Radiogenic Isotopes and Geochronology
CSIR NET Earth Sciences geochemistry: isochrons, Rb-Sr, Sm-Nd, epsilon Nd and model ages, Lu-Hf, Re-Os, U-Th-Pb concordia and Pb-Pb, Ar-Ar, U-series, mixing.
- 17 min
On this page · 15 sections
- Prerequisites
- The Hook: One Granite, Five Ages
- The Main Systems
- How the Systems Fractionate in the Earth
- Sm–Nd: Epsilon Notation and Model Ages
- The U–Th–Pb System
- K–Ar, Ar–Ar and U-Series in Brief
- Mixing of Two Sources
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- The decay law, the age equation, closure temperature and the basic dating methods from Part B: Geological Time Scale and Radiometric Dating. This page assumes that material and goes further.
- Compatible and incompatible elements: Atomic Structure, the Periodic Table and Ionic Substitution and Distribution Coefficients, Nucleation and Diffusion.
- Decay modes: Nuclear Stability and Radioactive Decay.
The Hook: One Granite, Five Ages
The Main Systems
| Parent → daughter (decay) | Decay constant \( \lambda \) (yr\(^{-1}\)) | Half-life | Reference (stable) isotope |
|---|---|---|---|
| \( ^{87}\text{Rb}\to{}^{87}\text{Sr} \) (\( \beta^- \)) | \( 1.42\times 10^{-11} \) (conventional) | about 48.8 Gyr | \( ^{86}\text{Sr} \) |
| \( ^{147}\text{Sm}\to{}^{143}\text{Nd} \) (\( \alpha \)) | \( 6.54\times 10^{-12} \) | about 106 Gyr | \( ^{144}\text{Nd} \) |
| \( ^{176}\text{Lu}\to{}^{176}\text{Hf} \) (\( \beta^- \)) | about \( 1.87\times 10^{-11} \) | about 37 Gyr | \( ^{177}\text{Hf} \) |
| \( ^{187}\text{Re}\to{}^{187}\text{Os} \) (\( \beta^- \)) | about \( 1.67\times 10^{-11} \) | about 42 Gyr | \( ^{188}\text{Os} \) |
| \( ^{238}\text{U}\to{}^{206}\text{Pb} \) (chain) | \( 1.55125\times 10^{-10} \) | 4.468 Gyr | \( ^{204}\text{Pb} \) |
| \( ^{235}\text{U}\to{}^{207}\text{Pb} \) (chain) | \( 9.8485\times 10^{-10} \) | 703.8 Myr | \( ^{204}\text{Pb} \) |
| \( ^{232}\text{Th}\to{}^{208}\text{Pb} \) (chain) | \( 4.9475\times 10^{-11} \) | about 14.0 Gyr | \( ^{204}\text{Pb} \) |
| \( ^{40}\text{K}\to{}^{40}\text{Ar} \) (EC, about 10.5%) and \( ^{40}\text{Ca} \) (\( \beta^- \)) | total \( 5.543\times 10^{-10} \); \( \lambda_e = 0.581\times 10^{-10} \) | 1.25 Gyr | \( ^{36}\text{Ar} \) (or \( ^{39}\text{Ar} \) after irradiation) |
The U, Th and K constants are the conventional values of Steiger and Jäger (1977). For Rb a slightly lower value (about \( 1.40\times 10^{-11} \)) is now often preferred; use whatever a question gives.
The general isochron
Every system above gives the same straight-line form (written for Sm–Nd):
$$\boxed{\left(\frac{^{143}\text{Nd}}{^{144}\text{Nd}}\right)_{\text{now}} = \left(\frac{^{143}\text{Nd}}{^{144}\text{Nd}}\right)_{0} + \left(\frac{^{147}\text{Sm}}{^{144}\text{Nd}}\right)_{\text{now}}\left(e^{\lambda t} - 1\right)}$$Slope \( m = e^{\lambda t} - 1 \Rightarrow t = \ln(1 + m)/\lambda \); intercept = initial ratio. When \( \lambda t \ll 1 \), \( t \approx m/\lambda \).
Requirements: cogenetic samples (same age, same initial ratio), a spread in parent/daughter, and a closed system since formation. Scatter beyond analytical error (expressed by the MSWD of the regression) means one of these failed: the line is then an errorchron.
