Primordial Nucleosynthesis

BBN theory Observational abundances BBN predictions versus data Non standard BBN

Astroparticle Course

Some bibliography
Primordial Nucleosynthesis: from precision cosmology to fundamental physics, F. Iocco, G. Mangano, G. Miele, O. Pisanti, P.D. Serpico. arXiv:0809.0631.

Astroparticle Course

BBN

0.738 t (s) 2 T ( MeV )

During the expansion after the Big Bang, for 1 s < t < 3600 s (1 MeV > T > 0.01 MeV), when the radiation still dominated the energy density, the universe behaved as a nuclear reactor, producing sensible amounts of light nuclei, H, 2H, T, 3He, 4He, 6Li, 7Li, 12C, 16O... Key pillar of the Hot Big Bang Model, BBN is an overconstrained scenario. Its theoretical predictions depends on two parameters:

b h

eff

Astroparticle Course

Why BBN is important?

Primordial Elements Observations Xi

Nuclear Astrophysics Nuclear rates

BBN

CMB

PDG and cosmo stuff: n , GN, , Physics...

Astroparticle Course 4

Why BBN is important?

Primordial Elements Observations Xi N,

Nuclear Astrophysics Nuclear rates

BBN

Burles, Nollett & Turner, astro-ph/9903300

CMB

PDG and cosmo stuff: n , GN, , Physics...

Astroparticle Course 5

Why BBN is important?

Primordial Elements Observations Xi N,

Nuclear Astrophysics Nuclear rates

BBN

Burles, Nollett & Turner, astro-ph/9903300

Barger et al., PL B569 (2003)

CMB

PDG and cosmo stuff: n , GN, , Physics...

Astroparticle Course 6

Why BBN is important?

Primordial Elements Observations Xi N,

Nuclear Astrophysics Nuclear rates

BBN

Burles, Nollett & Turner, astro-ph/9903300

Barger et al., PL B569 (2003)

CMB

PDG and cosmo stuff: n , GN, , Physics...

Astroparticle Course

new Particle Physics Properties deviations from FRW

7

History
1946, Gamow: nuclear reactions in the early universe might explain the abundances of elements. Fermi and Turkevich: lack of stable nuclei with mass 5 and 8 prevents significant production of nuclei more massive than 7Li. 1964, Peebles, Hoyle and Tayler: YP 0.25. 1967, Wagoner, Fowler and Hoyle: first detailed calculation of light nuclei abundances. 1988 & 1992, Kawano, release of a user friendly code. Schramm, Turner, Steigman, Olive, 2008, release of a new generation code, PArthENoPE.

Astroparticle Course

Equilibrium versus non-equilibrium

May we describe particle distribution in phase space via equilibrium FD or BE distribution functions? Formally no (but for massless particles), practically yes, to a very good precision. As long as particle interactions are fast with respect to the expansion rate: when radiation dominates

f = e

1
E T (t )

T (t )

1 MeV gi t (sec) 1

1/ 2

Remember that out-of-equilibrium phases are crucial in the history of the universe. As expansion goes on, interactions mantaining kinetic and chemical equilibrium between species may become less and less efficient: FREEZE OUT.
Astroparticle Course 9

Neutrino decoupling I
In the early stages of the expansion, neutrinos are kept in kinetic and chemical equilibrium with e, n, p via weak processes. We must solve -Boltzmann equations to get their momentum distributions. x=a(t)me, y=a(t)|p|

+ e + e + + + e + + e + +
d 1 f ( x, y ) = I ( f ( x, y )) dx xH

Neutrino distributions enter the conversion weak rates (only electronneutrinos), their overall energy density (which determines H) and pressure.

Astroparticle Course

10

Neutrino decoupling II
We search for solutions of the system of integro-differential equations having a distortion with respect to the equilibrium form.

where the initial conditions are fixed by thermodynamical equilibrium (satisfied for T > 10 MeV).
xin = me 20 10 MeV

f (20, y ) = 0

The asymptotic distribution functions are not perfectly thermodynamical distributions.

