The following are the leading quantum electrodynamics diagrams. Solid arrows indicate fermion-number flow, and wavy lines indicate photons. The heavy source line represents a charged nucleus; its recoil allows four-momentum conservation.
Figure 1.
Leading Compton, electron-line bremsstrahlung and nuclear-field pair-production diagrams
. Details.
For Compton scattering, , there are two diagrams: absorption of the incident photon followed along the electron line by emission of the outgoing photon, and the opposite ordering. These are the - and -channel electron propagators. Both scattering amplitudes are order .
For bremsstrahlung, , a virtual photon transfers momentum between the electron and the source. The real photon is emitted before or after this interaction along the electron line. These two dominant external-field diagrams have scattering amplitudes of order . A full finite-mass charged-target calculation also includes radiation from the target; it is suppressed in the heavy stationary-source approximation used here.
For nuclear-field pair production, , crossing the preceding diagrams gives two order- contributions, with the incoming real photon and the exchanged virtual photon in either order along the pair's fermion line. An isolated real photon cannot produce a massive pair in vacuum: its invariant squared mass is zero, whereas the pair has invariant squared mass at least . The source absorbs recoil momentum.
The decay is allowed in the Standard Model. Begin with ordinary muon decay, , . The outgoing electron can emit a virtual photon that converts to . The bottom panel of the figure in part (b) shows one representative diagram. Radiation from other charged lines and exchange of the identical final electrons complete the decay amplitude.
The final total electric charge is . Its electron-family lepton number is , and its muon-family lepton number is , matching the initial muon. The extra pair therefore violates no relevant Standard Model selection rule. It is also kinematically allowed since . The virtual pair costs two electromagnetic vertices: the decay amplitude is order , and the decay rate is suppressed by roughly two powers of the fine-structure constant, with additional phase space and logarithmic factors.
The observed dominant two-body transition is . Both baryons have valence quarks and the same strangeness; an electromagnetic interaction transition is therefore possible. Their mass difference is only about , below the lightest pion mass, so no additional on-shell strongly interacting particle fits. A photon is neutral and massless, and can carry the required angular momentum through a magnetic dipole transition between the two positive-parity spin- states. A single neutrino would violate lepton number and would not give the required fermion parity.
At rest, four-momentum conservation gives and . Thus
The result is slightly below because the Lambda baryon recoils; rounded masses give about .
The channels and are actually kinematically open: their total rest energies are about and , below . Their suppression is therefore not a decay threshold prohibition. Each changes strangeness from to zero and so requires a weak interaction. The much faster allowed electromagnetic transition makes such weak branching fractions extremely small. “Not observed” should be understood as overwhelming electromagnetic competition, not an exact ban in the Standard Model. The photon identification likewise uses the known low-energy spectrum and dominant transition mechanism, not neutrality alone.
Use metric and common spacetime origins. A four-vector transforms as under a Lorentz transformation. A Lorentz invariant is a scalar unchanged by that transformation. Since ,
The wave phase labels the same stage of a physical oscillation at a spacetime event, independently of the observer's coordinates. Equal-phase wavefronts and interference phase differences are physical; fixing a common phase origin makes the continuous phase a Lorentz scalar. For a plane wave,
The gradient of a scalar is a covector; raising its index gives the four-wavevector . Under an boost,
For vacuum light, , giving
A photon has four-momentum , with . For two photons separated by angle ,
In the centre-of-momentum frame the least possible pair energy is , when both final particles are at rest. Thus the threshold has ; above it, excess energy supplies their kinetic energy. Therefore
The kinematic equality is a limiting threshold; nonzero production phase space requires energies above it. For parallel photons no finite threshold is available, whereas head-on incidence minimizes the required energy:
The blackbody-peak photon pair-production estimate, using the frequency-spectrum blackbody peak, at , gives , or . If “peak” means the wavelength-spectrum peak, instead gives . The spectral convention changes the prefactor, not the conclusion that photons of order a few are needed for this idealized peak-photon threshold.
