{
  "id": "conservation-of-energy",
  "title": "Conservation of Energy",
  "lang": "en",
  "source_lang": "en",
  "summary": "Conservation of energy states that energy is transferred or transformed without net creation or destruction, subject to the symmetries and boundaries of the physical system.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [
    "Law of conservation of energy",
    "Energy conservation principle"
  ],
  "generated_at": "2026-11-02",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/conservation-of-energy/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/conservation-of-energy/",
      "json": "https://aiwiki.page/zh/conservation-of-energy.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Conservation of energy is the physical principle that the total energy of an isolated system remains constant through its internal changes. Energy may pass between objects, move across boundaries, or change between kinetic, thermal, electromagnetic, chemical, and other forms, but these processes do not create or destroy energy. The principle unifies phenomena as different as falling bodies, electric generators, metabolism, and nuclear reactions, and provides one of the central constraints used in physics.\n\nIts precise formulation depends on the theory and the system under consideration. In ordinary mechanics and thermodynamics, it takes the form of a balance between stored energy and transfers by work, heat, radiation, or matter. In theories with an appropriate time-translation symmetry, it follows mathematically from that symmetry. In general relativity, local energy–momentum balance remains fundamental, but a single conserved total energy cannot be defined for every possible spacetime. These distinctions explain both the extraordinary range of the principle and the care needed when applying it to quantum processes, gravitation, and the expanding universe."
    },
    {
      "heading": "Historical development",
      "text": "The modern principle emerged from several initially separate traditions: the mechanics of motion, the study of heat, electrical experimentation, and physiology. During the seventeenth and eighteenth centuries, investigators recognized quantities that remained constant in idealized mechanical systems. Christiaan Huygens studied collisions, while Gottfried Wilhelm Leibniz promoted vis viva, proportional to \\(mv^2\\), as a measure associated with motion. This was not yet the modern concept of energy: its normalization differed from present-day kinetic energy, and the relationships among motion, heat, and chemical change remained unclear.\n\nThe theory of heat presented a major obstacle. Caloric theories treated heat as a conserved substance that moved between bodies. Frictional heating was difficult to reconcile with a fixed stock of such a substance. In 1798, Benjamin Thompson, Count Rumford, described experiments on the heat produced during cannon boring and argued that its apparently inexhaustible production pointed toward a connection with motion. These observations did not establish the complete conservation law, but they weakened the separation between mechanical processes and heat.\n\nIn the 1840s, several researchers developed broader accounts of interconversion. Julius Robert von Mayer argued in 1842 for a quantitative equivalence between mechanical work and heat. James Prescott Joule investigated electrical heating and the mechanical production of heat, including paddle-wheel experiments in which descending weights stirred water. The essential comparison was between a measurable loss of mechanical energy and a measurable temperature rise. Modern units express this relationship through the thermochemical calorie, defined as exactly \\(4.184\\) joules, although historical calories and experimental estimates were not all identical.\n\nIn 1847, Hermann von Helmholtz presented a general account connecting conservation with mechanics, heat, electricity, magnetism, and physiology. Rudolf Clausius and William Thomson, later Lord Kelvin, subsequently helped formulate thermodynamics by distinguishing energy conservation from restrictions on the conversion of heat into work. The first law of thermodynamics expressed the balance of energy; the second law explained why energetically balanced processes could nevertheless be impossible.\n\nThe twentieth century extended rather than discarded the principle. Albert Einstein established the relation between mass and energy in 1905. In 1918, Emmy Noether connected conservation laws with continuous symmetries. Quantum theory introduced discrete energy levels and probabilistic measurements, while general relativity revealed that global energy conservation depends on spacetime structure. The result was a principle more precise than the unrestricted statement that a single universal sum must always remain constant."
    },
    {
      "heading": "Systems, boundaries, and energy accounting",
      "text": "An energy balance begins by specifying a system and its boundary. A thermodynamic system may be a gas in a cylinder, an electrical circuit, an organism, or a region of the atmosphere. An isolated system exchanges neither matter nor energy with its surroundings. A closed system exchanges energy but not matter, while an open system can exchange both. These distinctions determine which terms belong in the balance and prevent changes caused by an external influence from being mistaken for failures of conservation.\n\nFor a fixed control region, a general accounting statement is\n\n\\[\\frac{dE_{\\mathrm{stored}}}{dt} = \\dot E_{\\mathrm{in}}-\\dot E_{\\mathrm{out}}.\\]\n\nThe transfer terms may include mechanical work, heat conduction, electromagnetic radiation, and energy carried by flowing matter. More detailed formulations separate these contributions according to the problem. A turbine, for example, receives energy with incoming steam and exports both shaft work and outgoing steam. Treating only its rotating shaft as the system would omit the main energy supply.\n\nEnergy is a scalar quantity, measured in the International System of Units in the joule, with\n\n\\[1\\ \\mathrm{J}=1\\ \\mathrm{kg\\,m^2\\,s^{-2}}.\\]\n\nPower is the rate of energy transfer or transformation, measured in watts, where \\(1\\ \\mathrm{W}=1\\ \\mathrm{J\\,s^{-1}}\\). A kilowatt-hour is a unit of energy, equal to \\(3.6\\) million joules, not a unit of power. The distinction matters whenever a device’s operating rate is compared with its accumulated consumption.\n\nThe division into different forms of energy is partly a matter of description. A spring’s elastic energy can be represented macroscopically as potential energy or microscopically through changes in atomic interactions. Thermal internal energy includes microscopic motion and interactions that are not individually tracked. These descriptions must not be added as though they referred to independent stores: doing so would count the same energy twice.\n\nSome energy values depend on conventions or reference frames. Ordinary gravitational potential energy may be assigned zero at an arbitrary reference height, and kinetic energy changes when the observer’s velocity changes. Conservation concerns a consistently defined balance in a specified description, not an identical numerical energy assigned by every observer. Likewise, the energy of an interacting system need not be uniquely attributable to its separate parts. Interaction energy belongs to the combined arrangement and must remain in the accounting even when its allocation between components is conventional."
    },
    {
      "heading": "Mechanical energy and conservative forces",
      "text": "In classical mechanics, conservation appears most simply in the exchange between kinetic energy and potential energy. For a particle of constant mass moving much more slowly than light,\n\n\\[K=\\frac12 mv^2.\\]\n\nThe work–energy theorem follows from Newton’s laws of motion: the net work done on a particle equals the change in its kinetic energy,\n\n\\[W_{\\mathrm{net}}=\\Delta K.\\]\n\nThis theorem remains useful even when mechanical energy alone is not conserved.\n\nA conservative force admits a potential \\(V\\) such that \\(\\mathbf F=-\\nabla V\\), within a region where the potential is well defined. Work by that force depends only on the endpoints, and work around a closed path is zero. If the potential has no explicit time dependence and no other forces transfer energy, then\n\n\\[E_{\\mathrm{mech}}=K+V\\]\n\nis constant. Near Earth’s surface, \\(V=mgh\\) approximates gravitational potential energy when the gravitational field can be treated as uniform. For an ideal spring, \\(V=\\tfrac12 kx^2\\). A falling object or oscillating spring therefore provides a direct example of energy changing form while the total remains fixed.\n\nThe qualifications are important. A falling ball transfers energy to surrounding air through drag, and a bouncing ball transfers energy into sound, deformation, and microscopic motion. Friction reduces the mechanical energy of the visibly moving components, but does not remove energy from the larger system. An idealized friction law, such as kinetic friction with approximately constant magnitude, describes the transfer without resolving all the microscopic interactions responsible for it.\n\nAn external agent can also change a system’s mechanical energy without friction. If a spring’s stiffness is varied, or a gravitational source is deliberately moved, the potential depends explicitly on time. For a particle acted on only by \\(-\\nabla V(\\mathbf r,t)\\),\n\n\\[\\frac{d}{dt}(K+V)=\\frac{\\partial V}{\\partial t}.\\]\n\nThe missing balance lies in the mechanism that changes the potential. A subsystem in a prescribed time-dependent environment is generally not isolated.\n\nRotational and extended-body problems require corresponding additional terms. A rigid rotor has kinetic energy \\(\\tfrac12 I\\omega^2\\), involving its moment of inertia. Deformable solids store elastic energy and support internal vibrations. Mechanical conservation is therefore not restricted to particles, but the state description must include every dynamically relevant store and transfer."
    },
    {
      "heading": "Time-translation symmetry and Noether’s theorem",
      "text": "The deepest general connection is between energy conservation and the uniformity of the laws of motion in time. If an experiment is shifted to a different starting time while all relevant physical conditions remain the same, its governing laws ordinarily remain unchanged. This is time-translation symmetry. Noether’s theorem establishes that continuous symmetries of an action give rise to conserved quantities, subject to appropriate assumptions about the dynamics and boundary conditions. Time translations correspond to energy; spatial translations correspond to momentum; spatial rotations correspond to angular momentum.\n\nFor generalized coordinates \\(q_i\\), a Lagrangian \\(L(q_i,\\dot q_i,t)\\) defines an action through its time integral. Along solutions of the Euler–Lagrange equations, the quantity\n\n\\[E_L=\\sum_i\\dot q_i\\frac{\\partial L}{\\partial\\dot q_i}-L\\]\n\nsatisfies\n\n\\[\\frac{dE_L}{dt}=-\\frac{\\partial L}{\\partial t}.\\]\n\nThus, when the Lagrangian has no explicit time dependence, \\(E_L\\) is conserved. For the common form \\(L=T-V\\), with a quadratic kinetic energy and a velocity-independent potential, this conserved quantity becomes \\(T+V\\).\n\nIn Hamiltonian mechanics, the corresponding result is expressed using the Hamiltonian \\(H(q,p,t)\\). Along a Hamiltonian trajectory,\n\n\\[\\frac{dH}{dt}=\\frac{\\partial H}{\\partial t}.\\]\n\nA time-independent Hamiltonian is therefore constant. The Hamiltonian generates evolution in the chosen time coordinate, although its identification with a particular physical energy requires attention to coordinate choices, constraints, and external driving. Explicitly time-dependent changes of variables can alter the appearance of the Hamiltonian without creating or destroying physical energy.\n\nIn field theories, Noether’s theorem produces a current rather than merely a single number. Energy density and energy flux satisfy a local continuity equation. Integrating this equation over a region gives a conserved total only when the flux across its boundary vanishes or is properly included. The theorem therefore connects symmetry, local transport, and global bookkeeping.\n\nNoether’s result also clarifies the scope of the principle. A time-dependent external field can break the time-translation symmetry of a subsystem, while the larger system that produces the field may still conserve energy. In curved spacetime, a suitable global time-translation symmetry may not exist at all. Energy conservation is consequently more informative when stated with its symmetry conditions than when presented as an unrestricted numerical rule."
