This is a time line of life of the universe that ends in a Big Rip.
Credit: Jeremy Teaford, Vanderbilt University
The universe can be a very sticky place, but just how sticky is a matter of debate.
That is because for decades cosmologists have had trouble reconciling
the classic notion of viscosity based on the laws of thermodynamics
with Einstein's general theory of relativity. However, a team from
Vanderbilt University has come up with a fundamentally new mathematical
formulation of the problem that appears to bridge this long-standing
gap.
The new math has some significant implications for the ultimate fate
of the universe. It tends to favor one of the more radical scenarios
that cosmologists have come up with known as the "Big Rip." It may also
shed new light on the basic nature of dark energy.
The new approach was developed by Assistant Professor of Mathematics
Marcelo Disconzi in collaboration with physics professors Thomas Kephart
and Robert Scherrer and is described in a paper published earlier this
year in the journal Physical Review D.
"Marcelo has come up with a simpler and more elegant formulation that
is mathematically sound and obeys all the applicable physical laws,"
said Scherrer.
The type of viscosity that has cosmological relevance is different
from the familiar "ketchup" form of viscosity, which is called shear
viscosity and is a measure of a fluid's resistance to flowing through
small openings like the neck of a ketchup bottle. Instead, cosmological
viscosity is a form of bulk viscosity, which is the measure of a fluid's
resistance to expansion or contraction. The reason we don't often deal
with bulk viscosity in everyday life is because most liquids we
encounter cannot be compressed or expanded very much.
Disconzi began by tackling the problem of relativistic fluids.
Astronomical objects that produce this phenomenon include supernovae
(exploding stars) and neutron stars (stars that have been crushed down
to the size of planets).
Scientists have had considerable success modeling what happens when
ideal fluids -- those with no viscosity -- are boosted to near-light
speeds. But almost all fluids are viscous in nature and, despite decades
of effort, no one has managed to come up with a generally accepted way
to handle viscous fluids traveling at relativistic velocities. In the
past, the models formulated to predict what happens when these more
realistic fluids are accelerated to a fraction of the speed of light
have been plagued with inconsistencies: the most glaring of which has
been predicting certain conditions where these fluids could travel
faster than the speed of light.
"This is disastrously wrong," said Disconzi, "since it is well-proven
experimentally that nothing can travel faster than the speed of light."
These problems inspired the mathematician to re-formulate the
equations of relativistic fluid dynamics in a way that does not exhibit
the flaw of allowing faster-than-light speeds. He based his approach on
one that was advanced in the 1950s by French mathematician André
Lichnerowicz.
Next, Disconzi teamed up with Kephart and Scherrer to apply his
equations to broader cosmological theory. This produced a number of
interesting results, including some potential new insights into the
mysterious nature of dark energy.
In the 1990s, the physics community was shocked when astronomical
measurements showed that the universe is expanding at an
ever-accelerating rate. To explain this unpredicted acceleration, they
were forced to hypothesize the existence of an unknown form of repulsive
energy that is spread throughout the universe. Because they knew so
little about it, they labeled it "dark energy."
Most dark energy theories to date have not taken cosmic viscosity
into account, despite the fact that it has a repulsive effect strikingly
similar to that of dark energy. "It is possible, but not very likely,
that viscosity could account for all the acceleration that has been
attributed to dark energy," said Disconzi. "It is more likely that a
significant fraction of the acceleration could be due to this more
prosaic cause. As a result, viscosity may act as an important constraint
on the properties of dark energy."
Another interesting result involves the ultimate fate of the
universe. Since the discovery of the universe's run-away expansion,
cosmologists have come up with a number of dramatic scenarios of what it
could mean for the future.
One scenario, dubbed the "Big Freeze," predicts that after 100
trillion years or so the universe will have grown so vast that the
supplies of gas will become too thin for stars to form. As a result,
existing stars will gradually burn out, leaving only black holes which,
in turn, slowly evaporate away as space itself gets colder and colder.
An even more radical scenario is the "Big Rip." It is predicated on a
type of "phantom" dark energy that gets stronger over time. In this
case, the expansion rate of the universe becomes so great that in 22
billion years or so material objects begin to fall apart and individual
atoms disassemble themselves into unbound elementary particles and
radiation.
The key value involved in this scenario is the ratio between dark
energy's pressure and density, what is called its equation of state
parameter. If this value drops below -1 then the universe will
eventually be pulled apart. Cosmologists have called this the "phantom
barrier." In previous models with viscosity the universe could not
evolve beyond this limit.
In the Desconzi-Kephart-Scherrer formulation, however, this barrier
does not exist. Instead, it provides a natural way for the equation of
state parameter to fall below -1.
"In previous models with viscosity the Big Rip was not possible,"
said Scherrer. "In this new model, viscosity actually drives the
universe toward this extreme end state."
According to the scientists, the results of their pen-and-paper
analyses of this new formulation for relativistic viscosity are quite
promising but a much deeper analysis must be carried out to determine
its viability. The only way to do this is to use powerful computers to
analyze the complex equations numerically. In this fashion the
scientists can make predictions that can be compared with experiment and
observation.
The research was supported by National Science Foundation grant 1305705 and Department of Energy grant DE-SC0011981.
Story Source:
The above post is reprinted from
materials provided by
Vanderbilt University. The original item was written by David Salisbury.
Note: Materials may be edited for content and length.