Dark energy sound speed

 Posts: 1369
 Joined: September 23 2004
 Affiliation: University of Sussex
 Contact:
Dark energy sound speed
Are there any theoretical bounds on what sound speed (meaning [tex]c_s^2 \equiv \delta p / \delta \rho[/tex]) the dark energy could have? In particular [tex]c_s^2 <0[/tex] or [tex]c_s^2 > 1[/tex] are often not consisdered, but I don't see any immediate contradictions from the cosmological perturbation equations?

 Posts: 183
 Joined: September 24 2004
 Affiliation: Brookhaven National Laboratory
 Contact:
Dark energy sound speed
Well, isn't the sound speed just what is says on the can, i.e. the speed at which small perturbations move?
I would expecte that c_s^2>1 would violate causality, while c_s^2<0 would just give you evanescent waves, i.e. a dissipative media. So c_s^2<0 might actually be an interesting option after all.
I would expecte that c_s^2>1 would violate causality, while c_s^2<0 would just give you evanescent waves, i.e. a dissipative media. So c_s^2<0 might actually be an interesting option after all.

 Posts: 18
 Joined: September 25 2004
 Affiliation: ETH
 Contact:
Dark energy sound speed
I'm about to show my ignorance but I have no idea what an evanescent wave is. C_s^2<0 doesn’t seem to make much sense to me but it would be really interested to hear more if it does.
Adam
Adam

 Posts: 18
 Joined: September 25 2004
 Affiliation: ETH
 Contact:
Dark energy sound speed
Sorry, I've just been reading around. Just in case anyone else was as clueless as i was:
"An electromagnetic wave observed in total internal reflection, undersized waveguides, and in periodic dielectric heterostructures. While wave solutions have real wavenumbers k, k for an evanescent mode is purely imaginary. Evanescent modes are characterized by an exponential attenuation and lack of a phase shift."
Again, sounds interesting, but how does this apply to cosmology?
Adam
"An electromagnetic wave observed in total internal reflection, undersized waveguides, and in periodic dielectric heterostructures. While wave solutions have real wavenumbers k, k for an evanescent mode is purely imaginary. Evanescent modes are characterized by an exponential attenuation and lack of a phase shift."
Again, sounds interesting, but how does this apply to cosmology?
Adam

 Posts: 1369
 Joined: September 23 2004
 Affiliation: University of Sussex
 Contact:
Dark energy sound speed
How it applies to cosmology is really the question. If you give me a can of dark energy, then indeed [tex]c_s^2[/tex] is just what you think of as the sound speed: for a constant equation of state
[tex]\delta'' + k^2 c_s^2 \delta = 0[/tex]
(if the sound speed is negative, you immediately see that the solution is exponential rather than oscillating).
The question I'm asking is about dark energy in the universe. Here the full equations are relativistic and coupled to other matter by GR:
[tex]\delta' + 3H(\hat{c}_{s}^2w)(\delta +3H(1+w)v/k)+
(1+w)kv + 3H w' v/k= 3(1+w)h' [/tex]
[tex] v' + H(13\hat{c}_{s}^2)v + kA = k \hat{c}_{s}^2
\delta/(1+w) [/tex]
Much more complicated! (h' is a source from background expansion fluctuations, v is the velocity, A is the acceleration (zero in synchronous gauge), H is the conformal hubble rate, and derivatives are conformal time). The sound speed is defined in the rest frame of the dark energy as [tex]c_s^2 \equiv \delta p/ \delta \rho[/tex], which is not by definition positive despite the notation.
In general the equation of state [tex]w[/tex] is a function of time, but the sound speed can be a function of time and wavenumber k. So whether or not there are physical constraints on it  especially on superhorizon scales  is unclear to me. If the dark energy is some effective stuff, say a GR representation of some braneworld effect, I don't think you can argue constraints on its sound speed as though it were a simple fluid that could exist independent of cosmology?
[tex]\delta'' + k^2 c_s^2 \delta = 0[/tex]
(if the sound speed is negative, you immediately see that the solution is exponential rather than oscillating).
The question I'm asking is about dark energy in the universe. Here the full equations are relativistic and coupled to other matter by GR:
[tex]\delta' + 3H(\hat{c}_{s}^2w)(\delta +3H(1+w)v/k)+
(1+w)kv + 3H w' v/k= 3(1+w)h' [/tex]
[tex] v' + H(13\hat{c}_{s}^2)v + kA = k \hat{c}_{s}^2
\delta/(1+w) [/tex]
Much more complicated! (h' is a source from background expansion fluctuations, v is the velocity, A is the acceleration (zero in synchronous gauge), H is the conformal hubble rate, and derivatives are conformal time). The sound speed is defined in the rest frame of the dark energy as [tex]c_s^2 \equiv \delta p/ \delta \rho[/tex], which is not by definition positive despite the notation.
In general the equation of state [tex]w[/tex] is a function of time, but the sound speed can be a function of time and wavenumber k. So whether or not there are physical constraints on it  especially on superhorizon scales  is unclear to me. If the dark energy is some effective stuff, say a GR representation of some braneworld effect, I don't think you can argue constraints on its sound speed as though it were a simple fluid that could exist independent of cosmology?

