Proton decay
Proton decay refers to the hypothetical spontaneous decay of the proton, a fundamental baryon composed of two up quarks and one down quark, into lighter particles such as a positron and a neutral pion, violating the conservation of baryon number as predicted by the Standard Model of particle physics.[1] This process is a key prediction of grand unified theories (GUTs), which aim to unify the strong, weak, and electromagnetic fundamental forces into a single framework at high energies around 10^{16} GeV.[2] Despite extensive searches, no evidence for proton decay has been observed, placing stringent lower limits on the proton's partial lifetime, such as greater than 2.4 × 10^{34} years for the mode p → e⁺ π⁰ at 90% confidence level.[3] In theoretical models, proton decay arises from the exchange of heavy gauge bosons or other mechanisms that allow baryon number violation (ΔB = 1), with predicted lifetimes typically on the order of 10^{31} to 10^{36} years depending on the unification scale and model specifics, such as minimal SU(5) or supersymmetric extensions.[1] These theories, first proposed in the 1970s, resolve issues like the hierarchy of coupling constants observed at low energies and provide a natural explanation for the stability of matter, though the absence of detection challenges some minimal GUT variants.[2] Supersymmetric GUTs often favor alternative decay channels, like p → K⁺ ν̄ or p → μ⁺ K⁰, with branching ratios influenced by higher-dimensional operators suppressed by the Planck scale.[1] Experimental efforts to detect proton decay have primarily utilized large underground water Cherenkov detectors, which identify decay signatures through Cherenkov radiation from charged particles in a large volume of ultra-pure water.[1] The Super-Kamiokande experiment in Japan, operational since 1996 with a 50-kiloton fiducial volume, has set the world's most sensitive limits across multiple modes, including > 7.3 × 10^{33} years for p → μ⁺ η and > 4.5 × 10^{33} years for p → μ⁺ K⁰, based on exposures exceeding 0.40 megaton-years as of 2025.[4][3][5] Earlier experiments like IMB and Kamiokande contributed initial bounds in the 10^{32}-year range, but modern detectors have improved sensitivity by orders of magnitude through enhanced photon detection efficiency and background rejection techniques.[1] The non-observation of proton decay has profound implications for particle physics, constraining GUT parameters and motivating extensions like extra dimensions or string theory embeddings that suppress baryon violation.[2] If confirmed, it would confirm physics beyond the Standard Model and provide insights into the early universe, including mechanisms for baryogenesis that explain matter-antimatter asymmetry.[1] Future experiments, such as Hyper-Kamiokande with a 260-kiloton volume expected to begin operations in 2027 and the Deep Underground Neutrino Experiment (DUNE), aim to probe lifetimes up to 10^{35} years or beyond, potentially revolutionizing our understanding of unification.[1]Fundamentals
Definition and Hypotheses
In the Standard Model of particle physics, the proton is regarded as absolutely stable, as it is the lightest particle carrying nonzero baryon number B = +1, and baryon number along with lepton number L are conserved separately in all interactions. This conservation prevents the proton from decaying into any combination of lighter particles, such as leptons or mesons, without violating these quantum numbers. The stability of the proton underpins the long-term existence of ordinary matter, with no processes in the Standard Model allowing its spontaneous decay.[6] Proton decay emerges as a hypothesis in theories beyond the Standard Model, particularly grand unified theories (GUTs), which seek to unify the strong, weak, and electromagnetic forces under a single gauge symmetry at high energies. In these frameworks, baryon number B and lepton number L are not fundamental symmetries but arise from a deeper structure where they can mix, leading to processes that violate B and L individually while preserving \Delta(B - L) = 0. Such violations enable the proton, despite its stability in the Standard Model, to decay on extremely long timescales, potentially around $10^{34} years or longer depending on the unification scale. The conceptual foundation for baryon number violation traces back to Andrei Sakharov's 1967 proposal, where it was identified as one of three essential conditions—alongside C and CP violation and departure from thermal equilibrium—for successful baryogenesis to