spectrum is much steeper than the 369 Aeria spectrum. Mesosiderite NIR spectra will vary significantly depending on the proportions of pyroxene and metal (Burbine et al. 2007).

Vernazza et al. (2009) suggested 201 Penelope, 250 Bettina, and 337 Devosa as potential mesosiderite-like asteroids. NIR spectra of untreated and ion-irradiated samples of the mesosiderite Vaca Muerta were compared with spectra of the three asteroids. Associations were suggested based on the similarities between the continuum-removed spectra of the meteorite and asteroids, similar albedos, and the presence of weak pyroxene features in the approximately 0.9 and 1.9 μm regions. The Band I centers reported for 250 Bettina by Ockert-Bell et al. (2010), Fornasier et al. (2010), and this work are inconsistent with HED band parameters (Gaffey 1997). For 201 Penelope, the Band I and Band II centers reported by Hardersen et al. (2005, this work) are at wavelengths inconsistent with HED band parameters and chemistries (Gaffey 1997).

516 Amherstia exhibits a pyroxene chemistry somewhat more Fe-rich than 369 Aeria and has the deepest Band I feature of the pyroxene-dominated asteroids in Tables 2 and 3. Spectrally, the asteroid looks to be a composite of an S-type and an M-type spectrum, which suggest that a NIR spectral continuum may exist between these two taxonomic classes. The asteroid’s spectrally derived Fs content is more Fe-rich than ordinary chondrites (Rubin 1997), but overlaps the upper end of the Fs range for diogenites and mesosiderite pyroxenes (Weigand 1975; Hewins 1979; Mittlefehldt et al. 1998).

338 Budrosa and 497 Iva have larger calculated Fs values (approximately 39–40%), but their positions slightly above the pyroxene trend line in the band–band plot suggest additional mafic silicate phases (i.e., olivine) that may be causing an over-estimate of the Fs abundance. Shepard et al. (2010) suggest stony-iron meteorites (i.e., mesosiderites, pallasites) and CH chondrites as potential analogs for 497 Iva. Pallasites can be ruled out due to the lack of abundant pyroxene within pallasites (Mittlefehldt et al. 1998) and an NIR pallasite spectrum that should be dominated by olivine. The derived pyroxene chemistry for 497 Iva, even taking into account an olivine phase, will probably be too Fe-rich for CH chondrites. Mesosiderites remain a possible analog.

796 Sarita displays the lowest-Fe pyroxenes of these asteroids with a surface mineralogy that is dominated by orthopyroxene. The moderate radar albedo (0.25 ± 0.10) led Shepard et al. (2010) to suggest mesosiderites as a potential analog. The derived Fs contents (Fs11–18) for 796 Sarita in Table 3 overlap the most Fe-poor mesosiderite pyroxenes (Mittlefehldt et al. 1998) and could represent a mesosiderite-like object with diogenitic surface pyroxenes. Mesosiderite asteroids with higher-Fe pyroxenes, however, can be ruled out. Mesosiderites represent a wide range of metal abundances (17–90 wt%; Mittlefehldt et al. 1998). As 10/90 wt% mixtures of opx/metal produce a weak Band I pyroxene feature, this suggests that the vast majority of mesosiderite bodies in the asteroid belt will exhibit Band I/II pyroxene absorptions with varying band depths.

Olivine-Bearing Asteroids

766 Moguntia, 798 Ruth, and 1210 Morosovia, as well as the Eos family asteroids in Birlan et al. (2007) and Mothé-Diniz et al. (2008), display weak olivine ± pyroxene absorption features. Only 1210 Morosovia displays a weak Band II feature, which indicates a surface pyroxene component. Possible mechanisms for this band weakening include surface metal, opaque minerals, or space weathering (Cloutis et al. 1990; Clark et al. 2002; Birlan et al. 2007). Olivine/metal mixtures with olivine relative abundances of approximately 50–80 wt% reduce olivine band depths to approximately 4–13%. This is compared with a band depth of approximately 45% for a pure olivine NIR spectrum. This result is consistent with the band depths of the olivine-bearing asteroids in this work. Progressively increasing amounts of metal, beginning around 60/40 wt% olivine/metal mixtures, imparts a red slope to the NIR spectrum that is not observed in the asteroid NIR spectrum (Birlan et al. 2007; Mothé-Diniz et al. 2008; this work).

