COMET C/2004 Q2 (MACHHOLZ): PARENT VOLATILES, A SEARCH FOR DEUTERATED METHANE, AND CONSTRAINT ON THE CH₄ SPIN TEMPERATURE

The R0 and R1 lines (νにゅー₃ band) of CH₄ represent pure A and F transitions, respectively (A₂–A₁ and F₂–F₁; see Gibb et al. 2003 or Drapatz et al. 1987), and their flux ratio is sensitive to T_spin. The abundance ratio F/A increases with spin temperature until reaching the statistical equilibrium value of 9:5 at T_spin ≈ 45 K (see for example Figure 4 in Gibb et al. 2003). The strong R0 and R1 lines are detected simultaneously (Figure 2). Their flux ratio was determined with good precision and compared to the predicted flux ratios, that depend on both spin and rotational temperature (Gibb et al. 2003). We tested a range of spin temperatures, and (initially) adopted the rotational temperatures found for water measured simultaneously (Table 2). Measured and predicted line ratios are summarized in Table 5.

We found that the R0/R1 line ratio was consistent with spin temperatures higher than 35 K for both dates (2004 November 28 and 2005 January 19). Our January observations rule out T_spin below 38 K at the 99% confidence limit imposed by the 3σしぐま stochastic noise uncertainty (≈12%) in the measured line ratio. For November, the observed line ratio differs from the statistical equilibrium value by only 1.1σしぐま (the 3σしぐま stochastic noise limit is T_spin > 35 K). This result is consistent with the independently derived value by Kawakita & Kobayashi (2009), who report T_spin(CH₄) > 36 K at the 95% confidence limit, based on χかい² retrieval.

We verified that the assumed rotational temperature for methane does not impact our conclusion for T_spin ≳ 35 K. This conclusion holds for a wide range of T_rot as shown in Table 5 and Figure 4. Assuming a rotational temperature substantially lower than that found for H₂O reduces the agreement between observed and predicted line ratios for any value of the spin temperature. However, the predicted line ratio (R0/R1) for statistical equilibrium still provides the best match to the measurement. Assuming rotational temperatures higher than measured for water strongly favors T_spin ≳ 35 K.

Zoom In Zoom Out Reset image size

We also verified that the R0/R1 ratio (and respectively T_spin) is insignificantly affected by variation in the telluric transmittance function. The reason is that a change in the telluric transmittance produces a correlated change in the derived top-of-the-atmosphere fluxes of the two methane lines, so their relative intensities are only weakly affected. Finally, similar lower limits for T_spin are found at both blue (2004 November 28) and red (2005 January 19) Doppler shifts also in favor of the measurement's reliability.

The main limitations of the derived spin temperature are: (1) it is based on a single line of each spin species sampled (A and F for R0 and R1, respectively) and (2) a common adopted rotational temperature is assigned to each spin ladder. Most CH₄ lines are severely extinguished by telluric absorption at the Doppler shifts of our observations (given in Table 1). As a result, only R0 and R1 are detected with high signal to noise, required for a sensitive T_spin measurement.

Even if the Doppler shift was sufficient to obtain high signal to noise for a large number of methane lines, emissions other than R0 and R1 contain lines representing transitions from different spin ladders (Gibb et al. 2003) that remain blended at the spectral resolving power of NIRSPEC. In practice, it is more difficult to derive unique values for T_rot(CH₄) and T_spin(CH₄) from such blended transitions. Robust determination of T_rot(CH₄) and T_spin(CH₄) requires detections of multiple lines of spectrally resolved A-, F-, and E-type transitions that sample quantum states having a broad range of rotational energies. This would allow testing whether the rotational population distributions within a given spin ladder can be characterized by a single rotational temperature and whether the T_rot measured independently for the three spin ladders agree. Only then T_spin can be determined more reliably.

Our retrieval does not allow a robust line-by-line test of the CH₄ fluorescence model at hand, similar to the way it has been done for H₂O (e.g., Dello Russo et al. 2004; Bonev et al. 2007). For this reason, the reported error in T_spin(CH₄) does not include a potential model-related uncertainty.

We defer more detailed discussion on problems related to CH₄ spin temperature analysis in comets to a separate paper (E. L. Gibb et al. 2009, in preparation). Although potentially less robust, the value derived here for T_spin(CH₄) is in very good agreement with the result derived for H₂O in C/2004 Q2. Bonev et al. (2007) report T_spin(H₂O) > 34 K (this limit is dominated by the systematic uncertainty in line-by-line analysis which exceeds the uncertainty due to photon noise), while Kawakita & Kobayashi (2009) report T_spin(H₂O) > 27 K (95% confidence limit, based on χかい² retrieval).

