LOG#123. Basic Neutrinology(VIII).

There are some indirect constraints/bounds on neutrino masses provided by Cosmology. The most important is the one coming from the demand that the energy density of the neutrinos should not be too high, otherwise the Universe would collapse and it does not happen, apparently…

Firstly, stable neutrinos with low masses (about m_\nu\leq 1 MeV) make a contribution to the total energy density of the Universe given by:


and where the total mass is defined to be the quantity

\displaystyle{m_{tot}=\sum_\nu \dfrac{g_\nu}{2}m_\nu}

Here, the number of degrees of freedom g_\nu=4(2) for Dirac (Majorana) neutrinos in the framework of the Standard Model. The number density of the neutrino sea is revealed to be related to the photon number density by entropy conservation (entropy conservation is the key of this important cosmological result!) in the adiabatic expansion of the Universe:


From this, we can derive the relationship of the cosmic relic neutrino background (neutrinos coming from the Big Bang radiation when they lost the thermal equilibrium with photons!) or C\nu B and the cosmic microwave background (CMB):

T_{C\nu B}=\left(\dfrac{3}{11}\right)^{1/3}T_{CMB}

From the CMB radiation measurements we can obtain the value


for a perfect Planck blackbody spectrum with temperature

T_{CMB}=2.725\pm0.001 K\approx 2.35\cdot 10^{-4}eV

This CMB temperature implies that the C\nu B temperature should be about

T_{C\nu B}^{theo}=1.95K\approx 0.17meV

Remark: if you do change the number of neutrino degrees of freedom you also change the temperature of the C\nu B and the quantity of neutrino “hot dark matter” present in the Universe!

Moreover, the neutrino density \Omega_\nu is related to the total neutrino density and the critical density as follows:


and where the critical density is about

\rho_c=\dfrac{3H_0^2}{8\pi G_N}

When neutrinos “decouple” from the primordial plasma and they loose the thermal equilibrium, we have m_\nu>>T, and then we get

\Omega_\nu h^2=10^{-2}m_{tot}eV

with h the reduced Hubble constant. Recent analysis provide h\approx 67-71\cdot 10^{-2} (PLANCK/WMAP).

There is another useful requirement for the neutrino density in Cosmology. It comes from the requirements of the BBN (Big Bang Nucleosynthesis). I talked about this in my Cosmology thread. Galactic structure and large scale observations also increase evidence that the matter density is:

\Omega_Mh^2\approx 0.05-0.2

These values are obtained through the use of the luminosity-density relations, galactic rotation curves and the observation of large scale flows. Here, the \Omega_M is the total mass density of the Universe as a fraction of the critical density \rho_c. This \Omega_M includes radiation (photons), bayrons and non-baryonic “cold dark matter” (CDM) and “hot dark matter” (HDM). The two first components in the decomposition of \Omega_M


are rather well known. The photon density is

\Omega_rh^2=\Omega_\gamma h^2=2.471\cdot 10^{-5}

The deuterium abundance can be extracted from the BBN predictions and compared with the deuterium abundances in the stellar medium (i.e. at stars!). It shows that:

0.017\leq\Omega_Bh^2\leq 0.021

The HDM component is formed by relativistic long-lived particles with masses less than about 1keV. In the SM framework, the only HDM component are the neutrinos!

The simulations of structure formation made with (super)computers fit the observations ONLY when one has about 20% of HDM plus 80% of CDM. A stunning surprise certainly! Some of the best fits correspond to neutrinos with a total mass about 4.7eV, well above the current limit of neutrino mass bounds. We can evade this apparent contradiction if we suppose that there are some sterile neutrinos out there. However, the last cosmological data by PLANCK have decreased the enthusiasm by this alternative. The apparent conflict between theoretical cosmology and observational cosmology can be caused by both unprecise measurements or our misunderstanding of fundamental particle physics. Anyway observations of distant objects (with high redshift) favor a large cosmological constant instead of Hot Dark Matter hypothesis. Therefore, we are forced to conclude that the HDM of \Omega_M does not exceed even 0.2. Requiring that \Omega_\nu <\Omega_M, we get that \Omega_\nu h^2\leq 0.1. Using the relationship with the total mass density, we can deduce that the total neutrino mass (or HDM in the SM) is about

m_\nu\leq 8-10 eV or less!

Mass limits, in this case lower limits, for heavy or superheavy neutrinos (M_N\sim 1GeV or higher) can also be obtained along the same reasoning. The puzzle gets very different if the neutrinos were “unstable” particles. One gets then joint bounds on mass and timelife, and from them, we deduce limits that can overcome the previously seen limits (above).

There is another interesting limit to the density of neutrinos (or weakly interacting dark matter in general) that comes from the amount of accumulated “density” in the halos of astronomical objects. This is called the Tremaine-Gunn limit. Up to numerical prefactors, and with the simplest case where the halo is a singular isothermal sphere with \rho\propto r^{-2}, the reader can easily check that

\rho=\dfrac{\sigma^2}{2\pi G_Nr^2}

Imposing the phase space bound at radius r then gives the lower bound

m_\nu>(2\pi)^{-5/8}\left(G_Nh_P^3\sigma r^2\right)^{-1/4}

This bound yields m_\nu\geq 33eV. This is the Tremaine-Gunn bound. It is based on the idea that neutrinos form an important part of the galactic bulges and it uses the phase-space restriction from the Fermi-Dirac distribution to get the lower limit on the neutrino mass. I urge you to consult the literature or google to gather more information about this tool and its reliability.

Remark: The singular isothermal sphere is probably a good model where the rotation curve produced by the dark matter halo is flat, but certainly breaks down at small radius. Because the neutrino mass bound is stronger for smaller \sigma r^2, the uncertainty in the halo core radius (interior to which the mass density saturates) limits the reliability of this neutrino mass bound. However, some authors take it seriously! As Feynman used to say, everything depends on the prejudges you have!

The abundance of additional weakly interacting light particles, such as a light sterile neutrino \nu_s or additional relativistic degrees of freedom uncharged under the Standard Model can change the number of relativistic degrees of freedom g_\nu. Sometimes you will hear about the number N_{eff}. Planck data, recently released, have decreased the hopes than we would be finding some additional relativistic degree of freedom that could mimic neutrinos. It is also constrained by the BBN and the deuterium abundances we measured from astrophysical objects. Any sterile neutrino or extra relativistic degree of freedom would enter into equilibrium with the active neutrinos via neutrino oscillations! A limit on the mass differences and mixing angle with another active neutrino of the type

\Delta m^2\sin^2 2\theta\leq 3\cdot 10^{-6}eV^2 should be accomplished in principle. From here, it can be deduced that the effective number of neutrino species allowed by neutrino oscillations is in fact a litle higher the the 3 light neutrinos we know from the Z-width bound:

N_\nu (eff)<3.5-4.5

PLANCK data suggest indeed that N_\nu (eff)< 3.3. However, systematical uncertainties in the derivation of the BBN make it too unreliable to be taken too seriously and it can eventually be avoided with care.