The nature of bright C-complex asteroids

ABSTRACT

Most C-complex asteroids have albedo values less than 0.1, but there are some high-albedo (bright) C-complex asteroids with albedo values exceeding 0.1. To reveal the nature and origin of bright C-complex asteroids, we conducted spectroscopic observations of the asteroids at visible and near-infrared wavelengths. As a result, the bright B-, C-, and Ch-type (Bus) asteroids, which are subclasses of the Bus C-complex, are classified as DeMeo C-type asteroids with concave curvature, B-, Xn-, and K-type asteroids. Analogue meteorites and material (CV/CK chondrites, enstatite chondrites/achondrites, and salts) associated with these spectral types of asteroids are thought to be composed of minerals and material exposed to high temperatures. A comparison of the results obtained in this study with the SDSS photometric data suggests that salts may have occurred in the parent bodies of 24 Themis and 10 Hygiea, as well as 2 Pallas. The bright C-complex asteroids in other C-complex families were likely caused by impact heating. Bright C-complex asteroids that do not belong to any families are likely to be impact-metamorphosed carbonaceous chondrites, CV/CK chondrites, or enstatite chondrites/achondrites.

1 Introduction

In the 1970s, within several years of the first spectrophotometric study of asteroids (McCord et al. 1970), asteroids with flat reflectance spectra over the visible wavelength range were discovered (Johnson & Fanale 1973; McCord & Gaffey 1974; Chapman et al. 1975). Since these asteroids have spectral and physical properties similar to carbonaceous chondrites (e.g., Johnson & Fanale 1973; McCord & Gaffey 1974; Chapman et al. 1975) and interplanetary dust particles (e.g., Vernazza et al. 2015; Hasegawa et al. 2017), the asteroids are considered as parent bodies of these chondrites. These asteroids with flat reflectance spectra over the visible wavelength range are defined or categorized as C-complex/group/type asteroids (hereafter C-complex asteroids; Chapman et al. 1975; Tholen 1984; Bus & Binzel 2002b). The albedo of C-complex asteroids is typically below 0.1 (Chapman et al. 1975; Tedesco et al. 1989). The major trends in the distribution of the albedo of C-complex asteroids have not shown significant differences in recent studies that used infrared astronomical satellites dedicated to all-sky surveys, AKARI (Murakami et al. 2007) and WISE (Wright et al. 2010); however, small C-complex asteroids with albedos greater than 0.1 were found (Mainzer et al. 2011a, 2012; Usui et al. 2013).

To examine the characteristics of C-complex asteroids with geometric albedo of more than 0.1 (hereafter “bright” C-complex asteroids) in more detail, Kasuga et al. (2013, 2015) conducted near-infrared spectroscopic observations of bright C-complex asteroids in the outer main belt, and no signs of water ice were found in the bright C-complex asteroids. These studies argue that Mg-rich amorphous and/or crystalline silicates on the surface of the asteroids can be responsible for the high albedo values. The presence of crystalline silicate implies the occurrence of high-temperature environments during planetesimal formation, which was confirmed in some of the observed bright C-complex asteroids.

It is essential to classify the reflectance spectra of asteroids in order to know the properties of the surface materials on asteroids. The Bus taxonomy (Bus & Binzel 2002b) that classifies asteroids using visible reflectance spectra according to principal component analysis (PCA) and the Bus–DeMeo taxonomy (DeMeo et al. 2009; Binzel et al. 2019) that does the same using visible and near-infrared spectra according to PCA are the most popular and important classification methods for classifying asteroids. In particular, the Bus–DeMeo taxonomy is based on spectra with a wider wavelength range, so it is expected that more detailed asteroid properties will be extracted.

In this study, we made spectroscopic observations in visible and near-infrared wavelength regions, to obtain the reflectance spectra of bright C-complex asteroids. On the basis of the obtained near-infrared spectroscopic results of bright C-complex asteroids, we discuss the origin of bright C-complex asteroids using large amounts of spectrophotometric asteroid data. The subsequent sections of this paper describe the observations and data reduction procedures (section 2), present the spectroscopic and taxonomic results of bright C-complex asteroids (section 3) and the boundaries of albedos for bright C-complex asteroids (section 4), and discuss the nature and origin of the asteroids (section 5).

