Mass of the sigma resonance


One of the most hotly debated properties of the \sigma resonance is the exact determination of its mass. Difficulties arise from its broadness. Indeed, in \pi\pi scattering data this resonance appears with a very large peak that makes difficult a precise determination of the mass and, indeed,  a large body of data is needed to accomplish this. Initially, it was very difficult to accept the existence of this particle and, for some years, disappeared from particle listings of PDG. Recent papers, using Roy equation, proved without doubt the existence of this resonance and gave what appears the most precise determination of the mass and width so far (see here and here). This approach has been recently criticized (see here and, more recently, here) where is claimed that this approach currently underestimates the mass of the particle.

Due to such a situation, we prefer to consider another similar resonance, f0(980), whose mass is better determined giving

m_{f0(980)}=980\pm 10\ MeV.

Theoretically, we have built a full computation, starting with the spectrum of Yang-Mills theory, for the mass of all these resonances (see here). This paper does not use properly the mapping theorem but gives the right results. We have identified two kind of spectra (higher order spectra can also be obtained) giving

m_1(n)=1.198140235\cdot (2n+1)\sqrt{\sigma}


m_2(n,m)=1.198140235\cdot (2n+2m+2)\sqrt{\sigma}

being, as usual, \sigma the string tension, a parameter to be computed experimentally. So, one has the spectrum of the \sigma resonance and its excited states by simply taking m,n=0 giving



m_{\sigma^*}=2\cdot 1.198140235\sqrt{\sigma}.

So, taking \sqrt{\sigma}=410\ MeV we get easily m_{\sigma^*}=982\ MeV in close agreement with experiments, while m_\sigma=491\ MeV showing that, effectively, one has currently an underestimation of this quantity. With these values we will have from the width of the \sigma resonance the QCD constant \Lambda=285\ MeV (see here).

Finally, a derivation of \Lambda and string tension \sigma from other experimental data would be critical to obtain fixed all the constants of QCD, producing immediately a proper understanding of all physics about \sigma resonance.

QCD constants from sigma resonance


In our recent paper we were able to compute a relevant property of the \sigma resonance. We have obtained its decay width as (see here):

\Gamma_\sigma = \frac{6}{\pi}\frac{G^2_{NJL}}{4\pi\alpha_s} m_\sigma f_\pi^4\sqrt{1-\frac{4m_\pi^2}{m_\sigma^2}}

being G_{NJL}\approx 3.76\frac{4\pi\alpha_s}{\sigma}, \sigma the string tension that is one of the constants to be determined, and f_\pi\approx 93\ MeV the pion decay constant. From asymptotic freedom we know that


being n_f the number of flavors and \Lambda the scale where infrared physics sets in and is another constant to be computed. We have computed the mass of the \sigma resonance from the gluon propagator obtaining

m_\sigma\approx 1.198140235\sqrt{\sigma}

and we have all the theoretical data to compute \sqrt{\sigma} and \Lambda from experimental data. This can be accomplished using two main references (here and here) that give:

\sqrt{s_\sigma} = 441^{+16}_{-8}-i279^{+9}_{-14.5}\ MeV


\sqrt{s_\sigma} = 460^{+18}_{-19}-i255^{+17}_{-18}\ MeV

respectively. These produce the following values for QCD constants

\sqrt{\sigma}=368\ MeV\ \Lambda=255\ MeV


\sqrt{\sigma}=384\ MeV\ \Lambda=266\ MeV.

We have not evaluated the errors being this a back of envelope computation. A striking result is that the ratio of these constants is the same in both cases giving the pure number 1.44 that I am not able to explain.

These results, besides being truly consistent, are really striking making possible a deep understanding of QCD. I would appreciate any reference about these values and on their determination, both theoretical and experimental, from other processes in QCD. These values are critical indeed for strong interactions.

%d bloggers like this: