THE ASTROPHYSICAL JOURNAL, 491:L125–L128, 1997 December 20
© 1997. The American Astronomical Society. All rights reserved. Printed in U.S.A.
Up: Issue Table of Contents
Go to: Search Page | Previous Article | Next Article
Other formats: HTML (small files) | PDF (310 kb) | PostScript pages (113 kb)

Soft X-Rays From Four Comets Observed With EUVE

M. J. MUMMA

NASA Goddard Space Flight Center, Greenbelt, MD 20771

V. A. KRASNOPOLSKY

Catholic University of America/NASA GSFC, Greenbelt, MD 20771

AND

M. J. ABBOTT

Center for EUV Astrophysics, Berkeley, CA 94720

Received 1997 September 8; accepted 1997 October 16; published 1997 November 7


ABSTRACT

     EUVE observations of X-ray production rates, brightness distributions, and brightness maximum offsets from nuclei in four comets are compared with gas and dust production rates in those comets. This comparison favors charge transfer of solar wind heavy ions to cometary neutrals as a dominant process for X-ray excitation.

Subject headings: comets: general—comets: individual (Hyakutake, Hale-Bopp)—scattering—solar wind—X-rays: general

CONTENTS

§1. INTRODUCTION

     The discovery of X-ray emission from comet Hyakutake (C/1996 B2) by ROSAT (Lisse et al. 1996) and by EUVE (Mumma, Krasnopolsky, & Abbott 1997a) revealed a puzzling new X-ray phenomenon in the solar system: solar light scattered by comet Hyakutake in the visible region was weaker by a factor of 5 than that from the Moon, while X-rays from the comet exceeded those from the Moon by a factor of 600! The ensuing discovery of X-rays from comet Hale-Bopp (C/1995 O1) (Mumma, Krasnopolsky, & Abbott 1997b; Krasnopolsky et al. 1997; Owens et al. 1997) and five other comets (Dennerl, Engelhauser, & Trumper 1997) demonstrated that X-ray emission from comets is a general phenomenon. Here we discuss EUVE observations of X-rays in comets 6P/d'Arrest and Bradfield (C/1995 Q1) (Abbott, Krasnopolsky, & Mumma 1997), Hyakutake (C/1996 B2), and Hale-Bopp (C/1995 O1).

§2. OBSERVATIONS

     EUVE (Bowyer & Malina 1991) has two instruments sensitive to soft X-rays: the deep survey (DS) camera and the short-wavelength (SW) spectrometer. Both DS and SW cover a spectral range of 70–180 Å (180–70 eV). The DS half-maximum range is 100–165 eV. The DS peak effective area is 28 cm2 and is larger by a factor of 13 than that of SW. Only Hyakutake could have a signal measurable with SW. Here we discuss the DS measurements. The observing conditions are shown in Table 1.

     EUVE observations are processed in a standard manner at the Center for EUV Astrophysics (CEA) in Berkeley, using algorithms for targets with fixed positions. The work must be redone with additional steps for comets and other moving targets in order to map the detected photons relative to the position of the source. Software for moving targets did not exist at CEA, so we developed new software capable of performing this mapping while also keeping the comet-Sun direction fixed. However, this caused some delay in our data processing. In the analysis presented here, photons are remapped with corrections for cometary motion on the celestial sphere, the changing comet-Sun direction, the EUVE orbital parallax, and changes in the telescope pointing direction during the observations. The detected photon list is filtered to remove times when EUVE was in the South Atlantic Anomaly (which causes high background levels) and during occultations by Earth's limb below a tangent altitude of 200 km. Corrections for nonzero transmission of the DS optical filter at 1200–3200 Å (which is of the order of 10-8 of the peak value at 90 Å) are a few percent of the measured signal and are also made (Krasnopolsky et al. 1997).

