Open Collections

Application of the HF precision-velocity technique to the study of [delta] Scuti variables

University of British Columbia

In conventional radial-velocity techniques, wavelength calibration in stellar spectra is limited by collimation and guiding errors. These errors can be largely eliminated by imposing absorption lines of known wavelengths on the spectra. The most suitable natural absorption lines belong to the R-branch (3-0) vibration-rotation band of HF. The calibration is achieved by placing an absorption cell filled with HF gas into the stellar beam. The monel cell and the sapphire windows are maintained at 100°C in order to avoid polymerisation of the HF molecules. Connected to the cell is a reservoir of liquid HF maintained at 0°C and this would then maintain a constant pressure inside the cell. Meanwhile, the Reticon detector is utilised to provide the necessary high s/n spectra. In reducing the stellar spectra with the imposed HF lines, the effect of line blending between the two sets of lines can be minimised by performing numerical line cancellations using standard stellar or HF spectra. The relative positions of spectral lines are measured by means of the Fahlman-Glaspey difference technique. A modified Fahlman-Glaspey difference function can also be used to minimise the effect of small offset and scale differences between the line profiles. Simulation studies with artificial spectra have confirmed that the accuracy of line-position measurement would increase linearly with both s/n of the spectra and depth of the line profiles. Furthermore, the accuracy was found to decrease with increasing linewidths when the equivalent widths were held constant. A method has been devised to measure the HF gas temperature directly from the observed HF lines. This involves calculating the ratios between the observed HF line strengths. Theoretical line strengths for the (3-0) HF lines have also been calculated in order to study their dependence on the gas temperature. These theoretical values can also be used with the observed HF line strengths to calculate the relative gas pressure. A cutoff-free ATC theory on collisional line broadening was used to calculate HF linewidths and line shifts. Good agreement has been achieved between the calculated and published (1-0), (2-0), and (3-0) linewidths. This also enables a study on the temperature dependence of the (3-0) HF collisional self-broadened linewidths. Poor agreement was obtained between the published and calculated line shifts. Nevertheless, pressure-shift corrected wavelengths of the reference (3-0) lines have been derived. The δ Scuti stars are distinguished by pulsation periods of less than 0.3[sup d] and spectral type A or F. The small-amplitude δ Scuti stars typically have a light amplitude Δm[sub v] of 0.05[sup m] and a velocity 2K amplitude of less than 10kms⁻¹. Since the 2K/Δm[sub v] value is about 92kms⁻¹mag⁻¹ for most δ Scuti stars, a velocity precision of just ±0.1kms⁻¹ would already be equivalent to a precision of about ±0.001[sup m] in the light curve. Hence one can improve the study of δScuti pulsations with the use of precision radial-velocity techniques. The HF technique would also enable the study of individual spectral lines for profile variations. The star 20 CVn is a 0.122[sup d] δ Del type δ Scuti variable observed using the HF technique at CFHT. A 2K value of 1.4kms⁻¹ was obtained for the Ca II λ8662 line while a value of 1.4kms⁻¹ was obtained for the other lines. These give a 2K/Δm[sub v] of about 40kms⁻¹mag⁻¹ for 20 CVn which could imply nonradial pulsation. The star ρ Pup is a 0.141 δ Del type δ Scuti variable observed at CFHT. The line intensities of the stellar lines have been found to vary in phase with the light curve. These variations are at levels between 0.5% and 1% of the continuum. The lines are strongest near maximum light and weakest near minimum light. The variations can be considered as spectral-type variations caused by the variations of the effective temperature of the star over the pulsation cycle. The star o¹ Eri is a 0.082[sup d] broad-line δ Scuti variable observed at CFHT. A 2K amplitude of 4.3kms⁻¹ was measured for the Ca II λ8662 and H I λ8750 lines. Line-profile variations which are at a level of about 2% of the continuum, have been observed in the Ca II line. The variations can be characterised as the temporal movement of "features" across the broadened line profile. This would suggest the existence of nonradial pulsation in the star. The star β¹ Cas is a 0.104 broad-line δ Scuti variable observed at CFHT. The Ca II λ8662 line has a 2K value of about 7kms⁻¹while the value for the H I λ8750 and the other metallic lines is about 4.5kms⁻¹. The H I velocity curve was found to lag the other velocity curves by about 2% in phase. Line-profile variations which are at a level of 1% of the continuum, have been observed in the Ca II line. The variations can be characterised as linewidth variations. The broad-line phase coincides with the velocity minimum while the narrow-line phase coincides with the velocity maximum.

Text

2010-08-09

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/27223

Geophysics

University of British Columbia

Graduate