Laboratory spectra of brachinites (i.e., Brachina), ureilites (18 RELAB samples1), CO3 chondrites (ALH 82101, Frontier Mountain [FRO] 95002, FRO 99040, MET 00737), CK (3–5) chondrites (MET 01149, PCA 91470, DAV 92300, ALH 85002, Elephant Moraine [EET] 83311), a CV3 chondrite (Queen Alexandra Range [QUE] 93744), and an R3 chondrite (PRE 95404) produce weak Band I ± Band II features similar to the Eos family asteroids. Band areas and centers for these meteorite samples were measured to compare band parameters with the olivine-bearing asteroid spectra. CO3 chondrite Band I centers range from 1.059 to 1.069 μm and Band II centers range from 1.940 to 2.044 μm. About 8 vol% of bulk CO3 chondrites contain olivine and pyroxene, whereas 40–50 vol% of the matrix consists of olivine and pyroxene (Brearley and Jones 1998). Besides metal, opaque phases in the groundmass include magnetite and pyrrhotite (Brearley and Jones 1998). Olivine and pyroxene chemistries range from Fa∼1–60 and Fs<10 in unequilibrated CO3 chondrites, but Fa∼35–60 in more equilibrated CO3 chondrites (Brearley and Jones 1998). The Band II center for 1210 Morosovia, 1.883 ± 0.049 μm is consistent with an orthopyroxene component (Gaffey et al. 2002), but is at a somewhat shorter-λ compared with the CO3 chondrite Band II centers.

CV3 chondrite, QUE 93744, has a Band I center = 1.063 μm. CV chondrite chondrule olivine chemistries are generally Mg-rich as, for example, the mean Allende olivine chondrule chemistry is Fo762 (Brearley and Jones 1998). Fe-bearing pyroxenes have a wide chemistry range and include clinoenstatite, low-Ca pyroxene, and augite (Brearley and Jones 1998). Matrix olivine is generally ferroan (often mean Fa∼50), but the chemistry depends on the degree of equilibration and the amount of aqueous alteration (Brearley and Jones 1998). Meteorite albedo values (Gaffey 1976) are below those for all Eos family members except one (513 Centesima) (Mothé-Diniz et al. 2008).

CK chondrite Band I centers range from 1.064 to 1.068 μm. Four of the five samples have a Band II absorption, but two of the features are very weak. The Band II centers range from 1.980 to 2.075 μm. Band I depths vary from 9.4 to 15.1%. CK chondrites are thermally metamorphosed consisting mostly of matrix with an olivine chemistry of Fo67–71, a pyroxene chemistry of Fs23–29, and an opaque matrix component (mostly magnetite) of approximately 1–8 vol% (Mittlefehldt et al. 1998). The sole CK chondrite sample, EET 83311, lacking a Band II center has an olivine chemistry of Fo64±7 (Reddy et al. 2011).

A single R chondrite, PRE 95404, yields a Band I center = 1.064 μm, Band II center = 2.163 μm, and a BAR = 0.24. The band center data for PRE 95404 produces an olivine chemistry of Fo68±7, which overlaps the known olivine chemistry of R chondrites, Fo60–63 (Brearley and Jones 1998; Reddy et al. 2011). R chondrite pyroxene mineralogy includes low-Ca pyroxene, augites, and diopside; the Band II center suggests that the augite and diopside mineralogy is dominating the Band II center, and is inconsistent with the Band II center for 1210 Morosovia.

The single sample of Brachina has a Band I center = 0.04), ureilites (0.19–0.67), CO3 chondrites (0.58–0.85), and the CV3 chondrite (0.76).

Mothé-Diniz et al. (2008) suggest R chondrites as the best analog for the Eos family, while suggesting that CR and CV chondrites remain possible options. This analysis shows that the CO3 chondrite band parameters are most consistent with the Eos family band parameters (i.e., Band I/II centers, band depths, BAR, estimated mineral chemistries).

If the Eos family asteroids originate from a single parent body that formed in a specific region of the solar nebula and the family members now observed originate from that parent body, then that parent body should have an oxygen isotopic composition that is common among its family members. This would require meteorite analogs for the Eos family members to include only those analogs with similar oxygen isotope ratios. Oxygen isotopic ratios for the CO, CK, and CV chondrites overlap in a region of δ17O/δ18O space, but the R chondrites occupy a different region of this plot (Brearley and Jones 1998). This suggests that the CO/CK/CV chondrites could potentially compose the Eos parent body either singly or as a combination of two or more of these meteorite types. Recent work by Greenwood et al. (2010) suggests an affinity between the CK and CV chondrites, but not the CO chondrites. This suggests that the Eos family may have originated from a parent body with a CO chondrite, CV/CK-chondrite, or R chondrite origin, but not an original mixture of these meteorite types from the solar nebula.

