3- to 14-mm Spectroscopy of Comet C/1999 T1 (McNaught Hartley)

Icarus 159, 234–238 (2002)
doi:10.1006/icar.2002.6882

3- to 14-µm Spectroscopy of Comet C/1999 T1 (McNaught–Hartley)

David K. Lynch1 and Ray W. Russell1

The Aerospace Corporation, M2/266, P.O. Box 92957, Los Angeles, California 90009
E-mail: [email protected]

and
Michael L. Sitko1

Department of Physics, University of Cincinnati, Cincinnati, Ohio 45221-0011

Received November 2, 2001; revised March 4, 2002

We report spectroscopy of Comet C/1991 T1 (McNaught– Viewed from a distance, the Sun would have a weak and prob-
Hartley) at 3–13 µm on January 31.62 and February 1.7 2001 UT ably undetectable IR excess with current instruments (Leger
(delta = 1.29 AU, r = 1.40 AU) using the broadband array spectro- et al. 1996, Angel and Woolf 1997.) The Oort and Kuiper belts
graph system on the IRTF. The spectrum showed a silicate emission are nonetheless key components of our Sun’s disk because
feature extending about 20% above the continuum. Two emission ejected comet dust is thought to produce a significant part of
features at 10.3 and 11.2 µm appeared above the silicate band, the the zodiacal dust.
latter seemingly indicative of crystalline olivine. The 10.3-µm fea-
ture is only a 1–2 sigma detection but if real could indicate the Comets themselves were surely formed in the outer reaches
presence of hydrated silicates. The color temperature at 8–13 µm of the Solar System. This is evident from their composition of
was 260 ± 10 K, approximately 6% above the blackbody radiative lighter elements and volatiles, in parallel with the outer plan-
equilibrium temperature of 235 K. The magnitude at [N] was 3.13 ± ets. Though composed mostly of water ice, they have a compo-
0.02. On the second night, the comet had brightened slightly ([N] = nent of nonvolatile dust. The dust particles range from less than
2.98 ± 0.02) and the two prominent emission features were absent, a micrometer to many tens or hundreds of micrometers. This
although the silicate emission feature maintained its trapezoidal is known from the spectral structure seen in the infrared (e.g.,
shape with shoulders at 9.5 and 11.2 µm. c 2002 Elsevier Science (USA) Hanner et al. 1994).

Key Words: comets, composition; comets, origin; infrared obser- And yet, we still do not know the precise composition of comet
vations; interplanetary dust; spectrophotometry. dust. IR spectroscopy has provided strong evidence of silicate
minerals since the early 1970s and the dust collectors from the
I. INTRODUCTION comet fly-bys in the 1980s did indeed find the in situ atomic
building blocks of Fe, Mg, Si, O, etc. (Kissel et al. 1986a,b). By
With the growing number of disks (Aumann 1988, Knacke collecting interplanetary dust particles (Bradley 1984, Sanford
et al. 1993, Beckwith and Sargent 1996, Backman et al. 1997), and Walker 1985, Bradley et al. 1992, Brownlee et al. 1993),
planets (Mayor and Queloz 1995, Marcy and Butler 1996, Butler we have a chance to look closely at what may be actual dust
and Marcy 1996), and possibly even comets (Grady et al. 1997, particles from a comet. These have exhibited a wide variety of
2000) discovered around solar-type stars, the understanding of compositions, many of them unexpected, such as FeS (Bradley
comets in our own Solar System has become key to interpret- 1995), but most of them were siliceous at heart and frequently
ing these observations. Our Oort and Kuiper belts represent amorphous or “glassy.”
remnants of a disk left over from the Solar System’s forma-
tion and their relative sparseness presumably represents an ad- In the roughly 25 years since it has been possible to obtain
vanced, though probably normal, stage in the disk’s evolution. 10-µm spectra of comets, it has become clear that while there
are many aspects of comet dust that are similar, all comets are
1 Visiting Astronomer, NASA Infrared Telescope Facility, operated by the different. Some are quiescent and show no silicates; others are
University of Hawaii under contract with the National Aeronautics and Space active and the silicate feature changes, often within hours, which
Administration. is indicative of an inhomogeneous composition or particle size
distribution or both. Our best hope in identifying the types of
dust and paving the way toward understanding their role in the
Solar System is to observe these rare bodies as often as possible.
In 1999, one such comet was discovered.

