betaFeedback

WorksPhysics and Astronomy

Probing dark-matter scenarios of Belle II excess in B+→K+MXB^+\to K^+M_X with K+→π+MXK^+\to π^+ M_X decays

Abstract

The hypothesis of top-philic dark matter as the origin of the Belle II excess events in B→KMXB\to K M_X decays has direct impact on K→πMXK\to πM_X decays: Both of these decays proceed via loop diagrams~with a top quark emitting some mediator field RR and the subsequent decay of RR into a pair of dark-matter fermions χˉχ\barχχ. The Belle II results on B→KMXB\to K M_X allowed us to extract the parameters of the mediator RR with reasonable accuracy, whereas the sensitivity to the fermion mass mχm_χ proved to be rather low. In contrast to B→KMXB\to K M_X decays, the K→πMXK\to πM_X decays exhibit a strong sensitivity to mχm_χ; therefore, if - upon reducing theoretical and experimental uncertainties - the tension between standard-model predictions for the K→πMXK\toπM_X decays and the most recent NA62 measurements becomes pronounced the latter decays may either enable the extraction of the value of mχm_χ or provide a lower limit on~mχm_χ.

Introduction

In the Standard Model (SM), the missing-energy decay B+→K+MXB^{+} \to K^{+}M_{X} is effected by the decays B+→K+ννˉB^{+} \to K^{+}\nu\bar{\nu} with a pair of neutrino ν\nu and antineutrino νˉ\bar{\nu} in the final state. This flavour-changing neutral-current (FCNC) induced process is, in the SM, forbidden at tree level and proceeds via loops (penguins and boxes). For the resulting branching ratios, rather accurate theoretical predictions have been reported [1, 2, 3]:

Br(B+→K+ννˉ)=(4.44±0.30)×10−6,Br(B+→K∗+ννˉ)=(9.8±1.4)×10−6.(1)Br(B^{+} \to K^{+}\nu\bar{\nu}) = (4.44 \pm0.30) \times10^{-6}, \qquad Br(B^{+} \to K^{*+}\nu\bar{\nu}) = (9.8 \pm1.4) \times10^{-6}. \tag*{(1)}

In 2023, the Belle II collaboration reported [4] a surprising observation significantly exceeding the SM expectation, viz.,

Br(B+→K+ννˉ)=(2.3±0.7)×10−5=(5.4±1.5)Br(B+→K+ννˉ)SM,(2)Br(B^{+} \to K^{+}\nu\bar{\nu}) = (2.3 \pm0.7) \times10^{-5} = (5.4 \pm1.5)Br(B^{+} \to K^{+}\nu\bar{\nu})_{\mathrm{SM}}, \tag*{(2)}

as well as, in addition, the measurement of the differential distribution for the decay B→KMXB \to KM_{X}. If confirmed, these findings require the contribution of new physics beyond the SM. One of many new-physics scenarios for the B→KνˉνB \to K\bar{\nu}\nu enhancement is provided by the interpretation of the Belle II excess events as BB decays into invisible dark-matter (DM) particles [5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28]. Various specific scenarios have been considered, involving, for instance, a dark photon or decays into a DM fermion, χ\chi, and antifermion, χˉ\bar{\chi}, via some mediator field, in the following generically labelled RR. Taking into account the additional DM contribution, the B→KMXB \to KM_{X} decay width becomes

Γ(B→KMX)=Γ(B→Kνˉν)SM+Γ(B→RKχˉχ).(3)\Gamma(B \to KM_{X}) = \Gamma(B \to K\bar{\nu}\nu)_{\mathrm{SM}} + \Gamma(B \xrightarrow{R} K\bar{\chi}\chi). \tag*{(3)}

In the latter context, in our respective previous analyses [6, 14, 19, 20, 28] we studied B→K(∗)MXB \to K^{(*)}M_{X} decays within two specific top-philic DM scenarios defined by assuming the mediator field RR coupling both to the top quark tt and to a DM fermion-antifermion pair χχˉ\chi\bar{\chi}, for either a scalar mediator field RSR_{S} (labeled S-scenario) or a vector mediator field RVR_{V} (denoted V-scenario). For both scenarios, a good description of the Belle II excess data could be demonstrated and all relevant DM parameters (that is, the mediator mass MRM_{R}, the mediator decay width, and the couplings of the mediator to the top quark, gRttg_{Rtt}, and to the DM fermion, gRχχg_{R\chi\chi}) have been extracted.

Noteworthy, within top-philic scenarios the same mechanism contributes to any FCNC b→sb \to s and s→ds \to d transition, in particular, to the missing-energy decay K+→π+MXK^{+} \to\pi^{+}M_{X}, with the evident consequence for the latter’s decay width that

Γ(K→πMX)=Γ(K→πνˉν)SM+Γ(K→Rπχˉχ).(4)\Gamma(K \to\pi M_{X}) = \Gamma(K \to\pi\bar{\nu}\nu)_{\mathrm{SM}} + \Gamma(K \xrightarrow{R} \pi\bar{\chi}\chi). \tag*{(4)}

Both K+→π+νˉνK^{+} \to\pi^{+}\bar{\nu}\nu SM amplitude and corresponding differential distribution are known quite well (cf. e.g., the classical work [29] and its recent update [30]). For the additional contribution Γ(K+→Rπ+χˉχ)\Gamma(K^{+} \xrightarrow{R} \pi^{+}\bar{\chi}\chi), upon adopting the top-philic DM parameter values extracted from the successful description of the Belle II excess events in the B→KMXB \to KM_{X} decays, one readily obtains parameter-free predictions. We calculate the differential distribution of the decay K→RπχˉχK \xrightarrow{R} \pi\bar{\chi}\chi within both S- and V-scenarios and study further constraints on the DM sector, which may be obtained from a comparison of our theoretical predictions with the NA62 experimental results for K+→π+MXK^{+} \to\pi^{+}M_{X} decays.

Standard-Model and Dark-Matter Contributions to the K+→π+MXK^+ \to\pi^+ M_X Decay

K+→π+νˉνK^+ \to\pi^+ \bar{\nu}\nu Decays in the SM

The effective Hamiltonian HeffH_{\mathrm{eff}} describing the s→dνˉνs \to d\bar{\nu}\nu transition in the SM has been available for thirty years [29]:

Heff(s→dνˉν)=cLsˉγμ(1−γ5)d⋅νˉγμ(1−γ5)ν,(5)H_{\mathrm{eff}}(s \to d\bar{\nu}\nu) = c_L\bar{s}\gamma_\mu(1-\gamma_5)d \cdot\bar{\nu}\gamma^\mu(1-\gamma_5)\nu, \tag*{(5)}
cL=GF2αem2πsin⁡2θW∑l=e,μ,τ(Vcb∗VcsXl+Vtb∗VtsXt).(6)c_L = \frac{G_F}{\sqrt{2}}\frac{\alpha_{\mathrm{em}}}{2\pi\sin^2\theta_W}\sum_{l=e,\mu,\tau}\left(V_{cb}^*V_{cs}X^l+V_{tb}^*V_{ts}X_t\right). \tag*{(6)}

The corresponding differential K+→π+νˉνK^+ \to\pi^+\bar{\nu}\nu decay rate, with respect to the square q2q^2 of the difference of the involved meson momenta, q2=(pK+−pπ+)2q^2=(p_{K^+}-p_{\pi^+})^2, reads [31, 32]

dΓ(K+⟶SMπ+νˉν)dq2=cL232π3MK3λ3/2(MK2,Mπ2,q2)∣f+K→π(q2)∣2,(7)\frac{d\Gamma(K^+ \overset{\mathrm{SM}}{\longrightarrow} \pi^+\bar{\nu}\nu)}{dq^2}=\frac{c_L^2}{32\pi^3M_K^3}\lambda^{3/2}(M_K^2,M_{\pi}^2,q^2)\lvert f_+^{K\to\pi}(q^2)\rvert^2, \tag*{(7)}

where λ(a,b,c)=(a−b−c)2−4bc\lambda(a,b,c)=(a-b-c)^2-4bc. The analytic expressions for XlX^l and XtX_t can be found in Sec. XI of [29]. We use the numerical values of all the involved parameters from the comparatively recent analysis [30], which yield

cL=(3.1±0.14)×10−11 GeV−2,(8)c_L=(3.1\pm0.14)\times10^{-11}\,\mathrm{GeV}^{-2}, \tag*{(8)}

corresponding to

Br(K+→π+νˉν)=(7.71±0.61)×10−11.(9)Br(K^+\to\pi^+\bar{\nu}\nu)=(7.71\pm0.61)\times10^{-11}. \tag*{(9)}

The uncertainty in this estimate arises primarily from the uncertainty in cLc_L, since the form factor f+K→π(q2)f_+^{K\to\pi}(q^2) is known to a very good accuracy from lattice QCD [33]:

f+K→π(0)=0.9698±0.0017.(10)f_+^{K\to\pi}(0)=0.9698\pm0.0017. \tag*{(10)}

The uncertainty in the q2q^2-dependence of this form factor plays a minor role since the relevant range of q2q^2 is limited, 0<q2<(MK−Mπ)20<q^2<(M_K-M_{\pi})^2, and the maximal q2q^2 is much less than the mass of the lowest resonance in the q2q^2-channel, K∗K^*, that is, (MK−Mπ)2≪MK∗2(M_K-M_{\pi})^2\ll M_{K^*}^2.

K+→π+χˉχK^+ \to\pi^+\bar{\chi}\chi Decays in the DM S- and V-Scenarios

For the decay K+→π+MXK^+\to\pi^+M_X, we work out implications of the conjecture that the Belle II excess events actually arise from the decay into a pair of χˉχ\bar{\chi}\chi DM fermions that couple to the top quark via some mediator field RR. No interaction of the mediator RR with other SM particles is implied. In this case, both FCNC decays b→Rsχˉχb\overset{R}{\to}s\bar{\chi}\chi and s→Rdχˉχs\overset{R}{\to}d\bar{\chi}\chi proceed via the top–WW-boson loop depicted in Figure 1. Therefore, the amplitude of the K+→π+χˉχK^+\to\pi^+\bar{\chi}\chi decay is obtained from the amplitude of the B+→K+χˉχB^+\to K^+\bar{\chi}\chi decay by obvious replacements of the CKM factors, quark masses, and form factors.