How the Systems Fractionate in the Earth
Melting of the mantle separates parent from daughter, because one is more incompatible than the other. This is what makes isotopes tracers.
| System | More incompatible | Melt / continental crust gets | Depleted mantle residue gets | Over time, crust becomes |
|---|---|---|---|---|
| Rb–Sr | Rb (parent) | high Rb/Sr | low Rb/Sr | high \( ^{87}\text{Sr}/^{86}\text{Sr} \) |
| Sm–Nd | Nd (daughter) | low Sm/Nd | high Sm/Nd | low \( ^{143}\text{Nd}/^{144}\text{Nd} \) (negative \( \varepsilon_{\text{Nd}} \)) |
| Lu–Hf | Hf (daughter) | low Lu/Hf | high Lu/Hf | low \( ^{176}\text{Hf}/^{177}\text{Hf} \) (negative \( \varepsilon_{\text{Hf}} \)) |
| Re–Os | Re (parent); Os is compatible | high Re/Os | very low Re/Os | very radiogenic \( ^{187}\text{Os}/^{188}\text{Os} \) |
| U–Th–Pb | U, Th (parents) slightly more than Pb | high U/Pb (but Pb is fluid-mobile) | lower U/Pb | radiogenic Pb (complicated by Pb transport in fluids) |
Note the opposite senses of Rb–Sr and Sm–Nd: crust is high in \( ^{87}\text{Sr}/^{86}\text{Sr} \) but low in \( ^{143}\text{Nd}/^{144}\text{Nd} \). Plotted against each other, oceanic basalts form the negatively sloping mantle array.
Sr isotope evolution
- Basaltic achondrites give the solar system initial (BABI) of about 0.6990.
- Bulk Earth today is estimated at about 0.7045; MORB about 0.7025–0.703; old continental crust 0.71 and higher.
- Modern seawater \( ^{87}\text{Sr}/^{86}\text{Sr} \) is about 0.7092, a mix of river input (radiogenic, from continental weathering) and hydrothermal input (mantle-like). Its change through time is the basis of Sr isotope stratigraphy of marine carbonates; the steep Cenozoic rise is often linked to Himalayan weathering.
A granite with an initial ratio near 0.703 points to a mantle-derived source (I-type, juvenile); one near 0.710 or higher points to old crustal sources (typical S-type).
Sm–Nd: Epsilon Notation and Model Ages
Sm and Nd are both REE, so they are hard to separate by weathering, alteration or metamorphism. Whole-rock Sm–Nd therefore survives events that reset Rb–Sr.
\( \varepsilon_{\text{Nd}} \)
Differences in \( ^{143}\text{Nd}/^{144}\text{Nd} \) are small, so they are expressed in parts per 10,000 relative to the Chondritic Uniform Reservoir (CHUR), a model of the bulk silicate Earth:
$$\boxed{\varepsilon_{\text{Nd}}(t) = \left[\frac{(^{143}\text{Nd}/^{144}\text{Nd})_{\text{sample}}(t)}{(^{143}\text{Nd}/^{144}\text{Nd})_{\text{CHUR}}(t)} - 1\right]\times 10^{4}}$$Present-day CHUR values widely used: \( ^{143}\text{Nd}/^{144}\text{Nd} = 0.512638 \) and \( ^{147}\text{Sm}/^{144}\text{Nd} = 0.1967 \) (Jacobsen and Wasserburg, 1980; later revisions differ slightly in the fourth or fifth decimal). Both sample and CHUR must be calculated at the same time \( t \) by subtracting their ingrowth since \( t \).
| \( \varepsilon_{\text{Nd}} \) | Meaning | Typical |
|---|---|---|
| Positive | source had Sm/Nd above chondritic for a long time: depleted mantle | MORB about +8 to +10 |
| About 0 | chondritic source | some OIB, primitive estimates |
| Negative | source had low Sm/Nd: old continental crust or enriched mantle | old upper crust strongly negative (tens of units) |
Model ages
A model age is the time at which the sample’s Nd isotope ratio was equal to that of a model reservoir, assuming its Sm/Nd has not changed since it left that reservoir:
$$\boxed{T_{\text{CHUR}} = \frac{1}{\lambda}\ln\left[1 + \frac{(^{143}\text{Nd}/^{144}\text{Nd})_{\text{s}} - (^{143}\text{Nd}/^{144}\text{Nd})_{\text{CHUR}}}{(^{147}\text{Sm}/^{144}\text{Nd})_{\text{s}} - (^{147}\text{Sm}/^{144}\text{Nd})_{\text{CHUR}}}\right]}$$\( T_{\text{DM}} \) is the same formula with depleted-mantle values in place of CHUR (several parameterisations are in use, so \( T_{\text{DM}} \) values differ slightly between authors). For crustal rocks \( T_{\text{DM}} \) is the usual estimate of the crust-formation age: when the material was extracted from the mantle. If \( T_{\text{DM}} \) is much older than the crystallisation age, the rock is a melt of older crust; if they are similar, the crust was juvenile.