 Posts: 183
 Joined: September 24 2004
 Affiliation: Brookhaven National Laboratory
 Contact:
Re: Dark energy sound speed
Yes, I agree with you. My point was that locally the dark energy doesn't know about cosmology and hence [tex]c_s^2[/tex] cannot be greater than one, at least for large [tex]k[/tex]. If you take [tex]c_s^2[/tex] to be a function of [tex]k[/tex], then it can be anything on superhorizon scales... Do you buy this?Antony Lewis wrote:In general the equation of state [tex]w[/tex] is a function of time, but the sound speed can be a function of time and wavenumber k. So whether or not there are physical constraints on it  especially on superhorizon scales  is unclear to me. If the dark energy is some effective stuff, say a GR representation of some braneworld effect, I don't think you can argue constraints on its sound speed as though it were a simple fluid that could exist independent of cosmology?
Your complicated equations are perfect for beard stroking at 5pm with a cup of tea!

 Posts: 183
 Joined: September 24 2004
 Affiliation: Brookhaven National Laboratory
 Contact:
Re: Dark energy sound speed
Following Amara's suit, I will now ask a stupid question. Isn't it that [tex]{\rm d}p/{\rm d}\rho=w[/tex] by definition of [tex]w[/tex] (i.e. 1/3 for standard waves, etc...) How can you then define [tex]\delta p/\delta \rho[/tex] for a given [tex]k[/tex] mode and how does it relate to [tex]w[/tex]?Antony Lewis wrote:Much more complicated! (h' is a source from background expansion fluctuations, v is the velocity, A is the acceleration (zero in synchronous gauge), H is the conformal hubble rate, and derivatives are conformal time). The sound speed is defined in the rest frame of the dark energy as [tex]c_s^2 \equiv \delta p/ \delta \rho[/tex], which is not by definition positive despite the notation.

 Posts: 1369
 Joined: September 23 2004
 Affiliation: University of Sussex
 Contact:
Re: Dark energy sound speed
The equation of state is defined as [tex]w\equiv p'/\rho'[/tex]. This is not the same as the definition of [tex]c_s^2[/tex] so in general they are different (indeed for quintessence [tex]c_s^2=1[/tex] always). They are only equal if you apply adiabaticity: not the case for quintessence.
Some refs are astroph/0307100, astroph/0307104, astroph/0410680.
Some refs are astroph/0307100, astroph/0307104, astroph/0410680.

 Posts: 43
 Joined: November 11 2004
 Affiliation: LUTH, Observatoire de ParisMeudon
 Contact:
Re: Dark energy sound speed
This statement is too strong, whether scalar field perturbations are adiabatic or isocurvature depends on the initial conditions, which we think are set by inflation.Antony Lewis wrote: not the case for quintessence.
If the quintessence field is a decay product of the inflaton the perturbations in the quintessence will be adiabatic, which means that not only the perturbations of the quintessence field relative to the other fluids
(S_q f =0) initially vanish, but also the scalar field intrinsic entropy perturbation (Gamma_q =0) is initially zero.
If these conditions are verified at an initial time the conservation of the energy momentum tensor (which provide the equation of motion for the perturbations) impose that they are verified at any other time (no matter the gauge, actually everything can be formulated in gauge invariant language as S and Gamma are gauge invariant variables).
In other words adiabaticity is preserved by the flow equations and no matter what is the background scalar field evolution or the choice of the gauge, the scalar field perturbations will remain adiabatic (Gamma_q=0 always) as those in the other fluids.
This can be inferred from General Relativity arguments in the context of the Separate Universe Approach (Wands, Malik, Lyth and Liddle, astroph/0003278).
More specifically we have shown this by directly looking at the flow equations in the gauge invariant formulation (Bartolo, Corasaniti, Liddle, Malquarti, astroph/0311503).
Also we have found that if the initial conditions are isocurvature ones, which could be the case if the quintessence or kessence or whatever is not a decay product of the inflaton, then the evolution of the isocurvature quintessence mode depends on the dynamics of the background. In particular during kination regimes the isocurvature mode is amplified, while it decays during tracker ones. This implies that only those quintessence scenarios with initial isocurvature perturbations and with the field undergoing a long kination phase followed by a short period of tracking can give rise to a cosmologically relevant isocurvature perturbation that can survive at present time. (The specific time evolution of this isocurvature mode can be different in the case of kessence, since the two scenarios (Q <> K) can be mapped one into the other only at the homogeneus level).
I tend to accept the fact that c_s^2, as a free parameter for an effective fluid description of scalar field perturbations, is a simple way of parametrizing something which depends on initial conditions, however from a pure physical point of view I find this effective description rather misleading, so one should have always in mind the microscopic description of the scalar field whatsoever is its lagrangian.
PierStefano

 Posts: 1369
 Joined: September 23 2004
 Affiliation: University of Sussex
 Contact:
Re: Dark energy sound speed
I should probably have said internally adiabatic (following e.g. astroph/0410680). The rest frame sound speed of a quintessence field is always identically 1 independent of the initial conditions (a simple proof is given in e.g. astroph/0307104). Hence for [tex]w\ne 1[/tex] it is never internally adiabatic. I hope this is correct?
I take you point to be that for models in which [tex]c_s^2[/tex] is not fixed, it may depend on the initial conditions. This is an interesting point I'd not considered.
I take you point to be that for models in which [tex]c_s^2[/tex] is not fixed, it may depend on the initial conditions. This is an interesting point I'd not considered.