Astroparticle Course 11

BBN in four steps

1. initial conditions 2. n/p ratio freeze out 3. D bottleneck 4. nuclear chain T> 1 MeV T 1 MeV T 0.1 MeV 0.1 MeV > T > 0.01 MeV

Astroparticle Course

12

Initial conditions
Notations: nA= NA/V; = nB/n 274 10-10 Bh2 = nB = B = B c = c B n m n m n mB n B c B XA = nA/nB YP 4 n4He/nB For large temperatures all nuclear species are kept in chemical equilibrium (Nuclear Statistical Equilibrium, NSE).
A+ A m AT nA = g A e T T 2 m A = Zm p + ( A Z )mn BA 3/ 2 m

A3 / 2 2 nA = g A A 2 T
BA=binding energy

3( A1) / 2

n p nn

A Z

e BA / T

A = Z p + ( A Z ) n

n A 2 (3) = nB

A1

n g A 3/ 2 T A1 p A 2 m n N Course B Astroparticle

3( A1) / 2

nn n B

A Z

e BA / T
13

Binding energy

Astroparticle Course

14

n/p ratio freeze-out

The density ratio of n and p is kept in chemical equilibrium by weak processes:

e + n e + p e + p n + e+ e + e + p n

mn m p nn = exp np T

All other nuclei abundances are negligible for T>1 MeV. average over particle distribution

v n > H (t ) v n H (t )

equilibrium freeze out

15

Astroparticle Course

Weak rates
As for purely leptonic weak interactions, also np processes freeze out at T 1 MeV. In the standard calculation the thermal averaged weak rates are evaluated at tree level with V-A theory and in the infinite nucleon mass limit (Born approximation). Example: E0

GF (cV + 3c A ) 2 2 ( n p + e + e ) = dp p E e e ( E ) f ( E ) f ( Ee ) 3 2
GF, cV and cA are well know from muon decay, 0+0+ beta decays and neutron beta decay angular distribution, GF = 1.16639 10-5 GeV-2 cV = 0.9751 0.0006

cA/ cV= 1.2601 0.0025

Astroparticle Course 16

Weak rates

(a) e + n e + p (b) e + p e + n (c ) e + + n e + p (d ) e + p e + + n (e ) n e + e + p ( f ) e + e + p n
Detailed balance

Astroparticle Course

17

Improving precision
How good is the Born approximation? A possible check is the calculation of the neutron lifetime in vacuum n(Born) = 961 sec n(exp) = (885.7 0.8) sec

Very bad!
To improve the precision in n/p ratio, and eventually in 4He abundance at percent level, it is necessary to consider: QED radiative effects finite nucleon mass QED plasma corrections O() O(me/mN) O( T/me)
Astroparticle Course 18

The goodness of QED expansion

Up to the first order in n(Born) = 961 sec n(th) = 893.9 sec to be compared with n(exp) = (885.7 0.8) sec Thus QED corrections, up to order , reduces of 7% the neutron lifetime. Another 0.9% is needed, maybe provided by second order corrections?
Astroparticle Course 19
p n n p

Total corrections (%)

Helium mass fraction

Using weak rates and solving the set of Boltzmann equation for n and p densities:

Xn (1 + exp( (mn-mp)/T))-1 0.150

This ratio slightly decreases due to neutron decay:

Xn Xn exp(-t/n) 0.122
A first rough estimate: Since basically all neutrons are eventually captured in 4He nuclei (largest gain in energy), neglecting all other nuclei:

YP =

4n 4 He nB

4 n/2 2 n/ p = = 2 0.122 = 0.244 n + p 1+ n / p

n/p~1/7

This rough estimate turns out to be rather accurate indeed!

Astroparticle Course 20

Deuterium formation
Deuterium formation is crucial for triggering the complicated nuclear reaction chain.

2 n + 2 p 4He
disfavoured (low density)

Astroparticle Course

21

Toward 4He

Astroparticle Course

22

Nuclear chain
Once 2H is produced, 4He is rapidly formed, along with small fractions of 3H, 3He, 6Li, 7Li and 7Be.
7Be

eventually gives 7Li by electron capture

e- + 7Be e + 7Li
Though both 12C and 16O have larger binding energy than 4He, they are not produced in sensible amounts since: i) no tightly bound isotopes with A=5, 8 ii) Coulomb barrier starts to be significant iii) low baryon density suppresses triple processes (@ 0.1 MeV baryon density air density)
Astroparticle Course 23

Deuterium production
Two competing processes:

p+n +2H +2H n+p

fusion photo-dissociation

poorly known experimentally, theoretical calculation

Astroparticle Course

24

D synthesis & destruction

T9=T(K)/109

Astroparticle Course

25

Deuterium bottleneck
One would expect that when T just drops below BD= 2.23 MeV, photodissociation processes become ineffective. However: too many photons!!

XD 12 (3) T exp(BD / T ) = XnX p m p

Deuterium formation starts (rapidly leading to exp(BD/T*) 1. Since 10-10, T* 0.1 MeV.
4He)

3/ 2

only when

Astroparticle Course

26

BBN differential equations

How to evaluate nuclei yields? BBN code: solving a set of coupled differential equations.