A bare gauge boson mass would spoil the electroweak gauge structure. In the Standard Model a complex Higgs doublet instead has a symmetry-breaking potential. Choose the convention
The Higgs mechanism rearranges three scalar components into the longitudinal polarizations of the massive W bosons and Z bosons; the unbroken electromagnetic gauge symmetry leaves the photon massless. One physical scalar excitation, the Higgs boson , remains.[ref] Using tree-level Feynman diagrams,
These equations use natural units. Yukawa interactions generate charged-fermion masses and Higgs couplings proportional to those masses; the minimal model does not account for observed neutrino masses. The gauge kinetic term supplies Higgs couplings to W and Z pairs. The physical Higgs exchange also cancels the otherwise excessive high-energy growth of longitudinal vector-boson scattering. The mechanism does not fix the Yukawa couplings or the Higgs self-coupling, and most proton mass arises from strong-interaction dynamics rather than directly from the light-quark Higgs masses.
The mass-generation mechanism is a statement about the field and its vacuum expectation value; the observable boson is the excitation about that vacuum.
The initial strangeness is . The final state has strangeness , so its allowed nonleptonic amplitude is weak. One representative topology is followed by ; spectator combine with the produced into the , while contributes to the neutral pion. The available rest energy is .
The state would conserve strangeness and could have a strong vertex, but its threshold exceeds the parent mass by . It is kinematically forbidden. Coupling strength cannot rescue a closed decay channel.
The radiative channel also changes strangeness. It is a weak radiative hyperon decay, not an ordinary electromagnetic decay between states of the same flavour. A representative two-quark contribution is W exchange with a photon attached to a charged line; the other allowed attachments, hadronic intermediate states and short-distance loop contributions must be included for a complete amplitude. Its extra electromagnetic coupling and hadronic dynamics suppress it relative to the nonleptonic channel. Consequently
NA48 Collaboration measured a radiative branching fraction of about , corroborating the suppression rather than a purely electromagnetic interpretation.[ref]
For the neutral quark-antiquark mesons here, . Thus , while . The radiative mode conserves charge conjugation and admits a quark-line photon-emission diagram. Both components of the supplied wavefunction contribute: the charge-weighted flavour overlap is proportional to , so it does not cancel. Its photon energy is .
The mode changes charge conjugation and, more decisively, a spin-zero particle cannot decay into another spin-zero particle and one real photon. In the parent rest frame the photon has helicity , while the opposite-moving pion has helicity zero; the spin-zero two-body helicity selection rule cannot be satisfied. Its on-shell two-body amplitude vanishes even if a weak interaction violates C. A virtual photon subsequently making another pair is a different final state.
For , angular-momentum conservation requires relative orbital angular momentum . Its parity is then negative, matching the parent, but its charge-conjugation eigenvalue is : the two distinct self-conjugate particles are not interchanged by C, so there is no additional factor in this C eigenvalue. The decay is forbidden by strong and electromagnetic charge-conjugation conservation despite being above threshold. A weak contribution would require CP violation as well. At the strong/electromagnetic accuracy of the diagram classification,
If extremely small weak CP-violating amplitudes are retained, need not be exactly zero; the one-real-photon spin prohibition remains exact for .
For electron-positron annihilation into a muon pair the gauge diagrams are s-channel photon and Z exchange. For electron-positron annihilation into electron neutrinos there is s-channel Z exchange and t-channel W exchange, with interference. For electron-positron annihilation into W-boson pairs there are s-channel photon and Z diagrams and a t-channel electron-neutrino diagram. All three gauge amplitudes are required for the proper high-energy cancellation. Electron-Yukawa-suppressed s-channel Higgs diagrams can also produce the massive muon or W pair; they vanish in the usual zero-electron-mass approximation used for the leading gauge comparison.
Figure 1.
Lepton and W-pair gauge diagrams
. Details.
There is no energy-independent ordering. For unpolarized beams and inclusive Born cross sections the useful regimes are:
The Z-pole ratio follows . Treating every weak reaction as uniformly slower misses both the resonance and forward exchange.
The diphoton invariant mass follows by adding the two photon four-momenta: . Thus
For correlated diphoton mass resolution, at fixed angle, . If the two quoted fractional errors are independent one-standard-deviation uncertainties,
An unknown common energy-scale error would be correlated: with correlation coefficient , the fractional variance is . Fully correlated scale errors give 10%, rather than 7.1%. A mass consistent with a candidate is not by itself evidence establishing its identity.
