10.05.03 · em-sr / special-relativity

Four-velocity, four-momentum, and the relativistic energy-momentum identity

shipped3 tiersLean: none

Anchor (Master): Misner-Thorne-Wheeler, *Gravitation* (Freeman, 1973), §2; Weinberg, *Gravitation and Cosmology* (Wiley, 1972), §2; Landau-Lifshitz Vol. 2, *The Classical Theory of Fields*, 4e (Pergamon, 1980), §6-9

Intuition Beginner

In Newtonian physics, velocity is a three-vector $v$ with three independent components. When you change frames, the velocity changes by Galilean addition: walk forward at 5 km/h on a train moving at 100 km/h, and the ground sees you at 105 km/h. The three-vector is the natural object for ordinary speeds.

Special relativity replaces this picture. Space and time mix under a boost, so velocity must become a four-component object that includes the time direction. The new object is called the four-velocity $u^{μ} = (γ c, γ v)$ , with one time component $γ c$ and three space components $γ v$ . The Greek index $μ$ runs over 0, 1, 2, 3 — the four directions of spacetime.

The four-velocity has a remarkable property. Its Minkowski "length" is fixed at a universal value: it always squares to $- c^{2}$ in the signature where time carries a minus sign. Every massive particle's four-velocity has the same intrinsic magnitude. The only thing that changes between particles is the direction the four-velocity points in spacetime.

Multiplying the four-velocity by the rest mass $m$ gives the four-momentum $p^{μ} = m u^{μ} = (E / c, p)$ . The time component is the total energy divided by $c$ . The space components are the relativistic momentum $γ m v$ . Energy and momentum are not separate quantities in relativity — they are different components of one four-vector.

The fixed magnitude of the four-velocity translates into a fixed magnitude for the four-momentum:

E^{2} = (p c)^{2} + (m c^{2})^{2} .

This is the famous energy-momentum identity. The total energy squared equals the momentum-energy squared plus the rest-energy squared. It works like a Pythagorean theorem in energy-momentum space.

Three special cases illustrate the identity. A particle at rest has $p = 0$ , so $E = m c^{2}$ — the rest energy. A photon has $m = 0$ , so $E = p c$ — energy and momentum are proportional. A relativistic electron at $v = 0.99 c$ has Lorentz factor about 7, so its total energy is roughly seven times its rest energy of 0.511 MeV.

The identity controls every relativistic collision and decay. To find what comes out of a particle interaction, you add up the four-momenta of the incoming particles, then split the result among the outgoing particles in a way that conserves all four components. The "Pythagorean" identity then tells you the rest mass of any sub-system formed by combining particle four-momenta.

The picture above shows the geometric content of the energy-momentum identity. The hypotenuse is the total energy. The horizontal leg is the momentum contribution $p c$ . The vertical leg is the rest energy $m c^{2}$ . The triangle deforms as the particle's state changes, but the identity holds in every frame.

Worked example Beginner

A relativistic electron is accelerated to $v = 0.99 c$ . Compute its Lorentz factor, its total energy, its kinetic energy, and its momentum. The electron rest energy is $m c^{2} = 0.511$ MeV.

Step 1. Compute the Lorentz factor. With $v / c = 0.99$ , the ratio $v^{2} / c^{2} = 0.9801$ :

γ = \frac{1}{1 - 0.9801} = \frac{1}{0.0199} \approx 7.09.

Step 2. Total energy from $E = γ m c^{2}$ :

E = 7.09 \times 0.511 = 3.62 MeV .

The total energy is roughly seven times the rest energy because the Lorentz factor is roughly seven.

Step 3. Kinetic energy is total minus rest:

K = E - m c^{2} = 3.62 - 0.511 = 3.11 MeV .

Step 4. Momentum from the energy-momentum identity. Rearranging $E^{2} = (p c)^{2} + (m c^{2})^{2}$ :

p c = E^{2} - (m c^{2})^{2} = 3.6 2^{2} - 0.51 1^{2} = 13.10 - 0.26 = 12.84 \approx 3.58 MeV .

So the electron carries momentum $p \approx 3.58$ MeV/ $c$ . The ratio $p c / E = 3.58/3.62 \approx 0.989$ — nearly all the energy is in the momentum-energy contribution because the rest energy is small compared with the kinetic energy.

What this tells us: at $v = 0.99 c$ a relativistic electron is roughly seven times more energetic than at rest, and its energy is almost entirely momentum-energy. The photon limit is the case where the rest term goes to zero and $E = p c$ exactly.

Second worked example: pion decay at rest

A neutral pion ( $m_{π^{0}} c^{2} = 135$ MeV) decays at rest into two photons: $π^{0} \to γ + γ$ . Find the energy and momentum of each photon.

Step 1. Before the decay, the total energy is the pion rest energy: $E_{total} = 135$ MeV. The total momentum is zero (the pion is at rest).

Step 2. Conservation of four-momentum requires the two photons to share the total energy and to have equal and opposite momenta (so they sum to zero). By symmetry each photon takes half the energy:

E_{γ} = \frac{135}{2} = 67.5 MeV .

Step 3. Each photon has $E = p c$ (massless), so its momentum is:

p_{γ} = \frac{E _{γ}}{c} = 67.5 MeV / c .

The two photons fly apart in opposite directions, each carrying 67.5 MeV of energy and 67.5 MeV/ $c$ of momentum.

What this tells us: when a particle decays at rest, the daughter four-momenta are fixed by conservation alone. The decay products' kinematics are completely determined by the parent's rest mass and the daughters' masses — a fact used routinely in particle-physics experiments to identify parent particles from the kinematics of their decay products.

Check your understanding Beginner

Formal definition Intermediate+

Let a massive particle of rest mass $m$ trace out a timelike worldline $x^{μ} (τ)$ in Minkowski spacetime, parameterised by proper time $τ$ . Proper time is the time measured by a clock comoving with the particle, related to coordinate time $t$ by $d τ = d t / γ$ where $γ = (1 - v^{2} / c^{2})^{- 1/2}$ .

Definition (four-velocity). The four-velocity is $$ u^\mu := \frac{dx^\mu}{d\tau} = \gamma,(c, \mathbf{v}), $$ where the four spacetime components are $u^{0} = γ c$ , $u^{i} = γ v^{i}$ for $i = 1, 2, 3$ .

Throughout this unit we adopt the signature $(-, +, +, +)$ for the Minkowski metric $η_{μν} = diag (- 1, + 1, + 1, + 1)$ , following Misner-Thorne-Wheeler. Other texts (Landau-Lifshitz, Weinberg) use the opposite signature $(+, -, -, -)$ , which flips signs in the algebra below. The substantive content is signature-independent.

Proposition (norm of the four-velocity). The Minkowski norm of $u^{μ}$ is constant along any timelike worldline: $$ \eta_{\mu\nu},u^\mu u^\nu = -\gamma^2 c^2 + \gamma^2 \mathbf{v}\cdot\mathbf{v} = \gamma^2(-c^2 + v^2) = -c^2. $$ The final equality uses $γ^{2} (1 - v^{2} / c^{2}) = 1$ , so $γ^{2} (c^{2} - v^{2}) = c^{2}$ .

The four-velocity is timelike and future-directed for any massive particle. Its constancy is a kinematic constraint, not a physical assumption: the constraint follows from the definition $u^{μ} = d x^{μ} / d τ$ together with the relation $d τ^{2} = - d s^{2} / c^{2}$ between proper time and the spacetime interval.

Definition (four-momentum). The four-momentum is $$ \boxed{;;p^\mu := m,u^\mu = (\gamma m c, \gamma m \mathbf{v}) = (E/c, \mathbf{p}),;;} $$ where the energy is $E = γ m c^{2}$ and the relativistic three-momentum is $p = γ m v$ . The rest mass $m$ is a Lorentz invariant — the same in every inertial frame.

From the constancy of $u^{μ} u_{μ}$ and the definition $p^{μ} = m u^{μ}$ : $$ \eta_{\mu\nu},p^\mu p^\nu = m^2,\eta_{\mu\nu},u^\mu u^\nu = -m^2 c^2. $$

Writing this out using $p^{μ} = (E / c, p)$ : $$ -(E/c)^2 + |\mathbf{p}|^2 = -m^2 c^2, $$ which rearranges to the relativistic energy-momentum identity: $$ \boxed{;;E^2 = |\mathbf{p}|^2 c^2 + m^2 c^4.;;} $$

The quantity $m^{2} c^{2}$ is the invariant mass-squared of the particle. It has the same numerical value in every inertial frame, even though $E$ and $p$ separately transform under boosts.

Special cases. A particle at rest has $p = 0$ and $E = m c^{2}$ — the rest energy. A massless particle (photon, gluon, graviton) has $m = 0$ and $E = ∣ p ∣ c$ — energy proportional to momentum. The four-momentum of a massless particle is null: $η_{μν} p^{μ} p^{ν} = 0$ , and the worldline lies along the light cone.

Low-speed limit. For $v ≪ c$ , expand $γ = 1 + v^{2} / (2 c^{2}) + 3 v^{4} / (8 c^{4}) + \dots$ : $$ E = mc^2 + \tfrac{1}{2} m v^2 + \tfrac{3}{8} m v^4/c^2 + \cdots. $$ The rest energy $m c^{2}$ is a Lorentz scalar (frame-independent). The next term is the Newtonian kinetic energy $\frac{1}{2} m v^{2}$ . The subsequent terms are relativistic corrections suppressed by powers of $v^{2} / c^{2}$ .

Four-acceleration

Definition (four-acceleration). The four-acceleration is $$ \alpha^\mu := \frac{du^\mu}{d\tau}. $$ Differentiating the constancy of $u^{μ} u_{μ}$ gives $$ \frac{d}{d\tau}!\left(\eta_{\mu\nu} u^\mu u^\nu\right) = 2\eta_{\mu\nu} u^\mu \alpha^\nu = 0, $$ so the four-acceleration is orthogonal (in the Minkowski sense) to the four-velocity: $u^{μ} α_{μ} = 0$ . The four-acceleration is spacelike for any timelike worldline.

Four-force

The relativistic generalisation of Newton's second law is the four-force $$ f^\mu := \frac{dp^\mu}{d\tau} = m,\alpha^\mu. $$ The spatial part is $f = γ F$ where $F = d p / d t$ is the lab-frame three-force; the time part is $f^{0} = (γ / c) d E / d t$ . In the low-speed limit $γ \to 1$ and the spatial part of the four-force reduces to Newton's second law $F = d p / d t$ .

Counterexamples to common slips

"The rest mass changes with speed." The rest mass $m$ is a Lorentz invariant: $m^{2} = - p^{μ} p_{μ} / c^{2}$ has the same value in every frame. What grows with speed is $γ m$ , sometimes historically called the "relativistic mass." Modern usage reserves "mass" for the rest mass and lets $γ$ carry the speed dependence.
" $E = m c^{2}$ for any particle." The relation $E = m c^{2}$ applies only to a particle at rest. The general expression is $E = γ m c^{2}$ , or equivalently $E^{2} = ∣ p ∣^{2} c^{2} + m^{2} c^{4}$ from the energy-momentum identity.
"Photons have no momentum because they have no mass." Photons have $E = ∣ p ∣ c$ from the energy-momentum identity with $m = 0$ . Photon momentum is responsible for radiation pressure, Compton scattering, and photon-rocket propulsion.
"The four-acceleration of an inertial observer is non-zero because they are moving." An inertial observer has $u^{μ} =$ constant in any inertial frame, so $α^{μ} = d u^{μ} / d τ = 0$ . Four-acceleration measures deviation from geodesic motion, not coordinate velocity.

Key theorem with proof Intermediate+

Theorem (invariance of the energy-momentum norm). For any massive particle of rest mass $m$ and four-momentum $p^{μ} = (E / c, p)$ in some inertial frame, the quantity $$ M^2 c^4 := E^2 - |\mathbf{p}|^2 c^2 $$ is invariant under proper orthochronous Lorentz transformations and equals $m^{2} c^{4}$ .

Proof. Let $Λ \in SO^{+} (3, 1)$ be a proper orthochronous Lorentz transformation, and let $p^{' μ} = Λ^{μ}_{ν} p^{ν}$ be the transformed four-momentum. Compute the transformed norm using the Lorentz invariance of the Minkowski metric: $$ \eta_{\mu\nu},p'^\mu p'^\nu = \eta_{\mu\nu},\Lambda^\mu{}\alpha,\Lambda^\nu{}\beta,p^\alpha p^\beta = \eta_{\alpha\beta},p^\alpha p^\beta, $$ where the last step uses the defining property of Lorentz transformations: $η_{μν} Λ^{μ}_{α} Λ^{ν}_{β} = η_{α β}$ . The Minkowski norm of $p^{μ}$ is therefore frame-independent.

The norm has the explicit form $$ \eta_{\mu\nu},p^\mu p^\nu = -(p^0)^2 + |\mathbf{p}|^2 = -E^2/c^2 + |\mathbf{p}|^2 = -M^2 c^2, $$ so $M^{2} c^{4} = E^{2} - ∣ p ∣^{2} c^{2}$ is the invariant.

To identify $M$ with the rest mass $m$ , evaluate in the particle's rest frame, where $p = 0$ and $E = m c^{2}$ : $$ M^2 c^4 = (m c^2)^2 - 0 = m^2 c^4. $$ Hence $M = m$ for any massive particle. $□$

Bridge. The invariant-mass theorem is the foundational reason that high-energy physics computes invariant masses to identify particles. Putting these together with the conservation of four-momentum, an unstable resonance decaying into daughters $a, b, \dots$ has $M_{parent}^{2} c^{4} = (E_{a} + E_{b} + \dots)^{2} - ∣ p_{a} + p_{b} + \dots ∣^{2} c^{2}$ , and this combination is Lorentz invariant. The bridge is to detector analyses: measure the four-momenta of the daughters, reconstruct the parent invariant mass, and identify the resonance by its peak in the invariant-mass distribution. This is exactly how the J/ $ψ$ was discovered in 1974 and how the Higgs boson was identified in 2012. The pattern generalises to multi-body decays via Dalitz plots in 10.06.01 and builds toward the Mandelstam-variable formalism for 2-to-2 scattering covered in the Master tier below. Appears again in 12.11.01 when the Dirac equation factorises the energy-momentum identity at the operator level into a first-order spinor equation.

Worked example at intermediate level: Compton scattering

A photon of energy $E_{γ}$ scatters off an electron at rest. The photon is detected at angle $θ$ with energy $E_{γ}^{'}$ . Derive the wavelength shift $Δ λ = λ^{'} - λ$ .

Conservation of four-momentum: $p_{γ}^{μ} + p_{e}^{μ} = p_{γ}^{' μ} + p_{e}^{' μ}$ . Rearrange and contract with $η_{μν}$ : $$ (p_e'^\mu)(p_{e'\mu}) = (p_\gamma^\mu + p_e^\mu - p_\gamma'^\mu)(p_{\gamma\mu} + p_{e\mu} - p_{\gamma'\mu}). $$

The left side is $- m_{e}^{2} c^{2}$ . Expand the right side using $p_{γ}^{μ} p_{γ μ} = 0$ (photon is null), $p_{e}^{μ} p_{e μ} = - m_{e}^{2} c^{2}$ , $p_{γ}^{' μ} p_{γ^{'} μ} = 0$ : $$ -m_e^2 c^2 = -m_e^2 c^2 + 2 p_\gamma \cdot p_e - 2 p_\gamma \cdot p_\gamma' - 2 p_e \cdot p_\gamma'. $$

The $- m_{e}^{2} c^{2}$ terms cancel, leaving $p_{γ} \cdot p_{e} = p_{γ} \cdot p_{γ}^{'} + p_{e} \cdot p_{γ}^{'}$ . In the lab frame: $p_{e}^{μ} = (m_{e} c, 0)$ , $p_{γ}^{μ} = (E_{γ} / c) (1, \hat{n})$ , and $p_{γ}^{' μ} = (E_{γ}^{'} / c) (1, \hat{n}^{'})$ with $\hat{n} \cdot \hat{n}^{'} = cos θ$ . Computing the dot products: $$ -E_\gamma m_e = -E_\gamma E_\gamma'/c^2 + E_\gamma E_\gamma' \cos\theta/c^2 - E_\gamma' m_e. $$

Divide by $E_{γ} E_{γ}^{'} m_{e}$ : $$ -\frac{1}{E_\gamma'} = -\frac{1}{m_e c^2}(1 - \cos\theta) - \frac{1}{E_\gamma}. $$

Multiply by $- 1$ : $$ \frac{1}{E_\gamma'} - \frac{1}{E_\gamma} = \frac{1}{m_e c^2}(1 - \cos\theta). $$

Using $E_{γ} = h c / λ$ and $E_{γ}^{'} = h c / λ^{'}$ , this becomes the Compton formula: $$ \Delta\lambda = \lambda' - \lambda = \frac{h}{m_e c}(1 - \cos\theta). $$

The constant $h / (m_{e} c) = 2.43 \times 1 0^{- 12}$ m is the Compton wavelength of the electron. The wavelength shift depends only on the scattering angle, not on the incident wavelength — a directly verifiable prediction that won Compton the 1927 Nobel Prize.

Exercises Intermediate+

Exercise 5 (medium, symbolic).

Derive the threshold beam energy for the reaction $p + p \to p + p + π^{0}$ when one proton is at rest in the lab and the other is a beam of energy $E_{b}$ . The pion mass is $m_{π} c^{2} = 135$ MeV and the proton mass is $m_{p} c^{2} = 938$ MeV.

Hint

At threshold, all three final-state particles are at rest in the centre-of-momentum frame. Use the Mandelstam invariant $s = (p_{1}^{μ} + p_{2}^{μ})^{2} / c^{2}$ and require $s = 2 m_{p} c^{2} + m_{π} c^{2}$ .

Answer

In the lab frame: $p_{1}^{μ} = (E_{b} / c, p_{b})$ with $∣ p_{b} ∣ c = E_{b}^{2} - m_{p}^{2} c^{4}$ , and $p_{2}^{μ} = (m_{p} c, 0)$ . The Mandelstam $s$ :

s c^{2} = - (p_{1} + p_{2})^{μ} (p_{1} + p_{2})_{μ} = - (p_{1}^{μ} p_{1 μ} + p_{2}^{μ} p_{2 μ} + 2 p_{1}^{μ} p_{2 μ}) = m_{p}^{2} c^{4} + m_{p}^{2} c^{4} + 2 E_{b} m_{p} c^{2} .

So $s c^{2} = 2 m_{p}^{2} c^{4} + 2 E_{b} m_{p} c^{2}$ . At threshold $s c = 2 m_{p} c^{2} + m_{π} c^{2} = 2 (938) + 135 = 2011$ MeV. So $s c^{2} = (2011)^{2} = 4044121$ MeV $^{2}$ . Solve:

E_{b} = \frac{s c ^{2} - 2 m _{p}^{2} c ^{4}}{2 m _{p} c ^{2}} = \frac{4 044 121 - 2 ( 938 ) ^{2}}{2 ( 938 )} = \frac{4 044 121 - 1 759 688}{1876} \approx 1217.7 MeV .

Threshold kinetic energy $K_{b} = E_{b} - m_{p} c^{2} \approx 1218 - 938 = 280$ MeV. This is the minimum lab-frame kinetic energy required to produce a single $π^{0}$ in proton-proton collisions.

Exercise 7 (hard, symbolic).

Show that two photons in vacuum cannot annihilate into a single massive particle ( $γ + γ \to X$ ). Identify the kinematic obstruction.

Hint

Compute the invariant $s = - (p_{1}^{μ} + p_{2}^{μ}) (p_{1 μ} + p_{2 μ}) / c^{2}$ for two photons with opposite momenta versus parallel momenta.

Answer

The invariant mass of two photons is $M_{X}^{2} c^{4} = - (p_{1} + p_{2})^{μ} (p_{1} + p_{2})_{μ} c^{2} = - p_{1}^{μ} p_{1 μ} c^{2} - p_{2}^{μ} p_{2 μ} c^{2} - 2 p_{1}^{μ} p_{2 μ} c^{2}$ . The first two terms vanish (photons are null). The cross term is $- 2 p_{1}^{μ} p_{2 μ} c^{2} = 2 E_{1} E_{2} - 2 p_{1} \cdot p_{2} c^{2} = 2 E_{1} E_{2} (1 - cos θ)$ where $θ$ is the angle between the photon directions. For collinear photons ( $θ = 0$ ): $M_{X}^{2} = 0$ — only a massless particle can be the product. For antiparallel photons ( $θ = π$ ): $M_{X}^{2} c^{4} = 4 E_{1} E_{2}$ , which can equal any positive mass-squared depending on the photon energies. So $γ + γ \to X$ is allowed with non-collinear photons (used for high-energy photon-photon colliders and for two-photon production of Higgs-like states), but it is forbidden if the two photons are exactly collinear. The full kinematic obstruction is the requirement that $s > 0$ , which fails only in the degenerate collinear case.

Exercise 8 (hard, symbolic).

A particle of mass $M$ at rest decays into two particles of masses $m_{1}$ and $m_{2}$ . Find the energies $E_{1}$ and $E_{2}$ of the daughter particles in the rest frame of the parent.

Hint

Use conservation of four-momentum: $P^{μ} = (M c, 0) = p_{1}^{μ} + p_{2}^{μ}$ . Solve simultaneously $E_{1} + E_{2} = M c^{2}$ and $∣ p_{1} ∣ = ∣ p_{2} ∣$ (the two daughters fly apart with equal and opposite momenta).

Answer

From $p_{1} + p_{2} = 0$ : $∣ p_{1} ∣ = ∣ p_{2} ∣ := p$ . From the energy-momentum identity: $E_{1}^{2} = p^{2} c^{2} + m_{1}^{2} c^{4}$ and $E_{2}^{2} = p^{2} c^{2} + m_{2}^{2} c^{4}$ . So $E_{1}^{2} - E_{2}^{2} = (m_{1}^{2} - m_{2}^{2}) c^{4}$ , hence $(E_{1} - E_{2}) (E_{1} + E_{2}) = (m_{1}^{2} - m_{2}^{2}) c^{4}$ . Using $E_{1} + E_{2} = M c^{2}$ : $E_{1} - E_{2} = (m_{1}^{2} - m_{2}^{2}) c^{2} / M$ . Solving:

E_{1} = \frac{( M ^{2} + m _{1}^{2} - m _{2}^{2} ) c ^{2}}{2 M}, E_{2} = \frac{( M ^{2} - m _{1}^{2} + m _{2}^{2} ) c ^{2}}{2 M} .

The daughter momentum is

p c = \frac{c ^{2}}{2 M} M^{4} + m_{1}^{4} + m_{2}^{4} - 2 M^{2} m_{1}^{2} - 2 M^{2} m_{2}^{2} - 2 m_{1}^{2} m_{2}^{2} .

This is the standard two-body decay kinematics formula, used routinely in particle physics. The expression under the square root is the Källén triangle function $λ (M^{2}, m_{1}^{2}, m_{2}^{2})$ , which is positive precisely when $M > m_{1} + m_{2}$ — the energy-conservation threshold for the decay to be kinematically allowed.

Exercise 9 (hard, symbolic).

Define the rapidity $ϕ$ by $tanh ϕ = v / c$ (equivalently $γ = cosh ϕ$ , $γ v / c = sinh ϕ$ ). Show that under a boost of rapidity $ϕ$ along the $x$ -axis, the four-momentum components transform as

(E^{'} / c p_{x}^{'}) = (cosh ϕ - sinh ϕ - sinh ϕ cosh ϕ) (E / c p_{x}) .

