Observation of a narrow baryon resonance with positive strangeness formed in $K^{+}$Xe collisions

The charge-exchange reaction $K^{+}\mathrm{Xe}\to K^{0}p{\mathrm{Xe}}^{\prime }$ is investigated using the data of the DIANA experiment. The distribution of the $p{K}^0$ effective mass shows a prominent enhancement near 1538 MeV formed by nearly 80 events above the background, whose width is consistent with being entirely due to the experimental resolution. Under the selections based on a simulation of $K^{+}$Xe collisions, the statistical significance of the signal reaches 5.5$\sigma$. We interpret this observation as strong evidence for formation of a pentaquark baryon with positive strangeness, ${\Theta }^{+}\left(uudd\overline{s}\right)$, in the charge-exchange reaction $K^{+}n\to K^{0}p$ on a bound neutron. The mass of the ${\Theta }^{+}$ baryon is measured as $m\left({\Theta }^{+}\right)=1538\pm 2$ MeV. Using the ratio between the numbers of resonant and nonresonant charge-exchange events in the peak region, the intrinsic width of this baryon resonance is determined as $\Gamma \left({\Theta }^{+}\right)=0.34\pm 0.10$ MeV.

Figure 1

Incident $K^{+}$ momentum at collision point (a) and the opening angle between the $K_S^0$ and proton directions in the laboratory (b). The crosses show the simulated distributions that have been normalized to the number of live events (see Sec. 3).

Figure 2

Laboratory momenta of the $K^0$ (a) and proton (b) and the cosines of the $K^0$ (c) and proton (d) emission angles with respect to the beam. The crosses show the corresponding distributions of all simulated events that have been normalized to the number of measured events. Additionally shown by dots in (a) and (c) are the simulated spectra of unrescattered $K^0$ mesons, and in (b) and (d) are those of unrescattered protons.

Figure 3

The absolute laboratory momentum of the $p{K}^0$ system (a); the cosine of the $p{K}^0$ emission angle in the laboratory (b); and the transverse (c) and longitudinal (d) momenta of the $p{K}^0$ system. The crosses show the corresponding distributions of all simulated events that have been normalized to the number of live events. Depicted by dots are the simulated distributions of rescattering-free events.

Figure 4

In (a), the original $m\left(p{K}^0\right)$ distribution is fitted to a Gaussian plus a background function, obtained by scaling the simulated distribution by a factor which is a free parameter of the fit. The dashed line shows the null fit to the background function alone. Shown and fitted in (c) and (e) are the $p{K}^0$ effective-mass spectra under the selections ${\Theta }_K,{\Theta }_p<{100}^0$ and $p_L>120$ MeV, respectively (see Section 5). The selection $p_T<300$ MeV is additionally applied in the right-hand panels (b), (d), and (f).

Figure 5

The scatter plots in $m\left(p{K}^0\right)$ and $p\left(K^{+}\right)$ for all live (a) and simulated (b) events. Also shown in (a) is the fit to a two-dimensional Gaussian plus the background function. The ellipse in (a) is the 90% contour for the two-dimensional Gaussian.

Figure 6

Shown in (a) and (b) are the scatter plots in $m\left(p{K}^0\right)$ and $p\left(K^{+}\right)$ under the additional selections ${\Theta }_K,{\Theta }_p<{100}^0$ and $p_L>120$ MeV, respectively. Either scatter plot is fitted to a two-dimensional Gaussian plus a background function obtained by scaling the simulated distribution under similar selections (not shown). The ellipses are the 90% contours for the two-dimensional Gaussian.

Figure 7

Shown in (a) and (b) are the $p{K}^0$ effective-mass spectra under the selections ${\Theta }_K,{\Theta }_p<{100}^0$ and $p_L>120$ MeV plus the common selections $p_T<300$ MeV and $445<p\left(K^{+}\right)<535$ MeV. The signal and null fits are shown by the solid and dashed lines, respectively.

Figure 8

The simulated effective mass of the original $p{K}^0$ system prior to any intranuclear rescatterings, $m_0\left(p{K}^0\right)$, for the restricted fiducial volume $L\left(K^{+}\right)>520$ mm corresponding to the region of throughput measurements (a). The $K^0$ laboratory momentum is restricted to $p_K>140$ MeV which is the effective threshold for detecting $K_S^0\to {\pi }^{+}{\pi }^{-}$ decays in the scan, and the distribution is normalized to the total number of $K^{+}\mathrm{Xe}\to K_S^{0}X$ collisions with $K_S^0\to {\pi }^{+}{\pi }^{-}$ decays found by the scan in the aforementioned region $L\left(K^{+}\right)>520$ mm. Switching off Pauli blocking in the simulation and lifting the cut $p_K>140$ MeV results in the $m_0\left(p{K}^0\right)$ spectrum that is shown in (b). The open white corridor in the latter histogram depicts the mass region $1529<m_0\left(p{K}^0\right)<1547$ MeV.