& 8 a + e + b + ) = ( 2 3mPl a &b & n a = 3 nb a & = (t , X ) T a Qlepton ( e , T ) = Qbaryon ( X a )

Nc Nd ( ) ( ) X X d & = N (c + d a + b ) c X a a Nc! Nd ! b ,c , d N N ( X a ) a (X b ) b (a + b c + d ) N a ! Astroparticle N b! Course

=energy density of relativistic species (contains Neff)

e=electron chemical potential

27

Boltzmann equations
Nuclear processes during BBN proceed in an environment very different with respect to the stellar plasmas, where stellar nucleosynthesis takes place. In stars the plasma is dense and species are mostly in chemical equilibrium. For BBN we have a hot and low density plasma with a significant population of free neutrons, which expands and cools down very rapidly, resulting in peculiar out of equilibrium" nucleosynthetic yields. So, we need Boltzmann equations. For a nuclide a, interacting via a+b c+d,

Since
Astroparticle Course 28

Nuclide evolution
we get

By introducing

i + j k +l

i+ jk +l nB

1 4 (4 ) fin in ( ) d 2 p p T T T M ni n j

2 T

fi f j

one gets the well known form

d X a = cd ab X c X d abcd X a X b dt
Astroparticle Course 29

Revision of nuclear chain

1. Nuclear reate analysis shows differences for reactions
2H+2H

n+3He, 2H+2H H+3H, 2H+H +3He

2. the revision of the key reaction p+n+2H has allowed to reduce the theoretical uncertainty 3. Several reactions involving Tritium were missing, like 3He+3H 2H+ 4He. They are going to affect the amount of Deuterium

Astroparticle Course

30

BBN and baryon fraction

How 2H, 4He and 7Li depends on = 274 10-10 B h2? When increases:
2H

a larger baryon density shifts the onset of 2H production towards larger temperatures, burning into 4He is more efficient; weakly increasing for more efficient burning; for small , 7Li decreases as a result of the balance of the two processes
4He+3H+7Li 7Li+p4He+4He

4He 7Li

for larger , 7Li starts growing due to larger 7Be production, leading to 7Li via electron capture:
4He+3He+7Be 7Be+e- +7Li e

Astroparticle Course

31

BBN and baryon fraction

How 2H, 4He and 7Li depends on = 274 10-10 B h2? When increases:
2H

a larger baryon density shifts the onset of 2H production towards larger temperatures, burning into 4He is more efficient; weakly increasing for more efficient burning; for small , 7Li decreases as a result of the balance of the two processes
4He+3H+7Li 7Li+p4He+4He

4He 7Li

for larger , 7Li starts growing due to larger 7Be production, leading to 7Li via electron capture:
4He+3He+7Be 7Be+e- +7Li e

Astroparticle Course

32

BBN and relativistic DOF

Relativistic degrees of freedom are historically described as effective number of neutrinos, but they can account for: non instantaneous decoupling effects; non standard neutrino physics; extra relativistic degrees of freedom:

+ X N eff

7 4 8 11

4/3

How 2H, 4He and 7Li depends on Neff? When Neff increases:
2H, 7Li 4He

only slightly affected; a larger Neff implies a larger expansion rate, thus an earlier freeze out and a higher value of the n/p ratio: more neutrons more 4He;
Astroparticle Course 33

BBN code
Inputs: 1. nuclear rates (experimental values extrapolated in the relevant energy range) 2. baryon density (~B h2) 3. energy density in relativistic degrees of freedom, Neff Output: Xa (, Neff) All nuclides with A 7, stable or weak decaying are included in the code. The only exception is for 6He decaying in 8 sec in 6Li.

Astroparticle Course

34

Element observation
Main problem: We cannot observe directly primordial abundances, since stars have changed the chemical composition of the universe. Strategies: 1. observations in systems negligibly contaminated by stellar evolution 2. careful account for galactic chemical evolution

Astroparticle Course

35

Deuterium
2H

H ISM Interstellar and proto-solar observations: 2H 5 C/H 0.01-0.001 (C/H )solar = ( 2 . 1 0 . 4 ) 10 H psc
Observation of Lyman absorption lines by neutral hydrogen (HI) gas clouds placed along the line of sight of Quasar systems at large red-shift (z 2 3): very few candidates low statistics and possible local effects
Astroparticle Course

is easily destroyed by galactic chemical evolution. Two different class of measurements. FUSE observations 2 H C/H solar value = (1.56 0.04) 10 5