Conclude that the boost is a hyperbolic rotation in the $(E / c, p_{x})$ plane.

Hint

Apply the Lorentz boost matrix to $p^{μ} = (E / c, p_{x}, p_{y}, p_{z})$ and substitute $γ = cosh ϕ$ , $γ β = sinh ϕ$ .

Answer

The Lorentz boost matrix in $(γ, β)$ form acts on the time- $x$ block as $(p _{x}^{'} E ^{'} / c) = (- γ β γ γ - γ β) (p _{x} E / c)$ . Substituting $γ = cosh ϕ$ and $γ β = sinh ϕ$ :

E^{'} / c = cosh ϕ \cdot E / c - sinh ϕ \cdot p_{x},

p_{x}^{'} = - sinh ϕ \cdot E / c + cosh ϕ \cdot p_{x} .

The matrix has the form $(- s i n h ϕ c o s h ϕ c o s h ϕ - s i n h ϕ)$ , which is a hyperbolic rotation in the $(E / c, p_{x})$ plane. Successive boosts of rapidities $ϕ_{1}, ϕ_{2}$ along the same axis compose by rapidity addition: the total rapidity is $ϕ_{1} + ϕ_{2}$ , and the velocity-addition formula $v = (v_{1} + v_{2}) / (1 + v_{1} v_{2} / c^{2})$ is the image of this linear addition under $v / c = tanh ϕ$ . The Lorentz group restricted to collinear boosts is $SO (1, 1)$ , the hyperbolic-rotation group.

Exercise 10 (hard, symbolic).

Show that the Lagrangian $L = - m c^{2} 1 - v^{2} / c^{2}$ for a free relativistic particle yields the relativistic equation of motion $d p^{μ} / d τ = 0$ via the Euler-Lagrange equations. Verify that this Lagrangian reproduces $p = γ m v$ as the canonical three-momentum and $E = γ m c^{2}$ as the Hamiltonian.

Hint

Compute the canonical momentum $p_{i} = \partial L / \partial v^{i}$ , then $H = p \cdot v - L$ , and check that the Euler-Lagrange equations $d / d t (\partial L / \partial v^{i}) = \partial L / \partial x^{i}$ reduce to $d p / d t = 0$ for a free particle.

Answer

$L = - m c^{2} / γ$ where $γ = (1 - v^{2} / c^{2})^{- 1/2}$ . Compute $\partial L / \partial v^{i} = - m c^{2} \cdot \partial (1 - v^{2} / c^{2})^{1/2} / \partial v^{i} = - m c^{2} \cdot \frac{1}{2} (1 - v^{2} / c^{2})^{- 1/2} \cdot (- 2 v^{i} / c^{2}) = m v^{i} / 1 - v^{2} / c^{2} = γ m v^{i}$ . So the canonical three-momentum is $p = γ m v$ . The Hamiltonian is $H = p \cdot v - L = γ m v^{2} + m c^{2} / γ = m c^{2} (γ v^{2} / c^{2} + 1/ γ) = m c^{2} (γ^{2} v^{2} / c^{2} + 1) / γ = m c^{2} \cdot γ$ . The last step uses $γ^{2} v^{2} / c^{2} + 1 = γ^{2}$ which follows from $γ^{2} = 1 + γ^{2} v^{2} / c^{2}$ . So $H = E = γ m c^{2}$ . For a free particle the Euler-Lagrange equations give $d p / d t = 0$ , which combined with $d E / d t = 0$ yields $d p^{μ} / d τ = (1/ γ) d p^{μ} / d t = 0$ .

Advanced results Master

The four-momentum as a Noether charge

By Noether's first theorem applied to the spacetime translation symmetry $x^{μ} \to x^{μ} + a^{μ}$ of a Lorentz-invariant action $S = \int L d^{4} x$ , the conserved current associated with translations in the $ν$ -direction is the stress-energy tensor $$ T^{\mu\nu} = \frac{\partial \mathcal{L}}{\partial(\partial_\mu \phi)},\partial^\nu \phi - \eta^{\mu\nu}\mathcal{L}, $$ where $ϕ$ stands for the dynamical fields. The integrated charges $$ P^\nu = \frac{1}{c}\int T^{0\nu},d^3 x $$ form a four-vector and are conserved under translations: $\partial_{μ} T^{μν} = 0$ in flat spacetime. For a single point particle, this construction recovers $P^{μ} = p^{μ} = m u^{μ}$ . The Noether-charge interpretation makes the four-momentum a fundamental object of Lorentz-invariant field theory, not merely a kinematic auxiliary.

Theorem 1 (Noether, 1918). For a translation-invariant Lagrangian field theory in flat spacetime, the integrated stress-energy tensor $P^{μ} = (1/ c) \int T^{0 μ} d^{3} x$ is a conserved four-vector, and for a point particle it equals $m u^{μ}$ . (Noether 1918 Nachr. Königl. Ges. Wiss. Göttingen.)

Photons and the de Broglie-Einstein relation

For a plane electromagnetic wave with angular frequency $ω$ and wave vector $k$ satisfying $ω = ∣ k ∣ c$ , the photon four-momentum is $$ p^\mu = \hbar,k^\mu, \qquad k^\mu := (\omega/c, \mathbf{k}). $$ The wave four-vector $k^{μ}$ is null: $η_{μν} k^{μ} k^{ν} = - ω^{2} / c^{2} + ∣ k ∣^{2} = 0$ , consistent with $η_{μν} p^{μ} p^{ν} = 0$ for a massless particle. Einstein's 1905 photoelectric paper introduced $E = ℏ ω$ ; de Broglie's 1924 thesis introduced $p = ℏ k$ . The four-vector relation $p^{μ} = ℏ k^{μ}$ unifies them.

The Pauli-Lubanski four-vector and Wigner classification

For a particle of mass $m$ , define the Pauli-Lubanski four-vector $$ W^\mu := -\tfrac{1}{2},\varepsilon^{\mu\nu\rho\sigma},p_\nu,J_{\rho\sigma}, $$ where $J_{ρ σ}$ are the generators of Lorentz rotations and $ε^{μν ρ σ}$ is the totally antisymmetric Levi-Civita tensor. $W^{μ}$ commutes with $p^{μ}$ , and the two Casimir invariants of the Poincaré algebra are $p^{μ} p_{μ} = - m^{2} c^{2}$ and $W^{μ} W_{μ}$ .

Theorem 2 (Wigner, 1939). The unitary irreducible representations of the proper orthochronous Poincaré group are classified by the pair $(m^{2}, s)$ , where $m^{2} c^{4} = - p^{μ} p_{μ} c^{2} \geq 0$ is the mass-squared and $s$ is the spin. For $m > 0$ : $W^{μ} W_{μ} = - m^{2} s (s + 1) ℏ^{2} c^{2}$ with $s \in {0, 1/2, 1, 3/2, \dots}$ . For $m = 0$ : $W^{μ}$ is null and proportional to $p^{μ}$ , with helicity $h = S \cdot \hat{p}$ as the substantive label. (Wigner 1939 Ann. Math. 40, 149.)

This classification underpins the entire Standard Model: every elementary particle is identified by a $(m, s)$ pair labelling an irreducible representation of the Poincaré group.

