Special relativity

In short

Special relativity is Einstein's 1905 theory of how space, time, momentum and energy transform between inertial frames moving at constant relative velocity, built on just two postulates and valid for all speeds up to c. It replaced Galilean kinematics once the Michelson-Morley experiment showed no ether and no dependence of light speed on the observer's motion. B.Sc Modern Physics papers under UGC CBCS devote a full unit to it, and GATE/JAM/NET, JEST and TIFR papers ask numerical questions on time dilation, invariant mass and relativistic kinematics every year. It is applied knowledge for roles at the Department of Atomic Energy, BARC, RRCAT Indore, VECC Kolkata, IUCAA and ISRO - accelerators, nuclear reactions, GPS timing corrections and cosmic-ray detection all need it.

Explained

Inertial frames, Galilean failure and the two postulates

• An inertial frame is one in which a free body moves with constant velocity - Newton's first law holds. All frames moving uniformly with respect to an inertial frame are inertial.
• Galilean transformation: x′ = x − vt, y′ = y, z′ = z, t′ = t. It gives velocity addition u′ = u − v, so light would travel at c − v for an observer chasing it.
• Maxwell's equations predict c = 1/√(μ₀ε₀) ≈ 3 × 10⁸ m/s with no reference frame specified; they are not invariant under Galilean transformation. Something had to give.
• Michelson-Morley (1887): an interferometer with ~11 m arms should have shown a fringe shift of about 0.4 fringes as the Earth (30 km/s) swung through the supposed ether. The observed shift was essentially zero - a null result.
• Rescue attempts: ether drag (ruled out by stellar aberration and the Fizeau experiment), and the Lorentz-FitzGerald contraction hypothesis (correct formula, wrong physics - it assumed a real ether).
• Postulate 1 (principle of relativity): the laws of physics take the same form in all inertial frames; no experiment can detect absolute uniform motion.
• Postulate 2 (constancy of c): the speed of light in vacuum is c in every inertial frame, independent of the motion of source or observer.
• Consequence: there is no absolute time and no absolute simultaneity. Einstein's synchronisation convention - two clocks are synchronous if a light signal sent from A at t₁, reflected at B at t₂, returning at t₃ satisfies t₂ = (t₁ + t₃)/2.

Lorentz transformation and its immediate consequences

• Derivation outline (examined): demand (i) linearity, so uniform motion stays uniform, (ii) reduction to Galilean form as v/c → 0, (iii) invariance of the spherical light wavefront x² + y² + z² − c²t² = x′² + y′² + z′² − c²t′². Solving gives the boost along x.
• Define β = v/c and the Lorentz factor γ = 1/√(1 − β²) ≥ 1. Then x′ = γ(x − vt), t′ = γ(t − vx/c²); transverse coordinates unchanged. Inverse transformation: swap primes and replace v by −v.
• Relativity of simultaneity: two events simultaneous in S (Δt = 0) but separated by Δx are separated in S′ by Δt′ = −γvΔx/c². Simultaneity is frame-dependent; this is the root of most paradoxes.
• Time dilation: a clock at rest in S′ measures proper time τ; a laboratory observer records Δt = γτ. Moving clocks run slow - and the effect is reciprocal, each frame seeing the other's clocks slow.
• Length contraction: a rod of proper length L₀ at rest in S′ is measured in S (simultaneous end-position marking) as L = L₀/γ. Contraction is only along the direction of motion.
• Verification: cosmic-ray muons (proper lifetime 2.2 μs) reach sea level in large numbers only because of dilation; the Ives-Stilwell experiment (1938) confirmed the second-order time-dilation term in the *longitudinal* Doppler shift (the transverse shift was later checked by Mössbauer-rotor experiments); Rossi-Hall (1941) measured muon flux at two altitudes; particle lifetimes in every storage ring scale as γτ.
• No rigid-body paradox: the "pole and barn" and "ladder" paradoxes resolve entirely through simultaneity - the two ends are not marked at the same time in both frames.