account for the observed cosmic matter-antimatter asymmetry. While Sakharov's work did not specify proton decay, it highlighted the necessity of B-violating interactions in the early universe. The first explicit prediction of proton instability came in 1974 with Howard Georgi and Sheldon Glashow's SU(5) GUT, where the unification of quarks and leptons into common representations naturally induces dimension-6 operators mediating proton decay. A prototypical decay mode in minimal GUTs is p \to e^+ + \pi^0, where the proton transforms into a positron (lepton, L = -1) and a neutral pion (meson, B = 0), resulting in \Delta B = -1 and \Delta L = -1, consistent with \Delta(B - L) = 0. This process conserves electric charge (+1 to +1 + 0), total energy (proton mass exceeds the sum of daughter particles' masses), and momentum, while releasing the excess energy primarily as kinetic energy of the products. Observation of such a decay would confirm B- L mixing and validate GUT-scale physics.Possible Decay Modes
Proton decay, if it occurs, must satisfy basic conservation laws while incorporating baryon number violation. The process conserves total electric charge (final state charge summing to +1), violates baryon number by ΔB = -1, and typically violates lepton number by ΔL = -1 in the dominant channels, though ΔL = 0 modes are possible but suppressed. Spin conservation requires the final state to match the proton's spin of 1/2, often achieved through s-wave decays, while parity is violated due to the underlying weak interaction effective operators. These constraints limit the kinematically allowed final states to low-mass particles, primarily mesons and leptons.[7] The primary hypothetical decay modes are semileptonic, involving a charged or neutral lepton and a meson. The most prominent and widely searched mode is p → e⁺ + π⁰, predicted to be dominant in minimal grand unified theories due to favorable matrix elements and phase space. Other key semileptonic channels include p → μ⁺ + K⁰, which involves a heavier kaon and thus reduced phase space, and the neutrino-involving mode p → ν̄ + π⁺, which is challenging to observe owing to the absence of charged tracks and reliance on missing energy signatures.[8][7] In the minimal SU(5) grand unified theory, theoretical calculations yield branching ratios of approximately 60% for p → e⁺ π⁰ and 30% for p → ν̄ π⁺, with the remainder distributed among modes like p → μ⁺ K⁰ (around 8%) and others suppressed by Cabibbo mixing or higher masses. These ratios stem from the unification of quarks and leptons under SU(5) symmetry, where the decay proceeds via dimension-6 operators mediated by heavy gauge bosons.[7] Non-semileptonic modes, such as p → π⁺ + π⁰, are theoretically allowed but occur at much lower rates, suppressed by the need for ΔL = 0 (or even ΔL = 2 in some cases) and requiring additional quark rearrangements that are disfavored in standard models. These channels typically arise from higher-dimensional operators and contribute less than 1% to the total decay width in minimal theories.[8] Kaon-involving modes like p → μ⁺ K⁰ play a crucial role in model discrimination, exhibiting higher branching ratios in SO(10) grand unified theories (up to 20-30%) compared to SU(5) (under 10%), due to the inclusion of right-handed currents and unified fermion representations in the 16-plet. Observation of enhanced kaon modes would thus favor SO(10)-like structures over simpler SU(5).[8][7]Theoretical Foundations
Grand Unified Theories
Grand unified theories (GUTs) propose to unify the strong, weak, and electromagnetic forces into a single gauge interaction at high energies, typically around $10^{15} to $10^{16} GeV, where the Standard Model symmetries emerge from the breaking of a larger gauge group.[7] In these models, quarks and leptons are placed in common multiplets of the unified group, naturally incorporating baryon number minus lepton number (B - L) as a conserved quantum number within the gauge structure. The simplest such model is the Georgi-Glashow SU(5) theory, based on the gauge group SU(5), where each generation of fermions resides in the \overline{5} + 10 representations, unifying the SU(3)_C \times SU(2)_L \times U(1)_Y of the Standard Model.[9] Larger groups like SO(10) extend this by accommodating all fermions of one generation, including a right-handed neutrino, in a single 16-dimensional spinor representation, while E_6 further unifies SO(10) \times U(1) into a 27 representation per generation, offering richer structure for fermion masses and mixings.