RENOICONTI OELLA SQCIETA ITALlANA 01 MINERALOGIA E PETROLOGIA. 1988, Vol. U. pp. 8VN
Stable isotopes in petrology: a brief survey
BRUNO TURJ
Dipartimento di Sci~llU dd1a Terra. Univenita di Roma .. L. Sapienz.... Piazzal.~ A. Moro, OOl8S Roma
AIlSTRAcr. - The isotopic oomprniition of some elements
of low atomic number, notably H, C, 0, N. 5 and Si.
vary in nature as a oonseqUClKe of physicochemical
proccsses such as equilibrium exchange reactions and
kinetic phenomena. In general, the equilibrium constants
of the isotopic exchange reactions ~ tcrnperaturedependent; this is the basis of the isotope
gcothermomctry. Among the above elements, oxygccn
and suborWnately hydrogen arc the most useful in
pctrologkai studies. The 1B()f!40 and D/H ratiO$ (by
convention csptCSscd in oS-units, in parts per thousand
relative to the Standard Mean Ocean Water. SMOW)
arc currently being used to provide information on a
variety of problems. like I) the conditions and
mechanisms of minerals and rocks formation, 2) the
origin and evolution of magmas. 3) the interactions
betwccn magmas and oountry rocks, 4) the nature of the
fluids involved in gcolog.ica1 procC5SC5,.5) the evaluation
of the t:hcrmal history and the scale of fluid migration
during metamorphic processes. and 6) water-rod<:
intenctions. In partkular. oxygen and hydrogen isotope
analyses proved to be cl fundamental im.pon~ to
elucidate the rnctasomatic pnx:cucs occurring in the
upper mantle; they also demonstrated m.t high·and low·
temperature interactions betwccn meteoric waten and
rocks arc widespread and significant in Earth's crus!.
Basic principles of isotopic fractionalion
The abundance of the stable isotopes of
elements having low atomic numbers diplay
appreciable variations in nature as a
consequence of fractionation processes
occurring during chemical, physical, and
biological reactions. The term «isotope
fcactionation» refers to tM partitioning of the
isotopes of a given element between two (or
more) coexisting compounds or phases
containing that element.
In general, such variations of the isotopic
abundances. or isotope effects. are produced
either by kinetic processes or equiJibrium
isotope exchange reactions. Kinetic isotope
effects arise from differences in the rate of
reaction of isotopic molecules, and are
normally associated with unidirectional or fast
processes such as diffusion, evaporation, and
dissociation reactions. The magnitudes of
these effects ace comparable to. or even larger
than. those of the equilibrium processes.
Kinetic isotope effects, however, are relatively
rare in the high-temperature processes
occurring on Earth.
The isotope effects associated with
equilibrium exchange reactions are controlled
by the constant K of the reaction. Take for
example the equilibrium exchange of the
oxygen isotopes IS(} and 16() between carbon
dioxide and water vapor:
1/2CI6()2 + H/80~ 1/2C 180 2 + H/6()2 (1)
where 0 6°2 and C 180 2 mean that oxygen
atoms in the CO molecule are 160 and 180,
respectively(I). The equilibrium constant K
for this reaction is:
K. IC"0P"" . (H , ''0)
U)
(06()2)!/2 . (H 218())
where in parentheses are reported the
concentrations of the various species. K can
be also written in terms of the partition
funcion Q of the reactants and products:
{Il It should be noted that such molccules practically do
not exist in nature; this formalism, howev~r. islargdy
used for simplifying calculations.
84
K
B. TIJRl
&
QIIZ fC 1"O,l" QIU,''Ol. [QICI"O)QICIlQ) (3)
QI12 (C"O,l" Q(H,''OI
[Q(H.. "O/Q(I{ll*O~
Q is defined as follows:
Q = ~g,e-E~T
(4)
where g.:: statistical weight of the energy
level i; E; of the molecule; k: Bohzmann's
constant = 1.381 x 10- 16 erg/oK; T =
temperature in the Kelvin scale.
Q contanins all the energy information
about a given molecule; therefore, the
partition functions are normally used rather
than concentrations for the theoretical
calculation of the equilibrium constant for
isotope exchange reactions.