Velocity addition, Doppler effect and aberration

• Longitudinal addition: u = (u′ + v)/(1 + u′v/c²). Putting u′ = c gives u = c for any v - the postulate is built in. Two sub-light speeds can never combine to exceed c.
• Transverse components: u_y = u′_y /[γ(1 + u′_x v/c²)].
• Relativistic Doppler (source and observer approaching along the line of sight): ν = ν₀√[(1+β)/(1−β)]; receding gives the reciprocal square root. Receding light is redshifted - the basis of Hubble's law and of quasar/galaxy redshift measurement at IUCAA and GMRT.
• Transverse Doppler shift: at θ = 90°, ν = ν₀/γ - a pure time-dilation effect with no classical counterpart, which is why it is the cleanest test.
• Relativistic aberration: cos θ′ = (cos θ + β)/(1 + β cos θ); forward beaming of radiation from relativistic electrons is why synchrotron light sources (Indus-1 and Indus-2 at RRCAT) emit a narrow forward cone.

Relativistic dynamics: momentum, mass and energy

• Newtonian p = mv is not conserved in all frames under Lorentz transformation. The correct form is p = γm₀u, with m₀ the rest mass (invariant mass).
• The old "relativistic mass" m = γm₀ is still in many Indian textbooks and question papers; modern usage keeps mass invariant and puts γ with velocity. Know both, and state which convention you use.
• Force: F = dp/dt, not m a. For motion along a straight line, F = γ³m₀a; transverse to the velocity, F = γm₀a - so a particle resists acceleration differently along and across its motion.
• Total energy E = γm₀c², rest energy E₀ = m₀c², kinetic energy K = (γ − 1)m₀c². Expanding for β ≪ 1 gives K ≈ ½m₀v² + (3/8)m₀v⁴/c² - the classical result plus corrections.
• Energy-momentum relation: E² = p²c² + m₀²c⁴. For a massless particle (photon), E = pc and p = hν/c = h/λ.
• Mass-energy equivalence: E = mc². 1 u = 931.49 MeV/c²; mass defect Δm explains nuclear binding energy, fission (~200 MeV per U-235 fission) and fusion, and underpins the ~7,480 MW of installed nuclear capacity operated by NPCIL (figures change with new units; check the current NPCIL/AERB listing).
• Useful relation: v/c = pc/E, and for ultra-relativistic particles E ≈ pc.

Spacetime, intervals and four-vectors

• Spacetime interval: Δs² = c²Δt² − Δx² − Δy² − Δz² is Lorentz-invariant - the same number for every inertial observer.
• Classification: timelike (Δs² > 0) - events causally connectable, a frame exists where they occur at the same place; lightlike/null (Δs² = 0) - connected by a light signal; spacelike (Δs² < 0) - no causal connection, a frame exists where they are simultaneous.
• Light cone: future cone, past cone and the causally disconnected "elsewhere". Causality forbids signalling outside the light cone; this is why no information travels faster than c (phase velocity and group velocity of a wave may exceed c without carrying information).
• Minkowski diagram: ct against x; the x′ and ct′ axes of a moving frame tilt inward symmetrically about the 45° light line, which makes relativity of simultaneity visible as non-parallel lines of constant time.
• Four-vectors: position x^μ = (ct, x, y, z); velocity u^μ = γ(c, u); momentum p^μ = (E/c, p) with invariant p·p = m₀²c². Four-vectors transform by the Lorentz matrix, and any scalar product of two four-vectors is invariant - the quickest way to solve collision problems.
• Proper time dτ = dt/γ is itself invariant - the time read by a clock carried along the worldline.
• Twin paradox: the travelling twin is not in a single inertial frame (she turns around and accelerates), so the situation is not symmetric; her elapsed proper time ∫dt/γ is genuinely less. Not a contradiction, and not explained by "acceleration damages clocks".
• Relativistic transformations of fields: E and B mix under a boost - a pure electric field in one frame has a magnetic component in another. Charge is invariant; charge density and current form a four-vector. (Full treatment belongs to electrodynamics, but the idea is examined.)

