30 YEARS OF COSMIC STRINGS Alex Vilenkin Tufts Institute of Cosmology.

30 YEARS OF COSMIC STRINGS

Alex Vilenkin

Tufts Institute of Cosmology

30 YEARS OF COSMIC STRINGS

String evolution

Detection (bounds)

FOCUS ON:

30 YEARS OF COSMIC STRINGS

Superconducting strings VortonsSemilocal stringsString formation Strings in GUTsStrings in condensed matter …

LEAVE OUT:

0

20

40

60

80

100

120

1970 1975 1980 1985 1990 1995 2000 2005 2010

Strings areseeds ofgalaxies!

Strings are dead!

Cosmicsuper-strings!

A BRIEF HISTORYP

ub

lica

tio

ns

per

yea

r

Kibble 1976

OLD STRING EVOLUTION SCENARIO

• Distance between strings:

• Loop sizes:

• Loops decay by gravitational radiation:

ttd ~)(

ttl ~)(

Kibble (1976), A.V. (1981)

G

l~ Mass per unit

length of string

.1~~

THE FIRST COSMIC STRING REVOLUTION

High-resolution simulations: the loops are tiny

Bennett & Bouchet (1990)Allen & Shellard (1990)

310

(below theresolution)

Small-scale wiggles

0.6 3.0~ rad. matter

Loop sizes are set by the scale of wiggles.

SCENARIOS:

is determined by gravitational back-reaction:

G50~ Bennett & Bouchet (1990)

.2 ,)(~ nG n Siemens & Olum (2001)Siemens, Olum & A.V. (2002)

No scaling:

.0 Vincent, Hindmarsh & Sakellariadou (1997)

Observational predictions are sensitive to !

(“standard model”)

ttl ~)(

THE SECOND COSMIC STRING REVOLUTION

(still in progress!)

Small-scale wiggles and loops are resolved!

Ringeval, Sakellariadou & Bouchet (2005):

.17~/ 0max t

Olum & Vanchurin (2006):

.200~/ 0max t

Shellard & Martins (2005): .6~/ 0max t

Most of the energy goes Into loops with .

Requires a cutoff.

Loop formation on scales .

tl /

tl /

0~ l

.)( 2/5 ltnl

0l

Scaling peak in loop productiondevelops atafter a long transient regime.

tl 1.0~

Radiation era

1.0~ after a long transient regime.

ttl ~)(

Flat-space exact simulation

Vanchurin, Olum & A.V. (2005)

1000/ 0max t

tl /

The picture that seems to emergeis close to the old string scenario:

.1.0~ 0.3,~

Broad distribution of loops and small-scale wiggles.

(?)

Analytic models:

Kibble (1985)Bennett (1986)Copeland, Kibble & Austin (1992)Martins & Shellard (1996) Copeland, Kibble & Steer (1998)Polchinski & Rocha (2006)

To reach full understanding, we will need to combine numerical and analytic techniques.

COSMIC SUPERSTRINGS

Reconnection probability may be small: .110 3 p

Jackson, Jones & Polchinski (2004)

Witten (1985)Sarangi & Tye (2002)Majumdar & A. Davis (2002)

F, D and FD strings; FD networks.Copeland, Myers & Polchinski (2004)Dvali & A.V. (2004)

Metastable, but the lifetime can be >> 1010 yrs.

In models of brane inflation: .611 1010 G

Jones, Stoica & Tye (2003)

[Similar range in hybrid inflation GUT models]Jeannerot & Postma (2005)

How does affect string evolution?1p

Sakellariadou & A.V. (1990)Sakellariadou (2005)

.1 ps

Avgoustidis & Shellard (2006)

Simple argument suggests

Numerical evidence is inconclusive.

But in any case, for p << 1 there is a large number of strings per Hubble volume.

Direct observational test of string theory.

EVOLUTION OF FD-NETWORKSVachaspati & A.V. (1987)McGraw (1998)Tye, Wasserman & Wyman (2005)

Simple models

Scaling: ttd ~)(2/~/ GFD

depends on energy dissipation.

Spergel & Pen (1997)Copeland & Saffin (2005)Hindmarsh & Saffin (2006)

Global string network simulations

01.01.0~ 1/ FD

If the dominant energy lossis gravitational radiation:

G~1/1~/ GFD

Goldstone radiation

String domination

UrrestillaGauge strings

U(1)xU(1)

.25~/ 0max t

Scaling: Loop production?05.0~,~)( ttd

OBSERVATIONAL BOUNDS

STRING SIGHTINGS:

Sazhin et. al. (2003)

Cowie & Hu (1987)

Schild et. al. (2004)

GRAVITATIONAL RADIATION

Stochastic GW background & GW bursts from cusps.

Vachaspati & A.V. (1984) Hogan & Rees (1984) Caldwell & Allen (1992)Battye, Caldwell & Shellard (1996) …

Comparable power in bursts andin low harmonics.

1110~ GBursts may be detectable for .

Better for p << 1.

LIGO search is underway!

Damour & A.V. (2000,2005)Siemens et. al. (2006)Hogan (2006)

BOUNDS FROM PULSAR OBSERVATIONS

8 yrs: 710G Kaspi, Taylor & Ryba (1994)

17 yrs: 1010G Lommen (2002)Hogan (2006)

(disputed)

PTA: 2/38105.1 pG Jenet et. al. (2006)

Full PTA 1110~ G

(Pulsar Timing Array)

(20 pulsars for 5 yrs)

Implications of large loops )1.0~(

Nucleosynthesis bound:

.10 7G Vanchurin, Olum & A.V. (2005)

Reionization:

8103 G Olum & A.V. (2006)

loops seed early galaxy formation.

CMB BOUNDS

CMB anisotropies

7103 G Pogosian, Wasserman & Wyman (2006)

CMB polarization

B-type polarization due to vector perturbations induced by strings.

98 1010~ G may be detectable.Seljak & Slosar (2006)

Bad news: GUT-scale strings are ruled out.

)10( 8G

Good news: strings can be detected well belowthe GUT scale.

We are not likely to detect strings through gravitational lensing or CMB anisotropies.

Gravitational waves, CMB polarization

Constraint is much weaker for global strings:

.103 7G

CONCLUSIONS

A new generation of string simulations is underway.Strong indications of loop scaling; (?) important observational implications.

The strongest present bound on strings:

8105.1 G (PTA)

The most promising detection methods:pulsar timing, GW bursts, CMB polarization. May get to in ~ 5 yrs.1110~ G

1.0~

The field is as vibrant as ever!