    },
    {
      "heading": "The first law of thermodynamics",
      "text": "The first law extends energy accounting to processes in which microscopic degrees of freedom cannot be followed individually. The central quantity is internal energy, \\(U\\), a state function determined by the thermodynamic state. For a closed system, a common sign convention writes\n\n\\[\\Delta U=Q-W,\\]\n\nwhere \\(Q\\) is energy transferred into the system as heat and \\(W\\) is work done by the system on its surroundings. If overall motion or elevation changes, kinetic and gravitational potential energies must also appear.\n\nHeat and work are not substances stored inside a body. They name different modes of energy transfer across a boundary. Heating can increase internal energy, but a body does not contain a uniquely identifiable stock of “heat.” Likewise, work is associated with a process, not a state. Different paths between the same equilibrium states can involve different amounts of heat and work while yielding the same \\(\\Delta U\\).\n\nFor a simple compressible system, a quasistatic expansion against an external pressure gives boundary work\n\n\\[W=\\int p_{\\mathrm{ext}}\\,dV.\\]\n\nIn a reversible process, the external pressure differs only infinitesimally from the system pressure. In a sudden free expansion into a vacuum, by contrast, \\(p_{\\mathrm{ext}}=0\\), so no boundary work is done. For an isolated ideal gas, whose internal energy depends only on temperature, such an expansion leaves the temperature unchanged even though the volume increases. This illustrates why pressure, volume, work, and temperature cannot be related without specifying the process.\n\nOpen systems require the energy accompanying mass flow. The combination\n\n\\[h=u+pv\\]\n\nis specific enthalpy, where \\(u\\) is internal energy per unit mass and \\(v\\) is specific volume. The \\(pv\\) term accounts for the pressure work associated with moving material across the control surface. A steady-flow device with one inlet and one outlet obeys\n\n\\[\\dot Q-\\dot W_s = \\dot m\\left[ h_2-h_1+ \\frac{v_2^2-v_1^2}{2} +g(z_2-z_1) \\right],\\]\n\nwhere \\(v_1\\) and \\(v_2\\) here denote flow speeds and \\(\\dot W_s\\) is shaft work output.\n\nThe first law applies beyond equilibrium, but simple equilibrium variables may no longer describe every part of a system. Shock waves, rapid compression, reactive flows, and strong temperature gradients require local balances and constitutive relations. Conservation supplies the governing constraint; equations of state, heat-transfer laws, and material properties determine how a particular process unfolds."
    },
    {
      "heading": "Entropy, irreversibility, and available work",
      "text": "Conservation of energy does not imply that all energetically balanced transformations are possible. A warm object spontaneously transfers energy to a colder object, not the reverse in the absence of compensating changes. The second law of thermodynamics describes this directionality through entropy. For an isolated system,\n\n\\[\\Delta S\\geq 0.\\]\n\nReversible idealizations have zero total entropy production; real dissipative processes produce entropy. Energy remains conserved in both cases, but its capacity to produce useful work changes.\n\nA heat engine illustrates the distinction. An engine receiving heat \\(Q_h\\) from a hot reservoir and rejecting heat \\(Q_c\\) to a cold reservoir delivers work\n\n\\[W=Q_h-Q_c.\\]\n\nThe first law alone permits the formal possibility \\(Q_c=0\\). The second law excludes complete conversion of heat drawn cyclically from a single reservoir into work without other effects. For reservoirs at absolute temperatures \\(T_h\\) and \\(T_c\\), the ideal Carnot cycle has efficiency\n\n\\[\\eta_{\\mathrm{Carnot}}=1-\\frac{T_c}{T_h}.\\]\n\nAn engine operating between \\(600\\ \\mathrm{K}\\) and \\(300\\ \\mathrm{K}\\) therefore has a maximum efficiency of \\(50\\%\\), regardless of its working substance.\n\nThis limitation concerns energy quality rather than energy quantity. A joule of electrical energy can ideally be converted entirely into mechanical work, whereas a joule transferred as heat from a body only slightly warmer than its surroundings offers much less work potential. The maximum useful work obtainable as a system comes into equilibrium with a specified environment is called exergy. Unlike energy, exergy is destroyed by irreversibility. For an environment at temperature \\(T_0\\), the exergy destroyed is \\(T_0S_{\\mathrm{gen}}\\), under the assumptions of the standard exergy balance.\n\nThe distinction explains the technical language of “energy loss.” In a motor, electrical energy not delivered as shaft work may become internal energy in windings, bearings, and surrounding air. No energy disappears, but less remains available for the intended task. A refrigerator similarly requires work to transfer heat from a colder region to a warmer one. Its performance is constrained by both laws, not by conservation alone.\n\nA perpetual-motion machine of the first kind would supply net energy without a corresponding input or decrease in stored energy. A machine of the second kind would conserve energy while violating the second law, for example by cyclically converting heat from a single reservoir entirely into work. Separating these two claims is essential: energy balance is necessary for physical feasibility, but is not sufficient."
    },
    {
      "heading": "Electromagnetic energy and its transport",
      "text": "Electric and magnetic fields store energy and transport it through space. In vacuum, the classical electromagnetic energy density is\n\n\\[u_{\\mathrm{em}} = \\frac{\\varepsilon_0E^2}{2} + \\frac{B^2}{2\\mu_0},\\]\n\nwhere \\(\\varepsilon_0\\) and \\(\\mu_0\\) are the vacuum permittivity and permeability. The energy flux is described by the Poynting vector,\n\n\\[\\mathbf S=\\frac{1}{\\mu_0}\\mathbf E\\times\\mathbf B.\\]\n\nThese expressions show that energy transport is a property of the electromagnetic field, not merely of charged particles moving through a conductor.\n\nFrom Maxwell’s equations follows Poynting’s theorem,\n\n\\[\\frac{\\partial u_{\\mathrm{em}}}{\\partial t} +\\nabla\\cdot\\mathbf S =-\\mathbf J\\cdot\\mathbf E.\\]\n\nThe left side gives the local change in field energy plus outward flux. The right side accounts for work done by the electric field on matter, with \\(\\mathbf J\\) the electric current density. A positive \\(\\mathbf J\\cdot\\mathbf E\\) means that matter gains energy from the field. Combining this relation with the material energy balance yields conservation for the coupled system.\n\nA charging capacitor stores energy \\(\\tfrac12 CV^2\\), while an ideal inductor stores \\(\\tfrac12 LI^2\\). In a lossless oscillating circuit, energy moves between electric and magnetic storage. In a real circuit, electrical resistance converts part of that organized electromagnetic energy into microscopic motion through Joule heating. The familiar power relation \\(P=VI\\) is consistent with field transport: energy enters a resistor through the surrounding electromagnetic field rather than simply being carried along by individual electrons as a consumable cargo.\n\nRadiation adds another transport channel. Accelerating charges can emit electromagnetic radiation, and antennas exchange energy with propagating fields. A charged particle’s kinetic-energy decrease alone does not describe the full balance; radiation and, in detailed treatments, changes in the nearby field must be included. Absorption reverses the transfer, with field energy becoming material excitation, motion, or heat.\n\nIn matter, separating electromagnetic energy from material energy is more complicated than in vacuum. Polarization, magnetization, dispersion, and absorption involve microscopic responses. For ideal linear nondispersive media, familiar expressions involving \\(\\mathbf E\\cdot\\mathbf D\\) and \\(\\mathbf B\\cdot\\mathbf H\\) are useful, but they are not universal formulas for all materials. The conservation statement remains valid for the complete field–matter system even when assigning energy to its components requires a more detailed model."
    },
    {
      "heading": "Chemical reactions and biological transformations",
      "text": "In chemistry, energy conservation connects molecular structure with measurable heat and work. Reactants and products differ in electronic configurations, intermolecular interactions, and vibrational states. A chemical reaction can therefore transfer energy to or from its surroundings while conserving the total energy of the combined system. In ordinary chemical calculations, rest-mass changes are so small that masses are usually treated as separately conserved to excellent practical accuracy.\n\nA common misconception is that breaking a chemical bond releases energy. Separating bonded atoms requires energy relative to the bonded state. Energy is released when new bonds and interactions form, and an exothermic reaction releases more through product formation than it requires to disrupt the reactants. The net result depends on the complete reaction, including phase, solvent, pressure, and temperature. Bond-energy estimates are useful approximations, but precise thermochemical calculations use specified states and measured or calculated state functions.\n\nAt constant pressure, with only pressure–volume work, reaction heat equals the enthalpy change. Hess’s law follows from enthalpy being a state function: the enthalpy change between specified reactants and products is independent of the sequence of intermediate reactions. This permits reaction energies to be calculated from standard formation enthalpies. An activation energy, by contrast, concerns the kinetic barrier along a pathway, not the net energy difference between endpoints. A catalyst changes the reaction pathway and rate without changing that endpoint difference.\n\nBiological metabolism obeys the same accounting. In cellular respiration, the oxidation of organic molecules transfers part of their chemical free energy into electrochemical gradients and the synthesis of adenosine triphosphate. ATP hydrolysis can then be coupled to biosynthesis, transport, or mechanical motion. The phrase “energy-rich bond” is shorthand for a favorable overall reaction under specified conditions, not a claim that breaking one bond alone produces energy.\n\nPhotosynthesis couples absorbed light to chemical transformations. Photochemical charge separation initiates electron-transfer processes that support the production of reducing power and ATP; carbon fixation subsequently uses these resources to build organic compounds. Organisms neither create energy from nothing nor convert all absorbed energy into biomass.\n\nReaction spontaneity requires more than an energy balance. At constant temperature and pressure, Gibbs free energy combines enthalpic and entropic contributions. A process can be energetically balanced yet fail to proceed in a proposed direction without coupling to another process. Living systems maintain organization by exchanging matter and energy with their surroundings, while total entropy production remains nonnegative."
