an ion microprobe u-th-pb study of zircon xenocrysts … · methods imply that kimberlites emplaced...

AN ION MICROPROBE U-Th-Pb STUDY OF ZIRCON XENOCRYSTS FROMTHE LAHTOJOKI KIMBERLITE PIPE, EASTERN FINLAND

PETRI PELTONEN and IRMELI MÄNTTÄRI

PELTONEN, PETRI and MÄNTTÄRI, IRMELI 2001. An ion microprobe U-Th-Pb study of zircon xenocrysts from the Lahtojoki kimberlite pipe, easternFinland. Bulletin of the Geological Society of Finland 73, Parts 1–2, 47–58.

Eleven relatively large (diameter 1–2 mm) zircon grains extracted from theLahtojoki kimberlite pipe (Eastern Finland Kimberlite Province) have been an-alysed by the ion microprobe NORDSIM for their U- and Pb- isotopic compo-sition. The 207Pb/206Pb ages fall into two concordant age groups: 2.7 Ga and 1.8Ga. Discordant ages between these two groups are believed to result from par-tial resetting of Archaean grains in the 1.8 Ga thermal event. Since other datingmethods imply that kimberlites emplaced c. 0.6 Ga ago it is clear that the ana-lysed zircons are xenocrysts inherited from older sources and do not providethe age of the kimberlite magmatism. Their unusual size and morphology, to-gether with very low U- and Pb-concentrations, suggest, however, that these zir-con grains are not derived from typical Archaean gneisses. More likely, theyoriginate from lower crustal mafic pegmatites and from hydrous coarse-grainedveins within the uppermost lithospheric mantle. The predominance of 1.8 Gaold xenocrystic grains, together with the recovery of mafic granulite xenolithsof similar age in the kimberlites (Hölttä et al. 2000), emphasises the importanceof post-collisional lower crustal growth and reworking in central Fennoscandia.

Key words: kimberlite, pipes, zircon, xenoliths, crust, absolute age, U/Pb, Pre-cambrian, Lahtojoki, Finland

Petri Peltonen and Irmeli Mänttäri: Geological Survey of Finland, P.O. Box96, FIN-02151 Espoo, Finland

12496Bulletin73 29.1.2002, 15:3147

48 Petri Peltonen and Irmeli Mänttäri

INTRODUCTION

Zircon is an uncommon mineral in kimberlites.However, because of its physical properties it be-comes concentrated with diamonds and is thusoften available for study. The most characteristicfeatures of the kimberlitic zircons are their verylow U, Pb, Th and REE abundances, unusual largegrain size, colour and trace element compositions.In addition, many kimberlitic zircons contain man-tle minerals, such as olivine, Cr-diopside, Cr-spi-nel, magnesian ilmenite, K-richterite and phlogo-pite, as inclusions (Kresten et al. 1975, Lyakhov-ich 1996, Belousova et al. 1999). Zircon has alsobeen found as inclusions in diamonds (Meyer &Svisero 1973) and in metasomatised mantle xeno-liths (Konzett et al. 1998). These features implythat large zircon grains, i.e. megacrysts, in mostcases are xenocrysts from the mantle where theymost likely occur within coarse-grained metaso-matic veins (e.g. Kinny et al. 1989).

In many cases (Davies 1976, 1977, Davies et al.1980, Kinny et al. 1995) the U-Pb ages of thesemegacrysts are identical to the emplacement age oftheir host kimberlite determined by other methods,such as Ar-Ar from micas or U-Pb from perovskite.However, there is growing evidence that the mega-cryst series minerals are not cognate with the kim-berlite magma. This suggests that zircon megacrystscrystallised in the mantle immediately before theextraction of the kimberlite melt or, alternatively,they may represent much older grains of any ori-gin which effectively lost all their radiogenic leadwhile residing in the hot kimberlite magma.

In this paper we report single-grain U-Th-Pbisotope analyses of zircons recovered during test-scale mining operations of the Lahtojoki kimber-lite pipe and discuss their origin. The Lahtojokikimberlite pipe is one of the approximately 20kimberlitic bodies of the Eastern Finland Kimber-lite Province, which c. 600 Ma ago intruded thelate Archaean basement complex and its early Pro-terozoic metasedimentary cover rocks close to thecraton margin. This particular pipe is approximate-ly 200 m x 100 m in size and suboval in shape. Itis composed of macrocrystal heterolithic volcani-clastic kimberlite of the diatreme facies and mi-

nor hypabyssal facies kimberlite. On the basis ofpetrography and mineralogical and isotopic datait is most akin to Group I kimberlites (Griffin etal. 1995). The petrology of Eastern Finland Kim-berlite Province kimberlites and their mantle andlower crustal fragments has recently been exten-sively described by O´Brien and Tyni (1999), Pel-tonen et al. (1999), Peltonen (1999), Woodlandand Peltonen (1999), Kukkonen and Peltonen(1999), Woodland et al. (1999) and Hölttä et al.(2000).

ANALYTICAL TECHNIQUES

Eleven zircon grains were extracted by handpick-ing from the nonmagnetic heavy mineral concen-trate (d=2.9 g/cm3; 0.5–10 mm in diameter) of theLahtojoki kimberlite pipe. Zircons outside thisgrain-size range were not available for study, be-cause the sieved fraction smaller than 0.5 mm wasdumped in the process and coarse fractions larg-er than 10 mm were recycled to the crusher.

The selected zircons were mounted in epoxy,polished and coated with carbon for SEM and elec-tron microprobe study, and with gold for SIMSanalyses. Because of the different sizes of the zir-con grains, a small platform was applied beneathsmall grains so that the final cutting would inter-sect all grains exactly through their cores. The U-Pb analyses were run using the Nordic SIMS(NORDSIM) Cameca IMS 1270 at the SwedishMuseum of Natural History, Stockholm. The spot-diameter for the 4nA primary O2

- ion beam was ca.30 µm, and oxygen flooding in the sample cham-ber was used to increase the production of Pb+

ions. Four counting blocks including total twelvecycles of the Zr, Pb, Th and U species were meas-ured in each spot. The mass resolution was (M/∆M) of 5400 (10%). The raw data were calibrat-ed against a zircon standard (91500) and correct-ed for background (204.2) and age-related com-mon lead (Stacey & Kramers 1975). For a detaileddescription of the analytical procedure see White-house et al. (1999).

BSE images were obtained and inclusion phasesidentified by JEOL Scanning Electron Microscope

12496Bulletin73 29.1.2002, 15:3148

49An ion microprobe U-Th-Pb study of zircon xenocrysts...

at the Geological Survey of Finland (GTK). Zir-con compositions were determined by the wave-length dispersive technique using a Cameca Came-bax SX50 instrument at GTK. Acceleration volt-age of 25 kV or 15 kV, probe current of 50 nA or25 nA, and beam diameter of 10 mm were applied.Standards included both natural and synthetic min-erals, and data reduction was made using the PAPcorrection procedure supplied by Cameca.

ZIRCON MORPHOLOGY AND SEMOBSERVATIONS

Recovered zircons are 0.7-1.6 mm in length andthus much larger than normal crustal zircons. Onthe basis of their petrographic characteristics, theycan be divided into groups A, B and C (Table 1).Group A consists of four (grain codes 01-04) equi-dimensional non-prismatic grains having length towidth ratios close to 1:1. They are all non-zonedand enclose no inclusions (Fig. 1). The remain-ing grains are all prismatic with length to widthratios of c. 2:1. Prismatic grains can be either un-zoned (Group B: 05, 07 and 11) or zoned (GroupC: 06, 08, 09 and 10) and, with the exception ofone grain, contain inclusions of feldspars, amphi-boles and pyroxenes (Table 1). Group C grainsfrequently show typical igneous zoning or morediffuse banding in their back-scatter electron im-ages. Grain number 10 consists of three distinct

domains: the core domain being distinguished byits darker colour due to lower U-Th and Pb con-tents (Fig. 1). This domain is spatially associatedwith an area of intense fracturing and polyminer-alic rock fragment inclusions which consist ofhornblende + plagioclase + Ba-feldspar, horn-blende + Fe-Mn-Mg amphibole + plagioclase andhornblende + Fe-Mn-Mg amphibole. The bright-er, intermediate, domain shows growth zoning typ-ical of igneous zircons. Finally, the grain is man-tled by an up to 50 mm thick rim of darker (lowU-Th-Pb) zircon. Grains 06, 08 and 09 are alsozoned but their zoning is not as well defined asthat of grain 10. The core of the grain 06 is lightbrown and slighty turbid in transparent light be-ing surrounded by a 40 mm thick U-Th-Pb en-riched shell (Fig. 1). Most of the grain 06, how-ever, consists of homogeneous, pink, transparentand unzoned zircon. Central parts of the grains 08and 09 show ghost-like remnants of igneous zon-ing while large areas of the grain are composedof clear homogeneous zircon (Fig. 1).

MINERAL CHEMISTRY

The average Zr/Hf value (as determined by WDS)of the Finnish kimberlitic zircons is 45.5. Thisvalue is within the range reported for zircon mega-crysts in kimberlites (Kresten et al. 1975, Belous-ova et al. 1999). Zr/Hf is not indicative of zircon

Table 1. Summary of the petrographic and mineral chemical characteristics of the 11 xenocrystic zircon grainsrecovered from the Lahtojoki pipe, Eastern Finland Kimberlite Province.

Grain# size (mm) colour morphology zoning inclusions Zr/Hf Group

02 1.5 x 1.4 colourless clear equidimensional – – 48.6 A03 1.2 x 1.1 pink gemmy equidimensional – – 52.9 A04 1.1 x 1.0 pink clear equidimensional – – 34.9 A05 1.2 x 0.8 light brown – turbid – clear prismatic – labradorite, hypersthene 64.8 B

pink07 1.4 x 0.6 pink gemmy prismatic – andesine 38.8 B11 0.7 x 0.3 colourless clear prismatic – potassium feldspar, talc 41.7 B06 1.1 x 0.5 light brown – turbid – clear prismatic yes aegerine-augite, hypersthe- 47.6 C

pink ne, hornblende, actinolite08 1.1 x 0.6 pink clear prismatic yes – 39.4 C09 1.0 x 0.6 pink clear prismatic yes hornblende, hypersthene 43.3 C10 1.2 x 0.7 deep red turbid prismatic yes hornblende, Fe-Mg 37.5 C

amphibole, Ba-feldspar,andesine

12496Bulletin73 29.1.2002, 15:3149


Fig. 1. SEM images of selected zircon grains recovered from the Lahtojoki kimberlite pipe with points of NORD-SIM analyses and corresponding ages (Ma). All the grains that show intragrain heterogeneity with respect tochemical composition or ages, as well as representative examples of equidimensional Group A grains (03, 04)are shown.

12496Bulletin73 29.1.2002, 15:3150


Fig. 2. U-Th relationship of the xenocrystic zircons from the Lahtojoki pipe (data from Table 2) and zirconsextracted from two lower crustal xenoliths recovered from the same pipe (data from Hölttä et al. 2000). Fieldsfor zircon megacrysts from kimberlites (Ahrens et al. 1967, Krasnobayev 1980, Kinny et al. 1989, Belousova1999), metasomatised mantle xenoliths (Konzett et al. 1998) and granitic rocks (Ahrens et al. 1967) are indi-cated for comparison.

origin because zircon suites from felsic and ma-fic igneous rocks may have equal values (Heamanet al. 1990). The total ranges of U and Th withinthe analysed zircons are 8-588 ppm and 3-128ppm, respectively. Approximately half of thegrains – including both equidimensional and pris-matic – are characterised by anomalously low U(< 30 ppm) and Th (< 10 ppm) abundances simi-lar to those in zircon megacrysts from South Af-rican and Yakutian kimberlites. Thorium and Uabundances and Th/U alone are not, however, ad-equate parametes to determine zircon origin. Thisis because zircons from metasomatised mantlerocks may have crustal-like U-Th systematics, and

zircons from mafic lower crustal granulites maybe identical with zircon megacrysts of mantle or-igin (Fig. 2).

U-Pb DATING RESULTS

The analysed zircon grains yielded a large rangeof ages which, however, tend to correlate with thegeneral habit of the grains (Table 2, Fig. 3). Themost crustal-like grain, i.e. the dark-brown, pris-matic and strongly growth-zoned grain number 10,yielded the oldest ages. The U- and Pb-rich inter-mediate (10A) and outer zones (10B) record Ar-

12496Bulletin73 29.1.2002, 15:3151


chaean 207Pb/206Pb ages, while the core of the samegrain yields a younger age of c. 2.4 Ga. Grains06 and 08 also yield significantly older ages thanmost of the grains. These two grains are similar-ly zoned, but their zoning is less prominent com-pared to grain 10 giving an impression of partialdestruction of an original more profound growth-zoning. They yield a broad range of somewhatdiscordant ages between c. 2.1 and 2.6 Ga. Theyoungest ages in both grains are recorded by theirU- and Pb-depleted marginal zones.

Most of the grains record 207Pb/206Pb ages closeto 1.8 Ga. The weighted average of concordantzircons is 1789±24 Ma indicating a major thermalepisode at that time (Fig. 4). Importantly, thisyoung zircon population includes all equidimen-sional inclusion-free grains of Group A, all pris-matic unzoned grains of Group B and one pris-

matic zoned grain (09) with ghost-like remnantsof growth-zoning.

DISCUSSION

Age of the kimberlites

The first attempts to date the eastern Finland kim-berlites were K-Ar determinations of two wholerock samples from pipes number 1 and 2 whichgave ages of 430 Ma and 560 Ma, respectively(Tyni 1997). However, because pure mineral sep-arates were not analysed these ages were regard-ed only as rough estimate of the true age. Pelto-nen et al. (1999) provided an independent estimateof the emplacement age: garnet and clinopyrox-ene separated from four garnet-peridotite xenolithsfrom the Lahtojoki pipe yielded two-mineral Sm-

Fig. 3. U-Pb concordia diagram for the zircon xenocrysts from the Lahtojoki kimberlite pipe. Inset shows the 1.8Ga group analyses in more detail.

12496Bulletin73 29.1.2002, 15:3152


Tab

le 2

. U

-Th-

Pb

NO

RD

SIM

dat

a fo

r zi

rcon

xen

ocry

sts

from

the

Lah

tojo

ki k

imbe

rlit

e pi

pe.

deri

ved

ages

cali

brat

ed r

atio

sel

emen

tal

data

Zir

con

#Z

irco

n20

7 Pb

±σ

207 P

b±σ

206 P

b±σ

207 P

b±σ

207 P

b±σ

206 P

b±σ

Rho

Dis

c[U

][P

b][T

h]T

h/U

Th/

U20

6 Pb

spot

#gr

oup*

206 P

b23

5 U23

8 U20

6 Pb

%23

5 U%

238 U

%%

ppm

ppm

ppm

calc

.m

eas.

204 P

b

n158

-01a

A17

2941

1754

5317

7694

0.10

592.

234.

629

6.38

0.31

716.

070.

9525

1015

0.60

0.61

6.68

E+

03n1

58-0

2aA

1827

6417

3810

316

6617

30.

1117

3.54

4.54

112

.29

0.29

4811

.77

0.96

83

30.

340.

387.

02E

+07

n158

-03a

A18

0138

1813

6918

2312

50.

1101

2.10

4.96

38.

130.

3269

7.86

0.97

156

40.

260.

271.

84E

+08

n158

-04a

A17

4617

1715

3016

8952

0.10

680.

944.

412

3.60

0.29

953.

480.

9793

3662

0.61

0.67

3.88

E+

04n1

58-0

4bA

1787

1717

5433

1727

580.

1093

0.94

4.62

93.

910.

3073

3.80

0.97

102

4063

0.58

0.62

9.88

E+

04n1

58-0

5aB

1822

4617

5367

1695

114

0.11

142.

554.

619

8.03

0.30

087.

650.

9515

67

0.43

0.47

1.83

E+

04n1

58-0

6aC

2377

2322

5853

2129

104

0.15

281.

388.

243

5.89

0.39

135.

730.

972.

252

2953

0.88

1.03

5.78

E+

08n1

58-0

6bC

2131

1020

9021

2048

390.

1325

0.59

6.83

22.

290.

3740

2.22

0.97

3819

280.

700.

751.

42E

+04

n158

-06c

C24

876

2427

2123

5644

0.16

300.

379.

920

2.25

0.44

132.

210.

991.

945

825

010

40.

260.

238.

47E

+02

n158

-06d

C23

478

2276

2121

9741

0.15

010.

478.

403

2.27

0.40

612.

220.

983.

034

517

612

80.

350.

372.

50E

+03

n158

-07a

B17

5830

1779

5217

9894

0.10

751.

664.

769

6.17

0.32

175.

980.

9728

118

0.24

0.28

1.25

E+

04n1

58-0

8aC

2501

2524

0142

2284

840.

1644

1.46

9.63

84.

610.

4252

4.37

0.95

2.5

4224

340.

710.

825.

32E

+08

n158

-08b

C22

4310

2222

2121

9941

0.14

130.

587.

920

2.29

0.40

662.

220.

9787

4973

0.81

0.84

5.92

E+

04n1

58-0

8cC

2570

1325

6122

2549

470.

1713

0.77

11.4

572.

340.

4851

2.21

0.95

7050

711.

061.

021.

39E

+04

n158

-09a

C17

8941

1730

4916

8281

0.10

942.

234.

497

5.93

0.29

815.

500.

9327

1012

0.39

0.44

3.23

E+

08n1

58-0

9bC

1847

4318

3529

1824

360.

1129

2.40

5.09

23.

320.

3271

2.28

0.69

2410

130.

470.

536.

15E

+03

n158

-09c

C18

0015

1857

2019

0937

0.11

010.

815.

229

2.37

0.34

462.

220.

94–0

.815

561

210.

140.

134.

01E

+04

n158

-10a

C26

716

2665

2626

5859

0.18

200.

3412

.804

2.73

0.51

022.

710.

9958

836

789

0.15

0.15

1.02

E+

10n1

58-1

0bC

2662

1026

3826

2606

590.

1810

0.59

12.4

322.

810.

4982

2.75

0.98

178

111

460.

240.

261.

19E

+05

n158

-10c

C24

0212

2366

2823

2457

0.15

500.

719.

278

3.03

0.43

402.

940.

9794

5910

31.

051.

107.

81E

+04

n158

-11a

B18

6327

1850

4418

3980

0.11

391.

495.

184

5.22

0.33

015.

000.

9632

1528

0.85

0.88

3.51

E+

08

Deg

ree

of d

isc o

rda n

c e i

s c a

lcul

a te d

at

the

c los

e st

2 si

gma

lim

it. A

ll o

the r

err

ors

a re

a t 1

sig

ma

leve

l.*

Zir

c on

mor

phol

ogy

grou

ps:

see

text

for

de t

a ile

d e x

pla n

a tio

n.

12496Bulletin73 29.1.2002, 15:3153


Nd isochron ages between 525±10 and 607±20Ma. Because chemical composition of garnet andclinopyroxene suggests that they were in or closeto chemical equilibrium with respect to REE whenthe xenoliths were detached from the lithospher-ic mantle, the youngest isochron age of 525 Maprovides a close estimate for the emplacement ofthe kimberlite. The most appropriate method todate kimberlite magmatism is, however, by U-Pbmethod on perovskite, that is a mineral whichcrystallises from the kimberlite melt itself. Sucha work is in progress and the preliminary resultsindicate that the Finnish kimberlites were em-placed c. 600 Ma ago (Hugh O' Brien, personalcommunication 2000).

In the light of the above review, it is evidentthat the 207Pb/206Pb ages recorded by the zirconcrystals recovered from the Lahtojoki kimberlitepipe are much older than the emplacement age ofthe eastern Finland kimberlites. Clearly, the stud-ied zircon grains have not crystallised from theirhost kimberlite melt but occur as exotic xenocrystsderived from older sources.

Origin of the zircons in the Lahtojoki pipe

On the basis of their morphology, zoning and in-clusion suite, the studied grains were divided into

three subgroups A, B and C (Table 1). Group Cconsists of prismatic grains which exhibit zoningon the BSE images: they record a variety of agesbetween 1790–2670 Ma and exhibit considerableintragrain heterogeneity. Results from the grain 10are particularly important: low U and Pb contentsof the core, together with its highly fractured ap-pearance and intimate association with large sili-cate inclusions, suggest that the younger age of thecore results from U- and Pb-loss (accompanied byTh-gain) during alteration. Grains 06 and 08 alsoshow considerable intragrain heterogeneity ofages. Because the younging of the ages is accom-panied by weakening of the zoning and by theappearance of large domains of non-zoned zirconin these grains, we believe that the discordant agesbetween the 2.7 Ga and 1.8 Ga age groups do nothave any geological significance but represent old-er grains, probably Archaean, which have experi-enced partial lead-loss during a younger episode.Grain 09, in turn, is similar to the old Group Czircons in showing ghost-like remnants of igne-ous zoning but yields only young ages close to 1.8Ga within all domains. These features, togetherwith similar inclusions as in the old zircons, sug-gest that grain 09 represents a fairly advancedstage of resetting of old grains. The inclusion suiteof Group C zircons, dominated by pyroxenes andamphiboles (Table 1), differs from inclusions iden-tified in zircon megacrysts of kimberlites and zir-cons of metasomatised mantle xenoliths whichhost primary and metasomatic mantle mineralssuch as olivine, Cr-diopside, Cr-spinel, magnesianilmenite, K-richterite, rutile, apatite, phlogopite,Ni-Fe sulfides and diamond (Kresten et al. 1975,Lyakhovich 1996, Konzett et al. 1998). Most like-ly, all Group C zircons were derived from deep-seated Archaean rocks which have experiencedvariable U-Pb disturbance and recrystallisationprocesses in a thermal event at c. 1.8 Ga. The highU and Th contents and plagioclase and Ba-feld-spar bearing inclusion suite of grain 10 suggestgranitic protolith for that grain. The remainingGroup C zircons contain only pyroxene and am-phibole inclusions and are believed to originateeither from mafic lower crustal pegmatites or,more likely, from hydrous pyroxenitic mantle

Fig. 4. Concordant 207Pb/206Pb ages of the 1.8 Ga groupzircon grains from the Lahtojoki kimberlite pipe with2σ error bars.

12496Bulletin73 29.1.2002, 15:3154


dykes similar to those common in orogenic lher-zolite massifs (Bodinier et al. 1987, Peltonen etal. 1998) and upper mantle xenoliths erupted withalkali basalts (e.g. Wilshire et al. 1980).

Group B includes grains 05, 07 and 11 (Table1). These grains resemble those of Group C in be-ing prismatic and in containing some silicate inclu-sions. However, they differ from the Group Cgrains because they do not show any kind of intra-grain heterogeneity with respect to chemical com-position or age. Two alternative origins can be pos-tulated for the Group B grains. First, they may rep-resent new zircons that crystallised from magmasin the 1.8 Ga event or, alternatively, old grains sim-ilar to those of Group C but which have been com-pletely reset/recrystallised by protracted residenceat a high ambient (lower crustal?) temperature.Their inclusion suite consists of both plagioclaseand potassium feldspar together with orthopyrox-ene (Table 1), a feature which suggests a pyrox-ene granite protolith for the Group B grains.

Group A grains are distinct from all other grains(Table 1). They are non-prismatic, equidimension-al and inclusion-free grains which record agesclose to 1.8 Ga. Because of their morphology andabsence of inclusions we believe that they crys-tallised in an environment different from that ofGroup B and Group C zircons. Since there is noclear process by which Group B or C zircons couldbecome inclusion-free during recrystallisation,Group A grains most likely represent new grainscrystallised from magmas c. 1.8 Ga ago. Their lowU and Pb abundances and their large grain size,which is more than a magnitude larger than thatof zircons in lower crustal xenoliths (Hölttä et al.2000), suggest that they most likely were derivedfrom mafic to ultramafic lower crustal pegmatitesor coarse-grained mantle veins.

Implications

Although we failed to determine the age of thekimberlites by U-Pb analyses of zircons the studyturns out to have several implications. First, sev-eral of the grains have been reset in the 1.8 Gaevent but show no evidence for even minor dis-turbance caused by their residence in kimberlite

melt during eruption 600 Ma ago. We believe thatthis is due to the very low U concentrations, slowdiffusion of Pb, U and Th in zircon, and the highclosure temperature (of the order of 1000°C) forsuch large (1000 µm) grains at all relevant cool-ing rates (Lee et al. 1997) and due to the very shortresidence-time of the grains in their host kimber-lite. A similar conclusion was reached e.g. by Kin-ny et al. (1989) who reported the presence of Ar-chaean zircon grains in the Permian Jwaneng DK2kimberlite (Botswana). Whether these ArchaeanJwaneng grains were indeed derived from pegma-toidal veins in the uppermost lithospheric mantle,as argued by Kinny et al. (1989), or representxenocrysts from crustal sources, as their trace el-ement patterns would suggest (Belousova et al.1999), they do indicate that interaction of zirconwith kimberlite melt is incapable to reset zircon.

The predominance of c. 1.8 Ga old concordantzircons (Table 2, Fig. 4) in the Lahtojoki pipe isintriguing. This age clearly postdates the majorphases of the Svecofennian orogeny, and rocks ofthis age do not have surface expression close tothe kimberlites. However, similar ages close to 1.8Ga were obtained by Hölttä et al. (2000), whoextracted zircons from two lower crustal maficgranulite xenoliths recovered from the same kim-berlite pipe. One of their samples (L-94) gave con-cordant ages close to 1.8 Ga, while the other sam-ple (7HH) yielded a heterogeneous zircon popu-lation with the youngest ages close to 1.8 Ga. Thefact that both xenocrystic large zircons from kim-berlites and small zircons extracted from maficlower crustal xenoliths record similar post-colli-sional ages implies that such a thermal event hasbeen important in the lower crustal/uppermostmantle environment. Probably, a substantial partof the present lower crust in eastern Finland iscomposed of mafic granulites of magmatic origin(Hölttä et al. 2000). The intrusion of voluminousmafic melts at 1.8 Ga into the pre-existing lowercrust must have had a major thermal imprint re-sulting in local production of felsic magmas.

Elsewhere in Finland several post-collisionalgranitoids falling within this age group are indeedknown. These include the extensive Central-Lap-land granitoid complex (Lauerma 1982) and Nat-

12496Bulletin73 29.1.2002, 15:3155


tanen (Patchett et al. 1981) and Hetta (Meriläinen1976) granites in northern Finland. Interestingly,this is also the approximate age of the epigeneticAu-mineralisations in northern Finland as record-ed by monazite from the Saattopora Au-deposit andby the Pb-Pb ages of several other Au-occurrenc-es over a wide area (Mänttäri 1995). Several felsicintrusions in southern and central Finland, such asPuruvesi (Nykänen 1983), Ruokolahti (Nykänen1988), Mikkeli (Patchett & Kouvo 1986), Taki-ankangas (Kärki 1995), Anttola (Korsman 1984),Parkkila (Simonen 1982) and Hanko (Huhma1986), also belong to this age group. Furthermore,Eklund et al. (1998) have outlined a belt consist-ing of at least fourteen 1.8 Ga old post-collisionalshoshonitic intrusions in southern Finland and Rus-sian Karelia. These intrusions range from ultrama-fic, calc-alkaline potassium lamprophyres to pera-luminous granites and are believed to have beenderived from a lithospheric mantle source affectedby carbonate metasomatism. Eklund et al. (1998)concluded that the regional distribution of such in-trusions may indicate that large volumes of 1.8 Gajuvenile material resides in unexposed parts of thecrust. Direct evidence for the presence of lowercrust of this age is now available from eastern Fin-land (this study, Hölttä et al. 2000) and from theBelomorian Mobile Belt in Kola Peninsula(Downes et al. in prep).

CONCLUSIONS

Large zircon grains recovered from the Lahtojokipipe are not cogenetic with their kimberlite hostbut occur as xenocrysts derived from older sourcerocks. Prismatic grains are derived from Archae-an lower crustal or uppermost mantle sources. Inmost cases these grains bear evidence for partialU- and Pb-loss due to the post-collisional 1.8 Gathermal event. Equidimensional grains, in turn,originate from 1.8 Ga old lower crustal pegmatitesor hydrous veins in the uppermost mantle. Thiswork, together with recent data from mafic lowercrustal granulite xenoliths (Hölttä et al. 2000),suggests that the 1.8 Ga old post-collisional eventwas a major episode of crustal growth and rework-

ing in central Fennoscandian Shield.

ACKNOWLEDGEMENTS. Matti Tyni is thankedfor providing the zircons for study and the NORD-SIM staff at Stockholm for keeping the wheels run-ning. This work was supported by grant 40553from the Academy of Finland. We thank W.L. Grif-fin, E.A. Belousova, O. Eklund and editor Y.Kähkönen for constructive comments that led tothe improvement of the manuscript. This is NORD-SIM contribution number 48.

REFERENCES

Ahrens, L.H., Cherry, R.D. & Erlank, A.J. 1967. Obser-vations on the Th-U relationship in zircons from gra-nitic rocks and kimberlites. Geochimica et Cosmochim-ica Acta 31, 2379–2387.

Belousova, E.A., Griffin, W.L. & O'Reilly, S.Y. 1999.Cathodoluminescense and geochemical properties ofkimberlitic and lamproitic zircons. In: Gurney et al.(eds.) Proceedings of the 7th International KimberliteConference, Cape Town, South Africa, Vol. 1, 23–29.

Bodinier, J-L., Fabriès, J., Lorand, J-P., Dostal, J. & Du-puy, C. 1987. Geochemistry of amphibole pyroxeniteveins from the Lhertz and Freychinède ultramafic bod-ies Arièga, French Pyrenees. Bulletin de Minéralogie110, 345-358.

Davies, G.L., Krogh, T.E. & Erlank, A.J. 1976. The agesof zircon from kimberlites from South Africa. Carne-gie Institution of Washington Year Book 75, 821–824.

Davies, G.L. 1977. The ages and uranium contents of zir-cons from kimberlites and associated rocks. CarnegieInstitution of Washington Year Book 76, 631–635.

Davies, G.L., Sobolev, N.V. & Khar´kiv, A.D. 1980. Newdata on the age of Yakutian kimberlites obtained by theuranium-lead method on zircons. Doklady AkademiiNauk SSSR 254, 175–179.

Eklund, O., Konopelko, D., Rutanen, H., Frödjö, S. & She-banov, A.D. 1998. 1.8 Ga Svecofennian post-collision-al shoshonitic magmatism in the Fennoscandian shield.Lithos 45, 87–108.

Griffin, B.J., Rissanen, J., Pooley, G.D., Lee, D.C., Mac-Donald, I. & Kinny, P.D. 1995. A new diamondiferouseclogite-bearing kimberlitic occurrence from Finland. In:Sixth International Kimberlite Conference, Novosibirsk,Russia, Extended Abstracts, 198–200.

Heaman, L.M., Bowins, R. & Crocket, J. 1990. The chem-ical composition of igneous zircon suites: Implicationsfor geochemical tracer studies. Geochimica et Cosmo-chimica Acta 54, 1597–1607.

Hölttä, P., Huhma, H., Mänttäri, I., Peltonen, P. & Juhano-ja, J. 2000. Petrology and geochemistry of mafic gran-ulite xenoliths from the Lahtojoki kimberlite pipe, east-

12496Bulletin73 29.1.2002, 15:3156


ern Finland. Lithos 51, 109–133.Huhma, H. 1986. Sm-Nd, U-Pb and Pb-Pb isotopic evi-

dence for the origin of the Early Proterozoic Svecokare-lian crust in Finland. Geological Survey of Finland, Bul-letin 337. 48 p

Kärki, A. 1995. Palaeoproterozoic shear tectonics in thecentral Fennoscandian Shield, Finland. Res Terrae, Ser.A 10. 98 p.

Kinny, P.D., Compston, W., Bristow, J.W. & Williams, I.S.1989. Archaean mantle xenocrysts in a Permian kimber-lite: two generations of kimberlitic zircon in JwanengDK2, southern Botswana. In: Ross, J. (ed.) Kimberlitesand related rocks. Geological Society of Australia, Spe-cial Publication 14, 833–842.

Kinny, P.D., Griffin, B.J. & Brakhfogel, F.F. 1995.Shrimp U-Pb ages of perovskite and zircon from Yaku-tian Kimberlites. In: Sixth International KimberliteConference, Novosibirsk, Russia, Extended Abstracts,275–276.

Konzett, J., Armstrong, R.A., Sweeney, R.J. & Compston,W. 1998. The timing of MARID metasomatism in theKaapvaal mantle: An ion probe study of zircons fromthe MARID xenoliths. Earth and Planetary Science Let-ters 160, 133–145.

Korsman, K., Hölttä, P., Hautala, T. & Wasenius, P. 1984.Metamorphism as an indicator of evolution and struc-ture of the crust in Eastern Finland. Geological Surveyof Finland, Bulletin 328. 40 p.

Krasnobayev, A.A. 1979. Mineralogical-geochemical fea-tures of zircons from kimberlites and problems of theirorigin. International Geology Review 22, 1199–1209.

Kresten, P., Fels, P. & Berggren, G. 1975. Kimberlitic zir-cons - a possible aid in prospecting for kimberlites. Min-eralium Deposita 10, 47–56.

Kukkonen, I. & Peltonen, P. 1999. Xenolith-controlled ge-otherm for the central Fennoscandian Shield – implica-tions for the lithosphere-asthenosphere relation. Tectono-physics 304, 301–315

Lauerma, R. 1982. On the ages of some granitoid and schistcomplexes in northern Finland. Bulletin of the Geolog-ical Society of Finland 54, 85–100.

Lee, J.K.W., Williams, I.S. & Ellis, D.J. 1997. Pb, U andTh diffusion in natural zircons. Nature 390, 159–161.

Lyakhovich, V.V. 1996. Zircons in diamond-bearing rocks.Transactions (Doklady) of the USSR Academy of Sci-ences, Earth Science Section 347, 179–199.

Mänttäri, I. 1995. Lead isotope characteristics of epigeneticgold mineralization in the Palaeoproterozoic Laplandgreenstone belt, northern Finland. Geological Survey ofFinland, Bulletin 381. 70 p.

Meriläinen, K. 1976. The granulite complex and adjacentrocks in Lapland, northern Finland. Geological Surveyof Finland, Bulletin 281. 129 p.

Meyer, H.O.A. & Svisero, D.P. 1973. Mineral inclusionsin Brasilian diamonds. Physics and Chemistry of theEarth 9, 785–795.

Nykänen, O. 1983. Punkaharjun ja Parikkalan kartta-aluei-den kallioperä. Summary: Pre-Quaternary rocks of thePunkaharju and Parikkala map-sheet areas. Geological

map of Finland 1 : 100 000, Explanation to the mapsof Pre-Quaternary rocks, Sheets 4124+4142, 4123+4114.Geological Survey of Finland. 81 p.

Nykänen, O. 1988. Virmutjoen kartta-alueen kallioperä.Summary: Pre-Quaternary rocks of the Virmutjoki map-sheet area. Geological map of Finland 1 : 100 000, Ex-planation to the map of Pre-Quaternary rocks, Sheet4121. Geological Survey of Finland. 64 p.

O’Brien H. & Tyni M. 1999. Mineralogy and geochemis-try of kimberlites and related rocks from Finland. In:Gurney et al. (eds.) Proceedings of the 7th InternationalKimberlite Conference, Cape Town, South Africa,Vol. 2, 625–636.

Patchett, P.J., Kouvo, O., Hedge, C.E. & Tatsumoto, M.1981. Evolution of continental crust and mantle heter-ogeneity: evidence from Hf isotopes. Contributions toMineralogy and Petrology 78, 279–297.

Patchett, J. & Kouvo, O. 1986. Origin of continental crustof 1.9-1.7 Ga age: Nd isotopes and U-Pb zircon ages inthe Svecokarelian terrain of south Finland. Contributionsto Mineralogy and Petrology 92, 1–12.

Peltonen, P., Kontinen, A. & Huhma, H. 1998. Petrogene-sis of the mantle sequence of the Jormua ophiolite, Fin-land: melt migration in the upper mantle during Palae-oproterozoic continental break-up. Journal of Petrolo-gy 39, 297–329.

Peltonen, P. 1999. Silicification of garnet peridotite xeno-liths from the Lahtojoki kimberlite pipe, Finland. In:Gurney et al. (eds.) Proceedings of the 7th InternationalKimberlite Conference, Cape Town, South Africa,Vol. 2, 659–663.

Peltonen, P., Huhma, H., Tyni, M. & Shimizu, N. 1999.Garnet-peridotite xenoliths from kimberlites of Finland:nature of the continental mantle at Archaean craton –Proterozoic mobile belt transition. In: Gurney et al. (eds.)Proceedings of the 7th International Kimberlite Confer-ence, Cape Town, South Africa, Vol. 2, 664–676.

Simonen, A. 1982. Mäntyharjun ja Mikkelin kartta-aluei-den kallioperä. Summary: Pre-Quaternary rocks of theMäntyharju and Mikkeli map-sheet areas. Geologicalmap of Finland 1 : 100 000, Explanation to the mapsof Pre-Quaternary rocks, Sheets 3123, 4142. Geologi-cal Survey of Finland. 36 p.

Stacey, J.S. & Kramers, J.D. 1975. Approximation of theterrestrial lead isotope evolution by a two-stage model.Earth and Planetary Science Letters 26, 207–221.

Tyni, M. 1997. Diamond prospecting in Finland – a review.In: Papunen, H. (ed.) Mineral Deposits: Research andExploration, Where do They Meet? Proceedings of the4th biennal SGA Meeting, 789–791.

Whitehouse, M.J., Kamber, B. & Moorbath, S. 1999. Agesignificance of U-Th-Pb zircon data from early Archaeanrocks of west Greenland – a reassessment based on com-bined ion-microprobe and imaging studies. ChemicalGeology 160, 201–224.

Wilshire, H.G., Pike, J.E.N., Meyer, C.E. & Schwarzman,E.C. 1980. Amphibole-rich veins in lherzolite xenoliths,Dish Hill and Deadman Lake, California. AmericanJournal of Science 280-A, 576–593.

12496Bulletin73 29.1.2002, 15:3157


Woodland, A.B. & Peltonen, P. 1999. Ferric/ferrous ironcontents of garnet and clinopyroxene and calculated ox-ygen fugacities of peridotite xenoliths from the EasternFinland Kimberlite Province. In: Gurney et al. (eds.) Pro-ceedings of the 7th International Kimberlite Conference,Cape Town, South Africa, Vol. 2, 904–911.

Woodland, A.B., Seitz, H.-M. & Peltonen, P. 1999. Lightand transition element compositions of garnet andclinopyroxene from eclogite xenoliths from the EasternFinland Kimberlite Province. In: EUG 10, EuropeanUnion of Geosciences, 28.3.–1.4. 1999, Strasbourg(France), Abstract Volume, p. 821.

12496Bulletin73 29.1.2002, 15:3158

an ion microprobe u-th-pb study of zircon xenocrysts … · methods imply that kimberlites emplaced...

Documents