Abundance analysis of some bipolar Type I Planetary Nebulae
O. Lorente Espín1, A. Riera1,2, B. Balick3, G. Mellema4, K. Xilouri5 and Y. Terzian5
(1) Departament de Física i Enginyeria Nuclear, Universitat Politècnica de Catalunya, Spain. E-mail: oscar.lorente-espin@upc.es
(2) Departament d’Astronomia i Meteorologia, Universitat de Barcelona, Barcelona, Spain
(3) Department of Astronomy. University of Washington. USA
(4) Netherlands Foundation for Research in Astronomy. The Netherlands
(5) Department of Astronomy. Cornell University. USA
Introduction
Abundance variations within a single PN have been reported within several PNe containing FLIERs by Balick, Perinotto, Macioni, Terzian and Hajian(1994; BPMTH) , along the major axis of the bipolar PN M1-75(Guererro, Stanghellini and Manchado 1995), and along the PN IC 4406 (Corradi et
al. 1997). If real, these abundance variations imply that the parent nuclei of these nebulae eject highly processed stellar material which could be an important source of heavy element enrichment in the ISM.
Mass loss during the actual PN phase is observed, but the amount of mass lost at this phase is minor. Even so, it is possible that if the atmosphere consists largely of highly processed gas, then considerable enrichment of CNO in the nebula might occur if the heavy-element-rich winds mix into the
nebular gas ejected much earlier. The gas is accelerated by atmospheric radiation pressure to very high velocities (>1000 km s-1) and ejected in fast winds. However, the structure of the wind-driven bubbles is such that not much fast wind material is expected to end up in the nebula, let alone outside
it.
Alternatively, the abundances variations within a single PN are not real and appear as a result of the error inherent to the standard methodology used to determine the elemental abundances of photoionized nebulae (i.e. the "ionization correction factor" ("icf") method) (e.g., Alexander and Balick
1997; Moore, Hester and Dufour 2004). The "icf" method converts the observed nebular emission line fluxes into chemical abundances. Using the standard diagnostic techniques, the electron density and temperature, and the ionic abundances of species with observed optical lines are derived from
observed emission line ratios. Total chemical abundances are then derived by correcting the ionic abundances from the unseen ionization stages using the icf's.
To study the putative enrichment within a PN, we have undertaken deep long-slit spectroscopic observations of selected targets in which we expected to find clear evidence of abundance gradients.
NGC 6501
M 1-75
Observations
NGC 6537
Hb5
The present abundance study combines spatially resolved optical observations of several Type I planetary nebulae, covering the wavelength of 3500 to
7400 Å. The targets include objects with very large [NII]/Hα ratios (> 1) ( M 1-8, M 3-3, M 2-55, Hu 1-2, Sh 2-71, M 1-75); and several large bipolar PNe
(NGC 6537, NGC 650-1, Hubble 5) whose lobes exhibit [NII]/Hα ratios near unity or higher. Figure 1 shows the [N II] + Hα images of the PNe of our
sample. The sample includes nebulae of different excitation classes (from medium to high excitation) (Gurzadyan and Egikyan 1991).
The observations were conducted using the Palomar 200 inch (5.1 m) telescope and the double spectrograph on the night of 1996 March 26. Standard
gratings and spectrograph configurations were used to obtain two simoultaneous spectra in the blue and red. The dispersion is 2.1 Å pixel-1 in the blue and
3.1 Å pixel-1 in the red. The spatial scale is 0.78’’ pixel-1. In 1996 June several planetary nebulae were observed using the IDS (Intermediate Dispersion
Spectrograph) with the 235 mm camera at the 2.5 m Isaac Newton Telescope (Observatorio del Roque de los Muchachos, La Palma). The dispersion is 3.3
Å pixel-1, and the spatial scale was 0.33’’ pixel-1. Slit positions are shown in Figure 1 for each nebulae.
33
2
1
M 3-3
M 1-8
M 2-55
Hu 1-2
Analysis
A single extinction parameter was used to deredden all measurements made along the slit across one nebula. The c(Hβ) parameter was obtained from the
averaged Hγ/Hβ and Hα/Hβ ratios.
Density structure was measured using the density-sensitive ratio of the S+ forbidden lines at 6717, 6731 Å. These lines are quite bright and well measured.
Wherever possible we also used the ratios of Cl++ lines to guage the densities in regions of high ionization.
Temperatures were measured from the ratio of nebular and auroral forbidden lines of N+ and O++ in the usual manner (e.g. Kaler 1986). No small-scale
temperature fluctuations were observed. The comparison of fig. 2a and 2b show that, as expected, higher excitation nebulae have larger [OIII] temperatures
than medium-to-low excitation nebulae (e.g. Kaler 1986). The high excitation nebulae of our sample show [OIII] temperatures well above the T[NII] values.
It is interesting to note the extremely large [OIII] temperatures measured at the central regions of Hu1-2 with values above 16000 K (fig.2a).