For any isotope exchange betw~n two
compounds A and B we may define a
fractionation factor er as the quotient of the
heavy-to-light isotope ratios of the element
under study in A and B:
R,
(5)
,
O'A.B=-R-
where R = 180j1 60, Dep 2e, D/H, ).51'25,
and so forth.
Under equilibrium conditions, assuming
that the isotopes are randomly distributed
among all possibile sistes in the molecules, Cl
is related to K as follows:
a A.!., Kl/B
(6)
where n is the number of isotopes exhanged
in the reaction as written. For the reaction
(1), n .. 1 and therefore:
a(CO .11 0) =
1
I
( 180/ 160)
( 18
CO2
0/ 160) Hp
(7)
The general equation (6), however, does not
hold true for the compounds of hydrogen and
for those containing two or more isotopes of
the element is non-equivalent molecular
position. The equilibrium constants, and
therefore the fractionation factors, are
temperature-depeOOent; this is the basis of the
isotopic geothermometry. Their values are in
general very close to unity, typically within
the range from 1 ± O.Ox to 1 ± O.OOX. The
fractionation effects are directly proportional
to the relative mass difference between the
heavy (rare) and light (rommom) isotopes,
provided other parameters such as
temperature, bond strenght and oxidation
state are the same. For example, deuterium,
the heavier stable isotope of hydrogen (0 or
2H), has a mass twice that of the lighter lH
isotope, so that the relative mass difference
of these isotopes is almost 100% and the O/H
fractionations are accordingly substantially
larger than those of any other element of
goechemica! interest. For the stable isotope
of heavy elements like Sr, Nd, and Pb, the
relative mass differences are very small (e.g.
1.2% for the IJS r -86Sr pair); the natural
variations of their abundances are basically
related to radioactive phenomena and the
ratios of the parent to daughter elements, and
do not arise from physicochemical processes.
poly six elements of low atomic number,
namely H, C, N, 0, S, and Si display
variations in their isotope ratios useful in the
study of geological processes. These elements
(notably 0) are important constituents of a
large variety of naturally occurring solids and
fluids, and the abundance of their rare isotope
is sufficiently high to allow precise
determinations of the isotopic ratios by mass
spectrometry. Inasmuch as the difference in
the isotopic ratios of two substances is
measured far more precisely and easier than
are the absolute ratios, the isotopic
compositions of the above elements are
currently determinated by comparing the
heavy-ta-Iight isotope ratio of the element in
the sample (R~) with respect to the
corresponding ratio in a suitable reference
standard (R ,d). By convention, the
difference R.-~.td is expressed relative to the
standard ratio in the so-called {, units, in parts
per thousand Ot permil (°/00). {, is defined as
follows:
R - R""
R,
0.= ( •
) xlO'= ( - RJtd
R..d
1) x 10'(8)
where R., O/H, uCfl 2C, 18Qfl6(), and so
forth. 1£ R..4.. is known, the absolute ratio of
~y sample K~ can be readily obtained from
Its {, value.
The interrelationship between the
fractionation factor a ..... ! and the {, values of
A and B (expressed againist the same
85
STABLE ISOTOPES IN PETI10LOGY: A BRIEF SURVEY
standard) is given by:
1 + oJIOOO
!
1000 + 0,
(9)
,"
The isotopic effects observed in natural
samples are normally expressed as «permil
fractionations». For example, for the
18
0/ 160 exchange raction betweeq CaCO,
and H 2 0 under equilibrium conditions,
0' = 1.0056 at 300°C (FRIEDMAN and O'NElL,
1977), and the permil fractionation is
accordingly 5.6 permiL Inasmuch as 10' In
(1.00X);;;:X, the permil fractionation is
practically equivalent to 10' In Cl. .8' For 0'
values ~ 110 I, the approximate re1ationship
holds up:
"
et' A.B '"
1 + 0,,11000
'"
1000 + 0,
103InO'A.B;;:oA,oB
=
AA.B
(IO)
Thus, the permil fractionation can be
measured to a high degree of approximation
by simply subtracting the 0 values of the two
substances, if such values are close enough.