A nuclear excited state has internal energy above its ground state, with levels labelled by spin and parity and, where useful, approximate isospin. In the shell model, nucleons occupy orbitals split by spin-orbit coupling; nuclear particle-hole excitation and broken pairs produce excited configurations. An orbital has parity , and an even-even paired ground state is commonly . Residual interactions mix configurations, so shell spacings are not a complete spectrum.
Collective motion gives distinct patterns. A deformed nucleus can have a rotational band with , so a simple even-even band has . Near spherical shape, a quadrupole vibration gives a one-phonon excitation; two such phonons can couple to near twice that energy in the harmonic limit. Anharmonicity and coupling to single-particle motion modify these ideal patterns.
De-excitation occurs by gamma emission, internal conversion, or particle emission if energetically open. photon multipolarity must satisfy the angular-momentum triangle rule. Electric multipoles change parity by and magnetic multipoles by ; a single photon cannot connect . Large multipolarity or a small transition matrix element can make a long-lived nuclear isomer. Gamma energies, coincidence cascades, angular correlations and lifetimes constrain level schemes; inelastic scattering and transfer reactions supply complementary spin/configuration information. One observed line alone need not identify a unique level assignment.
The electroweak gauge group is . Left-handed fermions form weak-isospin doublets and right-handed charged fermions are singlets under . Before symmetry breaking the gauge fields are and , with couplings . A Higgs-doublet vacuum expectation value breaks the symmetry to electromagnetic .
In units , the physical combinations and tree-level relations are
The photon remains massless. charged currents change members of weak doublets, whereas the mediates weak neutral currents with parity-dependent couplings. At low energies exchange reduces to the Fermi contact interaction, ; the apparent weak suppression reflects the massive mediator. This unifies the gauge description, not the numerical equality of all low-energy couplings.[ref]
Neutral-current neutrino interactions and subsequent observation of massive weak vector bosons tested this structure. In a neutral-current neutrino experiment, an isolated hadronic interaction without an accompanying charged lepton is a candidate signal; entering neutral hadrons and missed charged-current leptons are important backgrounds. Gargamelle's neutral-current evidence supplied a direct experimental check beyond ordinary beta decay.[ref]
In time the particle advances , while a light wavelet travels . Coherent addition along the direction of emitted photons gives the Cherenkov cone angle
The angle is measured between photon propagation and the particle velocity. It is complementary to the angle between the conical wavefront surface and the trajectory; confusing those two angles exchanges sine and cosine. At threshold the photon cone closes, .
The radiative decay conserves strangeness and proceeds electromagnetically. A strong decay is closed because even the lighter neutral-pion threshold exceeds the supplied mass. The decay has about of available energy, but changes strangeness from to , so it requires the weak interaction. Its lifetime is therefore ordinarily much longer than the electromagnetic lifetime of ; no numerical lifetime follows from these masses alone.
Both particles are baryons with half-integer spin. A radiative transition to spin can couple through photon multipoles , giving . Electromagnetic parity is conserved, but without the electric/magnetic multipolarity neither nor its parity is fixed. In the weak decay, coupling orbital to the final spin gives ; parity conservation cannot be imposed on that weak amplitude. The supplied modes therefore do not determine unique spin-parity assignments.
The masses and flavours identify the familiar candidates as and . Their established ground-state assignments are both , consistent with the lowest-orbital quark model, but this is additional empirical/model information rather than a consequence of the two decay channels alone.[ref]
The symbols and mean electric and magnetic radiation multipoles of order . The orders are called dipole, quadrupole and octupole. They describe the angular structure of charge/current radiation and the total angular momentum carried by the radiation; they do not change the photon's intrinsic spin from one. A real single photon has no monopole radiation mode.
For a single-photon transition, angular momentum addition requires , , together with the electric/magnetic parity rules above. Thus a known final state and a measured multipole restrict both the spin difference and parity of the emitting level. For example, an E1 line reverses parity and allows except ; a mixed M1+E2 line preserves parity and must satisfy both angular triangles. An E2 line preserves parity and allows spin differences up to two.
