Feb 2004

Explaining Baryon 0.2 Dark through the synthesis of ordinary matter from mirror matter: a more general analysis
R. Foot and R. R. Volkas
School of Physics, Research Centre for High Energy Physics, The University of Melbourne, Victoria 3010, Australia The emerging cosmological picture is of a spatially at universe composed predominantly of three components: ordinary baryons (B 0.05), non-baryonic dark matter (Dark 0.22) and dark energy ( 0.7). We recently proposed that ordinary matter was synthesised from mirror matter, motivated by the argument that the observed similarity of B and Dark suggests an underlying similarity between the fundamental properties of ordinary and dark matter particles. In this paper we generalise the previous analysis by considering a wider class of eective operators that nongravitationally couple the ordinary and mirror sectors. We nd that while all considered operators imply Dark = fewB , only a subset quantitatively reproduce the observed ratio B /Dark 0.20. The 1 eV mass scale induced through these operators hints at a connection with neutrino oscillation physics.
PACS numbers: 95.35.+d,98.80.Cq,11.30.Er,12.90.+b

arXiv:hep-ph/0402267v2 25 Mar 2004

I.

INTRODUCTION

A variety of evidence, culminating in the Wilkinson Microwave Anistropy Probe measurements of the cosmic microwave background [1], points to a spatially at universe composed of approximately 5% baryons (B), 22% nonbaryonic dark matter (DM) and roughly 70% dark energy. These fractions, while dierent, are suspiciously similar in magnitude. From a theoretical point of view, this is surprising, as one might a priori expect the physics of each component to be rather dierent. This is a new naturalness puzzle. The similarity of the positive and negative pressure components, roughly 30% and 70% respectively, has received some attention in the literature. However, the similar magnitudes of B and Dark are also puzzling. (As usual, X denotes the ratio of the energy density of component X and the critical density.) In Ref. [2], we began an exploration of a possible solution. We proposed that the similar baryonic and dark matter densities suggest that the internal microphysics of each component are actually similar or identical. Specically, this will be case if the dark matter is identied with mirror baryonic matter, B .1 The purpose of this new paper is to extend the analysis of Ref. [2]. A theoretical motivation for mirror matter [4, 5] is to retain the full Poincar group, including improper Lorentz e transformations such as parity inversion and time reversal, as an exact symmetry group of nature, despite the V A character of weak interactions. Mirror matter theories have a G G gauge group structure, with ordinary and mirror particles assigned to (R, 1) and (1, R) representations, respectively. Under parity transformations, a given ordinary particle interchanges with its mirror partner. For fermions, a left-handed (right-handed) ordinary particle is paired with a right-handed (left-handed) mirror partner. Time-reversal invariance T follows from the CP T theorem, with T P CP T where P is the exact mirror parity operation dened above. We shall focus on the simplest mirror matter model, where G is the standard model gauge group, there are no exotic fermions besides the mirror partners of known quarks and leptons, and there is just one Higgs doublet paired with one mirror Higgs doublet. We shall consider only the case where the vacuum also respects improper Lorentz transformations, that is, the discrete spacetime symmetries are not spontaneously broken [5]. This means that an ordinary particle and its mirror partner have the same mass. A large region of Higgs potential parameter space allows this aesthetically appealing scenario [5]. Mirror particles interact amongst themselves through their own versions of the strong, weak and electromagnetic interactions. The two sectors are inevitably coupled by gravity, but various non-gravitational interactions can also connect them. Within the class of renormalisable, gauge-invariant and mirror-symmetry-invariant operators, these in-

Electronic Electronic 1

address: foot@physics.unimelb.edu.au address: r.volkas@physics.unimelb.edu.au The B Dark issue was very briey touched upon by Bento and Berezhiani [3] also in the context of mirror matter models. Note, though, that their mechanism produces B = for the symmetric mirror model. B

2 teractions are photonmirror-photon kinetic mixing and Higgs-bosonmirror-Higgs-boson coupling through a term (we denote mirror elds with a prime) [5]. The implications of these interactions have been explored in several papers [5, 6]. The mechanism under study here employs dimension-5 eective operators of the form lepton-Higgs-lepton -Higgs to couple the ordinary and mirror sectors. It also requires an assumption about the initial lepton and baryon asymmetries (prior to reprocessing). In Ref. [2], a particular such operator with natural initial asymmetries (see below for a review) was found to yield B 0.20 0.21, Dark (1)

in excellent agreement with the allowed range suggested by WMAP[1] 0.20 0.02. We would now like to know if this encouraging quantitative success is unique to the eective operator we happened to consider in Ref. [2]. The purpose of this paper is to examine a wider class of operators. We shall nd that while all selected operators yield B /Dark < 1, only a subset are quantitatively consistent with the observations, and that the operator of Ref. [2] is not uniquely favoured (from this cosmological perspective). The next section reviews the dynamical framework of Ref. [2] and the section after that presents our calculations. We end by discussing our results and drawing conclusions.
II. REVIEW OF THE DYNAMICAL FRAMEWORK

A fundamental feature of mirror matter cosmology is that the temperature of the mirror sector, T , is expected to be dierent from that of the ordinary sector, T , at least during certain epochs. For the late epoch during which big < bang nucleosynthesis takes place, T /T 0.5 should hold in order to constrain the expansion rate of the universe at that time. From a later period still, large scale structure formation with mirror DM suggests a slightly more stringent < constraint, T /T 0.2 [7]. The smaller this ratio is, the more does mirror DM resemble standard cold DM during the linear regime of density perturbation growth (they obviously must dier in the non-linear regime because mirror DM is self-interacting, chemically complex and dissipative) [7]. For earlier epochs, however, we have no observational constraints that T /T must satisfy. If Dark = , our fundamental hypothesis, then the inequality of and B B B strongly suggests that the temperatures were also dierent for at least part of the time during baryogenesis. A temperature dierence can be created, for example, through ination [8]. Imagine that there is an inaton and a mirror inaton, and that ination is seeded by a uctuation through whichever of these elds the uctuation favours. Upon reheating, the macroscopic ordinary-mirror asymmetry generated by the amplied uctuation will translate into T = T provided the two sectors are weakly enough coupled to each other. The subsequent evolution of T /T depends on the precise nature of the ordinary-mirror coupling terms. We emphasise that the macroscopic temperature asymmetry is not at all inconsistent with an exactly symmetric microphysics. We can now review the main dynamic events our scenario requires: Step 1. Suppose that reheating after ination leaves a universe with T T, (2)

that is, the universe is totally dominated initially by mirror matter. This would happen if ination was driven by the vacuum energy of a mirror inaton. Step 2. Around a certain temperature, T = T1 , mirror baryon and/or mirror lepton asymmetries are created. We prefer not to specify the mechanism. It might be the out-of-equilibrium decays of the heavy neutral mirror leptons as per leptogenesis [9], or something else. No signicant ordinary-sector asymmetries are generated a reasonable assumption in the limit of Eq.(2). Step 3. These initial asymmetries are chemically reprocessed. In particular, ordinary asymmetries are created through eective dimension-5 operators of the form L= 1 MN iL c + H.c., jR
ij

(3)

where L is a left-handed ordinary lepton doublet, is its mirror partner, is the ordinary Higgs doublet, and R is its mirror partner. The indices i, j = 1, 2, 3 denote the three families. These are the lowest dimension nonrenormalisable, gauge-invariant operators one can construct out of the fundamental elds. They may be generated from renormalisable operators through the exchange of gauge-singlet fermions, hence the notation MN . They also induce terms m 2 /MN in the light neutrino mass matrix. As well as contributing (with sphaleron and other

3 eects) to the chemical reprocessing of asymmetries, these interactions also thermally equilibrate the sectors, that is, induce T = T . They operate during the temperature regime, MN T T2 1010 (eV /m )2 GeV.
> > >

(4)

For T MN , the eective Lagrangian approach is not valid and the parent fundamental, renormalisable interactions < are slower than the expansion rate, while for T T2 the eective interactions are themselves slower than the expansion rate. The issue explored in this paper is how dierent family hierarchy assumptions for the (1/MN )ij aect B / . B Step 4. A second but relatively brief inationary episode (or some alternative process) must then occur (beginning at < some temperature, T3 ) in order to set up the mild hierarchy, T /T 0.2 0.5, as required for big bang nucleosynthesis and the later linear perturbation growth periods. It is interesting that this hierarchy is in the opposite sense to that created by the rst inationary episode [see Eq. (2)]. If the mirror inaton induces the earlier inationary burst, then it is tempting to ascribe the later burst to a largely failed attempt by the ordinary inaton to reciprocate.
III. CALCULATING B /Dark