    },
    {
      "heading": "Microscopic conservation and statistical mechanics",
      "text": "Statistical mechanics explains thermodynamic behavior through the dynamics and probabilities of microscopic states. In a classical isolated system described by a time-independent Hamiltonian, the trajectory remains on a surface of constant total energy in phase space. The microcanonical ensemble represents equilibrium by assigning probabilities to accessible states within a specified energy range, together with other conserved quantities and constraints.\n\nA subsystem can nevertheless have a fluctuating energy. If it exchanges energy with a much larger reservoir, the combined system may be isolated while the subsystem is not. Under equilibrium conditions, its state probabilities contain the Boltzmann factor,\n\n\\[p_i=\\frac{e^{-\\beta E_i}}{Z}, \\qquad \\beta=\\frac{1}{k_{\\mathrm B}T},\\]\n\nwhere \\(Z\\) is the partition function and \\(k_{\\mathrm B}\\) is the Boltzmann constant. The spread of subsystem energies is compatible with conservation because the reservoir undergoes compensating changes.\n\nFor the canonical ensemble, energy fluctuations obey\n\n\\[\\left\\langle(\\Delta E)^2\\right\\rangle = k_{\\mathrm B}T^2C_V,\\]\n\nunder the usual equilibrium conditions, with \\(C_V\\) the constant-volume heat capacity. For many ordinary macroscopic systems away from singular regimes, the relative size of these fluctuations decreases approximately as the inverse square root of particle number. This explains why a thermodynamic energy can appear sharply defined even though its microscopic constituents are continually exchanging energy.\n\nIrreversible macroscopic behavior does not require microscopic energy destruction. In an isolated gas, collisions conserve the appropriate microscopic totals while redistributing energy among many degrees of freedom. A prepared concentration of motion or a temperature difference can spread into an enormous number of microscopic arrangements. Statistical explanations of entropy increase depend on such multiplicities, the choice of macroscopic description, and initial conditions. Conservation alone does not select a direction of time.\n\nCoarse-grained models can make energy appear to disappear if unresolved degrees of freedom are omitted. A friction term in the equation for a large particle removes kinetic energy from that particle; a compatible thermal description attributes the transfer to the surrounding medium. In equilibrium, fluctuating forces from the medium accompany dissipation, as expressed by fluctuation–dissipation relations. The same microscopic interactions that damp organized motion also generate Brownian motion. This connection is important when modeling molecular machines, colloids, and nanoscale devices, where energy transfers fluctuate appreciably from one realization to another."
    },
    {
      "heading": "Quantum mechanics and energy measurements",
      "text": "In quantum mechanics, the Hamiltonian operator \\(\\hat H\\) governs time evolution and represents the system’s energy. For a closed system satisfying the Schrödinger equation, the expectation value obeys\n\n\\[\\frac{d}{dt}\\langle\\hat H\\rangle = \\left\\langle\\frac{\\partial\\hat H}{\\partial t}\\right\\rangle.\\]\n\nA Hamiltonian with no explicit time dependence therefore has a constant mean energy. More strongly, under this unitary evolution the probabilities associated with its energy eigenspaces remain fixed.\n\nAn energy eigenstate has a definite measured energy, while a quantum superposition of different energy eigenstates generally does not. Repeated measurements on identically prepared systems can yield different values according to their quantum probabilities. This spread is not evidence that an isolated system randomly gains and loses energy during undisturbed evolution. The distinction is between a distribution of possible measurement outcomes and an actual energy transfer.\n\nTransitions require the coupling responsible for them to be included. An excited atom can emit a photon as its internal energy decreases. In a complete treatment, conservation includes the photon, the atom’s recoil, and any relevant environment. For absorption, the incident radiation supplies energy and momentum. If a classical time-dependent laser field is treated as an external drive, the atom’s Hamiltonian is explicitly time dependent and its energy need not remain constant; the source provides the missing balance.\n\nThe energy–time uncertainty principle is often misinterpreted as permission to violate conservation for short intervals. Unlike position and momentum, time and energy do not have one universal operator relation covering every use of the uncertainty principle. Lifetime–linewidth relations concern the spectral width of unstable states, while other relations concern limits on the speed of quantum evolution. None supplies a general license to borrow energy from nothing. Likewise, a virtual particle in a perturbative calculation is not an observable object temporarily exempt from physical conservation laws.\n\nMeasurement and open-system evolution require further care. An apparatus interacts with the measured system and can exchange energy with it. Simplified collapse rules or reduced equations do not necessarily track that exchange explicitly. A consistent accounting must include the apparatus, environment, and work needed to control the interaction. Conservation is thus imposed on the physical process as modeled, not automatically on a selected subsystem after its surroundings have been discarded."
