Frontiers: Quantum Fields, the Standard Model & Quantum Gravity

For twenty-four chapters we have treated an electron as a thing with a wavefunction of its own, quietly assuming that the number of particles never changes. Einstein's E = mc² says otherwise: at high energies particles are created and destroyed constantly, as when a gamma-ray photon passing a nucleus becomes an electron and its antiparticle. The remedy, built from 1927 onward by Paul Dirac, Pascual Jordan, Werner Heisenberg, Wolfgang Pauli and others, is quantum field theory: the framework in which the fundamental entities are fields, quantities with a value at every point of space and time, and particles are their quantized excitations.

Chapter 13 already showed the central idea: each mode of the electromagnetic field is a quantum harmonic oscillator, and a photon is one rung of excitation. Quantum field theory says the same of everything: an electron is one quantum of an electron field that fills the universe, and likewise for quarks, neutrinos and the Higgs. Creation and annihilation operators, the ladder operators renamed, add and remove quanta. That is why every electron is perfectly identical, the fact behind Chapter 19's statistics: all are ripples of one field, as every ripple on a pond is made of the same water.

Fields also settle the wave-particle puzzle. The field is the wavy thing that spreads, diffracts and interferes; its quanta are why a detector always clicks with whole lumps of energy. Duality was never a contradiction between two kinds of object. It is one kind of object, a quantum field, showing two faces.

In 1928 Paul Dirac wrote an equation for the electron that obeyed both quantum mechanics and special relativity. Spin-½ emerged automatically, and the magnetic g-factor came out as exactly 2, the start of Chapter 1's precision story. But relativity fixes only the square of the energy, which equals the square of pc plus the square of the rest energy mc², and a square has two roots, so the equation also allowed negative energies. Nothing seemed to stop an electron from tumbling into them forever, which would make every atom unstable.

In 1930 Dirac imagined the negative-energy states as already filled, a bottomless sea in which a missing electron, a hole, would act as a positive particle. When Hermann Weyl showed that a hole must share the electron's mass, Dirac predicted an anti-electron in 1931. In 1932 Carl Anderson at Caltech found it in cosmic-ray tracks curving the wrong way in a magnetized cloud chamber and named it the positron. Quantum field theory later retired the sea: the electron field simply has two kinds of quanta, electrons and positrons, with equal mass and opposite charge.

Every particle has an antiparticle, and a few, such as the photon, are their own. A particle and its antiparticle can annihilate, turning all their rest energy into other particles: an electron and a positron at rest usually become two gamma-ray photons of 511 keV each, flying apart back to back. PET scanners, short for positron emission tomography, detect exactly these pairs from a tracer such as fluorine-18-labelled sugar. In 2023 CERN's ALPHA experiment showed that antihydrogen falls downward, consistent with ordinary gravity within the measurement's still-large uncertainty of roughly 20 to 30 percent.

Quantum electrodynamics, or QED, is the quantum field theory of electrons and light. Its first versions gave infinite answers. In the 1940s Sin-Itiro Tomonaga, Julian Schwinger and Richard Feynman tamed them, Freeman Dyson showed their methods equivalent, and the first three shared the 1965 Nobel Prize. The cure, called renormalization, expresses every prediction in terms of the electron's measured mass and charge, whereupon the infinities drop out of everything that can be measured.

Feynman's gift was a picture language. In a Feynman diagram, straight lines stand for electrons or positrons, wavy lines for photons, and each vertex for a charge emitting or absorbing a photon. Each diagram is shorthand for one mathematical term in the probability amplitude for a process, and the rule is the one from the double slit: add the amplitudes for every way the process can happen, then square the magnitude of the sum. Each extra pair of vertices brings another factor of the fine-structure constant α, about 1/137, so elaborate diagrams contribute less and less, an expansion called a perturbation series.

The electron's magnetism is the showcase. In 1948 Schwinger evaluated the single diagram in which the electron emits and reabsorbs one photon, finding that g/2 should exceed Dirac's value of 1 by α/2π, about 0.00116. The next orders take 7, 72, 891 and finally 12,672 diagrams, evaluated over seven decades by many physicists, the last two orders by Toichiro Kinoshita's team, and the result matches the 2023 Northwestern measurement to about a part in a trillion. QED also explains the Lamb shift, a 1,058 MHz splitting of two hydrogen levels that Dirac's theory said should coincide, measured by Willis Lamb and Robert Retherford in 1947.

Now some honesty. The internal lines of a diagram, which begin and end inside it, are called virtual particles. They do not obey the energy-momentum relation that every real particle must, no detector ever registers one, and lattice calculations, which put the fields on a grid instead of expanding in a small coupling, contain no diagrams at all. Virtual particles label terms in an approximation. The popular image of pairs popping in and out of empty space like tiny marbles is a cartoon. The vacuum is not a quiet nothing, for its fields keep the zero-point fluctuations of Chapter 13, but it is a state of the fields, not a box of marbles.

The Standard Model of particle physics, assembled between about 1961 and 1973, is the quantum field theory of every elementary particle ever detected and every known force except gravity. Its matter particles are spin-½ fermions in two families. Quarks feel the strong force: up, charm and top carry electric charge +2/3, in units of the proton's charge, while down, strange and bottom carry −1/3. Leptons do not: the electron, muon and tau carry −1, and each has a neutral partner, its neutrino. Each family comes in three generations, the charged members of each generation heavier than the last, and ordinary matter uses only the first: a proton is two up quarks and a down, a neutron one up and two downs. Masses are quoted in energy units via E = mc²: the electron's mass is 0.511 MeV.

The forces come from gauge symmetry. Demand that the equations keep their form when a field's internal phase is shifted differently at every point, and new fields are forced into existence; their quanta, the spin-1 gauge bosons, carry the forces. The photon carries electromagnetism. Eight gluons carry the strong force of quantum chromodynamics, or QCD, which confines quarks inside protons and neutrons. Two W bosons, one positive and one negative, and the neutral Z carry the weak force, which drives beta decay; Sheldon Glashow, Steven Weinberg and Abdus Salam unified it with electromagnetism.

Yet gauge symmetry seems to demand massless carriers, while the W and Z, found at CERN in 1983, weigh about 80 and 91 GeV. In 1964 three groups, Robert Brout and François Englert, Peter Higgs, and Gerald Guralnik, Carl Hagen and Tom Kibble, found the way out: a field filling space whose value in its lowest-energy state is not zero. Interacting with this Higgs field gives mass to the W and Z and, through separate couplings, to the quarks and charged leptons. Its ripple, the spin-0 Higgs boson, was announced by the ATLAS and CMS experiments at CERN's Large Hadron Collider on 4 July 2012. The Particle Data Group's 2026 average puts its mass at 125.13 ± 0.11 GeV, and Englert and Higgs shared the 2013 Nobel Prize.

Yet the Higgs does not give you most of your mass. The rest energies of a proton's three quarks add to about 9 MeV, roughly 1 percent of its 938 MeV; the other 99 percent is the energy of quark and gluon fields bottled up by the strong force, as lattice QCD calculations confirm. The Higgs matters to you differently: without the electron's mass, atoms could not form at all.

Despite half a century of collider triumphs, the Standard Model leaves out five big things. First, gravity: it has no graviton, and gravity is still described by Einstein's classical general relativity of 1915. Second, dark matter. Fritz Zwicky's 1933 study of the Coma galaxy cluster, Vera Rubin and Kent Ford's 1970s measurements of spinning galaxies, lensing by the Bullet Cluster and the cosmic microwave background all point to about five times as much unseen matter as ordinary matter. No Standard Model particle fits, and as of 2026 no detector has caught it.

Third, dark energy. In 1998 two teams studying distant exploding stars found that cosmic expansion is accelerating, earning Saul Perlmutter, Brian Schmidt and Adam Riess the 2011 Nobel Prize. The Planck satellite's 2018 data give a budget of about 5 percent ordinary matter, 27 percent dark matter and 68 percent dark energy. The obvious candidate is vacuum energy, but here is the debt Chapter 13 left you: adding the zero-point energies of all field modes up to the Planck scale overshoots the observed value by a factor of roughly 10¹²⁰. Since 2024 the DESI galaxy survey has hinted, at roughly three standard deviations depending on which other data are combined, that dark energy may be weakening: intriguing, not established.

Fourth, neutrino masses. The original Standard Model made neutrinos massless, but neutrinos change type in flight, as Super-Kamiokande showed in 1998 and the Sudbury Neutrino Observatory in 2001 and 2002, and such oscillation requires nonzero masses. In 2025 the KATRIN experiment, measuring the energies of electrons from tritium beta decay, capped the neutrino's effective mass below 0.45 eV, less than a millionth of the electron's.

Fifth, the matter-antimatter asymmetry. The Big Bang should have made equal amounts of each, which would have annihilated into light; instead, for roughly every billion quark-antiquark pairs there was one extra quark, and the leftovers became us. Andrei Sakharov showed in 1967 that this requires, among other conditions, matter and antimatter to obey slightly different laws, called CP violation. It exists: James Cronin and Val Fitch found it in kaons in 1964, and LHCb at CERN saw it in baryons in 2025. But the Standard Model's CP violation is far too weak to explain our universe.

At low energies gravity can be treated as a quantum field whose quantum, the still-undetected graviton, would have spin 2. At high energies the approach fails: computing with gravitons the way QED computes with photons produces infinities that renormalization cannot absorb, as Gerard 't Hooft and Martinus Veltman began to show in 1974. The breakdown scale combines ħ, c and Newton's constant G: the Planck length, about 1.6×10⁻³⁵ metres, and the Planck energy, about 1.2×10¹⁹ giga-electron-volts, nearly a million billion times the Large Hadron Collider's collision energy.

String theory replaces point particles with tiny vibrating strings whose patterns of vibration appear as different particles. It grew from a 1968 formula for the strong force, and in 1974 Joël Scherk and John Schwarz, and independently Tamiaki Yoneya, found that it contains a massless spin-2 vibration with exactly the graviton's properties. Juan Maldacena's 1997 holographic duality, linking gravity in a curved spacetime to a quantum field theory without gravity on its boundary, became a tool across physics. The costs: ten spacetime dimensions, six curled out of sight; a landscape of perhaps 10⁵⁰⁰ possible vacuum states, each with different low-energy laws, hence no unique prediction; and no sign at the LHC of the supersymmetry it requires at some energy.

Loop quantum gravity instead quantizes the geometry of spacetime itself. Built on Abhay Ashtekar's 1986 reformulation of general relativity and developed by Carlo Rovelli, Lee Smolin and others, it predicts that area and volume come in discrete quanta set by the Planck length, so space is granular. It keeps four dimensions but does not unify the forces and struggles to recover smooth spacetime at large scales. As of 2026 neither approach is confirmed by experiment, though observation can prune: in 2009 gamma rays of very different energies from a distant burst reached the Fermi telescope together, ruling out simple models in which granular space slows high-energy light.

Black holes are where the theories collide. In 1974 Stephen Hawking showed that quantum fields near a horizon make a black hole glow like a blackbody, at a temperature inversely proportional to its mass. A black hole of one solar mass sits near 60 billionths of a kelvin, far colder than the 2.7 K cosmic microwave background, so Hawking radiation from a real black hole has never been detected.

Then the paradox. If a black hole evaporates into exactly thermal radiation, a library and a star that fell in end as the same featureless glow, destroying information that unitary quantum evolution, Chapters 10 and 14, always preserves. In 2019 Geoff Penington, and independently Ahmed Almheiri, Netta Engelhardt, Donald Marolf and Henry Maxfield, showed in simplified models that information can escape, reproducing the curve Don Page had predicted in 1993. Exactly how it escapes remains open.

New physics travels through a pipeline. Most physics results appear first on arXiv, pronounced archive, the free preprint server Paul Ginsparg launched in August 1991, now run by Cornell University. arXiv screens submissions but does not peer review them: you see research the day its authors post it, before any independent expert has checked it. Peer review comes next, when a journal sends the paper to independent experts who can demand changes or reject it. Peer review filters errors; it does not certify truth. The final judge is replication by independent teams.

Particle physicists demand five sigma before claiming a discovery: odds below about 1 in 3.5 million that background noise alone would fake the signal. History explains the strictness. OPERA's faster-than-light neutrinos of 2011 were traced in 2012 to a loose fibre-optic connector. BICEP2's 2014 signal from the infant universe proved, once Planck data were added in 2015, compatible with dust in our own galaxy. And the muon's magnetic anomaly, trumpeted at 4.2 sigma in 2021, largely evaporated in 2025 when lattice QCD brought theory into line with Fermilab's final measurement. None of this was fraud; it was science correcting itself in public.

Quantum headlines need the same care. In 2022 news stories said physicists had made a wormhole in a quantum computer; the paper simulated, on a handful of qubits, a simplified model that mirrors a wormhole through a mathematical dictionary, and even that reading was disputed. So ask five questions. Does the headline match the abstract? Is it a preprint or peer reviewed? Has anyone reproduced it? Is it the thing itself or a simulation? Is the claim dated and modest?

That is why this course ends with a newspaper. The daily What's New in Quantum bulletin, on the academy's What's New tab, gathers each day's quantum stories from sources such as Nature, APS Physics and Quanta Magazine, links each one to the chapter that explains its physics, and I will read it aloud. Turn on alerts there and it reaches your phone as a push notification, an email, a text message or an ntfy alert.