1995 O1(Hale-Bopp)

PROCEEDINGS SUPPLEMENTS ELSEVIER

Nuclear Physics B (Proc. Suppl.) 69/1-3 (1998) 735-738

BeppoSAX LECS detection of comet C/1995 Ol(Hale-Bopp)* A. Owens, T. Oosterbroek, A. Orr, A.N. Parmar, R. Schulz a, L.A. Antonelli, F. Fiore b, G.P. Tozzic, M.C. Maccarone d, and L. Piro e aSpace Science Department of ESA, ESTEC, 2200 AG Noordwijk, The Netherlands bBeppoSAX Science Data Center, Nuova Telespazio, Via Corcolle 19 I00131 Roma, Italy COsservatorio Astro sico di Arcetri, Largo E. Fermi 5, 1-50125, Firenze, Italy dIFCAI/CNR, via U. La Malfa 153, 1-90146, Palermo, Italy eIAS/CNR, via Fosso del Cavaliere, 1-00133, Roma, Italy We report the detection of soft X-rays from comet C/1995 O1 (Hale-Bopp) by the LECS instrument on-board BeppoSAX on 1996 September 10-11. After correcting for the comets motion, a 7a enhancement was found within 2t of the expected location of the nucleus at much greater than 99.99% confidence. A weak 1.3a enhancement was also found in MECS data at the same location. The extracted LECS spectrum is well fit by a thermal bremsstrahlung model of temperature of 0.29 4- 0.06 keV - consistent with that observed in other comets. The 0.1-2.0 keV luminosity has a mean value of 5 x 1016 erg S - 1 and is a factor of -~ 8 greater than measured by the Extreme Ultraviolet Explorer (EUVE) 4 days later. This difference may be related to the emergence from the nucleus on 1996 September 9 of a dust-rich cloud which continued to expand rapidly over the next few days becoming tenuous by the start of the EUVE observation. There is no evidence for fluorescent carbon or oxygen emission with 95% confidence limits of 1.0 x 1015 and 7.8 x 1015 erg s -1 to narrow line emission at 0.28 and 0.53 keV, respectively. This implies that if such lines are present, their total luminosity must be must be <18% of the 0.1-2.0 keV continuum luminosity.

1. I N T R O D U C T I O N The detection of copious X-ray emission from comets C/1996 B2 (Hyakutake) and C/1990 N1 (Tsuchiya-Kiuchi) came as a considerable surprise [1,2]. The emissions were centered a few arcminutes (,--2 × 104 km) from the nucleus and had an extent of ,~10', or 5 × 104 km, being elongated normal to the Sun-nucleus line. The observed fluxes were a factor of -~103 greater than predicted by early models in which X-rays are generated by fluorescent emission and scattering of solar X-rays in the coma. Likewise, models which generate X-rays in the coma via solar wind proton, or electron, interactions also suffer from low efficiencies. In view of these difficulties, two models have since emerged which predict significantly

higher cometary X-ray intensities. In the solar wind model of H~iberli et al. [3] and Cravens [4], X-rays are generated following charge exchange excitations of highly ionized solar wind ions with neutral molecules in the comet's atmosphere. The attogram dust model of Wickramasinghe & Hoyle [5] and Krasnopolsky [6] produce X-rays from the scattering of solar X-rays in attogram (,~10 - i s g) dust particles, such as those detected in the wake of comet Halley [7]. Interestingly, :sub-micron sized grains have recently been observed in the coma of Hale-Bopp [8].