 Posts: 43
 Joined: November 11 2004
 Affiliation: LUTH, Observatoire de ParisMeudon
 Contact:
Re: Dark energy sound speed
If by internally adiabatic you mean that the intrinsic entropy perturbation [tex]\Gamma_q = 0[/tex], this is true as long as the initial conditions are adiabatic and on the large scales approximation. In fact inside the horizon the modes get mixed up even if you start from initial adiabatic conditions, as can be seen from Eq.(33) and (34) in astroph/0311503.Antony Lewis wrote:I should probably have said internally adiabatic (following e.g. astroph/0410680). The rest frame sound speed of a quintessence field is always identically 1 independent of the initial conditions (a simple proof is given in e.g. astroph/0307104). Hence for [tex]w\ne 1[/tex] it is never internally adiabatic. I hope this is correct?
I take you point to be that for models in which [tex]c_s^2[/tex] is not fixed, it may depend on the initial conditions. This is an interesting point I'd not considered.
Perhaps it is just a different language, [tex]c_s^2=1[/tex] it is used to define adiabatic scalar field perturbations in the effective fluid description, while [tex]c_s^2\ne 1[/tex] would correspond to isocurvature ones.
But I find this wording misleading anyway since as you say in the scalar field rest frame [tex]c_s^2=1[/tex] independently of the initial conditions, while adiabaticity has to do with the initial conditions independently of the gauge. And personally I prefer the microscopic description rather than the effective one, although the former requires to specify a lagrangian.

 Posts: 2
 Joined: November 12 2004
 Affiliation: INAFCNR/IASF, Sezione di Bologna
Re: Dark energy sound speed
Dear all,
I take this chance to communicate my ideas on the topic, which have generated
astroph/0211626, astroph/0304325, astroph/0405041.
My motivation was to understand if you can discriminate within different Dark Energy models with a different speed of sound: this was suggested by Kessence authors, since quintessence has always c_x^2=1. By c_X^2 I mean the quantity in the pressure
perturbations
\delta p_x = c_x^2 \delta \rho_x + f(\eta,w_x) \Theta_x
where the second term denotes the intrinsic nonadiabatic pressure perturbations,
\theta_x is the velocity potential for DE, w_x is the equation of state of w_x and \eta is the conformal time.
Within scalar field theories (where f(\eta) \ne 0)
it is difficult to discriminate between c_x^2=1 and generic c_x (unless c_x is drastically different from 1) as it can be seen in astroph/0405041.
For perfect fluid of model of DE (where f(\eta) = 0) , c_x^2 is locked in some way to w_x
and there is a chance to have stronger effects on CMB and LSS (as can be seen in
astroph/0211626, astroph/0304325). For instance, the Chaplygin gas is ruled out at more than 3 \sigma as a dark energy model (see astroph/0304325).
Cheers,
Fabio
I take this chance to communicate my ideas on the topic, which have generated
astroph/0211626, astroph/0304325, astroph/0405041.
My motivation was to understand if you can discriminate within different Dark Energy models with a different speed of sound: this was suggested by Kessence authors, since quintessence has always c_x^2=1. By c_X^2 I mean the quantity in the pressure
perturbations
\delta p_x = c_x^2 \delta \rho_x + f(\eta,w_x) \Theta_x
where the second term denotes the intrinsic nonadiabatic pressure perturbations,
\theta_x is the velocity potential for DE, w_x is the equation of state of w_x and \eta is the conformal time.
Within scalar field theories (where f(\eta) \ne 0)
it is difficult to discriminate between c_x^2=1 and generic c_x (unless c_x is drastically different from 1) as it can be seen in astroph/0405041.
For perfect fluid of model of DE (where f(\eta) = 0) , c_x^2 is locked in some way to w_x
and there is a chance to have stronger effects on CMB and LSS (as can be seen in
astroph/0211626, astroph/0304325). For instance, the Chaplygin gas is ruled out at more than 3 \sigma as a dark energy model (see astroph/0304325).
Cheers,
Fabio
Antony Lewis wrote:The equation of state is defined as [tex]w\equiv p'/\rho'[/tex]. This is not the same as the definition of [tex]c_s^2[/tex] so in general they are different (indeed for quintessence [tex]c_s^2=1[/tex] always). They are only equal if you apply adiabaticity: not the case for quintessence.
Some refs are astroph/0307100, astroph/0307104, astroph/0410680.