234

0019-1035/02 $35.00
c 2002 Elsevier Science (USA)
All rights reserved.

INFRARED SPECTROSCOPY OF COMET C/1999 T1 235

Comet C/1999 T1 was discovered by McNaught and Hartley
(1999) on October 7.64, 1999. It is a long-period (P ∼ 300,000
years), high-inclination (80◦) comet that reached perihelion on
December 14, 2000, and passed closest to Earth on February
2, 2001. In this paper we report and discuss spectra of Comet
C/1999 T1 (McNaught–Hartley) at 3–13 µm about a month and
a half after perihelion, when it was near its closest approach to
Earth.

II. OBSERVATIONS

The comet was observed on January 31.62 and February 1.7, FIG. 2. The spectrum changed significantly in 24 h. The peaks at 10.3
2001 UT at the NASA Infrared Telescope Facility (IRTF) on and 11.2 µm have disappeared. The 10-µm excess shows the now-familiar
Mauna Kea. At this time the comet was 1.4 AU from the Sun trapezoidal shape with breaks at around 9.6 and 12.2 µm and the 8- to 10-µm
and receding after perihelion (December 13.47, 2000). Observa- wing of the excess is depressed compared to the earlier spectrum. The 8- to
tions were made with Aerospace’s broadband array spectrograph 13-µm color temperature has increased to around 280 K.
system (BASS; Hackwell et al. 1990) with a 3.2-arcsecond-
diameter aperture, through which all wavelengths are measured mean of the spectra (9 spectra with 200 s of integration each on
simultaneously. the first night and 12 on the second night). Also shown is the
260-K gray body fit to the spectrum through the regions 8 and
The calibration source was Sirius (α CMa) on the first night 13 µm. The unusually large error bars and nonconforming points
and Arcturus (α Boo) on the second night, with α Boo being cal- between 9.3 and 9.7 µm are due to poor ozone cancellation
ibrated against α CMa on both nights. Both stars were observed and little can be learned about the comet from this wavelength
at about the same air mass as the comet so no extinction correc- interval.
tions were necessary. The flux model for α CMa was taken as a
Planck function at 11,070 K and normalized to magnitude −1.41 On the first night the spectrum showed a silicate emission
at 10 µm. For zero magnitude at all wavelengths we adopted a feature extending about 20% above the gray body continuum.
9700-K blackbody normalized to a value of 1.26 × 10−16 W Two emission features, one at 10.3 µm and the other at 11.2 µm,
cm−2 µm−1 at 10 µm. appeared above the silicate band. The 11.2-µm feature matched
the wavelength and width of the olivine feature seen in many
III. THE SPECTRA comets (Bregman et al. 1987, Campins and Ryan 1989, Lynch
Figure 1 shows the spectrum of the comet on January 31, et al. 1992, Hanner et al. 1994, Wooden et al. 1999). This is
2001. The error bars represent ±1 standard deviation of the also the highest contrast that this feature has ever displayed in a
comet and resembles a true emission feature more closely than it
FIG. 1. The spectrum shows a silicate emission feature extending from does a shoulder. The 10.3-µm feature is new and has never been
roughly 8.2 to 12 µm. Two features are evident at 10.3 and 11.2 µm (the feature seen in comets before. While this appears to be a firm detection,
at 9.7 is spurious due to poor ozone cancellation). The first peak at 10.3 µm is it is possible that it could be a statistical artifact, being only a
weak and, if real, is new. The second one at 11.2 µm has been seen in the spectra 2-sigma finding. It does, however, appear to be composed of two
of many comets before and is likely due to crystalline olivine. data points and not one, thereby supplying some confidence as
to its reality. It is also in a very clean and unstructured part of
the atmospheric transmission window. We cannot, however, rule
out the possibility that the feature is merely a statistical artifact
in the noise.

The comet was only marginally detected at short wavelengths
where it shone by reflected sunlight and, as a result, we were
unable to compute parameters relating to reflectivity. The L and
[N] (9.9–10.2 µm) magnitudes were 8.45 ± 0.16 and 3.13 ±
0.02, respectively.