2 Observations and data reduction procedures

In order to carry out the Bus–DeMeo taxonomy of bright C-complex asteroids for which visible spectra have already been published in the literatures, near-infrared spectroscopic observations for the asteroids were performed in this study. The main candidates for observations at near-infrared wavelengths were listed as bright C-complex asteroids on the basis of the Bus taxonomy (Bus & Binzel 2002b; Lazzaro et al. 2004)¹ in the AKARI asteroid catalogs (Usui et al. 2011; Hasegawa et al. 2013). A bright T-type asteroid, which typically has low albedo characteristics, was also observed (979 Ilsewa). Also, to increase the sample of bright C-complex asteroids, visible spectroscopic observations for bright “unknown-type” asteroids were carried out. Unknown spectral type candidates were selected from the bright asteroids in the outer main-belt region, where many C-complex asteroids are distributed, listed in the AKARI asteroid catalogs. Furthermore, spectroscopic observations of several “known as” C-complex asteroids with low albedo, which are located in the outer main-belt region, were also conducted to evaluate the reliability of the obtained reflectance spectra.

The spectroscopic observations for bright C-complex asteroids were conducted at two observatories: the Subaru Telescope, National Astronomical Observatory of Japan (NAOJ), National Institute of Natural Sciences (NINS) on Mount Mauna Kea in Hawaii, USA [Minor Planet Center (MPC) code: 568] from 2012 February to 2013 April, and the Okayama Astrophysical Observatory, NAOJ, NINS, in Okayama, Japan (MPC code 371) during 2013 December and 2014 January. Ten and eleven spectroscopic observations of the asteroids were made with the Subaru telescope and Okayama Astrophysical Observatory, respectively. The nightly observational details for the spectroscopy are listed in table 1.

2.1 Near-infrared spectroscopic observations

Spectroscopic data at near-infrared wavelengths were obtained using the infrared camera and spectrograph (IRCS; Tokunaga et al. 1998; Kobayashi et al. 2000) attached to the f/12 Cassegrain focus of the 8.2 m Subaru Telescope. The instrument detector has a 1024 × 1024 Aladdin-III InSb array, which provides a |${54^{\prime \prime }}$| × |${54^{\prime \prime }}$| field of view, with a pixel scale of |${0{^{\prime \prime }_{.}}052}$|⁠. The IRCS system is composed of a number of grisms. Spectroscopic data are obtained by dividing the spectral range into two wavelength regions: JH and HK. The two wavelength ranges determined using the JH grism with the order-sorting filter for JH mode and the HK grism with the order-sorting filter for HK mode were 1.0–1.6 μm and 1.45–2.5 μm, respectively. There is an overlap in wavelength of the two grism modes at 1.45–1.60 μm. The IRCS in low resolution has a dispersion of approximately 13 Å per pixel in the JH grism, and approximately 12 Å per pixel in the HK band grism, in the wavelength direction. The orientation of the slit of most observations was along the parallactic angle to prevent flux loss caused by atmospheric dispersion. To increase the signal-to-noise ratio and avoid flux loss due to the slit, a 188-element curvature sensor adaptive optics (AO188) system (Minowa et al. 2010) was used with IRCS observations. Since the target asteroids were too faint, the nearby bright stars were used as the natural guide star to operate the AO188 system. Typical seeing size with the AO188 system was improved to ∼|${0{^{\prime \prime }_{.}}2}$|–|${0{^{\prime \prime }_{.}}6}$| in the K band through the observations. Slits of width |${0{^{\prime \prime }_{.}}3}$|⁠, |${0{^{\prime \prime }_{.}}6}$|⁠, and |${0{^{\prime \prime }_{.}}9}$| were used according to the changes in the seeing size. The length of each width component of the long slit for IRCS was ∼|${15^{\prime \prime }}$| in the cross-wavelength direction. The spectral resolution in low dispersion in the JH and HK grisms was ∼100 Å when using the |${0{^{\prime \prime }_{.}}9}$| slit.