Figure 1. Feynman graphs underlying the FCNC decays b→sχˉχb\to s\bar{\chi}\chi (a) and s→dχˉχs\to d\bar{\chi}\chi (b) in top-philic DM scenarios. The Wilson coefficients corresponding to Diagram (a) and Diagram (b) are related to each other by the simple replacements VtbVts→VtsVtdV_{tb}V_{ts}\to V_{ts}V_{td} of the CKM factors and b→db\to d of the involved quark fields.

S-Scenario

A rather popular model [34, 35], referred to as SS-scenario, involves the interaction of some DM fermions χ\chi with the top quark tt by exchange of a scalar mediator field RsR_s, governed by the interaction Lagrangian

Lint=−gRsttmtvRstˉt−gRsχχRsχˉχ;(11)\mathcal{L}_{\mathrm{int}}=-\frac{g_{R_s tt}m_t}{v}R_s\bar{t}t-g_{R_s\chi\chi}R_s\bar{\chi}\chi; \tag*{(11)}

here v≃246 GeVv\simeq246\ \mathrm{GeV} is the vacuum expectation value of the SM Higgs field. The emerging effective Lagrangian encoding the FCNC vertex b→sRsb\to sR_s then reads [34, 35]

Ls→dRs=gsdRsRsdˉLsR+h.c.,gsdRs=gRsttmsv32GFmt2Vtd∗Vts16π2.(12)\mathcal{L}_{s\to dR_s}=g_{sdR_s}R_s\bar{d}_Ls_R+\mathrm{h.c.},\qquad g_{sdR_s}=\frac{g_{R_s tt}m_s}{v}\frac{3\sqrt{2}G_Fm_t^2V_{td}^{*}V_{ts}}{16\pi^2}. \tag*{(12)}

As depicted in Figure 1(b), the K+→π+χχˉK^+\to\pi^+\chi\bar{\chi} transition proceeds via the mediator field RsR_s. One of the ingredients of the amplitude of interest, describing the K+→π+K^+\to\pi^+ transition and expressed in terms of a form factor f0K→π(q2)f_0^{K\to\pi}(q^2), assumes the form

⟨π+∣dˉLsR∣K+⟩=12⟨π+∣dˉs∣K+⟩=12MK2−Mπ2ms−mdf0K→π(q2).(13)\langle\pi^+|\bar{d}_Ls_R|K^+\rangle=\frac{1}{2}\langle\pi^+|\bar{d}s|K^+\rangle=\frac{1}{2}\frac{M_K^2-M_\pi^2}{m_s-m_d}f_0^{K\to\pi}(q^2). \tag*{(13)}

For the differential distribution in the SS-scenario, upon neglecting mdm_d compared with msm_s, we obtain [19]

dΓ(K+→Sπ+χχˉ)dq2=λ1/2(MK2,Mπ2,q2)128π3MK3(MK2−Mπ2)2∣f0K→π(q2)∣24ms2gsdRs2gRsχχ2(MRs2−q2)2+MRs2ΓRs2(q2)q2(1−4mχ2q2)3/2,(14)\frac{d\Gamma(K^+\xrightarrow{S}\pi^+\chi\bar{\chi})}{dq^2}=\frac{\lambda^{1/2}(M_K^2,M_\pi^2,q^2)}{128\pi^3M_K^3}\frac{(M_K^2-M_\pi^2)^2|f_0^{K\to\pi}(q^2)|^2}{4m_s^2}\frac{g_{sdR_s}^2g_{R_s\chi\chi}^2}{(M_{R_s}^2-q^2)^2+M_{R_s}^2\Gamma_{R_s}^2(q^2)}q^2\left(1-\frac{4m_\chi^2}{q^2}\right)^{3/2}, \tag*{(14)}

with ΓRs(q2)\Gamma_{R_s}(q^2) representing the q2q^2-dependent width of the mediator RsR_s [14, 19]. This ΓRs(q2)\Gamma_{R_s}(q^2) width can be inferred from the imaginary part of the fermion-loop Feynman diagram with scalar vertices (cf. [36, 37]) in the form of

ΓRs(q2)=(q2−4mχ2MRs2−4mχ2)3/2MRsq2Θ(q2−4mχ2)ΓRs0,ΓRs0=gRsχχ28πMRs(1−4mχ2MRs2)3/2.(15)\Gamma_{R_s}(q^2)=\left(\frac{q^2-4m_\chi^2}{M_{R_s}^2-4m_\chi^2}\right)^{3/2}\frac{M_{R_s}}{\sqrt{q^2}}\Theta(q^2-4m_\chi^2)\Gamma_{R_s}^0,\qquad\Gamma_{R_s}^0=\frac{g_{R_s\chi\chi}^2}{8\pi}M_{R_s}\left(1-\frac{4m_\chi^2}{M_{R_s}^2}\right)^{3/2}. \tag*{(15)}

We emphasize that the ss-quark mass msm_s in the denominator of Eq. (14) cancels the factor msm_s in gsdRsg_{sdR_s}, Eq. (12); it thus drops out from the decay rate. The above expressions involve four DM parameters: MRsM_{R_s}, mχm_\chi, gsdRsg_{sdR_s}, and gRsχχg_{R_s\chi\chi}. Below, we assume that MRs>2mχM_{R_s}>2m_\chi and, furthermore, that the mediator RsR_s decays predominantly into the χˉχ\bar{\chi}\chi pair.