Figure 1. Hα + [NII] images of M 1-75, M 2-55, NGC 650-1, Hu 1-2, M 1-8 and M 3-3 were retrieved from “ The IAC
Morphological Catalog of Northern Galactic Planetary Nebulae” by A. Manchado, M.A. Guerrero, L. Stanghellini and M. Serra-Ricart.
The images of Hubble 5 and NGC 6537 were retrieved from the HST pictures site. The positions of the long-slit are shown in these plots.
With these physical properties mapped, it is straightforward to compute ionic abundance profiles. For this we made fits to the ionic abundances from 5level ionic computations using the NEBULAR package in IRAF.STSDAS.PLAYPEN (Alexander and Balick 1997).
(b)
(a)
Total abundances are computed from the sum of all observed ionization species multiplied by an ionization correction factor (icf) which accounts for
unobserved ionization stages. We used the icf s compiled by Kingsburgh and Barlow (1994) for the plots below. Other icf s (Köppen, Acker and Stenholm
1991, de Freitas Pacheco et al. 1993) give somewhat different, but not necessarily any better or more consistent results.
The various graphs show the ionic and total abundances (see Figure 3). An inspection of the figure shows that, for the most part, the abundance profiles are
constant or nearly so except where noise clearly affects the outcomes. This is a satisfying result. However, there are several cases that deserve close
scrutiny.
The analysis we present here on abundance variations within a single PN points out some unexpected and inexplicable results concerning the increase of
N/H, S/H and Ar/H abundances at the low-ionization edges or features of some nebulae (see the plots of Hu 1-2, M 2-55, M 1-75 and Hb 5 in Figure 3).
We conclude that the standard methodology to derive the total abundances - that is based on the assumption that N+ and O+ all have the same volume, and
also that occupy Ne++ and O++ the same region - cannot be applied straightforwardly to the low ionization regions. Our results are confirmed by the
investigations of the accuracy of the icf method to infer the chemical abundances from nebular spectra developed by Alexander and Balick (1997) and
Moore, Hester and dufour (2004). Alexander and Balick (1997) show that in the outer parts of the nebula some of the assumptions of the icf approximation
breaks down. Therefore, the use of spatially resolved observations and the icf method would lead to the erroneous conclusion that N/O and N/H increase at
the low ionization regons. Except for O, all icf abundances are vulnerable to large systematic errors with nebular radius (specially, for the bipolar nebulae).
Any conclusion about the abundance variations within a single nebula must rely on detailed photoionization models. For non-spherically symmetrical
objects a 2D or 3D photoionization codes must be used.
NGC 6501
Hu 1-2
M 1-75 (1)
0.20
0.16
0.20
He/H
0.15
0.40
He/H
He/H
0.16
0.30
0.10
0.10
0.08
0.10
He++
N/H
N/H
5E-4
2E-4
4E-4
1E-4
2.0E-3
6E-3
1.6E-3
N/H
2.0E-4
1.6E-4
7.5E-4
O/H
8E-4
2.5E-4
O+
O++
O+
4.0E-5
O/H
8E-4
O++
O0
2E-4
O/H
2E-4
4.0E-4
O/H
6E-4
3E-4
4E-4
2E-4
N+
N+
8E-4
O++
O++
2E-4
1E-4
O+
O+
O+
1E-4
1E-4
Ne/H
8E-5
8.0E-6
2E-4
4E-5
O0
4E-5
1E-4
Ne++
Ne++
Ne++
S/H
S/H
S++
2E-6
8E-6
3E-5
2E-5
S/H
6E-6
S/H
S++
S++
1E-5
5E-6
7.5E-6
5E-6
Ar/H
4E-6
7.5E-7
Ar++
150
200
250
2E-6
300
20
2E-6
4E-6
1E-6
S+
S+
5.0E-6
2E-6
S++
6E-6
S+
6E-6
Ar/H
6E-6
Ar/H
4E-6
8E-6
Ar/H
3E-6
Ar/H
4E-6
4E-6
2E-6
4E-6
1E-6
Ar3+
60
80
100
120
140
20
40
60
80
100
120
20
40
60
Ar++
2E-6
Ar++
2E-6