For reactions involving perfect gases,
10'lnO:' .B varies as lIT and 1/T2 at lowand hi~-temperatures, respectively. Similar
relationships have also been established for
several mineral-fluid pairs and mineral pairs.
Such fractionation curves are widely used in
stable isotope geothermometry. Over limited
temperature ranges, they are in general
represented by equations such as 10 31nO'&,8
= a + bIT
and
10 3InCl.".B = a + bri' 2
(depending on the temperature range) or by
a more encompassing equation like 1031nO'" B
=a+b/T+c/T2.
.
It is important to be aware, however, that
if extended over a wide range of temperatures,
the fractionation curves do not always show
such regular temperature dependencies;
observations have been made of crossovers
(change of sign of the fractionation factor)
inflections, maxima and minima (Fig. 2). The
knowledge of the shape of the fractionation
curve for a given system is essential in order
to avoid serious errors if the curve is used for
extrapolating data far beyond the temperature
range over which the curve has been
established.
At infinite temperature, the fractionation
factors between all substances become unity
(that is, the isotopic fractionations go to zero;
T 'C
, •
.
•
•
.•
~
~
•
,
0
-,
,
Fill. 1.
. . .
..
Expcrimcntally dClcrmincd 180f!60
for some mincral·water systems.
frll.CtiO~lion curvcs
BIGELEISEN and MAYER, 1947). Therefore, the
fractionations occurring in deep-seated
assemblages are typically smaller than those
occurring in supergenic environments.
Among the 6 light elements listed above,
oxygen and subordinately hydrogen proved
to be the most useful in petrological studies.
Oxygen is the most abundant element in
ordinary rocks, making up about 50% by
weight; it is therefore a major constituent of
most minerals present in igneous,
metamorphic, and sedimentary rocks, as well
as of a variety of fluids occurring in both
surface and deep-seated environments.
Hydrogen is by far less abundant than oxygen
in the rocks; its concentration varies from
trace amounts in many basaltic and ultramafic
rocks to about 1-15 wt.% in clays and
serpentinites.lt is, however, a constituent of
many important minerals (micas, amphiboles,
zeolites, and so forth) and fluids (H 2 0,
CH 4 , H 2 S, etc.).
s.
86
i
"
, ,
TURt
o
I 'C
Sedimentary rocks display the highest
8180 values, igneous rocks the lowest ones;
metamorphic rocks have in general 8180
values intermediate between those of the
sedimentary and igneous rocks. Such
differences in isotopic composition basically
reflect the different thermal features of the
geological environments where the three types
of rocks developed. For example, sedimentary
rocks and minerals are markedly enriched in
180 because of the large fractionations
permitted at the low temperatures prevailing
in the surface environments.
Within a given family of rocks, the 818()
c -
,
•
•
Fig. 2. - Calculated Il()pt() f~liofation curvd for
cl plotica! interest iIlustnting unusual
minima, maxima and infkctiOtlS.
SO~ S)'$1ems
Oxygen and hydrogen isotope analyses are
currently being used in petrology la provide
information on a variety of problems, like 1)
the conditions and mechanisms of minerals
and rocks formation; 2) the origin and
evolution of magmas; 3) the interactions
between magmas and country rocks; 4) the
nature of the fluids involved in geological
processes; 5) the evaluation of the thermal
history and the scale of fluid migration during
metamorphic processes; 6) water-rock
interactions.
18()f16Q variations in minerals and rocks
The 0 18 0 values of natural materials
(relative to the Standard Mean Ocean Water.
SMOW), vary within the wide range from
about-55 (glacier ice near the poles) [Q -+- 41
(atmospheric CO~; see Fig. 3.
