3 Results for 2 binary stars

 

3.1 Data reduction

 A malfunction in the camera control computer caused the loss of 80% to 90% of the short exposures. As a consequence, the speckle noise remains 3 times more intense than expected and dominates the photon noise.

The dark-speckle algorithm is described in Labeyrie (1995) and Boccaletti et al. (1998). The data reduction software provides two images. The first is equivalent to a long exposure, computed by co-adding all the short exposures containing the position of the photon-events. The second image, called the "cleaned map'', displays the result of the dark-speckle algorithm. The number of zero-photon events occurring in binned pixels containing $2\times 2$ camera pixels, is cumulated over successive frames. The cleaned map, thus obtained in negative form, can be inverted with a suitable gamma exponent. In principle, the cleaned image is more sensitive to the faint structures than the conventional long exposure.

Photon-counting data are affected by the so-called photon-centroiding (Tiébaut 1994) resulting from the limited resolution of the electronics in space and time. The counting behaviour is expected to improve with forthcoming detectors. The signal to noise ratio (SNR) is measured in the cleaned map according to Eq. (6) from Labeyrie (1995). The magnitude difference is derived from the CP20 long exposure by comparing the flux of the star outside the mask, and the flux of the companion when the star is under the mask.

To assess the performance of the system, one can compute the expected maximum brightness ratio theoretically reachable (Eq. (7) Boccaletti et al. 1998). This value only depends on the sampling (150 pixels/speckle area) and on the adaptive optics (AO) gain which is the brightness ratio between the peak of the corrected image and the halo level. The average AO gain obtained in the present run is about 15 at $0.5\hbox{$^{\prime\prime}$}$ from the star. The total integration time is assumed to be very large compared to the speckle lifetime. This leads to a 8.8 magnitude difference. However, as described in Sects. 3.2 and 3.3, this goal, although modest with respect to the $\Delta m=22$ goal attainable in principle with future adaptive optics, was not reached owing to practical problems. Among these was the loss of 90% of the observing data, due to a computer failure.

3.2 $\delta$ Per

 

  

Figure 4: a) Negative cleaned image of the binary star $\delta$ Per, obtained in H$\alpha$ ($\lambda=653.6\,{\rm nm}$, $\Delta\lambda=10\,{\rm nm}$) with the $0.45\hbox{$^{\prime\prime}$}$ mask. The intensity of the image represents the number of zero photon collected with 9503 short exposures of $20\,{\rm ms}$. The hypothetic companion (arrow) is visible at the edge of the mask (${\rm PA}=202\hbox{$^\circ$}$, $\rho=0.293\hbox{$^{\prime\prime}$}$) as a "hole'' in the cleaned map, amidst the fixed speckles generated by the spider structure. At this location, the zero photon-count is lower than elsewhere

 

Figure 4: b) The same image enhanced with an inverse square law, and smoothed to the speckle scale with a wavelet transform filter. The faint companion ($\Delta m=3.5$) emerges above the speckle noise, although no reference star was available in this case to subtract the fixed speckles

The first target star observed is $\delta$ Per (HD22928, m_v=2.99, ${\rm JD}=50725.646$), a binary system previously observed by Hipparcos (ESA 1997). We used the $0.45\hbox{$^{\prime\prime}$}$ occulting mask and $400\,\mu {\rm m}$ pupil stop. The turbulence conditions were characterized by r₀= 8 cm during observation. 9503 short exposures of $\delta$ Per have been acquired, corresponding to an observing time of 190s. We used an ${\rm H}\alpha$ filter ($653.6\,{\rm nm}$) of bandwidth $10\,{\rm nm}$. The faint companion appears at ${\rm PA}=(202\pm 1)\hbox{$^\circ$}$ and $\rho=(0.293\pm 0.027)\hbox{$^{\prime\prime}$}$. The sampling on the camera is 144 pixels per arcsecond. The image (Figs. 4a and 4b) presents some fixed patterns, mainly produced by the spider arm diffraction which is visible as four symmetrical bright speckles. An AO gain of 6, azimuthally averaged, is obtained at $0.3\hbox{$^{\prime\prime}$}$ from the star. An SNR of 53 is measured on the speckle size ($15\times 15$ pixels), while the model predicted an SNR of 133 (Boccaletti et al. 1998). The discrepancy with theory can be an indirect effect of the computer failure mentioned above; speckle noise becomes dominant owing to the small number of exposures. The intensity of the companion was derived from the CP20 long exposure and corresponds to a $\Delta m$ of $3.48\pm 0.20$.The second target star, $\eta$ Psc (HD 9270, m_v = 3.61, JD = 50730.507), was also selected from the Hipparcos binary star catalogue. Coronagraphic images of $\eta$ Psc were recorded without spectral filter, the bandpass being limited by the dichroic beam-splitter and the camera ($\Delta\lambda=650-850\,{\rm nm}$). These images illustrate the problem of the speckle noise in the search for faint companions. The Hipparcos data gives $\Delta m_v=3.14$, $\rho=0.33\hbox{$^{\prime\prime}$}$ and ${\rm PA}=221\hbox{$^\circ$}$, which is in good agreement with our results.

