betaFeedback

WorksPhysics and Astronomy

Quantum-correlation of neutral charmed mesons at BESIII

Abstract

Quantum-correlated DDˉD\bar D pairs give direct access to the strong-phase differences between D0D^0 and Dˉ0\bar D^0 decays, which are essential inputs to the CKM angle γγ and charm mixing. The BESIII experiment has accumulated the largest data sample at the ψ(3770)ψ(3770) resonance, corresponding to an integrated luminosity of 20.3 fb−120.3\,\text{fb}^{-1}. The neutral DDˉD\bar D pairs produced at ψ(3770)ψ(3770) are in a CC-odd state. We report new and improved measurements of the strong-phase differences in D→KS,Lπ+π−D\rightarrow K_{S,L}π^+π^-, D→K+K−π+π−D\rightarrow K^+K^-π^+π^-, D→K−π+π+π−D\rightarrow K^-π^+π^+π^- and D→K−π+π0D\rightarrow K^-π^+π^0 decays, alongside a measurement of the CPCP-even fraction in D→K+K−π+π−D\rightarrow K^+K^-π^+π^-. Furthermore, we present a novel method that studies the DDˉD\bar D quantum correlations using data collected at centre-of-mass energies from 4.134.13 to 4.23 4.23\,GeV, where the DDˉD\bar D pairs are produced in CC-even or CC-odd states, according to the number of accompanying photons. This gives the first observation of CC-even DDˉD\bar D pairs and a new measurement of the strong-phase difference in D→K−π+D\rightarrow K^-π^+ decays.

Introduction

The Cabibbo-Kobayashi-Maskawa (CKM) angle γ≡arg⁡(−VudVub∗/VcdVcb∗)\gamma\equiv\arg(-V_{ud}V_{ub}^{*}/V_{cd}V_{cb}^{*}) is one of the least constrained angles and is the only angle of the Unitarity Triangle that can be measured in tree-level decays. Its direct determination, (62.8±2.6)∘(62.8 \pm2.6)^\circ [1], is therefore a benchmark for the indirect determinations that involve loop processes, and the comparison between them can probe contributions from new physics. Such a direct measurement uses tree-level B±→DK±B^{\pm} \to D K^{\pm} decays, in which DD is a superposition of D0D^{0} and Dˉ0\bar{D}^{0} that decays to a common final state ff. The interference depends on γ\gamma and the strong-phase difference between the D0→fD^{0} \to f and Dˉ0→f\bar{D}^{0} \to f amplitudes. The strong-phase differences also enter measurements of charm mixing and CPCP violation, such as the observation of a non-zero mass difference between the neutral charm-meson eigenstates [2].

Quantum-correlated DDˉD\bar{D} pairs give access to the strong-phase differences directly. Hereafter, DD without superscript is a neutral DD of unspecified flavour, and charge conjugation is implied. At the ψ(3770)\psi(3770), the neutral DDˉD\bar{D} pair is produced in a CC-odd state, so the determination of the quantum state of one DD meson fixes the state of the other. The BESIII experiment has recorded the world’s largest data sample, 20.3 fb−120.3\ \mathrm{fb}^{-1}, at the ψ(3770)\psi(3770) resonance. Above the D∗DˉD^{*}\bar{D} threshold, BESIII has collected a data sample of 7.13 fb−17.13\ \mathrm{fb}^{-1} at centre-of-mass energies of 4.134.13 to 4.234.23 GeV, where the DDˉD\bar{D} pairs come from D∗Dˉ(∗)D^{*}\bar{D}^{(*)} production and so are in both CC-even and CC-odd states. These data samples make BESIII a unique laboratory to study the strong-phase differences.