*The SAX satellite is a joint Italian and Dutch programme. T Oosterbroek and A. Orr acknowledge ESA Fellowships. 0920-5632/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved.. PII S0920-5632(98)00277-1

2. C O M E T

HALE-BOPP

Comet C/1995 O1 ( H a l e - B o p p ) w a s discovered in mid 1995 [9]. At that time, it was 7 AU from the Sun and already 200 times brighter than comet Halley at the same distance [10] and intrinsically the brightest comet seen in over 400 years

736

A. Owens et al./Nuclear Physics B (Proc. Suppl.) 69/1-3 (1998) 735-738

[11]. It is a long period comet whose orbit is almost perpendicular to the Earth's orbital plane. It passed within 1.3 AU of the Earth in 1997 March near perihelion passage. This should be contrasted with the Earth-comet distance of ~0.1 AU for comet Hyakutake. However, although Hale-Bopp was about 15 times more distant than Hyakutake when observed by X-ray instruments, it is intrinsically much brighter. At the time of the BeppoSAX observations, the comet was 2.87 AU from the Earth and was in a period of relatively constant brightness of magnitude 6. The size of the coma was visually estimated to be about 20 ~ and some observers reported a short dust tail of up to 0.5-1 °, possibly even 1.5 ° . The diameter of the nucleus was estimated by HST to be between 27 and 42 km [12], about 4 times larger than comet Halley. CCD images showed up to seven straight jets emanating from the nucleus [10]. UV, IR, and submillimeter measurements during the first year of observation found evidence for the production of large quantities of dust, carbon monoxide and other gaseous emissions. The emissions were considerably more copious than is usually seen in comets at these distances (e.g. [13-15]) and make Hale-Bopp a good candidate for X-ray emission and in turn, an ideal target for the BeppoSAX Low Energy Concentrator Spectrometer (LECS). 3. O B S E R V A T I O N Hale-Bopp was observed by BeppoSAX between 1996 September 10 04:55 and September 11 04:21 UTC, yielding a total LECS exposure of 11.5 ks. A motion corrected X-ray image of the region of sky containing Hale-Bopp is shown in Fig. 1. The image has been re-binned to a 56" x 56" pixel size and smoothed using a Gaussian filter of width (a) 1.5 ~ (1.9 x 105 km). The direction of the comet's motion and the position of the Sun are indicated. A weak (2.1 ± 0.3) x 10 -12 ergs cm-2s-t; 0.1-2.0 keV) source is visible whose centroid position is 1.7'±1', or (2.1±1.3) x 105 km, from the position of the nucleus, in the general solar direction. The extent of the emission is consistent with the LECS point spread function (PSF) of 9.5' FWHM at the mean energy of the detected

Soft X-re image of comet Hale-Bopp (0.1-2.0 keV)

-05o30'00"

-05°45'00''

........

~

/

............

~ ~ , ',

~-~- e o m e ~ s -

J

...,,,-~/ velocity } vector

i -06"00'00" . . . . . . . . . . . . . .

Sur~ . . . .

--

-06°15'00"

i

-06o30'00'' 17"36~00=

i

i L

17h34m30*

i i

17~33m00= Right oscens~on

~ I

17n31~30"

17N30moo"

Figure 1. A 0.1-2.0 keV image of the sky containing comet Hale-Bopp. The image has been corrected for the motion of the comet (,,, 17" hr-1). For comparison, the FWHM of the PSF at the mean energy of the source is indicated by the circle in the lower left hand corner. The position of the nucleus is indicated by the cross.

emission, although the width normal to the Sunnucleus axis appears wider than in the direction of motion, similar to that observed in other comets. The 68% confidence limit to any source extent is <6 ~ or < 8.1 x 105 km. The 0.1-2.0 keV luminosity is 4.8 x 1016 erg s -1, or 2.1 x 10 -12 erg cm-2s -1, for the best-fit thermal bremsstrahlung spectrum and 5.9 x 1016 erg s -1, or 2.6 x 10-12 erg cm-2s -1, for the best-fit power-law spectrum. The derived luminosities are a factor ,,~8 greater than those measured by EUVE 4 days later [16]. The extracted spectrum was found to be equally well fit (with X2,s of 9 for 9 degrees of freedom (dof)) by a thermal bremsstrahlung model of temperature 0.29 4- 0.06 keV (see Fig. 2) or a power-law model of photon index 3.1+0:0 2. 6 A blackbody gives a less acceptable fit with a )/2 of 18 for 9 dof (P(>x2)=3%). The 0.1-2.0 keV spectrum predicted by the Solar wind model of H~iberli et al. [3] was also fit to the LECS data. This spectrum consists of a series of narrow lines resulting from the excitation of highly charged

