# Lithium-ion battery A **lithium-ion battery** is a rechargeable [[Electric_battery|battery]] in which [[Lithium|lithium]] ions move back and forth between two host materials through a non-aqueous [[Electrolyte|electrolyte]], each host taking the ions into its [[Crystal_structure|crystal structure]] without being destroyed by them. Nothing is plated and nothing dissolves: the cell works by intercalation, which is why it survives hundreds of cycles where a metal-plating chemistry would not. In the microsim below the reader turns one control, the C-rate, and reads three consequences off the intercalation cell — the open-circuit voltage against state of charge, the voltage sag that grows with current through `V = E_ocv(SoC) - b asinh(I/(2 I*)) - I R`, and the usable capacity that shrinks as the discharge is hurried.[^hav-bv][^hav-stack] Beside it sits the Ragone plane, with the pack's specific energy marked against the roughly 250 Wh/kg that current cells reach at pack level.[^visw2022] On the Energy flagship's spine this page is the main article for Part IV — Scientific use, section *The lithium-ion cell*, and it is the sibling of [[Electric_battery|the battery root]] that places the element [[Lithium|Li]] in the energy system. It hands on to [[Grid_energy_storage|grid energy storage]] for the stationary case and to [[Electric_vehicle|electric vehicles]] for the mobile one; the Chemistry flagship owns the element page itself and gets the Elements-program variant of this sim. Why lithium? Because the thermodynamic voltage of a cell is fixed by the reaction's [[Gibbs_free_energy|Gibbs free energy]] through `dG = -n F V_eq`, and lithium is both the most electropositive metal and among the lightest, so it buys a large voltage for very little mass.[^hav-thermo] The consequences run through this whole article: high voltage means water cannot be the solvent, a non-aqueous solvent means a flammable electrolyte, and a flammable electrolyte means that every gain in [[Energy_density|energy density]] is also a gain in the energy available to a failure. ## History The intercalation idea — storing charge by inserting guest ions into a layered host rather than by dissolving and re-depositing a metal — is what separates this chemistry from everything that came before it, and it was the route out of the dendrite problem that had blocked rechargeable lithium metal cells. Pairing a layered oxide positive electrode with a carbon negative electrode gave a cell in which lithium never exists as a metal at all during normal operation, and commercialisation followed in the early 1990s. The energy-system literature treats the result as the point at which electrochemical storage stopped being a niche and became a planning assumption for transport and the [[Electrical_grid|grid]].[^kerlin-ch9] ## Components Six parts matter. The positive electrode is a lithium-bearing transition-metal compound, built on some combination of [[Cobalt|cobalt]], [[Nickel|nickel]], [[Manganese|manganese]], or [[Iron|iron]] and [[Phosphorus|phosphorus]] in the phosphate chemistries. The negative electrode is usually [[Graphite|graphite]], sometimes with [[Silicon|silicon]] added for capacity. Between them is a porous separator wetted with the electrolyte: a lithium salt, commonly a [[Fluorine|fluorine]]-bearing one, dissolved in organic carbonates. Current is collected on [[Aluminium|aluminium]] foil at the positive side and [[Copper|copper]] at the negative. Finally there is the package, which is structural, thermal and safety hardware rather than active material. The electrodes are not solid slabs but porous composites — active particles, conductive additive and binder — and that geometry is the subject of the Portal Book's porous-electrode chapter. Porosity ε and tortuosity τ set the effective transport coefficients through `D = (eps/tau^2) D_m`, with the Bruggeman approximation `D ≈ eps^1.5 D_m` for the usual exponent, and specific surface area for spherical particles is `a = 6(1 - eps)/d`.[^hav-porous-geom] Halving the particle diameter doubles the reacting area, which is the main design lever for rate capability and a chief reason that electrode engineering, not new chemistry, delivers most year-on-year improvement. ## Electrochemistry On discharge, lithium leaves the negative host, crosses the electrolyte as [[Ion|ions]] and enters the positive host, while the matching electrons go the long way through the load. Charging drives the same reaction backwards — the cell is being run as an [[Electrolysis|electrolysis]] cell, with the loss terms added to the equilibrium voltage instead of subtracted from it.