Документ взят из кэша поисковой машины. Адрес оригинального документа : http://jet.sao.ru/hq/lizm/conferences/pdf/1987/1987_p27-p30.pdf Дата изменения: Wed Feb 24 18:49:34 2010 Дата индексирования: Tue Oct 2 09:02:06 2012 Кодировка: Поисковые слова: п п п п п п п п п п п п п

TH E ESTIMATIO N O F 5 3 Ca m MAGNETI C FIEL D VALU E BY MEAN S OF SPECTRA L LINE S WIDT H ANALYSI S

Yu. V. Glagolevskij, V. G. EL'kin, I. I. Romanyuk Special Astrophysical Observatory, Stavropolskij Kraj , 357147, Nizhnij Arkhyz, USSR N. E. Piskunov Astronomical Council of the USSR Academy of Sciences, Pyatnitskaya, 48 , 109017, Moscow, USSR

The systematical studies of 53 Cam spectral lines widening were carried out in order to find the possible structural features and the mean value of the surface magnetic field. The spectral region 4000 - 4700 A was registered on photoplates on 6-m telescope with the inverse dispersion 1.3 and 1.7 A/mm. Nearly 100 unblended lines were selected for analysis and for each of thee we have measured the total width on 10, 30 , 50 , and 70 % of central depth (Fig. la) . Unsecure measurements were not included in subsequent study. For every level we have plotted the dependence between width and Lande factor and obtained the linear regression parameters: the inclination, the zero Lande factor line width and the correlation. The results are listed in Table 1.

We know from the theory of magnetic intensification of tral lines that the line widths -ust be proportional surface magnetic field B , but our results have shown dependence from the level of width measurements (see Table Fig. 2) .
s

specto the strong 1 and

To control the was repeated for Line profiles were correlation between 32

reliability of the method the whole procedure the spectrophotometry standard star Procyon. taken from Griffin (1979). No significant Lande factor and line width has been found.


The systematical increase of the surface field value with the level of width measurements is quite unbelievable, so we tried to find some spectroscopical effect like a fine structure, which was not yet accounted. The importance to account for the fine structure was proved by numeric experiments. Fig. 1. presents a model components, the intensity of which decreases (b) line with 10 and increases (c) to the center of spectral line (the field is 12 k6s). Zeeman effect is working in a different way for different splitting patterns. So for the correct use of line intensification theory for every depth level we need to use a special "effective" Lande factor.

To account for the fine structure we have divided the full list of lines on 4 parts corresponding to the Zeeman components pattern: 33


1) pure triplets; 2) the intensity of 3) the intensity of 4) all other cases.

components increases to the line center; components decreases to the line center;

For every group w carried out the independent regression analysis and numerically calculated line width on different levels for some typical Zeeman patterns using effective Lande factor for each level. The effective Lande factor for 70 % level was calculated as intensity weighted mean of all - components which are stronger than 70X of maximal intensity. The comparison between calculations and regression parameters (Fig. 3) provided the BS value for a given rotation phase.

34


RESULTS 1. Our results show strong dependencies of line width on the level of width measurements. 2. The magnetic field of 53 Cam in rotation phase 0.75 is properly described by magnetic dipole model with B =11.96 0.56 kGs. Deviations froe dipole are smaller than the observational errors. The proposed method proved to be usefull for statistical surface field determination for magnetic stars;
S

3. Using direct splitting measurements Preston (1969) found B = 15 kGs which is quite close to our estimation 12.7 kGs for the same phase. So the same dipole model is proved by different type of observations;
S

4. Better Lande factor calibration procedure is required for this method because for some Zeeman patterns the agreement was rather poor. An influence of possible toroidal component is not also excluded, but it is the problem of future investigations.

REFERENCES 1.Griffin, R. and Griffin, R.: 1979, A Photometric Atlas of the Spectrum of Procyon, R. and R. Griffin, Cambridge, England. 2.Preston, G. W., 1969, Astrophys. J., v. 157, 247. 35