We argue that the possibility of having infinite energy in the center-of-mass frame of colliding particles is a generic property of rotating black holes. We suggest a general model\char21{}independent derivation valid for dirty black holes. The earlier observations for the Kerr or Kerr-Newman metrics are confirmed and generalized.

Recently, in the series of works a new effect of acceleration of particles by black holes has been found. Under certain conditions, the energy in the center-of-mass system can become infinitely large. The essential ingredient of such effect is the rotation of a black hole. In this work, it has been argued that the similar effect exists for a nonrotating but charged black hole even for the simplest case of radial motion of particles in the Reissner-Nordström background. All main features of the effect under discussion due to rotating black holes have their counterpart for the nonrotating charged ones.

Black hole mimickers are possible alternatives to black holes; they would look observationally almost like black holes but would have no horizon. The properties in the near-horizon region where gravity is strong can be quite different for both types of objects, but at infinity it could be difficult to discern black holes from their mimickers. To disentangle this possible confusion, we examine the near-horizon properties, and their connection with far away asymptotic properties, of some candidates to black mimickers. We study spherically symmetric uncharged or charged but nonextremal objects, as well as spherically symmetric charged extremal objects. Within the uncharged or charged but nonextremal black hole mimickers, we study nonextremal $\epsilon$-wormholes on the threshold of the formation of an event horizon, of which a subclass are called black foils, and gravastars. Within the charged extremal black hole mimickers we study extremal $\epsilon$-wormholes on the threshold of the formation of an event horizon, quasi-black holes, and wormholes on the basis of quasi-black holes from Bonnor stars. We elucidate whether or not the objects belonging to these two classes remain regular in the near-horizon limit. The requirement of full regularity, i.e., finite curvature and absence of naked behavior, up to an arbitrary neighborhood of the gravitational radius of the object enables one to rule out potential mimickers in most of the cases. A list ranking the best black hole mimickers up to the worst, both nonextremal and extremal, is as follows: wormholes on the basis of extremal black holes or on the basis of quasi-black holes, quasi-black holes, wormholes on the basis of nonextremal black holes (black foils), and gravastars. Since in observational astrophysics it is difficult to find extremal configurations (the best mimickers in the ranking), whereas nonextremal configurations are really bad mimickers, the task of distinguishing black holes from their mimickers seems to be less difficult than one could think of it.

A new simple and general explanation of the effect of acceleration of particles by black holes to infinite energies in the center of mass frame is suggested. It is based on kinematics of particles moving near the horizon. This effect arises when particles of two kinds collide near the horizon. For massive particles, the first kind represents a particle with the generic energy and angular momentum (I call them ``usual''). Near the horizon, such a particle has a velocity almost equal to that of light in the frame that corotates with a black hole (the frame is static if a black hole is static). The second kind (called ``critical'') consists of particles with the velocity v<c near the horizon due to special relationship between the energy and angular momentum (or charge). As a result, the relative velocity approaches the speed of light c, and the Lorentz factor grows unbound. This explanation applies both to generic rotating black holes and charged ones (even for radial motion of particles). If one of the colliding particles is massless (photon), the critical particle is distinguished by the fact that its frequency is finite near the horizon. The existence (or absence) of the effect is determined depending on competition of two factors\char22{}gravitational blue shift for a photon propagating towards a black hole and the Doppler effect due to transformation from the locally nonrotating frame to a comoving one. Classification of all possible types of collisions is suggested depending on whether massive or massless particle is critical or usual.