Phyllosilicate-Bearing Asteroids

This article presents the first NIR spectral evidence in the wavelength range of approximately 0.8–2.5 μm for phyllosilicate ± hydroxide minerals on asteroid surfaces, although absorptions at visible and 3 μm wavelengths have been previously reported (Vilas 1994; Rivkin et al. 1995, 2000). The suggestion of Fe-rich chamosite or chlorites, more generally, is consistent with the hydrous mineralogy within CI and CM chondrites (Brearley and Jones 1998). CI/CM chondrites are not, however, meteorite analogs for 22 Kalliope, 55 Pandora, 132 Aethra, and 504 Cora because of albedo and/or overall spectral slope dissimilarities (Gaffey 1976; Tedesco et al. 2002).

The combination of multiple NIR absorption features attributable to phyllosilicates and hydroxides, moderate albedos significantly brighter than CI/CM chondrites, and variably red NIR spectral slopes consistent with laboratory phyllosilicate NIR spectra suggests that these asteroids represent a population of extensively hydrated asteroids that do not have meteorite analogs in the terrestrial collection. This suggestion is supported by these asteroids’ locations beyond the 2:1 mean-motion resonance at 2.5 AU.

The 26Al accretionary heating model of Grimm and McSween (1993) predicts a heliocentric-based temperature gradient throughout the asteroid belt based on asteroid size and semimajor axis. Using current asteroid diameters can begin to constrain the greatest semimajor axis where an asteroid should reside based on the inferred temperature environment that it experienced in this model. For example, 22 Kalliope has an IRAS diameter of 181 km. Assuming T = 300 K for the hydrous and anhydrous models of Grimm and McSween (1993), 22 Kalliope should reside at a semimajor axis of approximately 2.86 AU and 3.00 AU, respectively. 22 Kalliope’s current semimajor axis is at 2.92 AU. Increasing this asteroid’s diameter will not change its semimajor axis in these models.

55 Pandora, 132 Aethra, and 504 Cora have current diameters ranging from approximately 30 to 70 km, but they are probably fragments from larger parent asteroids. Using current diameters results in predicted minimum semimajor axes of approximately 2.6–2.7 AU; their maximum semimajor axes will be no more than approximately 2.9–3.0 AU, depending on the parent asteroids’ original diameters. The location of these four potentially hydrous asteroids is consistent with the predicted locations of extensive asteroid belt hydrothermal alteration in the models of Grimm and McSween (1993). It also suggests the approximately 2.6–3.0 AU region in the main asteroid belt may be a good location to look for hydrous asteroids with detectable NIR absorption features.

Summary of Existing Data

A summary of the derived band centers from the works of Hardersen et al. (2005), Ockert-Bell et al. (2008, 2010), Fornasier et al. (2010), and this article are presented in Table 7. A comparison of the data in Table 7 shows that the results from different research groups are often divergent when reporting the presence of NIR absorption features and the derived band center(s) for individual asteroids. The data in Table 7 shows differing results for 22 Kalliope, 69 Hesperia, 77 Frigga, 110 Lydia, 125 Liberatrix, 129 Antigone, 135 Hertha, 136 Austria, 201 Penelope, 216 Kleopatra, 250 Bettina, 338 Budrosa, 347 Pariana, 369 Aeria, 382 Dodona, 441 Bathilde, 497 Iva, 498 Tokio, 516 Amherstia, 558 Carmen, 758 Mancunia, 785 Zwetana, and 872 Holda. The possible causes for the differences are numerous and could include compositional variations with rotation, measurement difficulties due to absorption band weakness, telluric effects, data reduction procedures, and observational circumstances (Ockert-Bell et al. 2008).

