{
  "id": "radiocarbon-dating",
  "title": "Radiocarbon Dating",
  "lang": "en",
  "source_lang": "en",
  "summary": "Radiocarbon dating estimates the age of carbon-bearing materials from radioactive carbon-14, with calibration translating measurements into calendar dates.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [
    "Carbon-14 dating",
    "carbon dating",
    "radiocarbon age determination"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/radiocarbon-dating/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/radiocarbon-dating/",
      "json": "https://aiwiki.page/zh/radiocarbon-dating.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Radiocarbon dating is a method of radiometric dating that determines when carbon in a material ceased exchanging with a particular environmental reservoir. It relies on the decay of carbon-14, a radioactive isotope of carbon continually produced in the atmosphere and distributed through the carbon cycle. The method is especially important for dating once-living materials from approximately the last 50,000 years, although its practical range depends on sample preservation, measurement technique, and laboratory background.\n\nRadiocarbon dating transformed archaeology by providing a numerical chronology largely independent of artifact styles and written records. It also became a major tool in environmental science, Earth-system research, and the study of recent geological history. A radiocarbon measurement does not, however, directly yield an exact calendar year. Reliable dating requires identifying what event the sample represents, removing contaminants, accounting for its carbon source, and comparing the measurement with an appropriate calibration curve."
    },
    {
      "heading": "Physical basis and the radiocarbon clock",
      "text": "Carbon has two stable isotopes, carbon-12 and carbon-13, and the naturally occurring radioactive isotope carbon-14. Carbon-14 contains six protons and eight neutrons. It forms primarily when secondary neutrons, generated by interactions between cosmic rays and the upper atmosphere, strike nitrogen-14 nuclei. The principal production reaction is\n\n\\[{}^{14}_{7}\\mathrm{N}+{}^{1}_{0}\\mathrm{n} \\rightarrow {}^{14}_{6}\\mathrm{C}+{}^{1}_{1}\\mathrm{p}.\\]\n\nThe newly formed carbon is oxidized and eventually enters atmospheric carbon dioxide. Plants incorporate it through photosynthesis, and animals acquire it by eating plants or other animals. Exchange among the atmosphere, biosphere, soils, and ocean distributes radiocarbon unevenly across reservoirs with different mixing and residence times.\n\nCarbon-14 undergoes beta decay to nitrogen-14, emitting an electron and an electron antineutrino. Its physical half-life is approximately 5,730 years. If a sample becomes effectively closed to carbon exchange, the number of carbon-14 atoms remaining follows the exponential decay law:\n\n\\[N(t)=N_0e^{-\\lambda t}, \\qquad \\lambda=\\frac{\\ln 2}{T_{1/2}}.\\]\n\nHere \\(N_0\\) is the number present at closure, \\(N(t)\\) is the number remaining after time \\(t\\), and \\(T_{1/2}\\) is the half-life. After one half-life, half remains; after ten half-lives, approximately one-thousandth remains. The progressively smaller signal makes old samples increasingly sensitive to background and contamination.\n\nThe clock does not necessarily start at the death of an entire organism. In a tree, each annual growth ring incorporates carbon during its formation, and much of that structural carbon subsequently remains fixed. A piece of heartwood can therefore predate the tree’s felling by centuries. Animal tissues turn over at different rates, so their carbon may represent food consumed over different intervals. A shell records carbon incorporated during shell formation, but its initial radiocarbon content also depends on the surrounding water.\n\nThe principal complication is that \\(N_0\\) cannot ordinarily be measured directly in an ancient sample. Initial radiocarbon abundance has varied with solar activity, Earth’s magnetic field, carbon exchange, and human activities. Moreover, organisms do not all draw carbon from the same reservoir. The physical decay law is well established, but translating a measured isotope ratio into elapsed calendar time requires independent information about the initial conditions. This distinction separates a radiocarbon measurement from a fully interpreted date."
