Combined	
  Chemical,	
  Isotopic	
  and	
  
Geochronological	
  Analysis	
  of	
  Accessory	
  
Minerals	
  at	
  High	
  Spa<al	
  Resolu<on	
  –	
  
Insights	
  into	
  Mantle	
  Sources	
  and	
  
Melt	
  Crystalliza<on	
  History	
  	
  
	
  
ANTONIO	
  SIMONETTI	
  
	
  
Dept.	
  Civil	
  &	
  Environmental	
  Engineering	
  &	
  Earth	
  Sciences	
  
University	
  of	
  Notre	
  Dame	
  
Notre	
  Dame,	
  Indiana	
  46556,	
  USA	
  
Introduc<on	
  
•  LA-­‐(MC)-­‐ICP-­‐MS	
  techniques	
  now	
  make	
  it	
  possible	
  to	
  obtain	
  
reliable	
  radiogenic	
  (e.g.,	
  Sr,	
  Nd,	
  Pb,	
  Hf)	
  isotope	
  signatures	
  
and	
  U-­‐Pb	
  ages	
  on	
  individual	
  mineral	
  grains	
  
•  Suitable	
  accessory	
  minerals	
  (e.g.,	
  apa#te,	
  zircon,	
  #tanite	
  
perovskite)	
  may	
  crystallize	
  early	
  and	
  during	
  a	
  significant	
  
period	
  of	
  melt	
  crystallizaPon	
  
•  Combined	
  major	
  &	
  trace	
  element	
  data,	
  isotope	
  ra<os,	
  and	
  
ages	
  yield	
  valuable	
  insights	
  in	
  relaPon	
  to	
  mantle	
  source	
  
region(s)	
  and	
  melt	
  differenPaPon	
  processes	
  
Why	
  accessory	
  minerals	
  in	
  carbona<tes?	
  
•  Carbona<tes	
  –	
  mantle-­‐derived,	
  igneous	
  rocks	
  (intrusive	
  or	
  
extrusive)	
  that	
  contain	
  >50%	
  modal	
  carbonate	
  minerals	
  (e.g.,	
  
calcite,	
  dolomite)	
  
Why	
  accessory	
  minerals	
  in	
  carbona<tes?	
  
	
  
•  Although	
  scarce	
  in	
  number	
  (527	
  occurrences),	
  these	
  occur	
  on	
  
all	
  con#nents	
  and	
  span	
  in	
  age	
  from	
  ~3.0	
  Ga	
  to	
  present-­‐day	
  
(Oldoinyo	
  Lengai,	
  Tanzania)	
  
•  Low	
  viscosity	
  and	
  unique	
  geochemistry	
  with	
  extremely	
  high	
  
abundances	
  of	
  Sr,	
  Nd,	
  REEs	
  buffer	
  their	
  inherited	
  mantle	
  
isotope	
  signature	
  from	
  possible	
  contaminaPon	
  during	
  ascent	
  
and	
  emplacement	
  
	
  
Woolley & Kjargaard (2008)
527 occurrences worldwide
Why	
  accessory	
  minerals	
  in	
  carbona<tes?	
  
•  Important	
  for	
  understanding	
  mantle	
  metasomaPc	
  
processes	
  
•  Calcite	
  and	
  a	
  variety	
  of	
  accessory	
  minerals	
  (e.g.,	
  apa#te,	
  
perovskite,	
  niocalite)	
  can	
  be	
  examined	
  for	
  their	
  chemical	
  &	
  
isotopic	
  composiPons	
  and	
  U-­‐Pb	
  ages	
  at	
  high	
  spaPal	
  
resoluPon	
  
•  Apa<te:	
  Ca5(PO4)3F	
  
•  Perovskite:	
  CaTiO3	
  
•  Niocalite:(Na0.34Fe0.06Mn0.19Mg0.09Ca13.40REE0.15Ti0.02)total=14.25Nb2.12Ta0.06(Si2O7)4O7.65F2.44	
  
Chen	
  et	
  al.	
  (in	
  press,	
  Can.	
  Mineral.)	
  