§3. RESULTS

     EUVE data on X-ray radiation in the comets are summarized in Table 2. Two parameters are retrieved: the X-ray photon production rate (luminosity, QX) in the instrumental bandpass (FWHM of 100–165 eV or 123–75 Å) for a given aperture radius (ρ), and the sunward offset of the brightness maximum from the nucleus. In calculating the X-ray luminosities, we assume that radiation within the DS bandpass is spectrally uniform. We measured the offset of the brightness maximum relative to the nucleus in the sky plane and estimated the absolute offset (ρB) by assuming that the brightness maximum lay on the Sun's azimuth. The position uncertainty of stars observed with EUVE is 30&arcsec; (Abbott et al. 1996), and we adopted this value as the positional uncertainty of the nucleus.

     Images of Hyakutake and d'Arrest in soft X-rays are shown in Figure 1 (Plate L13). These images were constructed by convolving the measured cometary emission with a Gaussian having half-maximum radius of 5300 and 32,000 km, respectively. A similar image of Hale-Bopp was given in Krasnopolsky et al. (1997). D'Arrest looks rather symmetric relative to the brightness maximum, which is close to the nucleus (the offset is smaller than 30&arcsec;). The symmetry is partly due to a small phase angle: if a comet is axially symmetric relative to the Sun-comet line, then its image should be concentric at small phase angles regardless of the detailed intensity distribution. In contrast, Hyakutake was observed at small distances Δ≈0.12 AU and at a phase angle of 50°, and its crescent-like shape shows a large offset from the nucleus in a direction very close to that of the Sun—revealing the morphology of excitation of X-rays in the coma.

Fig. 1

     The EUVE X-ray luminosity of Hyakutake Q$\mathstrut{_{{\rm X}}}$=(7.5±1.5)×10$\mathstrut{^{24}}$ photons s-1 is in excellent agreement with the exposure-mean luminosity of 1.0×10$\mathstrut{^{25}}$ photons s-1 from revised analysis (Krasnopolsky 1997c) of observations with the ROSAT wide field camera (WFC) (Lisse et al. 1996) in the same aperture and spectral range. Our value of the sunward offset is larger by a factor of 2 than the ROSAT value obtained with the high-resolution imager (HRI). Temporal variations and the different spectral ranges may be considered to explain this difference. Recently, Krasnopolsky (1997c) reanalyzed the ROSAT data for Hyakutake and found that the exposure-mean emission was equal to 1.3×10$\mathstrut{^{16}}$ ergs s-1 in the ROSAT HRI range of 90–2000 eV. The soft X-rays in the DS spectral range constitute a third of the total luminosity seen by HRI.

     No emission was observed from Bradfield with a 2 σ upper limit of 57 photons for ρ = 120,000 km. In an aperture of the same angular size, 550 photons were detected from d'Arrest, 8100 photons from Hyakutake, and 480 photons from Hale-Bopp. The upper limit for Bradfield corresponds to the X-ray production rate of 2.8×10$\mathstrut{^{23}}$ photons s-1.

     Figure 2 shows azimuthally averaged brightness distributions of X-rays in d'Arrest and Hale-Bopp. We also show Hyakutake's distributions in the antisolar direction and in a direction that is normal to the Sun-comet line. We refer radius ρ to the brightness maximum. For uniform spherical outflow, gas and dust column densities in comets are proportional to ρ$\mathstrut{^{-1}_{0}}$ (ρ0 is the distance from the nucleus). If the X-ray brightness were proportional to the gas and/or dust column density, then a slope (α) of the curves in Figure 2 would be -1. Intervals where α=-1±0.5 are given in Table 2. Corrections of these intervals for phase angles and observing geometry are more complicated than those for ρB and are not considered here.

Fig. 2

     X-ray luminosities of three comets as functions of aperture radius are shown in Figure 3. This figure is helpful for comparison with other measurements made with different apertures.

