Two-Dimensional Partial-Wave Analysis of Exclusive 190 GeV $\pi^- p$ Scattering into the $\pi^-\pi^-\pi^+$ Final State at COMPASS (CERN)}

The dynamics of strong interaction in the regime of low energies, i.e. large distances, is still not understood. Given its simplicity the non-relativistic simple quark model (SQM) describes successfully the observed hadronic spectra. QCD-inspired models, however, predict hadronic states where the gluonic content contributes to the hadron quantum numbers. These so-called hybrids cannot be explained within the SQM. A solid experimental proof of the existence of such systems would be the observation of spin-exotic states, with spin-parity quantum numbers, not allowed in the SQM. The study of mesons, the simplest hadrons, permits to gain insight into the realm of strong interaction where hadrons are the relevant degrees of freedom. The most promising spin-exotic meson candidate is the $\pi_1(1600)$, which was claimed in several experiments and in particular in data taken during a previous hadron campaign of the COMPASS experiment. The hadron spectroscopy program of the COMPASS experiment at CERN focuses on the investigation of the light-meson spectrum in order to enlighten this rarely understood regime of strong interaction. During the 2008 data taking an unprecedented statistical precision has been reached in peripheral interactions of 190 GeV/$c$ pions with a proton target leading to the $\pi^-\pi^-\pi^+$ final state. A spin-parity analysis in the kinematical region of the squared four-momentum transfer $0.1 \leq t' \leq 1.0$ GeV$^2/c^2$ was carried out based on a model of 88 partial waves up to a total angular momentum of 6. Besides the precise determination of properties of known resonances, a new axial-vector state, the $a_1(1420)$, was observed for the first time in a mass region where neither model nor lattice calculations predict mesons with this quantum numbers. Noteworthy is the very small intensity of this signal and that it only couples to the $f_0(980)$ isobar which is assumed to have a large strangeness content. The spin-exotic $\pi_1(1600)$ was observed albeit as a broad component dominated by non-resonant contributions. The simultaneous analysis in bins of the three-pion invariant mass and the squared four-momentum transfer $t'$ allowed to study resonant and non-resonant contributions in more detail. Of special interest is the Deck effect, a non-resonant production mechanism, leading to the same final state and assumed to contribute to the partial waves that also contain the $a_1(1260)$ and the spin-exotic $\pi_1(1600)$. The hybrid nature of the two established states $\pi(1800)$ and $\pi_2(1880)$ was investigated. In addition the $\pi_2(1880)$ was confirmed as a resonance separate from the $\pi_2(1670)$. For the first time a study of the $t'$ dependence on the level of partial waves was undertaken. These findings will serve as input for future models of light-meson production. The same spin-parity analysis was applied to a smaller data set taken in 2009 with a lead target. The aim was to compare the result with the 2004 analysis and to learn more about the production by a direct comparison with the 2008 proton data set. In order to improve the tracking close to the high-intensity hadron beam, five GEM (Gas Electron Multiplier) detectors with pixelised readout were developed and constructed for the 2008 data-taking. Similar detectors with strip readout and inactive areas around the beam axis were already installed in COMPASS and proved their high-rate capability in combination with a small interaction length. These new detectors replaced parts of the inner tracking system and were the backbone of the Small Angle Tracker (SAT) system during the 2008/2009 hadron campaign.