7 The FIR emission  

NGC 128 is associated in both the IRAS Point Source Catalogue (IPSC) and Faint Source Catalogue (FSC) to the IRAS source 00266+0235. At 12 and 25 $\mu$m the IRAS catalog gives only upper limits, while high quality fluxes are available at 60 and 100 $\mu$m respectively in the FSC and IPSC (see Table 1). The total FIR luminosity, following Helou et al. (1985) and adopting our heliocentric recession velocity and Hubble constant, turns out of $4.2\ 10^9~L_{\hbox{$\odot$}}$.

Knapp et al. (1989) derived the fluxes at 60 and 100 $\mu$m for NGC 128 by averaging the IRAS data at the galaxy position. Using a similar technique, Bally & Thronson (1989) obtained flux values which are slightly different. Their results are in agreement, within the errors, with the PSC values reported in Table 1.

The 60 and 100 $\mu$m IRAS data were processed using the method described by Assendorp et al. (1995). After the standard co-addition of the images, we used the Maximum Entropy (ME) method described by Bontekoe et al. (1994). The resulting 60 and 100 $\mu$m high-resolution (about 1') infrared maps are overlaid onto a deep V-band image of NGC 128, taken with CAFOS at the 2.2 m telescope of Calar Alto (Figs. 14 and 15).

  

Figure 14: The group of NGC 128 with superposed the contours of the emission at 60 $\mu$m detected by IRAS. North is up and East to the left

The peak of the FIR radiation is situated in the area where the galaxy undergoes interaction with NGC 127. The shift between the 60 and 100 $\mu$m peak intensity position is 42'' in the IRAS cross scan (SW) direction and therefore not significant compared to the 1' resolution achieved with the ME method. Taking into account this uncertainty, we cannot associate the FIR emission to NGC 128 instead of NGC 127. Both these galaxies may contribute to the FIR emission.

On the other hand, the shift in the NW scan direction is much smaller ($\sim$5'') implying that the FIR radiation has its peak in the North semiaxes, so that it is probably connected to the interaction of the two galaxies.

We note in passing that the companion galaxy NGC 125 is not seen at 100 $\mu$m, and that the ring-like structure visible at 60 $\mu$m is approximately in the same position of the optical ring detected by BC77. The presence of such peculiar feature leads to suspect that NGC 128 and NGC 125 possibly underwent an interaction in the past.

  

Figure 15: The group of NGC 128 with superposed the contours of the emission at 100 $\mu$m detected by IRAS. North is up and East to the left

The FIR thermal emission can be used to measure the dust mass. The resulting dust mass depends on the physical-chemical properties of the grains (i.e. grain radius, density and emissivity) and on the adopted dust temperature.

In principle we can derive the dust temperature ($T_{\rm d}$) by fitting the FIR data with a grey-body characterized by a spectral trend $\lambda^{-\alpha}$. Since only the 60 and 100 $\mu$m fluxes are available, we used the relation obtained by Henning et al. (1990) to compute three dust color temperatures, obtaining respectively $T_{\rm bb}=39$ K ($\alpha=0$ black-body), $T_{\rm gb}=33$ K($\alpha=1$ grey-body), $T_{\rm gb}=28$ K($\alpha=2$ grey-body). The values chosen for $\alpha$ are suggested by the current dust models (Mathis & Whiffen 1989; Wright 1989; Désert et al. 1990; Draine & Malhotra 1993).

Following Hildebrand (1983) we computed the dust mass adopting a spectral index $\alpha=1$ (except with the formula of Thuan & Sauvage 1992 where we used $\alpha=1.5$). The derived masses are listed in Table 4. Notice that, considering the flux uncertainty of the IRAS data ($\sim$10%), our values are in good agreement with those available in literature. The differences with Bally & Thronson (1989) and Roberts et al. (1991) are due both to differences in the adopted fluxes and in the dust temperature evaluations.

  

Table 4: Dust masses with a single temperature model

The dust masses in Table 4 are derived using a single temperature model. This is a rough approximation for the condition found in the galactic environment. The dust is in fact heated by the radiation field, which in turn depends on the sources of luminosity and their spatial distribution in the galaxy. The total FIR emission is likely due to the contribution of dust at different temperatures. Moreover, the IRAS FIR measurements are not adequate to detect the emission coming from cold dust ($10\div20$ K) which peaks at a wavelength between 200 and 300 $\mu$m.

We thus estimate the total dust mass by introducing a dust temperature distribution depending on two free parameters which affect the shape of the function (see Kwan & Xie 1992 and Merluzzi 1998 for details).

We choose a range of temperatures that contribute to the IRAS emission between 7 and 60 K and we select the proper values of the free parameters which reproduce the ratio of the flux densities at 60 and 100 $\mu$m and takes into account the value of the FIR color temperature. The computed dust masses for a family of temperature distributions which satisfy the previous constraints are comparable within the flux and dust parameters uncertainties.

With $\alpha=1$ and the fluxes given in Table 1, we estimate the dust mass associated to the galaxies NCG 128 and NGC 127, to be in the range $6\ 10^6\ M_{\hbox{$\odot$}} \div 1\ 10^7\ M_{\hbox{$\odot$}}$. This value is larger than those obtained by the single temperature models, because the temperature distribution accounts for the contribution of the colder dust.

It has to be noticed that, since our constraint on the peak temperature is derived from the 60-100 $\mu$m data, the computed mass may be biased by the presence of the warm dust even if the temperature distribution estimates the different contributions. In particular, this happens if a significant fraction of cold dust is present in the source. In this case, we can consider our dust mass evaluation a lower limit for the dust content of NGC 128 and NGC 127. On the other hand, the NIR dust mass belongs entirely to NGC 128, and must be considered an upper limit, since it was derived using the R_V galactic value, i.e. the absorption of a spiral galaxy for the early-type NGC 128.

Taking into account the high uncertainty of the NIR dust mass ($\sim$40%), we thus suggest to adopt the dust mass of $6-10\ 10^6$ $M_{\hbox{$\odot$}}$ for the pair of galaxies NGC 128 and NGC 127, and a dust mass lower limit of $\sim$$6\ 10^6~M_{\hbox{$\odot$}}$ for NGC 128.