    },
    {
      "heading": "Relativity, mass, and nuclear processes",
      "text": "Special relativity joins energy and momentum into a four-vector. For a free particle with invariant mass \\(m\\),\n\n\\[E^2=p^2c^2+m^2c^4,\\]\n\nwhere \\(c\\) is the speed of light. In the particle’s rest frame,\n\n\\[E_0=mc^2.\\]\n\nThis mass–energy equivalence means that rest mass contributes to total energy. It does not mean that energy has mass in every sense, or that the separate rest masses of all components must be conserved in an interaction.\n\nThe invariant mass of a composite system includes its internal kinetic and interaction energies, evaluated through its total energy and momentum. Heating a sealed object increases its mass slightly if the added energy remains inside. Conversely, a bound system has a lower rest energy than its separated constituents when energy has been released during assembly. The corresponding mass difference is the binding energy divided by \\(c^2\\).\n\nIn nuclear reactions, these differences are large enough to be central to the calculation. Nuclear fission of heavy nuclei and nuclear fusion of suitable light nuclei can produce more tightly bound final arrangements. The reduction in total rest energy appears as kinetic energy of products, radiation, and sometimes energy carried by weakly interacting particles. A typical uranium-235 fission releases energy of order \\(200\\ \\mathrm{MeV}\\), with the exact accounting depending on products and subsequent decays. Energy per reaction is commonly expressed in the electronvolt and its multiples.\n\nConservation also helped reveal unseen particles. The continuous electron spectrum in beta decay was difficult to reconcile with a simple two-body decay carrying a fixed energy release. In 1930, Wolfgang Pauli proposed an additional neutral particle, later named the neutrino, to account for missing energy, momentum, and angular momentum. Enrico Fermi developed a theory of beta decay in 1933–1934, and the neutrino was experimentally detected in 1956.\n\nIn particle physics, energy conservation is inseparable from momentum conservation. A single photon cannot create an electron–positron pair in otherwise empty space because the complete four-momentum balance cannot be satisfied. A nucleus, another photon, or another participant can make pair production possible by taking recoil or supplying momentum. Such examples show why an adequate energy budget alone does not establish that a reaction can occur."
    },
    {
      "heading": "Local conservation in curved spacetime",
      "text": "General relativity describes gravitation through the geometry of spacetime. Matter and nongravitational fields are represented by the stress–energy tensor \\(T^{\\mu\\nu}\\), whose components include energy density, momentum density, and stresses. Their local balance is expressed by\n\n\\[\\nabla_\\mu T^{\\mu\\nu}=0,\\]\n\nwhere \\(\\nabla_\\mu\\) is the covariant derivative compatible with the spacetime metric. This equation follows consistently with the Einstein field equations and the geometric identities they satisfy.\n\nThe equation is not simply the ordinary divergence-free condition of a flat-space current. The covariant derivative contains connection terms that account for spacetime geometry. Consequently, integrating a coordinate energy density over a region does not automatically yield a globally conserved quantity. In curved spacetime, vectors at different points cannot generally be compared without specifying how they are transported, and there may be no preferred global time coordinate.\n\nA spacetime symmetry can nevertheless produce an energy current. If \\(\\xi^\\mu\\) is a Killing vector representing an appropriate time-translation symmetry, then\n\n\\[J^\\mu=T^{\\mu\\nu}\\xi_\\nu\\]\n\nhas vanishing covariant divergence. With suitable hypersurfaces and boundary conditions, integrating this current gives a conserved charge. Stationary spacetimes therefore admit useful conserved energies even though arbitrary evolving spacetimes may not. For a freely moving test particle, the corresponding symmetry gives a constant along its trajectory.\n\nGravitational energy creates an additional complication. The equivalence principle prevents the construction of a generally applicable local gravitational energy density that behaves as an ordinary tensor and captures all gravitational effects. Coordinate-dependent pseudotensors, boundary-based definitions, and quasi-local constructions serve different purposes. For isolated asymptotically flat systems, ADM energy describes total energy at spatial infinity, while Bondi energy describes energy at null infinity and decreases as radiation escapes.\n\nGravitational waves have measurable energetic effects despite this limitation on local representation. In the appropriate approximation they carry a well-defined averaged energy flux, and asymptotic methods describe their contribution to energy balance. The observed orbital decay of the Hulse–Taylor binary pulsar agrees with the loss predicted from gravitational radiation. The absence of a universal local gravitational energy tensor therefore does not imply that gravitational radiation is physically energyless; it identifies a restriction on how that energy can be represented."
    },
    {
      "heading": "Expansion of the universe and cosmological energy",
      "text": "The expanding universe makes the distinction between local balance and global conservation especially important. In a homogeneous and isotropic cosmological model, local stress–energy conservation leads to\n\n\\[\\dot\\rho+3H\\left(\\rho+\\frac{p}{c^2}\\right)=0,\\]\n\nwhere \\(\\rho\\) is mass-equivalent energy density, \\(p\\) is pressure, and \\(H=\\dot a/a\\) is the expansion rate defined by the scale factor \\(a\\). Equivalently, for a comoving volume \\(V\\propto a^3\\),\n\n\\[d(\\rho c^2V)=-p\\,dV.\\]\n\nThis resembles the first law for adiabatic expansion, but interpreting it does not require treating the universe as a gas pushing against an external container.\n\nFor nonrelativistic matter with negligible pressure, \\(\\rho\\propto a^{-3}\\). Its energy density decreases mainly because the same particles occupy a larger volume. Radiation behaves differently: its pressure is \\(p=\\rho c^2/3\\), so its energy density decreases as \\(a^{-4}\\). Three powers come from dilution and one from the redshift of individual photons, whose energies scale inversely with \\(a\\).\n\nThis raises the question of where redshifted photon energy goes. In a general expanding spacetime there is no global time-translation symmetry guaranteeing a conserved total energy that includes all such photons. The local conservation equation remains satisfied, but no universal additional reservoir must exist to receive a globally defined missing sum. Descriptions involving gravitational work can be useful in particular formulations, yet they should not be mistaken for a unique, coordinate-independent transfer account for the entire universe.\n\nA cosmological constant provides the opposite-looking example. Its energy density remains constant as space expands, while its pressure is negative,\n\n\\[p=-\\rho c^2.\\]\n\nThe energy associated with an expanding comoving volume therefore increases. Substituting this pressure into the local balance equation makes the increase consistent with the same conservation framework. Again, an unrestricted global mechanical analogy would be misleading.\n\nThese distinctions are central to the Lambda–Cold Dark Matter Model and to discussions of dark energy. Claims that the total energy of the universe is exactly zero, or that all positive matter energy is uniquely canceled by negative gravitational energy, depend on additional assumptions and definitions. They are not direct consequences of the ordinary conservation law. Cosmology requires attention to spacetime geometry, global topology, boundaries, and the precise energy quantity under discussion."