The abundances of C₂H₆, CH₃OH, and HCN relative to H₂O in C/2004 Q2 are similar to those observed in several Oort Cloud comets, tentatively referred to as "organics-normal" (see Mumma et al. 2008, DiSanti & Mumma 2008, Mumma et al. 2003). The abundance of native H₂CO is at the lower end of the range observed in Oort Cloud comets (∼0.1%–∼0.8%).

The mixing ratios CO/H₂O and CH₄/H₂O were found to vary by an order of magnitude (or more in the case of CO) among comets (Mumma et al. 2003; Gibb et al. 2003, 2007b). However, there is no correlation between the CO and CH₄ abundances. This notion is supported by our observations of Q2/Maccholz where the mixing ratio CO/H₂O (≈5%) is intermediate in comparison with other comets (abundances vary between <0.1% and ∼15%), while CH₄/H₂O (∼1.45%) is near the high end of the observed range (< ∼0.1%–∼2.0%).

Interestingly, acetylene is low in abundance relative to both water and ethane. The C₂H₆/C₂H₂ ratio (∼5) in Q2/Machholz is substantially larger than observed in other comets. The relative abundance CH₄/HCN (∼10) is also distinctly larger than in previous comets observed (∼2.5 to ∼5.5).

More detailed intercomparison among comets observed in the IR (including C/2004 Q2) will be presented in a future review dedicated to the emerging taxonomic classification of comets based on parent volatile composition (see Mumma et al. 2008 for preliminary results). Here, we focus our discussion on two particular topics: evidence for compositional homogeneity in Q2/Machholz (Section 6.2) and cosmogonic implications of our measured CH₃D/CH₄ upper limit (Sections 6.3–6.5).

Infrared measurements of parent volatiles are rarely obtained over a large range of heliocentric distances for a given comet (see Section 3), owing to the difficulty in scheduling adequate observing time. Comets 1P/Halley and C/1995 O1 (Hale-Bopp) are exceptions. Both featured sufficient advance notice to permit planned campaigns. For Halley, water was studied from 1.16 AUえーゆー (preperihelion), through perihelion (0.58 AUえーゆー), and out again to 1.10 AUえーゆー (postperihelion). The ortho-para ratio of water remained unchanged over that interval (Mumma et al. 1993).

In Hale-Bopp, CO and C₂H₆ were measured from, respectively, 4.1 AUえーゆー and 3.01 AUえーゆー preperihelion, through perihelion (0.91 AUえーゆー), and out again to 2.83 AUえーゆー postperihelion (DiSanti et al. 2001, Dello Russo et al. 2001). The specific mass loss over this period (kg s⁻¹ * R_h², with R_h in AUえーゆー) was discussed and compared with similar measures of dust and other parent volatiles (Mumma et al. 2003). For comparison, in the radio Biver et al. (1997) studied extensively the long-term evolution of outgassing in Hale-Bopp between 7.0 AUえーゆー (preperihelion) and 4.0 AUえーゆー (postperihelion). These authors investigated the evolution of production rates and relative abundances of a number of species, founding that the latter did not exhibit substantial changes at R_h < 1.5 AUえーゆー. Such studies address (at least) two important questions as follows.

• 1.  
  Is the comet nucleus compositionally homogeneous (or heterogeneous) on a bulk scale?
• 2.  
  Are the observed chemical abundances of comet volatiles affected by heliocentric evolution of outgassing?

Several other comets were sampled at IR wavelengths at diverse heliocentric distances. In Comet C/2001 A2 (LINEAR), the H₂CO/H₂O ratio varied day-to-day by a factor of 4 at 1.16 AUえーゆー (while the comet was in outburst), while the CH₄/H₂O ratio increased by a factor of 2 as the comet moved from ∼1.16 AUえーゆー and 1.55 AUえーゆー (Gibb et al. 2007b). In 9P/Tempel-1, "Deep Impact" returned evidence of dissimilar CO₂/H₂O ratios in individual vents (Feaga et al. 2007). Such variability might be attributed to internal heterogeneity in volatile composition caused by radial dynamical mixing of cometesimals prior to forming a cometary nucleus.

By contrast, Comet 8P/Tuttle was observed over a range of heliocentric distances (1.16–1.08 AUえーゆー, spanning 38 days) during its favorable 2007–2008 apparition and very similar relative abundances resulted from all IR observations (Bonev et al. 2008, Böhnhardt et al. 2008), consistent with homogeneous volatile composition.

