THE UNEXPECTEDLY BRIGHT COMET C/2012 F6 (LEMMON) UNVEILED AT NEAR-INFRARED WAVELENGTHS^*

Comets visit the inner solar system from at least two long-term reservoirs—the scattered disk/Kuiper Belt (KB) and the Oort cloud (OC). But they actually agglomerated in other regions during the formative phase of the solar system. Indeed, icy bodies (such as comets and trans-Neptunian objects) are believed to preserve material from that time. While some may have experienced subsequent processing owing to the effects of thermal warming, exposure to energetic radiation, and/or collisions since their formation, the relatively pristine composition of others may reveal the physicochemical conditions that prevailed 4.6 billion years ago.

Although an ongoing subject of discussion, current evidence suggests that most comets might have formed in an extensive region near or around the CO₂ and CO snow lines (over the range in heliocentric distance R_h ∼ 5–30 AU; Dodson-Robinson et al. 2009; A'Hearn et al. 2012; Qi et al. 2013) before migrating to their current reservoirs (OC and KB; e.g., Gomes et al. 2005; Brasser & Morbidelli 2013). Further lines of evidence also suggest that a stationary formation (i.e., one not accounting for radial mixing) might not fully account for the diversity of volatile abundances and other cosmogonic parameters in icy bodies, such as spin temperatures and isotopic ratios (see Paganini et al. 2012, 2013a, and references therein). Indeed, prior to outward migration, small bodies could have experienced possible chemical processing as a result of dynamical translation within the disk driven by the inward and outward migration of Jupiter and Saturn in the first 100,000 yr after formation of the Sun (the so-called Grand Tack event; cf. Walsh et al. 2011 and references therein).

Thus far, the analysis of primary volatiles in cometary comae, i.e., species directly released from the nucleus (traditionally termed parent species), represents unique evidence for (relatively) unaltered material. The importance of such measurements is twofold: it might help us unveil the primordial conditions that occurred during the early solar system, and it might help us prove whether dynamical conditions played an important role during planetary formation. In addition to studies of product species at optical wavelengths (e.g., A'Hearn et al. 1995; Fink 2009; Schleicher & Bair 2010; Langland-Shula & Smith 2011; Cochran et al. 2012), radio and infrared surveys of primary volatiles seek to build taxonomical databases for comets (for reviews, see Bockelée-Morvan et al. 2004; Mumma & Charnley 2011). Most of these studies show compositional diversity amongst objects from the OC and KB, thus demonstrating the need for more comprehensive statistical evaluations. Aided by dynamical (theoretical) studies and sample analyses of solid material (e.g., of returned dust samples; Brownlee et al. 2006), our ongoing IR survey is providing clues into the possible origins of the early solar system by analyzing a growing database of comets from both the OC and KB.

On 2012 March 23, Alex Gibbs of the Mount Lemmon Survey discovered an inbound faint object with a visual magnitude of ∼20 at 5 AU from the Sun, which subsequently was designated C/2012 F6 (Lemmon; hereafter C/2012 F6). Its eccentricity (e = 0.998), original semimajor axis⁶ (a = 458.08 AU), Tisserand parameter (T_j = 0.147), and inclination to the ecliptic (i = 826) classify it as belonging to the nearly isotropic (long-period) comet group originating from the OC (Nakano 2013). In January 2013, the comet underwent significant brightening. This prompted several optical observations, which revealed a greenish coma, indicative of a volatile-dominated spectrum (e.g., atomic oxygen and various carbon- and nitrogen-bearing species). We initiated a worldwide campaign using different facilities, including VLT/CRIRES, Keck-2/NIRSPEC, and IRTF/CSHELL, in the infrared. Here, we present results obtained for several volatiles in C/2012 F6 during this campaign, including production rates, rotational temperatures, spatial profiles, and cosmogonic parameters (ortho–para ratio and D/H in water).

We obtained high-resolution near-IR spectra of comet C/2012 F6 on five dates spanning February through June 2013 (bracketing perihelion, q = 0.73 AU on 2013 March 24), using three separate spectrometer–telescope combinations. Table 1 presents a log of our observations.