Limits, scope and where relativity hands over

• Special relativity is valid only for inertial frames and flat spacetime; accelerated frames and gravity need general relativity (1915), which adds gravitational time dilation.
• GPS/NavIC: satellite clocks lose ~7 μs/day to special-relativistic motion and gain ~45 μs/day from weaker gravity, a net ~38 μs/day correction applied on board. ISRO's NavIC/IRNSS system makes the same corrections; without them positions would drift kilometres per day.
• At v ≤ 0.1c the relativistic correction is under about 1%, so classical mechanics is adequate for all engineering speeds - hence "relativistic" effects show up only in accelerators, nuclear physics, astrophysics and precision timekeeping.
• Quantum mechanics plus special relativity gives the Dirac equation and quantum field theory; the non-relativistic Schrödinger equation is a low-speed limit.
• SI units follow the 2019 redefinition: c is exactly 299,792,458 m/s by definition, so the metre is defined from c and the second; India's primary time scale is maintained by CSIR-NPL, New Delhi, and the Legal Metrology Act, 2009 makes SI mandatory for trade.

Key points

• Postulate 1 - laws of physics identical in all inertial frames; absolute uniform motion is undetectable.
• Postulate 2 - c is the same in all inertial frames, independent of source or observer motion.
• Michelson-Morley null result - no ether; the expected ~0.4-fringe shift was absent.
• Lorentz factor γ = 1/√(1 − v²/c²); γ = 1 at rest, γ → ∞ as v → c.
• Time dilation - Δt = γτ; a moving clock runs slow, reciprocally for each observer.
• Length contraction - L = L₀/γ, only along the direction of relative motion.
• Proper time / proper length - measured in the frame where the clock or rod is at rest; both are the extreme values (largest length, smallest time).
• Relativity of simultaneity - Δt′ = −γvΔx/c²; events simultaneous in one frame are not in another.
• Velocity addition - u = (u′+v)/(1+u′v/c²); never exceeds c.
• Relativistic momentum p = γm₀v; it is p, not mv, that is conserved.
• Total energy E = γm₀c²; kinetic energy K = (γ−1)m₀c², which is not ½mv² at high speed.
• E² = p²c² + m₀²c⁴ - the master relation; photons have m₀ = 0 and E = pc.
• Invariant interval Δs² = c²Δt² − Δr²; classifies events as timelike, lightlike or spacelike.
• Rest mass is invariant - the same in every frame; "relativistic mass" is an older convention.
• Transverse Doppler shift ν = ν₀/γ - purely relativistic, no classical analogue.
• Twin paradox - resolved by the traveller's non-inertial turnaround, not by symmetry.
• 1 u = 931.49 MeV/c² - the working conversion for all nuclear mass-defect numerics.
• Classical limit - every relativistic formula reduces to the Newtonian one as v/c → 0.

Formulas and examples

Example 1 - Time dilation and length contraction at 0.8c. A spaceship of proper length 100 m passes Earth at v = 0.8c. Find γ, its length as measured from Earth, and the time an Earth observer assigns to a 1-hour process on board.

• β = 0.8, β² = 0.64, 1 − β² = 0.36, √0.36 = 0.6 → γ = 1/0.6 = 1.667.
• Length measured from Earth: L = 100/1.667 = 60 m.
• Earth time for the onboard hour: Δt = 1.667 × 1 h = 1.667 h = 1 h 40 min.
• The crew see Earth distances contracted by the same factor - the effect is symmetric for uniformly moving frames.

Example 2 - Cosmic-ray muon reaching sea level. A muon (proper lifetime τ = 2.2 μs) is created 10 km up and travels at v = 0.995c. Does it reach the ground?