2.2 Visible spectroscopic observations

Optical spectroscopic data were recorded using the Kyoto Okayama Optical Low-dispersion Spectrograph (KOOLS; Ohtani et al. 1998; Ishigaki et al. 2004) installed at the f/18 Cassegrain focus of the 1.88 m telescope at the Okayama Astrophysical Observatory. The 2048 × 4096 SITe ST-002A CCD installed on KOOLS with 15 μm square pixels produces a |${5^{\prime }}$| × |${4{^{\prime }_{.}}4}$| field of view with a pixel scale of |${0{^{\prime \prime }_{.}}3}$|⁠. A grism with the 6563 Å mm⁻¹ blaze and a Y49 order-sorting filter were used, and the spectral range covered is 0.47–0.88 μm. KOOLS has a dispersion of approximately 3.8 Å pixel⁻¹ in the wavelength direction. The length of the long slit for KOOLS was |${4{^{\prime }_{.}}4}$| in the cross-wavelength direction. Since the 1.88 m telescope did not have an atmospheric dispersion corrector to prevent flux loss errors, it was necessary to select a wide slit for KOOLS observations; thus, a large width of |${6^{\prime \prime }}$| was utilized, which yields a spectral resolution of 32 Å.

2.3 Data acquisition methods for asteroids

To acquire an asteroid image as a point source, non-sidereal tracking was employed for the Subaru and 1.88 m telescopes. As a typical method to obtain the reflectance spectrum of an asteroid, the asteroid flux spectrum is divided by the solar flux spectrum. Therefore, solar analogue stars classified as G or F8 stars in the SIMBAD Astronomica database³ or Drilling and Landolt (1979) were used as standard stars. Observations of the solar analogue stars were coordinated so that the airmass difference between the asteroid and the solar analogue star was less than 0.1 in each case. Spectroscopic observations were performed with an airmass lower than 1.4 to prevent airmass mismatch between the asteroid and solar analogue star. The asteroids and solar analogue stars were situated in the center of the slit during spectroscopic observations to avoid flux envelope loss due to the slits. To reduce noise due to atmospheric emission lines and thermal emissions, spectroscopic data were alternated between two different slit positions, A and B, on the detector. An observation set for the asteroids was composed of four positions, A–B–B–A. The A–B–B–A sequence for IRCS and KOOLS observations was conducted using two-position nodding patterns with intervals of |${7^{\prime \prime }}$| and |${20^{\prime \prime }}$| along the cross-wavelength direction of the slit. The spectrum of the obtained image was recurrently located on the same column to ensure signal reproducibility. For IRCS and KOOLS observations, flat-fielding images were obtained with a halogen lamp. Wavelength calibration frames for IRCS and KOOLS were acquired regularly at night with light from argon and iron–neon–argon hollow cathode lamps, respectively.

2.4 Data reduction procedures

Reduction procedures for all spectroscopic data were conducted using the Image Reduction and Analysis Facilities (IRAF) software. For sensitivity correction of each pixel, each object frame for asteroids and solar analogue stars was divided by the flat-fielding frame. The spectra were reduced by subtracting A and following B images to eliminate the sky background, including the emission lines and thermal emissions. The two-dimensional spectrum on the image was collapsed to a one-dimensional spectrum using the apall IRAF task. Next, wavelength reference positions were determined by adapting the wavelength calibration frame to the object frame. The reflectance of the asteroid was calculated by dividing the spectrum of the asteroid into that of the solar analogue star. Individual visible and near-infrared spectra were smoothed by taking the median of the final reflectance spectrum of each frame, which was normalized to 0.55 and 1.60 μm, respectively.

3 Obtained spectra and taxonomic classification of asteroids

The spectra of the asteroids taken in this study are shown in figures 1–3. The figures also show the spectra obtained in Kasuga et al. (2013, 2015). Spectroscopic and spectrophotometric data in visible and near-infrared wavelength regions reported in past studies (Zellner et al. 1985; Sykes et al. 2000;⁴ Baudrand et al. 2001, 2004; Bus & Binzel 2002a; Lazzaro et al. 2004; Carvano et al. 2010; Clark et al. 2010; Fornasier et al. 2014; Popescu et al. 2016) are also displayed in figures 1–3. Spectral data with wavelength gaps were interpolated to match both levels. The dotted line in the figure is the line used for interpolation between the individual wavelength gap data.

The validity of the method for acquiring the visible spectrum of asteroids using KOOLS is shown by comparing the results with data from other studies (Hasegawa et al. 2014). The consistency between the visible spectrum of 3037 Alku in this study (figure 3) and that in Bus and Binzel (2002a) also supports the validity of the method. However, the legitimacy of obtaining near-infrared spectral data of asteroids using IRCS with the AO188 system was not disclosed. To validate the observational methods using IRCS with the AO system, observations for 419 Aurelia, which has a continuous spectrum from 0.3–3.5 μm (Clark et al. 2010; Usui et al. 2019), were taken, although not in a photometric condition. The spectra of 419 Aurelia in this study (figure 3) are consistent with those in Clark et al. (2010), which indicates that obtaining the spectra using IRCS with the AO188 system is the proper method.

The observed asteroids were classified to determine the nature of the bright C-complex asteroids. To classify the asteroids with only visible data and both visible and near-infrared data, the Bus taxonomy (Bus & Binzel 2002b) and Bus–DeMeo taxonomy (DeMeo et al. 2009; Binzel et al. 2019) were employed, respectively. Based on the Bus taxonomy, nine asteroids were classified by best fits of the mean visible spectra, which are listed in table III of Bus and Binzel (2002b), to those data. The Bus–DeMeo taxonomy for 18 asteroids was determined with a taxonomic tool available on the web.⁵ When classifying, obvious false data in the spectra due to strong atmospheric absorption between the Y and J bands, between the J and H bands, and between the H and K bands were corrected by linear interpolation using the appropriate spectra around them. The Bus–DeMeo classification for five asteroids with visible spectra and near-infrared photometric data in J, H, and K_s bands were conducted, extrapolating the value in the K_s band to the value at 2.45 μm. Note that both taxonomies are based on the spectral shape only, and geometric albedo is not taken into account for the classification.

The classification results for bright C-complex, T-type, and unknown spectral type asteroids and for dark (albedo values less than 0.1) asteroids in the Bus and Bus–DeMeo taxonomy are shown in table 2. The results of the Bus classifications for 2250 Stalingrad, 2310 Olshaniya, 2670 Chuvashia, and 4925 Zhoushan could not be narrowed down to one spectral type due to the poor quality of the observed data. Using the Bus–DeMeo classification, 555 Norma was classified as “indeterminate” using both visible and near-infrared spectra, but was also classified as C-type using only the near-infrared spectrum.

The DeMeo C-type asteroids include asteroids classified as Bus B-type (see table 3 in this study; DeMeo et al. 2009; de León et al. 2012). In this work, these objects are called C-type asteroids with concave curvature. About a third of the known C-type (DeMeo) asteroids (DeMeo et al. 2009; Clark et al. 2010; Ostrowski et al. 2011; de León et al. 2012; Hasegawa et al. 2017, 2018; Binzel et al. 2019) are C-type with concave curvature. It is possible that these data are subject to selection bias; nevertheless, it is worth noting that Bus B-type asteroids are not always classified as DeMeo B-type asteroids. This is because the Bus B-type asteroids do not necessarily have concave curvature over the range from visible to near-infrared. The Bus B-type asteroids with a flat or convex curvature in the near-infrared are typically classified as DeMeo C-type.

The Bus–DeMeo taxonomic results from both visible and near-infrared spectra for bright C-complex asteroids show the existence of DeMeo X-complex and C-type with concave curvature, B-, Cb-, K-, and Sa-type (see the top of table 2). Bright T-type (Bus) asteroids were classified as L-type (DeMeo) in this study (see the second panel of table 2). Bright asteroids of unknown spectrum type were identified as DeMeo X-complex, K-, Sq-, and C-type with concave curvature (see the third panel of table 2).