	
  	
  
Why investigate apatite?
•  Apatite is also employed in low-temperature
thermochronology studies in relation to apatite fission
track and apatite (U–Th)/He thermochronometers
•  Apatite has also been investigated in high temperature
thermochronology studies - U–Pb apatite system has a
closure temperature of ca. 450–550 °C (Chamberlain and
Bowring, 2000; Schoene and Bowring, 2007).
In-situ U-Pb dating of accessory
minerals by LA-(MC)-ICP-MS method
•  The advent of U-Pb geochronology by LA-ICPMS offers lowcost,
rapid data acquisition and sample throughput compared
to the ID-TIMS or ion microprobe U–Pb methods (e.g., Košler
and Sylvester, 2003).
•  However, few studies demonstrated the applicability of the LAICPMS
to U–Pb apatite, perovskite, niocalite geochronometry.
•  The methodology for apatite was recently validated by the
detailed study by the group at Memorial University (Chew et
al., 2011, Chem. Geol.) by dating apatite standards of known
crystallization age
U-Th-Pb apatite, perovskite, niocalite
dating by LA-ICP-MS - advantages
•  Accessory minerals that contain both
abundant U and Th, hence there are
several independent geochronometers
that can be used simultaneously, and
serve to cross-validate results
U-Th-Pb geochronometers
EHePbU +++→ −
β684
2
206
82
238
92
EHePbU +++→ −
β474
2
207
82
235
92
EHePbTh +++→ −
β464
2
208
82
232
90
However, in almost all instances, 207Pb/235U ratios (ages)
are not measured but rather calculated.
207Pb/235U = 207Pb/206Pbmeas x 206Pb/238Umeas x 238U/235U(137.88)
U-Th-Pb dating by LA-ICP-MS - problems
•  However, precise and accurate U–Pb and Th–Pb
apatite age determinations are commonly
hindered by low U, Th and Pb concentrations
and high common Pb contents, which requires
a common Pb correction.
•  Since apatite, perovskite, niocalite can
accommodate a significant amount of initial Pb
into its crystal structure, this places limitations on
the accuracy and precision of age
determinations.
U-Th-Pb dating by LA-ICP-MS - problems
Thus, strategies to curtail or address the
common Pb issue include:
– 1) Define concordia or isochron plots on a
suite of cogenetic apatite grains with a large
spread in common Pb/radiogenic Pb ratio; or
–  2) to undertake a Pb correction based on an
appropriate choice of initial Pb isotopic
composition.
U-Th-Pb dating by LA-ICP-MS –
possible solutions
•  A variety of methods do not require an estimate of the
initial Pb isotope composition
•  These require several analyses of a suite of cogenetic
apatite grains with a significant variation in common Pb/
radiogenic Pb ratios, which in turn define a wellcontrained
linear array on a Concordia diagram or
isochron.
•  E.g., Total-Pb/U isochron, a three-dimensional 238U/
206Pb vs. 207Pb/206Pb vs. 204Pb/206Pb plot (Ludwig, 1998)
U-Th-Pb dating by LA-ICP-MS –
possible solutions
•  An alternative approach often employed in the LA-ICP-MS U–
Pb dating of high common Pb phases, such as perovskite (Cox
and Wilton 2006; Simonetti et al. 2008) and titanite (Simonetti
et al. 2006; 2008) involves projecting an intercept through the
uncorrected data on a Tera–Wasserburg Concordia to
determine the common Pb-component (y-intercept) on the
207Pb/ 206Pb axis.
•  The 238U/206Pb age can then be calculated as either a lower
intercept age on the 238U/206Pb axis (x-intercept) or as a
weighted average of 207Pb-corrected ages (see below) using
the Concordia 207Pb/206Pb intercept as an estimate of the initial
Pb isotopic composition.
•  This approach also assumes that the U–Pb* data are
concordant and equivalent.
Common Pb correction schemes-
207Pb method
f = proportion of common Pb
From Simonetti et al. (2008)
Common Pb correction schemes-
207Pb method
From Simonetti et al. (2008)
Common Pb correction schemescorrection
for initial Pbcommon
•  Estimates of initial Pb isotopic
compositions are typically derived from Pb
evolution models (e.g., Stacey and
Kramers, 1975) or by analyzing a low-U
co-magmatic phase (e.g., K-feldspar or
plagioclase), which exhibits negligible
ingrowth of radiogenic Pb.
Common Pb correction schemes-
204Pb correction method
•  This method relies on the measurement of the
low abundance non-radiogenic 204Pb isotope.
•  Measured Pb isotopic signals are corrected
using the assumed 206Pb/204Pb, 207Pb/204Pb and
208Pb/204Pb ratios of the common Pb to extract
net signal intensities of the radiogenic daughter
206Pb*, 207Pb* and 208Pb* isotopes.
Common Pb correction schemes-
204Pb correction method
Common Pb correction schemes-
208Pb correction method
•  Assumes that the ratio of 232Th to the
parent U isotope in the analyzed sample
has not been disturbed (concordant)
•  Any excess 208Pb (e.g., 208Pbmeasured -
208Pb*) can be attributed to the assumed
common Pb component.
Common Pb correction schemes-
208Pb correction method
•  The proportion of common 206Pb in this
case can be calculated as:
This formulation works particularly well for targets with very low 232Th/
238U (b0.5), but is not suitable for high Th/U targets
Common	
  Pb	
  correc<on	
  schemes-­‐	
  
208Pb	
  correc<on	
  method	
  
In-situ U-Pb dating by LA-ICP-MS
method
•  In addition to the incorporation of common Pb into the
apatite crystal lattice, U–Th–Pb dating of apatite by LAICPMS
also presents the problem of laser-induced U–
Th–Pb fractionation.
•  A matrix-matched standard is usually required for
external calibration of down-hole fractionation of Pb, Th
and U in LA-ICPMS dating of accessory minerals since
different minerals (e.g., apatite, titanite, perovskite, and
zircon) typically show different time-resolved Pb/U and
Pb/Th spectra during lasering (e.g., Gregory et al.,
2007).
Laser Ablation U-Pb dating Analysis
– Brief Outline
•  Analytical protocol is detailed in Simonetti & Neal (2010, EPSL);
•  UP213 laser ablation system coupled to an Thermo Scientific
Element2 HR-ICP-MS
•  Apatite crystals were ablated using a 55 micron spot size (stationary
mode), fluence of 3-4 J/cm2, repetition rate of 4Hz
•  Madagascar apatite standard was used for external calibration
purposes (i.e. monitoring of instrumental drift and Pb-U laser induced
elemental fractionation) with a ‘standard-sample bracketing’ technique
•  Data Reduction- 207Pb- and 208Pb common Pb correction was
performed (e.g., Simonetti et al., 2006; Chew et al., 2011, Chem. Geo.)
Method
NuPlasmaII	
  -­‐2nd	
  GeneraPon	
  
NuPlasma	
  II	
  Instrument	
  Outline	
  –	
  “Double-­‐focusing”	
  
mass	
  spectrometer	
  
Plasma	
  &	
  
Introduc<on	
  system	
  
Transfer	
  lenses	
  
Electrosta<c	
  
analyser	
  
Laminated	
  
Magnet	
  
Zoom	
  lens	
  
Detector
s	
  
Nu	
  Plasma	
  II	
  –	
  MC-­‐ICP-­‐MS	
  
•  Equipped	
  with	
  21	
  collectors	
  in	
  total	
  
– 16	
  Faraday	
  cups	
  
– 5	
  ion	
  counters	
  (discrete	
  dynode	
  SEMs)	
  
238U 232Th
208Pb
207Pb
206Pb
204Pb,
204Hg
202Hg
In-situ common Pb – LA-MC-ICP-MS
low abundance (≤ 1-2 ppm) samplesMultiple
ion counter configuration
Simonetti et al. (2008)
Simonetti et al. (2008)
Faraday – ion counter
calibration
Time-Resolved Analysis (TRA) Software – NuPlasma II
In-­‐situ	
  common	
  Pb	
  –	
  
Faradays	
  
•  Samples	
  containing	
  more	
  abundant	
  Pb	
  contents,	
  typically	
  
>10	
  ppm	
  
•  Lots	
  of	
  flexibility	
  for	
  the	
  cup	
  configuraPon,	
  i.e.	
  more	
  
Faraday	
  collectors	
  available	
  than	
  needed	
  
•  The	
  instrumental	
  mass	
  bias	
  can	
  be	
  corrected	
  using	
  
simultaneous	
  aspiraPon	
  of	
  Tl	
  (via	
  desolvaPong	
  nebuliser),	
  
or	
  using	
  the	
  “Standard-­‐Sample”	
  brackePng	
  technique	
  
Madagascar apatite standard
478
480
482
484
486
488
490
492
0.0768
0.0772
0.0776
0.0780
0.0784
0.0788
0.0792
0.50 0.54 0.58 0.62 0.66 0.70 0.74
Weighted Mean 206Pb/238U Age
485.18±0.81/0.85/0.99 Ma
MSWD=5.0, n=4
(internal errors/with tracer calibration errors/
with tracer calibration and decay constant errors)
206Pb/238U
The apatite standard was supplied by Stuart N. Thomson, University of Arizona
207Pb/235U
ID-TIMS
Madagascar
apatite standard
Emerald Lake
apatite standard
Emerald Lake Apatite –
Memorial University (Chew et al., 2011)
‘Anchored’	
   ‘Unanchored’	
  
Emerald Lake Apatite –
University of Notre Dame
800
0.0
0.2
0.4
0.6
0.8
1.0
0 20 40 60
238
U/206
Pb
207
Pb/206
Pb
Emerald Lake, apatite
Intercept at
93.0 ± 2.7 Ma
MSWD = 0.88
'anchored'
data-point error ellipses are 2σ
800
0.0
0.2
0.4
0.6
0.8
1.0
0 20 40 60
238
U/206
Pb
207
Pb/206
Pb
Emerald Lake, apatite
Lower intercept at
75 ± 32 Ma
MSWD = 0.88
'unanchored'
data-point error ellipses are 2σ
25
35
45
55
65
75
85
95
105
115
125
135
145
207
Pbcorrectedage
Emerald Lake, apatite
Weighted Mean age= 92.6 ± 1.8 Ma 95% conf.
Wtd by data-pt errs only, 0 of 30 rej.
MSWD = 3.0, probability = 0.000
(error bars are 2σ)
data-point error symbols are 2σ
Memorial	
  University	
   University	
  of	
  Notre	
  Dame	
  
Chew	
  et	
  al.	
  (2011)	
  
CASE	
  STUDY-­‐	
  
	
  
	
  OKA	
  CARBONATITE	
  COMPLEX,	
  QUEBEC	
  (CANADA)	
  
Monteregian	
  Igneous	
  Province	
  (MIP)	
  –	
  
Plume	
  rela<onship?	
  
Monteregian	
  Igneous	
  Province	
  (MIP)	
  
1-­‐	
  Oka;	
  2-­‐Royal;	
  3-­‐Bruno;	
  4-­‐St.	
  Hilaire;	
  5-­‐Rougemont;	
  6-­‐Johnson;	
  7-­‐Yamaska;	
  8-­‐Shefford;	
  
9-­‐Brome;	
  
Oka	
  Carbona<te	
  Complex	
  
Field	
  Rela<ons	
  
CarbonaPte	
  &	
  ijolite	
  breccia	
  
Niocalite?	
  
Chen	
  et	
  al.	
  (in	
  press,	
  Can.	
  Mineral.)	
  
Source:	
  Webminerals	
  
OBJECTIVES	
  
•  Conduct	
  a	
  detailed	
  chemical,	
  isotopic	
  and	
  
geochronological	
  invesPgaPon	
  of	
  the	
  Oka	
  
carbonaPte	
  complex	
  using	
  consPtuent	
  and	
  
accessory	
  minerals	
  at	
  high	
  spaPal	
  resoluPon	
  –	
  
outline	
  petrogene<c	
  history	
  
•  Mantle	
  sources	
  –	
  plume	
  involvement?	
  
Methodology	
  
•  Major	
  element	
  chemistry	
  &	
  
cathodoluminescence	
  imaging	
  of	
  apa#te,	
  calcite,	
  
perovskite,	
  and	
  niocalite	
  conducted	
  by	
  Electron	
  
Microprobe	
  (EMP)	
  analysis	
  
•  In-­‐situ	
  trace	
  element	
  chemistry	
  obtained	
  by	
  LA-­‐
HR-­‐ICP-­‐MS	
  (Element2)	
  
•  In-­‐situ	
  Sr,	
  Nd,	
  Pb	
  and	
  U-­‐Pb	
  age	
  daPng	
  conducted	
  
by	
  LA-­‐MC-­‐ICP-­‐MS	
  (NuPlasmaII)	
  &	
  HR-­‐ICP-­‐MS	
  
(Element2)	
  
Apa<te	
  -­‐	
  Major	
  &	
  Trace	
  Element	
  Chemistry	
  
Chen	
  &	
  Simonem	
  (2013,	
  Chem.	
  Geol.)	
  
Trace	
  Element	
  Chemistry	
  –	
  Apa<te	
  &	
  Niocalite	
  
Chen	
  et	
  al.	
  (in	
  press,	
  Can.	
  Mineral.)	
  
Equilibrium	
  &	
  Frac<onal	
  Crystalliza<on	
  Modeling	
  
Chen	
  &	
  Simonem	
  (2013,	
  Chem.	
  Geol.)	
  
U-­‐Pb	
  age	
  da<ng	
  
Chen	
  &	
  Simonem	
  (2013,	
  Chem.	
  Geol.)	
  
U-­‐Pb	
  Age	
  vs.	
  Chemistry	
  
Numbers	
  (n)	
  Numbers	
  (n)	
  
206Pb/238U	
  age	
  (Ma)	
   206Pb/238U	
  age	
  (Ma)	
  
ApaPte	
  Total	
  
Perovskite	
  Niocalite	
  
0.5122
0.5124
0.5126
0.5128
0.513
0.5132
0.5134
0.701 0.702 0.703 0.704 0.705 0.706 0.707 0.708
143Nd/144Nd
87Sr/86Sr
Our study
Oka_TIMS
East African
	
  
	
  
	
  
	
  
	
  
	
  
	
  
	
  
DMM
HIMU
EM I
EM II
Nd	
  vs.	
  Sr	
  isotopes	
  
Nd	
  vs.	
  Sr	
  isotopes	
  
0.5126
0.51265
0.5127
0.51275
0.5128
0.51285
0.5129
0.51295
0.513
0.7028 0.703 0.7032 0.7034 0.7036 0.7038 0.704
143Nd/144Nd
87Sr/86Sr
Wen et al., 1987;
whole rock ID-
TIMS
Our Study
Apatite Sr	
  
Nd	
  
AP	
  
CPX	
   CC	
  
Pb	
  isotopes	
  
Chen	
  &	
  Simonem	
  (submiped,	
  Geology)	
  
Pb	
  isotopes	
  –	
  modeling	
  
x	
  
Chen	
  &	
  Simonem	
  (submiped,	
  Geology)	
  
Inherent	
  mantle	
  heterogeneity	
  
or	
  
Subducted	
  component?	
  
Pb	
  vs.	
  Carbon	
  
isotopes	
  
Chen	
  &	
  Simonem	
  (submiped,	
  Geology)	
  
Chen	
  &	
  Simonem	
  (submiped,	
  Geology)	
  
Conclusions	
  
•  Combined	
  chemical	
  and	
  radiogenic	
  isotope	
  data	
  indicate	
  that	
  
Oka	
  underwent	
  a	
  complex	
  crystallizaPon	
  history	
  involving	
  
repeated	
  incursion	
  of	
  small	
  volume	
  melts	
  –	
  not	
  related	
  to	
  one	
  
parental	
  magma!	
  
•  What	
  is	
  the	
  real	
  meaning	
  of	
  WHOLE	
  ROCK	
  data??	
  
•  Radiogenic	
  isotopes	
  (and	
  carbon)	
  indicate	
  derivaPon	
  from	
  
mulPple	
  mantle	
  sources	
  –	
  mantle	
  plume	
  with	
  HIMU-­‐EMI	
  
mixture?	
  and	
  presence	
  of	
  an	
  Ancient	
  Depleted	
  Mantle	
  (ADM)	
  
•  In-­‐situ	
  U-­‐Pb	
  ages	
  indicate	
  a	
  protracted,	
  bimodal	
  crystallizaPon	
  
history	
  for	
  Oka	
  that	
  spanned	
  between	
  ~127	
  Ma	
  and	
  ~115	
  Ma,	
  
as	
  indicated	
  by	
  results	
  from	
  all	
  the	
  accessory	
  minerals	
  studied	
  
here	
  
Acknowledgements	
  
Wei	
  Chen,	
  PhD	
  student