• γ = 1/√(1 − 0.995²) = 1/√(1 − 0.990025) = 1/√0.009975 = 1/0.09987 = 10.01.
• Lab lifetime: Δt = 10.01 × 2.2 μs = 22.0 μs.
• Distance covered in lab frame: d = 0.995 × 2.998 × 10⁸ × 22.0 × 10⁻⁶ = 6,560 m ≈ 6.56 km mean travel, so a large fraction survives the 10 km - classically, 0.995c × 2.2 μs = only 657 m, and almost none would arrive.
• In the muon's own frame the atmosphere is contracted: 10 km/10.01 = 999 m, covered in 999/(0.995 × 2.998×10⁸) = 3.35 μs ≈ 1.5 proper lifetimes. Both frames agree on the observed fact.

Example 3 - Kinetic energy and the classical error. An electron is accelerated to v = 0.9c. Find γ, K, total energy and momentum (m₀c² = 0.511 MeV).

• 1 − 0.81 = 0.19, √0.19 = 0.4359 → γ = 2.294.
• E = 2.294 × 0.511 = 1.172 MeV; K = (2.294 − 1) × 0.511 = 0.662 MeV.
• pc = √(E² − E₀²) = √(1.172² − 0.511²) = √(1.3736 − 0.2611) = √1.1125 = 1.055 MeV, so p = 1.055 MeV/c.
• Classical estimate: ½m₀v² = 0.5 × 0.511 × 0.81 = 0.207 MeV - only 31% of the true value. Never use ½mv² above about 0.1c.

Example 4 - Velocity addition. Two probes leave Earth in opposite directions, each at 0.75c as seen from Earth. What is their relative speed?

• Work in the frame of probe A: Earth moves at v = 0.75c, probe B moves at u′ = 0.75c in the same direction relative to Earth.
• u = (0.75 + 0.75)c/(1 + 0.5625) = 1.5c/1.5625 = 0.96c - not 1.5c.
• Corresponding γ = 1/√(1 − 0.9216) = 1/√0.0784 = 1/0.28 = 3.571.

Example 5 - Mass defect and fusion energy. In deuterium-tritium fusion, ²H + ³H → ⁴He + n releases a mass defect of 0.018883 u. Find the energy per reaction and per gram of fuel mixture.

• E = 0.018883 × 931.49 = 17.59 MeV per reaction.
• In joules: 17.59 × 10⁶ × 1.602 × 10⁻¹⁹ = 2.818 × 10⁻¹² J.
• Reacting mass per event = 5 u = 5 × 1.6605 × 10⁻²⁷ = 8.303 × 10⁻²⁷ kg, so 1 g of D-T mixture gives 10⁻³/8.303×10⁻²⁷ = 1.204 × 10²³ reactions.
• Energy per gram = 1.204 × 10²³ × 2.818 × 10⁻¹² = 3.39 × 10¹¹ J ≈ 94,000 kWh - about ten million times the ~0.008 kWh (~30 kJ) from burning a gram of coal. This is the physics behind India's ITER participation through the Institute for Plasma Research, Gandhinagar.

Where it is used

• Accelerators and synchrotron sources: Indus-1 and Indus-2 at RRCAT Indore, the medical cyclotron at VECC Kolkata and India's LHC collaboration work - beam energy, bending magnets, RF frequency and radiation beaming are all computed relativistically.
• Nuclear power and radiation safety: NPCIL and BARC use mass-energy equivalence for fission/fusion energy balance, reactor physics and AERB dose calculations.
• Satellite navigation and timekeeping: ISRO's NavIC and GPS receivers apply combined special- and general-relativistic clock corrections; CSIR-NPL maintains Indian Standard Time with caesium clocks and hydrogen masers.
• Astrophysics and radio astronomy: redshift of galaxies and quasars, relativistic jets and cosmic-ray showers studied at IUCAA Pune, GMRT Khodad, TIFR and the Major Atmospheric Cherenkov Experiment at Hanle.
• Medical and industrial electron beams: linear accelerators for radiotherapy (Tata Memorial, AIIMS) and e-beam sterilisation run electrons above 6 MeV, where γ ≈ 13 and only relativistic dynamics applies.
• Exam and teaching roles: compulsory in B.Sc/M.Sc physics, GATE PH, JAM, CSIR-UGC NET and JEST; a standing requirement for scientist entry at DAE, ISRO and the defence research labs.