Mandelstam variables for $2 \to 2$ scattering

For a process $1 + 2 \to 3 + 4$ with four-momenta $p_{1}, p_{2}, p_{3}, p_{4}$ , define the Mandelstam invariants: $$ s := -(p_1 + p_2)^\mu(p_1 + p_2)\mu c^2, $$ $$ t := -(p_1 - p_3)^\mu(p_1 - p_3)\mu c^2, $$ $$ u := -(p_1 - p_4)^\mu(p_1 - p_4)_\mu c^2. $$

These are Lorentz invariants: $s$ is the square of the centre-of-momentum energy, $t$ is the momentum-transfer squared (relevant to forward scattering), and $u$ is the cross-channel variable. They satisfy the constraint $$ s + t + u = \sum_{i=1}^{4} m_i^2 c^4. $$

Theorem 3 (Mandelstam, 1958). For a $2 \to 2$ scattering process, the on-shell amplitude $M (s, t, u)$ is a function of two of the three Mandelstam variables, with the third determined by the constraint. Crossing symmetry relates $M$ in the $s$ -, $t$ -, and $u$ -channels via analytic continuation in the complex $s$ -, $t$ -, $u$ -plane. (Mandelstam 1958 Phys. Rev. 112, 1344.)

Mandelstam-variable analysis is the workhorse of perturbative quantum field theory: scattering amplitudes are expressed as functions of $(s, t)$ , and the analytic structure encodes crossing symmetry, unitarity, and dispersion relations.

Thomas precession

A non-collinear sequence of Lorentz boosts does not return the same frame: it composes to a boost followed by a rotation. The angle of the residual rotation is the Thomas rotation, and the rate at which the spin axis of an accelerated electron precesses relative to the lab frame is the Thomas precession rate.

Theorem 4 (Thomas, 1926). An electron in a circular orbit of angular frequency $ω_{orb}$ at speed $v$ experiences a Thomas precession of its spin axis with angular frequency $$ \omega_T = (\gamma - 1),\omega_{\rm orb}. $$ For non-relativistic speeds, $ω_{T} \approx (v^{2} /2 c^{2}) ω_{orb}$ , which produces the factor-of-1/2 correction (the "Thomas factor") in atomic fine structure. (Thomas 1926 Nature 117, 514.)

The Thomas factor reconciles the spin-orbit coupling computed in the electron rest frame with the experimentally observed atomic fine-structure splittings. Before Thomas's 1926 paper, atomic spectroscopy disagreed with the naive spin-orbit Hamiltonian by a factor of 2; Thomas's relativistic kinematics resolved the discrepancy.

Relativistic action and the principle of extremal proper time

The Lorentz-invariant action for a free relativistic point particle is $$ S = -mc^2\int d\tau = -mc\int\sqrt{-\eta_{\mu\nu},dx^\mu,dx^\nu}. $$

The Euler-Lagrange equations from this action give the geodesic equation $d^{2} x^{μ} / d τ^{2} = 0$ , corresponding to inertial (straight) worldlines. The principle of extremal action becomes the principle of extremal proper time: the classical worldline of a free particle is the one that maximises proper time between two events. For a charged particle in an electromagnetic field with four-potential $A^{μ} = (ϕ / c, A)$ , the action acquires a coupling term: $$ S = -mc\int\sqrt{-\eta_{\mu\nu},dx^\mu,dx^\nu} + \frac{q}{c}\int A_\mu,dx^\mu. $$

The Euler-Lagrange equations now give the covariant Lorentz force $$ \frac{dp^\mu}{d\tau} = \frac{q}{c},F^{\mu\nu},u_\nu, \qquad F^{\mu\nu} := \partial^\mu A^\nu - \partial^\nu A^\mu. $$

The spatial part recovers the familiar Lorentz force $F = q (E + v \times B)$ with relativistic momentum, while the time part gives the power equation $d E / d t = q E \cdot v$ .

The stress-energy tensor of a continuous medium

For a perfect fluid with energy density $ρ c^{2}$ (measured in its rest frame), pressure $p$ , and four-velocity $u^{μ}$ , the stress-energy tensor is $$ T^{\mu\nu} = (\rho c^2 + p),u^\mu u^\nu/c^2 + p,\eta^{\mu\nu}. $$

Conservation $\partial_{μ} T^{μν} = 0$ reduces in the non-relativistic limit to the continuity equation $\partial_{t} ρ + \nabla \cdot (ρ v) = 0$ and the Euler equation $ρ (\partial_{t} v + v \cdot \nabla v) = - \nabla p$ . The integrated four-momentum $P^{μ} = \int T^{0 μ} d^{3} x$ is conserved.

Theorem 5. For a closed system in flat Minkowski spacetime, the integrated four-momentum $P^{μ} = \int T^{0 μ} d^{3} x$ is conserved and transforms as a four-vector. In general relativity, the conservation law generalises to $\nabla_{μ} T^{μν} = 0$ with the covariant derivative replacing the partial derivative, and $T^{μν}$ becomes the source of the Einstein field equations $G^{μν} = 8 π G T^{μν} / c^{4}$ .

Cosmological redshift of photon four-momentum

In a Friedmann-Lemaître-Robertson-Walker spacetime with scale factor $a (t)$ , the four-momentum of a photon is parallel-transported along the null geodesic, and its frequency redshifts as $ω \propto 1/ a (t)$ . The observed redshift $z$ for light emitted at scale factor $a_{emit}$ and observed at $a_{obs} = 1$ is $$ 1 + z = \frac{a_{\rm obs}}{a_{\rm emit}} = \frac{\omega_{\rm emit}}{\omega_{\rm obs}}. $$

The four-momentum machinery developed here is the kinematic foundation of observational cosmology: cosmic-microwave-background photons emitted at $z \approx 1100$ have $ω_{obs} / ω_{emit} \approx 1/1100$ , corresponding to a temperature drop from approximately 3000 K at the surface of last scattering to 2.7 K today. The Sunyaev-Zel'dovich effect (1972) — a Compton-scattering shift in the CMB spectrum through hot galaxy clusters — is computed directly from the four-momentum kinematics of the Compton-scattering formula derived in the Intermediate tier above.

Centre-of-momentum frame and the Lorentz boost to it

For a system of $N$ particles with four-momenta $p_{a}^{μ}$ , the total four-momentum $P^{μ} = \sum_{a} p_{a}^{μ}$ is conserved and transforms as a four-vector under Lorentz transformations. The centre-of-momentum (CM) frame is the inertial frame in which the spatial total momentum vanishes: $\sum_{a} p_{a} = 0$ . The Lorentz boost from the lab to the CM frame has velocity $$ \mathbf{v}{\rm CM} = \frac{c^2,\mathbf{P}{\rm tot}}{E_{\rm tot}}, $$ which is always subluminal because $∣ P_{tot} ∣ c < E_{tot}$ for any system with at least one massive particle (or any non-collinear collection of photons). In the CM frame $P^{μ} = (s / c, 0)$ where $s$ is the invariant centre-of-momentum energy.

For a fixed-target experiment in which a beam of energy $E_{b}$ and momentum $p_{b}$ strikes a target of mass $m_{t}$ at rest, the invariant CM energy grows only as the square root of the beam energy: $$ \sqrt{s} = \sqrt{m_b^2 c^4 + m_t^2 c^4 + 2 m_t c^2 E_b} \approx \sqrt{2 m_t c^2 E_b}\quad (E_b \gg m_b c^2, m_t c^2). $$

For colliding beams of equal energy and mass, $s = 2 E_{beam}$ — linear in the beam energy. This factor-of- $E$ disadvantage of fixed-target experiments is why high-energy physics has used colliders since the 1970s.

Geodesics, the equivalence principle, and the lift to general relativity

The free-particle action $S = - m c^{2} \int d τ$ generalises to a Lorentzian manifold $(M, g)$ by replacing the flat-spacetime proper time with the curved-spacetime arc length: $S = - m c \int - g_{μν} d x^{μ} d x^{ν}$ . The Euler-Lagrange equations of this action give the geodesic equation $$ \frac{d^2 x^\mu}{d\tau^2} + \Gamma^\mu{}{\nu\rho},\frac{dx^\nu}{d\tau}\frac{dx^\rho}{d\tau} = 0, $$ where $\Gamma^\mu{}{\nu\rho} $a r e t h e C h r i s t o f f e l sy mb o l so f t h e L e v i - C i v i t a co nn ec t i o n . T h e f o u r - m o m e n t u m$ p^\mu = m,dx^\mu/d\tau $i s p a r a l l e l - t r an s p or t e d a l o n g t h e g eo d es i c :$ \nabla_u p^\mu = 0 $, w i t h$ \nabla_u $t h eco v a r ian t d i r ec t i o na l d er i v a t i v e a l o n g$ u^\mu $. T h e f l a t - s p a ce t im eco n ser v a t i o n l a w$ dp^\mu/d\tau = 0 $b eco m es t h ec u r v e d - s p a ce t im es t a t e m e n tt ha t$ p^\mu$ has zero covariant derivative along the worldline.

