A Contribution from Richard White
Note: The following narration was produced by Richard White in response
to a request from the British Broadcasting Company. They were preparing a
larger retrospective featuring Stephen Hawking. The work White discusses
is connected with the first computer calculation ever done that treated the
phenomenon now known as "Black Holes." As such, it is an important
artifact of computational history, and certainly worthy of consideration
independently of other aspects in the developments within Relativistic
- From the early 1950s through the 1960s, the U.S. Atomic Energy Commission
laboratories, especially those at Livermore and Los Alamos, were the world's
leading centers for scientific computing. We had a near monopoly on the largest
and fastest computers. In connection with weapons development, scientists at these
labs invented techniques for using computers to solve the complex problems of
weapons physics. We had computer programs that calculated the physics of mass
and energy transport as well as thermonuclear reactions that drive both nuclear
physics and stars. There were then few centers in the entire world that had these
- In 1961, Stirling Colgate proposed to me that we use the Lawrence Livermore
Lab's computers to study the propagation of shock waves through the atmosphere
of an exploding supernova. Colgate believed that the shock from the explosion
would drive particles of matter in the tenuous envelope of gas surrounding the star
to velocities very near that of light. He proposed that this was a primary source of
extra-solar cosmic rays.
- We set out to model the explosion of a supernova. I adapted techniques originally
applied to weapons calculations, adding the effects of gravity, as well as the exotic
material properties expected in the dense core of the pre-supernova star.
Conventional Wisdom (an oxymoron)
- At this time, the standard model for supernovae was one proposed by Geoff and
Margaret Burbidge, William Fowler, and Fred Hoyle. They showed that at the
endpoint of billions of years of evolution, any star of mass several times that of the
sun, forms a central core made primarily of iron. Though it may contain even more
mass than our sun, this iron core is compacted into a sphere slightly smaller than
the earth, giving it a density of around 10 million grams per cubic centimeter.
- As the pre-supernova star radiates energy into space, the iron core contracts,
becomes denser and hotter until a critical point is reached where the iron atoms
bang into one another with such great energy that they split apart. Nuclear physics
predicts they will split into helium atoms. The splitting of the iron into heliums
sops up energy that previously provided pressure to balance the gravitational
forces that pull the matter toward the center of the star. Without this pressure, the
core dramatically collapses under the force of its own gravity.
- According to Fowler and Hoyle, the collapse was supposed to stop, at least
temporarily, when the compression of the helium core drove the pressure up
enough to again balance the gravitational force. The collapse and abrupt stop
would perturb the oxygen envelope. Thermonuclear reactions in the oxygen would
then cause the envelope to explode, making a supernova.
- The explosion of the oxygen envelope was crucial. It was thought that through this
explosion, the star would throw off mass and leave behind a core of mass less than
or equal to about one sun. It was important to have this low mass because
Chandrahsekhar had calculated, using Einstein's theory of gravitation, that larger
mass cores would collapse in their own gravitational field, apparently to infinite
density. This was thought to be untenable by most physicists.
- Fowler and Hoyle developed their model without benefit of computer calculations.
Their arguments were based upon sound physical principles and calculations up to
the point where the iron atoms split into heliums. But they relied on some plausible
guesses, after the collapse began, because the details of the collapse are far too
complex to calculate without the aid of computers.
First Computer Calculations
- We were surprised when the results of our computer calculations did not follow the
script written by Fowler and Hoyle. Following the break up of the iron atoms into
helium, the core continued to collapse. The pressure in the helium never became
large enough to stop the collapse. Furthermore, when we simulated the
thermonuclear explosion of the oxygen envelope, the explosion was directed
primarily inward; that is, it was an implosion not an explosion.
- The collapse of the core continued until it reached densities where nuclear physics
indicated the helium atoms would now break up into free electrons and protons.
We revised our computer calculations to include this break up into electrons and
protons, thinking that the pressure developed as this electron-proton gas
compressed would stop the collapse. It didn't. The core densities became so large
that nuclear physics told us that the electrons would be crowded and pressed into
the protons, to make neutrons.
- The collapse would not be stopped until the core, now made of neutrons, reached
nuclear densities, about 1015 g/cc. It had collapsed from an earth sized chunk
of iron to a ball of neutrons, a mere 10 km. in diameter. All of this occurred in a few
tenths of a second. So long as we used Newton's gravitation, the core collapse
would be stopped by the very high pressures of the dense neutron matter.