[^hav-sign] Both hosts change [[Oxidation_state|oxidation state]] as their lithium content changes; neither dissolves. Two features of the intercalation mechanism show up directly in the discharge curve. Where the host takes lithium as a continuous [[Solid_solution|solid solution]], the activity of lithium in the host varies with composition and the voltage slopes, following [[Nernst_equation|Nernst-type]] behaviour, `E_eq = E0' + (R T/(n F)) ln(c_O/c_R)`.[^hav-nernst] Where the host instead splits into two coexisting phases over a composition range, the activities are pinned by the phase rule and the voltage runs flat — the plateau that makes phosphate chemistries so easy to use and so hard to gauge, because a flat curve carries almost no information about how much charge is left. The rate at which lithium crosses each particle surface follows [[Butler–Volmer_equation|Butler–Volmer]] kinetics; for a symmetric transfer coefficient the [[Overpotential|overpotential]] needed for a given current is `eta = b asinh(j/(2 jstar))` with `b = 2 R T/F`, about 51 mV at 298 K, so each decade of current costs roughly 118 mV.[^hav-bv] Beneath about 26 mV the interface is effectively a resistor; well above it, it is exponential. The `b asinh(...)` term in the sim's equation is exactly this, and the asinh rather than the logarithm is the Portal Book's own device for keeping the overpotential from going negative below the exchange current.[^hav-stack] ## Charging and discharging The asymmetry between the two directions is the operational heart of the cell. Discharging is limited by what the load asks for; charging is limited by what the cell will accept without plating lithium metal on the negative electrode, which is both a capacity loss and a safety hazard. ### Constant-current constant-voltage charging The standard profile has two phases. In the first, current is held constant and the terminal voltage climbs, partly because the state of charge is rising and partly because the loss terms `b asinh(I/(2 I*)) + I R` sit on top of the equilibrium voltage during charge.[^hav-stack] When the terminal voltage reaches the chemistry's upper limit, the charger switches to holding that voltage constant, and the current decays as the cell fills and the required overpotential shrinks. The charge is called complete when the current falls below a small fraction of its constant-current value. The reason the second phase exists is worth stating precisely: because part of the terminal voltage is loss rather than stored energy, a cell that has *reached* the voltage limit has not yet *stored* the charge that corresponds to it. Holding voltage while current decays lets the loss terms fall away so that the equilibrium voltage can rise into the gap. In the microsim, pushing the C-rate up makes the constant-current phase shorter and the constant-voltage tail longer, without much changing the total time — which is why fast charging is mostly a thermal and transport problem, not a thermodynamic one. ### Self discharge Even at rest a cell loses charge, through slow side reactions at electrode surfaces that consume lithium into films rather than returning it to the other electrode. The rate rises with temperature roughly as an [[Arrhenius_equation|Arrhenius]] process, so storage temperature matters more than storage time over ordinary intervals. Compared with the [[Electric_battery|older aqueous chemistries]] the loss per month is small, which is why lithium-ion is usable in equipment that sits unused for months. ## Battery designs and formats Cells reach the user in three geometries — cylindrical, prismatic and pouch — and the choice trades mechanical robustness against packing efficiency and heat removal. A cylindrical can is self-supporting and tolerates internal pressure; a pouch has almost no dead volume but needs external restraint and a path for swelling. ### Cells Inside every format the structure is the same sandwich repeated: coated positive foil, separator, coated negative foil, either wound into a spiral or stacked in layers. Because the electrodes are thin and the sandwich is repeated, the useful design variable is coating thickness, and it has an optimum rather than a maximum. Thicker coatings raise the fraction of the cell that is active material but lengthen the transport path, and past a Thiele modulus of about 3 only the leading 1/M of the electrode reacts at all; the Portal Book puts the optimal effectiveness factor at about 1/3 rather than close to 1.