[7] Proton decay arises in GUTs through the exchange of heavy gauge bosons at the unification scale, which mediate interactions violating baryon and lepton numbers. In the SU(5) model, the leptoquark gauge bosons X and Y (transforming as (3,2) and (\overline{3},2) under SU(3)_C \times SU(2)_L) couple quarks to leptons, inducing effective four-fermion operators with \Delta B = 1/3, \Delta L = 1/3 per quark-lepton vertex after integrating out these bosons at energies around $10^{15}–$10^{16} GeV.[7] These interactions lead to baryon number-violating processes like proton decay via dimension-6 operators, such as the dominant mode p \to e^+ \pi^0 in minimal SU(5). The unification scale M_{\rm GUT} is determined by the renormalization group evolution of the gauge couplings from low energies to the unification point, yielding M_{\rm GUT} \sim 10^{16} GeV in typical models.[7] In the minimal nonsupersymmetric SU(5) model, the predicted proton lifetime is \tau_p \sim 10^{31}–$10^{32} years for the p \to e^+ \pi^0 mode, based on the unification scale and coupling strengths, but this has been ruled out by experimental lower limits exceeding $10^{34} years.[7] To reconcile with observations, extensions incorporate supersymmetry, which alters the running of couplings and raises the effective scale via heavy superpartners, or use higher-dimensional representations for Higgs fields to suppress decay rates. Following the seminal 1974 Georgi-Glashow SU(5) proposal, GUT development evolved to include SO(10) models that naturally explain neutrino masses via the seesaw mechanism.[9] Later variants, such as the flipped SU(5) \times U(1) model proposed in 1982, modify particle assignments to avoid certain mass relations while preserving unification and proton decay predictions, and the Pati-Salam SU(4)_C \times SU(2)_L \times SU(2)_R model from 1974 unifies quarks and leptons through colored leptons, serving as an intermediate step toward full SO(10) unification.Baryon Number Violation
In the Standard Model of particle physics, baryon number B, which assigns +1 to quarks and -1 to antiquarks, is conserved as an accidental global U(1)_B symmetry. This conservation arises because the model's Lagrangian contains no operators that violate B at the perturbative level up to dimension four, and the symmetry is anomaly-free due to the cancellation of quantum anomalies within each generation of fermions; the original formulation lacked right-handed neutrinos, further ensuring this structure.[10] However, non-perturbative effects in the electroweak sector, mediated by sphaleron processes, do violate baryon number, though in a specific manner: these processes change B by \Delta B = 3 (for three generations) while preserving B - [L](/page/L'), where L is lepton number, resulting in \Delta (B + L) = 6.[11][12] In contrast, proton decay requires \Delta B = 1, which necessitates a violation of B - L by \Delta (B - L) = \pm 1 or \pm 2, depending on the decay mode; for instance, the mode p \to e^+ \pi^0 has \Delta L = -1 and \Delta (B - L) = 0, while p \to \bar{\nu} K^+ has \Delta L = -1 and \Delta (B - L) = 0.[13] Such \Delta B = 1 violations can occur through instanton effects in grand unified theories (GUTs) or string theory compactifications, where non-perturbative configurations generate effective operators at high energy scales, typically around $10^{15} GeV or above.[12][14] In minimal models predicting proton decay, these processes typically violate baryon and lepton numbers equally (\Delta B = -\Delta L), thereby preserving B - L. This distinction highlights why sphaleron-induced violations alone cannot mediate proton decay, as they maintain B - L invariance. The necessity of baryon number violation was first emphasized in the context of explaining the observed baryon asymmetry of the universe. In 1967, Andrei Sakharov outlined three conditions for baryogenesis: processes that violate baryon number, charge-parity (CP) violation, and departure from thermal equilibrium to allow an asymmetry to develop.[15] These conditions underscore the fundamental role of B violation in beyond-Standard-Model physics, extending beyond proton stability searches.Experimental Status
Historical Searches