V-Scenario

A widely discussed top-philic vector-mediator scenario, involving a vector mediator field RvμR_v^\mu that couples both to the top quark tt and to some DM fermions χ\chi, is defined by the Lagrangian [38, 39, 40]

Lint=gRvttRvμtˉγμ(1+γ5)t+gRvχχRvμχˉγμχ.(16)\mathcal{L}_{\mathrm{int}}=g_{R_vtt}R_v^\mu\bar{t}\gamma_\mu(1+\gamma_5)t+g_{R_v\chi\chi}R_v^\mu\bar{\chi}\gamma_\mu\chi. \tag*{(16)}

The coupling of the mediator RvμR_v^\mu to the s→ds\to d FCNC occurs by means of the penguin loop diagram shown in Figure 1(b). Integrating out the heavy top and WW fields leads to the s→dRvs\to dR_v effective Lagrangian [41]

Ls→dRv=gsdRvdˉγμ(1−γ5)sRvμ.(17)\mathcal{L}_{s\to dR_v}=g_{sdR_v}\bar{d}\gamma_\mu(1-\gamma_5)sR_v^\mu. \tag*{(17)}

The resulting Wilson coefficient gsdRv≡gˉsdRv(MW)g_{sdR_v}\equiv\bar{g}_{sdR_v}(M_W), where MWM_W denotes the mass of the WW boson in the penguin loop of Figure 1(b), has the form

gsdRv=−gRvtt16π2GF2VtsVtd∗mt2F(xt),xt=mt2MW2,F(xt)=7−8xt+xt2+2[4+(xt−2)xt]log⁡xt(1−xt)2=0.19.(18)g_{sdR_v}=-\frac{g_{R_vtt}}{16\pi^2}\frac{G_F}{\sqrt{2}}V_{ts}V_{td}^{*}m_t^2F(x_t),\qquad x_t=\frac{m_t^2}{M_W^2},\qquad F(x_t)=\frac{7-8x_t+x_t^2+2[4+(x_t-2)x_t]\log x_t}{(1-x_t)^2}=0.19. \tag*{(18)}

The associated differential distribution dΓ(K−→Vπ−χχˉ)/dq2d\Gamma(K^-\xrightarrow{V}\pi^-\chi\bar{\chi})/dq^2, involving the width ΓRv(q2)\Gamma_{R_v}(q^2) of the mediator RvR_v, reads [19]

dΓ(K−→Vπ−χχˉ)dq2=λ3/2(MK2,Mπ2,q2)192π3(1+2mχ2q2)1−4mχ2q2gsdRv2gRvχχ2(MRv2−q2)2+ΓRv2(q2)MRv2∣f+K→π(q2)∣2.(19)\frac{d\Gamma(K^-\xrightarrow{V}\pi^-\chi\bar{\chi})}{dq^2}=\frac{\lambda^{3/2}(M_K^2,M_\pi^2,q^2)}{192\pi^3}\left(1+\frac{2m_\chi^2}{q^2}\right)\sqrt{1-\frac{4m_\chi^2}{q^2}}\frac{g_{sdR_v}^2g_{R_v\chi\chi}^2}{(M_{R_v}^2-q^2)^2+\Gamma_{R_v}^2(q^2)M_{R_v}^2}\left|f_+^{K\to\pi}(q^2)\right|^2. \tag*{(19)}

Under the assumption that MRv>2mχM_{R_v} > 2m_\chi, the q2q^2-dependent decay width ΓRv(q2)\Gamma_{R_v}(q^2) can be read off from the imaginary part of the two-point Feynman diagram with fermions χ\chi in the loop and vectorial vertices [37]:

ΓRv(q2)=MRvq2q2−4mχ2MRv2−4mχ2q2+2mχ2MRv2+2mχ2Θ(q2−4mχ2)ΓRv0,ΓRv0=gRvχχ2MRv12π1−4mχ2MRv2(1+2mχ2MRv2).(20)\Gamma_{R_v}(q^2) = \frac{M_{R_v}}{\sqrt{q^2}} \sqrt{\frac{q^2 - 4m_\chi^2}{M_{R_v}^2 - 4m_\chi^2}} \frac{q^2 + 2m_\chi^2}{M_{R_v}^2 + 2m_\chi^2} \Theta(q^2 - 4m_\chi^2) \Gamma_{R_v}^{0}, \qquad\Gamma_{R_v}^{0} = \frac{g_{R_v\chi\chi}^2 M_{R_v}}{12\pi} \sqrt{1 - \frac{4m_\chi^2}{M_{R_v}^2}} \left(1 + \frac{2m_\chi^2}{M_{R_v}^2}\right). \tag*{(20)}

These expressions involve the four DM parameters MRvM_{R_v}, mχm_\chi, gsdRvg_{sdR_v}, and gRvχχg_{R_v\chi\chi}:

DM CONTRIBUTIONS TO THE OBSERVABLES

For most of the DM parameters of relevance for the present analysis, their values have been extracted in our earlier studies [14, 19, 20] by fitting the Belle II observation [4] of missing-energy excess events within both S- and V-scenarios. The only DM parameter yet to be pinned down is the mediator–quark coupling gsdRg_{sdR}, R=RsR = R_s or R=RvR = R_v, respectively.