4 Boundaries of albedos for bright C-complex asteroids

In order to find appropriate albedo boundaries between the dark and the bright C-complex asteroids, the distributions of the geometric albedo for B-, Ch-, and C-types (Bus) were derived. The geometric albedo values of asteroids are available in the literature (AKARI: Usui et al. 2011; Hasegawa et al. 2013; IRAS: Tedesco et al. 2002; WISE: Grav et al. 2011, 2012; Mainzer et al. 2011b; Masiero et al. 2011, 2012, 2014). The classification results based on the Bus taxonomy are quoted from Binzel et al. (2001, 2004), Bus and Binzel (2002b), Rivkin et al. (2003), Yang et al. (2003), Lazzaro et al. (2004), Lazzarin et al. (2004a, 2004b, 2005), Marchi et al. (2005), Davies et al. (2007), Moskovitz et al. (2008a, 2008b), Mothé-Diniz and Nesvorný (2008a, 2008b), Roig et al. (2008), Duffard and Roig (2009), and Iwai et al. (2017). Most asteroids in this study are main-belt asteroids with diameters larger than 10 km. The albedo distributions are shown in figure 4.

To start with, we focus on the B-type asteroids. Figure 4 (left) indicates the existence of bright B-type asteroids. Alí-Lagoa et al. (2016) reported that the albedo of the members of the Pallas family occupied by B-type asteroids is 0.141 ± 0.043. The presence of bright B-type asteroids (Bus) in the Pallas family (table 3) supports the result of Alí-Lagoa et al. (2016). As a result of applying the two-component Gaussian distribution to B-type asteroids, the mean value of the high-albedo part for B-type asteroids is 0.164, which is consistent with the result of Alí-Lagoa et al. (2016). The boundary value of the albedo between the bright and dark B-type asteroids is derived as 0.098. All B-type asteroids in the Bus taxonomy with albedo values larger than 0.1 are listed in table 3.

Next, the albedo distribution for Bus Ch- and C-type asteroids was statistically evaluated. In order not to underestimate the values of the standard deviation for Ch- and C-type (Bus) asteroids, they were obtained without using Gaussian distribution fitting. The mean and standard deviation of log(albedo) for Ch-type asteroids are −1.261 = log(0.065) and 0.160, respectively. The mean and standard deviation of log(albedo) for C-type asteroids are −1.225 = log(0.060) and 0.207, respectively. Even if the asteroid has an albedo value exceeding 0.1 into the 2 σ range (Ch-type: less than 0.115; C-type: less than 0.154), it is considered to be a normal Ch- or C-type asteroid. Figure 4 (middle and right) shows the results of the albedo distribution for Ch- and C-type (Bus) asteroids. Almost no Ch- or C-type (Bus) asteroids with an albedo value below −2 σ exist, but there are several Ch- and C-type (Bus) asteroids with albedo values larger than +2 σ. Since Bus Ch- and C-type asteroids with an albedo value within +2 σ can be considered normal Ch- and C-type (Bus) asteroids, we next consider the nature of bright Ch- and C-type (Bus) asteroids with an albedo value exceeding +2 σ. The boundary values of the albedos between the bright and dark Ch-type asteroids and between the bright and dark C-type asteroids are derived as 0.115 and 0.154. All Ch- and C-type asteroids in the Bus taxonomy with albedo values larger than 0.115 and 0.154 are listed in tables 4 and 5, respectively.

5 Discussion

5.1 Analogue material and meteorites

On the basis of visible and near-infrared spectroscopic results in the literature as well as this study, bright C-complex asteroids are classified as B, C with concave curvature, Xn, and Sq (Bus B-type), Sa (Bus Ch-type), and Xn, K, and B (Bus C-type)—see tables 3, 4, and 5. There are some outliers in these results. For example, Ch-type (Bus) asteroid 3104 Durer is classified as Sa-type in this study, which is consistent with A_p-type in Carvano et al. (2010). Meanwhile, the spectrum of 3104 Durer in Lazzaro et al. (2004) showed Ch-type (Bus) characteristics. This spectral mismatch may be explained by a misidentification of the target in one of their surveys. For another example, B-type (Bus) asteroid 8567 1996 HW₁ is classified as Sq-type in DeMeo, Binzel, and Lockhart (2014), which is inconsistent with B-type in Lazzarin et al. (2004b). This spectral discrepancy may also be due to misidentification of the target. Except for such S-complex asteroids, these spectral type asteroids are candidates for bright C-complex asteroids.

It is known that the spectra of meteorites become bluer with an increase in the particle size of the regolith layer (e.g., Johnson & Fanale 1973; Vernazza et al. 2016). Hasegawa et al. (2019) argued that Q-type spectra can be explained by a lack of fine particles even if space weathering matures their surface. It is expected that originally bright X-complex asteroids with reddish spectra appear to be bright C-complex asteroids due to the spectral bluing effect caused by the absence of regolith on their surface. Solar radiation pressure and electrostatic force are known to be responsible for the removal of fine particles from the asteroid surface layer (Hasegawa et al. 2019). However, these effects can only work effectively on near-earth asteroids less than 0.5 km in diameter. The bright C-complex (Bus) asteroids that are the focus of this study are main-belt asteroids with diameters of more than 10 km, so that fine particles on their surface cannot escape since the influence of these forces is low due to the asteroids’ high gravity and/or low solar flux compared to near-earth space. Therefore, the existence of bright C-complex (Bus) asteroids cannot be attributed to spectral changes.

Based on meteorite studies (e.g., Hiroi et al. 1993, 1996), bright B-type asteroids are likely to be thermally metamorphosed carbonaceous chondrites. de León et al. (2012) showed that the spectra of B-type asteroids are similar to those of CV3 and CK4 carbonaceous chondrites. Meanwhile, Marsset et al. (2020) showed that the cause of the brighter surface on 2 Pallas may be the distribution of salts in the surface layer. Salts have already been found on 1 Ceres (De Sanctis et al. 2016), and their presence has also been proposed for the Trojan asteroids (Yang et al. 2013). Xn-type (DeMeo) asteroids are mineralogically associated with enstatite chondrites and/or achondrites (e.g., Lucas et al. 2019), as also reported by Kasuga et al. (2013, 2015). Enstatite chondrite and/or achondrites are formed in a high-degree reductive environment and a high-temperature environment such as in internal igneous or external large-scale impact melting processes (e.g., Keil 1989, 2010). Neeley et al. (2014) suggested that DeMeo Xk-type asteroids are related to iron meteorites, enstatite chondrites, and stony-iron meteorites, all of which originate from high-temperature parent bodies. DeMeo K-type asteroids are linked to CK chondrites (Mothé-Diniz et al. 2008; Clark et al. 2009), which are known to be highly oxidized and thermally metamorphosed meteorites (e.g., Greenwood et al. 2010). As described above, we speculate that the bright C-complex asteroids originate from parent bodies whose interiors experienced high temperatures. To confirm this hypothesis, we investigate the correlation by adapting the results in this study to the known spectrophotometric data as follows.

5.2 Sources of bright C-complex asteroids

We examine the spectral distribution of some collisional families including bright C-complex asteroids and correlate them with the spectra of the bright C-complex asteroids in order to understand the origin of these asteroids. Since members of asteroid families are considered to be fragments ejected from parent bodies by catastrophic disruption or cratering, it is possible to estimate the composition of the parent bodies from those spectra. The number of spectroscopic data points acquired in both the visible and near-infrared wavelength regions for bright C-complex asteroids is still too small and there are not enough asteroids, even those with visible spectra, to compare the spectroscopic properties of family members (tables 3, 4, and 5). To examine the spectral properties of family members, we use the classification based on the Sloan Digital Sky Survey (SDSS) multicolor photometric data in Carvano et al. (2010). The C-complex asteroids in Carvano et al. (2010) are classified into one class, C_p-type (photometric C-type). In this study, we define C_p-complex asteroids as asteroids not only of C_p-type, but also CD_p-, CL_p-, and CX_p-types, which are intermediate types between C_p and the other spectral types. Bright C_p-complex asteroids are defined as those with an albedo of more than 0.115 (see section 4), which is the threshold value for bright Ch-type (Bus) asteroids. We also add the definition of B_p-type, as B-type asteroids in Alí-Lagoa et al. (2016) among C_p-complex asteroids. The mean spectral distribution and abundance of families with bright C-complex asteroids present in their members (tables 3, 4, and 5) and families having mean low albedo values are shown in figure 5 and table 6, respectively.