The equivalence principle — that gravitational and inertial forces are locally indistinguishable — translates kinematically into the statement that, in any sufficiently small spacetime region around a point, one can choose coordinates in which the metric reduces to $η_{μν}$ and the Christoffel symbols vanish to first order. In such local inertial coordinates the special-relativistic four-momentum formalism applies exactly to leading order, with curvature effects appearing at second order through the Riemann tensor.

Synthesis. The four-velocity, four-momentum, and energy-momentum identity build toward the entire edifice of relativistic dynamics. The central insight is that energy and momentum are not independent quantities but different projections of a single Lorentz-covariant object, and this is exactly the structure that identifies kinematic quantities with Noether charges of spacetime symmetries. Putting these together with Wigner's classification, every elementary particle is labelled by an irreducible representation $(m, s)$ of the Poincaré group, and the energy-momentum identity $E^{2} = ∣ p ∣^{2} c^{2} + m^{2} c^{4}$ is the Casimir invariant $p^{μ} p_{μ} = - m^{2} c^{2}$ in disguise.

The Mandelstam-variable structure generalises the energy-momentum identity from a single-particle invariant to a 2-to-2-scattering invariant set $(s, t, u)$ , and the bridge is between high-energy collider phenomenology and the abstract analytic structure of scattering amplitudes. The action principle $S = - m c^{2} \int d τ$ identifies the classical worldline with the proper-time-extremising path; the foundational reason this works is that proper time is the only Lorentz-invariant scalar one can construct from a single worldline, and the principle of extremal action then forces inertial motion. The pattern recurs in general relativity: the geodesic equation in curved spacetime is the Euler-Lagrange equation of $S = - m c^{2} \int d τ$ with proper time replaced by the arc length of the curved-spacetime metric.

The Thomas precession identifies a non-Abelian feature of the Lorentz group hidden in non-collinear boosts: the composition of two boosts in different directions is a boost plus a rotation, and this rotation produces measurable atomic fine-structure splittings. The foundational reason fine structure is reduced by a factor of 2 compared with the naive computation is precisely this kinematic compositional twist. Appears again in 13.02.01 when the tensor formalism on smooth manifolds generalises the flat-spacetime four-momentum apparatus to general spacetimes, where the conservation law becomes $\nabla_{μ} T^{μν} = 0$ with covariant derivatives.

Full proof set Master

Proposition 1 (Källén triangle function and two-body decay phase space). For a particle of mass $M$ at rest decaying into two particles of masses $m_{1}$ and $m_{2}$ , the magnitude of the daughter three-momentum in the parent rest frame is $$ |\mathbf{p}|c = \frac{\sqrt{\lambda(M^2 c^4, m_1^2 c^4, m_2^2 c^4)}}{2 M c^2}, \qquad \lambda(a,b,c) := a^2 + b^2 + c^2 - 2ab - 2ac - 2bc. $$ The decay is kinematically allowed if and only if $M \geq m_{1} + m_{2}$ .

Proof. By conservation $p_{1}^{μ} + p_{2}^{μ} = P^{μ} = (M c, 0)$ with $p_{1} = - p_{2} := p$ . Define $∣ p ∣ = p$ . From the energy-momentum identity in the parent rest frame: $E_{1} = p^{2} c^{2} + m_{1}^{2} c^{4}$ , $E_{2} = p^{2} c^{2} + m_{2}^{2} c^{4}$ . Conservation of energy: $E_{1} + E_{2} = M c^{2}$ . So $$ \sqrt{p^2 c^2 + m_1^2 c^4} + \sqrt{p^2 c^2 + m_2^2 c^4} = M c^2. $$

Isolate one radical and square: $$ p^2 c^2 + m_1^2 c^4 = (M c^2 - \sqrt{p^2 c^2 + m_2^2 c^4})^2 = M^2 c^4 - 2 M c^2 \sqrt{p^2 c^2 + m_2^2 c^4} + p^2 c^2 + m_2^2 c^4. $$ The $p^{2} c^{2}$ terms cancel: $$ m_1^2 c^4 - M^2 c^4 - m_2^2 c^4 = -2 M c^2 \sqrt{p^2 c^2 + m_2^2 c^4}. $$

Divide and square: $$ 4 M^2 c^4 (p^2 c^2 + m_2^2 c^4) = (M^2 c^4 + m_2^2 c^4 - m_1^2 c^4)^2. $$ Expand and rearrange: $$ 4 M^2 c^4 \cdot p^2 c^2 = (M^2 c^4 + m_2^2 c^4 - m_1^2 c^4)^2 - 4 M^2 c^4 \cdot m_2^2 c^4. $$

The right side simplifies to the Källén function $λ (M^{2} c^{4}, m_{1}^{2} c^{4}, m_{2}^{2} c^{4})$ , so $$ p^2 c^2 = \frac{\lambda(M^2 c^4, m_1^2 c^4, m_2^2 c^4)}{4 M^2 c^4}. $$ Taking the square root gives the claimed formula. The Källén function factorises as $λ (a, b, c) = (a - (b + c)^{2}) (a - (b - c)^{2})$ for positive arguments; positivity requires $a > (b + c)^{2}$ , i.e., $M > m_{1} + m_{2}$ . The decay is kinematically allowed precisely in this regime. $□$

Proposition 2 (orthogonality of four-acceleration and four-velocity). Along any timelike worldline, $u^{μ} α_{μ} = 0$ .

Proof. The four-velocity satisfies $η_{μν} u^{μ} u^{ν} = - c^{2}$ , a constant. Differentiating with respect to proper time: $$ \frac{d}{d\tau}(\eta_{\mu\nu} u^\mu u^\nu) = \eta_{\mu\nu},\frac{du^\mu}{d\tau},u^\nu + \eta_{\mu\nu},u^\mu,\frac{du^\nu}{d\tau} = 2\eta_{\mu\nu},\alpha^\mu u^\nu = 2,u_\mu \alpha^\mu. $$ The total derivative vanishes because the magnitude is constant, so $u_{μ} α^{μ} = 0$ . $□$

Proposition 3 (rapidity additivity for collinear boosts). For two boosts of velocities $v_{1}, v_{2}$ along the same spatial axis, the composite boost has velocity $v = (v_{1} + v_{2}) / (1 + v_{1} v_{2} / c^{2})$ , equivalently rapidity $ϕ = ϕ_{1} + ϕ_{2}$ where $tanh ϕ_{i} = v_{i} / c$ .