    },
    {
      "heading": "Discovery and development",
      "text": "The experimental foundation of radiocarbon dating emerged from nuclear research in the first half of the twentieth century. Martin Kamen and Samuel Ruben identified carbon-14 in 1940 through accelerator experiments. Its comparatively long half-life subsequently made it a plausible tracer of environmental carbon over archaeological timescales. Willard F. Libby developed the dating concept during the 1940s, arguing that atmospheric production and biological uptake should establish measurable radiocarbon in living matter and that its decline after carbon exchange ceased could provide a clock.\n\nLibby’s group at the University of Chicago, including Ernest C. Anderson and James R. Arnold, developed sensitive counting methods and tested the assumptions experimentally. A central challenge was detecting very low levels of radioactivity against natural radiation backgrounds. Early work established that contemporary organic materials contained measurable radiocarbon and compared measurements from historically dated objects with their expected ages. Arnold and Libby’s 1949 publication on samples of known age helped establish the method’s credibility. Libby received the 1960 Nobel Prize in Chemistry for his development of radiocarbon age determination.\n\nAdoption was rapid because the method addressed a fundamental chronological problem. Archaeologists could often order artifacts by style or place deposits in relative sequence, but numerical ages remained uncertain outside regions with written chronologies. Radiocarbon dating made it possible to compare developments across continents without assuming that similar technologies or artistic forms were contemporary. Its early results also revised estimates of the timing of postglacial environmental changes and the duration of archaeological sequences. These changes were not simply substitutions of numbers for labels: they altered interpretations of cultural contact, migration, and the independent emergence of innovations.\n\nThe recognition that atmospheric radiocarbon had varied was equally consequential. Comparisons with independently dated tree rings showed that radiocarbon years were not identical to calendar years. Work by Hans Suess and subsequent researchers helped establish systematic tree-ring calibration. In 1977, several research groups demonstrated practical radiocarbon measurement using accelerator mass spectrometry, or AMS. Rather than waiting for radioactive atoms to decay, AMS counted rare carbon-14 atoms relative to stable carbon isotopes. This reduced sample requirements dramatically and extended dating to small seeds, individual fragments, and selected chemical fractions.\n\nLater development emphasized internationally coordinated standards, interlaboratory comparisons, improved chemical preparation, and increasingly comprehensive calibration datasets. The history of the method is therefore not a progression toward a single universally accurate machine reading. It is a combination of advances in nuclear measurement, environmental understanding, sample chemistry, and chronological modeling. Each addressed a different source of uncertainty, and all remain necessary in present-day practice."
    },
    {
      "heading": "Datable materials and sample preparation",
      "text": "A material is suitable for radiocarbon dating when it contains carbon whose origin and subsequent history can be understood. Common samples include wood, charcoal, seeds, nutshells, textiles, paper, peat, bone collagen, and shells. Some carbonates, dissolved inorganic carbon, and specific organic compounds can also be measured, but they require different interpretive assumptions. Ordinary geological rocks are not dated simply because they contain carbon: carbon inherited from ancient limestone, petroleum, or other geological sources may be effectively radiocarbon-free and unrelated to the event being investigated.\n\nSample selection begins with the relationship between material and event. A charred annual seed in a sealed hearth may closely date a burning episode, whereas timber reused in a building may substantially predate construction. Charcoal can persist in soils and later be redeposited. Shells can be transported from older deposits into younger ones. Dating a sediment’s bulk organic carbon may combine recently deposited plant matter, older eroded soil carbon, and aquatic material. In such cases the measurement can be analytically sound without representing a single meaningful age.\n\nChemical pretreatment seeks to isolate an appropriate carbon fraction while removing introduced material. Wood and charcoal often undergo acid–base–acid treatment: acid removes carbonates, alkali removes some mobile humic substances, and a final acid step eliminates carbon dioxide absorbed during treatment. More resistant charcoal procedures may include additional oxidation. Wood can be processed to isolate cellulose, reducing contributions from resins, conservation chemicals, or mobile compounds. The treatment must be matched to preservation, because aggressive chemistry may destroy fragile material or leave too little carbon for analysis.\n\nFor bone, laboratories commonly isolate collagen or a collagen-derived gelatin fraction. Mineral components and soil contaminants are removed through controlled chemical steps; some protocols also use ultrafiltration to separate molecular-size fractions. Collagen yield, elemental composition, and the atomic carbon-to-nitrogen ratio help assess preservation, although no single screening criterion guarantees an uncontaminated result. Burnt bone requires separate consideration: calcined bone mineral can sometimes be dated, but its carbon may reflect exchange during heating rather than straightforward preservation of original biological carbon.\n\nConservation materials are a particular problem for museum objects. Waxes, adhesives, consolidants, and biocides may contain modern or fossil-derived carbon. Their removal can require solvent extraction and material-specific testing. Compound-specific dating uses chromatography and related separation methods to isolate individual molecules or molecular classes, such as particular amino acids or lipid compounds. This can improve chemical specificity but increases the importance of procedural blanks because the isolated carbon mass may be very small. Sample preparation is consequently part of the measurement itself, not merely a cleaning step preceding it."