The Eos family is known to be bright, and have a mean albedo of more than 0.1 (Masiero et al. 2011) and a mixture of members with various spectra such as K-, Ld-, Xc-, X-, C-, and B-types (Mothé-Diniz et al. 2005). Some Eos family members have a flat visible spectrum, e.g., 4455 Ruriko, which is identified as a C-complex asteroid by Doressoundiram et al. (1998), have albedo values of more than 0.1 (Masiero et al. 2011). In this study, bright C_p-complex asteroids were found to be abundant in the Eos family (table 6). The mean spectra of bright C_p-complex asteroids in the Eos family tend to be slightly redder than those in other families, and also the bright C-complex asteroids in this study (figure 5). Within the Eos family, 55% of bright C_p-complex asteroids are of C_p-type, 28% of CX_p-type, and 14% of CL_p-type. This indicates that the bright C_p-complex asteroids in the Eos family contain X_p-type and the L_p-type asteroids, and is consistent with the fact that the family members have a diverse spectrum (Mothé-Diniz et al. 2005). The Eos family asteroids are thought to be related to the R-chondrites and CK chondrites (Mothé-Diniz et al. 2008), and these meteorites are known to have formed at high temperature (Weisberg et al. 2006). Thus, one of the origins of the bright C-complex asteroids is likely to be from asteroids that were formed at high temperature in the interior of the parent bodies. From the models proposed by Greenwood et al. (2010) for the parent bodies of CV and CK chondrites, it is preferable to adopt one with a single parent object and a single collisional event to explain diverse spectra and the power law of the size distribution of the Eos family members.

Bright C_p-complex asteroids in the Pallas, Themis, and Hygiea families are found to have concave curvature over 0.5–0.9 μm, similar to the bright B-type asteroids (figure 5). Members of these families also include bright B_p-type asteroids (table 6). Marsset et al. (2020) indicated the possibility of salt deposits on 2 Pallas and on the Pallas family members as well. The Themis and Hygiea families are located in the outer main belt, and most of their members have a low albedo of less than 0.1, unlike the Pallas family members (table 6). The Hygiea family is thought to have been formed by a cratering event based on the size distribution of its members (Carruba 2013), while the Themis family was formed by a catastrophic disruption event (Marzari et al. 1995), both of which originated from parent bodies of similar size to 2 Pallas (Marzari et al. 1995; Vernazza et al. 2020). The composition of these parent bodies could be CK and/or R chondrite analogues, such as the Eos family, or salt deposits, such as 2 Pallas (Marsset et al. 2020). Also, (a) salts may be present in asteroids in the outer main belt (Poch et al. 2020), (b) theory and observations suggest that salts may be present in the Themis parent body (Castillo-Rogez & Schmidt 2010; Marsset et al. 2016), (c) salt precipitation can be explained by small percentages (Castillo-Rogez et al. 2018), thus it is very likely that the parent bodies of the Themis and Hygiea family were as likely to have precipitated salt as 1 Ceres and 2 Pallas. However, the proportion of salts produced in the Themis and Hygiea parent bodies is considered to be lower than in 2 Pallas, because the fraction of bright C-complex asteroids among members of the Themis and Hygiea family is smaller than in the Pallas family (table 6).