The variability of 8 18 0 in igneous,
metamorphic, and sedimentary rocks is much
narrower (Fig. 4).
variation displays some regularities due to the
chemical and mineralogical features of the
rock. All else being equal, the isotopic
propenies of a substance depend on the nature
of its chemical bonds. [n general, bonds to
ions with a high ionic potential and low atomic
mass tend to incorporate the heavy isotope
preferentially. This is basically the reason why
in igneous and metamorphic rocks under
equilibrium conditions quartz is always the
most 18()·rich mineral and Fe- Ti oxides are
always the most 180·poor minerals (TAYLOR
and EPSTEIN, 1962; GARUCK and EPSTEIN,
1967; O'NElL, 1977). After quartz, alkali
feldspar and plagioclase are the most 180_
rich minerals; such decrease in the 18 0
content of framework silicates results from the
progressive replacement of Si·O bonds by
Al-O bonds.
Thus, one should expect to observe a
regular order of 18 0 enrichement in
equilibrium mineral assemblage. For example,
in granitic rocks, coexisting minerals always
concentrate 180 in the sequence magnetite
..... biotite ..... hornblende ..... muscovite .....
plagioclase -- alkali feldspar -+ quartz.
Sometimes, however, feldspar is enriched in
18 0 with respect to quartz (isotopic
reversal), and in other cases the 180/ 160
fractionation between quartz and feldspar
displays values like 5 or 6, that is appreciably
higher than the normal ones (1.0-2.5; TAYLOR,
1978). Such granites were conceivably
«dhturbed» by some later event such as an
isotopic exchange with external fluids at low
and high temperatures, respectively; the
mineral assemblages did not retain the original
87
STABLE ISOTOPES IN PETROLOGY, A BRIEF SURVEY
r
:!
. ,- TI'
_
0_..
S .. OlMENTS
C ... IIIION ...TES
,""0"
~._
0...... ""...
_........
..~~
-" ..
........
•
IGN"OUS ROCKS
/~
OCE"N W"TER
Fm..
::
~.r
,
\r
.,'U_
....._.V
!':::
~
_ '......m'..e
u ...
_'"~
M...G.....TlC:
w...T"RS •
I
_.....".
'0,
1···~'t1
U.LI.
...... / ~
I.. _....,.
-0
D••U"OM ,M'"
..... _'0
0,.Me,.1I0M
fRESH W...TEIII
PHOTOSYNTHESIS
2H,O_(CH,O)+H,O+O,
co."
~
Fig. J. -
...c'·"."OM ,.
....... '.O'OM'
Oxygen isolope v.riations in nature. (Modified atfer RoATO, 1961)
I
CIIIl.T.
..... "lllll
SEDIMENTARY
ROCKS
I
1.1I0.TOIOII
I
I
I
L
_
G".Nu...l.
-l
~"Bij
.TlT
r.• "OROTNE".......T
••••••,.,.,c_,-- O~~~.,.,.~.['~",.~=i
r
,-nO'..
Il~T''''
O
D····..·
"~"""_""'=I
II
o
FiB. 4. -
8
IGNEUS ROCKS
I
1
LUII.. R ROU.
4
I
,
~~"c.~~:'~.
o
11
1
I
AltOESIU.
UIA...
METAMORPHIC
ROCKS
.CHI.".PNI.UTE •
.......III.O.,Yl.
. . . .RIlO OR.IIITES
I
.,.IlUO,,1l1
12
6 '"0 ,...,
16
20
24
28
32
Oxygen isotope composilions of ignenus, metamorphic and scdimcmary rocks.
88
8. nJRI
isotopic equilibrium b«ause [he various
minerals exchanged to diffettnt extents with
the fluids. From the above discussion it
-,
becomes clear why, as far as .normal. igneous
rocks is concerned, the more SiC?i"rich clans
are progressively enriched in I"V, for the
.,
o
."
."
M~ltO"'fS
(uc!. carbo choM"'n)
Lu~or
I ~g~I~"
i ;:::J----.
rocks
,
.1
Sould-er bolhOI,1h
..
'I'
···..3a··
O+---'-9=--~---'~f=~~
\
~o;,:..
Son Juan Mtns.
'.1.'
-'0
.
SCOtl'Sh HttM'Cln
-.
,',
~ .' :~;.
,.J:.."7
••• '.. ". • • • '}
:1:',
'~~':~/'i
:;:lff' .••
. i ' _.....