    },
    {
      "heading": "Engineering balances and energy efficiency",
      "text": "Engineering applies conservation through explicit balances for machines, buildings, industrial processes, and networks. The usefulness of the principle lies in its independence from many internal details. If a steady device receives energy at one rate and stores none, its total output rate must match its input rate. Measurements of fuel flow, electrical power, fluid enthalpy, temperatures, and mechanical output can therefore expose missing transfer paths or inconsistent operating data.\n\nFor a heat engine, thermal efficiency is the useful net work output divided by heat input. For an electric motor, efficiency is shaft power divided by electrical input. A motor receiving \\(10\\ \\mathrm{kW}\\) and delivering \\(9\\ \\mathrm{kW}\\) has an efficiency of \\(90\\%\\); the remaining \\(1\\ \\mathrm{kW}\\) leaves mainly through heating and other non-shaft outputs. The energy balance does not identify exactly where the dissipation occurs, but it fixes its total once storage and all other outputs are accounted for.\n\nHeat pumps use a different performance measure. Their heating coefficient of performance is\n\n\\[\\mathrm{COP}_{\\mathrm{heating}}=\\frac{Q_h}{W},\\]\n\nwhere \\(Q_h\\) is heat delivered to the warm space and \\(W\\) is work supplied. This ratio can exceed one because the device transfers environmental heat as well as converting input work into heat:\n\n\\[Q_h=Q_c+W.\\]\n\nA coefficient of performance of four means that four units of heat are delivered per unit of work, with three units drawn from the colder source. It does not mean that four units of energy are created from one.\n\nStorage technologies likewise change energy’s form rather than preserve an abstract substance without losses. Batteries alter chemical states; pumped-storage facilities raise water; flywheels store rotational kinetic energy; capacitors store electric-field energy. Round-trip efficiency compares recovered useful energy with input energy over a specified cycle. Self-discharge, resistance, friction, auxiliary equipment, and thermal management can all affect that ratio.\n\nIn an electrical grid, supply and demand must be balanced continuously, allowing for transmission losses and storage changes. Short-term mismatches affect rotating kinetic energy, electromagnetic states, and system frequency. In buildings, thermal storage in walls and furnishings can make instantaneous heating input differ substantially from outward heat flow.\n\nThe everyday phrase “energy conservation” also means reducing energy consumption through efficiency or changed activity. That usage is distinct from the physical law. An efficient process uses less high-quality input for a specified service; it does not make the fundamental conservation principle more valid. Engineering evaluation therefore separates energy quantity, useful output, conversion losses, and the thermodynamic quality of available resources."
    },
    {
      "heading": "Planetary and astrophysical energy budgets",
      "text": "On planetary scales, energy balances connect radiation, circulation, storage, and geological activity. Earth’s energy budget is dominated by incoming solar radiation and outgoing thermal infrared radiation. Averaged over the entire surface, incident sunlight at the top of the atmosphere is approximately one quarter of the solar irradiance measured on a surface facing the Sun near Earth’s orbit. With solar irradiance near \\(1{,}361\\ \\mathrm{W\\,m^{-2}}\\), the geometric average is about \\(340\\ \\mathrm{W\\,m^{-2}}\\).\n\nA fraction is reflected by clouds, atmospheric particles, and the surface. Using a planetary albedo near \\(0.30\\) gives absorbed solar energy of roughly \\(240\\ \\mathrm{W\\,m^{-2}}\\). These rounded figures vary with time and observing method. A planet in long-term radiative equilibrium emits the same average power that it absorbs, but an imbalance causes energy storage or depletion. The rate of storage is constrained by the difference, not by surface temperature alone.\n\nThe atmosphere complicates the route by which energy escapes. Greenhouse gases absorb and emit infrared radiation; convection and evaporation transport energy upward; clouds affect both sunlight and infrared exchange. Surface radiation can therefore exceed the planet’s outgoing radiation to space without violating conservation, because the surface also receives downward infrared energy from the atmosphere. Internal exchanges must not be confused with the net flux across the planetary boundary.\n\nWithin Earth, the ocean stores large amounts of thermal energy because of its mass and heat capacity. Melting ice stores energy as latent heat, and atmospheric and oceanic motions redistribute energy geographically. Geothermal heat, produced partly by radioactivity and partly by cooling from earlier stages of planetary history, is important for geology but much smaller than absorbed sunlight in the global surface budget.\n\nFor a star, the balance involves nuclear energy release, gravitational contraction, internal storage, radiation, and neutrino emission. The virial theorem explains why a self-gravitating object can heat as it loses total energy: contraction makes gravitational potential energy more negative while increasing internal kinetic energy. This behavior differs from that of many familiar laboratory systems and is associated with negative heat capacity in appropriate self-gravitating regimes.\n\nThe Sun currently obtains most of its sustained luminosity from hydrogen fusion in its core. A complete balance includes both electromagnetic output and neutrino energy. For rapidly evolving stars, collapsing objects, and accreting systems, changes in stored thermal and gravitational energy can be as important as nuclear reactions."