[^hav-porous] ## Uses The cell's applications sort by which of its two figures of merit dominates. Where energy per unit mass rules — phones, laptops, drones, [[Electric_vehicle|electric vehicles]] — lithium-ion displaced everything else. Where energy per unit cost rules, as in [[Grid_energy_storage|stationary storage]], it competes with [[Pumped-storage_hydroelectricity|pumped hydro]] and increasingly wins on siting rather than on price. Where power per unit mass rules and total energy is small, [[Supercapacitor|supercapacitors]] remain better. And where the requirement is seasonal rather than daily storage, no battery is competitive, because the cost scales with stored energy rather than with power, which is the structural argument for chemical carriers such as [[Hydrogen_storage|stored hydrogen]]. ## Performance This is the section the article's microsim belongs to. The label quantities are capacity in ampere-hours and nominal voltage, and their product is the energy, `Ah × V = Wh`.[^murphy-elec] For a cell whose voltage runs between roughly 3.0 V empty and 4.2 V full, the nominal figure quoted is about 3.7 V, and it is a weighted average over the discharge, not a measured constant. The sim's open-circuit curve is an **ILLUSTRATIVE** display fit built around that nominal value, not measured data for any particular chemistry — its purpose is to show the *shape* of the state-of-charge relationship and how sag stacks on top of it. The table below is the fit the sim draws: | State of charge | 100 % | 80 % | 50 % | 20 % | 5 % | 0 % | |---|---|---|---|---|---|---| | Open-circuit voltage (ILLUSTRATIVE fit) | 4.20 V | 4.00 V | 3.70 V | 3.50 V | 3.30 V | 3.00 V | The reader's control is the C-rate: 1C is the current that would empty the rated capacity in one hour, so a 50 Ah cell at 2C is drawing 100 A. Raising the C-rate does three things at once on screen. The terminal voltage drops by the loss stack, so the whole discharge curve slides down. The energy delivered falls faster than the charge delivered, because energy is the area under a voltage that is now lower everywhere. And the usable capacity itself contracts, because the reaction front cannot penetrate the full electrode thickness at high current — the effectiveness-factor argument again, running in the discharge direction.[^hav-porous] Heat rises with the square of current through the ohmic term and roughly linearly through the kinetic term, so a doubling of C-rate more than doubles the cooling problem.[^hav-stack] The second panel is the Ragone plane, specific energy against specific power, on which every store is a region rather than a point. The mark to steer by is about 250 Wh/kg at pack level for current lithium-ion technology, the figure the battery-powered-flight literature uses as its present-day baseline.[^visw2022] An electric-vehicle readout converts that to range by dividing pack energy by the vehicle's consumption per kilometre, which the sim exposes as a user input rather than asserting, since consumption depends on mass, speed and terrain far more than on the cell. ### Round-trip efficiency Round-trip efficiency is the energy out divided by the energy in over a full charge and discharge, and it is bounded by the ratio of the two terminal voltages: the cell is charged above its equilibrium voltage and discharged below it, so the loss stack is paid twice.[^hav-sign] Because those losses grow with current, efficiency is a function of rate and not a property of the chemistry — a cell that is 96 % efficient at C/10 can be well below 90 % at 3C, with the difference appearing as heat. Lithium-ion's advantage over thermal storage routes is exactly this: it converts chemical energy to work directly, with no [[Heat_engine|heat engine]] and therefore no Carnot ceiling. ## Lifespan Two clocks run together. Calendar ageing proceeds at rest through the same side reactions that cause self-discharge, consuming lithium into surface films. Cycle ageing adds mechanical damage: the host lattices expand and contract as lithium enters and leaves, which cracks particles, breaks conductive contact and exposes fresh surface for more film growth. Both raise internal resistance, which raises heat at a given current, which accelerates both — a feedback rather than a countdown.