We study the X-ray spectra of tenuous, two-temperature accretion flows using a model involving an exact, Monte Carlo computation of the global Comptonization effect as well as general relativistic description of both the flow structure and radiative processes. In our previous work we found that in flows surrounding supermassive black holes, thermal synchrotron radiation is not capable of providing a sufficient seed photons flux to explain the X-ray spectral indices as well as the cut-off energies measured in several best-studied AGNs. In this work we complete the model by including seed photons provided by nonthermal synchrotron radiation and we find that it allows to reconcile the hot flow model with the AGN data. We take into account two possible sources of nonthermal electrons. First, we consider e+- produced by charged-pions decay, which should be always present in the innermost part of a two-temperature flow due to proton-proton interactions. We find that for a weak heating of thermal electrons (small delta) the synchrotron emission of pion-decay e+- is much stronger than the thermal synchrotron emission in the considered range of bolometric luminosities, L Ledd. The small-delta model including hadronic effects in general agrees with the AGN data, except for the case of a slowly rotating black hole and a thermal distribution of protons. For large-delta, the pion-decay e+- have a negligible effect and then in this model we consider nonthermal electrons produced by direct acceleration. We find an approximate agreement with the AGN data for the fraction of the heating power of electrons which is used for the nonthermal acceleration eta~0.1.

We study radiative properties of hot accretion flows in a general relativistic model with an exact treatment of global Comptonization, developed in our recent works. We note a strong dependence of electron temperature on the strength of magnetic field and we clarify that the underlying mechanism involves the change of the flow structure, with more strongly magnetised flows approaching the slab geometry more closely. We find that the model with thermal synchrotron radiation being the main source of seed photons agrees with the spectral index vs Eddington ratio relation observed in black hole transients below 1 per cent of the Eddington luminosity, LEdd, and models with a weak direct heating of electrons (small delta) are more consistent with observations. Models with large delta predict slightly too soft spectra, furthermore, they strongly overpredict electron temperatures at ~0.01 LEdd. The low-luminosity spectra, at <0.001 LEdd, deviate from a power-law shape in the soft X-ray range and we note that the first-scattering bump often resembles a thermal like component, with the temperature of a few hundred eV, superimposed on a power-law spectrum. The model with thermal Comptonization of thermal synchrotron radiation does not agree with well studied AGNs observed below ~0.01 LEdd, for which there is a substantial evidence for the lack of an inner cold disc. This indicates that the model of hot flows powering AGNs should be revised, possibly by taking into account an additional (but internal to the flow) source of seed photons.

Radiatively inefficient, hot accretion flows are widely considered as a relevant accretion mode in low-luminosity AGNs. We study spectral formation in such flows using a refined model with a fully general relativistic description of both the radiative (leptonic and hadronic) and hydrodynamic processes, as well as with an exact treatment of global Comptonization. We find that the X-ray spectral index--Eddington ratio anticorrelation as well as the cut-off energy measured in the best-studied objects favor accretion flows with rather strong magnetic field and with a weak direct heating of electrons. Furthermore, they require a much stronger source of seed photons than considered in previous studies. The nonthermal synchrotron radiation of relativistic electrons seems to be the most likely process capable of providing a sufficient flux of seed photons. Hadronic processes, which should occur due to basic properties of hot flows, provide an attractive explanation for the origin of such electrons.

We present a model of gamma-ray emission through neutral pion production and decay in two-temperature accretion flows around supermassive black holes. We refine previous studies of such a hadronic gamma-ray emission by taking into account (1) relativistic effects in the photon transfer and (2) absorption of gamma-ray photons in the radiation field of the flow. We use a fully general relativistic description of both the radiative and hydrodynamic processes, which allows us to study the dependence on the black hole spin. The spin value strongly affects the gamma-ray emissivity within ~10 gravitational radii. The central regions of flows with the total luminosities L < 0.001 of the Eddington luminosity are mostly transparent to photons with energies below 10 GeV, permitting investigation of the effects of space-time metric. For such L, an observational upper limit on the gamma-ray (0.1-10 GeV) to X-ray (2-10 keV) luminosity ratio of L_{0.1-10 GeV}/L_{2-10 keV} << 0.1 can rule out rapid rotation of the black hole; on the other hand, a measurement of L_{0.1-10 GeV}/L_{2-10 keV} ~ 0.1 cannot be regarded as the evidence of rapid rotation, as such a ratio can also result from a flat radial profile of gamma-ray emissivity (which would occur for nonthermal acceleration of protons in the whole body of the flow). At L > 0.01 of the Eddington luminosity, the gamma-ray emission from the innermost region is strongly absorbed and the observed gamma-rays do not carry information on the value of the black hole spin. We note that if the X-ray emission observed in Centaurus A comes from an accretion flow, the hadronic gamma-ray emission from the flow should contribute significantly to the MeV/GeV emission observed from the core of this object, unless it contains a slowly rotating black hole and protons in the flow are thermal.