The spectral trend of bright C_p asteroids in other families shown in table 6 is similar to the spectral trend for all bright C_p asteroids. This might be caused by a common process, rather than common composition; one of the candidates for the common process is thermal metamorphism due to frequent impact events. Carbonaceous chondrites are known to get brighter albedo values upon thermal metamorphism (Hiroi et al. 1993; Cloutis et al. 2012). The area where an impact event would raise the temperature sufficiently to cause such a metamorphism is limited to several times as large as the size of the projectile at most (e.g., Davison et al. 2014; Kurosawa & Genda 2018). In addition, the area around the impact point becomes fine ejecta, so impact-metamorphosed asteroids may not remain as large objects. Thus, the ratio of bright C_p-class asteroids in a family composed of low-albedo members is considered to be as small as a few percent (table 6).

The ratio of all bright C_p-complex asteroids in the total population is different from that in low-albedo families. This might be caused by other factors than impact thermal metamorphism. Since DeMeo K-type and Xn-type asteroids were identified as the spectral type of bright C-complex asteroids in this study, bright C_p-complex asteroids not belonging to any families are considered to be composed of CV/CK chondrites, enstatite chondrites/achondrites, or impact-metamorphosed carbonaceous chondrites.

6 Conclusions

Spectroscopic observations of bright C-complex asteroids were carried out to unravel the origin and properties of such asteroids. The results suggest that most bright C-complex (Bus) asteroids are composed of DeMeo C-type with concave curvature, B-type, Xn-type, and K-type asteroids. The meteorites associated with these spectral asteroids have been found to be composed of minerals that have experienced high temperatures. Based on a comparison of observations in this study with the spectra of families, the origin of bright C-complex asteroids can be inferred as follows:

• Members of the Eos family, including the bright C-complex asteroids, are composed of CV/CK chondrite analogues.

• Salts may have been deposited in the Pallas, Themis, and Hygiea parent bodies. It is likely that not much salts were present in the parent body of the Themis and Hygiea family due to the paucity of bright C-complex asteroids in the families. On the contrary, it is possible that 2 Pallas had more salts than the Themis and Hygiea parent bodies since the Pallas family is composed of more bright C-complex asteroids.

• For other families with dark C-complex asteroids, the bright C-complex asteroids were most likely formed by impact thermal metamorphism.

• Bright C-complex asteroids that do not belong to any families could be linked with CV/CK chondrites and/or enstatite chondrites/achondrites, as well as impact-metamorphosed carbonaceous chondrites.

Acknowledgment

We are grateful to Dr. Víctor M. Alí-Lagoa and Dr. Marcela I. Cañada Assandri for sharing their valuable SDSS photometric data. We would like to thank the referee for her/his careful review and constructive suggestions, which helped us to improve the manuscript significantly. We are grateful to Dr. Masateru Ishiguro and Dr. Katsuhito Ohtsuka for insights and fruitful discussions. This study was based in part on data collected at Okayama Astrophysical Observatory and using the Subaru Telescope, which is operated by NAOJ. Data collected at Okayama Astrophysical Observatory were used, as well as data obtained from SMOKA, which is operated by the Astronomy Data Center, NAOJ. Taxonomic type results presented in this work were determined, in whole or in part, using a Bus–DeMeo Taxonomy Classification web tool by Dr. Stephen M. Slivan, developed at MIT with the support of National Science Foundation Grant 0506716 and NASA Grant NAG5-12355. This study has utilized the SIMBAD database, operated at CDS, Strasbourg, France, and the JPL Small-Body Database Browser, operated at JPL, Pasadena, USA. We would like to express our gratitude to the staff members of Okayama Astrophysical Observatory and the Subaru Telescope for their assistance. DK is supported by the Optical & Near-Infrared Astronomy Inter-University Cooperation Program, MEXT, Japan. This study was supported by JSPS KAKENHI (grant nos. JP15K05277, JP17K05636, JP18K03723, JP19H00719, JP19H00725, JP20H00188, and JP20K04055) and by the Hypervelocity Impact Facility (former facility name: the Space Plasma Laboratory), ISAS, JAXA.

Footnotes

References