The main issue is the family structure assumed for Eq. (3). We shall take for simplicity and deniteness that one avour combination for the eective operators dominates [through having the largest (1/MN )ij ], with all the others too small to aect the chemical reprocessing. Given the connection with neutrino mass, we shall assume the relevant MN is of order 2 /(1 eV ), since the 1 eV scale is a reasonable upper bound on neutrino mass, and is even directly suggested by the LSND anomaly [10] [see also Ref. [11] for discussions of ordinary-mirror neutrino mixing]. In Ref. [2], we examined only one case, where the i = j = 2 term dominates. Assuming that there is just one eV mass term connecting the ordinary and mirror sectors and that it is approximately avour diagonal, then there are just 6 distinct cases to consider (including i = j = 2 to be called case 1 from now on): case 1 : case 2 : case 3 : case 4 : case 5 : case 6 : L= L= L= L= L= L= 1 c 2L 2R + H.c. MN 1 c 1L 1R + H.c. MN 1 c 3L 3R + H.c. MN 1 c 1 c 2L 3R + 3L + H.c. MN MN 2R 1 c 1 c 1L 3R + 3L + H.c. MN MN 1R 1 c 1 c 1L 2R + 2L + H.c. MN MN 1R (5) (6) (7) (8) (9) (10)

These operators aect chemical reprocessing by constraining the chemical potentials of the species concerned. We now need to review how the reprocessing is analysed. For temperatures below about 1012 GeV , QCD [12] and electroweak [13] non-perturbative processes plus the Yukawa interactions for the fermions c, , b and t are faster than the expansion rate of the universe [14]. The Yukawa interactions for eR , uR , dR , R and sR do not become fast enough until the temperature drops below about 1010 GeV . Diagonal and o-diagonal weak interactions involving left-handed quarks are also happening rapidly. By T2 1010 GeV [see Eq. (4)], the selected eective operator, one of Eqs. (5-10), has also been inducing rapid interactions, aecting the chemical composition and inducing T = T . At about T = 1010 GeV , all the quarks, leptons, Higgs bosons and their mirror partners are in thermal equilibrium with distribution functions governed by the temperature and chemical potentials for all the involved species. We denote the chemical potential for species X by X (X by ). The rapid processes listed above relate the s. In addition, we impose electric charge or X hypercharge neutrality, and mirror electric/hypercharge neutrality, for the universe. The chemical constraint equations are [14]
3

9q +
i=1 3

i = 0

(Electroweak non-perturbative), (QCD non-perturbative),

6q
i=1

(ui + di ) = 0

4
3

3q + 2 +
i=1

(2ui di i ei ) = 0 q d3 q + u2 q + u3 3 e3 = = = = 0 0 0 0

(Electric charge neutrality), (b-quark Yukawa), (c-quark Yukawa), (t-quark Yukawa), ( -lepton Yukawa),

(11)

plus the corresponding seven equations from the mirror sector. The simplied notation here is: q q1L = q2L = q3L , where the equalities are enforced by o-diagonal weak interactions; ui uiR , di diR and i iL . We are working in the Yukawa-diagonal basis (so u3R becomes the right-handed component of the mass eigenstate t-quark after the electroweak phase transition, and so on). Then there are one or two more constraint equations, induced by the dimension-5 operator(s): case 1 : 2 + + = 0; 2 case 2 : 1 + + = 0; 1 case 3 : 3 + + = 0; 3 case 4 : 2 + + = 0; 3 case 5 : 1 + + = 0; 3 case 6 : 1 + + = 0; 2 2 1 1 + 3 + = 0; + 3 + = 0; + 2 + = 0. (12) (13) (14) (15) (16) (17)