Table 7.   Compilation of reported M-/X-asteroid band centers from Hardersen et al. (2005), Fornasier et al. (2010), Ockert-Bell et al. (2008, 2010), and this work.
AsteroidHardersen et al. Band I center (μm)Hardersen et al. Band II center (μm)Ockert-Bell et al. Band I center (μm)Ockert-Bell et al. Band II center (μm)Fornasier et al. All band centers
16 Psyche0.932 ± 0.0080.95 ± 0.010.430 ± 0.0040.949 ± 0.008
21 LutetiaNo data
22 Kalliope0.94-(?)0.90 ± 0.01 0.434 ± 0.005; 0.903 ± 0.008
55 Pandora0.85-(?), 0.92-(?), 0.97-(?) 1.100.93 ± 0.011.94 ± 0.020.91 ± 0.010
69 Hesperia0.923 ± 0.011; 0.909 ± 0.0091.71–1.85No dataNo data0.430 ± 0.004; 0.951 ± 0.009
71 NiobeNo dataNo dataNo data
77 Frigga0.87 ± 0.01No data
97 Klotho
110 Lydia0.914 ± 0.006; 0.914 ± 0.004; 0.903 ± 0.0051.71–1.900.88 ± 0.011.750.942 ± 0.008
125 Liberatrix0.920 ± 0.006No dataNo data
129 Antigone0.918 ± 0.003; 0.930 ± 0.0080.89 ± 0.011.028 ± 0.010
132 Aethra0.91-, 1.10-, 1,40-, 2.28–2.50+No dataNo data0.498 ± 0.004
135 Hertha0.91 ± 0.010.515 ± 0.005; 0.905 ± 0.008
136 Austria0.85 ± 0.01No data
161 AthorNo dataNo dataNo dataNo data
184 Dejopeja0.931 ± 0.004No dataNo dataNo data
201 Penelope0.932 ± 0.005; 0.945 ± 0.008; 0.917 ± 0.010No dataNo data
216 Kleopatra0.923 ± 0.0030.91 ± 0.011.99 ± 0.020.429 ± 0.004; 0.969 ± 0.008
224 Oceana
250 Bettina0.914 ± 0.0070.91 ± 0.010.885 ± 0.010
325 HeidelbergaNo dataNo data
338 Budrosa0.939 ± 0.0081.933 ± 0.018No dataNo data0.425 ± 0.004; 0.876 ± 0.010
347 Pariana0.919 ± 0.0080.94 ± 0.011.79 ± 0.020.871 ± 0.008
369 Aeria0.920 ± 0.0031.865 ± 0.023No dataNo data0.884 ± 0.008
382 Dodona0.934 ± 0.007No dataNo data
413 EdburgaNo dataNo dataNo data
417 SueviaFlat feature from ∼0.82 to 1.17No dataNo dataNo data
418 Alemannia0.922 ± 0.010; 0.910 ± 0.005No dataNo data
441 Bathilde0.87 ± 0.01
497 Iva0.941 ± 0.0061.918 ± 0.0190.90 ± 0.01No data
498 TokioNo dataNo data0.430 ± 0.005; 1.159 ± 0.008
504 Cora0.85-, 0.91-, 1.07-No dataNo dataNo data
516 Amherstia0.927 ± 0.0031.911 ± 0.039No dataNo data0.965 ± 0.008; 1.949 ± 0.010
558 Carmen0.940 ± 0.006No dataNo data
678 FredegundisNo dataNo data0.91 ± 0.01No data
739 MandevilleNo dataNo dataNo data
755 QuintillaNo dataNo dataNo dataNo data0.904 ± 0.010; 1.369 ± 0.010; 1.610 ± 0.008; 1.864 ± 0.010
758 Mancunia0.87 ± 0.011.90 ± 0.02No data
766 Moguntia1.068 ± 0.004No dataNo dataNo data
771 LiberaNo dataNo data0.90 ± 0.01No data
779 NinaNo dataNo data0.93 ± 0.011.78 ± 0.02No data
785 Zwetana0.62 ± 0.011.68 ± 0.02
796 Sarita0.914 ± 0.0081.846 ± 0.027No dataNo dataNo data
798 Ruth1.056 ± 0.004No dataNo data
849 AraNo dataNo dataNo dataNo data
857 GlasenappiaFlat feature from ∼0.94 to 1.081.958 ± 0.028No dataNo dataNo data
860 UrsinaWeak inflection?No dataNo data
872 HoldaWeak inflection?0.95 ± 0.010.965 ± 0.020
931 WhittemoraWeak inflection?No dataNo dataNo data
1210 Morsovia1.047 ± 0.009No dataNo dataNo data
1461 Jean-JacquesNo dataNo dataNo data