Line energies identify connected levels, while calibrated intensities and coincidences test cascade flow and branching. If a proposed assignment permits a fast low-order transition to a lower state, an experimentally stringent absence of that line argues against the assignment. Nevertheless, an allowed transition can have a small or vanishing nuclear matrix element. Pure E2 radiation alone does not prove that M1 is forbidden by angular momentum. Detection sensitivity, efficiency and overlap must be considered, and in real nuclei internal conversion can remove photons; this question explicitly excludes that competing channel. Lifetimes or reduced-strength estimates make a missing-line argument substantially stronger than a bare selection-rule statement.
Evidence for quarks comes from several kinds of observation that test different aspects of their interpretation. The light meson and baryon spectra organize into approximate flavour multiplets with the charges and strangeness expected for and . In particular, the baryon octet and baryon decuplet encode the three light quark flavours and their spin couplings. This is an economical account of hadron quantum numbers, but spectroscopy alone would not establish that the constituents behave as localized scattering centers.
Deep inelastic scattering supplies that second test. In energetic electron–proton scattering the outgoing lepton measures the energy and momentum transferred through a virtual photon, while the broken-up target is summed over. Approximate Bjorken scaling means the structure functions depend primarily on rather than varying rapidly with the resolving momentum scale . It indicates nearly pointlike constituents. In the leading spin-half parton model,
The charge-squared weights and the second relation test the constituents' fractional charges and spin-half nature. Scaling violations reflect quark and gluon radiation rather than a failure of the constituent picture.[ref]
Particle jets give complementary final-state evidence. A high-energy quark produces a collimated spray of hadrons because confinement prevents it from emerging as an isolated particle. Back-to-back jets fit quark–antiquark production; three-jet events show hard gluon radiation. Their angular correlations test a spin-one radiated parton. This distinguishes constituents of hadrons from a literal observation of free fractional-charge particles.
Three independent arguments identify an additional threefold colour degree of freedom. First, the spin-aligned state of a ground-state would have symmetric space, spin and flavour factors. The Pauli exclusion principle requires an antisymmetric additional factor. The three-quark colour singlet supplies it. The same construction accommodates the spin-aligned state of the without altering its measured quantum numbers.
Second, the inclusive hadronic R ratio, away from thresholds and narrow resonances, is at leading electromagnetic order
Each quark flavour has distinguishable colour final states. With , the predicted plateaux for three, four and five accessible flavours are , and ; without the factor three they would be a third as large. The measured inclusive normalization is therefore a colour-counting test, with radiative and threshold corrections included in precision comparisons.[ref]
Figure 1.
Leading colour-counting predictions for the hadronic R ratio; these are model values, not plotted measurements
. Details.
Third, the neutral pion's two-photon amplitude contains a quark-loop colour sum, proportional to ; its rate and the other evidence are consistent with three colours. Quantum chromodynamics identifies this degree of freedom as a gauge charge, with eight gluons. Observed hadrons are colour singlets; the evidence for colour is its quantitative and symmetry consequences, not visible colours or isolated coloured quarks.
The common energetic principle behind nuclear fission and nuclear fusion is the nuclear binding energy. If the products are more tightly bound in total than the reactants, their smaller rest mass releases
The two binding-energy expressions assume that the same numbers of protons and neutrons are counted on both sides; emitted free nucleons have zero nuclear binding. The energy appears mainly as product kinetic energy, with radiation and subsequent decays contributing according to the reaction.
The binding energy per nucleon rises rapidly through light nuclei, has a broad maximum in the iron/nickel region, and decreases more slowly for heavier nuclei. Combining suitable light nuclei can therefore increase their total binding, whereas splitting a sufficiently heavy nucleus into moderately heavy fragments can do the same. This is an energy comparison, not a rule that every fusion or fission channel is exothermic: the actual nuclear masses fix the sign of .
Figure 1.
Schematic binding-energy trend explaining why light-nucleus fusion and heavy-nucleus fission can release energy
. Details.
In nuclear fission, the liquid-drop model illustrates the competition between attractive short-range nuclear cohesion and long-range proton repulsive Coulomb interaction. Deformation increases surface area and costs surface energy, but separating the charge can reduce nuclear Coulomb energy. Their competition creates a fission barrier. An excited nucleus may pass over the barrier; spontaneous fission can occur through quantum tunnelling. The resulting fragments are often excited and neutron-rich, so they may emit prompt neutrons and photons and undergo later beta decay. The typical total energy released by a heavy-nucleus fission is of order a few hundred MeV, rather than an electronic-scale energy. Its size follows from an approximately MeV-per-nucleon binding gain summed over a heavy nucleus. A neutron from one fission can induce another; a nuclear chain reaction is the repeated physical process, distinct from the energy-release condition for one event.
In nuclear fusion, the initial charged nuclei repel one another. Exothermicity does not remove this entrance Coulomb barrier. A reaction requires sufficient overlap of their wavefunctions, achieved by energetic encounters and quantum tunnelling. For nonresonant low-energy charged-particle fusion, a useful separation is
The astrophysical S factor isolates the nuclear part from much of the rapidly varying Coulomb penetration probability. In a thermal population, the competition between the decreasing high-energy population and increasing penetration produces the Gamow peak. This explains how fusion can proceed below the classical barrier without requiring the mean thermal energy to equal its height.
Both processes conserve total energy, charge, baryon number and angular momentum. Their very different entrance barriers and mass ranges explain why a favorable value by itself does not specify a reaction rate. Short-range attraction gives approximately saturated nuclear binding, while surface, Coulomb, symmetry and pairing terms account for the broad trend and its isotope-dependent departures.
In the Breit-Wigner resonance cross section, is the total centre-of-momentum energy, the resonance energy, and either incident particle's center-of-momentum momentum magnitude. The total decay width is , while and are partial widths into the entrance and specified exit channels. Their ratios to are the corresponding branching fractions. The factor averages over incident spin states and sums over the resonance's magnetic substates.
In natural units, , the resonant partial-wave cross-sectional scale is an area proportional to the squared reduced de Broglie wavelength, . Equivalently a partial-wave scattering amplitude carries one factor , and its squared modulus supplies . In SI area units the same scale is ; if is expressed in energy units, multiply the final inverse-energy-squared answer by .
For unpolarized spin-half incident muons, the unpolarized resonance spin factor is
Hence and . The outgoing tau spins do not supply another independent multiplicity here: their allowed final states are already included in .
Near a narrow resonance, treat the smooth prefactor and the partial widths as constant across the line. At the peak the denominator is . Half its height occurs when , namely at . Thus the full width at half maximum is within the Breit-Wigner narrow-line approximation. A varying prefactor or convolution with the beam-energy distribution can modify an observed line shape.
Figure 1.
Tree production and decay, fermion triangles, and the W-loop topologies for Higgs diphoton decay
. Details.
The production amplitude has a muon pair joined to a Higgs propagator through their Yukawa coupling, followed by the bottom-quark pair at a second Yukawa vertex. The separate graph is a tree-level Higgs–fermion vertex. There are no tree-level or vertices in the renormalizable Standard Model: their lowest nonzero graphs are loops. A quark triangle, dominated by the top contribution, produces two gluons. Charged-fermion triangles contribute to two photons; charged-W loops also contribute and are dominant in the Standard Model amplitude. Photon permutations and the W contact topology belong to the full amplitude; the diagram shows these topologies rather than claiming one triangle alone is the entire gauge-invariant W contribution.[ref]
The dominant interaction is electromagnetic. The color singlet charm-anticharm component of the J/psi meson annihilates through a virtual photon, which creates a muon and an antimuon:
Panel (i) shows the two electromagnetic vertices. Vector quantum numbers allow the one-photon channel. A Z boson contribution is negligible at this mass scale; a single gluon cannot connect a color singlet initial state to colorless leptons. The dominant interaction for this specified final state need not dominate the meson's total decay width.
In neutron-proton radiative capture, the initial pair at rest has total energy , where . The emitted photon carries momentum, requiring equal and opposite nuclear recoil. Some binding energy becomes deuteron kinetic energy.
Exact two-body kinematics in natural units gives
Since ,
With the stated rounded masses,
The photon energy is . At leading order the difference equals the nonrelativistic recoil estimate ; exact kinematics fixes the small distinction between using and in that approximation.