Fig. 3

§4. GAS AND DUST PRODUCTION RATES

     Correlations of properties of X-ray emissions with gas and dust production rates may be used to identify a process of X-ray excitation. The water production rate in d'Arrest was measured by Mumma, DiSanti, & Xie (1995) at 8×10$\mathstrut{^{27}}$ s-1 during the EUVE observation. Production rates for OH and dust in d'Arrest were measured by D. G. Schleicher (1997, private communication): Q$\mathstrut{_{{\rm OH}}}$=9×10$\mathstrut{^{27}}$ s-1 and Afρ = 0.73 m on 1995 August 5–6, decreasing to 4×10$\mathstrut{^{27}}$ s-1 and 0.5 m, respectively, on 1995 November 18. Here Afρ is a photometric value that is proportional to dust production rate (A'Hearn et al. 1984); A is the particle albedo, f is the filling factor, and ρ is the aperture radius. The interpolated value Q$\mathstrut{_{{\rm OH}}}$=7×10$\mathstrut{^{27}}$ s-1 corrected for the yield of OH from H2O implies a water production rate coincident with that measured by Mumma et al. The obtained Afρ/QOH coincides with a value from previous passages of d'Arrest (A'Hearn et al. 1995). IUE observations on 1995 August 30 and September 22 gave Q$\mathstrut{_{{\rm H}_{2}{\rm O}}}$=1.8×10$\mathstrut{^{28}}$ and 1.4×10$\mathstrut{^{28}}$ s-1, respectively (M. C. Festou, G. P. Tozzi, & A. Talavera 1997, private communication). Then a log-mean Q$\mathstrut{_{{\rm H}_{2}{\rm O}}}$ and Qgas are equal to 1028 and 1.2×10$\mathstrut{^{28}}$ s-1, respectively, during our observation. Here we assume that water accounts for 80% of the total gas production.

     Q$\mathstrut{_{{\rm OH}}}$=2×10$\mathstrut{^{27}}$ s-1 and Afρ = 0.63 m were measured in Bradfield on 1995 November 18 (D. G. Schleicher 1997, private communication). We extrapolate these values to November 6–7 using Q=Ar$\mathstrut{^{k}}$ with kOH = -3.32 and k$\mathstrut{_{Af{\rho}}}$ = -2.39 for young, long-period comets (A'Hearn et al. 1995).

     Qgas and Afρ for Hyakutake are means of measurements by Mumma et al. (1996), Hicks & Fink (1996), and Millis et al. (1996), and for Hale-Bopp are from Weaver et al. (1997), Crovisier et al. (1997), Rauer et al. (1997), Schleicher et al. (1997), and Biver et al. (1997).

§5. EXCITATION PROCESS

     Of many processes that can produce X-rays in comets (Krasnopolsky 1997a, 1997b), four mechanisms were suggested to be significant: (1) charge transfer of solar wind heavy ions to cometary neutrals followed by X-ray emission (first suggested by Cravens 1997 and subsequently investigated by Haberli et al. 1997, Krasnopolsky 1997b, and Ip & Shemansky 1997), (2) scattering of solar X-rays by very small (10-19 g) dust particles (Wickramasinghe & Hoyle 1996; Krasnopolsky 1996, 1997a, 1997b), (3) spectral line radiation from electron impact and recombination excitation (Bingham et al. 1997), and (4) electron bremsstrahlung (Northrop et al. 1997; Northrop 1997).

     Only charge transfer and scattering by very small, so-called attogram dust will be considered here. Electrons captured in charge transfer may have an excess energy up to the ionization potential of the product ion (for example, 739 eV for O6+), and this energy is subsequently radiated. Attogram particles discovered by Utterback & Kissel (1990) with particle impact analyzers during the Vega and Giotto flybys of comet Halley, may be very efficient scatterers of solar X-rays. Krasnopolsky (1997c) established that the line radiation processes considered by Bingham et al. (1997) are either ineffective or refer to the ultraviolet. The only important process of electron impact is the excitation of K lines of O 525 eV and C 277 eV, at a level of 2% of the total emission. This value was calculated with the use of the Vega 2 electron flux spectrum (Gringauz & Verigin 1990). Krasnopolsky (1997c) found two errors in the calculations of electron bremsstrahlung by Northrop et al. (1997). His corrected value for the Vega 2 electron flux agrees with that from Bingham et al. (1997) and is equal to 0.5% of the total emission.

     The brightness offset ρB may be used to identify the process. In the case of attogram dust, the dust coma is optically thin, and hence the offset is independent of the dust production rate. The offset is related to asymmetric outflow into the sunward hemisphere, and to dust velocity, to the effects of solar radiation pressure on dust, and to charging of very small dust particles and removal by electromagnetic forces. If solar radiation pressure is the main removal process, then the offset is proportional to v$\mathstrut{^{2}_{d}}$/a and hence to r (v$\mathstrut{_{d}}$ is the dust velocity, which is expected to be proportional to gas velocity v=0.85r$\mathstrut{^{-1{/}2}}$ km s-1 [Cochran & Schleicher 1993] and a is the solar pressure deceleration). If charging of attogram dust and removal by electromagnetic forces are more important, then a lifetime t associated with these processes is proportional to r2, and ρ$\mathstrut{_{B}}$=vt is proportional to r3/2. Table 3 shows that attogram dust fails to explain the observed offset in d'Arrest, although it is consistent with the offset seen for Hale-Bopp. If charge transfer is a main excitation process, then a column density from infinity to ρB [N=Q$\mathstrut{_{{\rm gas}}}$/(4πvρ$\mathstrut{_{B}}$)] should be the same in all comets. This mechanism is consistent with the offsets seen for both comets (Table 3).

     Correlation of X-ray production rates with gas and dust production rates may also be used to choose between the excitation processes. With our restricted data, we assume a linear response of X-rays to gas or dust production. To reduce errors associated with this assumption, we use maximum apertures corresponding to the linear response (α≃-1 in Fig. 2 and Table 2): 7×10$\mathstrut{^{4}}$ km for d'Arrest (Q$\mathstrut{_{{\rm X}}}$=4.8×10$\mathstrut{^{23}}$ photons s-1), 1.3×10$\mathstrut{^{5}}$ km for Hyakutake (Q$\mathstrut{_{{\rm X}}}$=8.1×10$\mathstrut{^{24}}$ photons s-1, see a footnote to Table 2), and 4×10$\mathstrut{^{5}}$ km for Hale-Bopp (Q$\mathstrut{_{{\rm X}}}$=7×10$\mathstrut{^{24}}$ photons s-1). A comparison between the measured values and those calculated under the assumptions of correlations between X-ray production and dust or gas for dust scattering and charge transfer, respectively, is given in Table 4. The last column shows uncertainty factors that are based on mean log differences between the measured and calculated values. Evidently, excitation by charge transfer is in much better agreement with the observed luminosities, than is scattering from attogram dust, especially for d'Arrest. The X-ray luminosity of Bradfield is expected to be 1.2 and 0.3 that of d'Arrest for attogram dust scattering and charge transfer, respectively. Again, the measured upper limit favors charge transfer.

     One may expect 4πI$\mathstrut{_{B}}$r$\mathstrut{^{2}}$ to be constant for charge transfer if the solar wind variations are low and comets are collisionally thick. Here 4πI$\mathstrut{_{B}}$ is the X-ray brightness near the brightness maximum. Indeed, this value is 23.5 mR in d'Arrest, 30 mR in Hyakutake, and 40 mR in Hale-Bopp. The smaller value for d'Arrest may be due to the fact that the comet is not collisionally thick. This approach also suggests charge transfer as the dominant mechanism.

     Thus, the measured X-ray brightness offsets, production rates, and maximum brightnesses in four comets favor charge transfer of solar wind heavy ions to cometary neutrals as a dominant process of X-ray excitation in comets.

ACKNOWLEDGMENTS

     We are grateful to David Schleicher and Michel Festou for their unpublished data on water and dust production rates in d'Arrest and Bradfield. This work was supported by the EUVE Guest Observer Program.

REFERENCES

FIGURES


Full image (67kb) | Discussion in text

     FIG. 1.—X-ray images of comets Hyakutake (C/1996 B2) (A) and 6P/d'Arrest (B) observed with EUVE on 1996 March 21–24 and 1995 September 4–5, respectively. North is at the top, and east is to the left.