Interview questions

• Q: What exactly did the Michelson-Morley experiment prove? A: It found no fringe shift as the Earth moved through the supposed ether, showing that the speed of light does not depend on the observer's motion and that no ether frame exists.
• Q: If each observer sees the other's clock running slow, who is actually younger? A: For two frames in uniform relative motion neither is - the question has no frame-independent answer. Only if one observer turns around, as in the twin paradox, does the comparison become unambiguous, and that twin is younger.
• Q: Why can nothing with mass reach the speed of light? A: Its energy γm₀c² and momentum γm₀v diverge as v → c, so infinite energy would be needed; only massless particles travel at c.
• Q: Is length contraction a real physical shrinkage or an illusion? A: It is a real measurement result, not an optical illusion - but it is frame-dependent, because the two ends are located simultaneously only in the measuring frame.
• Q: What is proper time and why does it matter? A: The time measured by a clock at rest with respect to the events, dτ = dt/γ; it is Lorentz-invariant, so every observer agrees on it, which makes it the right variable for particle lifetimes.
• Q: State the energy-momentum relation and use it for a photon. A: E² = p²c² + m₀²c⁴; setting m₀ = 0 gives E = pc, so a photon carries momentum hν/c despite having no rest mass.
• Q: Does E = mc² mean mass can be converted into energy? A: It means rest mass is a form of energy; in nuclear reactions the mass defect appears as kinetic energy and radiation while total energy and momentum are conserved.
• Q: Can anything travel faster than light? A: No information or energy can. Phase velocity in a medium, the sweep of a laser spot or the expansion of space can exceed c without transmitting a signal.
• Q: Why is the invariant interval more useful than the time or distance separately? A: Δs² = c²Δt² − Δr² is the same for all inertial observers, so it classifies event pairs as causally connected or not and makes collision problems easy.
• Q: How do GPS satellites use relativity? A: Their orbital speed slows their clocks by about 7 μs/day while weaker gravity speeds them by about 45 μs/day; the net ~38 μs/day is corrected on board, else positions would err by kilometres each day.
• Q: At 0.1c, how much error does classical kinetic energy give? A: About 0.75% low - which is why classical mechanics suffices for everything below roughly 10% of c.
• Q: Why is "relativistic mass" discouraged today? A: Because mass as an invariant property is cleaner - γ belongs with velocity in p = γm₀v - and the longitudinal and transverse "masses" differ, so a single mass that grows with speed is misleading.

Common mistakes

• Using Δt = γΔτ the wrong way round, so the moving clock comes out fast; always identify which frame measures proper time first.
• Applying length contraction to the transverse dimensions, or contracting the proper length in the frame where the rod is at rest.
• Using ½mv² for kinetic energy at relativistic speeds, or writing K = γm₀c² and forgetting to subtract the rest energy.
• Adding velocities classically and getting a result above c, instead of using (u′+v)/(1+u′v/c²).
• Treating the twin paradox as symmetric and concluding it is a contradiction, or blaming acceleration itself rather than the broken inertial symmetry.
• Forgetting that simultaneity is frame-dependent - the source of every apparent paradox and of lost marks in Minkowski-diagram questions.
• Mixing units: using v in km/s with c in m/s, or u (atomic mass unit) with kg in a mass-defect calculation without the 931.49 MeV/c² conversion.

Related topics

Quantum mechanics basics · Atomic and molecular physics · Nuclear and particle physics · Classical mechanics and Lagrangian dynamics · Electromagnetic theory and Maxwell's equations · Astrophysics and cosmology