3 μm Correlations

Three pyroxene-bearing M-asteroids: 110 Lydia, 129 Antigone, and 201 Penelope exhibit both approximately 0.9 ± 1.9 μm and approximately 3.0 μm absorptions, whereas seven asteroids (16 Psyche, 125 Liberatrix, 184 Dejopeja, 216 Kleopatra, 369 Aeria, 497 Iva, 796 Sarita) have approximately 0.9 ± 1.9 μm features, but no approximately 3.0 μm feature. 857 Glasenappia has a complex olivine/pyroxene NIR spectrum, but does not have an approximately 3.0 μm feature with some uncertainty (Rivkin et al. 2000). Four asteroids (21 Lutetia, 77 Frigga, 135 Hertha, 136 Austria) have an approximately 3.0 μm feature, but are spectrally featureless in the NIR. 758 Mancunia and 785 Zwetana lack both NIR spectral features and a approximately 3.0 μm absorption. One of the potentially phyllosilicate-bearing M-asteroids, 22 Kalliope, has an approximately 3.0 μm absorption.

Any number of scenarios can account for these combinations of absorption features and spectral shapes in the VNIR region. Assuming that an approximately 3.0 μm feature is due to a hydrated mineral phase on an asteroid’s surface, collisional mixing can account for the presence of abundant hydrous and anhydrous phases on an asteroid’s surface. A noncollisional mechanism could involve a parent body with internal ice that heated and melted during the early solar system heating event (Herbert et al. 1991; Grimm and McSween 1993). The internal water could have mobilized and aqueously altered only portions of the interior prior to disruption. The coexistence of moderately reducing minerals (i.e., low-Fe pyroxenes) with ice seems unlikely, but could be facilitated in middle regions of the main asteroid belt (approximately 2.6–3.0 AU) where the chemical and mixing conditions are quite variable (Grimm and McSween 1993).

Implications of Space Weathering

An uncertainty in our results involve the effects of space weathering on the asteroid NIR spectra and, hence, the resulting interpretations. Space weathering is generally defined as a spectral effect that causes either VNIR spectral slope reddening, absorption band weakening, and surface material darkening due to either micrometeorite impacts causing the deposition of nano-phase Fe or solar wind sputtering (Clark et al. 2002). A variety of laboratory and observational efforts have been underway to try to better understand and constrain the effect as it relates to asteroids (Moroz et al. 1996; Hiroi et al. 2006; Lazzarin et al. 2006; Gaffey 2010 and references therein).

Gaffey (2010) showed that asteroids might be experiencing multiple types of space weathering processes. Plots of band depth versus albedo and spectral slope versus albedo for 243 Ida and 433 Eros differ both from lunar-style space weathering trends and each other’s trends. Band analysis techniques (i.e., band centers) and mineralogical determinations (i.e., Gaffey et al. 2002) appear to be immune from whatever space weathering effects are occurring (Gaffey 2010). It is also important to note that VNIR spectral slopes and mineral band depth changes can be attributed to a wide variety of observational and compositional effects, which are not easy to distinguish (Cloutis et al. 1990; Birlan et al. 2007; Ockert-Bell et al. 2008).

The pyroxene-only bearing asteroids in this article display systematically weaker Band I ± Band II absorptions compared with the olivine-bearing asteroids. A pure orthopyroxene sample will display a Band I depth of approximately 50%, whereas the band depths of the pyroxene-bearing asteroids range from approximately 1 to 5%. No obvious trend is apparent when plotting Band I depth versus semimajor axis for these asteroids, although the asteroids with the mildly deeper Band I depths (approximately 3–5%) are located at semimajor axes from 2.6 to 2.7 AU, whereas those asteroids with Band I depths of 1–2% are found along the entire 2.6–3.2 AU semimajor axis range.

A pure olivine sample will display a Band I depth of approximately 45%, whereas the band depths of the olivine-bearing asteroids range from approximately 4 to 10%. The band depths of the olivine-bearing asteroids, however, are consistent with the band depths for the CO, CK, and R chondrites in this study. The opaque mineral component of these meteorite types (discussed above) is the likely cause of the resulting spectral band depth weakness and does not require a space weathering mechanism.

The four potentially phyllosilicate-bearing asteroids exhibit one or more weak absorption features, but these features are similarly weak in laboratory VNIR spectra of the same minerals (Calvin and King 1997). The weakness of these features is attributed to crystal field transitions (Burns 1993).

The cause(s) of mineral absorption band weakness (or absence) in asteroid NIR spectra include space weathering, surface opaque minerals, the absence of Fe2+-bearing minerals, and metal. If space weathering is the cause of the band weakness for these asteroids, then the mechanism must be able to explain differential, mineral-dependent space weathering effects. Continuing work is necessary to disentangle these variables to enable a more robust understanding of this mechanism and its effects on NIR spectra.