J. , • •
.
',-.:
,
'-~
.
, '.*=,,~~
.
~'"
••
GO
. ' ..
.
-~J'":J.!!--
-~
8"'d', lonn, oa<on",c plul0ns
,,)~o' :~.
I
,~~~.
-
I
~ '1.'0 .' •
Rhy01.r,C 0''''110'' lulls
......... ,
PlutOflle 9,on"II,
ll'onOO'o"lu 8 lonol.1U
ScUOllS 6 9Obb'O$
• wrwle
rock
• Quo'"
4no<l"os"n
• rtlclWClt
• PyrouM
-,
o
.,
."
Fig, ,. _ 6 ll1().values of igf1COus rocks and minenls from various localities throughout the world. (From T AVLOR,
1974; ~prodoccd by permission of the: Ee:onomic: Geology Publishing Company).
STABLE IsaroPES IN PETROlOGY, A BRIEF SURVEY
1bis compilation rep~nts the state of the
art in 1974; since then, of course, a huge
amount of new analyses have been produced,
and the data base is now so voluminous to
make it pratically impossible to synthetize it
in a single diagram. In spite of this, Fig. 5 still
represents the main features of the 18Q
variations in igneous rocks.
sequence from ultramafic rocks through
basalts and gabbros, andesites, dacites,
tonalites, granodiorites, to rhyolites and
granites.
In marine sedimentary rocks, a trend to
concentrate ISO similar to that outlined
above is noticeable among authigenic
minerals, resulting in a progressive decrease
in 180/ 160 in going from cherts through
carbonates and shales to ferromanganese
nodules (KNAUTH and EpSTEIN, 1976;
KOLODNY and EpSTEIN, 1976; SAVIN and
EpsTEIN, 1970; VElZER and HOEFs, 1976;
FtUD et al., 1983). Note that, with respect
to other classes of minerals, carbonates are
enriched in Il10 because their oxygen is
bonded to the small, highly charged C 4 • ion
(O'NElL, 1977).
1. Basaltic rocks
The most abundant type of igneous rocks
produced at present on Earth is represented
by the mid-ocean ridge basalts (MORB).
Modern MORB are remarkably uniform in
8 18Q (5.7 ± 0.3); very similiar values are
observed in many ultramafic products (e.g.
TAYLOR, 1968; KYSER et al., 1981, 1982).
This strongly suggest the existence of a large
reservoir in the upper mantle very uniform
in l8Qf16Q; as shown by radiogenic isotope
studies, however, this reservoir is «depleted»
in LIL elements and ndiogenic 87Sr.
The average 8180 of MORB is nicely
Some applications of oxygen isotopes
in igneous petrology
The oxygen isotopic composition of igneous
rocks is portrayed in Fig. 5.
• Tholeiile
6 Alkalie
OK~lava
89
BASIC LAVAS
:
:
: .1
.11 .n.t
MAR
i.
:iil .
.....
,
EPR
.f.... •• •
IIin5lit
I 1•••• '"
6 ••• 6
Oceani
Is.
6
I>
iut
I>
Continental
4
Fig. 6. - D lSQ·valun of fresh, uncontaminated tho1c:iites, alkali bas.ln and potassic lavas from continental and
oceanic areas. EPR. East Pacific Rise; MAR. Mid·Atlantic Ridge. (From KY$1t, 1986; rCpmducuJ by permission
of the Mineralogical Society of America).
90
B.
comparable to that obtained for lunar basalts.
The extreme uniformity in 0180 of the
Moon is attributed to the lack of water and
consequently to the absence of lowtemperature alteration processes.
It is therefore conceivable to cooclude that,
in terms of oxygen isOlOpeS, only minor
differences exist between the Earth and the
Moon (which appeared to have formed from
the same bulk reservoir), as well as between
the Earth-Moon system and the ordinary
chondrites, which represent the most
abundam types of meteorites
(CLAYTON,
1976; TAYLOR and SHEPPARD, 1986).