  • For the S-scenario, the “best” parameter values of BB-decay origin, corresponding to the global minimum of χ2\chi^2, are

MRs=2.4±0.4 GeV,ΓRs0=2.9−0.9+1.1 GeV,mχ=0.42−0.4+0.2 GeV,gRsχχ≈6.08.(21)M_{R_s} = 2.4 \pm0.4\ \mathrm{GeV}, \qquad\Gamma_{R_s}^{0} = 2.9^{+1.1}_{-0.9}\ \mathrm{GeV}, \qquad m_\chi= 0.42^{+0.2}_{-0.4}\ \mathrm{GeV}, \qquad g_{R_s\chi\chi} \approx6.08. \tag*{(21)}

In addition, the S-scenario entails a simple relation between BB-decay relevant and KK-decay relevant parameters:

gsdRs=gbsRsVtdVtbmsmb.(22)g_{sdR_s} = g_{bsR_s} \frac{V_{td}}{V_{tb}} \frac{m_s}{m_b}. \tag*{(22)}

Recalling the earlier obtained result gbsRs≈5.6×10−8g_{bsR_s} \approx5.6 \times10^{-8} and inserting it in Eq. (22) yields for its sdRssdR_s counterpart

gsdRs≈2.4×10−11.(23)g_{sdR_s} \approx2.4 \times10^{-11}. \tag*{(23)}
  • For the V-scenario, proceeding in a manner similar to the S-scenario implies, for the BB-decay relevant parameters,

MRv≤3 GeV,ΓRv0=4.0−1.5+2.0 GeV,mχ=0.6−0.18+0.10 GeV,gRvχχ≈7.13.(24)M_{R_v} \le3\ \mathrm{GeV}, \qquad\Gamma_{R_v}^{0} = 4.0^{+2.0}_{-1.5}\ \mathrm{GeV}, \qquad m_\chi= 0.6^{+0.10}_{-0.18}\ \mathrm{GeV}, \qquad g_{R_v\chi\chi} \approx7.13. \tag*{(24)}

In the V-scenario, the relationship between BB-decay governing and KK-decay governing parameters is still simpler,

gsdRv=gbsRvVtdVtb,(25)g_{sdR_v} = g_{bsR_v} \frac{V_{td}}{V_{tb}}, \tag*{(25)}

and trivially enables us to deduce the value of gsdRvg_{sdR_v} from the previously extracted parameter gbsRv≈1.92×10−8g_{bsR_v} \approx1.92 \times10^{-8}:

gsdRv≈1.65×10−10.(26)g_{sdR_v} \approx1.65 \times10^{-10}. \tag*{(26)}

Figure 2 summarizes the differential distributions for the missing-energy decay K+→π+MXK^+ \to\pi^+ M_X as obtained within the SM and the S- and V-scenarios for the DM parameters from, respectively, Eqs. (21) and (23), or (24) and (26). We separately present the theoretically predicted differential distributions in (a) the SM vs the DM S-scenario and (b) the SM vs the DM V-scenario. Moreover, the binned experimentally measured differential distribution from Ref. [44] vs the SM contribution and the additional DM contribution are shown for both the S-scenario (c) and the V-scenario (d).

Figure 2. Glance at missing-energy decay K+→π+MXK^{+}\to\pi^{+}M_{X}: Confrontation of the differential distributions composed of SM and DM contributions (top row)—with dash-dotted green lines indicating the SM prediction and dash-dotted blue or red lines highlighting the additional contributions caused by the K→πχˉχK\to\pi\bar{\chi}\chi decay—and of the degree of agreement between the binned experimentally measured differential distribution—extracted from Fig. 8(a) of Ref. [44]—with the expectation arising from combining the SM and the additional DM contributions (bottom row), for both the DM S-scenario (left column) and the DM V-scenario (right column).

Various brief comments on our insights gained, and the conclusions inferred, from the outcomes of this investigation, collectively presented in Figure 2, are in order:

  • The B+→K+MXB^+ \to K^+M_X decays proved, as revealed in earlier studies [14, 19, 20], to be not particularly sensitive to the DM fermion mass mχm_\chi. For the S-scenario, practically any mass from the range 0<mχ<0.6 GeV0 < m_\chi< 0.6\ \mathrm{GeV} is admissible. For the V-scenario, the mass constraint was more stringent: 0.2−0.3 GeV<mχ<0.7−0.8 GeV0.2 - 0.3\ \mathrm{GeV} < m_\chi< 0.7 - 0.8\ \mathrm{GeV}. In contrast to this, the K+→π+MXK^+ \to\pi^+M_X decays turn out to be very sensitive to the precise value of mχm_\chi for the case of light DM fermions, satisfying mχ<(MK−Mπ)/2m_\chi< (M_K - M_\pi)/2, for which the decay K+→π+χχˉK^+ \to\pi^+\chi\bar{\chi} is kinematically allowed.

  • For the S-scenario, the maximal DM contribution to the K+→π+MXK^{+}\to\pi^{+}M_{X} differential distribution corresponds to a vanishing DM fermion mass, mχ=0m_{\chi}=0. There, it amounts to some 30 – 40% excess, more or less homogeneously smeared over the full allowed q2q^{2}-region. This result is well compatible with the recent NA62 measurement of the differential distributions in K+→π+MXK^{+}\to\pi^{+}M_{X} [42, 43, 44]. Thus, the DM contribution to K+→π+MXK^{+}\to\pi^{+}M_{X} in the S-scenario is well compatible with the DM parameters deduced earlier from the Belle results for the decay B+→K+MXB^{+}\to K^{+}M_{X}.