Strong-phase differences in D→KS,Lπ+π−D \to K_{S,L}\pi^{+}\pi^{-} decays

Updated model-independent binned measurement

The strong phase of D→KSπ+π−D \to K_{S}\pi^{+}\pi^{-} varies significantly across the Dalitz plot, which makes this decay the most sensitive DD mode for γ\gamma and the charm-mixing parameter measurements. The BESIII experiment has measured the strong-phase parameters (ci,si)(c_i,s_i) with a data sample of 7.93 fb−17.93\ \mathrm{fb}^{-1} in three binning schemes with 2×82 \times8 bins [3]. The D→KLπ+π−D \to K_{L}\pi^{+}\pi^{-} decay serves as both a signal and a tag mode in this study. The earlier study linked the (ci′,si′)(c_i',s_i') in D→KLπ+π−D \to K_{L}\pi^{+}\pi^{-} with (ci,si)(c_i,s_i) by introducing a difference from amplitude model prediction [4], which brings a weak model assumption here. In this analysis, a larger data sample is used and both results with and without this model assumption are presented, both provide the most precise measurement to date. In Figure 1 (a) and (b), the optimal binning scheme and the result are shown. With the study of B±→DK±B^{\pm} \to D K^{\pm}, D→KSπ+π−D \to K_{S}\pi^{+}\pi^{-} simulation samples that have negligible BB statistical uncertainty, the propagated uncertainty from the (ci,si)(c_i,s_i) measured from optimal binning scheme to γ\gamma is found to be 0.9∘0.9^\circ (1.5∘1.5^\circ) based on constrained (unconstrained) results. The first γ\gamma measurement with the upgraded LHCb detector [5] has already used the parameters without the model assumption, so its D→KSπ+π−D \to K_{S}\pi^{+}\pi^{-} input is free from amplitude-model dependence.

Figure 1. For D→KSπ+π−D \to K_S\pi^+\pi^- strong-phase analysis, from left to right: (a) Optimal binning scheme (m±2≡m2(KSπ±)m^2_{\pm} \equiv m^2(K_S\pi^{\pm})), and (b) (ci,si)(c_i,s_i) in this scheme from 7.93 fb−1^{-1}, without (red) and with (green) the model link to KLπ+π−K_L\pi^+\pi^-, and the model prediction (blue) [3]. (c) Strong-phase difference ϕ(z)\phi(z) and (d) optimal weight wopt(z)w^{\mathrm{opt}}(z), which define the per-event weights [6].

Unbinned measurement: determination of optimal Fourier weights

Binning averages the interference inside each bin, and the current binning schemes keep about 85%85\% of the statistical sensitivity to γ\gamma. The joint BESIII and LHCb analysis uses the optimal Fourier weights woptcos⁡[kϕ(z)]w^{\mathrm{opt}}\cos[k\phi(z)] and woptsin⁡[kϕ(z)]w^{\mathrm{opt}}\sin[k\phi(z)] [7]. Here, ϕ(z)\phi(z) is the strong-phase difference at a Dalitz plot point zz, and the optimal weights woptw^{\mathrm{opt}} include the amplitude ratio, efficiency and background effects. The Dalitz plot distribution of ϕ(z)\phi(z) and woptw^{\mathrm{opt}} are shown in Figure 1 (c) and (d) respectively. Both optimal weight and ϕ(z)\phi(z) are predicted from an amplitude model, however, the weighted integrals are determined from data. A poor model only dilutes the sensitivity but does not bias the result itself. At BESIII, these weighted integrals have been measured with 8 fb−1^{-1} D→KS,Lh+h−D \to K_{S,L}h^+h^- decays. LHCb can then use this information to measure the weighted B±→Dh±B^{\pm} \to Dh^{\pm} (h=K,πh=K,\pi) yields with 9 fb−1^{-1} of pppp collision data. A simultaneous fit with Fourier expansion k≤2k \leq2 for D→KSπ+π−D \to K_S\pi^+\pi^- and k≤1k \leq1 for D→KSK+K−D \to K_SK^+K^- gives γ=(71.3±5.0)∘\gamma=(71.3\pm5.0)^\circ [6] (Figure 2(a)), which is the most precise single measurement to date. Compared with the previous measurement that uses the same LHCb data [8], this measurement improves the statistical precision on γ\gamma by 5%, and reduces the uncertainty due to the BESIII input to 0.6°.

Figure 2. From left to right: (a) two-dimensional 68% and 95% confidence-level (CL) contours of (γ,rBDK)(\gamma,r_B^{DK}) from the joint BESIII and LHCb unbinned analysis [6]. (b) Measured (ci,si)(c_i,s_i) of D→K+K−π+π−D\to K^+K^-\pi^+\pi^- in the four bin pairs (black, 68% CL, statistical only) compared with the amplitude-model prediction (blue crosses) [10]. (Rf,δDf)(R_f,\delta_D^f), for decay modes ff of (c) D→K−π+π+π−D\to K^-\pi^+\pi^+\pi^- and (d) D→K−π+π0D\to K^-\pi^+\pi^0, from the global fit. The black lines show the earlier 2.93 fb−1^{-1} result [12].