The split comet 73P/Schwassmann-Wachmann 3 (73P/SW3) provided the strongest evidence for compositional homogeneity. The mixing ratios of multiple parent volatiles measured in fragments B and C were remarkably similar (heliocentric distances 1.06–0.99 AUえーゆー; Dello Russo et al. 2007). The mixing ratios C₂H₆/H₂O and HCN/H₂O measured independently in fragment C (Villanueva et al. 2006; based on measurements at heliocentric distances 1.27 and 1.19 AUえーゆー) and in fragment B (heliocentric distance 1.03 AUえーゆー; Kobayashi et al. 2007) agree with one another and with those reported by Dello Russo et al. thereby increasing the evidence for compositional homogeneity of the parent body on bulk scales.

The mixing ratios of parent volatiles in C/2004 Q2 measured at heliocentric distance of 1.5 AUえーゆー agree with those measured at 1.2 AUえーゆー (Figure 3), although separated in time by nearly two months. During this interval, the comet's gas productivity increased by a factor of about 2 (Tables 1 and 2), while the abundances relative to H₂O remained (nearly) unchanged. The similarity in relative abundances measured at 1.5 AUえーゆー and 1.2 AUえーゆー contrasts strongly with the degree of chemical diversity observed among Oort Cloud comets (see Section 1).

Our results support chemical homogeneity in the mean volatile release from C/2004 Q2, but do not exclude the possibility that the nucleus has more than one active region (perhaps many) as suggested by studies of daughter fragments (Farnham et al. 2007; Lin et al. 2007). Even if different active regions are characterized by distinct mixing ratios among volatiles, the "mean" release exhibits nearly identical chemistry suggesting that roughly the same proportions of release are maintained. Such compositional similarities for a given comet strongly suggest that the heliocentric evolution of outgassing (whether gradual, e.g., 8P/Tuttle and Q2/Machholz, or abrupt, from outburst or splitting as in 73P/SW3) has not affected the observed abundances. The most plausible hypothesis is that these abundances are representative of the bulk volatile composition and reflect the early history of cometary ices.

Searches for deuterated species are a key part in testing the "interstellar-comet connection" (Charnley & Rodgers 2008a, 2008b). In particular,

• 1.  
  What is the processing history experienced by interstellar organic matter as it is incorporated into the disk of a (low mass) protostar?
• 2.  
  How (and to what extent) is the deuterium fraction modified (with respect insterstellar values) prior to incorporation of ices in cometary nuclei?
• 3.  
  Does "unprocessed" (without change in isotopic signatures) interstellar volatile material exist in comets?

Models that incorporate gas phase chemistry and trace the evolution of D/H from cold (10 K) molecular cloud, through core collapse, to formation of protoplanetary disk predict CH₃D/CH₄ ≥ 0.10 (Aikawa & Herbst 1999, 2001; see also the nice discussion in Kawakita et al. 2005). On the other hand, our measured upper limit (CH₃D/CH₄ < 0.02; 3σしぐま) agrees with predictions for methane formation via ion–molecule reactions in relatively warm (>25–30 K) gas (Millar et al. 1989; Charnley & Rodgers 2008a). Models of gradients in temperature and composition in protostellar cloud cores indeed predict low D/H for these outer regions—thought to be the main source for material in the disk at the time when comets form (Charnley & Rodgers 2008b).¹¹ Thus, the measured upper limit for CH₃D/CH₄ is consistent with, but (quite possibly) does not unambiguously imply formation of cometary methane exclusively via gas phase chemistry at temperatures exceeding 25–30 K.

An important question is whether our measurement can also be explained within the interpretation of Boogert et al. (2004), who detected both gas phase and solid phase CH₄ along the line of sight of the massive protostar NGC 7838 IRC 9 and suggested that methane is formed by H-atom addition reactions on grain surfaces and is intimately mixed with water in the protostellar envelope.

Gas-grain chemistry and the effects of three desorption mechanisms (thermal-, photo-, and cosmic ray induced desorption) were incorporated by Willacy (2007), who modeled the chemical evolution in a static (i.e., w/o mixing processes, see Section 6.4) T Tauri disk formed after the collapse of a molecular cloud. This model predicts CH₃D/CH₄ ratios in the range ∼0.07–0.20. The inconsistency with our upper limit might be related to the initial condition (CH₃D/CH₄ ≈ 0.07), resulting from methane formation in molecular cloud of temperature equal to 10 K.