The prediction of proton decay in grand unified theories following the 1974 proposal by Georgi and Glashow spurred the development of dedicated experimental searches in the late 1970s, as these models forecasted lifetimes around 10^{31} years, accessible with emerging large-scale detectors.[16] One of the earliest efforts was the Mont Blanc experiment (NUSEX), operational from 1973 to 1983 in the Mont Blanc tunnel at a depth of about 5,000 meters water equivalent (m.w.e.), utilizing an iron tracking calorimeter with 140 tons of iron and flash chambers for particle tracking. In 1983, the collaboration reported a candidate event interpreted as a proton decay in the mode p → μ⁺ K⁰ with an energy of approximately 0.8 TeV, generating initial excitement; however, subsequent analysis and lack of confirmation from other experiments led to its retraction by 1984 as likely background.[17][18] In the 1980s, the Kolar Gold Fields (KGF) experiment in India, located at 2,300 m underground, employed a 140-ton detector with iron plates and proportional counters to search for nucleon decay modes, setting early lower limits on the proton lifetime exceeding 10^{30} years based on null results over its operational period starting in 1980. Similarly, the Irvine-Michigan-Brookhaven (IMB) detector in the United States, a 3,300-ton water Cherenkov system with 2,048 photomultiplier tubes at 1,570 m.w.e. depth, began operations in 1982 and established initial limits of τ > 10^{30} years for the mode p → e⁺ π⁰ by the mid-1980s through observation of no decay candidates amid cosmic-ray-shielded data. These water Cherenkov techniques, which detect Cherenkov radiation from charged particles, marked a shift toward larger fiducial volumes on the order of 10^3 m³ to monitor vast numbers of nucleons while minimizing backgrounds from cosmic rays.[19][20] By the 1990s, the Soudan II experiment in Minnesota, USA, at 2,100 m.w.e. depth, utilized a 700-ton iron scintillator tracking calorimeter to probe various nucleon decay modes, including p → e⁺ π⁰ and p → ν K⁺, and contributed to tightening limits to around 10^{32} years without observing signals, further emphasizing the need for enhanced shielding and tracking resolution in underground facilities. These historical searches, driven by the expectation of short lifetimes near 10^{29} years, ultimately revealed much longer scales, prompting refinements in detector technologies like improved photomultiplier coverage and event reconstruction to distinguish rare decays from atmospheric neutrino backgrounds.[21]Current Limits and Experiments
The primary ongoing experiment searching for proton decay is Super-Kamiokande, a 50-kiloton water Cherenkov detector located in Japan and operational since 1996. This detector observes Cherenkov radiation produced by charged particles traversing the ultrapure water, allowing reconstruction of decay events through characteristic light patterns. For instance, the positron in the p → e⁺π⁰ mode produces a fuzzy ring due to electromagnetic showers, while the neutral pion decays into two photons that are identified via their conversion to electron-positron pairs, enabling kinematic reconstruction of the decay. As of 2025, Super-Kamiokande has accumulated over 450 kiloton-years of exposure without observing any proton decay candidates, setting stringent lower limits on the proton lifetime at 90% confidence level. The limit for the mode p → e⁺π⁰ is τ > 2.4 × 10^{34} years, while for p → νK⁺ it is τ > 5.9 × 10^{33} years.[22][23] These bounds are derived using Poisson statistics assuming zero signal events, accounting for backgrounds primarily from atmospheric neutrino interactions, with the effective exposure corresponding to approximately 3 × 10^{31} nucleon-years. Recent analyses in 2024–2025 have refined background modeling for neutrino-induced events, further tightening these limits without evidence of signals. Other current or recently active experiments contribute supplementary limits, though less stringent than Super-Kamiokande's. The KamLAND liquid scintillator detector in Japan has searched for p → νK⁺ using 8.97 kiloton-years of data, establishing a limit of τ > 5.4 × 10^{32} years at 90% confidence level, limited by its smaller fiducial volume and higher backgrounds from scintillator alpha decays. Borexino, which ceased data-taking in 2021 after operating in Italy's Gran Sasso laboratory, analyzed its full dataset for invisible nucleon decay modes, setting limits such as τ > 2.8 × 10^{29} years for nn → invisible at 90% confidence, but its proton decay sensitivities were minor compared to water-based detectors due to quenching effects in scintillator. Similarly, the SNO+ experiment in Canada, using linear alkylbenzene scintillator, has provided limits on invisible modes like nn → invisible at τ > 1.3 × 10^{28} years from early water-phase data, with ongoing tellurium-loaded phases focusing more on neutrinoless double-beta decay but contributing to baryon-number-violating searches.