With CRIRES we used the 04 slit, which delivers high spectral resolving power (λ/Δλ ∼ 50,000) and spatial coverage of 40''. The adaptive optics module (Multi-Applications Curvature Adaptive Optics) provided seeing correction and image stabilization, thereby delivering excellent spatial registration. On both nights, we oriented the slit at position angle 290°, i.e., close to the Sun–comet direction (262° on February 2 and 288° on February 4). Atmospheric conditions were stable on both nights: relative humidity was ∼20%, wind speeds were below 7 m s⁻¹, and seeing was in the range of 10–15. Standard star HR 5336 (located near the comet's trajectory) allowed flux calibration and measures of column content of absorbing species in the terrestrial atmosphere.

With CSHELL we used the 2'' wide slit, maximizing the flux collection while still delivering moderate resolving power (λ/Δλ ∼ 20,000), which was adequate for our molecular studies. This slit width was necessary because C/2012 F6 was positioned 20°–25° south of the Sun and so was available only during daytime, which precluded use of the CCD guider in CSHELL. The 30'' long slit was oriented north–south on March 31, close to the projected Sun–comet direction (position angle 199°; Table 1), and east–west on April 1. Flux levels the terrestrial contribution were established through observations of the standard star HR 8728. For our CSHELL observations, three distinct but related science investigations were awarded eight consecutive days (UT March 25 through April 1), totaling 64 hr of clock time. However, problems associated with weather, observatory hardware, and high winds restricted our available time to the last two dates, resulting in a total of approximately 10 hr of clock time on the comet. The lack of active guiding during daytime⁷ also reduced the observing efficiency (e.g., see DiSanti et al. 2009 for details). Our measurement strategy balanced volatiles measured in all observing epochs (H₂O, C₂H₆, and CH₃OH) against those that were not measured with CRIRES or NIRSPEC (CO and CH₄), owing to inadequate Doppler shift on those dates. The resulting production rates and abundance ratios are listed in Table 2.

Table 2. Molecular Abundances for Volatiles in Comet C/2012 F6 (Lemmon)^a

Species	Date	νb	Linesc	Trot	Global Qd	Abundance
	(UT)	(cm−1)		(K)	(1027 s−1)	(Relative %)
2013 Feb 2e
f Rh = 1.20 AU; Δ = 0.99 AU; Δ-dot = −2 km s−1; P.A. = 2624; α = 534
OH*	Feb 2.31	mult.	24	(70)	187.88 ± 27.56	99
CH3OH	Feb 2.39	2840.61	10	$66_{ - 9}^{ + 11}$g	3. 50 ± 0.47	1.85 ± 0.27
C2H6	Feb 2.35	2992.68	6	$69_{ - 7}^{ + 9}$	0.49 ± 0.06	0.26 ± 0.03
C2H2	Feb 2.35	3293.80	9	(70)	<0.10	<0.05
HCN	Feb 2.31	3299.52	3	(70)	0.21 ± 0.08	0.11 ± 0.05
H2CO	Feb 2.31	2811.05	7	(70)	<0.22	<0.12
NH2	Feb 2.31	3310.10	4	(70)	0.58 ± 0.13	0.31 ± 0.07
NH3	Feb 2.31	3295.39	1	(70)	<1.20	<0.63
2013 Feb 4
fRh = 1.18 AU; Δ = 0.99 AU; Δ-dot = −4 km s−1; P.A. = 2876; α = 523
H2O	Feb 4.23	3014.67	8	$69_{ - 5}^{ + 8}$	189.05 ± 11.04	100
OH*	Feb 4.23	3397.83	3	(70)g	168.39 ± 14.98	89
HDO	Feb 4.25	2683.36	11	(70)	<0.93 ± 0.31	<0.49
2013 Mar 31h
fRh = 0.75 AU; Δ = 1.55 AU; Δ-dot = +16 km s−1; P.A. = 1993; α = 324
H2O	Mar 31.86	mult.	6/15	$80_{ - 10}^{ + 10}$	459 ± 65	100
CO	Mar 31.82	2152.64	2	(80)	19.5 ± 2.2	4.25 ± 0.64
C2H6	Mar 31.85	2981.86	2	(80)	1.80 ± 0.24	0.39 ± 0.07
CH3OH	Mar 31.85	2984.34	1	(80)	6.16 ± 0.68	1.34 ± 0.24
CH4	Mar 31.93	3040.64	1	(80)	3.06 ± 0.57	0.67 ± 0.11
H2CO	Mar 31.98	2783.60	3	(80)	2.46 ± 0.45	0.54 ± 0.12
2013 Apr 1h
fRh = 0.75 AU; Δ = 1.56 AU; Δ-dot = +15 km s−1; P.A. = 2018; α = 318
H2O	Apr 1.90	2984.34	6/8	$100_{ - 10}^{ + 10}$	661 ± 92	100
CO	Apr 2.03	2152.64	2	(100)	25.2 ± 2.3	3.82 ± 0.63
C2H6	Apr 1.81	2985.00	2	(100)	2.18 ± 0.22	0.33 ± 0.13
CH3OH	Apr 1.84	mult.e	3	(100)	6.76 ± 0.51	1.02 ± 0.39
2013 Jun 20
fRh = 1.74 AU; Δ = 1.79 AU; Δ-dot = +5 km s−1; P.A. = 2563; α = 334
H2O	Jun 20.50	3441.46	11	$50_{-6}^{+7}$	105.4 ± 6.6	100
C2H6	Jun 20.50	2945.67	13	$59_{-10}^{+14}$	0.31 ± 0.03	0.29 ± 0.03
CH3OH	Jun 20.50	2908.89	17	$50_{-5}^{+7}$	1.56 ± 0.14	1.48 ± 0.16
HCN	Jun 20.55	3299.54	5	(50)	0.23 ± 0.02	0.22 ± 0.03
NH3	Jun 20.55	3295.42	1	(50)	0.64 ± 0.22	0.61 ± 0.21
NH2	Jun 20.55	3301.84	2	(50)	0.12 ± 0.03	0.12 ± 0.03
C2H2	Jun 20.55	3295.94	8	(50)	<0.06	<0.06
H2CO	Jun 20.55	2851.41	6	(50)	<0.25	<0.24