2E-6
Ar++
1E-6
Ar++
Ar++
2E-6
1E-6
Ar3+ Ar4+
40
Ar++
Ar++
2E-6
Ar4+
1E-6
100
4E-6
S+
Ar/H
2E-6
2.5E-6
Ar++
Ar4+
50
Ar/H
4E-6
Ar++
Ar++
60
3E-6
3E-6
2E-6
50
Ar/H
Ar/H
Ar/H
1.0E-5
6E-6
5E-6
3E-6
Ar3+
40
6E-6
S+
3E-6
5.0E-6
3E-6
30
6E-6
S++
2E-6
S+
S+
S+
4E-6
2E-6
S++
2E-6
S+
Ar/H
20
S/H
1.5E-5
S++
S++
S/H
8E-6
S/H
S/H
O0
4E-6
6E-6
4E-6
2E-5
2.5E-7
8E-6
6E-6
2E-5
1E-6
5.0E-7
2.0E-5
1E-5
4E-6
S+
O+
O+
8E-6
Ne++
4E-5
4.0E-6
S++
8E-6
S/H
2E-4
6E-5
2E-5
1E-5
O++
O++
2E-4
2E-5
6E-5
5E-5
S/H
1.0E-6
4E-4
Ne/H
Ne++
S+
4E-4
O0
3E-4
Ne++
1E-4
S/H
3E-6
6E-4
O0
4E-5
1E-4
O/H
6E-4
4E-5
2E-4
8E-4
O/H
O++
6E-5
2E-4
3E-4
1.2E-5
Ne++
8.0E-4
4E-4
O/H
Ne/H
Ne/H
O0
2E-5
8E-4
8E-5
6E-5
Ne/H
Ne++
2E-4
O+
N+
N+
O0
3E-4
4E-4
4E-4
N+
2E-4
O+
Ne/H
O++
N/H
1.2E-3
2.5E-4
4E-4
O+
5E-4
Ne/H
1.6E-3
N/H
6E-4
2E-4
O++
O+
3E-4
Ne/H
4E-5
6E-4
3E-4
1E-3
4E-4
O0
O0
4E-4
2E-4
6E-4
8.0E-5
O++
6E-5
He++
N/H
7.5E-4
5.0E-4
O/H
O0
N+
1E-3
1.2E-4
4.0E-4
He++
1.0E-3
N/H
N/H
0.05
N+
O/H
1.2E-3
6E-4
0.05
He++
9E-4
He+
He+
0.05
He++
He+
0.10
5.0E-4
2E-4
2E-4
1.0E-3
8.0E-4
O++
8E-4
He+
He+
4E-4
8E-4
1E-4
N/H
He+
4E-4
N+
O/H
N+
3E-4
0.15
6E-4
N+
1E-4
0.10
0.05
1E-3
8.0E-4
1.6E-3
2E-4
6E-4
6E-4
4.0E-4
O/H
0.04
1.2E-3
2E-3
4E-4
0.10
He+
He++
8E-4
4E-4
2E-4
0.15
He/H
He/H
6E-4
4E-3
N/H
3E-4
N+
He+
8E-4
N/H
He/H
0.15
0.15
0.20
0.10
0.08
He++
He+
0.05
He+
0.20
He/H
He/H
0.10
He+
0.04
He++
0.20
M 2-55
M 1-8
0.12
He++
He+
0.04
3E-4
0.15
0.08
0.05
0.16
M 3-3
Hb 5(3)
0.20
He/H
He/H
0.20
He+
0.05
He+
0.20
0.12
He++
Hb 5 (2)
Hb 5 (1)
0.20
He/H
0.15
0.12
NGC 6537
M 1-75 (2)
Figure 2: The [OIII] temperature vs. [N II] temperature for: (a) high excitation PN and (b) medium excitation PN. Labels in the
figure indicate the name of the nebula and the slit position. We have considered different spatial regions along the slit (with
different temperatures).
80
100
Figure 3. Ionic abundances for several ionization stages and total abundances of He, N, O, Ne, S and Ar of our PN sample as a
function of the position along the slit (in pixels). The name of the nebula and the slit position number are also shown. In the plot of
He, the blue dashed line corresponds to He+/H+ inferred from the HeI 5876 Å and the red dashed line to the He+/H+ abundance inferred
from HeI 6678 Å. In the plot of O, the blue dashed line corresponds to the O+/H+ relative abundance calculated from the [O III] 3727
Å, and the red dashed line corresponds to the relative abundance inferred from the [O III] 7325 Å. The icfs of N, Ne and S include the
O+/H+ relative abundance. The blue solid line in these plots implies the use of the O+/H+ relative abundance inferred from the [O III]
3727 Å, while the red solid line implies the use of the O+/H+ relative abundance inferred from the [O II] 7325 Å.
0
20
40
60
80
0
20
40
60
80
0
20
40
60
80
0
10
20
30
40
50
0
20
40
60
0
20
40
60
80
References:
Alexander, J., Balick, B. 1997, AJ, 114, 713;
Balick, B., Perinotto, M., Maccioni, A., Terzian, Y.,Hajian, A. 1994, ApJ, 424, 800;
Corradi, R. L. M., Perinotto, M., Schwarz, H. E., Claeskens, J.F. 1997, A&A, 322, 975;
de Freitas-Pacheco, J. A., Barbuy, B., Costa, R. D. D., Idiart, T. E. P. 1993,A&A, 271, 429;
Guerrero, M. A., Stranghellini, L., Manchado, A. 1995, ApJ, 444, L49;
Gurzadyan, G.A., Egikyan, A.E. 1991, Ap&SS, 181, 73;
Kaler, J.B. 1986, ApJ, 308, 322;
Kingsburgh, R. L., Barlow, M. J. 1994, MNRAS, 271, 257;
Koppen, J., Acker, A., Stenholm, B. 1991, A&A, 248, 197;
Moore, B.D., Hester, J.J., Dufour, R.J. 2004, AJ, 127, 3484
100