All Figures

$\begin{figure} \par\includegraphics[width=7.7cm,clip]{0569figa.eps}\vspace*{1cm} \par\includegraphics[width=7.7cm,clip]{0569figb.eps} \end{figure}$ Figure 1: The representative 34.6 ${\mu }$ m absorption line profiles for the 5 detection objects. Panel a) shows the best spectra among the SWS 01, 02 or 06 ones of each object; panel b) shows the best SWS 07 spectrum of each object. The line centers or the expected line centers of the 34.6 ${\mu }$ m doublet in panel b) are marked by short vertical lines. The spectra in both panels have been shifted arbitrarily for the sake of convenience. The TDT identification of the spectra shown in the figure are as follow: IRAS 07209-2540 (73402218, 73601963); IRAS 17424-2859 (09401504, 69602001); IRAS 17441-2822 (28702002, 46001217); IRAS 19244+1115 (36401631, 36401613); NML Cyg (52200719, 52200201).
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig2.eps} \end{figure}$ Figure 2: The 34.6 ${\mu }$ m absorption line depths measured in the spectra versus the expected values for these spectra (both divided by continuum flux $F^{\rm c}_{34.6}$ ). The location of the spectra of IRAS 17424-2859, IRAS 17441-2822 and IRAS 06053-0622 are indicated explicitly. The adopted classification of the spectra (a, t) is explained in Sect. 3.1.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig3.eps} \end{figure}$ Figure 3: The expected 34.6 ${\mu }$ m absorption line depth estimated using OH 1612 MHz maser blue peak flux shown against the 3 $\sigma$ noise level (both divided by continuum flux $F^{\rm c}_{34.6}$ ) for all the sample spectra. The solid and dotted lines mark where the expected 34.6 ${\mu }$ m absorption line depth equals to the 3 $\sigma$ noise and tenth of the 3 $\sigma$ noise, respectively. The location of the "a'' and "t'' spectra of IRAS 17424-2859, IRAS 17441-2822 and IRAS 06053-0622 are indicated explicitly. The adopted classification of the spectra (a, t, u, n) is explained in Sect. 3.1.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig4.eps} \end{figure}$ Figure 4: Counts of ISO SWS spectra of different resolutions for galactic OH-IR objects. The two lowest resolutions are combined into a single column "400 & 500''. The "at'' and "un'' type spectra are counted separately.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig5.eps} \end{figure}$ Figure 5: Counts of ISO SWS spectra of different noise levels for galactic OH-IR objects. The negative flux level of the leftmost column is due to wrong dark subtraction from the spectra, the effect not corrected during our analysis. The "at'' and "un'' type spectra are countedseparately.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig6.eps} \end{figure}$ Figure 6: Counts of sources in 5 intervals of the mean continuum flux around 34.6 ${\mu }$ m. The leftmost bin corresponds to the sources which have only spectra with wrong dark current subtraction.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig7.eps} \end{figure}$ Figure 7: The IRAS two color diagram for all our sample OH-IR objects with good quality fluxes at 12, 25 and 60 ${\mu }$ m. The boxes marked on the plot are after van der Veen & Habing (1988).
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig2.eps} \end{figure}$	Figure 2: The 34.6 ${\mu }$ m absorption line depths measured in the spectra versus the expected values for these spectra (both divided by continuum flux $F^{\rm c}_{34.6}$ ). The location of the spectra of IRAS 17424-2859, IRAS 17441-2822 and IRAS 06053-0622 are indicated explicitly. The adopted classification of the spectra (a, t) is explained in Sect. 3.1.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig4.eps} \end{figure}$	Figure 4: Counts of ISO SWS spectra of different resolutions for galactic OH-IR objects. The two lowest resolutions are combined into a single column "400 & 500''. The "at'' and "un'' type spectra are counted separately.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig5.eps} \end{figure}$	Figure 5: Counts of ISO SWS spectra of different noise levels for galactic OH-IR objects. The negative flux level of the leftmost column is due to wrong dark subtraction from the spectra, the effect not corrected during our analysis. The "at'' and "un'' type spectra are countedseparately.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig6.eps} \end{figure}$	Figure 6: Counts of sources in 5 intervals of the mean continuum flux around 34.6 ${\mu }$ m. The leftmost bin corresponds to the sources which have only spectra with wrong dark current subtraction.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=7.8cm,clip]{0569fig7.eps} \end{figure}$	Figure 7: The IRAS two color diagram for all our sample OH-IR objects with good quality fluxes at 12, 25 and 60 ${\mu }$ m. The boxes marked on the plot are after van der Veen & Habing (1988).
Open with DEXTER