Full image (22kb) | Discussion in text

     FIG. 2.—X-ray brightnesses in three comets as functions of distance from the brightness maxima. Azimuthally averaged brightnesses are shown for d'Arrest and Hale-Bopp to improve the statistics. For Hyakutake, brightnesses in the antisolar direction (squares) and in the direction normal to the Sun-comet line (circles) are shown.


Full image (18kb) | Discussion in text

     FIG. 3.—X-ray luminosities (production rates) of three comets as functions of aperture.

TABLES

TABLE 1
OBSERVING CONDITIONS
Parameter a6P/d'ArrestC/1995 Q1 BradfieldC/1996 B2 HyakutakeC/1995 O1 Hale-Bopp
Dates...1995 Sep 4–51995 Nov 6–71996 Mar 21–241996 Sep 14–19
r (AU)...1.421.501.073.07
Δ (AU)...0.471.260.122.91
&phis; (deg)...21415019
ε (deg)...1208312488
τ (s)...4.5 × 1044.2 × 1041.06 × 1051.4 × 105
τeff (s)...4.2 × 1043.3 × 1041.0 × 1059.6 × 104

     a r and Δ are heliocentric and geocentric distances, &phis; is phase angle (Sun-comet-Earth angle), ε is elongation (Sun-Earth-comet angle), and τ and τeff are total and effective exposure times.

Image of typeset table | Discussion in text
TABLE 2
SUMMARY OF EUVE OBSERVATIONS OF SOFT X-RAYS IN FOUR COMETS
Parameter a6P/d'ArrestC/1995 Q1 BradfieldC/1996 B2 HyakutakeC/1995 O1 Hale-Bopp
r (AU)...1.421.501.073.07
QX (1023 photons s-1)...7≤2.87570
ρ (104 km)...12121240
ρSP (104 km)...≤1…4.414
ρB (104 km)...≤3…6 ± 127 ± 12
ρ(α = -1 ± 0.5) (104 km)...1.5–7…6 b15–40
Qgas (1028 s-1)...1.20.42060
Afρ (m)...0.650.8572630

     a r is the heliocentric distance, QX is the X-ray production rate within an aperture of radius ρ, ρSP and ρB are the brightness maximum offset from the nucleus in the sky plane and the corrected value, ρ(α=-1±0.5) (see § 3), and Qgas and Afρ are the gas and dust production rates.
     b The upper boundary is not covered by our observation.

Image of typeset table | Discussion in text
TABLE 3
BRIGHTNESS MAXIMUM OFFSET RATIOS ρB/ρBHya MEASURED AND CALCULATED FOR TWO EXCITATION PROCESSES
ρB/ρBHya ad'ArrestHale-Bopp
Measured...≤0.54.5 ± 2.3
Dust b...1.32.9
Dust c...1.54.9
Charge transfer...0.075.1

     a ρBHya is the offset in Hyakutake.
     b If solar radiation pressure dominates in dust removal.
     c If electromagnetic forces dominate in dust removal.

Image of typeset table | Discussion in text
TABLE 4
X-RAY PRODUCTION RATIO QX/QXHya a MEASURED AND
CALCULATED FOR TWO EXCITATION PROCESSES
QX/QXHyad'Arrest
(ρ = 7 × 104 km)		Hale-Bopp
(ρ = 4 × 105 km)		Uncertainty
Factor b
Measured...0.0590.86…
Dust...0.00283.39
Charge transfer...0.0181.12

     a QXHya = 8.1 × 1024 photons s-1 is the Hyakutake X-ray production for ρ = 1.3 × 105 km.
     b This factor is equal to 10$\mathstrut{^{{(}{\mid}{\rm log}x_{1}{\mid}+{\mid}{\rm log}x_{2}{\mid}{)}{/}2}}$; x1 and x2 are the values for d'Arrest and Hale-Bopp.

Image of typeset table | Discussion in text
Up: Issue Table of Contents
Go to: Top of This Article | Search Page | Previous Article | Next Article