A. Owens et al./Nuclear Physics B (Proc. Suppl.) 69/1-3 (1998) 735-738

posures revealed no "hot spots", or other instrumental features at this position. The Medium Energy Concentrator Spectrometer [24] (MECS) detected excess 1.3-2.0 keV emission at a position consistent with the LECS detection (at 1.3a confidence) with an intensity of i3.3 + 2.5) × 10 -14 erg cm -2 S-1. This is in goo~i agreement with LECS 1.3-2.0 keV intensity of (1.8 + 1.1)× 10 -14 erg cm -2 s -1. Based on the above, the positional coincidence of the source and the spatial distribution of events (Sect. 4), i we identify Hale-Bopp as the source of the detected X-rays.

LECS observotions of comet Hcle-Bopp

o

c5

I(],I

02

0.5 Energy, keY

1

737

2

4. D I S C U S I O N

Figure 2. The observed LECS Hale-Bopp spectrum and best-fit thermal bremsstrahlung model of temperature kT = 0.29 keV. The lower panel shows the residuals.

states of O, C, and Ne ions (see Table 2 of ref. [3]). The line species should not vary appreciably from comet to comet, and so the line energies were fixed at the values given in [3]. The best-fit )/2 is 44 for 6 dof, with 3 of the lines requiring zero flux. The 95% confidence limits to C and O narrow line emission at 0.28 and 0.53 keV are 1.0× 1015 and 7.8× 1015 erg s -1. This implies that any such narrow line luminosity must be <18% of the 0.1-2.0 keV continuum luminosity. 3.1. A s s o c i a t i o n o f t h e o b s e r v e d e x c e s s with Hale-Bopp An examination of the R O S A T All Sky Sur-

vey catalog [20] revealed no known X-ray sources at this position. Using the R O S A T logN-logS relation [21] the probability of randomly detecting an X-ray source as bright as this within the extraction region is 6 x 10 -3. We exclude a UV leak as being responsible for the counts since the LECS has made a number of deep observations of UV bright stars such as Polaris and Capella and no unexpected features were found. In addition, there are no unusual objects in the SIMBAD database and inspection of several deep LECS ex-

Krasnopolosky et al. [17] have shown that the attogram dust model can successfully explain the measured fluxes from comet Hyaku~ake. Scaling the Hyakutake X-ray flux by the quiescent gas and dust production rates and heliospheric distances of Hyakutake and Hale-Bopp, the predicted Hale-Bopp X-ray flux is in good agreement with that measured by E U V E , under the assumption that the X-ray emissivities are similar. Since X-ray intensity is expected to scale with the amount of dust, a factor of ,~8 more dust is required to account for the X-ray intensity observed by the LECS assuming the dust size distribution and solar X-ray flux remain the same. Contemporary optical observations were carried out on the ESO 2.2 m telescope at La Silla using the multi-mode EFOSC II instrument in 3 narrow bands to differentiate between dust and gas emissions [18]. Hale-Bopp was iobserved at 01:00 UTC on the nights of 1996 Sept 10 to Sept 17 with exposure times of 180 s (dust), 180 s (C2) and 300 s (CN). Each image waS divided by the previous one so that temporal changes in the coma become apparent. This technique removes the strong smooth intensity profile of the coma so that superimposed time variable s~ructures become visible [19]. The quotient images show that a large dust cloud was ejected from the nucleus which expanded outwards and appeared as a shell when integrated over the line of sight. The dust production rate before the outburst is estimated to be ---4 × 104 kg s -1 and the ratio of dust to gas production ,,~3 [22]. The estimated