The oxygen ismopic composition of fresh,
uncontaminated tholeUtes and alkali basalts
from oceanic islands (OIB) and continental
basalts and potassic lavas are shown in
Fig. 6. Also shown for comparison are many
0 18 0 analyses of basahs (mostly tholeiites)
from the Mid Atlantic Ridge (MAR) and East
Pacific Rise (EPR); note that these two groups
of MORS are pratically indistinguishable.
A considerale overlap is evident in the
ip 80 values of OIB and MORS; however,
tholeiites from oceanic islands tend to be
lower in h l8() than those from oceanic ridge
systems, averaging 5.4, whe~as alkali basalts
tend to have higher 018() values, averaging
6.1, relative to tholciiric basalts, whether they
come from oceanic ridges or islands (KYSER
et al., 1982). lboleiitic and alkali basalts from
continental areas are quite variable in
1ll()/I6() with h l8 0 mostly betw«:n about 5
and 7.5 (TAYLOR. 1968; KYSER et al., t982).
Cuntinental potassic lavas have typically
values relatively high, from about 6 to
8 (TAYLOR et al., 1984; FERRARA et al., 1985,
1986). These variations in OHIO attest to the
18 0/ 160 heterogeneity in the upper "mantle
beneath the continents. The most plausible
processes responsible of such heterogeneity are
1) subduction processes; 2) high-temperature
merasomatic exhange of upper mantJe
minerals with external, oxygen-bearing fluids
(C0 2, 1-!l...0' etc.). Leucite-bearing lavas can
show 01"U values as high as about 10 (TuRl
and TAYLOR, 1976; TAYLOR et al., 1979;
FERRARA et al., 1986); much of this IS().
enrichment, however, is due to interactions
with the 180_rich continental crust. The
01~0
TURI
highest 0'80 values of leucitites or
nephelinites attributable to upper mantle
processes are very likely around 7.5-8.0
(KYSER. et al., 1982). The primitive high
ISO-contents of some Si01-undersaturated
aIkalic magmas can be attributed to a
precursor metasomauc activity that brought
about an enrichment in IS() (and in LIL
elements as well) of their source regions in the
upper mantle. The relatively high-IS() alkali
basalts (0180 ~ 6.5 or somewhat higher)
probably derived from partial melting of
merasomatically IllQ-enriched peridotites
(GREGORY and TA nOR, 1986).
2. \Y/atcr-rock interactions
Volcanic and plutonic rocks with 0 180
values much lower than 5 have been found
in various localities throughout the world.
Some of them, such as some low- l8() basahs
from Iceland (MuEHLENBAcHs et al., 1974)
and the Seychelles granite pluton (TAYLOR,
1977), definitely crystallized from
anomalously IOW· 180 magmas; the majority
of such rocks are instead the results of subsolidus meteoric-hydrothermal alteration.
Meteoric waters are typically depleted in
both 180 and D relative to seawater (CRAIG,
1961). Isotopic exchange between such waters
and rocks at moderate to high temperatures
results in appreaciable reductions of the
OlllO and oD of the rocks, with
corresponding enrichmenrs in the 0180 and
oD of the water. These types of hydrothermal
systems, which represent the « fossill)
equivalents of the deep portions of modern
geothermal systems, occur wherever igneous
intrusions are emplaced into permeable watersaturated rocks near the earth's surface. Silicarich, calcalkaline magmas are most commonly
involved in such systems, probably because
viscous magmas lend to form stubby
intrusions and because their explosive
character promotes fracturing (TAYLOR, 1974,
1977).
So far, ~ than 50 meteoric-hydrothermal
systems, very variable in size, have bttn
recognized (CRlSS and TA-YLOR, 1986). The
~Iatively smaller ones Oess than 100 km 2 ,
such as Tonopah, Nevada (TAYLOR, 1973),
STABLE ISOTDPES
IN
PETROlOGY, A BRIEF SURVEY
Bodie, California (O'NElL et al., 1973),
Skaergaard, Greenland (TAYLOR and
FORESTER, 1979), Ardnamurchan, Scotland
(TAYLOR and FORESTER, 1971) in general
developed around small stocks and volcanic
centers, whereas much larger systems, ranging
from lOO's to 1000's km 2 in extent, are
associated with large calderas and batholiths;
examples are the Idaho batholith (TAYLOR and
MAGARITZ, 1978; CRlSS and TAYLOR, 1983)
the Mull and Skye volcanic centers, Scotland
(FORESTER and TAYLOR, 1976, 1977), and the
Lake City caldera, Colondo (URSEN and
TAYLOR, 1986).