However, matter from further out in the star continued to rain down upon the
collapsed core, adding to the mass that had initially collapsed so the neutron core
might contain several solar masses.
The Neutrino Prediction
- An exciting prediction came from all this: The combining of an electron and a
proton to form a neutron is accompanied by the emission of a neutrino. Therefore,
the dramatic collapse and transformation of the iron core into a highly-condensed
neutron core releases about 1057 (a billion trillion trillion trillion trillion)
neutrinos. We predicted that supernovae would be accompanied by an enormous release of
neutrinos into space. Our computer calculations also predicted that some of the
neutrinos in the core would be deposited in the envelope and that they played an
important role in the explosion of the stellar envelope.
Switching to Einstein's Physics
- Overly massive neutron cores presented a problem. If the mass were too large, e.g.:
several solar masses, then the stronger gravitation predicted by Einstein's general
relativity, might cause the core to collapse further. The core remnants of some of
our supernovae were massive enough and condensed enough that the differences
between Einstein's Relativistic Gravitation and Newton's simpler inverse-square
law were significant. We needed to include relativity if our calculations were to be
- In the summer of 1963, we organized an institute at the Lawrence Livermore
Laboratory. John Wheeler came from Princeton and brought a number of his
students: Charles Mizner, Kip Thorne, David Sharp, Jim Bardeen, Richard
Lindquist, and others, all of whom have very solid reputations in general relativity.
>From this institute came a number of important papers and the beginnings of an
understanding of how to convert our computer calculations from Newton's physics
to the physics of Einstein.
- Colgate left Livermore to become president of the New Mexico Institute of Mining
and Technology. I began a collaboration with Michael May that led us to the first
computer calculations of what we called, "Continued Gravitational Collapse." This
is the same phenomenon that the genius of John Wheeler later named "black
holes." I think if John had not come up with this wonderful name that has so
caught the public fancy, we would not be discussing these things on the television.
- By the end of 1964, we had built a computer program that could calculate the
stellar collapse using relativistic physics. We showed that the remnant neutron
core, in one of our earlier Newtonian calculations, would contract a bit more under
the stronger relativistic gravitational fields but that it would come to rest.
- However, more massive stars produced more massive cores and we showed that
these cores could enter a state of continuing collapse in which material densities
actually went to infinity. Further, we showed that light emitted from the center of
these overly massive collapsed cores never reached the surface of the star no
matter how long we extended our calculations. However, these calculations left
unanswered what would happen if some physics occurred at very high mass
densities that we had failed to include in our calculations. Could some
unanticipated physics stop the collapse or were black holes inevitable?
- In 1965, Roger Penrose, from Cambridge, discovered and published a famous
abstract mathematical theorem about general relativity. When applied to our
calculations, it said that once one of our cores reached a high enough density, no
matter and no light could ever escape from that core. This was the crucial piece of
physics needed to establish the existence of black holes. Our calculations showed
that the critical densities could be reached, and would be reached as the expected
outcome of the stellar collapse. Penrose's theorem told us that a black hole was the
- In a sense, all of this was speculation. Though physics told us they should occur, no
black hole or neutron star had ever been detected. We had no direct evidence of
their formation. Then, in February 1987, an extraordinary event gave dramatic
experimental confirmation of the scenario leading to the formation of neutron stars
and black holes. For the first time in 383 years, a supernova exploded in the
vicinity of our galaxy. Fortunately, there were several neutrino detector
experiments operating around the world. Though we are some 170,000 light years
from the star that spawned this supernova, the number of neutrinos emitted was so
enormous that approximately 10 billion passed through every square centimeter of
the earth in a fraction of a second. Neutrinos are elusive, hard to stop, hard to
detect. Most of these neutrinos passed right on through, not only through our
bodies, but through the entire earth. However, the detectors around the world,
altogether collected about a dozen and a half neutrinos that arrived in an
extraordinary pulse simultaneous with the optical signal from the exploding
supernova. Though the number detected was small it is consistent with the
predicted number of neutrinos arriving at earth and with the sensitivity of the
detectors. The violent collapse of the star and formation of a neutron core was
For information about this page, contact us at: firstname.lastname@example.org