News

Monday September 25th, 2006

First Observation of Particle-Antiparticle Oscillation of Bs0 Meson

Bs_box_diagram

Figure 1 : A diagram of the particle-antiparticle oscillation of Bs0.

Bs_flavor_tag

Figure 2 : Flavor tagging of Bs0 production.

phipipi_mass

Figure 3 : Invariant mass distribution reconstructed from Bs0 → Ds+(ϕπ+.

tbl_Bs_reco

Table 1 : Event numbers of  Bs0 in which the reconstruction succeeded.

all_unblind_ampscan_wsyst

Figure 4 : Angular frequency of the particle-antiparticle oscillation of Bs0 .

likelihood_widerange

Figure 5 : Likelihood ratio between the case with an assumption that oscillation occurred and not occurred.

asymmetry

Figure 6 : Time variation of amplitude obtained from the measured particle-antiparticle oscillation frequency.

Neutral B mesons undergo “particle-antiparticle oscillation” due to weak interactions that change the flavor of their constituent quarks. To explain in more detail, the mass eigenstates of the quarks and the eigenstates of the weak interaction Hamiltonian differ, causing oscillation between particles and antiparticles at an angular frequency of Δm = (m(BH0m(BL0)). Here, BH0 and BL0 represent the “heavy” and “light” mass eigenstates of the neutral B meson, respectively, and Δm is the mass difference. While particle-antiparticle oscillation of Bd0 (consisting of a (db) pair) had been observed previously, the particle-antiparticle oscillation of Bs0 (consisting of an (sb) pair) was observed for the first time in the CDF experiment, in which Osaka City University participates, and its angular frequency was measured. Viewing the Bs0 particle-antiparticle oscillation microscopically, within the meson, as shown in Figure 1, two W bosons are exchanged between the two quarks, causing the transformation from particle to antiparticle (box diagram).

In practice, to observe the oscillation, a Bs0 or its antiparticle Bs0 is first produced in a proton-antiproton collision. Since these particles are unstable, they decay after a certain time. By knowing whether it was a Bs0 or Bs0 at production, the proper time until decay, and whether it was a Bs0 or Bs0 at decay, we can determine if oscillation occurred and, if so, measure the oscillation frequency. Considering the production of Bs0, in proton-antiproton collisions, b (b) quarks are mainly produced in bb pair production, and each b and b quark independently hadronizes and decays. If a b quark combines with an s quark to form a Bs0, the s quark is produced in ss pair production, so the remaining s quark usually forms a K, which flies out in the same direction as the Bs0. Conversely, if a Bs0 is formed, it is accompanied by a K+. Therefore, by observing the charge sign of this K±, we can tag the initial state of the Bs meson (whether it was Bs0 or Bs0). This method is called “flavor tagging.” Besides the method mentioned above, there is another method for flavor tagging, which involves capturing the semileptonic decay of the opposite side b (b) quark in the initially produced bb pair. The charge of the lepton preserves the charge sign of the b quark before decay, allowing for flavor tagging (Figure 2).

Next, to identify the particle-antiparticle state when the Bs meson decays, we reconstruct the decay products of the Bs meson. The CDF detector is equipped with high-performance tracking detectors, calorimeters, time-of-flight detectors, and muon detectors, enabling the reconstruction of many decay modes from Bs0. Figure 3 shows the reconstruction result of one such decay mode, Bs0 → Ds+(ϕπ+)π. Including partially successful reconstructions, 8700 hadronic decays and 61500 semileptonic decays of Bs mesons were obtained, as shown in Table 1. Furthermore, the proper time until Bs meson decay can be calculated from the distance between the point of proton-antiproton collision and the point of reconstructed Bs meson decay, as well as the momentum of the Bs meson.

The data on the number of produced Bs mesons and the proper time until decay obtained in this manner represent the “waveform” of particle-antiparticle oscillation. Thus, by performing a Fourier transform considering decay attenuation on this waveform, if oscillation occurs at a constant angular frequency, a sharp peak should appear at that frequency. Figure 4 shows this, with a peak observed around 17 ps−1. Figure 5, although difficult to explain briefly, indicates the probability of an oscillation-like feature occurring due to “random fluctuations”. Near 17 ps−1, Λ drops to −15, corresponding to a probability of 5.7 × 10−7 for occurrence due to random fluctuations. From this, the existence of Bs0 meson particle-antiparticle oscillation is confirmed, with an angular frequency measured as

Δms = 17.77 ± 0.10(stat.) ± 0.07(sys.) ps−1.

Additionally, viewing this oscillation more straightforwardly, Figure 6 shows the amplitude at each time point by dividing the period obtained from the measured oscillation frequency into five. The experimental data neatly align with a cosine waveform, indicating oscillation between particle and antiparticle occurs about 3 trillion times per second.

This result of Bs0 particle-antiparticle oscillation yields important information on the transition probability between quark generations, specifically the value of the Cabibbo-Kobayashi-Maskawa matrix elements. Using the measured Δms, the following value is obtained:

|Vtd/Vts| = 0.2060 ± 0.0007(exp) +0.0081 (theory) .
−0.0060

This value records the highest measurement precision in the world as of September 2006. Another significant point is that the experimental measurement error is an order of magnitude smaller than the error from theoretical calculations. In this measurement, the experiment has surpassed the theory.