So cases 1-3 have 15 constraint equations, while cases 4-6 have 16. With 28 chemical potential variables, this leaves 13 free s for cases 1-3, and 12 for cases 4-6. The free variables correspond to conserved quantities. The solution strategy is simply to (i) select the appropriate number of s (13 or 12) as independent variables, (ii) solve for the remaining chemical potentials in terms of them, (iii) identify the independent, conserved quantities for each case and write the associated chemical potential as a linear combination of the independent chemical potentials chosen at step (i). Finally, (iv) solve for the baryon/lepton and mirror baryon/lepton asymmetries at T 1010 GeV in terms of the conserved charges. Below 1010 GeV , other processes come into play in determining the nal, low temperature values of the baryon and mirror baryon asymmetries via a by now standard procedure (see below). We now examine each case dened by Eqs. (5-10) separately. In the main text, we shall write down the conserved charges and give the expressions for B, L, B and L at T 1010 GeV in terms of those charges. In the Appendix, we give the algebraic details for case 1 to illustrate the methodology.
A. Cases 1 & 2

Considering rst case 1, at T 1010 GeV there are 13 conserved charges:

1 L0 = 3 B L2 + 1 B L 2 3

L1 =

L3 , L2 = L3 = Le1R , L4 = Le2R , L5 = Bu1R Bd1R , L6 = Bd1R Bd2R ,

1 3B 1 3B

L1 ,

L 1 L 2

1 = 3 B L , 1 1 = 3 B L , 3 L3 = L 1R , e L = L 2R , 4 e L = Bu1R Bd1R , 5 L = Bd1R Bd2R . 6

(18)

The notation is: Li is the lepton number of family i, B (L = L1 + L2 + L3 ) is total baryon(lepton) number, BqiR is the charge for right-handed quark q from family i, and LeiR is the charge for the right-handed charged lepton of family i. The respective quantities from the mirror sector carry a prime. In terms of these quantities,
6 6

B = 0 L0 +
i=1 6

i Li +
i=1 6

L i , i
i L i . i=1

L = 0 L0 +
i=1

i Li +

(19)

5
TABLE I: The and coecients in the expansions of B and L for cases 1 & 2 as dened in Eqs. (19) and (20).
69 316 43263 98276 7050 24569 45189 98276 23385 98276 963 24569 963 49138 89 316 1 = 445 98276 534 24569 7743 98276 20381 98276 3649 24569 3649 49138

0 = 1 = 2 = 3 = 4 = 5 = 6 =

0 = = 1 = 2 = 3 = 4 = 5 = 6
345 98276 414 24569 6003 98276 15801 98276 2829 24569 2829 49138

1 = 2 = 3 = 4 = 5 = 6 =

55803 98276 16571 24569 31443 98276 59567 98276 5515 24569 5515 49138

2 =

3 =

4 =

5 =

6 =

TABLE II: The and coecients in the expansions of B and L for case 3.
12 79 735 1738 71 158 55 158 42 79 67 237 1 = 2 = 5 158 55 474

0 =

0 = = = 1 2 = = 3 4 = 5
112 869 12 869 4 79

1 = 2 = 3 = 4 = 5 =

1 = 2 = 3 = 4 = 5 =

3 = 4 =

46 869

88 237

5 =

70 237

6 = 5 /2

= /2 6 5

6 = 5 /2

6 = 5 /2

Under mirror symmetry, B B , L L , Li L (and L0 L0 ). Hence, i

6 6

B = 0 L0 +
i=1 6

i L i +
i=1 6

Li , i
i Li . i=1

L = 0 L0 +
i=1

i L i +

(20)

The values of the and parameters are given in Table I. Results for case 2 follow from case 1 with the replacements: L2 L1 , L L in L0 and L1 (L ) L2 (L ) in L1 2 1 1 2 (L ). Because both muon- and electron-Higgs Yukawa interactions are negligible above 1010 GeV , the , coecients 1 for case 2 are trivially the same as case 1 which were given in Table I.
B. Case 3

In this case, we again have 13 conserved charges, as in Eq. (18), but with L3 L2 (L L ) in L2 (L ), and 3 2 2 L2 L3 , L L in L0 . The and values are listed in Table II. 2 3

6
TABLE III: The and coecients in the expansions of B and L for cases 4 & 5 as dened in Eqs. (22) and (23).

D = 328321 0 = 53484 D 1 = 2 = 3 = 4 =
140874 D 165018 D 90276 D 48288 D 4 2

= 0 = 1 = 2 = 3 = 4

66132 D 4908 D

0 = 1 =

123061 D 177434 D 72793 D 164583 D 156188 D 4 2

0 =

85644 D 5105 D 58037 D 59919 D 126284 D

4 2

1 =

34188 D 14388 D 78192 D 4 2

2 = 3 = 4 =

2 =

3 =

4 =

5 =

= 5

5 =

5 =

C.

Cases 4 & 5

Considering rst case 4, at T 1010 GeV, there are 12 conserved charges:

1 1 L0 = 3 B L3 + 1 B L , L = 1 B L + 3 B L2 , 2 0 3 3 3 1 L1 = 1 B L1 , L = 3 B L , 1 1 3 L2 = Le1R , L = L 1R , 2 e L3 = Le2R , L = L 2R , 3 e L4 = Bu1R Bd1R , L = Bu1R Bd1R , 4 L5 = Bd1R Bd2R , L5 = Bd1R Bd2R .

(21)

In terms of these quantities,

5

B =
i=0 5

(i Li + L i ) , i
(i Li + i L i ) . i=0

L =

(22)

Under mirror symmetry, B B , L L , Li L . Hence, i

5

B =
i=0 5

(i L i + Li ) , i
(i L i + i Li ) . i=0

L =

(23)

The values of the and parameters are given in Table III. Results for case 5 follow from case 4 with the replacements: L L (L2 L1 ) in L0 (L ) and L1 L2 2 1 0 (L L ) in L1 (L ). Because both muon- and electron-Higgs Yukawa interactions are negligible above 1010 GeV , 1 2 1 the , coecients for case 5 are trivially the same as those of case 4 (given in Table III).
D. Case 6

This case is similar to case 4, except that L1 L3 (L L ) in L1 (L ) and L3 L1 (L L ) in L0 (L ). The 1 3 1 3 1 0 , coecients are given in Table IV

7
TABLE IV: The and coecients in the expansions of B and L for case 6. 0 = = 0 = 1
46 D 1633 12D 3519 12D 0 = 0 = 13617 12D2

D = 1343 1 =
362 D 4793 12D

D2 3D = 4029 1 =
2627 D2 25597 12D2

1 =

178 D2 6319 12D2

2 = 3 = 4 =

= = 2 3 = 4 = 5

2 = 3 = 4 = 5 =

2 = 3 =

637 3D

943 3D 4 2

4567 3D2 4 2

4 =

3649 3D2
4 2

5 =

4 2

5 =

IV.

RESULTS AND CONCLUSIONS

The values of the conserved charges, Li , L and L0 depend on the initial asymmetry generation mechanism. As in i Ref. [2], we consider, for deniteness, the simple case of non-zero B and/or L : B = X0 , L = Y0 , B = L = 0 (with L1 = L2 = L3 Y0 /3 and Bu1R = Bd1R = Bd2R ). The only nonzero conserved charges are then L = L = L0 = 1 2 L 1 1 (X Y0 ) Z 3 0 = L = L0 Z cases 4 6. 0 cases 1 3, (24)

After chemical processing, the baryon and mirror baryon asymmetries at T 1010 GeV are then B = Z(0 + + ), B = Z(0 + 1 + 2 ), cases 1 3 1 2 B = Z(0 + + ), B = Z(0 + + 1 ), cases 4 6. 0 1 0 The ordinary matter/dark matter ratio at that temperature is therefore B B B B B B B B 0 + + 1 2 0 + 1 + 2 0 + + 1 2 = 0 + 1 + 2 0 + + 0 1 = 0 + + 1 0 0 + + 0 1 = 0 + + 1 0 = 3795 0.25 cases 1 & 2, 15487 52 = 0.18 case 3, 289 20754 = 0.48 cases 4 & 5, 43415 1265 = 0.67 case 6. 1897 =

(25)

(26)

The value of this ratio changes to some extent at lower temperatures as dierent chemical processes become important. However, at temperatures near that of the electroweak phase transition, T = TEW 200 GeV, the values of B and B depend only on the values of B L and B L . These charges are separately conserved for T 1010 GeV, because the interactions in Eqs. (5-8) chemically connecting the ordinary and mirror sectors are slower than the expansion rate. This yields the well-known relation between B and B L [15], B= 28 (B L), 79 (27)

with an identical relation for B in terms of B L . Below the electroweak phase transition temperature, there are no processes fast enough to further aect B and B . Thus the nal, low temperature value of the ratio B/B = B / B is simply given (B L)/(B L ) evaluated at T 1010 GeV :
55 0 + + 0 1 2 B 1 2 = = 0.22, B 0 + 1 + 2 0 1 2 256

cases 1 & 2;

8
13 0 + + 0 1 2 B 1 2 = = 0.25, case 3; B 0 + 1 + 2 0 1 2 53 214 0 + + 0 0 1 B 0 1 = cases 4 & 5; + = 409 0.52, B 0 + 0 1 0 1 0 B 55 0 + + 0 0 1 0 1 = case 6. + = 98 0.56, B 0 + 0 1 0 1 0

(28)

If there is some brief period of ination between T 1010 GeV and TEW , as suggested by Step 4, then the results depend on when this second period of ination occurs, T = T3 , as well as the subsequent ordinary and mirror sector reheating temperatures, TRH and TRH respectively (with TRH < TRH required for successful big bang nucleosynthesis and large scale structure formation). At one extreme, if T3 TEW , then the results of Eq. (28) hold irrespective of the reheating temperatures. The other extreme case, T3 = 1010 GeV , allows three outcomes. If TRH > TEW and TRH > TEW , then Eq. (28) again holds. In the opposite situation, TRH < TEW and TRH < TEW , the nal values are simply given by Eq. (26), because no further reprocessing can take place. The acceptable intermediate situation, TRH > TEW and TRH < TEW , has further ordinary sector reprocessing but a frozen mirror sector. For this situation, the nal ratios are given by B B B B B B B B
28 0 + + 0 1 2 1 2 79 0 + 1 + 2 28 0 + + 0 1 2 1 2 = 79 0 + 1 + 2 28 0 + + 0 0 1 0 1 = 79 0 + + 1 0 28 0 + 0 + 0 0 1 1 = 79 0 + + 1 0

3080 0.20, cases 1 & 2; 15487 182 = 0.21, case 3; 867 1578892 = 0.46, cases 4 & 5; 3429785 440 = 0.54, case 6. 813 =

(29)

These results all reproduce the qualitative observation that there is more dark matter than ordinary matter. Intuitively this is because the FRW universe is born full of mirror matter (under our assumptions), and only some of the net mirror baryon/lepton number is chemically reprocessed into an ordinary baryon asymmetry. But it is also interesting that there is a subset of eective dimension-5 operators which are quantitatively successful, namely cases 1-3 [with case 3 marginal unless the circumstances leading to Eq. (29) obtain]. Given that the eective operators also contribute to the light neutrino mass matrix through ordinary-mirror neutrino mixing, a tentative connection between the dark matter problem and neutrino oscillation physics can be made. The connection must be tentative, because some important assumptions lie behind our results, especially the microphysical desert between the electroweak scale and physics at 101012 GeV (as emphasised in Ref. [2]). In conclusion, we have shown that the ratio of baryonic to non-baryonic dark matter, b /dark = 0.200.02, inferred by WMAP[1], can be quantitatively explained if mirror matter is identied with the non-baryonic dark matter. Our explanation involves a set of assumptions about the physics governing the early evolution of the Universe, which are not unique but are nevertheless plausible. Our approach also has important implications for neutrino physics, suggesting eV scale neutrino masses, which can be tested/constrained from upcoming neutrino experiments such as miniBooNE.
Acknowledgments

This work was supported by the Australian Research Council.

APPENDIX A: SOLVING THE CASE 1 EQUATIONS

We show the algebraic technique for solving case 1. The other cases follow similarly. The aim is to use Eqs. (11) and (12) to extract the total baryon number B, total lepton number L, and their mirror matter analogues B and L at T 1010 GeV . These 15 equations reduce the 28 variables, q , ui , di , i , ei , and their primed counterparts to 13 independent variables. The number of independent variables corresponds to the number of conserved charges, Eq. (18). The problem at hand is to nd B, L, B and L in terms of the conserved charges.

9 One systematic way of doing this is the following. First identify 13 independent variables. One possible choice is the following: q , 2 , 3 , e1 , e2 , d1 , d2 , , 3 , 1 , 2 , 1 and 2 . Then use Eqs. (11) and (12) to write q e e d d the 15 dependent variables in terms of the chosen independent variables. Doing this we have: 1 = 9q 2 3 1 (21q + 3d1 + 3d2 + e1 + e2 + 3 ) = 6 1 d3 = (27q 3d1 3d2 e1 e2 3 ) 6 1 e3 = (21q 3d1 3d2 e1 e2 + 53 ) 6 1 u3 = u2 = (15q + 3d1 + 3d2 + e1 + e2 + 3 ) 6 1 u1 = (39q 9d1 9d2 e1 e2 3 ) 6 2 = 2 + 1 = 9 2 + 3 q 1 = 21 + 3 1 + 3 2 + 1 + 2 + 3 e e d q d 6 1 3 = 27 3 1 3 2 1 2 3 d e e d q d 6 1 21 3 1 3 2 1 2 + 5 3 3 = e e d q d e 6 1 3 = 2 = 15 + 3 1 + 3 2 + 1 + 2 + 3 u u e e d q d 6 1 1 = (A1) 39 9 1 9 2 1 2 3 , u e e d q d 6 where it is understood that in the 1,2 equations, and are to be substituted with the respective righthand sides above. In terms of the chosen independent i , the baryon and lepton number are:
3

B = 6q +
i=1

(ui + di )

= 12q
3

L =
i=1

(2i + ei )

29 1 5 5 5 1 (A2) q d1 d2 + e1 + e2 + 3 2 2 2 6 6 6 The next step in the calculation is to write the conserved charges, Li in Eq. (18), in terms of the 13 independent variables. For example:2 = L1 = 1 B L1 3 1 3
3

= 2q +

(ui + di ) 21 e1
i=1

= 22q + 22 + 23 e1 , where Eq. (A1) has been used in the last step. The results for the other Li are given below: 1 1 1 2 1 4 L0 = 11q 42 e2 e1 3 d1 d2 3 + 1 + 2 + 1 2 + 3 d q d e e 3 3 3 3 3 3

(A3)

The conventional denition of, say, the baryon number of the universe is the ratio nB /s where nB is the net baryon number per unit volume, while s is entropy density. There is proportionality factor relating this denition of baryon number to the simple one convenient for our application (just the appropriate linear combination of chemical potentials).

10 L2 = L3 = L4 = L5 = L6 = L = 1 L = 2 L = 3 L = 4 L = 5 L = 6 1 (3q + 3d1 + 3d2 + e1 + e2 173 ) 6 e1 e2 1 (39q 15d1 9d2 e1 e2 3 ) 6 d1 d2 1 4 1 1 1 5 29 7q + d1 + d2 + e1 + e2 + 22 + 3 1 2 1 2 + 3 q d d e e 3 3 3 3 3 3 1 3 + 3 1 + 3 2 + 1 + 2 17 3 e e d q d 6 e1 2 e 1 39 15 1 9 2 1 2 3 e e d q d 6 d1 d2

(A4)

For the given set of values for the conserved quantities, Li and L , the baryon number can be found by solving the i identities:
6

B = 12q = 12 q + 0 2 + 0 3 + . . . 0 L0 +
i=1

(i Li + L ) . i i

(A5)

where it is understood that the L s are also functions of the 13 independent variables [using Eq.(A4)]. By equating coecients of each of the 13 independent variables a set of 13 simultaneous equations for the s results: 1 13 12 = 110 + 221 + 2 + 5 7 1 2 2 0 = 21 40 + 21 17 1 1 1 0 = 21 2 5 0 + 6 6 3 3 1 1 1 1 1 0 = 1 + 2 + 3 5 0 + 6 6 3 3 1 1 1 4 1 0 = 2 + 4 5 0 + 6 6 3 3 1 5 1 0 = 2 5 + 6 0 + 1 2 2 3 1 0 = 2 5 6 0 + 1 2 2 13 1 0 = 30 + 29 + + 1 2 2 2 5 5 17 1 1 0 = 0 + 3 3 1 6 2 6 5 1 4 1 1 0 = 0 + + 3 3 3 1 6 2 6 5 2 1 1 1 0 = 0 + + 4 3 3 1 6 2 6 5 1 5 0 = 0 + + 1 6 2 2 2 5 3 1 0 = 0 + 6 1 2 2 2 5 These 13 equations can easily be solved for the 13 s; the results are as displayed in table I.

(A6)

11 A similar procedure with L instead of B in Eq. (A5) yields results for the s.

[1] [2] [3] [4]

[5] [6]

[7]

[8] [9] [10] [11] [12] [13] [14] [15]

D. N. Spergel et al, (WMAP Collaboration) Astrophys. J. Suppl. 148, 175 (2003), and references therein for earlier work. R. Foot and R. R. Volkas, Phys. Rev. D68, 021304 (2003). L. Bento and Z. Berezhiani, hep-ph/0111116; see also, Phys. Rev. Lett. 87, 231304 (2001). T. D. Lee and C. N. Yang, Phys. Rev. 104, 256 (1956); I. Kobzarev, L. Okun and I. Pomeranchuk, Sov. J. Nucl. Phys. 3, 837 (1966); M. Pavsic, Int. J. Theor. Phys. 9, 229 (1974); S. I. Blinnikov and M. Yu. Khlopov, Sov. J. Nucl. Phys. 36, 472 (1982); Sov. Astron. 27, 371 (1983). R. Foot, H. Lew and R. R. Volkas, Phys. Lett. B272, 67 (1991). B. Holdom, Phys. Lett. B166, 196 (1985); S. L. Glashow, ibid. B167, 35 (1986); E. D. Carlson and S. L. Glashow, ibid. B193, 168 (1987); R. Foot and X.-G. He, ibid. B267, 509 (1991); S. N. Gninenko, ibid. B326, 317 (1994); R. Foot and S. Gninenko, ibid. B480, 171 (2000); A. Yu. Ignatiev and R. R. Volkas, ibid. B487, 294 (2000); R. Foot and R. R. Volkas, ibid. B 517, 13 (2001); R. Foot, A. Yu. Ignatiev and R. R. Volkas, ibid. B503, 355 (2001); R. Foot, Acta Phys. Polon. B32, 3133 (2001); R. Foot and T. L. Yoon, ibid. B33, 1979 (2002); R. Foot and S. Mitra, Phys. Rev. D66, 061301 (2002); Astropart. Phys. 19, 739 (2003); Phys. Lett. B558, 9 (2003); Phys. Lett. A315, 178 (2003); Phys. Rev. D68, 071901 (2003); Z. Silagadze, astro-ph/0311337; R. Foot, hep-ph/0308254; astro-ph/0309330. Z. Berezhiani, D. Comelli and F. L. Villante, Phys. Lett. B503, 362 (2001); A. Yu. Ignatiev and R. R. Volkas, Phys. Rev. D68, 023518 (2003); Z. Berezhiani, P. Ciarcelluti, D. Comelli and F. L. Villante, astro-ph/0312605; P. Ciarcelluti, astro-ph/0312607. E. W. Kolb, D. Seckel and M. S. Turner, Nature (London) 314, 415 (1985); H. M. Hodges, Phys. Rev. D47, 456 (1993); Z. G. Berezhiani, A. D. Dolgov and R. N. Mohapatra, Phys. Lett. B375, 26 (1996); V. Berezinsky and A. Vilenkin, Phys. Rev. D62, 083512 (2000). M. Fukugita and T. Yanagida, Phys. Lett. B174, 45 (1986). C. Athanassopoulos et al., Phys. Rev. C54, 2685 (1996); Phys. Rev. Lett. 81, 1774 (1998); A. Aguilar et al., Phys. Rev. D64, 112007 (2001); see also, G. Drexlin, Nucl. Phys. B (Proc. Suppl.) 118, 146 (2003). R. Foot, H. Lew and R. R. Volkas, Mod. Phys. Lett. A7, 2567 (1992); R. Foot, ibid. A9, 169 (1994); R. Foot and R. R. Volkas, Phys. Rev. D52, 6595 (1995). R. N. Mohapatra and X. Zhang, Phys. Rev. D45, 2699 (1992). V. A. Kuzmin, V. A. Rubakov and M. E. Shaposhnikov, Phys. Lett. 155B, 36 (1985). See W. Buchmller, hep-ph/0101102 for a review. u S. Yu. Khlebnikov and M. E. Shaposhnikov, Nucl. Phys. B308, 885 (1988). There are small higher-order corrections to this result, see S. Yu. Khlebnikov and M. E. Shaposhnikov, Phys. Lett. B387, 817 (1996); M. Laine and M. E. Shaposhnikov, Phys. Rev. D61, 117302 (2000).