Notes. ^aUncertainties represent 1σ, and upper limits represent 3σ. The reported error in production rate includes the line-by-line scatter in measured column densities, along with photon noise, systematic uncertainty in the removal of the cometary continuum, and (minor) uncertainty in rotational temperature. ^bMean wavenumber of all emission lines (used for this reduction) from a particular species. ^c"Lines" refers to the number of spectral intervals sampled. For the CSHELL water observations, the first entry indicates the number of intervals in the H2O_3A setting that were used to determine T_rot. The second entry indicates the total number of intervals used to calculate Q(H₂O), including water lines in the H2O_3A and CO_K settings, and OH* in the CH4_E and H2CO_B settings on March 31 and in the CH3OH_A setting on April 1, as indicated in Table 1. ^dGlobal production rate, after applying a measured growth factor of (1.7 ± 0.2) and (1.7 ± 0.1) to the nucleus-centered production rates of volatiles measured on UT 2013 February 2 and February 4, respectively. For CSHELL (March 31 and April 1), growth factors were: H₂O (2.0 ± 0.2), CO (1.9 ± 0.1), C₂H₆ (1.8 ± 0.05), CH₃OH (2.3 ± 0.1), CH₄ (2.2 ± 0.1), and H₂CO (1.8 ± 0.2). For NIRSPEC (June 20), GFs were: 3.1 ± 0.2 for H₂O and 1.8 ± 0.1 for minor species (based on measurements of C₂H₆, CH₃OH, and HCN). ^eWe adopt Q(H₂O) from February 4, which is consistent with production rates from (mean) Q(H₂O) and Q(OH*) from the combined CH, HCN, and CH₃OH settings on February 2. ^fR_h: heliocentric distance; Δ: geocentric distance; Δ-dot: geocentric velocity; P.A.: position angle of the extended Sun–comet vector; α: solar phase (Sun–comet–Earth) angle for the midpoint of data acquisition. ^gWe adopted T_rot (shown in parentheses) based on values measured for C₂H₆, CH₃OH, and H₂O from CRIRES and NIRSPEC spectra, and for H₂O from CSHELL spectra when calculating the NC production rates for species whose T_rot determination was not possible. ^hFor trace species measured on both dates, we obtained the following weighted mean values, 10²⁷ s⁻¹ (%): C₂H₆: 2.0 ± 0.2 (0.38 ± 0.06), CH₃OH: 6.0 ± 0.4 (1.25 ± 0.20), CO: 2.2 ± 0.2 (4.03 ± 0.45).

Download table as:  ASCIITypeset image

NIRSPEC is a cross-dispersed echelle grating spectrometer that samples a wide range of spectral orders with relatively high spectral resolving power. On June 20, we used a 0432 × 24'' slit configuration (delivering λ/Δλ ∼ 25,000) for observations of the comet, with the slit positioned along the projected Sun–comet direction. Flux calibration was achieved through observations of the IR standard star HR 6707 with a 0720 × 24'' slit configuration.

Comet spectra were acquired in our standard four-step sequence (ABBA), with accumulated times of 30, 60, or 120 s per stored A or B frame (depending on the instrument used and wavelengths targeted). In both the CRIRES and CSHELL cases, the telescope was nodded along the slit by 15'' between the A and B positions, and with NIRSPEC the telescope was nodded by 12''.

Data reduction and analysis of the spectral frames included flat fielding, removal of pixels affected by high dark current and/or cosmic-ray hits, spatial and spectral rectification, and spatial registration of individual A and B beams. After combining the A and B beams from the (processed) difference frames, which further assists in removing any residual background emission, we extracted spectra by summing 15 spatial pixels (13) centered on the nucleus for the CRIRES observations, 10 pixels (20) for the CSHELL observations, and 9 pixels (13) for the NIRSPEC observations. We synthesized a transmittance function for the terrestrial atmosphere by fitting the absorptions observed in the standard star spectra, and adjusting the estimated atmospheric contribution until an optimum fit was obtained. We used a multiple layer atmosphere and the LBLRTM model (Clough et al. 2005) that accessed the HITRAN 2008 molecular database modified with our custom updates (see, e.g., Villanueva et al. 2011b for details). We convolved our synthetic transmittance function to the resolving power of the comet observations, scaled it to the cometary continuum, and subtracted it from the measured spectrum. This isolated the net emission spectrum in excess of the cometary continuum. We then generated a fully resolved transmittance function for use in extracting the true transmittance experienced by individual spectral lines at their Doppler-shifted wavelengths.

These observations allowed robust measures of pre-perihelion rotational temperatures and production rates for multiple species (Paganini et al. 2013b). The simultaneous measurement of H₂O and its direct proxy (OH*) in each instrument setting permitted accurate characterization of the abundance ratio for each trace species. Spectra of H₂O and OH* are shown in Figure 1, and of other primary volatiles in Figure 2. Spectra resulting from the HDO setting are shown in Figure 3, along with a synthetic spectrum of expected HDO lines. With CRIRES, six volatiles (H₂O, OH*, C₂H₆, HCN, CH₃OH, and NH₂) were securely detected, and upper limits were retrieved for H₂CO, NH₃, C₂H₂, and HDO. Three of these species (H₂O, C₂H₆, and CH₃OH) were also measured with CSHELL and NIRSPEC. With CSHELL we also detected H₂CO, CO, and CH₄ on March 31/April 1 (Figure 4; DiSanti et al. 2013b). The latter two molecules require a geocentric velocity (Δ-dot) of ∼10 km s⁻¹ or higher to displace comet lines from their (opaque) terrestrial counterparts (Figure 4). Comet C/2012 F6 presented small Doppler shifts during the CRIRES and NIRSPEC observations, preventing detection of CO and CH₄ (Table 1).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

We determined the rotational temperature (T_rot) for an individual species, and we use this to quantify the production rate from the measured line fluxes. Results are given in Table 2, and the abundances obtained during our entire observing campaign are summarized in Table 3. The detailed procedures are discussed below.

