Note to users. If you're seeing this message, it means that your browser cannot find this page's style/presentation instructions -- or possibly that you are using a browser that does not support current Web standards. Find out more about why this message is appearing, and what you can do to make your experience of our site the best it can be.
Heterogeneous Hadean Hafnium: Evidence of Continental Crust at 4.4 to 4.5 Ga
T. M. Harrison,1,2*J. Blichert-Toft,3W. Müller,1,4F. Albarede,3P. Holden,1S. J. Mojzsis5
The long-favored paradigm for the development of continentalcrust is one of progressive growth beginning at 4 billion yearsago (Ga). To test this hypothesis, we measured initial 176Hf/177Hfvalues of 4.01- to 4.37-Ga detrital zircons from Jack Hills,Western Australia. Hf (deviations of 176Hf/177Hf from bulk Earthin parts per 104) values show large positive and negative deviationsfrom those of the bulk Earth. Negative values indicate the developmentof a Lu/Hf reservoir that is consistent with the formation ofcontinental crust (Lu/Hf 0.01), perhaps as early as 4.5 Ga.Positive Hf deviations require early and likely widespread depletionof the upper mantle. These results support the view that continentalcrust had formed by 4.4 to 4.5 Ga and was rapidly recycled intothe mantle.
1 Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia. 2 Department of Earth and Space Sciences and Institute of Geophysics and Planetary Physics, University of California at Los Angeles, Los Angeles, CA 90095, USA. 3 Ecole Normale Supérieure, CNRS Unité Mixte de Recherche 5570, 69364 Lyon Cedex 7, France. 4 Department of Geology, Royal Holloway University of London, Egham, TW20 0EX, UK. 5 Department of Geological Sciences, University of Colorado, Boulder, CO 80309, USA.
* To whom correspondence should be addressed. E-mail: mark.harrison{at}anu.edu.au
A fundamental question of Earth's evolution is: When did thegrowth of continental crust begin? One model is that the firstcrust formed after 4 Ga and grew slowly until the present day(1, 2). This view reflects the absence of a >4-Ga rock record(3) and the broadly coherent post–4 Ga evolution of depletedmantle 143Nd/144Nd (4) and 176Hf/177Hf (5). Long-standing observationsof early Nd (6) and Hf (7, 8) depletions, however, leave openthe possibility of even earlier global fractionations. Anotherview (9, 10) is that continental crust was widespread duringthe Hadean Eon [the first 500 million years (My) of Earth history].In such a scenario, the lack of direct evidence of earlier depletionevents reflects subsequent remixing. Detrital zircons from JackHills, Western Australia, with 4.0- to 4.4-Ga U-Pb ages (11–13)represent pieces of crust that have been sequestered for upto 4.4 Ga. Hf isotopic compositions vary because of radioactivedecay of 176Lu, and such variations in zircons constitute anexcellent tracer of Earth's crust/mantle differentiation. Thisis because zircons have very low Lu/Hf ratios and thus recordnear-initial 176Hf/177Hf at the time given by their U-Pb age.Amelin and co-workers (14) investigated Hf isotopes in JackHills zircons as old as 4.14 Ga and inferred the existence ofreworked Hadean crust. We have now extended this applicationby undertaking Lu-Hf analyses of grains ranging in age up to4.37 Ga, thereby narrowing the gap to less than 200 My fromthe end of Earth's accretion to the first mineral record. Wedocument significant Hf isotopic heterogeneity during the earlyHadean and conclude that major differentiation of the silicateEarth, possibly the formation of continental crust with a volumesimilar in magnitude to the present day, may have occurred by4.4 to 4.5 Ga.
Using the multicollector Sensitive High Resolution Ion MicroprobeII, we have surveyed the radiogenic 207Pb/206Pb (207Pb/206Pb*)ratio of over 50,000 Jack Hills zircons separated from a largeconglomerate sample (JH992) obtained from the original localityof Compston and Pidgeon (11). Hadean grains thus identifiedwere dated by the U-Pb method (11–13). Zircons between4.01 and 4.37 Ga were then analyzed for Lu/Hf and Hf isotopiccomposition (tables S1 and S2). Although 207Pb/206Pb* ages canbe variable within Hadean zircons (11–13), it is rarelythe case (15) that an individual grain exhibits more than twogenerations of crystal growth. In the cases of concordant ornearly concordant (<12% discordant) results, we assume thatthe 207Pb/206Pb* age records the time of crystallization, whichwe then use to calculate Hf(T), where T is age. For grains thatshowed evidence of zoning or yielded apparent Hf(T) values thatstrongly deviated from those of bulk Earth, we undertook additionalion microprobe dating investigations to assess whether multiplegenerations of zircon growth were present (table S2 and fig.S2).
Of the 104 multicollector–inductively coupled plasma–massspectrometry (MC-ICP-MS) zircon 176Hf/177Hf results we present,44 were measured by solution MC-ICP-MS (table S1) (16) and 60by laser ablation MC-ICP-MS (table S2) (16). We deliberatelyused both approaches, because each has potential limitationsthat are in part compensated for by the other. Solution MC-ICP-MSprovides a bulk Hf isotope composition, but does not sufferfrom isobaric interferences on 176Hf from Yb and Lu. Althoughlaser ablation MC-ICP-MS measurements require peak strippingto correct for Yb and Lu interferences on 176Hf, they allowspatially resolved analysis using spots of 60 to 80 µmin diameter, which facilitates the identification of zirconszoned with respect to 176Hf/177Hf. Because these analyses retainsome of the zircon under analysis, subsequent ion microprobeinvestigations of potential age heterogeneity are possible.
We calculated Hf(T) using the "terrestrial" 176Lu decay constant(176) of 1.867 ± 0.008 x 10–11 year–1 [(17)and references therein] and the present day chondritic parameters176Lu/177Hf = 0.0332 ± 0.0002 and 176Hf/177Hf = 0.282772± 29 (18) [176 and solar system initial 176Hf/177Hf deducedfrom meteorite Lu-Hf isochrons appear to reflect irradiationeffects in the solar nebula (19)]. Using the chondritic averagesof Patchett and co-workers (20) instead would not significantlychange our calculated values of Hf(T), because their 176Hf/177Hfand Lu/Hf values all fall on the same 4.5-Ga isochron as thosein (18).
Our data (tables S1 and S2) show a high degree of Hf isotopeheterogeneity, with variation in Hf(T) values at 4.2 Ga of over20 units. Before we examine the geological significance ofthese data, we investigate possible artifacts of combined interpretationof Lu-Hf and U-Pb systematics in a complex zircon populationthat could lead to incorrect inferences of early deviationsfrom chondritic evolution. A key link in using zircon to assessinitial Hf isotopic ratios is that the age ascribed accuratelyreflects the time at which the Hf isotopic composition was incorporatedinto the zircon. In the case of a zircon zoned in U-Pb age,relating an older age component to the bulk Hf isotope compositioncould result in an incorrect estimate of Hf(T).
For example, consider the case of grain CU05 8.5, which yieldsan ion microprobe age of 4123 ± 18 Ma and age-correctedsolution MC-ICP-MS 176Hf/177Hf = 0.280415 ± 0.000015(2 SE) (table S1), corresponding to an apparent Hf(T) = +10.7± 0.5. However, because the bulk solution ICP-MS 207Pb/206Pbage is only 3624 Ma (21), we infer that this grain is substantiallyheterogeneous with respect to age and that the 4.1-Ga ion microprobedate represents only a small portion of the total U-Pb system.If we instead use the bulk 3.62-Ga age, the apparent Hf dropsto a near–bulk Earth value of +1.2 ± 0.5. We thuseliminated this sample from further consideration.
Less obvious is that mixtures of concordant components withcontrasting U concentrations each plotting on bulk Earth couldalso yield substantial positive and negative deviations in Hf.This is because of the strongly nonlinear relationship between207Pb/206Pb* (or Pb/U) and age, whereas the long half-life of176Lu resulted in essentially linear growth in 176Hf/177Hf overthe past 4.56 Gy. To assess this effect, we developed a modelin which concordant core (T1) and rim (T2) ages are characterizedby Hf(T) at T1 and T2, respectively. If the bulk age of thiscomposite is characterized by a U-Pb date obtained by ion microprobespot analysis in the core, but associated with a 176Hf/177Hfratio that is a mixture of the two end-members, an incorrectvalue of Hf results. Even in the case for which U-Pb age andHf isotopes are measured on the same spot overlapping thesetwo zones, aberrant values can result. Figure 1 illustratespossible trajectories in Hf(T) resulting from mixing two suchcomponents. Because variations in U concentration are far greaterthan those in Hf concentration (22), we only varied U concentrations.
Fig. 1. Results of a model illustrating possible trajectories in Hf(T) versus age space ensuing from mixing zircon components of differing age, U concentration, and Hf isotope composition. The model assumes concordant ages of T1 (core) and T2 (rim), which are characterized by the Hf isotope composition of bulk Earth at T1 and T2, respectively. Mixtures of the two components are shown adjacent the relevant curve in % of the rim component for the case of a 4.4-Ga core and 3.7-Ga rim. Where the bulk age of a mixture is characterized by a U-Pb age in the core, but incorrectly associated with a 176Hf/177Hf ratio that is a mixture of the two end-members, the potential exists for incorrect Hf values to be calculated. More surprising is that in the case where U-Pb age and Hf isotopes are measured on the same spot overlapping these two zones, aberrant values result, as shown in the mixing curves.
[View Larger Version of this Image (20K GIF file)]
Consider the case of a 4.4-Ga core (176Hf/177Hf4.4 Ga = 0.279930)and a 3.7-Ga rim (176Hf/177Hf3.7 Ga = 0.280398) (i.e., an agecontrast of 1 half-life of 235U). Mixtures of these two componentsbetween 0 and 100%, shown in 10% increments, are indicated bythe black symbols (Fig. 1). The variation in symbol shape representsdiffering U concentrations of the 4.4-Ga core (T1) and the 3.7-Garim (T2). For the case of equal U concentration (diamonds),mixing produces a roughly symmetric trajectory in positive Hfspace, reaching an Hf value of +2 at 4 Ga. Doubling the U concentrationin the core relative to the rim (squares) increases the apparentHf value to close to +5. This is expected because the Hf isotopecomposition at every intermediate position is associated withan older apparent 207Pb/206Pb* age and thus appears to originatein a higher Lu/Hf environment. Where the rim U concentrationis a factor of 1.6 greater than in the core (circles), no variationfrom bulk Earth is seen (because 207Pb/206Pb* drops by a factorof 1.6 for each half-life of 235U). Thus, the reduced rate ofgeneration of Pb* at 3.7 Ga is in this case exactly offset bythe 160% greater U content of the rim. Where U in the core is10 times less than in the rim (triangles), Hf values as negativeas –7 result.
For many grains analyzed by laser ablation MC-ICP-MS, furtherage investigations are possible, because the method is not completelydestructive, compared with solution chemistry. Grains with calculatedHf that differ from bulk Earth by more than 5 units were, wheresufficient sample remained, redated near their rims to assessage homogeneity. In the majority of cases, 207Pb/206Pb* ageswere reproduced at a satisfactory level (table S2). In one case(RSES17 2.9), we were able to resolve a time-dependent 176Hf/177Hfsignal that dropped by 4 units upon penetrating a 3469 ±12–Ma rim from a 4127 ± 64–Ma core, yieldingHf(4.13 Ga) = 10 ± 1.3 (table S2 and fig. S1).
Where rims are demonstrably younger than core ages, we wereable to rule out mixing between domains by the lack of a time-resolvedvariation in the 176Hf/177Hf signal and by visual evidence fromimaging studies. Even where mixing may have occurred below ourability to resolve time variations in 176Hf/177Hf, we have ruledout this effect as a cause for overestimating variations inHf(T) from bulk Earth by using our mixing model (table S2, comments).For example, grain ANU39 15.15 yields a 207Pb/206Pb age of 4234± 10 Ma (0% discordant), which corresponds to a highapparent Hf(T) of 15.3 ± 1.8 (table S2 and fig. S2).Both the ion microprobe spot for the U-Pb age and the laserablation pit for the Hf isotope analysis were obtained fromthe central portion of the grain, which was imaged as a homogenouscore (fig. S2). Although the overgrowth region was later foundto yield an age of 3548 ± 10 Ma (table S2 and fig. S2),our modeling shows that the relationship between U concentrationsin the core [82 parts per million (ppm)] and rim (440 ppm) wouldlikely increase the estimated value of Hf if a portion of therim had inadvertently been included in the analysis. Thus, thevalue of +15 is a minimum estimate of the true Hf of the graincore.
In four cases, measured 174Hf/177Hf and 178Hf/177Hf were outsideaccepted values (16), and these data were eliminated from consideration.Figure 2 shows the results for which we are confident that allthe aforementioned effects have not affected the true Hf values.Concordant or near-concordant results are shown as solid symbols,whereas >12% discordant results are shown as open symbols.The diagonal line extending into the negative Hf field froman age of 4.5 Ga, labeled Lu/Hf = 0, demarcates the "forbiddenregion," corresponding to Lu/Hf < 0. Data plotting near thisline must have been derived from an environment in which Luwas essentially completely fractionated from Hf shortly afterEarth formation. If that event was the formation of continentalcrust, then the average continental crust 176Lu/177Hf valueof 0.01 (23) is a more appropriate comparator. The many resultsthat plot along this line require formation of an enriched reservoirby 4.4 to 4.5 Ga.
Fig. 2. Plot of Hf(T) versus age for new MC-ICP-MS Hf isotope analyses together with results of Amelin and co-workers (14) recalculated using the "terrestrial" 176Lu decay constant (17). Solid symbols are concordant to <12% discordant, whereas open symbols are >12% discordant. The line marked Lu/Hf = 0 corresponds to 176Hf/177Hf ratios equivalent to the bulk silicate Earth value at 4.5 Ga. The stippled region indicates values not attainable (i.e., lower than solar system initial 176Hf/177Hf). Error bars indicate 2 SE.
[View Larger Version of this Image (34K GIF file)]
Positive Hf(T) deviations observed in the same age intervalhave not previously been documented (14) and imply derivationfrom reservoirs with high Lu/Hf ratios. These data almost surelyreflect the rapid development of depleted mantle with single-stage176Lu/177Hf ratios within the possible range of values (0.05and 0.2) for rocks residual from mantle melting (24). Whereasthe depleted signature we observe is roughly symmetrical withthe enriched reservoir (Fig. 2), negative Hf values outnumberpositive Hf by 2.5:1. This is in part expected, because rocksoriginating from a source with a depleted signature are lesspetrogenetically suited to crystallizing zircon.
Most models for the growth of continental crust have emphasizeddelayed, slow growth (1, 2, 4). The existence of Hadean JackHills zircons has been known for 20 years (11) but has beenlargely considered as a curiosity rather than a fundamentalrecord of the origin of continental crust (25). Investigationsof Jack Hills zircons [18O, inclusion assemblages, zircon thermometry(12, 13, 26–28)] indicate that the vast majority formedin a continental environment (26) characterized by two formsof convergent margin magmatism (crustal anatexis and calcalkalinemagmatism at or close to water saturation) throughout the Hadean.Together, these data indicate that Earth was experiencing continentalcrust formation during the Hadean and that a mature sedimentrecycling system similar to that of the known era of plate tectonicshad developed by 4.4 Ga.
Jack Hills zircons are largely of continental origin (12, 13,26–28), and our preferred interpretation of the variationof Hf(T) (Fig. 2) is that a major differentiation event occurredat 4.4 to 4.5 Ga, producing continental crust and a complementarydepleted mantle reservoir. Because the production of moderncontinental crust is intimately connected with orogenic magmas,our interpretation implies that plate boundary interactionsmay have begun within the first 100 My of Earth history. Ifthe relative fraction of depleted versus enriched samples (Fig. 2)is representative of the general ongoing process, the volumeof depleted mantle must have been a substantial fraction ofthe silicate Earth.
Positive Hf(T) deviations of +15 at 4.2 Ga imply a reservoirwith Lu/Hf of 0.1 (Fig. 2). This extrapolates to Hf of over+200 today, whereas values as negative as –7 at 4.2 Gaproject to Hf(0) of about –100. No evidence for thesereservoirs has yet been recognized in the post-Hadean rock record(5, 29). Although it is conceivable that such reservoirs existbut have remained hidden for 4 Gy (30), their disappearanceis well explained by the recycling of continental crust duringthe Hadean at rates 10 times greater than those at present (9,31), coupled with vigorous stirring in a hotter and thus less-viscousmantle.
Armstrong (9) emphasized that all large terrestrial planets,including Earth, must have immediately differentiated into relativelyconstant volume core, depleted mantle, enriched crust, and fluidreservoirs. Recent investigations of 142,143Nd/144Nd variationsindicate that a major silicate differentiation event occurredwithin 50 to 150 My (32, 33), and possibly even within 30 My(30) of Earth's formation. Our results support the view thatcontinental crust was at least a component of the enriched counterpartthat formed at 4.5 Ga, but this original crust was largely recycledback into the mantle by the onset of the Archean (<4 Ga).
References and Notes
1. J. Veizer, S. L. Jansen, J. Geol.87, 341 (1979).
2. S. R. Taylor, S. M. McLennan, The Continental Crust: Its Composition and Evolution (Blackwell Scientific, Oxford, 1985).
31. G. F. Davies, Geochim. Cosmochim. Acta66, 3125 (2002).
32. M. Boyet et al., Earth Planet. Sci. Lett.214, 427 (2003).
33. G. Caro, B. Bourdon, J. L. Birck, S. Moorbath, Nature423, 428 (2003). [CrossRef] [Medline]
34. We thank M. McCulloch, P. Lanc, Z. Bruce, A. Schmitt, T. Ireland, and S. Mussett for their major contributions to this project. The work was supported at the Australian National University by Australian Research Council grant DP0342709 to T.M.H., by the French Institut National des Sciences de l'Univers and the Programme National de Planétologie grants to J.B.T. and F.A., and by NASA Exobiology grant NAG5-13497 to S.J.M.
Received for publication 25 July 2005. Accepted for publication 2 November 2005.
The editors suggest the following Related Resources on Science sites:
In Science Magazine
TECHNICAL COMMENTS
John W. Valley, Aaron J. Cavosie, Bin Fu, William H. Peck, and Simon A. Wilde (26 May 2006) Science312 (5777), 1139a.
[DOI: 10.1126/science.1125301] |Abstract »|Full Text »|PDF »
TECHNICAL COMMENTS
T. M. Harrison, J. Blichert-Toft, W. Müller, F. Albarede, P. Holden, and S. J. Mojzsis (26 May 2006) Science312 (5777), 1139b.
[DOI: 10.1126/science.1125408] |Abstract »|Full Text »|PDF »
PERSPECTIVES
Yuri Amelin (23 December 2005) Science310 (5756), 1914.
[DOI: 10.1126/science.1121536] |Summary »|Full Text »|PDF »
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES:
Archean Accretion and Crustal Evolution of the Kalahari Craton--the Zircon Age and Hf Isotope Record of Granitic Rocks from Barberton/Swaziland to the Francistown Arc.
P. A. Cawood, A. Kroner, W. J. Collins, T. M. Kusky, W. D. Mooney, and B. F. Windley (2009)
Geological Society, London, Special Publications
318, 1-36
|Abstract »|Full Text »|PDF »
R. M. Hazen, D. Papineau, W. Bleeker, R. T. Downs, J. M. Ferry, T. J. McCoy, D. A. Sverjensky, and H. Yang (2008)
American Mineralogist
93, 1693-1720
|Abstract »|Full Text »|PDF »
Archaean to Proterozoic Crustal Evolution in the Central Zone of the Limpopo Belt (South Africa-Botswana): Constraints from Combined U-Pb and Lu-Hf Isotope Analyses of Zircon.
A. Zeh, A. Gerdes, R. Klemd, and J. M. Barton Jr (2007)
J. Petrology
48, 1605-1639
|Abstract »|Full Text »|PDF »
Temperature spectra of zircon crystallization in plutonic rocks.
Crustal recycling at modern subduction zones applied to the past--Issues of growth and preservation of continental basement crust, mantle geochemistry, and supercontinent reconstruction.
D. W. Scholl and R. von Huene (2007)
Geological Society of America Memoirs
200, 9-32
|Abstract »|Full Text »|PDF »
4 billion years ago (Ga). To test this hypothesis, we measured initial 176Hf/177Hf values of 4.01- to 4.37-Ga detrital zircons from Jack Hills, Western Australia. 
Hf (deviations of 176Hf/177Hf from bulk Earth in parts per 104) values show large positive and negative deviations from those of the bulk Earth. Negative values indicate the development of a Lu/Hf reservoir that is consistent with the formation of continental crust (Lu/Hf 
0.01), perhaps as early as 4.5 Ga. Positive 
176) of 1.867 ± 0.008 x 10–11 year–1 [(
18O, inclusion assemblages, zircon thermometry (