First model-independent strong-phase measurement of DK+K−π+π−DK^+K^-\pi^+\pi^- decay

The D→K+K−π+π−D \to K^+K^-\pi^+\pi^- decay has been studied by the LHCb Collaboration and found to provide interesting sensitivity to γ\gamma [9]. Similar to the D→KSπ+π−D \to K_S\pi^+\pi^- decay, the strong-phase differences of D→K+K−π+π−D \to K^+K^-\pi^+\pi^- decay can be measured in phase-space bins. The binning has been done by defining: ΔδD≡arg⁡(AD0/AD‾0)\Delta\delta_D \equiv\arg(\mathcal{A}_{D^0}/\mathcal{A}_{\overline{D}^0}), rD≡∣AD0/AD‾0∣r_D \equiv|\mathcal{A}_{D^0}/\mathcal{A}_{\overline{D}^0}|, and therefore the five-dimensional phase space can be projected into a two-dimensional surface. With 20.3 fb−1^{-1} of ψ(3770)→DD‾\psi(3770)\to D\overline{D} data, the first model-independent measurement of the binned strong-phase difference (ci,si)(c_i,s_i) for D→K+K−π+π−D\to K^+K^-\pi^+\pi^- has been reported [10] (Figure 2(b)), this is consistent with the model predicted value. This analysis also gives the CPCP-even fraction F+=0.754±0.010±0.008F_+=0.754\pm0.010\pm0.008, where the first uncertainty is statistical and the second is systematic. For a BB sample with a size equivalent to LHCb Run 1 and Run 2, the contribution to γ\gamma is about 10°, as a reference the statistical uncertainty of the γ\gamma itself is 12.7°. The LHCb experiment reported its first model-independent binned analysis of B±→[K+K−π+π−]Dh±B^{\pm}\to[K^+K^-\pi^+\pi^-]_D h^{\pm} and B±→[π+π−π+π−]Dh±B^{\pm}\to[\pi^+\pi^-\pi^+\pi^-]_D h^{\pm} decays using this result, and gives γ=(53.9−8.9+9.5)∘\gamma=(53.9^{+9.5}_{-8.9})^\circ [11].

Coherence factors and strong-phase differences in DK±nπDK^{\pm}n\pi (n≥2n\geq2) decays

The D→K−π+π+π−D\to K^-\pi^+\pi^+\pi^- decay and D→K−π+π0D\to K^-\pi^+\pi^0 decay are an important category of charm decays and they provide unique sensitivity for γ\gamma measurement. These quasi-flavour decay modes 񟿿 combine both Cabibbo-favoured and doubly Cabibbo-suppressed amplitudes, and the interference can be measured as RfeiδDfR_f e^{i\delta_D^f}. Here, RfR_f is the coherence factor and δDf\delta_D^f is the strong-phase difference for decay mode ff. With 7.93 fb−1^{-1} of quantum-correlated DDˉD\bar{D} data, BESIII obtains RK3π=0.51±0.04R_{K3\pi}=0.51\pm0.04, δDK3π=(182−13+14)∘\delta_D^{K3\pi}=(182^{+14}_{-13})^\circ, RKππ0=0.75±0.03R_{K\pi\pi^0}=0.75\pm0.03 and δDKππ0=(209−8+7)∘\delta_D^{K\pi\pi^0}=(209^{+7}_{-8})^\circ [12]. The uncertainty on RK3πR_{K3\pi} is two to three times smaller than in the CLEO-c and earlier BESIII results (Figure 2(c,Figure 2d)).

The K−π+π+π−K^-\pi^+\pi^+\pi^- parameters are also measured in four phase-space bins, this is benefiting from the large local coherence effect up to 0.80−0.06+0.070.80^{+0.07}_{-0.06}, from the overlapping resonance contributions. This enables a stand-alone determination of γ\gamma. With the 9 fb−1^{-1} LHCb sample, its expected uncertainty is 8.5∘8.5^\circ, of which 3.5∘3.5^\circ comes from the BESIII inputs.