    },
    {
      "heading": "Experimental tests, missing energy, and computational practice",
      "text": "Energy conservation is tested through quantitative closure: independently measured inputs, outputs, and storage changes are compared within their uncertainties. Calorimetry measures energy transfers through temperature changes, phase changes, or calibrated electrical compensation. Mechanical experiments compare work with changes in motion or internal energy. Precision tests require corrections for heat leakage, radiation, evaporation, friction, electrical leads, and the energy stored in apparatus.\n\nAn apparent discrepancy is not automatically evidence against conservation. It may reveal an omitted subsystem, an incorrect model, an unmeasured particle, or an instrumental error. The history of the neutrino is an important example of missing energy leading to new physics while retaining the conservation framework. Such reasoning is not circular if the proposed explanation makes independent predictions that can be tested.\n\nAt particle colliders, detectors reconstruct the energies and momenta of visible products. Because the initial longitudinal momenta of colliding partons are not fully known, hadron-collider analyses often emphasize missing transverse momentum. A nonzero result can indicate neutrinos or other undetected particles, but detector acceptance, calibration, resolution, and mismeasurement can produce similar signatures. Events are interpreted using complete statistical and detector models rather than energy arithmetic alone.\n\nMeasurement uncertainty determines how closely a balance can be expected to close. A small residual below the combined uncertainty is not a measured violation. Conversely, persistent discrepancies beyond uncertainty require investigation of both physical assumptions and measurement procedures. Metrology supplies traceable standards and calibration methods that make comparisons between mechanical, electrical, and thermal measurements meaningful.\n\nNumerical calculations introduce another distinction. A physical equation may conserve energy exactly while a discretized approximation does not. Some integration schemes accumulate artificial energy drift; others preserve selected invariants or maintain a nearby effective Hamiltonian over long intervals. Symplectic methods are particularly useful for many Hamiltonian problems, although they do not generally conserve the original energy exactly at every step. Fluid and plasma simulations often use conservative discretizations so that flux leaving one computational cell enters another consistently.\n\nThe status of energy conservation therefore depends on a clearly specified level of description. It is exact in many closed theoretical models, approximate for deliberately simplified subsystems, and constrained by symmetry and geometry in relativistic settings. Proposed extensions of established physics must state their energy accounting explicitly: which degrees of freedom are included, whether the dynamics depend on time, and what observable departures distinguish the proposal from ordinary transfer to an unobserved environment."
    }
  ],
  "links": [
    "energy",
    "physics",
    "thermodynamics",
    "general-relativity",
    "james-prescott-joule",
    "first-law-of-thermodynamics",
    "albert-einstein",
    "thermodynamic-system",
    "international-system-of-units",
    "joule",
    "classical-mechanics",
    "kinetic-energy",
    "potential-energy",
    "newtons-laws-of-motion",
    "work-physics",
    "force",
    "friction",
    "moment-of-inertia",
    "noethers-theorem",
    "momentum",
    "angular-momentum",
    "lagrangian",
    "action-physics",
    "hamiltonian-mechanics",
    "internal-energy",
    "state-function",
    "heat",
    "pressure",
    "ideal-gas",
    "enthalpy",
    "second-law-of-thermodynamics",
    "entropy",
    "carnot-cycle",
    "maxwells-equations",
    "poynting-theorem",
    "electric-field",
    "electrical-resistance",
    "joule-heating",
    "electromagnetic-radiation",
    "chemistry",
    "chemical-reaction",
    "chemical-bond",
    "activation-energy",
    "metabolism",
    "cellular-respiration",
    "adenosine-triphosphate",
    "photosynthesis",
    "gibbs-free-energy",
    "statistical-mechanics",
    "phase-space",
    "partition-function",
    "boltzmann-constant",
    "heat-capacity",
    "brownian-motion",
    "quantum-mechanics",
    "hamiltonian-operator",
    "schrodinger-equation",
    "quantum-superposition",
    "photon",
    "uncertainty-principle",
    "virtual-particle",
    "special-relativity",
    "mass",
    "speed-of-light",
    "mass-energy-equivalence",
    "nuclear-fission",
    "nuclear-fusion",
    "electronvolt",
    "beta-decay",
    "neutrino",
    "particle-physics",
    "spacetime",
    "einstein-field-equations",
    "gravitational-wave",
    "redshift",
    "cosmological-constant",
    "lambda-cdm-model",
    "dark-energy",
    "electrical-grid",
    "earth-energy-budget",
    "albedo",
    "ocean",
    "radioactivity",
    "star",
    "sun",
    "calorimetry",
    "measurement-uncertainty",
    "metrology"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}