3.1. Rotational Temperatures

3.2. Production Rates and Molecular Abundances

Using the optimum value for T_rot, an apparent production rate is determined from the flux of each rovibrational transition detected within our sampled aperture, F_line (W m⁻²). Our methodology considers the molecular lifetime τ(s), the molecular fluorescence efficiency or g-factor (g_i) at the appropriate T_rot and heliocentric velocity (v), terrestrial transmittance (T_i), geocentric distance (Δ), and the fraction of the total molecular content in the coma sampled by each pixel f(x). The nucleus-centered production rate is calculated using

$$\begin{equation*} Q_{{\rm nc}} = \frac{{4\pi \Delta ^2 F_{{\rm line}} }}{{\tau g_i T_i f(x)}}. \end{equation*}$$

Our IR observations are impacted by slit losses (i.e., loss of flux) that result from atmospheric "seeing" and slight aperture effects (e.g., imperfect centering of the cometary photocenter in the slit). To correct for these effects, we obtain a growth factor (GF) from the observed spatial profiles after averaging Q at diametrically opposite positions along the slit, as this averages asymmetries in outflow (Xie & Mumma 1996). For our infrared observations, multiplying the Q_nc by GF results in total (or "global") production rates, Q_tot.

Details of fluorescence models for each molecule can be found as follows: H₂O, HDO (Villanueva et al. 2012b), OH* (Bonev et al. 2006), C₂H₆ (Villanueva et al. 2011a), CO, C₂H₂, and CH₄ (Paganini et al. 2013a; Villanueva et al. 2011b; Gibb et al. 2003), NH₃, HCN (Villanueva et al. 2013; Lippi et al. 2013), NH₂ (Kawakita & Mumma 2011), H₂CO (DiSanti et al. 2006), and CH₃OH (Villanueva et al. 2012a; DiSanti et al. 2013a).

Obtaining reliable spatial profiles for observed volatiles requires adequate signal-to-noise ratios. CRIRES achieved such profiles for H₂O and C₂H₆, and for the continuum (see Figure 5(A)). These spatial profiles (pre-perihelion) reveal a fairly symmetric distribution of outgassing around the nucleus out to projected distances near 2000 km (as projected onto the sky plane), with some excess in the range ∼1000–1500 km, perhaps indicating the presence of icy grains toward the northwest (i.e., relative anti-sunward direction).

Zoom In Zoom Out Reset image size

With CSHELL (Figure 5(B)), we obtained spatial profiles for H₂O, C₂H₆, CO, and H₂CO. Compared to profiles obtained with CRIRES, the spatial distribution of H₂O, C₂H₆, CO, and the continuum extended to significantly larger nucleocentric distances (to ∼4000 km), owing to the larger production rates at 0.75 AU versus 1.2 AU (with CRIRES). Formaldehyde extended to ∼3500 km and displayed stronger intensities toward the anti-sunward direction. The continuum showed a rather symmetric distribution about the nucleus, while CO was enhanced in the sunward direction and C₂H₆ showed some excess flux in the anti-sunward direction. In agreement with CRIRES, the H₂O profile is rather symmetric with some enhancement in the anti-sunward direction at about 2000 km, similar to C₂H₆. The different configurations of CO, H₂O, and C₂H₆ could be related to separate polar and apolar release of ices in the nucleus (see Section 4.3). NIRSPEC profiles also revealed enhancement of C₂H₆ in the anti-sunward direction (similar to HCN and H₂O; Figure 5(C)). On the other hand, methanol shows a rather symmetric distribution about the nucleus, and the continuum was stronger in the sunward direction. The observed (flux) enhancement of H₂O relative to other minor volatiles could be related to the release of icy grains at R_h = 1.78 AU.

3.2.1. CRIRES Observations

The production rates and mixing ratios for water and seven trace species are given in Table 2. Our value for water production on February 4 (Q_tot = 1.9 ± 0.1 ×10²⁹ molecules s⁻¹) agrees with estimates of H₂O production using OH prompt emission lines (OH*, a direct proxy for water) detected in the CH, CH₃OH, and HCN settings on February 2, and in the HDO and H₂O settings on February 4 (Table 2). Our results show no evidence for significant variation in water production on February 2 and 4; we therefore conclude that adopting this value for both nights provides an accurate measure of molecular abundances.

3.2.2. CSHELL Observations

CSHELL observations occurred shortly after perihelion, when the gas production was larger than when measured pre-perihelion with CRIRES. The water production rate on April 1 (6.6 ± 0.9 × 10²⁹ s⁻¹) was marginally higher than on March 31 (4.6 ± 0.6 × 10²⁹ s⁻¹), an increase by a factor of 1.5 ± 0.4, or ∼1.5σ based on their combined uncertainties. The production rates for C₂H₆, CH₃OH, and CO were also somewhat higher on April 1, but the abundance ratios for trace gases were systematically lower on April 1 than on March 31 (see Table 2). However, both quantities were unchanged within their 1σ uncertainties, even though the slit was oriented north–south on March 31 and east–west on April 1 (see Table 2). This suggests a non-varying composition for the responsible source region(s).

3.2.3. NIRSPEC Observations

Our observing campaign concluded with a run on June 20, when comet C/2012 F6 was at 1.74 AU from the Sun and 1.79 AU from Earth. The water production rate was 1.1 ± 0.1 × 10²⁹ s⁻¹, and we securely detected C₂H₆, CH₃OH, HCN, NH₃, and NH₂. The results are listed in Table 2.

4.1. Water Production

The H₂O production rates we measured with CRIRES and CSHELL follow a heliocentric power law of $R_{\rm h}^{-2.2 \,{\pm}\, 0.8}$, which is consistent with insolation-limited outgassing, while a comparison of CSHELL and NIRSPEC observations resulted in a power law of $R_{\rm h}^{-1.9 \,{\pm}\, 0.4}$. A possible explanation for the relative increase in water production between March 31 and April 1 is the presence of strong jets near perihelion; however, in this case we would expect pronounced asymmetries that are not seen in our H₂O profiles (Figure 5). The most straightforward explanation favors an increase in overall gas production on April 1, perhaps due to preferential exposure of a larger or more active vent as a result of nucleus rotation, although the limited data sample restricts any conclusive interpretation.

In Figure 6, we compare our measured H₂O production rates with those inferred from Lyα using the SWAN all-sky camera on the Solar and Heliospheric Observatory (SOHO; Combi et al. 2013), and preliminary results inferred from the optical OH using the TRAPPIST telescope (revised estimates; Opitom et al. 2013a, 2013b; C. Opitom 2013, private communication). We do observe a systematic difference between the TRAPPIST results (from measurements of OH using Q(H₂O) = 1.1 Q(OH)) and those obtained with other techniques.

Zoom In Zoom Out Reset image size

We also compare with Q(H₂O) estimated from visual magnitudes from Seiichi Yoshida's webpage⁸ using the empirical relation log Q(H₂O) = 30.675 − 0.2453(m_v − 5 log Δ) given by Jorda et al. (2008). In early February, all three techniques (CRIRES, SWAN, and empirical estimates) provided similar results. However, the Jorda relation should be used with caution, since it does not account for the large differences seen among comets in C₂ emission intensities, in grain size distributions, and also in dust/gas ratios, all of which can influence the visible magnitude and thus introduce significant systematic uncertainties. Bearing in mind these limitations, these optical magnitudes provide estimates of Q(H₂O) in the range of 1.2–2.2 × 10²⁹ s⁻¹, consistent with SOHO/SWAN values (1.6–2.1 × 10²⁹ s⁻¹) and also with our CRIRES value (1.9 ± 0.1 × 10²⁹ s⁻¹).

Our CSHELL production rate for H₂O on March 31 (4.6 ± 0.6 × 10²⁹ s⁻¹) is lower than the prediction based on the Jorda estimate (∼(6.2–11.2)×10²⁹ s⁻¹); however, our value on April 1 (6.6 ± 0.9 × 10²⁹ s⁻¹) is consistent with the predicted value at the 1σ level. The SWAN values show better agreement with the predicted estimates. However, although each SWAN measurement represents the mean water production averaged over ∼10 days, the day-to-day variations near perihelion exhibit significant excursions—relatively larger than the (marginal) increase we measured between March 31 and April 1. Similar variations are not seen in the SWAN results in early February, around the time of our CRIRES observations. This could point to substantially more variable gas production around perihelion.

The water production rate we obtained with NIRSPEC (1.1 ± 0.1 × 10²⁹ s⁻¹) on June 20 (R_h = 1.74 AU) allows us to quantify a rather comprehensive evolution pre- and post-perihelion of comet C/2012 F6. At the 2σ level, our measurements display a behavior similar to estimates derived from the SOHO/SWAN results and the Jorda approximation. Even though our IR observations provide a general overview of the overall evolution of water production and abundance of minor volatiles (relative to water), upcoming measurements from other facilities will add key information to the evolution of water production rates, and thus the real behavior of comet C/2012 F6.

4.2. Abundances of Trace Volatiles—Placing C/2012 F6 (Lemmon) in Context

4.3. Possible Interpretation of Chemical Abundances in C/2012 F6

4.4. D/H in Water

4.5. Ortho–Para Ratio in Water

We investigated the production of primary volatiles in comet C/2012 F6 (Lemmon) using VLT/CRIRES on 2013 February 2 and 4 (pre-perihelion; R_h = 1.2 AU), IRTF/CSHELL on March 31 and April 1 (post-perihelion; R_h = 0.75 AU), and Keck-2/NIRSPEC on June 20 (post-perihelion; R_h = 1.74 AU). We detected 10 volatile species (H₂O, OH* prompt emission, C₂H₆, CH₃OH, H₂CO, HCN, CO, CH₄, NH₃, and NH₂), and obtained upper limits for two others (C₂H₂ and HDO). We quantified the OPR of H₂O using ortho and para emission lines (OPR = 3.31 ± 0.33), which constrains the spin temperature to be >37 K at the 95% confidence limit but is consistent with statistical equilibrium. Our measured production rate for H₂O and upper limit for HDO provided an upper limit for D/H in water (D/H < 2.45 × 10⁻³, i.e., <16 VSMOW).

With CSHELL we obtained a marginally higher water production rate on April 1 compared with March 31, but the values on both dates showed a reasonable excess compared with CRIRES results in February (and with NIRSPEC results in June), in accord with the increased (decreased) insolation between these epochs.

The obtained volatile abundance ratios are summarized in Table 3. Comet C/2012 F6 is rather depleted in C₂H₆ and CH₃OH; HCN, CH₄, and CO displayed normal abundances, and H₂CO was the only volatile showing an enhancement relative to water. These volatile abundances suggest a certain degree of chemical post-processing and/or formation within warm regions in the nebula. However, the normal abundance of HCN and of the hypervolatiles CH₄ and CO, plus the enhancement of H₂CO, argues against formation under such conditions, and suggests possible heterogeneity in this comet. Further evidence on this matter will be revealed as additional results from radio, infrared, and optical wavelengths become available.

We thank the VLT science operations team of the European Southern Observatory, Keck Observatory, and NASA's InfraRed Telescope Facility for efficient operations of the observatories. We gratefully acknowledge support by the NASA Postdoctoral Program (Paganini), by NASA's Planetary Astronomy (Paganini, Mumma, DiSanti, Villanueva) and Astrobiology (Mumma, DiSanti, Bonev) Programs, by NSF's Astronomy and Astrophysics Research Grants Program (Bonev), and by the Max Planck Gesellschaft (Böhnhardt). This material is based upon work supported by the National Aeronautics and Space Administration through the NASA Astrobiology Institute under Cooperative Agreement No. NNA09DA77A (Keane, Meech) issued through the Office of Space Science. L.P. also acknowledges Michael R. Combi for fruitful discussions on C/2012 F6 (Lemmon) and for sharing the SWAN results prior to final publication. We also acknowledge Emmanuel Jehin and Cyrielle Opitom for sharing pre-publication results from the TRAPPIST project, and the referee for helpful suggestions. The NASA-IRTF is operated by the University of Hawaii under Cooperative Agreement NCC 5–538 with the NASA-OSS Planetary Astronomy Program. The authors acknowledge the very significant cultural role and reverence that the summit of Mauna Kea has always had within the indigenous Hawaiian community. We are most fortunate to have the opportunity to conduct observations from this mountain.