To create a black hole a body must occupy a radius that is smaller than its Schwarzschild limit. If a body has a mass greater than 3.2 solar masses then the theoretical behaviour of the neutron is that the force acting upon the neutrons due to the pressure of the collective mass of the star would cause the neutron to collapse. Then the way that a neutron star would form a black hole is by having a companion star which looses matter to the neutron star. If the mass of the neutron star is close to the mass of 3.2 solar masses then the star will theoretically collapse into a black hole when enough matter is accreted to exceed the critical limit.
The caveat is that there is no force in the universe that can accelerate energy faster than the speed of light. When neutron collapses it’s surface must accelerate at a rate that exceeds the speed of light, so this in fact is not allowed. An alternate theory is that the matter does not collapse into a black hole nut rather exists at an energy density that is just above its Schwartzschild limit. In fact recent observations made by Ligo of colliding black holes and by the recent observations of Sagittarius A*'s G2 gas cloud collision in fact fits much more closely with this model than these object being in fact true black holes. When the G2 gas cloud interacted with Sagittarius A* it did not create a large output of energy, but rather there was a just small release of energy consistent with a black star that exists at an energy state just above its true event horizon. These objects do not emit light and so are virtually indistinguishable from black holes, except when they interact with other objects.
When LIGO first observed a gravitational wave there was a small gamma ray burst 0.04 seconds after the gravitational wave struck. There should be no Gamma Ray Bursts when black holes merge. There was a paper written in June of 2016 that used statistical models to suggest that the GRB did not occur. (Someone's job was at stake.) When you look at the method they used it virtually eliminates any possible observation of a GRB that would be associated with a dark star merger due to the weak GRB it would produce. Thus, no black hole merger could be observed in the future associated with a gravitational wave observation, unless it as a neutron star merger, which produces a much larger GRB. Recently LIGO announced just such an event where two neutron stars merged and this coincided with a GRB. When asked why this has not been observed before the officials at LIGO said it was because the merging of black holes do not produce GRBs.
The problem with creating black holes is that the matter at the event horizon, or simply the energy, must accelerate faster than the speed of light, which is not allowed by relativity. The comeback is that spacetime itself is collapsing faster than the speed of light. One must then ask how this could happen if the relativity does not allow super luminous speeds. To accelerate spacetime to super luminous speeds something must have dragged spacetime to this acceleration. But, that is not allowed. In fact what the observations are really showing is that it is most probable that black holes do not exist in nature. To back this up the search for extraterrestrial planets has found zero black holes orbiting other stars. Since 1968 only about 20 stellar mass black hole candidates have been found anywhere. If they exist where are they? Every observation made is consistent with black stars and only mathematical models are consistent with black holes.
WE must also consider that the present theory predicts that black holes will explode at the end of their lifetimes as they evaporate. No spikes of energy of this nature has been observed and yet there should some if the theory is correct. We must also consider that CERN should have produced black holes at the energy levels they have employed and no black holes have been created. This was not expected.
Black stars paper.
A better solution to the runaway endpoint problem
The statistical trick and logic used to eliminate future black star merger observations
The non-event of the Sagittarius A* G2 collision