Search for CP-Violation in D→ π π π Decays

I have searched for a CP violating asymmetry in the neutral charm meson decay of D -> π 0 π 0 π 0 . I present here a brief description of CP violation and the CLEO II.V detector, the methods used to conduct this search, and the initial measured branching ratio of the studied decay, as well as the initial measured asymmetry of the final state of the decay. Background information: Symmetries and their conservation laws form, together with elementary particles and their interactions, the basis of the fundamental physical description of nature. Until approximately 1956 it was assumed that the laws of physics remained unchanged when one changes the sign of spatial coordinates (i.e. changing x, y, or z into –x, -y, or –z) in a given system. This mirroring is called the parity operation P. However, C.S. Wu observed that the mirror image of the left-handed neutrino, the right-handed neutrino, does not exist and that therefore the symmetry of the weak interaction is broken by the P operation [1]. The symmetry is restored when the P operator is not applied alone, but when the combined operation CP is applied. C, charge conjugation, is the operation that transforms a particle into its anti-particle. CP transformation changes a left-handed neutrino into a right-handed anti-neutrino, which does exist. In 1964, a symmetry violation of the CP-transformation was observed by James Cronin and Val Fitch in the case of the neutral K-meson at a level of 0.2% [2]. Since then precise measurements have taken place to determine the origin of CP violation in the Kmeson system. However, the K-meson effects due to the strong interaction are too large to draw any definitive conclusions about the origin of CP violation. Thus finding and studying CP violating decays of other mesons is important to better understand this phenomenon. An interesting side note on the subject of CP violation is that it is considered a leading idea in the explanation of the baryon asymmetry in the universe, that is, why our universe contains more matter than antimatter. The decay that I studied is of interest because no one has ever looked for it, meaning any results I find are the first evidence of this decay occurring. Another interesting aspect of this decay is due to the fact that the π 0 is its own antiparticle, so the decay of D (composed of cu quarks) to three π 0 s should happen exactly the same amount of time as the decay of D (composed of uc quarks) to three π 0 s. Thus, if CP violation occurred the branching ratio of D to three π0s would NOT equal the branching ratio of D to three π 0 s. The CLEO II.V detector: The Cleo II.V detector is used at the Cornell Electron Storage Ring (CESR). CESR is a ring in which an electron beam is circulated in one direction and a positron beam is circulated in the opposite direction. The beams are kept on a circular path by magnets. After reaching a certain energy, the electrons and positrons collide inside the CLEO detector and the energy caused by their annihilation creates different particles. The CLEO detector is a multipurpose high energy physics detector incorporating charged and neutral particle detection and measurement, used to analyze electron-positron collision events generated by CESR. The detector itself is about 6 meters on a side, containing about 900,000 kilograms of iron and over 25,000 individual detection elements [3]. A picture of a side cross-section of the detector is shown in Figure 1, with the electronpositron beam passing through the center of the detector in the plane of the figure. Figure 1: the CLEO II.V detector Methods: There are several steps to complete in order to look for CP violation in a decay mode. These steps are: 1. Writing (or modifying) a computer program that reconstructs the decays and extracts the data; 2. Using that program on Monte Carlo simulated data; 3. Using that program on real data taken from the CLEO detector; and 4. Comparing the branching ratio of the decay of the particle to the branching ratio of the decay of the antiparticle. I have completed the first three of these steps and have done initial calculations of the fourth step this summer, and I describe them in more detail below. Step one: Writing code in FORTRAN To begin, a computer program must be created that will collect the desired data and present it in a way that can be analyzed. I modified a program written in FORTRAN 77 to reconstruct the decays I was studying. Charge conjugation is implied throughout unless explicitly stated otherwise. In order to reconstruct these decays, I had to select photons that came from π 0 s. I placed certain requirements (or “cuts”) on what the code should accept as a photon. These cuts are designed to remove photons which may have come from particles similar to π 0 s, but which are not. One cut was that the mass of the particle formed by the di-photon combination had to be within 2.5 standard deviations from the nominal π 0 mass of 135 MeV/c [4]. Another cut required that one of the two photons from which the particles were built had to lie within the main section of the detector. The accepted pions were then kinematically combined to form the D candidates. Cuts are placed on the D later on, but not at this point because we want to test what sort of features all the D candidates exhibit and to know how many Ds are found. There is another step to the reconstruction process, namely finding out whether the π 0 s came from a D or a D. We do this by requiring the Ds to have occurred in the decay chain D-> D π slow . Measuring the charge of the slow pion produced in the decay will identify the flavor of the meson (either a D or a D). If the pion is a π+ we have a D, which came from a D. If the pion is a πwe have a D, which came from a D. The decay processes are shown in figure 2. Step two: Using code on Monte Carlo simulated data Before being able to use the code on real data from the CLEO detector, the code must be used on simulated data to test the efficiency of the code. First, a dataset is created where all decays are of the mode being studied; this is called “signal” Monte Carlo. Thus, if the code and the detector are both 100% efficient (which they aren’t), all the decays should be measured. Out of 1000 signal events created, my code measured 82 events, which means that the efficiency of my code is approximately 8%. This efficiency is used D D π π π π D D π π π π Figure 2: A D decays into a D and a pion, and the D then decays into three neutral pions. If CP symmetry is conserved, decays A and B should happen in equal amounts.