ISO observations

36

Deuterium

Astroparticle Course

37

Deuterium

2H + 0.22 5 = ( 2 . 87 ) 10 0.21 H QAS

our analysis

Astroparticle Course

38

3He
In stellar interior 3He can be either produced by 2H-burning or destroyed in the hotter regions. All the 3He nuclides surviving the stellar evolution phase contribute to the chemical composition of the ISM. Stellar and galactic evolution models are necessary to track back the primordial 3He abundance from the post-BBN data, at least in the regions where stellar matter is present. The most accurate value was measured in Jupiter's atmosphere by the Galileo Probe, 3He/4He = (1.660.05)10-4. It supports the idea of a conversion of 2H initially present in the outer parts of the Sun into 3He via nuclear reactions. By counting the helium ions in the solar wind, the Ulysses spacecraft has measured 3He/4He = (2.48+0.68-0.62)10-4. The values found in Planetary Nebulas result one order of magnitude larger than in Proto Solar Material and Local Interstellar Medium, 3He/H =(2 - 5) 10-4, confirming a net stellar production of 3He in at least some stars.
Astroparticle Course 39

3He

No 3He dependence on environment metallicity, as predicted by chemical evolution models of Galaxy: known as the 3He problem. Thus a conservative approach: 3He/H < (1.1 0.2) 10-5.
Astroparticle Course 40

4He
evolution can be simply understood in terms of nuclear stellar processes which, through successive generations of stars, have burned H into 4He and heavier elements, hence increasing the 4He abundance above its primordial value. Since the history of stellar processing can be tagged by the metallicity (Z) of the particular astrophysical environment, YP can be derived by extrapolating the YP-O/H and YP-N/H correlations to O/H and N/H 0. Observation in ionized gas regions (He II HeI recombination lines) in low metallicity environments (BCG), with O abundances 0.02 0.2 times those in the sun. YP in different galaxies plotted as function of O and N abundances. regression to zero metallicity.
Astroparticle Course 41
4He

4He
Most recent estimates: 1. Izotov et al. (2004) reported the estimate YP = 0.24210.0021 2. Olive et al. (2004) quoted the value YP = 0.2490.009 (small sample size and large uncertainties) 3. Fukugita et al. (2006), based on a reanalysis of a sample of 33 HII regions from 1) determined a value of YP = 0.2500.004 4. Peimbert et al. (2007) presented a new 4He mass fraction determination yielding YP = 0.24770.0029 (new atomic physics computations together with observations and photoionization models of metal-poor extragalactic HII regions) All recent estimates are dominated by systematics. One can take, as central value of YP, the average (without weights) of the four determinations, while the systematic error is estimated as the semi-width of the distribution of the four best values, YP = 0.2470.002stat0.004syst.
Astroparticle Course 42

7Li
Method: observation of absorption lines in spectra of halo stars of POP II (low heavy element abundances: 0.03 - 0.0003 of solar ones). Very similar abundances of 7Li (Spite plateau) which should be close to the primordial value. However, possible increase in the abundance of 7Li with the iron abundance indicates that the 7Li of the plateau stars is not primordial. Source of systematics: 1) galactic chemical evolution of Li before star birth from the interstellar medium; 7 +0.44 10 Li / H = (2.19 0.37 ) 10 2) presence of anomalous stars in the sample; 7 + 0.36 10 Li / H = (2.07 0.23 ) 10 3) modelling of the depletion of the initial surface 7Li Astroparticle Course 43

Likelihood analysis
run a BBN code to get the abundances; construct the relative likelihood function:

2H 4He

( x Y ex ) 2 ( x Yi th ( N eff , )) 2 1 i exp Li ( N eff , ) = dx exp 2 th ex th 2 ex N 2 i ( N eff , ) i 2 ( , ) 2 i eff i

define a total likelihood function L=LD L4He plot the CL contours in the (Neff, B h2) plane

Astroparticle Course

44

Likelihood analysis
Only 2H cannot fix both Neff and B h2. the deuterium likelihood shows a degeneracy

Astroparticle Course

45

Likelihood analysis

the combined 2H-4He likelihood breaks the degeneracy

Astroparticle Course

46

Degenerate BBN
From the neutrality of the universe

e
T

10 10

but not severe bounds on neutrino-antineutrino asymmetry, =/T.

7 2 4 = T 120

non degenerate

T 4 e

slightly degenerate

1 8
2

strongly degenerate
Astroparticle Course 47

BBN and chemical potential

How 4He depends on i? When i increase: i contribute to Neff giving a larger expansion rate, thus more 4He these two effects imply a 2 4 30 i 15 i degenerate BBN region N = 3 + + + ... 2 in the (Neff, B h2) plane 7 4 i 7

e favours n p processes with respect to p n ones, thus less 4He Neutrino oscillations mix e, , : we can take them equal. From YP:

e = 0.004 0.009 stat 0.017 syst

Astroparticle Course 48