Significant advances in understanding the mineralogical and compositional nature of the M-/X-asteroid population has been made in the past 15 yr. This article reports significant mineralogical and spectral diversity among a group of 45 M-/X-asteroids. The primary results include:

  • 1 Weak NIR absorption features in the approximately 0.9 ± 1.9 μm region are common and include 42% of the asteroids in this study. These asteroids are found throughout the main asteroid belt, exhibit band depths from approximately 1 to 5%, and Band I centers typically at λ < 0.93 μm. These asteroids are dominated by surface orthopyroxene, whereas a few asteroids with Band I > 0.94 μm includes orthopyroxene and Type B clinopyroxene. Meteorite analogs for these asteroids include mesosiderites, silicate-bearing iron meteorites, CB/CH chondrites, and metallic cores of differentiated asteroid parent bodies with remnant mantle surface pyroxene.
  • 2 The three olivine-bearing asteroids reported here are restricted to the Eos dynamical family. As a group, these olivine ± pyroxene-bearing asteroids have possible affinities to the CO chondrites, CK chondrites, CV chondrites, ureilites, brachinites, and R chondrites. The band parameters for several Eos family asteroids and the above meteorite types suggest that CO chondrites are most consistent with the spectral traits of the measured Eos family asteroids.
  • 3 Four potentially phyllosilicate ± hydroxide-bearing asteroids are located in the central region of the main asteroid belt. We report what appears to be the first NIR spectral absorption features in the approximately 0.8–2.5 μm range consistent with phyllosilicate or hydroxide minerals. Fe-rich chamosite is suggested as a primary surface mineral for some regions on 132 Aethra. The locations of these asteroids are consistent with the predicted locations of hydrous asteroids heated by 26Al via the model of Grimm and McSween (1993). CI/CM chondrites are not meteorite analogs due to the much higher albedos reported for these asteroids. This suggests that these asteroids are from objects not represented in the terrestrial meteorite collection.
  • 4 M-/X-asteroids with featureless NIR spectra compose 40% of our sample, but display noticeable variations in spectral slope. Enstatite chondrites and NiFe meteorites remain viable interpretations for most of the asteroids in this group.
  • 5 While taxonomies are useful tools to categorize asteroids into similar groups, their utility in exploring the mineralogical diversity of a taxonomic group’s members is limited. Higher-quality instrumentation and data reduction protocols, along with the ability to reliably detect weak NIR spectral absorption features, allow enhanced opportunities to discover mineralogical diversity within a single taxonomic class.
  • 6 The asteroids in this article are, for the most part, genetically unrelated. Interpretations and potential meteorite analogs for these asteroids should be decoupled; each asteroid should be considered individually and usually assumed to derive from a unique parent body.


  • 1

    EET 87720, MET 01085, Graves Nunatak 95205, EET 87517, Y-791538, ALHA77257, Lewis Cliff 88201, ALH 82130, ALHA81101, PCA 82506, EET 96042, Grosvenor Mountains 95575, Dar al Gani 319, Novourei, Kenna, META78008, NWA 1500, Goalpara.

  • 2

    Figure 45 from Brearley and Jones (1998) incorrectly shows the mean Allende chondrule olivine chemistry as Fa76. It should read Fo76, which is correctly discussed in the text on page 3–53.

Acknowledgments–– The authors thank reviewers Lucy McFadden, Taki Hiroi, Maureen Ockert-Bell, and the MAPS Associate Editor, Beth Clark, for comments that improved this manuscript. This work has been supported by NASA Planetary Astronomy Program grant NNG05GH01G. The contributions of Driss Takir, Sherry Fieber-Beyer, Beth Reynolds, Paul Abell, and Michael Gaffey to this work are noted and appreciated. The authors thank the NASA Infrared Telescope Facility, Alan Tokunaga, Bobby Bus, John Rayner, and telescope operators Bill Golisch, David Griep, Paul Sears, and Eric Volquardsen for facilitating the results of this paper. Please forgive the lead author (PSH) for nearly burning out the brakes on an IRTF vehicle during his first-ever observing run in April 2001.

Thanks to the Canadian Space Agency, the Canada Foundation for Innovation, the Manitoba Research Innovations Fund, NSERC, and the University of Winnipeg for providing funding to EAC to establish and operate the Planetary Spectrophotometer Facility at the University of Winnipeg.

Author T. Mothé-Diniz was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico––CNPq/Brasil and by the Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro––FAPERJ.

Editorial Handling–– Dr. Beth Ellen Clark