Tuesday, May 24, 2005

Wormholes, NEC and all that

The BBC and New Scientist articles seem to have generated a lot of interest in this topic. Odd how my colleagues can hear me loudly discussing this stuff for six months with my postdoc and grad student, but only after the BBC decides to write about it do they want to know more :-)

The original papers are listed below. Both have been revised since posting on arxiv.org - if you want a more up to date version please contact me.

http://arxiv.org/abs/hep-th/0504003 (wormholes)
http://arxiv.org/abs/hep-th/0502203 (instability and NEC)

There is a longer version of the instability analysis forthcoming, by Buniy, Hsu and Murray.

Let me make some comments here for physics readers:

1) Our original interest was in dark energy. The observational data suggest (although not strongly - see comments) that w = p/rho < -1, which violates various energy conditions. We wanted to understand how easy or hard it is to build models with w < -1. With some collaborators at Caltech, I had obtained a result in classical scalar models that w < -1 implies instability. We wanted to generalize this result.

2) Our strongest results are in the contexts of classical field theory (including both gauge and scalar fields) and perfect fluids. There is a quantum loophole involving renormalization that allows for small violations of the NEC (well-known examples are the Casimir effect and black hole spacetimes).

3) When applying this to wormholes, we are considering the exotic (NEC-violating) matter necessary to stabilize the wormhole. This matter must have large energy-momentum tensor T_mn. We focus on wormholes which have nearly-classical spacetimes (the other type is less useful for Sci Fi). We show that this condition is strong enough to require that the exotic matter evolves semi-classically - i.e., it is subject to our results in classical field theory.

4) Some readers (esp. from the relativity community) have misinterpreted our results as claiming that the Casimir or black hole vacuum is unstable, but this is not the case (see point (2) above). In the wormhole case, the key point is that semi-classical wormholes cannot result from exotic matter which violates the NEC via quantum effects.


Dave Bacon said...

Steve, you say observational data suggest that p/rho<-1, but your paper shows that many models with p/rho<-1 are not stable. Is your result really in direct conflict with the experimental data?

steve said...


No, there is a big allowed range for w. (Roughly, -1.6 < w < -.7 or so at 95% CL.)

Some have interpreted the fact that there is more allowed parameter space with w < -1 to be a hint that w *is* less than -1. But it seems unlikely to me. A relativist who is familiar with the energy conditions might suggest to the cosmo guys that they adopt w > -1 as a *prior* in their analysis, but that is somewhat controversial.

Our work suggests it is quite hard to produce a stable model with w < -1. So, as observations get better we would expect the value of w to converge on something > -1.

The wormhole stuff, which was kind of a side observation, somehow gets all the attention :-)

BTW, welcome back to the pacific NW!

Dave Bacon said...

Oh, I see. I wonder how close to -1 w gets. It seems to me that an observational value of w=-1 would be very strange and might even be part of the key to understand what the heck this dark energy is. Ah well, one can dream, right?

Thanks for the welcome. Now if I can just let myself readjust to the laid back nature of the PNW.

steve said...


Actually w = 1 corresponds to a cosmological constant (or false vacuum)! So, that is a very plausible value. It saturates a number of the energy conditions from GR without violating them.

Blog Archive