Old Galactic Novae Shells
An Honours/1st year Project Proposed by : Mike Dopita (RSAA) & Bob Williams (STScI)
 
Objectives
We propose to make WiFeS observations of four old nova shells to investigate their kinematics and physical conditions and to derive the degree of mixing and the chemical abundances from the analysis of their recombination line spectra.
Background
Nova explosions result from the accretion of hydrogen-rich material derived from a binary companion onto the surface of a white dwarf star. When enough gas has been accreted, thermonuclear hydrogen burning ignites once more. However, if the accretion rate has been low enough, the accreted material is strongly electron-degenerate. In these circumstances, the increase in temperature produced by thermonuclear burning produces almost no change in the pressure, which is dominated by the degeneracy pressure of the electrons. This initiates a thermonuclear runaway until the thermal pressure of the gas breaks the degeneracy, by which time the gas reaches temperatures of up to 108K. The enormous and sudden release of thermonuclear energy causes the star to shine briefly  at a luminosity of more than the Eddington value for the electron scattering opacity, and the radiation pressure drives a shell of gas out at velocities of hundreds to thousands of kilometers per second. During the ejection phase, the energy input from beta-unstable nuclei such as 13N, 14O, 15O, and 17F is important in powering the luminosity and the outflow.
The nature of the explosion and the thermonuclear products produced depend on the mass of the white dwarf. If this mass is high (>1 solar mass, approximately), less mass can accumulate before thermonuclear runaway is initiated. As a consequence, the ejected mass is low (of order 10-5 solar mass), the velocity of ejection is high (several thousand km/s), and the variation in the emitted light is rapid (days). Such fast novae eject material rich in O, Ne, and Mg. Low-mass white dwarfs produce slow novae, with light curves decaying over months, lower ejection velocities (~1000 km/s, or even less), with relatively large ejected masses rich in the products of partial hydrogen burning.
The nebular optical spectra of the fast O, Ne, Mg novae are dominated by lines such as [Ne V] 3300, 3426Å, [Ne III] 3868,3876Å, [O III] 4959,5007Å, [O II] 7218,7328Å and recombination lines of HeII and of H. The slow novae have spectra much more like those of PNe, although the nitrogen lines are much stronger. The line profiles of both classes usually show the double-horn structure expected for an expanding shell, but there is often a lot of sub-structure caused by individual condensations. This proposal is to study these condensations at very late phases, when the shell covers an appreciable angular extent, suitable for observation with WiFeS, see fig 8.1.
In the visible, such old nova shells display a curious spectrum dominated by hydrogen and helium recombination lines, as well as permitted lines of various abundant heavy elements which also arise from recombination (Williams et al. 1978; Williams, 1982). Weak [O II] and [N II] lines are visible, but these probably arise in a separate component with more normal abundances. The relative absence of the forbidden lines arises because the electron temperature is far too low to excite the visible forbidden lines. The high chemical abundances cause recombination to occur at a very low temperature. As a result, the Balmer continuum decreases very rapidly above the Balmer limit. This has been used by Williams to measure the electron temperature; of order 500 K for DQ Her and 800 K for CP Pup. Because all of the emission lines are the result of recombination, the ratio of the heavy element lines to the Balmer lines can be used to obtain the abundances, once the temperature is known. This needs low-temperature recombination coefficients, which were calculated and tabulated by Smits (1991). Calculation of absolute abundances requires a time-dependent photo ionization modelling, so that ionization correction factors can be properly calculated.
 
Proposed Observations
The Wide Field Spectrograph (WiFeS) is particularly suited to study the nebular and late phases of Nova eruptions, when the ejected nebular shell becomes spatially resolved. During this nebular phase and the final decline, the emission lines can be used to derive the nebular abundances in the same way as is done with PNe.
We propose to observe four old nova shells, to be selected according to the time of year of the observations from the list below. Typical surface brightness in H-Beta are log(F) ~ 10-16 erg cm-2 s-1 arc sec-2 , so 2.5 hrs of observation on each object will provide a S/N in the bright lines of  ~30. This will allow the recombination lines of the heavy elements to be detected with adequate signal to noise. Figure 2 demonstrates the quality of data we would expect in a single slice with WiFeS, and shows how we can separate condensations not only in terms of their spatial coordinates but also in terms of their radial velocities.
Figure 8.1. Old Nova remnants, from Tim O’Brien’s web pages. Top left to bottom right these are: RR Pictoris 1925, HR Delphini 1967, GK Persei 1901, DQ Herculis 1934. The image sizes here range from 5 to 60 arc sec. across.
 
Figure 8.2. Long-slit spectra of FH Serpentis in the region of [N II] 6548,84Å and H-Alpha, showing the velocity ellipsoid associated with the expansion of the ejecta,  and emphasizing their knotty nature.
 
Table 8.1 Candidate Nova Targets
 
SNR    
R.A.  (J2000.0)
Dec  (J2000.0)
RR Pict    1925
06:35:36.10
- 62:38:23.7
V351 Pup1991
08:11:38.40
- 35:07:30.2
CP Pup    1942
08:11:46.01
 -35:21:06.1
T Pyx   (recurr.)    
09:04:41.53
- 32:22:47.2
V382 Vel 1999
10:44:48.40
- 52:25:31.1
GQ Mus  1983
11:52:02.40
- 67:12:21.6
FH Serp   1970
18:30:46.71
+02:36:50.7
QU Vul    1984
20:24:40.50
+27:40:48.0
HR Del    1967
20:42:20.20
+19:09:39.1
 
References
Smits, D. P. 1991, MNRAS, 248, 217
Williams et al. 1978, ApJ, 224, 171
Williams, 1982, ApJ, 257, 170