These comparisons with chemical models suggest that the methane ice released from the nucleus of Q2/Machholz is not dominated by a component that keeps a chemical "memory" of cold (≪ 25 K) environments in the natal cloud or in the outer regions of the protoplanetary disk (where similar chemistry can occur despite higher densities; see Aikawa & Herbst 2001). The ${\rm (D/H)}_{{\rm H}_{\rm 2} {\rm O}}$ ratio (<2.3 × 10⁻⁴, preliminary result) in C/2004 Q2 (Biver et al. 2005) supports a similar conclusion for water. The spin temperatures of H₂O (>34 K, Bonev et al. 2007) and CH₄ (>35–38 K, this work) in C/2004 Q2 also support a relatively warm formation environment, providing T_spin is indeed a measure of the chemical formation temperature of the corresponding molecule.

These conclusions are tentative because the current models for evolution of deuterium enrichments in comet volatiles are still under development. One of the principal challenges is to investigate the possibility for nebular (versus interstellar) origin of volatile matter incorporated into comet nuclei.

An alternative pathway that can explain our CH₃D/CH₄ upper limit involves hydrocarbons that formed in the inner nebula. Based on laboratory simulations, Nuth et al. (2000) predicted that hydrogenated species can be synthesized along with crystalline dust in the hot environments of the inner solar nebula. Recent laboratory work (Nuth et al. 2008) demonstrated that an organics-rich macromolecular coating forms on various grain surfaces via reactions that reduce CO to produce hydrocarbons. In this process molecules, like CH₄ and C₂H₆ are released in the gas phase, while forming layers of carbon-rich macromolecular residue further providing a catalytic surface for continuing efficient synthesis of organic material. A similar process could produce hydrocarbons efficiently in the hot inner protoplanetary disk.

Outward radial transport of matter could bring these products of high-temperature chemistry to environments where ices can form. Several models (Dubrulle 1993, Drouart et al. 1999, Shu et al. 2001, Bockelee-Morvan et al. 2002, Gail 2002) explored mechanisms for radial mixing in the nebula, and predicted "grand-scale" transport of dust and gas within the nebular disk allowing high-temperature chemical products to be ultimately incorporated into comet nuclei. These predictions were resoundingly confirmed by analysis of refractory minerals in 81P/Wild 2 returned by the "Stardust" mission (Brownlee et al. 2006, Zolensky et al. 2006).

We expect that material synthesized in the hot inner nebula would be characterized by the protosolar value of D/H. Our observed upper limit for CH₃D/CH₄ corresponds to D/H (<5 × 10⁻³; 3σしぐま) exceeding the protosolar value ([2.35 ± 0.3] × 10⁻⁵)¹² by two orders of magnitude. Therefore, our result does not rule out the hypothesis that (some) methane (and other organics) in C/2004 Q2 was synthesized in the hot inner nebula. Under this scenario, the amount of inner-nebula organics stored in comet nuclei would be controlled by

• 1.  
  the time-dependent abundances of various hydrocarbons resulting from inner-nebula chemistry,
• 2.  
  the efficiency and distance scale (neither are well-understood at present) of outward radial transport in the protoplanetary disk at a particular time, and
• 3.  
  the conditions in outer ice-forming regions, where comet nuclei could accrete.

Since crystalline dust is also produced in the inner nebula (see review of Wooden 2008), the ratio of crystalline-to-total dust content in C/2004 Q2 would provide an additional constraint on the hypothesis of nebular origin for a fraction of organic material stored in the nucleus of this comet (Drouart et al. 1999). A high fraction of crystalline dust is predicted to correlate with low CO-to-hydrocarbons (e.g., CO/C₂H₆) and N₂/NH₃ ratios (Nuth et al. 2000).

Our upper limit for CH₃D/CH₄ (<0.020, 3σしぐま) implies that methane ice released from the nucleus of Q2/Machholz is not dominated by an unprocessed component formed in extremely cold (∼10 K) environments either in the inner portions of the protosolar molecular cloud or (at a later evolutionary stage) in the outer regions of the protoplanetary disk. We discussed three formation pathways for methane stored in the nucleus of Q2/Machholz:

• 1.  
  Formation of methane in a relatively warm (>25 K) gas via ion–molecule chemistry (consistent with the observed CH₃D/CH₄ upper limit).
• 2.  
  Formation of methane simultaneously with H₂O via H-atom addition reactions on interstellar icy grains in the protosolar envelope.
• 3.  
  Synthesis of methane and other hydrocarbons in the hot inner solar nebular, followed by outward radial transport of matter (characterized by protosolar D/H) to environments where comet nuclei can form.

Evaluating the relative contributions of these pathways and the extent to which each pathway is consistent with both the measured upper limit for CH₃D/CH₄ and the observed volatile abundances relative to H₂O requires further modeling of chemistry including both gas-phase and gas-grain processes in the natal interstellar cloud, during core collapse, and in the protoplanetary disk.