Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://star.arm.ac.uk/iaj/online/v26n1p1.ps Äàòà èçìåíåíèÿ: Thu Feb 25 16:50:25 1999 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 02:40:41 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: m 63

Irish Astr. J., 26(1), 11­16, (1999) SPECTRUM OF RR TELESCOPII F. L. CRAWFORD
THE OPTICAL SPECTRUM OF THE SYMBIOTIC NOVA RR TELESCOPII
F. L. CRAWFORD
Dept. of Pure and Applied Physics, The Queen's University of Belfast, Belfast BT7 1NN, Northern Ireland
email: fergal.crawford@qub.ac.uk
ABSTRACT. The slow symbiotic nova RR Telescopii has been observed with the 3.9­m telescope of the Anglo­Australian
Observatory (AAO), using the University College London Echelle Spectrograph (UCLES) in conjunction with a Tek CCD.
The optical spectrum of RR Tel is very rich in emission. By comparing the results of this study with previous publications
on the subject, it is found that the RR Tel system is advancing towards higher degrees of excitation. It is also shown that
several nebular lines (for example, [O iii] –4363 š A and Ne iv –4714 š A) demonstrate component structure, perhaps caused by
the different densities of the emitting plasmas.
1. INTRODUCTION
The physical nature of RR Tel is highly complex; involved are
a white dwarf, a Mira variable of spectral type M5, a plan­
etary nebula­like plasma and an additional hot plasma that
could be due to colliding winds from the two stellar compo­
nents (Feast et al. 1983; Penston et al. 1983; M¨urset et al.
1991; Jordan, M¨urset & Werner 1994; Heck & Manfroid 1985).
The optical spectrum consists almost entirely of emission lines,
and an emission spectrum has been recorded which progresses,
with increasing time, to higher excitation. This progression is
presumably caused by the evolution of the hot component to
higher effective temperatures and smaller radii, which leads
to the photoionization of the outer atmosphere of the giant
(Kenyon et al. 1993). Thus, in addition to being a fascinating
and complex astronomical object, RR Tel is recognized as an
excellent source of emission lines that may be used to confirm
theoretical predictions of wavelengths. Many emission lines,
especially from high ionization stages of iron and nickel, have
been identified for the first time in the optical spectrum of
RR Tel (see, for example, Raassen 1985).
In an attempt to probe the optical spectrum of RR Tel
as deeply as possible, Crawford et al. (1998) obtained high
spectral resolution (¸ 6 km s \Gamma1 FWHM) observations, with
which they were able to measure reliable emission line inten­
sities for weaker features than was previously possible. The
present study uses the same data to illustrate the progres­
sion towards higher degrees of ionization and to investigate the
claims (e.g. Schild & Schmid 1996) that the nebular emission
shows density­dependent component structure.
2. OBSERVATIONS AND DATA REDUCTION
Spectroscopic observations of RR Tel were obtained using the
3.9­m telescope at the Anglo­Australian Observatory on 22
July 1996. The University College London Echelle Spectro­
graph (UCLES) was employed with the 31 lines mm \Gamma1 grating
and 700 mm camera, along with the Tek CCD as a detector.
The result was a high resolution, high signal­to­noise (S/N)
spectrum of RR Tel over the near­UV, optical and near­IR re­
gions, 3100 -- 9800 š A. The resolution was –/\Delta– ' 50,000, a
factor of 2 greater than for the previous data from CTIO
(McKenna et al. 1997), and the S/N was ' 20 pixel \Gamma1 in the
continuum, similar to the CTIO observations. This allowed
a greater dynamic range than with earlier observations, and
enabled the measurement of reliable line intensities for much
weaker features than was previously possible. The high resolu­
tion data is shown in Figure 1, with a labelled section of the
same appearing in Figure 2.
Fig. 1. The flux­calibrated high resolution optical spectrum of RR
Tel, where the flux is in units of 10 \Gamma11 ergs cm \Gamma2 s \Gamma1 š A \Gamma1 .
In order to provide absolute flux calibration of our data,
it was necessary to obtain low resolution spectra of RR Tel.
These were obtained on 2 August 1996 with the Australian
National University 2.3­m telescope and double­beam spectro­
graph, which gave a resolution of ' 2 š A pixel \Gamma1 . However, this
11

Irish Astr. J., 26(1), 11­16, (1999) SPECTRUM OF RR TELESCOPII F. L. CRAWFORD
Fig. 2. The flux­calibrated high resolution optical spectrum of RR Tel in the wavelength range 4850 -- 4980 š A, where the flux is in units
of 10 \Gamma12 ergs cm \Gamma2 s \Gamma1 š A \Gamma1 , featuring lines of the highly ionized species [Fe vii] at 4893.90 š A and 4942.30 š A and [O iii] at 4931.00 š A and
4958.91 š A, as well as the strong Hfi line at 4861.33 š A.
Fig. 3. The flux­calibrated low resolution optical spectrum of RR Tel, where the flux is in units of 10 \Gamma11 ergs cm \Gamma2 s \Gamma1 š A \Gamma1 .
12

Irish Astr. J., 26(1), 11­16, (1999) SPECTRUM OF RR TELESCOPII F. L. CRAWFORD
Fig. 4. Titanium oxide absorption bands detected in the spectra of RR Tel at 6651 š A, 7053 š A, 7666 š A and 8206 š A.
instrumental setup only gave useful data for wavelengths in the
range 3490 Ÿ – Ÿ 5530 š A. Line fluxes for wavelengths outside
this range were derived from the H i and He ii recombination
line spectrum (see e.g. Crawford et al. 1998). The low resolu­
tion spectrum is shown in Figure 3, where the accuracy of the
flux calibration is estimated at 10%.
All CCD images were reduced using the Image Reduction
and Analysis Facility (IRAF) version 2.3 implemented at the
Queen's University, Belfast starlink node. Standard proce­
dures were followed in the reduction of the images to one di­
mensional spectra. Low resolution spectra were optimally ex­
tracted and flux calibrated using standards from AAO lists.
The densely­packed 'echelle spectra were extracted using cos­
mic ray rejection algorithms for sky regions only, as cosmic ray
hits in the spectra themselves could be confused with real emis­
sion features. Multiple high resolution 'echelle images made it
possible to median filter the extracted spectra to subsequently
eliminate cosmic rays. Once reduced to one dimensional spec­
tra, the data were input into the user­friendly spectrum analy­
sis program DIPSO (Howarth et al. 1996) for further analysis.
3. LINE IDENTIFICATIONS
A complete list of all emission features in the RR Tel spec­
trum from –3183 --- 9455 š A is presented in Tables 2 and 3 of
Crawford et al. (1998). The methods employed in identifying
the many emission lines measured are also described therein.
It is noted that many previously unresolved lines have become
more easily identifiable, due to the increase in efficiency of de­
tectors. An example of this is Thackeray's (1977) –4488.93 š A
line, which he associated with both [Fe ii] –4488.75 š A and Fe
ii –4489.19 š A; both iron lines are clearly resolved in the RR
Tel spectrum used by Crawford et al. (1998), although not all
of their lines were fully resolved.
It is possible that there is some O vi emission at ––3811.36
and 3834.24 š A. The 3811.36 š A line was listed by McKenna
et al. (1997), although its doublet partner at 3834.24 š A was
not. Attempts were made by Crawford et al. (1998) to as­
certain whether or not these lines are real, as both features
were blends, by comparing the profiles of the O iii (multi­
plet 2) lines at ––3759.87, 3774.00 and 3791.26 š A to that of
the line at –3811.21 š A, which they identified as a blend of O
iii –3810.96 š A and O vi –3811.36 š A. However, this method
proved inconclusive.
There is also some evidence of TiO absorption in the
RR Tel spectrum. Thackeray (1977) noted that the presence
of the late­type star in the RR Tel system is confirmed by the
detection of TiO bandheads in the near infrared and red re­
gions of the spectrum. McKenna et al. (1997) were unable to
find any such evidence in their spectra from CTIO, but with
their higher resolution, Crawford et al. (1998) made defini­
tive identifications of the TiO absorption bands (see Figure 4).
13

Irish Astr. J., 26(1), 11­16, (1999) SPECTRUM OF RR TELESCOPII F. L. CRAWFORD
Fig. 5. The [O iii] 4363 š A emission line profile in RR Tel (flux measured in arbitrary units). The line profile has been fitted using two
components, separated by ¸ 28 km s \Gamma1 , and indicated by dashed and dotted lines, while the solid line shows the sum of these.
Identification IP (eV) I (1968) a I (1993) b I (1996) c
Fe ii (3779.58) 7.9 0.170 0.212 0.088
Fe ii (3824.91) 7.9 0.594 0.480 0.265
Fe ii (3914.48) 7.9 0.453 0.184 0.078
Si ii (3856.02) 8.2 0.453 0.324 0.170
Si ii (3862.59) 8.2 0.665 0.548 0.265
He ii (3923.48) 24.6 0.764 1.164 1.139
[Fe iv] (4206.30) 30.7 0.363 0.232 0.308
[Ne iii] (3868.74) 41.0 60.830 57.200 75.380
[Fe v] (3895.70) 54.8 2.122 2.552 2.678
[Ne iv] (4724.15) 63.5 5.374 3.032 3.378
[Fe vii] (4942.30) 100.0 6.046 6.240 8.092
[Ca vii] (4940.30) 108.8 --- 6.040 10.030
Table 1. An illustration of the changes occurring in the intensities of lines as RR Tel progresses towards a higher degree of
ionization. All intensities are given relative to Hfi = 100. Notes: (a) Aller et al. (1973), (b) McKenna et al. (1997), (c) Crawford
et al. (1998).
14

Irish Astr. J., 26(1), 11­16, (1999) SPECTRUM OF RR TELESCOPII F. L. CRAWFORD
Fig. 6. The [O iii] line profile at (a) 4959 š A and (b) 5007 š A, showing a four­component structure; (c) O ii –4072 š A showing two component
structure; (d) three­component structure in Ne iv –4714 š A (flux measured in arbitrary units).
These features are known to be prominent in mid­M­type stars
(Kirkpatrick, Henry & McCarthy 1991; Vardya 1992).
4. LINE INTENSITIES
Aller et al. (1973) list line intensities relative to Hfi. Thus,
an approximate comparison between the intensity of RR Tel's
lines in 1968 (Aller et al. 1973), 1993 (McKenna et al. 1997)
and 1996 (Crawford et al. 1998) was possible, by scaling all line
intensities relative to Hfi. It was found that the strengths of the
lines for low ionization species are decreasing, while the line in­
tensities of species of higher degrees of ionization are increasing
(see Table 1). For example, the line intensities of species such
as He i, [O i], Fe ii and O iii are decreasing, while the intensi­
ties of [Fe vii] lines are increasing. This is the expected result,
as RR Tel's emission spectrum has been well documented as
advancing towards higher excitation (Kenyon et al. 1993; Allen
1980).
5. COMPONENT STRUCTURE IN THE NEBULAR LINES
The data used in this investigation are of sufficiently high res­
olution to show component structure in many of the nebular
lines of RR Tel. For example, the [O iii] –4363 š A line shows
definite evidence of a two component structure in the nebular
emission (see Figure 5). Schild & Schmid (1996) note that this
may illustrate a combination of a high density component with
a logarithmic electron density (Ne in cm \Gamma3 ) of ¸ 8, along with
a low density component (log Ne ¸ 5.5), which was found to
be blue­shifted by ¸ 28 km s \Gamma1 with respect to the high den­
sity feature (see Figure 5), in comparison to the blue­shift of
¸ 20 km s \Gamma1 found by Schild & Schmid (1996).
The [O iii] lines at ––4959 and 5007 š A also show compo­
nent structure. However, in these cases, the line profiles com­
prise at least three components, making the modeling process
much more complicated (see Figure 6). A detailed analysis of
these lines may be used to demonstrate additional physical
characteristics of the RR Tel system, such as its winds (Nuss­
baumer & Dumm 1997). Similar structural attributes to those
of the [O iii] lines may also be found in the profiles of O ii, N
ii, N iii and Ne iv.
6. CONCLUSION
The number of lines measured by McKenna et al. (1997) and
not listed in Thackeray (1977) is 70, including [Cu ii] at
–3806.34 š A, which was predicted by Thackeray to appear in
RR Tel, due to its presence in the spectrum of j Car (Thack­
eray 1953). In Crawford et al. (1998), a further 333 lines were
15

Irish Astr. J., 26(1), 11­16, (1999) SPECTRUM OF RR TELESCOPII F. L. CRAWFORD
measured, also some of the lines listed in McKenna et al. (1997)
were discarded and others re­identified. This was made possible
by the improved resolution of that study.
Crawford et al. (1998) further investigated the changes
that are continually occurring in the emission spectrum of
RR Tel. They found that the tendency towards higher levels
of ionization has continued, as noted by Kenyon et al. (1993)
and Allen (1980).
TiO absorption bandheads may be seen in the near in­
frared and red regions of the spectrum, which indicate the
presence of the late­type star in the RR Tel system, as noted
by Thackeray (1977). These features are prominent in mid­M­
type stars (Kirkpatrick, Henry & McCarthy 1991). Figure 4
shows definite identifications of these bands.
Acknowledgements
The author is grateful to the Department of Education for
Northern Ireland for the award of a research studentship. This
work was supported by NATO travel grant CRG/930722 and
the National Science Foundation through Grants Nos. AST
90­14133, AST 93­13991, and AST 94­16985 to the University
of California, Los Angeles. The author is also grateful to Alan
Fitzsimmons, QUB, for obtaining the observations presented in
this paper. This research has made use of NASA's Astrophysics
Data System Abstract Service.
References
Allen D. A., 1980, MNRAS, 192, 521.
Aller L. H., Polidan R. S., Rhodes E. J., Wares G. W., 1973,
ApSS, 20, 93.
Crawford F. L., McKenna F. C., Keenan F. P., Aller L. H.,
Feibelman W. A., Ryan S. G., 1998, ApJS, submitted.
Feast M. W., Whitelock P. A., Catchpole R. M., Roberts G.,
Carter B. S., 1983, MNRAS, 202, 951.
Heck A., Manfroid J., 1985, A&A, 142, 341.
Howarth I. D., Murray J., Mills D., Berry D. S., 1996, DIPSO
(V3.4), Starlink User Note, No. 50.20.
Jordan S., M¨urset U., Werner K., 1994, A&A, 283, 475.
Kenyon S. J., Mikolajewska J., Mikolajewski M., Polidan R.
S., Slovak M. H., 1993, AJ, 106, 1573.
Kirkpatrick J. D., Henry T. J., McCarthy D. W., 1991, ApJS,
77, 417.
McKenna F. C. et al., 1997, ApJS, 109, 225.
M¨urset U., Nussbaumer H., Schmid H. M., Vogel M., 1991,
A&A, 248, 458.
Nussbaumer H., Dumm T., 1997, A&A, 323, 387.
Penston M. V. et al., 1983, MNRAS, 202, 833.
Raassen A. J. J., 1985, ApJ, 292, 696.
Schild H., Schmid H. M., 1996, Nebular Density Profiles of
PU Vul and RR Tel, in Mikolajewska J. (ed) Physical Pro­
cesses in Symbiotic Binaries and Related Systems.
Thackeray A. D., 1953, MNRAS, 113, 211.
Thackeray A. D., 1977, Mem. RAS, 83, 1.
Vardya M. S., 1992, Long Period Variable Stars, in Maran S.
P. (ed.) The Astronomy and Astrophysics Encyclopedia,
Cambridge University Press, Cambridge.
16