[24] The Jiangmen Underground Neutrino Observatory (JUNO) in China, a 20 kiloton liquid scintillator detector, began data taking in August 2025 and is expected to search for proton decay modes such as p → ν̄ K⁺, with projected sensitivity of τ / B(p → ν̄ K⁺) > 9.6 × 10^{33} years after 10 years of operation, benefiting from high light yield for kaon identification and background mitigation.[25] In the United States, the ProtoDUNE liquid argon time projection chamber prototype at CERN, tested in 2018–2020, has validated tracking and particle identification capabilities informing the Deep Underground Neutrino Experiment (DUNE), particularly for low-energy kaon reconstruction in proton decay channels, though full DUNE operations are not yet online. These efforts collectively emphasize zero-event Poisson limits and background rejection via particle identification, with no confirmed proton decay signals reported as of 2025.Future Detectors
Several next-generation experiments are under construction or in advanced planning stages to extend the search for proton decay to lifetimes approaching or exceeding predictions from grand unified theories, typically in the range of $10^{34} to $10^{36} years. These detectors aim to achieve this through significantly larger fiducial volumes, enhanced reconstruction capabilities, and reduced backgrounds compared to current facilities.[26][27] Hyper-Kamiokande, located in Japan, is a water Cherenkov detector with a 260 kiloton fiducial volume, approximately ten times larger than that of Super-Kamiokande; cavern excavation was completed in July 2025 and it is scheduled to begin operations around 2027. It is projected to reach a sensitivity of \tau \sim 10^{35} years for the p \to \pi^0 e^+ mode and \tau \sim 3 \times 10^{34} years for p \to \bar{\nu} K^+ after 20 years of data collection, benefiting from improved \pi^0 reconstruction due to higher granularity in photon detection.[26][28][29][30] The Deep Underground Neutrino Experiment (DUNE) in the United States features a 40 kiloton liquid argon time projection chamber (TPC) detector, with full operations expected around 2030, offering superior tracking resolution for identifying decay products like kaons. Projections indicate sensitivities exceeding \tau > 10^{34} to $10^{35} years for key modes, leveraging the TPC's ability to reconstruct low-energy events with high precision.[31][32][33] Other proposed or developing facilities include the European Spallation Source Neutrino Super Beam (ESSnuSB), a proposed long-baseline experiment with large water Cherenkov far detectors, offers strong potential for proton decay searches by combining high beam intensity with underground shielding to suppress cosmogenic backgrounds.[34][27] Technological advancements enabling these sensitivities include scaled-up detector volumes for increased event rates, improved photon detection systems such as ARAPUCA devices in DUNE for efficient light collection in liquid argon, and machine learning algorithms for event classification to distinguish rare signals from backgrounds. These experiments also hold potential for novel searches, such as dark matter-induced proton decays, as outlined in 2025 theoretical proposals that explore baryon number violation mediated by dark sector particles.[33][32][35]Implications
Baryogenesis and Matter-Antimatter Asymmetry
In 1967, Andrei Sakharov outlined three fundamental conditions necessary for baryogenesis: baryon number (B) violation, charge conjugation (C) and charge-parity (CP) violation, and departure from thermal equilibrium.[36] These conditions must be met to generate the observed baryon asymmetry parameter η ≈ 6 × 10^{-10}, which quantifies the matter-antimatter imbalance in the universe. Proton decay, as a process that violates B by ΔB = 1, provides a low-energy manifestation of the required B violation, linking high-scale physics to the cosmic asymmetry.[37] Grand unified theories (GUTs) offer a natural framework for baryogenesis through the out-of-equilibrium decays of heavy gauge bosons, such as X and Y bosons, at energies around 10^{15}-10^{16} GeV. These decays produce a primordial baryon asymmetry of order η ~ 10^{-10} via CP-violating interactions, satisfying Sakharov's conditions during the early universe's rapid expansion. However, electroweak sphaleron processes, which violate B + L but conserve B - L, would erase this asymmetry unless B - L is preserved in the model, as in SO(10) GUTs where leptoquarks couple to both baryons and leptons.