    },
    {
      "heading": "Measurement techniques and laboratory standards",
      "text": "Radiocarbon can be measured by observing its radioactive decays or by counting carbon-14 atoms. In decay-counting methods, a purified sample is converted into a suitable counting medium. Gas proportional counters commonly use a carbon-bearing gas, while liquid scintillation counting detects light pulses produced when beta particles interact with a scintillating medium. Both approaches require strong background control, often including shielding and auxiliary detectors. Because only a small proportion of carbon-14 atoms decay during a laboratory measurement, counting may require relatively large samples and extended observation periods.\n\nAMS instead measures isotope ratios. Samples are usually combusted to carbon dioxide, purified, and converted to graphite, although some systems accept carbon dioxide directly. An ion source produces negative carbon ions, which enter an accelerator and undergo charge-changing collisions in a stripper. This process breaks apart molecular ions that could mimic mass-14 carbon. Nitrogen, a potentially important atomic isobar, does not form a stable negative ion in the way required for ordinary carbon AMS measurement, providing an additional means of discrimination.\n\nElectric and magnetic elements separate ions according to their motion and mass-to-charge characteristics, while detectors identify carbon-14. Carbon-12 and carbon-13 measurements establish reference isotope currents or ratios. This is a specialized form of mass spectrometry, distinguished by the accelerator and its suppression of molecular interferences. Many routine AMS analyses use about a milligram of carbon or less; specialized facilities can measure much smaller amounts. The smallest feasible sample is not necessarily the smallest reliably datable sample, because contamination and background become increasingly important as carbon mass decreases.\n\nA laboratory run includes standards of known or assigned radiocarbon content, process blanks, and quality-control materials. A process blank ideally contains no measurable original carbon-14 but undergoes the same preparation as an unknown sample, revealing carbon introduced during chemistry, graphitization, and measurement. Reference materials establish consistency with internationally accepted radiocarbon conventions. Replicate measurements and interlaboratory comparisons test reproducibility. Laboratories must also correct for isotope fractionation, because physical and biological processes discriminate slightly among carbon isotopes and thereby change isotope ratios independently of radioactive decay.\n\nReported precision depends on more than counting statistics. Relevant contributions include instrumental stability, standard normalization, blank correction, preparation reproducibility, and sample size. For favorable recent material, analytical uncertainties of a few tens of radiocarbon years are common, but that figure is not automatically the uncertainty of the corresponding calendar date. Near the practical age limit, the measured signal approaches the laboratory background, and results may be reported as minimum ages rather than finite dates. A claim that a sample is “older than” a specified limit is analytically different from assigning it a well-constrained age beyond that limit."
    },
    {
      "heading": "Conventional ages and calendar calibration",
      "text": "Laboratories traditionally express results as conventional radiocarbon ages in years before present, abbreviated BP. By definition, “present” is the year 1950 CE, not the year of measurement. This reference date became established during the method’s early development and also precedes the major atmospheric radiocarbon increase caused by nuclear weapons testing. Conventional ages retain the historical Libby half-life of 5,568 years, even though the accepted physical half-life is approximately 5,730 years. Retaining this convention keeps measurements comparable across laboratories and publication dates.\n\nIn simplified form, the conventional age is\n\n\\[t_{\\mathrm{RC}}=-8033\\ln F,\\]\n\nwhere \\(F\\) is the sample’s normalized radiocarbon abundance relative to the conventional modern reference and 8,033 years is the mean life corresponding to the Libby half-life. Normalization includes correction for isotope fractionation, conventionally to a carbon-13 composition of \\(\\delta^{13}\\mathrm{C}=-25\\) per mille for samples. A conventional age is therefore a standardized expression of measured radiocarbon content, not an elapsed calendar time obtained by using the best current physical half-life.\n\nRadiocarbon calibration compares that measurement with independently dated records of past radiocarbon abundance. Annual tree rings provide especially strong evidence because their calendar positions can be established by matching overlapping growth patterns. For older periods and some environments, calibration also uses corals, cave deposits, lake sediments, and marine records whose chronologies are independently constrained. The internationally coordinated IntCal20, SHCal20, and Marine20 curves, published in 2020, represent Northern Hemisphere atmospheric, Southern Hemisphere atmospheric, and marine conditions respectively. IntCal20 extends to 55,000 calibrated years BP.\n\nCalibration is not equivalent to subtracting a fixed correction. The curve contains slopes, wiggles, and plateaus produced by changes in radiocarbon production and carbon cycling. If radiocarbon abundance changed little over several calendar centuries, samples from those centuries may have nearly indistinguishable conventional ages. Conversely, a steep curve segment can provide comparatively precise calendar placement. One measured age may intersect the calibration curve in several places, yielding a calendar-age probability distribution with multiple separate ranges.\n\nThe uncertainties of both the measurement and the calibration curve must be propagated. A result such as \\(3000\\pm30\\) BP cannot be converted simply into a calendar year by subtracting 3,000 from 1950. Calibrated results are commonly reported as ranges with stated probability coverage, often approximately 68.3% or 95.4%, and identified as cal BP, cal BCE, or cal CE. Where a distribution has several separated intervals, their individual probabilities matter. A midpoint and symmetrical error bar can conceal the actual shape of the evidence and imply precision that the calibration process does not support."