  • For the V-scenario, the emerging overall picture is quite different: There, a pronounced sensitivity of the excess events induced by some DM contribution in K+→π+MXK^{+}\to\pi^{+}M_{X} is observed. The region mχ<0.11m_{\chi}<0.11 GeV is practically excluded by the NA62 measurements, in excellent agreement with the constraints inferred from the Belle II data on B+→K+MXB^{+}\to K^{+}M_{X}. A highly interesting case is the region mχ>0.11−0.12m_{\chi}>0.11-0.12 GeV: There, one will note a strong enhancement of the differential distribution in the region q2>0.07q^{2}>0.07 GeV2^{2}. Optimists may see—or even interpret such behavior as—some indications of an enhancement of the experimental results compared to the SM predictions in this region. If this enhancement is established for greater statistics, an explanation of these findings within the DM V-scenario will be a viable possibility.

Summary

In conclusion, we analyzed the consequences, for the missing-energy decay K+→π+MXK^{+}\to\pi^{+}M_{X}, of the conjecture of some top-philic DM origin of the Belle II excess events observed for the decay B+→K+MXB^{+}\to K^{+}M_{X}. Within the framework of this conjecture, the branching ratio of the decay under study receives two contributions:

Br(K+→π+MX)=Br(K+→π+ννˉ)SM+Br(K+→Rχπ+χˉχ)DM.(27)Br(K^{+}\to\pi^{+}M_{X})=Br(K^{+}\to\pi^{+}\nu\bar{\nu})_{\mathrm{SM}}+Br\left(K^{+}\xrightarrow{R_{\chi}}\pi^{+}\bar{\chi}\chi\right)_{\mathrm{DM}}. \tag*{(27)}

Herein, the DM contribution involves precisely the same DM parameters as those determined previously [19] from the analysis of the decay B+→K+MXB^{+} \to K^{+}M_{X}. Therefore, the decay K+→π+MXK^{+} \to\pi^{+}M_{X} provides an unambiguous probe of the two top-philic scenarios employed.

The present experimental results reported by NA62 [44] are compatible within uncertainties with the SM predictions. Nevertheless, an optimistic eye may tend to see some excess events in the data. If (ultimately) established at a higher statistics, such excess events might be a useful additional probe of DM (as well as of any other new physics) scenarios.

Our conclusions may be summarized as follows:

  1. The mediator mass MRM_{R} was determined from the Belle II B+→K+MXB^{+} \to K^{+}M_{X} data with reasonable accuracy, whereas the DM fermion mass mχm_{\chi} extracted from these data had a larger uncertainty. The decay K+→π+MXK^{+} \to\pi^{+}M_{X}, however, provides a very promising possibility to probe the DM fermion mass mχm_{\chi} in the region mχ<(MK−Mπ)/2m_{\chi} < (M_{K}-M_{\pi})/2, where the decay K+→π+χχˉK^{+} \to\pi^{+}\chi\bar{\chi} is kinematically allowed: The differential distribution of K+→π+χχˉK^{+} \to\pi^{+}\chi\bar{\chi} is sensitive to the precise value of mχm_{\chi}, thus potentially providing constraints on mχm_{\chi}.

  1. For the S-scenario, the allowed range of mχm_{\chi} extracted from the Belle II data on B→KMXB \to KM_{X} essentially tolerates for the mass mχm_{\chi} any value in accordance with the bounds 0≤mχ<0.70 \leq m_{\chi} < 0.7 GeV. The maximal DM contribution, of branching ratio Br⁡(K+→Sπ+χχˉ)=2.83×10−11\operatorname{Br}(K^{+} \overset{S}{\to} \pi^{+}\chi\bar{\chi}) = 2.83 \times10^{-11}, is expected for mχ=0m_{\chi} = 0. Figure 2(a) provides the theoretical differential distributions for both the SM and the DM S-scenario, whereas Figure 2(c) shows the binned differential distribution measured by the NA62 experiment vs the SM and DM S-scenario contributions. The access events are distributed smoothly over the kinematically allowed range q2<(MK−Mπ)2q^{2} < (M_{K}-M_{\pi})^{2}. So, any smoothly distributed excess events would speak in favour of the S-scenario with a small value of the DM fermion mass parameter mχm_{\chi}.

  1. For the V-scenario, the allowed range of mχm_{\chi} demanded by the Belle II data on B→KMXB \to KM_{X} is mχ=0.6±0.2m_{\chi} = 0.6 \pm0.2 GeV. However, the region of smaller mχm_{\chi} is likewise not totally excluded. Figure 2(b) provides the theoretical differential distributions for both the SM and the DM V-scenario, and Figure 2(d) presents the binned differential distribution measured by the NA62 experiment vs the SM and DM V-scenario contributions. In this case, a rather particular DM contribution to the decay K+→π+MXK^{+} \to\pi^{+}M_{X} emerges for mχ>0.11m_{\chi} > 0.11 GeV: the contribution is peaked in the region q2≃0.04−0.07q^{2} \simeq0.04 - 0.07 GeV2^{2} and may be sufficiently large. So, if higher-statistics data establish excess events peaked at large q2q^{2}, this behavior will give a strong support to the V-scenario with DM fermion mass mχ≃0.11−0.15m_{\chi} \simeq0.11 - 0.15 GeV.