[37][38] As an alternative to direct GUT baryogenesis, leptogenesis generates a lepton asymmetry (L) in the decays of heavy right-handed neutrinos in the seesaw mechanism, which sphalerons then partially convert to a baryon asymmetry with η_B ≈ (28/79) η_L. This process occurs at scales ~10^{9}-10^{15} GeV and conserves B - L, making it robust against sphaleron washout. Proton decay serves as an indirect probe of B - L stability in these models, as unobserved decay modes (e.g., p → e^+ π^0) impose lower limits on the unification scale, testing whether B - L violating operators are sufficiently suppressed.[37][39] The experimental lower limits on the proton lifetime, exceeding 10^{34} years for key modes like p → e^+ π^0, constrain baryogenesis models by requiring suppression of B-violating operators at low energies, often necessitating fine-tuning of GUT parameters or additional symmetries like R-parity in supersymmetric extensions. This suppression implies that primordial asymmetries generated at the GUT scale must be protected from dilution, demanding precise balancing of decay rates and couplings.[37] As of 2025, electroweak baryogenesis within the Standard Model remains insufficient to produce the observed η due to the weak first-order electroweak phase transition driven by the 125 GeV Higgs mass, which fails to isolate baryon-violating sphalerons from equilibrium. Recent developments favor high-scale mechanisms, including GUT baryogenesis.Beyond-Standard-Model Physics
Proton decay serves as a critical probe for physics beyond the Standard Model, particularly in extensions that address shortcomings of the minimal Standard Model, such as the lack of baryon number violation and unification of forces. In supersymmetric grand unified theories (SUSY GUTs), such as SUSY SU(5) and SO(10), proton decay arises primarily through dimension-5 operators mediated by sfermion exchanges in the superpotential, which integrate out heavy colored Higgsinos and gauginos.[40] These operators can enhance decay rates compared to dimension-6 contributions in non-supersymmetric models, but the rates are suppressed by the masses of Higgsinos and gauginos, typically pushing predicted lifetimes beyond 10^{34} years to evade current experimental limits.[41] In SUSY SU(5), the dominant mode is often p → \bar{\nu} K^+, while SO(10) models incorporate additional structure from right-handed neutrinos, linking decay signatures to fermion mass patterns.[40] Beyond SUSY GUTs, other beyond-Standard-Model frameworks predict proton decay through distinct mechanisms. In left-right symmetric models, baryon number violation with ΔB=1 can occur via exchanges of right-handed W_R bosons coupled to leptoquark scalars, enabling modes like p → e^+ π^0 or three-lepton final states, with lifetimes depending on the right-handed scale around 10^{10}-10^{15} GeV.[42] Models with extra dimensions, such as orbifolded SUSY SU(5) on S^1/Z_2, allow proton decay via Kaluza-Klein modes or boundary interactions, potentially lowering the effective unification scale and yielding lifetimes near 10^{35} years for modes involving pions and positrons. Similarly, composite models where quarks and leptons emerge as bound states of preons at a scale Λ_pre ~ 10^{12}-10^{15} GeV induce proton decay operators at the compositeness threshold, predicting rates suppressed by 1/Λ_pre^2 and favoring semileptonic modes like p → e^+ K^0.[43] A recent development in 2025 proposes that dark matter particles can induce proton decay without invoking traditional GUT scales, where dark sector mediators violate B+L symmetry at one loop, linking baryon decay to dark matter stability.[44] In SO(10) GUTs, the seesaw mechanism for generating small neutrino masses through right-handed neutrino singlets naturally ties proton decay rates to neutrinoless double beta decay processes, as both probe similar dimension-5 operators involving left-right mixing; enhanced proton decay in these models could correlate with signals in experiments like CUORE or KamLAND-Zen.| Model | Predicted Lifetime (years) | Dominant Decay Mode |
|---|---|---|
| Non-SUSY SU(5) | 10^{31}-10^{34} | p → e^+ π^0 |
| SUSY SU(5) | >10^{34} | p → \bar{\nu} K^+ |
| Non-SUSY SO(10) | 10^{33}-10^{35} | p → e^+ π^0 |
| SUSY SO(10) | 10^{34}-10^{36} | p → \bar{\nu} K^+ or μ^+ K |
| Left-Right Symmetric | 10^{32}-10^{35} | p → e^+ π^0 or 3 leptons |
| Extra Dimensions | ~10^{35} | p → e^+ π^0 |
| Composite Preons | 10^{34}-10^{36} | p → e^+ K^0 |
| DM-Induced (2025) | Model-dependent | Semileptonic modes |