    },
    {
      "heading": "Reservoir effects, contamination, and practical limits",
      "text": "A radiocarbon reservoir effect occurs when a sample obtains carbon whose radiocarbon content differs systematically from that of the contemporaneous atmosphere. Marine organisms are the best-known example. Surface seawater exchanges carbon with the air but also mixes with deeper water that has been isolated from the atmosphere for long periods. Marine carbon can consequently appear several hundred radiocarbon years older than terrestrial carbon of the same calendar age. The magnitude varies geographically and through time, particularly where upwelling brings old deep water toward the surface.\n\nMarine calibration curves represent broad ocean behavior, while a regional correction, commonly written \\(\\Delta R\\), describes local departures from the marine reference. This correction is not interchangeable with a single universal subtraction from the laboratory age. Its value and uncertainty must relate to the relevant region, period, and calibration framework. Dietary mixing complicates human and animal remains: bone collagen from an individual who consumed both marine and terrestrial foods may require a model combining carbon sources rather than calibration against one curve alone.\n\nFreshwater systems can produce even less predictable offsets. Water draining limestone terrain can acquire dissolved carbon from ancient carbonate rocks, creating a hard-water effect. Aquatic plants, mollusks, and fish may incorporate that depleted carbon. In some settings, freshwater foods also contribute reservoir-aged carbon to human tissues or charred food residues on pottery. Groundwater radiocarbon interpretation is similarly complicated by mineral dissolution, mixing, and exchange during subsurface movement; an apparent age is not automatically the time since recharge. Volcanic or geothermal carbon dioxide can also cause locally depleted radiocarbon in plants growing near emissions.\n\nContamination differs from a reservoir effect because it introduces carbon after, or outside, the process intended to be dated. Modern contamination generally makes an old sample appear younger, while radiocarbon-free carbon makes it appear older. The asymmetry becomes striking near the method’s limit. A truly radiocarbon-dead sample containing just 1% modern carbon would have an apparent conventional age of about 37,000 BP:\n\n\\[-8033\\ln(0.01)\\approx37{,}000.\\]\n\nThe result would not indicate a small error around an ancient age; it would be largely a measurement of the contaminant.\n\nThere is therefore no universal maximum age applicable to every material and laboratory. Around 50,000 calendar years, only a few tenths of one percent of the original carbon-14 remains, and small blank errors become consequential. Exceptionally clean samples and specialized procedures can support measurements beyond the routine range, but numerical precision alone does not demonstrate accuracy. Radiocarbon is also poorly suited to some very recent pre-bomb intervals because calibration-curve structure can produce broad or ambiguous ranges. Its limits arise from a combination of nuclear statistics, environmental history, chemistry, and the chronological question being asked."
    },
    {
      "heading": "Building archaeological and environmental chronologies",
      "text": "Radiocarbon dates acquire historical meaning through their association with events and deposits. Stratigraphy can establish that one layer predates another, but a sample within a layer may have been redeposited or introduced by later disturbance. Burrowing animals, roots, sediment reworking, and excavation errors can separate the age of a carbon-bearing object from the age of its present context. Dating several well-chosen samples from a sequence generally provides a stronger chronology than obtaining an extremely precise measurement from a single uncertain object.\n\nShort-lived terrestrial materials often provide a close connection to archaeological activity. Annual seeds, small twigs, and identifiable food remains reduce the risk that biological age substantially predates use. Long-lived wood creates the “old-wood” problem: an inner growth ring dates when that ring formed, not when the timber was cut, burned, or deposited. Reuse adds another interval. The same principle applies to heirlooms, curated bones, and excavated materials incorporated into later structures. A date can provide a maximum age for an event without directly dating that event.\n\nBayesian inference allows radiocarbon measurements to be combined formally with independent chronological information. A model may specify that samples occur in stratigraphic order, belong to a phase, or precede a boundary. The radiocarbon evidence supplies likelihoods, while archaeological knowledge supplies prior constraints. The resulting posterior distributions can estimate event dates, phase durations, or gaps between occupations. Such modeling can sharpen a chronology, but the improvement depends on the validity of the assumptions; an incorrect sequence or unjustified phase structure can produce precise-looking but misleading results.\n\nWiggle matching uses samples separated by known time intervals, particularly tree rings. Instead of calibrating each measurement independently, the sequence is shifted against the calibration curve while preserving the known spacing. A distinctive pattern can identify the calendar position more effectively than one date alone. Sharp atmospheric carbon-14 increases recorded in tree rings, including those associated with the events of 774–775 and 993–994 CE, can sometimes provide annual chronological anchors. Their use requires identifying the relevant signal securely and linking it to a correctly counted ring sequence.\n\nEnvironmental archives present related but distinct problems. Lake and marine sediment cores commonly require age–depth models that interpolate between dated levels while accounting for accumulation rates and interruptions. Bulk sediment dates may reflect mixed sources, so terrestrial plant macrofossils are often preferred where available. Erosion surfaces, sediment mixing, and abrupt deposition can invalidate simple interpolation. Good chronological reporting therefore distinguishes the measured sample age, its calibrated distribution, the modeled age of a deposit, and the inferred timing of an environmental or cultural event. These are connected quantities, not interchangeable labels."