C-even DDˉD\bar{D} above threshold: a new source of correlated pairs

On and above the D∗DˉD^*\bar{D} threshold, DDˉD\bar{D} pairs are produced with photons and neutral pions. When these consist of nn photons or any number of the π0\pi^0, the DDˉD\bar{D} pairs are then expected to have an eigenvalue of C=(−1)n+1C=(-1)^{n+1} [13]. The BESIII experiment fully reconstructed both DD mesons and explored five different production mechanisms, using 7.13 fb−1^{-1} data sample at s=4.13\sqrt{s}=4.13–4.234.23 GeV, for the quantum-correlation studies. The productions of DDˉD\bar{D}, π0DDˉ\pi^0D\bar{D}, γγ/π0π0DDˉ\gamma\gamma/\pi^0\pi^0D\bar{D} with C=−1C=-1, and γDDˉ\gamma D\bar{D}, γπ0DDˉ\gamma\pi^0D\bar{D} with C=+1C=+1 are studied with the missing energy and mass. As shown in Figure 3, the efficiency-corrected yields match the prediction well, which leads to the first observation of the CC-even quantum-correlation in the e+e−→XDDˉe^+e^-\to XD\bar{D} processes [13, 14].

Figure 3. The ratios of efficiency-corrected yields observed in data to those expected in the absence of correlations and mixing for each DDˉ\mathit{D}\bar{\mathit{D}} final state originating from each production mechanism. The displayed errors include systematic uncertainties [13].

The strong-phase differences δDKπ\delta_D^{K\pi} can then be measured with these data set, giving δDKπ=(192.8−12.4−2.4+11.0+1.9)∘\delta_D^{K\pi}=(192.8^{+11.0+1.9}_{-12.4-2.4})^\circ, where the first uncertainty is statistical and the second is systematic [14]. This result agrees with the previous BESIII measurement [15], with a three times smaller systematic uncertainty, and the combination of the two measurements is δDKπ=(189.2−7.4−3.8+6.9+3.4)∘\delta_D^{K\pi}=(189.2^{+6.9+3.4}_{-7.4-3.8})^\circ. Moreover, the CC-even decay rates have a linear relation to the mixing parameters xx and yy, which gives a promising sensitivity for the time-integrated mixing measurements at e+e−e^+e^- colliders. The study with a predicted 1 ab−1^{-1} data at each of four energies from 4.01 to 4.26 GeV at a future super tau-charm facility gives σ(x)≈5.6×10−4\sigma(x)\approx5.6\times10^{-4} and σ(y)≈1.5×10−4\sigma(y)\approx1.5\times10^{-4} [13], comparable to the present world averages.

Summary

The BESIII experiment has provided important DD strong-phase inputs to CPCP-violation studies, most notably the determination of the CKM angle γ\gamma. Novel methods such as the unbinned strong-phase parametrisation and the exploitation of CC-even correlations in D∗Dˉ∗D^{*}\bar{D}^{*} data significantly improve both statistical and systematic uncertainties. A summary of the results quoted in this report is shown in Table 1. Future measurements with improved analysis methods and increased datasets are expected to push the precision further.

ModeObservableIntegrated luminosity (fb−1\mathrm{fb}^{-1})σγ\sigma_{\gamma}
D→KS,Lπ+π−D \to K_{\mathrm{S,L}}\pi^{+}\pi^{-}(ci,si)(c_i,s_i) in 3×(2×8)3 \times(2 \times8) bins7.930.7∘0.7^{\circ}–1.5∘1.5^{\circ}
D→KS,Lh+h−D \to K_{\mathrm{S,L}}h^{+}h^{-}Optimal Fourier weights8 (+LHCb 9)0.6∘0.6^{\circ}
D→K+K−π+π−D \to K^{+}K^{-}\pi^{+}\pi^{-}(ci,si)(c_i,s_i) in 2×42 \times4 bins, F+F_{+}20.3≈10∘\approx10^{\circ}
D→K−π+π+π−D \to K^{-}\pi^{+}\pi^{+}\pi^{-}RR, δD\delta_D (global and in 4 bins)7.933.5∘3.5^{\circ}
D→K−π+π0D \to K^{-}\pi^{+}\pi^{0}RR, δD\delta_D7.93–
D→K−π+D \to K^{-}\pi^{+}δD\delta_D7.13–

Table 1. Results reviewed here. σγ\sigma_{\gamma} is the contribution of the BESIII inputs to the uncertainty on γ\gamma, as estimated in each paper.