Acknowledgments

The research of A.B. and D.M. was carried out within the scientific program of the National Center for Physics and Mathematics, Section 5 Particle Physics and Cosmology. Stage 2026–2027.

References

  1. [1]D. Bečirević, G. Piazza, and O. Sumensari, Revisiting \(B \to K^{(*)}\nu\bar{\nu}\) decays in the Standard Model and beyond, Eur. Phys. J. C 83, 252 (2023).
  2. [2]W. G. Parrott, C. Bouchard, and C. T. H. Davies [HPQCD collaboration], Standard Model predictions for \(B \to K\ell^+\ell^-\), \(B \to K\ell_1^+\ell_2^-\), and \(B \to K\nu\bar{\nu}\) using form factors from \(N_f = 2 + 1 + 1\) lattice QCD, Phys. Rev. D 107, 014511 (2023); 107, 119903(E) (2023).
  3. [3]L. Allwicher, D. Bečirević, G. Piazza, S. Rosauro-Alcaraz, and O. Sumensari, Understanding the first measurement of \(B(B \to K\nu\nu)\), Phys. Lett. B 848, 138411 (2024).DOI
  4. [4]I. Adachi et al. (Belle-II Collaboration), Evidence for \(B^+ \to K^+\nu\nu\) decays, Phys. Rev. D 109, 112006 (2024).
  5. [5]M. Abdughani and Y. Reyimuaji, Constraining light dark matter and mediator with \(B \to K\nu\nu\) data, Phys. Rev. D 110, 055013 (2024).
  6. [6]A. Berezhnoy and D. Melikhov, \(B \to K^*M_X\) vs \(B \to KM_X\) as a probe of a scalar-mediator dark matter scenario, EPL 145, 14001 (2024).
  7. [7]A. Datta, D. Marfatia, and L. Mukherjee, \(B \to K\nu\bar{\nu}\), MiniBooNE and muon \(g-2\) anomalies from a dark sector, Phys. Rev. D 109, L031701 (2024).
  8. [8]L. Calibbi, T. Li, L. Mukherjee, and M. A. Schmidt, Is Dark Matter the origin of the \(B \to K\nu\nu\) excess at Belle II?, Phys. Rev. D 112, 075020 (2025). arXiv:2502.04900 [hep-ph].arxiv.org/abs/2502.04900
  9. [9]S.-Y. Ho, J. Kim, and P. Ko, Recent \(B \to K\nu\nu\) Excess and Muon \(g-2\) Illuminating Light Dark Sector with Higgs Portal, Phys. Rev. D 111, 055029 (2025).arxiv.org/abs/2401.10112
  10. [10]X.-G. He, X.-D. Ma, M. A. Schmidt, G. Valencia, and R. R. Volkas, Scalar dark matter explanation of the excess in the Belle II \(B \to K+\) invisible measurement, JHEP 07, 168 (2024).
  11. [11]B.-F. Hou, X.-Q. Li, M. Shen, Y.-D. Yang, and X.-B. Yuan, Deciphering the Belle II data on \(B \to K\nu\nu\) decay in the (dark) SMEFT with minimal flavour violation, JHEP 06, 172 (2024).
  12. [12]E. Gabrielli, L. Marzola, K. Muursepp, and M. Raidal, Explaining the B → Kνν excess via a massless dark photon, Eur. Phys. J. C 84, 460 (2024).
  13. [13]X. G. He, X. D. Ma, J. Tandean and G. Valencia, B → K+invisible, dark matter, and CP violation in hyperon decays, JHEP 07, 078 (2025).
  14. [14]A. Berezhnoy, W. Lucha, and D. Melikhov, Analysis of q2-distribution for B → KMX and B → K*MX in a scalar-mediator dark matter scenario, Phys. Rev. D 111, 075035 (2025).
  15. [15]L. Kloy and S. Nandi, Flavour and Electroweak Precision Constraints on a Simplified Dark Matter Model with a Light Spin-0 Mediator, arXiv:2503.15609 [hep-ph].
  16. [16]T. M. Aliev, A. Elpe, I. Turan and L. Selbuz, Explaining Belle Data on B → K(*)νν̄ Decays via Dark Photon Resonances, Phys. Rev. D 112, 015025 (2025).
  17. [17]K. Ding, Y. Li, X. Liu, Y. Liu, C. T. Lu and B. Zhu, Resonant ALP-Portal Dark Matter Annihilation as a Solution to the B± → K±νν̄ Excess, Phys. Rev. D 112, 115034 (2025).DOI
  18. [18]X. G. He, X. D. Ma, J. Tandean and G. Valencia, Light dark-matter window in K+ → π+ℓ̸, Phys. Rev. D 112, 055025 (2025).
  19. [19]A. Berezhnoy, W. Lucha and D. Melikhov, Probing vector- versus scalar-mediator dark-matter scenarios in B → (K, K*)MX decays, Eur. Phys. J. Plus 141, 495 (2026).DOI
  20. [20]A. Berezhnoy, W. Lucha, and D. Melikhov, Scrutinizing Dark-Matter Scenarios with B → (K, K*)νν̄ Decays, Universe 11 (2025) 12, 385.DOI
  21. [21]Z. K. Liu, Y. Li, B. F. Hou and Q. Chang, Search for light dark matter in rare meson decays, Phys. Rev. D 113, 035030 (2026).