Figure 2 shows the spectrum of the comet on the following
night, February 1, 2001 (12 spectra with 200-s integrations),
along with a 280-K gray body fit to the spectrum through the

236 LYNCH, RUSSELL, AND SITKO

regions 8 and 13 µm. Like many comets, Comet McNaught– does not produce a peak in another part of our observable spec-
Hartley changed overnight (compare Figs. 1 and 2). In addition trum, except perhaps at 11.2 µm, where its contribution would
to brightening slightly (L and [N] were 8.18 ± 0.1 and 2.98 ± blend with and be indistinguishable from that of olivine.
0.02, respectively), the possible emission feature at 10.3 µm
has disappeared and the 11.2-µm emission feature’s contrast is Comets are composed largely of water, a volatile, and it would
somewhat diminished, leaving a shoulder at 11.2 µm. The net seem reasonable to suppose that in the warm early presolar days
result is a spectrum showing a trapezoidal continuum that is when comets formed or with the chemical mobility that oc-
reminiscent of many comets (Hanner et al. 1994). We believe curs with periodic heating of a comet as it passes perihelion,
that the difference in the spectra between the two nights is real. some more complex and possibly even hydrated minerals could
The seeing was fairly good on both nights and the absolute signal be produced. The possibility that hydrated minerals could be
levels of the calibrators were the same to within 1–2%. Neither found in comets has been touched upon by Campins and Swindle
the 10.3-µm nor the 11.2-µm feature is in a region of signi- (1998) and recently H. Campins and co-workers (private com-
ficant atmospheric attenuation. The telluric opacity here is due munication) have found a spectral feature at 2.2 µm in Comet
primarily to water vapor, and the DC levels on the sky were the 28P/Neujmin 1 tentatively identified as the Al–OH bond that
same on the two nights to within our ability to measure them. resembles a feature found in smectite, a hydrated silicate. Smec-
tite does not have an emission feature that matches the 10.3-µm
The January 31 spectrum was reasonably well fit by a sin- feature.
gle Planckian gray body whose color temperature TC was 260 ±
10 K, about 10% above the radiative equilibrium blackbody tem- Minerals, especially hydrated minerals with cosmically abun-
perature TBB of 235 K (=278/R1/2 K) for a perfect blackbody dant compositions, that have single, isolated peaks at 10.3 µm in-
at a distance of R = 1.4 AU from the Sun. On the next night the clude natrolite [Na2(Al2Si3O10) · 2H2O], hackmanite [Na4Al3S
color temperature was 280 ± 10 K but owing to the increased (SiO4)3Cl], and sodalite (Na4Al3Si3O12Cl). Probably less likely
water vapor that night, uncertainty in the 8-µm point raises the but nonetheless spectrally similar minerals include crocidolite
error bars to ±25 K. Thus we could not be certain that any real [Na2Fe2+3Fe3+2Si8O22(OH)2], and carpholite [MnAl2Si2O6
change in the color temperature occurred between the two nights. (OH)4]. The features in both sodalite and natrolite are somewhat
wider than the 10.3-µm feature we report and are not at precisely
To the extent to which the “nonsilicate” continuum can be the same wavelength. In view of the fact that feature shapes and
characterized by a gray body that passes through the 8- and locations can vary according to particle size, we would not want
13-µm parts of the spectrum (Fig. 2), two aspects of the silicate to rule these out as candidate minerals. We caution, however,
emission can be seen: (1) It comes up slowly from around 8 µm that these possible identifications are based on a single spectral
and contains no features until the 10.3-µm feature, which itself feature.
begins to rise at around 10.1 µm., and (2) After the 11.2-µm
olivine feature, the silicate emission drops rapidly and essen- V. THE COMET–DISK CONNECTION
tially disappears by 12 µm.
In the past few years, there has been a growing understanding
IV. POSSIBLE ORIGIN OF THE 10.3-µm FEATURE of the mineralogical and chemical connection between comets
and circumstellar disks, especially in young stars. Such a con-
The first possibility is that both peaks were produced by the nection is not, perhaps, unexpected because current thinking
same mineral. The likeliest candidates are those already be- suggests that comets either are a component of circumstellar
lieved to have been identified in comets: olivine and pyroxene disks or are one possible end product of the disks’ evolution. The
(Wooden et al. 1999). Both produce two features at about the most compelling chemical connection is found in the optical and
right wavelengths but neither mineral seems likely to be able to UV spectra of certain main sequence stars where there are rapid
produce a peak at precisely 10.3 µm. In most spectra their short- variations in absorption line spectra of neutral and singly ionized
wavelength peak is closer to 10.1 µm. Thus we can tentatively species (Grady et al. 2000). These rapid changes are most easily
rule out both minerals as the source of both features with one interpreted as follows: Comets that are falling into or passing
caveat: Both are actually “mineral groups” with a common un- close to the star begin to evaporate; as the molecules and miner-
derlying structure but with significant compositional and molec- als dissociate into their individual atomic components, it is these
ular disorder variations under the same name. When these varia- components that we are seeing in absorption. To cause hourly
tions are taken with the effects of particle size, there are usually or daily spectral variations, such stars must have a dynamic and
small but significant wavelength shifts that could account for the very extensive comet cloud.
10.3-µm feature. Such effects can be easily seen by looking at
spectra of the same mineral at different particle sizes. The other piece of evidence linking comets and disks together
is found in their 10-µm spectra, such as the ones shown in Figs. 1
The other possibility is that two different minerals are respon- and 2. The silicate feature, at least in its unstructured form, has
sible for the two peaks, one for each. If we accept that olivine is been long known in both comets and circumstellar shells, but re-
producing the 11.2-µm peak, then we must find a mineral which cent advances in spectral resolution in the 10-µm window have
(1) produces a peak at 10.3 µm and with a suitable width and (2) allowed us to parse the silicate feature and discern narrower,

INFRARED SPECTROSCOPY OF COMET C/1999 T1 237

ACKNOWLEDGMENTS

We are grateful to Dave Griep at the IRTF for expert telescope operation and
to Daryl Kim and Suellen Brafford for help in obtaining the observations. This
work was supported by the Aerospace Corporation’s Independent Research and
Development program and by the United States Air Force Space and Missile
Systems Center through the Mission Oriented Investigation and Experimentation
program, under Contract F4701-00-C-0009. The data shown in Figs. 1 and 2 are
available from the author as digital ASCII (text only) files.

REFERENCES

FIG. 3. The spectrum of the Comet McNaught–Hartley (MH; see Fig. 1) Angel, J. R. P., and N. J. Woolf 1997. An imaging nulling interferometer to study
along with the spectra of Comet Hale–Bopp (HB; Hanner et al. 1999), HD35187 extrasolar planets. Astrophys. J. 475, 373–379.
(HD35, M. Sitko, in preparation), HD163296 and HD31648 (HD16 and HD36;
Sitko et al. 1999). All the spectra show some similarities which are related to Aumann, H. H. 1988. Spectral class distribution of circumstellar material in
a presumed common mineralogy. They all show broad emission between 8 and main-sequence stars. Astron. J. 96, 1415–1419.
13 µm extending above the blackbody continuum. Most evident are an emission
feature or inflection point at around 11.2 µm and another near 9.3 µm. The Backman, D. E., M. W. Werner, G. H. Rieke, and J. E. Van Cleve 1997. Ex-
scatter and larger error bars between 9.2 and 9.8 µm are due to incomplete ploring planetary debris disks around solar-type stars. In From Stardust to
ozone cancellation. Planetesimals, ASP Conference Series, Vol. 122, pp. 49–66. Astron. Soc. of
the Pacific, San Francisco.
lower contrast features that are useful in determining the min-
eralogy. Figure 3 shows a selection of silicate features in the Beckwith, S. V., and A. Sargent 1996. Circumstellar disks and the search for
spectra of comets and circumstellar shells. The most frequently neighbouring planetary systems. Nature 383, 139–144.
occurring such narrow feature is the one at 11.2 µm, most prob-
ably due to olivine in one crystalline form or another. Thus there Bradley, J. P. 1984. Nanometer-scale mineralogy and petrography of fine-grain
is strong evidence that in some cases and perhaps all, comets and aggregates in anhydrous interplanetary dust particles. Geochim. Cosmochim.
remnant circumstellar disks have a common mineralogy and al- Acta 58, 2123–2134.
most certainly a common evolutionary link.
Bradley, J. P. 1995. GEMS and new pre-accretionally irradiated relict
It is perhaps relevant to note that the 11.2-µm feature that grains in interplanetary dust—The plot thickens. Meteoritics 30,
is so common in comets is not found in most circumstellar 491–491.
shells. Among the HAeBes, T Tauri, and other YSOs, only
those HAeBe stars with thin shells and centrally cleared disks— Bradley, J. P., H. J. Humecki, and M. S. Germani 1992. Combined infrared
probably indicating more advanced disk age—display the and analytical electron microscope studies of interplanetary dust particles.
11.2-µm feature (Sitko et al. 1999) and the other forsterite fea- Astrophys. J. 394, 643–651.
tures at 19.5–33.5 µm (Malfait et al. 1998, van den Ancker
et al. 2000, Hill et al. 2001). As these disks are mostly likely to Bregman, J., H. Campins, F. C. Witteborn, D. H. Wooden, D. M. Rank,
have dust that has been heated over a long period, this is consis- L. J. Allamandolla, M. Cohen, and A. G. G. M. Tielens 1987. Airborne and
tent with laboratory studies of the transition from amorphous to groundbased spectrophotometry of Comet P/Halley from 5–13 µm. Astron.
crystalline silicates (Hallenback et al. 1998). Astrophys. 187, 616–620.