Lu–Hf in zircon
Zircon contains about 1 wt% Hf but very little Lu, so its \( ^{176}\text{Hf}/^{177}\text{Hf} \) barely changes after crystallisation. A zircon dated by U–Pb and analysed for Hf isotopes on the same spot gives both a crystallisation age and an initial \( \varepsilon_{\text{Hf}} \): the source signature. Detrital zircon U–Pb–Hf data are the main record of crustal growth through time.
Re–Os
Os is highly siderophile and compatible in the mantle (it stays in sulphides and alloys during melting), while Re is moderately incompatible. So:
- Mantle peridotites lose Re during melting and their Os isotopes freeze. The rhenium-depletion model age (\( T_{\text{RD}} \)) of a mantle xenolith gives a minimum age for melt extraction: the standard tool for dating the lithospheric mantle under cratons.
- Re–Os on molybdenite (Re-rich, almost no initial Os) dates ore deposition directly; Re–Os on sulphides and organic-rich shales dates mineralisation and deposition.
The U–Th–Pb System
Concordia and discordia
For a closed mineral with no initial Pb:
$$\frac{^{206}\text{Pb}^*}{^{238}\text{U}} = e^{\lambda_{238}t} - 1 \qquad \frac{^{207}\text{Pb}^*}{^{235}\text{U}} = e^{\lambda_{235}t} - 1$$The curve traced by the pair as \( t \) increases is the concordia (Wetherill diagram: \( ^{206}\text{Pb}^*/^{238}\text{U} \) against \( ^{207}\text{Pb}^*/^{235}\text{U} \); the Tera–Wasserburg diagram plots \( ^{238}\text{U}/^{206}\text{Pb} \) against \( ^{207}\text{Pb}/^{206}\text{Pb} \) and is preferred for young samples with common Pb).
- Concordant grain: both ages agree; it lies on the curve.
- Pb loss in a later event moves grains towards the origin of the Wetherill diagram along a straight chord, the discordia. Upper intercept = crystallisation age; lower intercept = age of Pb loss (if episodic).
- Inheritance: old cores in younger zircon rims give mixed analyses; in-situ methods (SHRIMP, LA-ICP-MS) date cores and rims separately.
Zircon, monazite, baddeleyite (in mafic rocks), titanite and rutile are the usual U–Pb minerals; their closure temperatures differ (zircon is among the highest; titanite and rutile are much lower), so they can record different parts of a P–T–t path.
Pb–Pb ages
Combining the two decay equations, with the present \( ^{238}\text{U}/^{235}\text{U} = 137.88 \) (the conventional value), the slope of an isochron of cogenetic samples on a \( ^{207}\text{Pb}/^{204}\text{Pb} \) against \( ^{206}\text{Pb}/^{204}\text{Pb} \) plot is
$$\boxed{\frac{\Delta(^{207}\text{Pb}/^{204}\text{Pb})}{\Delta(^{206}\text{Pb}/^{204}\text{Pb})} = \left(\frac{^{207}\text{Pb}}{^{206}\text{Pb}}\right)^* = \frac{1}{137.88}\cdot\frac{e^{\lambda_{235}t} - 1}{e^{\lambda_{238}t} - 1}}$$The age needs no U measurement, so it is robust against recent U loss. The equation is solved numerically or from a table:
| \( t \) (Ga) | 0.5 | 1.0 | 1.5 | 2.0 | 2.5 | 3.0 | 3.5 | 4.0 | 4.5 |
|---|---|---|---|---|---|---|---|---|---|
| radiogenic \( ^{207}\text{Pb}/^{206}\text{Pb} \) | 0.0572 | 0.0725 | 0.0936 | 0.1230 | 0.1643 | 0.2227 | 0.3058 | 0.4250 | 0.5967 |
The ratio rises with age because \( ^{235}\text{U} \) was relatively more abundant in the past. Patterson’s meteorite Pb–Pb isochron (the geochron) gave the age of the Earth and solar system.
