A Model for the X-Ray and UV Emission from Seyfert Galaxies

We propose that the X--ray emission from radio quiet AGN and galactic black holes is due to Comptonization of soft thermal photons emitted by the underlying accretion disk in localized structures (blobs). The power per unit area produced by the blobs, impi

a r X i v :a s t r o -p h /9405059v 1 26 M a y 1994A MODEL FOR THE X-RAY AND UV EMISSION FROM SEYFERT GALAXIES

AND GALACTIC BLACK HOLES

Francesco Haardt 1,2,Laura Maraschi 3and Gabriele Ghisellini 4

1:ISAS/SISSA,Via Beirut 2–4,34014Trieste,Italy,E-mail:38028::haardt

2:present address:STScI,3700San Martin Dr.,Baltimore,MD 21218,E-mail:6559::haardt

We propose that the X--ray emission from radio quiet AGN and galactic black holes is due to Comptonization of soft thermal photons emitted by the underlying accretion disk in localized structures (blobs). The power per unit area produced by the blobs, impi

1.INTRODUCTION

The recent observations by OSSE show spectral“breaks”or cut–offs in the hard X–ray spectra of some Seyfert1galaxies(e.g.Maisack et al.1993,Cameron et al.1993),sup-porting thermal or quasi–thermal models for the X–ray emission,in which a population of semi–relativistic electrons Comptonizes the available soft photons(e.g.Shapiro,Lightman &Eardly1976,Podzniakov,Sobol&Sunyaev1980,Sunayev&Titarchuk1980),and rein-forcing the analogy with the high energy spectra of galactic black hole candidates(GBH) (e.g.Done et al.1990,Johnson et al.1993,and Grebenev et al.1993).

A second implication of the OSSE observations is that the luminosity L X in the medium and hard X–ray range can be reliably estimated.If the“X–ray bolometric correc-tion”established for the few galaxies detected by OSSE is adopted as a general property, the X–ray to UVflux ratios measured with ROSAT for a large sample of objects(Walter &Fink1993)leads to an estimate of the X–ray luminosity in many cases definitely smaller than that at the UV bump,L UV.On the other hand in the best studied objects,the Seyfert galaxies NGC4151and NGC5548,the UV and X–rayfluxes are comparable,and vary in a correlated fashion on time-scales of weeks(Perola et al.1989,Clavel et al.1992, hereinafter C92).This led to the suggestion that in these objects the UV emission is due to reprocessing of the higher frequency radiation,implying L X>∼L UV.

We have proposed(Haardt&Maraschi1991,Haardt&Maraschi1993,hereinafter Papers I and II respectively)that the main features of the high energy emission from radio quiet active galactic nuclei(AGN),and GBH(Haardt et al.1993,Ueda,Ebisawa&Done 1994),can be explained by the interplay of a hot active corona with a colder accretion flow.Soft thermal photons with an energy of few tens(hundreds for GBH)eV are Comp-tonized by mildly–relativistic electrons in the hot corona,leading to the formation of a power–law spectrum with a high energy cut–off.In steady state,the equilibrium temper-ature of the electron distribution(assumed to be Maxwellian)can be computed balancing heating and radiative cooling,and depends only on the electron scattering optical depthτ. Furthermore,when the compactnessℓof the source(proportional to the luminosity to size ratio)is large,electron–positron pair production yields a lower limit toτand an upper limit to the temperature kT.Quite remarkably,for10<∼ℓ<∼100the theoretical values 300>∼kT>∼50keV are close to thefirst results of the OSSE experiment(e.g.Maisack et al.1993,Cameron et al.1993).

The average observed X–ray spectrum can be reproduced if the Comptonized–to–soft luminosity ratio L C/L S is≃2.In fact this value leads to a Compton y parameter close to1,and to a spectral index of the Comptonized spectrumαx≃1.In Paper I and II the condition L C/L S≃2was achieved by assuming that the entire gravitational power is released in the hot corona.Therefore the soft emission L S derived only from absorption and reprocessing of the high energyflux impinging on the disk.In that model,roughly half of the Comptonized photons is absorbed in the cold disk and reemitted as thermal radiation, while half is radiated away.Any further local dissipation within the disk would produce additional soft photons,lowering the coronal temperature and hence giving rise to steeper X–ray spectra.The model is then tightly constrained,and predicts nearly equal UV and X–ray luminosities.As was illustrated in Paper II,the different angular distributions of the Comptonized photons with respect to the thermal ones can give rise to a UVflux larger

2

We propose that the X--ray emission from radio quiet AGN and galactic black holes is due to Comptonization of soft thermal photons emitted by the underlying accretion disk in localized structures (blobs). The power per unit area produced by the blobs, impi

than the X-rayflux for viewing angles close to face on,but it seems difficult to achieve soft–to–hard ratios greater than5.Another prediction is a tight correlation between the hard Comptonized photons(X–rays)and the soft photons emitted by the cold disk(UV or soft X–rays for GBH).For NGC4151and NGC5548,the UV and X–rayfluxes simultaneously observed in different periods are indeed correlated,but variations(up to30%,Nandra et al.1990)in X–rays on timescales of hours are not accompanied by similar variations in the UV.Furthermore,for large UVfluxes,the correlation between X–rays and UV breaks down(Perola et al.1986,C92).

In Papers I and II we considered a uniform,plane parallel model which allowed us to minimize the parameters of the problem.Here we wish to relax these assumptions and adopt a more physical description of the energy dissipation process in order to account for a broader set of observational results,including X–ray and UV variability.

We maintain the assumption that the disk magneticfield can drain a fract …… 此处隐藏：19363字，全部文档内容请下载后查看。喜欢就下载吧 ……