References

  1. [1]LHCb Collaboration, “Simultaneous determination of the CKM angle γ and parameters related to mixing and CP violation in the charm sector“, LHCb-CONF-2025-003, doi:10.17181/CERN.GFHE.LM6W.DOI
  2. [2]LHCb Collaboration, “Observation of the mass difference between neutral charm-meson eigenstates“, Phys. Rev. Lett. 127 (2021) 111801.DOI
  3. [3]BESIII Collaboration, “Updated model-independent measurement of the strong-phase differences between D⁰ and D̄⁰ → K⁰S/π⁺π⁻ decays,” JHEP 06 (2025) 086.
  4. [4]BESIII Collaboration, “Model-independent determination of the relative strong-phase difference between \(D^0\) and \(\bar D^0 \to K_S^0\pi^+\pi^-\) and its impact on the measurement of the CKM angle \(\gamma/\phi_3\),” Phys. Rev. D 101 (2020) 112002; BESIII Collaboration, “Determination of strong-phase parameters in \(D \to K_S^0\pi^+\pi^-\),” Phys. Rev. Lett. 124 (2020) 241802.
  5. [5]LHCb Collaboration, “Measurement of \(\gamma\) using \(B^\pm \to DK^\pm\) and \(B^\pm \to D\pi^\pm\) decays with \(D \to K_S^0\pi^+\pi^-\) and \(D \to K_S^0K^+K^-\),” JHEP 08 (2026) 037.
  6. [6]BESIII and LHCb Collaborations, “Measurement of the CKM angle \(\gamma\) in \(B^\pm \to D(\to K_S^0h'^+h'^-)h^\pm\) decays with a novel approach,” Phys. Rev. D (accepted), arXiv:2604.05701; BESIII and LHCb Collaborations, “Precise measurement of the CKM angle \(\gamma\) with a novel approach,” Phys. Rev. Lett. (accepted), arXiv:2604.05712.arxiv.org/abs/2604.05701
  7. [7]A. Poluektov, “Unbinned model-independent measurements with coherent admixtures of multibody neutral \(D\) meson decays,” Eur. Phys. J. C 78 (2018) 121.DOI
  8. [8]LHCb Collaboration, “Measurement of the CKM angle \(\gamma\) in \(B^\pm \to DK^\pm\) and \(B^\pm \to D\pi^\pm\) decays with \(D \to K_S^0h^+h^-\),” JHEP 02 (2021) 169.DOI
  9. [9]LHCb Collaboration, “A study of \(CP\) violation in the decays \(B^\pm \to [K^\mp K^\mp\pi^\pm\pi^\mp]_Dh^\pm\) (\(h=K,\pi\)) and \(B^\pm \to [\pi^+\pi^-\pi^+\pi^-]_Dh^\pm\),” Eur. Phys. J. C 83 (2023) 547.
  10. [10]BESIII Collaboration, “Measurement of the strong-phase difference between \(D^0\) and \(\bar D^0 \to K^+K^-\pi^+\pi^-\) in bins of phase space,” Phys. Rev. D 112 (2025) 012015.DOI
  11. [11]LHCb Collaboration, “A model-independent measurement of the CKM angle \(\gamma\) in the decays \(B^\pm \to [K^+K^-\pi^+\pi^-]_Dh^\pm\) and \(B^\pm \to [\pi^+\pi^-\pi^+\pi^-]_Dh^\pm\) (\(h=K,\pi\)),” JHEP 01 (2026) 062.
  12. [12]BESIII Collaboration, “Improved measurements of the coherence factors and strong-phase differences in \(D \to K^-\pi^+\pi^+\pi^-\) and \(D \to K^-\pi^+\pi^0\) with quantum-correlated \(D\bar D\) decays,” JHEP 08 (2026) 038.DOI
  13. [13]BESIII Collaboration, “First observation of quantum correlations in \(e^+e^- \to XD\bar D\) and \(C\)-even constrained \(D\bar D\) pairs,” Phys. Rev. Lett. 135 (2025) 171901.
  14. [14]BESIII Collaboration, “Novel measurement of the strong-phase difference between \(D^0 \to K^-\pi^+\) and \(\bar D^0 \to K^-\pi^+\) decays using \(C\)-even and \(C\)-odd quantum-correlated \(D\bar D\) pairs,” Phys. Rev. D 112 (2025) 072006.
  15. [15]BESIII Collaboration, “Improved measurement of the strong-phase difference \(\delta_D^{K\pi}\) in quantum-correlated \(D\bar D\) decays,” Eur. Phys. J. C 82 (2022) 1009.

Paper details

vlnc:W_mwv2nk7j
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 7

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents