The case of the Supergiant shell in IC2574

The supergiant shell in IC2574 was first seen in high resolution VLA H I observations (Walter & Brinks 1998a; see also Fig. 3, left panel, which is a scaled-down version of Fig. 1 (top right). The shell has a linear size of about 1000pc x 500pc ( $60''\times30''$ ) and is expanding at $\sim$ 25kms^-1. It is therefore an ideal target to study expansion models since despite its size it has not stalled yet (as most of the supergiant shells in the LMC have). The elliptical shape of the H I shell is indicated in Figs. 3 and 4. The kinematic age based on the observed size and expansion velocity is estimated at 14 Myr.

Deep narrow-band H $\alpha$ -imaging revealed that current star formation (SF) regions within IC2574 are predominantly situated on the rim of the H I shell (Fig. 3, right panel, greyscale). This suggests that we are witnessing triggered star formation on the rim due to the expansion of the H I-shell (see, e.g., Elmegreen 1994). Follow-up radio continuum observations showed that these starforming regions are the main source of the radio continuum emission (see the contours in Fig. 4 for a map of the $\lambda$ 6cm emission).

Various theories on the creation and formation of supergiant shells (SGSs) predict that the cavity within the shell should be filled with hot gas (see, e.g., Cox & Smith 1974, Weaver et al. 1977, Chu et al. 1995). A pointed ROSAT observation towards IC2574 (Walter et al. 1998) revealed that the supergiant shell is indeed filled with extended hot X-ray gas (see the contours in Fig.3). This makes the supergiant shell in IC2574 a truly unique region and suggests that we have caught this SGS at an auspicious moment. Assuming a Raymond-Smith (1977) plasma temperature of log( $T\rm [K])\,=\,6.8\pm0.3$ and an internal density of $(0.03\,\pm\,0.01)$ cm^-3 we derive an internal pressure of $P\approx 4 \times10^5$ Kcm^-3. This pressure is much higher than the pressure of the ambient warm ionized medium ( $P\approx 10^3-10^4$ Kcm^-3) suggesting that it is probably this hot gas which is still driving the expansion of the shell (see, e.g., Weaver et al. 1977).

**Figure 3:** Left: IC2574 in the 21cm line of neutral hydrogen (H I). Right: Blowup of the supergiant shell in IC2574. The ellipses plotted in both maps indicate the size of the expanding H I shell (linear size $\approx 1000\times 500$ pc). The greyscale is a representation of the H $\alpha$ emission coming from the rim of this shell. The contours represent the X-ray emission coming from the inside of the shell as observed with the ROSAT PSPC camera (for details see Walter et al. 1998). Coordinates are given for B1950.0.
$\begin{figure} \par\centerline{\psfig{figure=walter_ver2_fig3.eps,width=13cm}} \par\end{figure}$

**Figure 4:** The supergiant shell in IC2574, showing the same region as in Fig. 3 (right panel). The plotted ellipse again indicates the size of the expanding H I shell. The greyscale is a representation of our deep R-band image showing the central stellar association. The superimposed contours represent $\lambda$ 6cm radio continuum emission.
$\begin{figure} \par\centerline{\psfig{figure=walter_ver2_fig4.eps,width=9cm}} \par\end{figure}$

We have just been granted observing time during Cycle 1 with the Advanced X-ray Astrophysics Facility (AXAF) so we will soon be able to derive the spatial extent and the temperature of the X-ray gas to a much higher accuracy. The AXAF observations will also allow us to determine the contribution to the X-ray flux by point sources (e.g., X-ray binaries and supernovae). Note that the X-ray source is resolved in the ROSAT observations, indicating that at least a significant fraction of the X-ray emission is extended.

>From ground based R-band imaging, a giant stellar association is readily visible within the IC2574-SGS (see Fig. 4, greyscale). We speculate that this stellar association is in fact responsible for the formation and expansion of the shell as well as for the heating of the X-ray gas. Unfortunately, the evidence is still largely circumstantial.

The wealth of observations which is available for this supergiant shell suggests that this central stellar association is the powering source for the formation and expansion of the shell as well as for the heating of the X-ray gas. Based on our H I observations and using the models of Chevalier (1974), we derive that the energy required to produce the shell must be of order 10⁵³ ergs or the equivalent of about 100 Type II SNe. This would mean that the least massive stars that go off as SN are most probably still present in the central stellar association since their lifetimes ( $\sim$ 50 Myr) are somewhat longer then the dynamical age of the hole ( $\sim$ 14 Myr, as derived from the H I observations).