  22. [22]J. Kim and P. Ko, B+ → K+νν̄ excess and DM semi-annihilation, Phys. Rev. D 113, 075023 (2026).DOI
  23. [23]F. Mescia, S. Okawa, J. Swallow and C. Toni, Dark matter emission at Belle II and NA62 in the minimal flavor violation framework, Phys. Rev. D 113, 095014 (2026).
  24. [24]P. D. Bolton, J. F. Kamenik and M. Novoa-Brunet, Dark light shining on B → K(*)Emiss, Phys. Rev. D 113, 115001 (2026).DOI
  25. [25]S. Jahedi, J. H. Liang, Y. Liao, X. D. Ma and Y. Uchida, Systematic study of lepton-flavor-violating dark matter interactions via indirect detection in effective field theories, Phys. Rev. D 114, 035007 (2026).
  26. [26]L. Gärtner, N. Krug, T. Kuhr, M. A. Schmidt, S. Stefkova and B. Yabsley, Constraints on invisible B+ → K+X decays from the Belle II B+ → K+νν̄ measurement, Phys. Rev. D 114, no.3, 032003 (2026).
  27. [27]S. Y. Ho, J. Kim and P. Ko, A Unified Dark Matter Explanation for B+ → K+νν̄ and the Super-Kamiokande Antineutrino Excess, [arXiv:2607.03206 [hep-ph]].
  28. [28]A. Berezhnoy, W. Lucha and D. Melikhov, B(s) → V MX decays as probes of dark-matter scenarios of Belle II enhancement in B → KMX decays, [arXiv:2609.13069 [hep-ph]].arxiv.org/abs/2609.13069
  29. [29]G. Buchalla, A. J. Buras and M. E. Lautenbacher, Weak Decays beyond Leading Logarithms, Rev. Mod. Phys. 68, 1125 (1996).DOI
  30. [30]J. Brod, M. Gorbahn and E. Stamou, Updated Standard Model Prediction for K → πνν̄ and εK, PoS BEAUTY2020, 056 (2021).
  31. [31]P. Colangelo, F. De Fazio, P. Santorelli and E. Scrimieri, Rare B → K(*) neutrino anti-neutrino decays at B factories, Phys. Lett. B 395 339 (1997).
  32. [32]D. Melikhov, N. Nikitin and S. Simula, Rare exclusive semileptonic b → s transitions in the standard model, Phys. Rev. D 57, 6814 (1998).
  33. [33]Y. Aoki et al. [Flavour Lattice Averaging Group (FLAG)], FLAG review 2024, Phys. Rev. D 113, 014508 (2026).
  34. [34]B. Batell, M. Pospelov, and A. Ritz, Multi-lepton Signatures of a Hidden Sector in Rare B Decays, Phys. Rev. D 83, 054005 (2011).DOI
  35. [35]K. Schmidt-Hoberg, F. Staub, and M. W. Winkler, Constraints on light mediators: confronting dark matter searches with B physics, Phys. Lett. B 727, 506 (2013).
  36. [36]G. J. Gounaris and J. J. Sakurai, Finite width corrections to the vector meson dominance prediction for ρ → e+e−, Phys. Rev. Lett. 21, 244 (1968).DOI
  37. [37]D. Melikhov, O. Nachtmann, V. Nikonov, and T. Paulos, Masses and couplings of vector mesons from the pion electromagnetic weak, and pi gamma transition form-factors, Eur. Phys. J. C 34, 345 (2004).
  38. [38]P. Langacker, The Physics of Heavy Z′ Gauge Bosons, Rev. Mod. Phys. 81, 1199 (2009).DOI
  39. [39]P. Cox, A. D. Medina, T. S. Ray and A. Spray, Novel collider and dark matter phenomenology of a top-philic Z′, JHEP 06, 10 (2016).DOI
  40. [40]Y. Hu, Y. Liu and Y. Liu, A study on vector mediator top-philic dark matter, Commun. Theor. Phys. 76, 085201 (2024).DOI
  41. [41]T. Inami and C. S. Lim, Effects of Superheavy Quarks and Leptons in Low-Energy Weak Processes KL → μμ̄, K+ → π+νν̄ and K0 ↔ K̄0, Prog. Theor. Phys. 65, 297 (1981) [erratum: Prog. Theor. Phys. 65, 1772 (1981)].
  42. [42]E. Cortina Gil et al. [NA62], Observation of the K+ → π+νν̄ decay and measurement of its branching ratio, JHEP 02, 191 (2025).
  43. [43]X. Chang [NA62], New measurement of K+ → π+νν̄ branching ratio at the NA62 experiment, [arXiv:2604.12649 [hep-ex]].arxiv.org/abs/2604.12649
  44. [44]S. Alibocus et al. [NA62], Measurement of the branching ratio of the K+ → π+νν̄ decay, [arXiv:2607.16413 [hep-ex]].

Paper details

vlnc:W_3s4edrau
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

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