MSSM Higgs results by D0 February 9, 2007Posted by dorigo in news, physics, science.
A few days ago D0 released their own result on the search for neutral Higgs boson decays to tau lepton pairs. This decay can only be seen at the Tevatron if the Higgs belong to a minimal SuperSymmetrical extension of the standard model, because only in that context its rate can be high enough.
The MSSM explains the spontaneous breaking of electroweak symmetry by introducing two complex doublets of Higgs bosons, leading to five observable particles: a pair of charged states H+, H–, and three neutral states h, H, and A. Two free parameters fully describe the spectrum of these particles: the mass of the A M(A), and the ratio of vacuum expectation values of the two doublets, tan(beta). If the A is not heavy and tan(beta) is large, the rate of Higgs bosons produced at the Tevatron is much larger than that of the regular neutral Higgs predicted by the Standard Model.
Neutral Higgs bosons in the MSSM like to decay to pairs of b-quarks or to pairs of tau leptons. However, the final state of two b-quarks is too hard to separate from backgrounds, which are dominated by strong interaction processes. Tau pairs are less problematic, because strong interactions do not produce leptons, and the largest background is then the decay Z->tau tau – which is manageable. It is for that reason that both CDF and D0 have been studying the latter final state.
Searching for neutral MSSM Higgs decays to tau-lepton pairs is still a tricky business. One needs to devise a suitable online selection of tau decays, and an optimized offline identification of taus. The tau is the heaviest of the three charged leptons, with a mass of 1.77 GeV. Because its mass is larger than that of many light hadrons, the tau can choose to decay to one of his lighter brothers (the electron or the muon), which are easy to see in the detector, or to a few light hadrons – which are harder to tag since they mimic the signature of a jet, produced at high rate by strong interactions.
D0 did all of that. They used a trigger that selects events with muons, which provide a possible tau->muon decay signature, and they devised a well-tuned neural network trained to distinguish hadronic decays of the tau from the much more common hadronic jets. Then, they studied the background processes that could contribute to events required to contain two tau candidates. Finally, they produced for each selected event an estimate of the mass that a hypothetical particle would have had to produce the two observed tau leptons with the measured energy, and set out to study the resulting mass distribution.
The mass distribution they obtain is shown in the plot above. You see the data points as black crosses, and the various backgrounds piled up with different colors. The largest of these is the already mentioned Z decay to tau pairs, but also strong-interaction processes (labeled QCD in the plot) do contribute with jet events where both jets fake the tau signature. Courtesy D0, you also get to see the expected Higgs contribution in grey, if the mass of the neutral higgses is 160 GeV. A response to the CDF excess ?
The agreement of data and expected backgrounds seems in fact to tell us that there is little space left for a Higgs decay, and a statistical analysis allows the extraction of limits on the existence of neutral MSSM higgs bosons as a function of the two parameters M(A), tan(beta) mentioned above.
The limits depend on a few assumptions on an additional parameter which I will not discuss here. An example is shown in the plot on the left, where you can see that D0 does exclude a large chunk of the parameter space (the blue region). Previous limits by CDF and D0 are shown with red and blue lines, while the LEP limits on the bottom of the plot are given in light green. The black line shows the limit D0 expected a priori to be able to place in the no-MSSM case, with the data analyzed and the chosen methodology.
It appears that D0 is excluding a region of the parameter space where CDF seemed to have seen a possible hint of a higgs signal. But is it really so ? I will address this question in the next post.