Common Pb and its evolution
\( ^{204}\text{Pb} \) is not radiogenic, so it measures non-radiogenic (“common”) Pb. Source U/Pb is expressed as \( \mu = {}^{238}\text{U}/^{204}\text{Pb} \) and Th/U as \( \kappa = {}^{232}\text{Th}/^{238}\text{U} \). In the Holmes–Houtermans single-stage model, all terrestrial Pb evolved from a primordial composition (Canyon Diablo troilite) in reservoirs of constant \( \mu \); a galena’s Pb isotope ratio then gives a model age. The Stacey–Kramers two-stage model refines this and is the usual common-Pb correction.
Mantle reservoirs
Oceanic basalts define a few isotopic end-members (Zindler and Hart, 1986):
| End-member | Isotope character | Usual interpretation |
|---|---|---|
| DMM (depleted MORB mantle) | low \( ^{87}\text{Sr}/^{86}\text{Sr} \), high \( \varepsilon_{\text{Nd}} \), unradiogenic Pb | residue of continental-crust extraction |
| HIMU (high \( \mu \)) | very radiogenic \( ^{206}\text{Pb}/^{204}\text{Pb} \), low Sr | recycled oceanic crust that lost Pb in subduction |
| EM1 | moderately high Sr, low \( \varepsilon_{\text{Nd}} \), low \( ^{206}\text{Pb}/^{204}\text{Pb} \) | recycled lower crust, lithosphere or pelagic sediment |
| EM2 | high \( ^{87}\text{Sr}/^{86}\text{Sr} \), low \( \varepsilon_{\text{Nd}} \) | recycled continental (terrigenous) sediment |
K–Ar, Ar–Ar and U-Series in Brief
K–Ar age:
$$\boxed{t = \frac{1}{\lambda}\ln\left[1 + \frac{\lambda}{\lambda_e}\cdot\frac{^{40}\text{Ar}^*}{^{40}\text{K}}\right]} \qquad \frac{\lambda}{\lambda_e} = \frac{5.543}{0.581} = 9.54$$In \( ^{40}\text{Ar}/^{39}\text{Ar} \) dating, neutron irradiation converts \( ^{39}\text{K} \) to \( ^{39}\text{Ar} \); a standard of known age irradiated alongside gives the irradiation parameter \( J \), and \( t = \frac{1}{\lambda}\ln\left(1 + J\frac{^{40}\text{Ar}^*}{^{39}\text{Ar}}\right) \). Step-heating gives an age spectrum: a flat plateau suggests a closed system; a rising spectrum suggests Ar loss from grain margins; an inverse isochron identifies excess (non-atmospheric) argon.
U-series disequilibrium. In a closed system older than several half-lives of the intermediate daughters, every member of a decay chain has the same activity (\( \lambda N \)): secular equilibrium. Weathering, melting or precipitation separates the members (U is soluble as U(VI), Th is insoluble). Return to equilibrium is a clock:
- Corals and speleothems take up U but no Th, so \( ^{230}\text{Th} \) grows towards equilibrium with \( ^{234}\text{U} \) (half-life of \( ^{230}\text{Th} \) about 75 kyr): ages up to about 500 kyr.
- Excess \( ^{230}\text{Th} \) and \( ^{210}\text{Pb} \) in sediments decay with depth and give accumulation rates.
- \( (^{230}\text{Th}/^{238}\text{U}) \) disequilibria in young lavas record melting rates and the presence of garnet in the source.
Mixing of Two Sources
When two components A and B (concentrations \( C_A \), \( C_B \) of the element; isotope ratios \( R_A \), \( R_B \)) mix in proportion \( f \) of A, the isotope ratio of the mixture is the concentration-weighted mean:
$$\boxed{R_M = \frac{fC_AR_A + (1 - f)C_BR_B}{fC_A + (1 - f)C_B}}$$(strictly with concentrations of the reference isotope, but element concentrations are adequate for Sr and Nd). On an isotope ratio against \( 1/C \) plot, mixtures lie on a straight line; on a plot of two isotope ratios with different elements, the mixing curve is a hyperbola whose curvature depends on the ratio of the two elements’ concentration ratios. Crustal contamination of a magma that is also crystallising (AFC) adds a further term.
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| Isochron age (crystallisation or closure) | Model age \( T_{\text{CHUR}} \), \( T_{\text{DM}} \) (separation from mantle) | |
| \( \varepsilon_{\text{Nd}} > 0 \): depleted mantle source | \( \varepsilon_{\text{Nd}} < 0 \): old crustal or enriched source | |
| Concordia (curve of agreeing U–Pb ages) | Discordia (chord of Pb-loss arrays) | |
| Pb–Pb age: no U needed | U–Pb age: needs U/Pb ratio | |
| Isotope ratios (dimensionless) | Isotope activities (decays per time) in U-series | |
| Mixing line in ratio vs \( 1/C \) | Isochron line in ratio vs parent/daughter |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper. Use the constants in the tables above.
Level 1: Foundation
- Which reservoir has positive \( \varepsilon_{\text{Nd}} \)?
- What is the upper intercept of a discordia with concordia?
- Why is \( ^{204}\text{Pb} \) used as the reference isotope for Pb?
Answers
- Depleted mantle (MORB source).
- Usually the crystallisation age of the zircon population.
- It has no radioactive parent, so it measures common (initial) Pb.
Level 2: Part B Style
Practice Q1. A Sm–Nd mineral isochron (garnet, cpx, whole rock) for an eclogite has slope 0.0130. Its age is closest to:
(a) 0.92 Ga (b) 1.47 Ga (c) 1.97 Ga (d) 2.99 Ga
Answer: (c) 1.97 Ga
\( t = \ln(1.0130)/6.54\times 10^{-12} = 0.012916/6.54\times 10^{-12} = 1.97\times 10^{9} \) yr. The shortcut \( m/\lambda = 1.99 \) Ga is close.
Practice Q2. A zircon has radiogenic \( ^{206}\text{Pb}^*/^{238}\text{U} = 0.50 \). Its \( ^{206}\text{Pb}/^{238}\text{U} \) age is:
(a) 1.21 Ga (b) 2.23 Ga (c) 2.61 Ga (d) 3.22 Ga
Answer: (c) 2.61 Ga
\( t = \ln(1.50)/1.55125\times 10^{-10} = 0.4055/1.55125\times 10^{-10} = 2.61\times 10^{9} \) yr.
Practice Q3. A volcanic sanidine has \( ^{40}\text{Ar}^*/^{40}\text{K} = 0.0100 \). Its K–Ar age is closest to:
(a) 18 Ma (b) 164 Ma (c) 1.21 Ga (d) 1.64 Ga
Answer: (b) 164 Ma
\( t = \frac{1}{5.543\times 10^{-10}}\ln(1 + 9.54\times 0.0100) = \frac{\ln 1.0954}{5.543\times 10^{-10}} = \frac{0.0911}{5.543\times 10^{-10}} = 1.64\times 10^{8} \) yr. Option (a) leaves out the factor 9.54.
Practice Q4. A basalt has \( ^{87}\text{Sr}/^{86}\text{Sr} = 0.7028 \) and \( \varepsilon_{\text{Nd}} = +9 \). Its source is most likely:
(a) old continental crust (b) depleted upper mantle (c) recycled terrigenous sediment (EM2) (d) a chondritic reservoir
Answer: (b)
Low Sr isotope ratio and strongly positive \( \varepsilon_{\text{Nd}} \) both indicate a long-term depletion in Rb/Sr and enrichment in Sm/Nd: the MORB-source mantle. A chondritic source would have \( \varepsilon_{\text{Nd}} \approx 0 \).
Level 3: Part C Style
Practice Q5. A granite crystallised at 1.0 Ga (zircon U–Pb). Today it has \( ^{143}\text{Nd}/^{144}\text{Nd} = 0.511900 \) and \( ^{147}\text{Sm}/^{144}\text{Nd} = 0.1100 \). With CHUR (0.512638, 0.1967) and \( \lambda = 6.54\times 10^{-12} \) yr\(^{-1}\), which statements are correct?
- \( \varepsilon_{\text{Nd}} \) at 1.0 Ga is about −3.3.
- \( T_{\text{CHUR}} \) is about 1.30 Ga, older than the crystallisation age.
- The magma was derived entirely from depleted mantle at 1.0 Ga.
- The rock’s present \( \varepsilon_{\text{Nd}} \) (about −14) is more negative than its initial value because its Sm/Nd is lower than CHUR.
(a) 1 and 3 only (b) 1, 2 and 4 only (c) 2 and 3 only (d) 1, 2, 3 and 4
Answer: (b) 1, 2 and 4 only
\( e^{\lambda t} - 1 \) at 1.0 Ga = \( e^{0.00654} - 1 = 0.006561 \).
- Sample at 1.0 Ga: \( 0.511900 - 0.1100\times 0.006561 = 0.511178 \). CHUR at 1.0 Ga: \( 0.512638 - 0.1967\times 0.006561 = 0.511347 \).
- 1 true: \( (0.511178/0.511347 - 1)\times 10^4 = -3.3 \).
- 2 true: \( T_{\text{CHUR}} = \ln(1.008512)/6.54\times 10^{-12} = 1.30 \) Ga.
- 3 false: a negative initial \( \varepsilon_{\text{Nd}} \) needs an older, low-Sm/Nd crustal component; pure depleted-mantle melt would be strongly positive.
- 4 true: with \( ^{147}\text{Sm}/^{144}\text{Nd} = 0.110 < 0.1967 \), the sample’s Nd ratio grows more slowly than CHUR, so \( \varepsilon_{\text{Nd}} \) becomes more negative: −3.3 then, −14.4 now.
Practice Q6. Cogenetic feldspar and sulphide samples from one ore field fall on a line in \( ^{207}\text{Pb}/^{204}\text{Pb} \) against \( ^{206}\text{Pb}/^{204}\text{Pb} \) with slope 0.164. Separately, a magma with 400 ppm Sr and \( ^{87}\text{Sr}/^{86}\text{Sr} = 0.7030 \) is contaminated by an equal mass of crust with 100 ppm Sr and 0.7200. Which pair of answers is correct? (Use the Pb–Pb table above.)
(a) Pb–Pb age about 2.5 Ga; mixture ratio about 0.7064 (b) Pb–Pb age about 2.5 Ga; mixture ratio about 0.7115 (c) Pb–Pb age about 1.6 Ga; mixture ratio about 0.7064 (d) Pb–Pb age about 4.0 Ga; mixture ratio about 0.7115
Answer: (a)
- A radiogenic \( ^{207}\text{Pb}/^{206}\text{Pb} \) slope of 0.164 matches the table at 2.5 Ga (0.1643).
- Mixing with \( f = 0.5 \): \( R_M = \frac{0.5\times 400\times 0.7030 + 0.5\times 100\times 0.7200}{0.5\times 400 + 0.5\times 100} = \frac{140.6 + 36.0}{250} = 0.7064 \).
- 0.7115 is the simple average of 0.7030 and 0.7200, which ignores the fourfold difference in Sr content.
Quick Revision Table
| Concept | Key point |
|---|---|
| Isochron | slope \( e^{\lambda t} - 1 \); intercept = initial ratio; MSWD tests fit |
| \( \lambda \) values | Rb 1.42e−11; Sm 6.54e−12; \( ^{238}\text{U} \) 1.55125e−10; \( ^{235}\text{U} \) 9.8485e−10; K 5.543e−10 |
| Fractionation | Rb, Re, U parents incompatible; Nd, Hf daughters incompatible |
| Crust | high \( ^{87}\text{Sr}/^{86}\text{Sr} \), negative \( \varepsilon_{\text{Nd}} \), \( \varepsilon_{\text{Hf}} \) |
| Seawater Sr | about 0.7092 today; Sr isotope stratigraphy |
| CHUR | 0.512638; \( ^{147}\text{Sm}/^{144}\text{Nd} \) 0.1967 |
| \( \varepsilon_{\text{Nd}} \) | (sample/CHUR − 1) × 10\(^4\) at the same \( t \) |
| Model age | \( T = \frac{1}{\lambda}\ln[1 + \Delta(\text{Nd})/\Delta(\text{Sm/Nd})] \) |
| Re–Os | Os compatible; \( T_{\text{RD}} \) of mantle xenoliths; molybdenite ages |
| U–Pb | concordia; discordia upper/lower intercepts; zircon, monazite, baddeleyite |
| Pb–Pb | slope \( = \frac{1}{137.88}\frac{e^{\lambda_5t} - 1}{e^{\lambda_8t} - 1} \) |
| K–Ar | \( t = \frac{1}{\lambda}\ln(1 + 9.54\,\text{Ar}^*/\text{K}) \) |
| U-series | secular equilibrium; \( ^{230}\text{Th} \) about 75 kyr; to about 500 kyr |
| Mantle end-members | DMM, HIMU, EM1, EM2 |
| Mixing | concentration-weighted ratios; straight line vs \( 1/C \) |
Key Takeaways
What’s Next
Radiogenic isotopes vary because of decay; stable isotopes vary because of mass-dependent fractionation. Continue with Stable Isotopes and Geochemical Cycles. For the geological time scale that these ages calibrate, return to Geological Time Scale and Radiometric Dating.