Proof. The Lorentz boost matrix along the $x$ -axis with velocity $v$ has the form $$ \Lambda(\phi) = \begin{pmatrix} \cosh\phi & -\sinh\phi & 0 & 0 \ -\sinh\phi & \cosh\phi & 0 & 0 \ 0 & 0 & 1 & 0 \ 0 & 0 & 0 & 1 \end{pmatrix}, $$ with $tanh ϕ = v / c$ . The product of two such boosts is $$ \Lambda(\phi_1)\Lambda(\phi_2) = \begin{pmatrix} \cosh\phi_1\cosh\phi_2 + \sinh\phi_1\sinh\phi_2 & -(\cosh\phi_1\sinh\phi_2 + \sinh\phi_1\cosh\phi_2) & 0 & 0 \ -(\sinh\phi_1\cosh\phi_2 + \cosh\phi_1\sinh\phi_2) & \cosh\phi_1\cosh\phi_2 + \sinh\phi_1\sinh\phi_2 & 0 & 0 \ 0 & 0 & 1 & 0 \ 0 & 0 & 0 & 1 \end{pmatrix}. $$ By the hyperbolic addition formulas $cosh (ϕ_{1} + ϕ_{2}) = cosh ϕ_{1} cosh ϕ_{2} + sinh ϕ_{1} sinh ϕ_{2}$ and $sinh (ϕ_{1} + ϕ_{2}) = sinh ϕ_{1} cosh ϕ_{2} + cosh ϕ_{1} sinh ϕ_{2}$ , this product equals $Λ (ϕ_{1} + ϕ_{2})$ . The corresponding velocity is $v = c tanh (ϕ_{1} + ϕ_{2}) = c (tanh ϕ_{1} + tanh ϕ_{2}) / (1 + tanh ϕ_{1} tanh ϕ_{2}) = (v_{1} + v_{2}) / (1 + v_{1} v_{2} / c^{2})$ . $□$

Connections Master

SR postulates and Lorentz transformations 10.05.01. Supplies the framework of Lorentz transformations $Λ \in SO^{+} (3, 1)$ acting on Minkowski spacetime, on which the four-velocity and four-momentum live as covariant tensors. The transformation law $p^{' μ} = Λ^{μ}_{ν} p^{ν}$ is a direct consequence of the postulates, and the invariance of $p^{μ} p_{μ}$ is the algebraic statement that Lorentz transformations preserve the Minkowski inner product.
Relativistic kinematics and dynamics 10.05.02. Direct prerequisite. The relativistic kinematics unit develops $p = γ m v$ , $E = γ m c^{2}$ , and the energy-momentum relation in three-vector language. The present unit lifts these to the covariant four-vector formalism and derives the energy-momentum identity as the Minkowski-norm constancy of $p^{μ}$ . The Compton-scattering and threshold-energy computations there are recovered here as applications of four-momentum conservation and Mandelstam-variable analysis.
Covariant electrodynamics 10.06.01. The covariant Lorentz-force law $d p^{μ} / d τ = (q / c) F^{μν} u_{ν}$ uses the four-momentum and four-velocity defined here as the dynamical variables. The Faraday tensor $F^{μν}$ acts on $u^{μ}$ to produce $d p^{μ} / d τ$ , and the spatial components recover the three-vector Lorentz force $F = q (E + v \times B)$ . The minimal-coupling action $S = - m c^{2} \int d τ + (q / c) \int A_{μ} d x^{μ}$ unifies free-particle motion with electromagnetic interaction.
Dirac equation 12.11.01. The quantum-mechanical analog of the energy-momentum identity is the Dirac equation $γ^{μ} p_{μ} = m c$ , obtained by "taking the square root" of $p^{μ} p_{μ} = - m^{2} c^{2}$ at the operator level. The requirement that the square root yield a first-order linear differential equation forces the introduction of $4 \times 4$ Dirac matrices and four-component spinor wavefunctions, predicting antimatter and spin-1/2 as kinematic consequences of relativistic invariance. The energy-momentum identity is the classical foundation on which relativistic quantum mechanics rests.
Tensors on smooth manifolds 13.02.01. The general-relativistic generalisation replaces flat-spacetime four-vectors with tensors on a Lorentzian manifold and the conservation law $\partial_{μ} T^{μν} = 0$ with the covariant version $\nabla_{μ} T^{μν} = 0$ . The four-momentum is parallel-transported along geodesics, generalising free-particle inertial motion to curved spacetime. The four-momentum framework lifts essentially unchanged from special to general relativity, with only the geometric setting modified.

Historical and philosophical context Master

Einstein 1905 "Zur Elektrodynamik bewegter Körper" ^{[Einstein1905a]} derived the Lorentz transformation from the postulates of relativity and the constancy of the speed of light, then deduced relativistic momentum addition and the velocity-addition law. The companion paper ^{[Einstein1905b]} "Ist die Trägheit eines Körpers von seinem Energieinhalt abhängig?" derived $E = m c^{2}$ as the inertia of energy: a body emitting electromagnetic radiation of energy $L$ loses mass $L / c^{2}$ . The 1907 review article ^{[Einstein1907]} "Über das Relativitätsprinzip und die aus demselben gezogenen Folgerungen" (Jahrbuch der Radioaktivität und Elektronik 4, 411-462) consolidated the rest-energy formulation. The four-momentum as a unified geometric object appeared with Minkowski's 1908 Cologne lecture ^{[Minkowski1908]} "Raum und Zeit": "Von Stund' an sollen Raum für sich und Zeit für sich völlig zu Schatten herabsinken, und nur noch eine Art Union der beiden soll Selbständigkeit bewahren."

Minkowski's reformulation cast energy and momentum as a single four-vector $p^{μ} = (E / c, p)$ , and the invariant $η_{μν} p^{μ} p^{ν} = - m^{2} c^{2}$ unified the separate Newtonian conservation laws into a single covariant identity. The geometric viewpoint was initially resisted (Einstein himself was skeptical) but became essential when general relativity required curved-spacetime tensor analysis.

The Noether-charge interpretation ^{[Noether1918]} of four-momentum awaited Emmy Noether's 1918 theorem, which identified the integrated stress-energy tensor as the conserved current of spacetime translation symmetry. This connection made four-momentum a fundamental object of Lorentz-invariant field theory rather than a kinematic auxiliary, and it is the modern starting point for relativistic quantum field theory.

Thomas's 1926 paper ^[Thomas1926] on the spinning electron solved a long-standing puzzle in atomic spectroscopy: the spin-orbit interaction computed naively (in the electron rest frame, then transformed back to the lab) gave a fine-structure splitting twice as large as observed. Thomas showed that the non-collinear boost composition contributes a kinematic precession at half the rate, exactly cancelling the discrepancy. The "Thomas factor" of 1/2 in atomic fine structure is a relativistic kinematic effect with no Newtonian analog.

The Wigner 1939 classification ^[Wigner1939] of unitary irreducible representations of the Poincaré group made $(m, s)$ the universal labels of elementary particles. The Pauli-Lubanski four-vector $W^{μ}$ introduced by Pauli 1939 and Lubanski 1942 ^{[Lubanski1942]} is the second Casimir of the Poincaré algebra, and its squared norm for massive particles equals $- m^{2} s (s + 1) ℏ^{2} c^{2}$ . Every elementary particle in the Standard Model is identified by its Wigner $(m, s)$ pair: photon $(0, 1)$ with helicity $\pm 1$ , electron $(0.511 MeV / c^{2}, 1/2)$ , Higgs $(125 GeV / c^{2}, 0)$ , graviton (postulated) $(0, 2)$ .

The Mandelstam 1958 paper ^{[Mandelstam1958]} introduced the $(s, t, u)$ invariants for $2 \to 2$ scattering and the corresponding analytic continuation, which became foundational for $S$ -matrix theory and dispersion relations. Mandelstam's framework underlies the analytic-bootstrap programme of the 1960s and the modern amplitude-based approach to quantum field theory.

The interplay between $E = m c^{2}$ and nuclear physics emerged in the 1930s with the discovery of nuclear binding-energy defects. A deuteron $^{2}$ H has rest energy approximately $2.224$ MeV less than the sum of its constituent proton and neutron rest energies, and this mass defect equals the deuteron binding energy. Uranium-235 fission releases roughly $200$ MeV per nucleus, corresponding to a mass defect of about 0.1% of the total nuclear mass. The Bethe-Weizsäcker semi-empirical mass formula (Bethe and Bacher 1936 Rev. Mod. Phys. 8, 82; Weizsäcker 1935 Z. Phys. 96, 431) systematised the dependence of binding energy on nuclear mass number, identifying iron-56 as the most tightly bound nucleus per nucleon. The applications to nuclear reactors and weapons followed during the 1940s as direct macroscopic realisations of $E = m c^{2}$ .