3.3 $\eta$ Psc

 The second target star, $\eta$ Psc (HD 9270, m_v=3.61, JD = 50730.507), was also selected from the Hipparcos binary star catalogue. Coronagraphic images of $\eta$ Psc were recorded without spectral filter, the bandpass being limited by the dichroic beam-splitter and the camera ($\Delta\lambda=650-850$ nm). These images illustrate the problem of the speckle noise in the search for faint companions. Indeed, for the same reason as above the number of short exposures is poor (9013). Altough the companion emerges easily above the photon noise, the residual speckle noise inhibits its detection on the raw data (Fig. 5a). However, a nearby reference star ($\eta$ And, m_v = 4.42), has been observed immediately afterwards for map subtraction (Fig. 5b). The efficiency of the subtraction is limited for several reasons:

(i)

The conditions of turbulence, and thus the correction applied by the AO system, are different for both stars ($r_0=8\,{\rm cm}$ for $\eta$ And, and $r_0=6.2\,{\rm cm}$ for $\eta$ Psc).

(ii)

The coronagraphic images are not invariant to translation (Malbet 1996), and in fact highly sensitive to tracking drift. Any centering difference between both sequences causes spurious speckle noise in the subtracted image.

(iii)

A 76 Hz vibration of the telescope drive motor, seen by the Shack-Hartmann wave sensor, is left uncorrected by the adaptive system.

To overcome these problems, both star and reference have been corrected from Flat-Field and smoothed with wavelet transform. Then, to account for the lower speckle contrast of $\eta$ Psc, presumably caused by unequal seeing lifetime during both observations, the reference star was convolved with a gaussian shape of 4 pixels width. Finally, $\eta$ And was shifted to overlap $\eta$ Psc in the field, and scaled in intensity (Fig. 5c). After subtraction of the two cleaned negative images, the remaining bright and dark speckles belong respectively to the reference star ($\eta$ And) and the target star ($\eta$ Psc).

The measurement of the SNR, on the cleaned map, is made difficult by the crowded field. Although the companion remains undetected on the initial image (Fig. 5a), in terms of speckle noise, the comparison of both images eliminates all features common to the target (Fig. 5a) and the reference (Fig. 5b), so that the only residual speckle indicates the companion position ($\rho=(0.507\pm 0.01)\hbox{$^{\prime\prime}$}$, ${\rm PA}=(94\pm 1)\hbox{$^\circ$}$). This process leads to an SNR of 137, instead of 250 as predicted by the model. In this case, the SNR is naturally decreased by the companion's smearing, induced by the Wynne correctors.

The long exposure allows to derive a $\Delta m=4.15\pm 0.20$ in the R band. The Hipparcos V-band data gives $\Delta m_v=3.7$, $\rho=0.64\hbox{$^{\prime\prime}$}$ and ${\rm PA}=47\hbox{$^\circ$}$. A difference of spectral types for both components can explain the apparent discrepancy. We must also notice a discrepancy in the companion position. As the orbital solution is unknown, further observing runs will be needed to confirm the detection of the companion.

  

Figure 5: a) Negative cleaned map of the binary star $\eta$ Psc, observed with the $0.45\hbox{$^{\prime\prime}$}$ mask and the Wynne corrector's spectral band ($\lambda_0 = 635 \,{\rm nm}$, $\Delta\lambda=650-850\,{\rm nm}$). Due to the strong turbulence ($r_0=6.2\,{\rm cm}$) and the small number of exposures (9013), the faint companion (arrow) remains buried among the speckles. A flat field correction has been applied to the map

 

Figure 5: b) Reference star ($\eta$ And) observed in the same way just before $\eta$ Psc with r0 = 8 cm, 3645 short exposures. The suspected "companion" peak of Fig. 5a is here absent, although the static residual speckles are similar. The scattered halo level is significantly lower than in Fig. 4a

 

Figure 5: c) Difference of $\eta$ Psc and $\eta$ And cleaned maps. The companion now appears as a dark feature (arrow). The image has been smoothed with a wavelet transform filter

 

Figure 5: d) Positive cleaned image obtained with an inverse square law, and a threshold at 3$\sigma$ level

 

Table 1: Summary of our results compared to the HIC data