VI. SUMMARY AND CONCLUSIONS Brownlee, D. E., D. J. Joswiak, and S. G. Love 1993. Identification of cometary
and asteroidal particles in stratospheric IDP collections. Lunar Planet. Sci.
Comet C/1999 T1 showed relatively strong, structured silicate 24, 205–206.
emission with variable strength peaks at 10.3 and 11.2 µm. Its
color temperature was about 6% above the blackbody radiative Butler, R. P., and G. W. Marcy 1996. A planet orbiting 47 Ursae Majoris. Astro-
equilibrium temperature and showed a small brightness increase. phys. J. 464, L153–L156.
It seems evident that small silicate grains were a significant
component of the dust, and based on the 11.2-µm feature it is Campins, H., and E. Ryan 1989. The identification of crystalline olivine in
likely that crystalline olivine was one of them. The origin of the cometary silicates, Astrophys. J. 341, 1059–1066.
possibly new 10.3-µm feature is not known.
Campins, H., and T. D. Swindle 1998. Expected characteristics of cometary
meteorites. Meteoritics Planet. Sci. 33, 1201–1211.

Grady, C. A., M. L. Sitko, K. S. Bjorkman, M. R. Pe´rez, D. K. Lynch, R. W.
Russell, and M. S. Hanner 1997. The star-grazing extra-solar comets in the
HD100546 system. Astrophys. J. 483, 449–456.

Grady, C. A., M. L. Sitko, R. W. Russell, D. K. Lynch, M. S. Hanner, M. R. Perez,
K. S. Bjorkman, and D. de Winter 2000. Infalling planetesimals in pre-main
sequence stellar systems. In Protostars and Planets IV (V. Mannings, A. P.
Boss, and S. S. Russell, Eds.), pp. 613–638. Univ. of Arizona Press, Tucson.

Hackwell, J. A., D. W. Warren, M. Chatelain, Y. Dotan, P. Li, D. K. Lynch,
D. Mabry, R. W. Russell, and R. Young 1990. A low resolution array spectro-
graph for the 2.9–13.5 µm spectral region. In Proc. SPIE Conference 1235
on Instrumentation in Astronomy VII, pp. 171–180.

Hallenbeck, S. L., J. A. Nuth, and P. L. Daukantas 1998. Mid-infrared spectral
evolution of amorphous magnesium silicate smokes annealed in vacuum:
Comparison to cometary spectra. Icarus 131, 198–209.

Hanner, M. S., D. K. Lynch, and R. W. Russell 1994. The 8–13 µm spec-
tra of comets and the composition of silicate grains. Astrophys. J. 425,
274–285.

238 LYNCH, RUSSELL, AND SITKO

Hanner, M. S., R. D. Gehrz, D. E. Harker, T. L. Hayward, D. K. Lynch, C. G. Malfait, K., C. Waelkens, L. B. F. M. Waters, B. Vandenbussche, E. Huygen, and
Mason, R. W. Russell, D. M. Williams, D. H. Wooden, and C. E. Woodward M. S. de Graauw 1998. The spectrum of the young star HD 100546 observed
1999. Thermal emission from the dust coma of Comet Hale–Bopp and with the Infrared Space Observatory. Astron. Astrophys. 332, L25.
the composition of the silicate grains. Earth Moon Planets 79, 247–
264. Mannings, V., and A. Sargent 1997. A high-resolution study of gas and dust
around young intermediate-mass stars: Evidence for circumstellar disks in
Hill, H. G. M., C. A. Grady, J. A. Nuth III, S. L. Hallenbeck, and M. L. Herbig AE systems. Astrophys. J. 490, 792–802.
Sitko 2001. Constraints on nebular dynamics and chemistry based
on observations of annealed magnesium silicate grains in comets and Marcy, G. W., and R. P. Butler 1996. A planetary companion to 70 Virginis.
in disks surrounding Herbig Ae/Be stars. Publ. Natl. Acad. Sci. 98, Astrophys. J. 464, L147–L151.
2182–2187.
Mayor, M., and D. Queloz 1995. A Jupiter-mass companion to a solar-type star.
Kissel, J., and 22 colleagues 1986a. Composition of Comet Halley dust particles Nature 378, 355–359.
from Vega observations. Nature 321, 280–281.
McNaught, H., and M. Hartley 1999. Comet C/1999 T1 (McNaught–Hartley).
Kissel, J., and 18 colleagues 1986b. Composition of Comet Halley dust particles IAU Circ. 7273.
from Giotto observations. Nature 321, 336–337.
Sanford, S. A., and R. M. Walker 1985. Laboratory infrared transmission spectra
Knacke, R., S. Fajardo-Acosta, C. M. Telesco, J. A. Hackwell, D. K. Lynch, and of individual interplanetary dust particles from 2.5 to 25 microns. Astrophys.
R. W. Russell 1993. The silicates in the disk of ß Pictoris. Astrophys. J. 418, J. 291, 838–851.
440–450.
Sitko, M., C. Grady, D. K. Lynch, R. W. Russell, and M. S. Hanner 1999.
Leger, A., J. M. Mariotti, B. Mennesson, M. Ollivier, J. L. Puget, D. Rouan, and Cometary dust in debris disks of HD 31648 and HD 163296: Two “Baby” β
J. Schneider 1996. Could we search for primitive life on extrasolar planets in Pics. Astrophys. J. 510, 408–412.
the near future? Icarus 123, 249–255.
Sitko, M. L., D. K. Lynch, and R. W. Russell 2000. Silicate emission in TW the
Lynch, D. K., and S. Mazuk 2000. Adventures in modeling thermal emis- Hydrae Association. Astron. J. 120, 2609.
sion spectra of comets. Proceedings of NASA conference. In Thermal Emis-
sion Spectroscopy and Analysis of Dust, Disks and Regoliths (M. L. Sitko, van den Ancker, M. E., J. Bouwman, P. R. Wesselius, L. B. F. M. Waters, S. M.
A. Sprague, and D. K. Lynch, Eds.), ASP Conference Series, Vol. 196, Dougherty, and E. F. van Dishoeck 2000. ISO spectroscopy of circumstellar
pp. 127–166. Astron. Soc. of the Pacific, San Francisco. dust in the Herbig Ae systems AB Aur and HD 163296. Astron. Astrophys.
357, 325.
Lynch, D. K., R. W. Russell, J. A. Hackwell, M. S. Hanner, and H. B. Hammel
1992. 8- to 13-µm spectroscopy of Comet Levy 1990 XX. Icarus 100, 197– Wooden, D. H., D. E. Harker, C. E. Woodward, H. M. Butner, C. Koike,
202. F. C. Witteborn, and C. W. McMurty 1999. Silicate mineralogy of the dust in
the inner coma of Comet C/1995 O1 (Hale–Bopp) pre- and post-perihelion.
Astrophys. J. 517, 1034–58.