[^hav-batt] ### Recommendations The practical advice follows from the mechanisms rather than from folklore. Keep the cell cool, because every degradation pathway is thermally activated. Avoid long storage at full charge, where the positive electrode is most oxidising and the negative most reducing. Avoid deep discharge, which drives the negative electrode to potentials where its collector can corrode. And prefer moderate C-rates, since both the heat and the mechanical strain scale with current. ## Safety The hazard specific to this chemistry is that the electrolyte is a flammable organic liquid held next to electrodes that store a large amount of energy and oxidising capacity in a small volume. Any fault that converts stored energy to heat faster than the package can remove it — internal short, external short, overcharge, crush, or a manufacturing inclusion — raises the temperature, and the reactions that follow generate their own heat. ### Fire hazard The dangerous feature is self-acceleration. Above a threshold temperature, decomposition reactions release heat faster than it can be conducted away, so the temperature rises, which speeds the reactions further; the released gases vent and can ignite. Mitigation therefore attacks the feedback rather than the ignition: separators that shut down ionically when hot, vents that relieve pressure before rupture, pack designs that keep one failing cell from heating its neighbours, and management electronics that hold every cell inside its voltage and temperature window. ## Supply chain The supply chain is a materials question with a geographic answer. Lithium comes from brines and hard-rock ores; cobalt and nickel from a small number of producing regions; graphite largely from a small number of processing ones. Because a cell's metals are neither consumed nor chemically destroyed by cycling, the long-run supply picture depends on how much of the installed stock returns for recovery.[^hav-batt] ### Solid waste and recycling A spent cell is a concentrated ore: its lithium, cobalt and nickel contents are far above anything mined. That arithmetic, rather than waste-volume concern, is the strongest case for collection. The engineering constraint is that recovery must separate materials that were deliberately engineered into intimate contact — a coated foil with binder and conductive carbon is not easy to take apart — so recycling economics turn on cell design as much as on process chemistry.[^kerlin-ch9] ## Research Current work runs along four lines that the modelling literature makes legible. Higher-capacity negative electrodes, principally silicon-bearing ones, trade capacity against the volume change that drives mechanical ageing. Solid electrolytes aim to remove the flammable liquid altogether and, if lithium metal can be used safely, to raise specific energy substantially. Cobalt-free and sodium-based chemistries trade energy density for supply security. And electrode architecture — porosity, tortuosity and particle size — remains the cheapest available lever, because the effectiveness-factor relations say directly how much of an electrode is doing work at a given rate.[^hav-porous][^hav-transport] ## See also - [[Electric_battery]] - [[Electric_vehicle]] - [[Grid_energy_storage]] - [[Lithium]] - [[Energy_storage]] - [[Supercapacitor]] - [[Fuel_cell]] ## References [^hav-thermo]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 1 Electrochemistry, pp. 25–26 (`ΔG = −nFV_eq`, F ≈ 96,485 C/mol e⁻; ΔH = ΔG + TΔS). https://open.umn.edu/opentextbooks/textbooks/electrolysers-fuel-cells-and-batteries-analytical-modelling [^hav-nernst]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 1 Electrochemistry, p. 32 (the single-electron Nernst form `E_eq = E0' + (RT/nF) ln(c_O/c_R)`). [^hav-bv]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 1 Electrochemistry, pp. 30, 33–35 (Butler–Volmer; the α = ½ form `j = 2j* sinh(η/b)` with b = 2RT/F and its inverse `η = b asinh(j/(2j*))`; linear regime for η ≲ RT/F ≈ 25 mV; one decade of current costs b_a ln 10 ≈ 120 mV. The exact 118 mV at 298.15 K, with b = 51.4 mV and RT/F = 25.7 mV, is derived from the book's forms in the Wikitube extract). [^hav-sign]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 1 Electrochemistry, pp. 