    },
    {
      "heading": "Applications beyond conventional archaeological dating",
      "text": "Radiocarbon dating underpins chronologies of human settlement, agriculture, monument construction, and cultural change. It permits comparison between archaeological sequences and independently reconstructed environmental events. It also provides age control for research involving ancient DNA, allowing genetic observations to be placed in time rather than ordered only by inferred relationships. Directly dating a bone or associated material can be important where the age of its surrounding sediment is uncertain. Nevertheless, a date establishes temporal placement, not by itself the cause of migration, technological change, or population replacement.\n\nIn geology and paleoclimatology, radiocarbon constrains the timing of glacier retreat, wetland development, coastal change, and late Quaternary ecosystem shifts. Organic material beneath or above a deposit may provide limiting ages for floods, landslides, or volcanic events. Peat and lake sequences preserve carbon suitable for age–depth modeling, while marine measurements help reconstruct past ocean circulation. Comparisons between contemporaneous atmospheric and deep-water radiocarbon reveal aspects of ventilation history, although converting these differences into circulation rates requires models of mixing and carbon exchange.\n\nAtmospheric nuclear weapons tests during the 1950s and early 1960s produced a large artificial increase in carbon-14. Northern Hemisphere atmospheric levels reached a peak in 1963, after which exchange with the ocean and biosphere, together with fossil-carbon emissions, reduced the excess. This “bomb pulse” supplies a tracer for recent carbon movement. Suitable materials can sometimes be assigned formation periods within the second half of the twentieth century or later. Because similar radiocarbon levels occur on the rising and falling limbs of the curve, a measurement may have more than one possible date unless other evidence resolves the ambiguity.\n\nBomb radiocarbon is used to investigate carbon turnover in soils, long-lived plants, biological tissues, and other reservoirs. Measurements of selected cellular components can reveal when their constituent carbon was synthesized, provided the relationship between carbon incorporation and tissue formation is understood. Forensic applications may distinguish pre-bomb from post-bomb material or constrain the period when a biological substance formed. These are not always direct estimates of an individual’s birth or death: tissue turnover and carbon-source history determine what the measurement represents.\n\nRadiocarbon also distinguishes fossil carbon from recently exchanged biological carbon. Petroleum, coal, and geologically old natural gas contain negligible measurable carbon-14, whereas contemporary biomass contains atmospheric-derived radiocarbon. This contrast supports studies of fossil-fuel contributions to atmospheric carbon dioxide, source attribution in aerosols, and measurement of biobased carbon in products. In art and cultural heritage research, radiocarbon can test whether a canvas, parchment, wooden support, or organic binding medium is compatible with a proposed period. It dates the selected carbon fraction, however, rather than the act of painting, writing, or assembly; old materials and later restoration can separate those events by substantial intervals."
    }
  ],
  "links": [
    "radiometric-dating",
    "isotope",
    "carbon",
    "carbon-cycle",
    "archaeology",
    "neutron",
    "cosmic-ray",
    "carbon-dioxide",
    "photosynthesis",
    "atmosphere-of-earth",
    "ocean",
    "beta-decay",
    "electron",
    "half-life",
    "magnetic-field",
    "radioactivity",
    "nobel-prize-in-chemistry",
    "dendrochronology",
    "cellulose",
    "collagen",
    "solvent",
    "chromatography",
    "graphite",
    "mass-spectrometry",
    "probability-distribution",
    "stratigraphy",
    "bayesian-inference",
    "ancient-dna",
    "geology",
    "paleoclimatology",
    "cultural-heritage"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}