738

A. Owens et al./Nuclear Physics B (Proc. Suppl.) 69/1-3 (1998) 735-738

emission reported in Mumma et al. [23]. The bulk of the dust outburst occurred towards the North and North-West of the nucleus. It is interesting to note that the majority of the X-rays observed by E U V E also originated to the North of the nucleus (see Fig. 1 of ref. [16]). In the absence of a more convincing explanation, we attribute the reduced flux observed by E U V E to dust fragmentation to sizes which are increasingly inefficient in producing X-rays. This presumably occurred on a time scale of days. REFERENCES

Figure 3. The temporal evolution of the dust shell observed in Hale-Bopp between 1996 September 10 and 17. Each image is the ratio between an image and that obtained the previous day and thus highlights changes between consecutive days. The images have the same orientation as Fig. 1. The intensity coding has been modified to enhance the resulting shell in each case. White indicates a large change and black no change.

dust production rate during the outburst is at least 3 x 105 kg s -1, an increase by a factor of >7 than before the outburst. This is sufficient to account for the observed X-ray flux with the attogram dust model. We cannot directly measure the corresponding change in the amount of gas surrounding the comet. However during previous outbursts the gas production rate increased by a factor of only 1-3, depending on species [12]. Since the shell in Fig. 3 is thickening and expanding with time, its density is expected to vary as ,,,v -3, rather than v -2 expected for a shell of constant thickness. Thus, assuming constant gas and dust velocities and spherical outflow, the gas density will decrease --,200 times more rapidly than the dust due to its higher expansion velocity. By the start of the E U V E observation the dust density had decreased to a maximum of twice its quiescent level and the cloud had a size of ,-,2' or -~3 x 105 km. By the end of the E U V E observation, the cloud filled most of the aperture providing an explanation for the source of extended

1. C.M. Lisse et al., Science, 274 (1996) 205. 2. K. Dennerl, K. Englhauser & J. Triimper, IAU Circ. (1997) 6404. 3. R.M. H~iberli et al., Science, 276 (1997) 939. 4. T. Cravens, Geophys. R. Lett., 24 (1997)105. 5. N. Wickramasinghe & F. Hoyle, Ap. Space Sci., 239 (1996) 121. 6. V.A. Krasnopolsky, Icarus, 128 (1997) 368. 7. N.G. Utterback & J. Kissel, A J, 100 (1990) 1315. 8. D.M. Williams et al., ApJ, 489 (1997) L91. 9. A. Hale & T. Bopp, IAU Circ. (1996) 6187. 10. M.R. Kidger et al., ApJ (1996) submitted. 11. B.G. Marsden, IAU Circ. (1995) 6191. 12. H.A. Weaver et al., Science, 275 (1997) 1900. 13. A.W. Davis et al., Nature 380 (1996) 157. 14. N. Biver et al., Nature 380 (1996) 137. 15. D. Jewitt, M. Senay & H. Matthews, Science 271 (1996) 1110. 16. V.A. Krasnopolsky et al., Science, 277 (1997) 1488. 17. V.A. Krasnopolsky et al., Annales Geophysicae, 15, supp. III (1997) C802. 18. R. Schulz et al., to be presented at the 1st Int. Conf. on Hale-Bopp, Teneriffe, 1998. 19. R. Schulz, in Proc. 3rd ESO/ST-ECF Data Analysis Workshop, P. Grossbol & R. Warmels (eds), Garching: ESO (1991) 73. 20. W. Voges et al., A&A, submitted, 1996. 21. G. Hasinger et al., A&A, 275 (1993) 1. 22. G. Boella, et al., 1997, A&A, 122, 299 23. H. Rauer, C. Arpigny & H. Boehnhardt, Science, 275 (1997) 1909. 24. M. Mumma et al., IAU Circ. (1997) 6625.