25, 27 (sign convention — an electrolytic cell has V_cell < V_eq < 0, so losses are added on charge and subtracted on discharge; voltage efficiency φ_e = V_eq/V_cell; the ΔH/ΔG bookkeeping ratios). [^hav-stack]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 1 Electrochemistry, pp. 36–38 (the stacked loss equation and its Eq. 1.36 form; footnote 19's replacement of ln(x) by asinh(x/2); activation losses dominate at low current density and the ohmic term at high current density). [^hav-porous]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 5 Porous electrodes, pp. 96, 99–100 (the effectiveness factor `E = tanh(M)/M`, falling to 1/M for M ≫ 1; the combined-limitation form with at most 12 % error; optimal thickness at E_opt ≈ 1/3 rather than E ≈ 1). [^hav-porous-geom]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 5 Porous electrodes, pp. 78–84 (porosity ε = V_pore/V, tortuosity τ ≥ 1, specific area a = 6(1 − ε)/d for spheres, effective transport `D = (ε/τ²)D_m` and the Bruggeman form `D ≈ ε^1.5 D_m`; the fitted exponent is 1.58 against a nominal 1.6). [^hav-transport]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 3 Transport, pp. 46–65 (page to pin) (diffusion in the electrolyte and the limiting-current condition that caps rate). [^hav-batt]: Haverkort, Willem (2024). *Electrolysers, Fuel Cells and Batteries: Analytical Modelling*. Portal Book 053, Chapter 7 Batteries, pp. 110–129 (page to pin) (the battery chapter, including energy density and the electrode/electrolyte limits on rate and life). [^murphy-elec]: Murphy, Thomas (2021). *Energy and Human Ambitions on a Finite Planet*. Portal Book 097, Chapter 5 Energy and Fossil Fuels, pp. 97–98 (`P = IV`, 1 A = 1 C/s, and Ah × V = Wh; the 9 V, 0.5 Ah = 4.5 Wh = 16.2 kJ reference case). https://open.umn.edu/opentextbooks/textbooks/energy-and-human-ambitions-on-a-finite-planet [^murphy-app]: Murphy, Thomas (2021). *Energy and Human Ambitions on a Finite Planet*. Portal Book 097, Chapter 8 Appendices, pp. 378–431 (page to pin) (the appendix treatment of battery storage in the whole-system energy accounting). [^kerlin-ch9]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*. Portal Book 048, Chapter 9, pp. 354–466 (page to pin) (electrical storage in the energy-system context, including its role for transport and the grid). https://open.umn.edu/opentextbooks/textbooks/future-energy-opportunities-challenges [^visw2022]: Viswanathan, Venkatasubramanian; Epstein, Alan H.; Chiang, Yet-Ming; Takeuchi, Esther; Bradley, Marty; Langford, John; Winter, Michael (2022). "The challenges and opportunities of battery-powered flight." *Nature*. https://doi.org/10.1038/s41586-021-04139-1 — cited here, as in the Wikitube Aviation plan, for the ≈250 Wh/kg pack-level specific-energy mark for current lithium-ion technology. ## External links - *Electrolysers, Fuel Cells and Batteries: Analytical Modelling* (2024), Portal Book 053 — Open Textbook Library record: https://open.umn.edu/opentextbooks/textbooks/electrolysers-fuel-cells-and-batteries-analytical-modelling - *Energy and Human Ambitions on a Finite Planet* (2021), Portal Book 097 — Open Textbook Library record: https://open.umn.edu/opentextbooks/textbooks/energy-and-human-ambitions-on-a-finite-planet - The Wikipedia pair's External links section lists the pair's own links, including the standards and recall notices behind the Safety and Supply chain sections. <!-- MATTERSIM:BEGIN g33 — Matter & Energy Cluster microsim (framework build, specs/sims/Lithium-ion_battery.json); do not hand-edit inside --> **Microsim — three.js (Wikitube framework):** *Lithium-ion battery* <div class="wt-sim" data-src="https://wikitube-3d-microsims.netlify.app/matter/Lithium-ion_battery.html" data-title="Lithium-ion battery"></div> *Built from `MICROSIM_GUIDE/specs/sims/Lithium-ion_battery.json`; part of the [[PORTAL_Matter|Matter portal]] spine (section sims and See-also variants).* <!-- MATTERSIM:END --> ## Wikipedia : Wikitube **Strict pair:** [Wikipedia](https://en.wikipedia.org/wiki/Lithium-ion_battery) : [Wikitube](https://en.wikitube.io/wiki/Lithium-ion_battery) · pinned revision [1374040253](https://en.wikipedia.org/w/index.php?oldid=1374040253) · 2026-09-11 ## Previous hub tags Hubs: `Life_Physics`. Portals: [[PORTAL_Energy]]. --- *Matter & Energy Cluster child articles, wave 1 · 2026-09-11 · drafted · Energy row E20 · sim pending (matter/Lithium-ion_battery).*