Antimatter prediction and discovery provided dramatic confirmation of the energy-momentum identity in quantum field theory. Dirac's 1928 equation, derived by demanding a first-order linear wave equation consistent with $E^{2} = ∣ p ∣^{2} c^{2} + m^{2} c^{4}$ , predicted negative-energy solutions that Dirac interpreted as a sea of filled states with vacancies behaving as positively-charged particles of electron mass. Anderson's 1932 Phys. Rev. 41 discovery of the positron in cosmic-ray cloud-chamber tracks confirmed this prediction, and the subsequent observation of pair production $γ + Z \to e^{+} + e^{-} + Z$ (with the nucleus absorbing recoil momentum) demonstrated direct conversion of electromagnetic energy into rest mass. The threshold $E_{γ} > 2 m_{e} c^{2} = 1.022$ MeV follows immediately from four-momentum conservation and the energy-momentum identity, with the nucleus required to satisfy the kinematic constraint that single-photon pair production in vacuum is forbidden because two massive particles cannot have total four-momentum equal to that of a single photon.

The lineage Einstein-Minkowski-Noether-Thomas-Wigner-Mandelstam traces a single thematic arc: the progressive identification of relativistic kinematic quantities with the representation theory of the Poincaré group. The energy-momentum identity is the Casimir constraint of the simplest non-vanishing unitary representation of the spacetime symmetry group, and the entire subsequent development of relativistic quantum field theory rests on this identification.

Bibliography Master

@article{Einstein1905a,
  author = {Einstein, Albert},
  title = {Zur Elektrodynamik bewegter K\"orper},
  journal = {Annalen der Physik},
  volume = {17},
  pages = {891--921},
  year = {1905},
}

@article{Einstein1905b,
  author = {Einstein, Albert},
  title = {Ist die Tr\"agheit eines K\"orpers von seinem Energieinhalt abh\"angig?},
  journal = {Annalen der Physik},
  volume = {18},
  pages = {639--641},
  year = {1905},
}

@misc{Minkowski1908,
  author = {Minkowski, Hermann},
  title = {Raum und Zeit},
  note = {Lecture at the 80th meeting of German Natural Scientists, Cologne, September 21},
  year = {1908},
}

@article{Einstein1907,
  author = {Einstein, Albert},
  title = {\"Uber das Relativit\"atsprinzip und die aus demselben gezogenen Folgerungen},
  journal = {Jahrbuch der Radioaktivit\"at und Elektronik},
  volume = {4},
  pages = {411--462},
  year = {1907},
}

@article{Thomas1926,
  author = {Thomas, L. H.},
  title = {The motion of the spinning electron},
  journal = {Nature},
  volume = {117},
  pages = {514},
  year = {1926},
}

@article{Wigner1939,
  author = {Wigner, Eugene P.},
  title = {On unitary representations of the inhomogeneous {L}orentz group},
  journal = {Annals of Mathematics},
  volume = {40},
  pages = {149--204},
  year = {1939},
}

@article{Mandelstam1958,
  author = {Mandelstam, S.},
  title = {Determination of the pion-nucleon scattering amplitude from dispersion relations and unitarity},
  journal = {Physical Review},
  volume = {112},
  pages = {1344--1360},
  year = {1958},
}

@article{Lubanski1942,
  author = {Lubanski, J. K.},
  title = {Sur la th\'eorie des particules \'el\'ementaires de spin quelconque. I},
  journal = {Physica},
  volume = {9},
  pages = {310--324},
  year = {1942},
}

@article{Noether1918,
  author = {Noether, Emmy},
  title = {Invariante Variationsprobleme},
  journal = {Nachrichten der K\"oniglichen Gesellschaft der Wissenschaften zu G\"ottingen, Mathematisch-physikalische Klasse},
  pages = {235--257},
  year = {1918},
}

@book{MisnerThorneWheeler1973,
  author = {Misner, Charles W. and Thorne, Kip S. and Wheeler, John Archibald},
  title = {Gravitation},
  publisher = {W. H. Freeman},
  year = {1973},
}

@book{Weinberg1972,
  author = {Weinberg, Steven},
  title = {Gravitation and Cosmology: Principles and Applications of the General Theory of Relativity},
  publisher = {Wiley},
  year = {1972},
}

@book{Rindler1991,
  author = {Rindler, Wolfgang},
  title = {Introduction to Special Relativity},
  edition = {2},
  publisher = {Oxford University Press},
  year = {1991},
}

@book{LandauLifshitz1980,
  author = {Landau, L. D. and Lifshitz, E. M.},
  title = {The Classical Theory of Fields},
  edition = {4},
  publisher = {Pergamon},
  year = {1980},
}

@book{Hartle2003,
  author = {Hartle, James B.},
  title = {Gravity: An Introduction to {E}instein's General Relativity},
  publisher = {Pearson},
  year = {2003},
}

Prerequisites

10.05.01
10.05.02

Used in

10.06.01
12.11.01

Tier anchors

beginner: Hartle, *Gravity* (Pearson, 2003), §4-5; Resnick, *Introduction to Special Relativity* (Wiley, 1968)
intermediate: Rindler, *Introduction to Special Relativity*, 2e (Oxford UP, 1991), §4; Hartle, *Gravity*, §5
master: Misner-Thorne-Wheeler, *Gravitation* (Freeman, 1973), §2; Weinberg, *Gravitation and Cosmology* (Wiley, 1972), §2; Landau-Lifshitz Vol. 2, *The Classical Theory of Fields*, 4e (Pergamon, 1980), §6-9

References

Einstein, A. — Zur Elektrodynamik bewegter Körper, Annalen der Physik 17 (1905) · pp. 891-921
Einstein, A. — Ist die Trägheit eines Körpers von seinem Energieinhalt abhängig?, Annalen der Physik 18 (1905) · pp. 639-641
Minkowski, H. — Raum und Zeit, Cologne lecture (1908) · 80th meeting of German Natural Scientists
Einstein, A. — Über das Relativitätsprinzip und die aus demselben gezogenen Folgerungen, Jahrbuch der Radioaktivität und Elektronik 4 (1907) · pp. 411-462
Thomas, L. H. — The motion of the spinning electron, Nature 117 (1926) · p. 514
Wigner, E. P. — On unitary representations of the inhomogeneous Lorentz group, Ann. Math. 40 (1939) · pp. 149-204
Mandelstam, S. — Determination of the pion-nucleon scattering amplitude from dispersion relations and unitarity, Phys. Rev. 112 (1958) · pp. 1344-1360
Lubanski, J. K. — Sur la théorie des particules élémentaires de spin quelconque, Physica 9 (1942) · pp. 310-324
Misner, C. W., Thorne, K. S., & Wheeler, J. A. — Gravitation (Freeman, 1973) · §2
Weinberg, S. — Gravitation and Cosmology (Wiley, 1972) · §2
Rindler, W. — Introduction to Special Relativity, 2e (Oxford UP, 1991) · §4
Landau, L. D. & Lifshitz, E. M. — The Classical Theory of Fields, 4e (Pergamon, 1980) · §6-9

Estimated time

beginner: 15m
intermediate: 40m
master: 70m

Intuition Beginner

Worked example Beginner

Second worked example: pion decay at rest

Check your understanding Beginner

Formal definition Intermediate+

Four-acceleration

Four-force

Counterexamples to common slips

Key theorem with proof Intermediate+

Worked example at intermediate level: Compton scattering

Exercises Intermediate+

Advanced results Master

The four-momentum as a Noether charge

Photons and the de Broglie-Einstein relation

The Pauli-Lubanski four-vector and Wigner classification

Mandelstam variables for 2→2 scattering

Thomas precession

Relativistic action and the principle of extremal proper time

The stress-energy tensor of a continuous medium

Cosmological redshift of photon four-momentum

Centre-of-momentum frame and the Lorentz boost to it

Geodesics, the equivalence principle, and the lift to general relativity

Full proof set Master

Connections Master

Historical and philosophical context Master

Bibliography Master

Mandelstam variables for $2 \to 2$ scattering