Rationalizing the baryon excess away

Before we turn to the pressing matter of explaining whence the baryon excess arises, we need to discuss its robustness to arguments that the universe can, despite the experimental evidence, be brought into matter-antimatter balance. It turns out that the baryon excess is quite robust.

Cosmic diffuse gamma rays

Cohen, de Rujula and Glashow (CDG) have explored a model in which matter and antimatter segregate into domains during the evolution of the universe [4]. Such a model has been offered as explanation for the strong evidence that there are no sizeable nearby accumulations of antimatter -- if the antimatter isn't nearby, then it must be far away. The CDG paper examines the observable effects of matter and antimatter domains of linear size $d_0$ and concludes that this size has to be approximately the Hubble radius to agree with observational evidence. Their argument is sketched in the following paragraphs.

If matter and antimatter domains indeed exist, then at some time since the last scattering of the CMB they must have been in contact. We can infer this from the isotropy of the CMB: if there had been voids between matter and antimatter domains, then these voids would be observable in the CMB.

If the domains have been in contact or are still in contact, then the nucleons in the matter domain should be annihilating the antinucleons in the antimatter domain. The main reaction is proton-antiproton annihilation to several pions, which further decay to photons or electrons:

$$p + \overline p \rightarrow \pi^0 (\rightarrow \gamma\gamma),$$

$$ \pi^\pm (\rightarrow e\nu_e\nu_\mu\overline {\nu}_\mu)$$

The annihilation products should lead to an observable effect. The annihilation electrons will Compton-scatter off CMB photons; they will also heat the interstellar medium. By both these mechanisms they will alter the CMB spectrum. The annihilation photons will contribute to the cosmic diffuse gamma radiation (CDG) reaching the Earth. A quantitative analysis shows that the effects due to the annihilation electrons do not lead to a strengthening of existing limits on the size of the putative antimatter domains. The prediction for the CDG contribution, however, leads to a very strong limit. Fig. 3 shows the observed CDG spectrum superimposed on the expected contribution from annihilation photons produced by 20 Mpc domains (top curve) and 1 Gpc domains (bottom curve). The data rules out domain sizes smaller than a Gpc. In effect, there is no space in our universe for even one of these antimatter domains.

Figure 3: CDG data and expected photon spectra from proton-antiproton annihilation. The top curve results from 20 Mpc domains, the bottom curve from 1 Gpc domains. The annihilation photons peak near the energy one would expect if all pions from the $p\overline p$ annihilation are produced at rest ($\sim$ 70 MeV); from that energy they are redshifted by a factor up to $\sim 1000$, leading to an energy cutoff of a few MeV.

Experimental search for antimatter

Despite the bleak CDG predictions for finding sizable antimatter accumulations, there are nevertheless experiments looking for antinuclei. These experiments need to be conducted outside the atmosphere because antimatter is copiously produced when cosmic rays collide with atmospheric nuclei; they are balloon-, satellite- or spacecraft-borne.

One of these experiments, the Anti-Matter Spectrometer (AMS), is scheduled to be installed on the International Space Station. Its principle of operation is as follows. A charged particle in a magnetic field will have a curved trajectory. By measuring the radius of the trajectory, the momentum $p = \gamma \beta m$ can be inferred. If the detector also records the energy loss of the particle (which is a well-understood function of $\beta$) or the time the particle takes to traverse the detector (``Time of flight'', TOF), which is proportional to $\beta^{-1}$, we can determine the mass of the particle. The direction in which the track bends in the magnetic field tells us the charge. If we find a particle with negative charge and mass $4m_p$ or greater, we have detected an antinucleus. Whether this antinucleus indicates that there are enough other antinuclei to balance the observed asymmetry is not clear, although observing for example an anticarbon nucleus would indicate the existence of antistars. Fig. 4 shows the results of the search for antihelium in an AMS test flight aboard the Space Shuttle; so far there is no evidence for antinuclei.

Figure 4: Results of an antihelium search in the AMS during a test flight aboard the Space Shuttle [5]. $\left \vert R\right \vert$ is the rigidity, $R = p / q$. Particles of mass $4m_p$ lie along the white band. Particles with $q < 0$ are indicated as black triangles. There are no antihelium candidates in this data set.