Even larget systems (10,000 km 2 ?) occur
whete oceanic spreading centers impinge into
subaerial environments, like at Jabal at Tirf,
Saudi Arabia (TAYLOR, 1980).
The application of the stable isoto~
techniques to the fossil meteoric-hydrothermal
systems represents a powerful and unique tool
fot eluddating the fluid/rock processes in the
inaccessible parts of the crust. The stable
isotope analyses allow us to estimate the initial
isotopic composition of the fluid and the
relative prop:>rtions of fluid and rock involved
in the exchange. It is now known that fluid
circulation is usually concentrated along major
structures, although substantial quantities of
fluid can also migrate along minor fractures
and grain boundaries. NORTON and TAYLOR
(1979) made a computer simulation of the
Skaergaard meteoric-hydrothermal system.
They were able to demonstrate that most of
the sub-solidus hydrotermal exchange in the
intrusion took place at very high tem~ra~
(400-800°C) and that, over the 500,OOO-year
lifetime of the system, integrated amounts of
100 to 5000 kg of water have flowed trought
each cm 2 cross-section of rock above the
level of the unconformity sc=parating a 9 kmthick pile of permeable Early Tertiary basalts
from the underlying Precambrian gneiss, much
less permeablc=.
Fluid circulation can often attain depths of
5 to 10 km (c=.g. in the Skaergaard systc=m).
Taking into account the areas of these systems
and the volumes of rocks involved, it turns
out that a significant portion of the earth's
crust has experienced such a type of meteorichydrothermal altc=ration.
91
Meteoric-hydrothc=rmal fluids play an
important rolc= in ore deposition (TAYLOR,
1974, 1979; O'NElL and Sll.BERMAN, 1974;
BETHKE and RYE, 1979; CASADEWALL and
OHMOTO, 1977; and others). Such type of
fluids are almost invariably tesponsible for the
formation of the epithermal ore deposits, a
class of shallow-level mineralizations
developed in enviroments with striking
analogies with modern geothermal systems
(FIELD and FIFAllEK, 1985). Meteoric waters
also partidpate in the genesis of numerous
deeper, higher temperature ore dc=posits,
usually together with fluids of different origin
(e.g. magmatic). To Conclude this sc=ction, it
is worth noting that if the interactions with
meteoric waters takc= place at low
temperatures, like those occurring during
weathering, glass hydration, and in the lowtemperature portions of some mc=teorichydrothermal systems (TAYLOR, 1968;
LAWRENCE and TAYLOR, 1971; CKlSS et al.,
1984), the (VS() of the altered rocks can
actually be increased, because of the large
values of the mineral-water fractionation
factors at low temperatures.
J. High- 180 igneous rocks
The majority of the igneous rocks on Earth
have 8 1s O values lower than about 10,
typically between 5 and 10 (Fig. 5). Igneous
rocks with OIS() > 10, however, are not so
rare, and their widespread occurrence has
progressivdy been recognized in the last tc=n
years. This explains why such rocks are
scarcely represented in Fig. 5. The high-IS()
igneous rocks are mostly represented by
granitoids and their volcanic equivalents.
They can be divided into 3 subgroups,
depending upon their anomalously high
ISO/I6() ratios are a) inherited from the
original magmas; 2) the rc=suJt of hightemperature interactions with high-ISO
country rocks, or c) the result of a secondary
event like weatheting or low-temperature
hydrothermal alteration, as discussed in the
previous section (TAYLOR, 1978).
Subgroup a) includc=s plutonic and volcanic
rocks crystallized ftom magmas of Ctustal
origin. High_ISO, alumina·rich sedimentary
92
B. mRI
and metasedimentary rocks played an
important role in the genesis of such magmas.
Most, but not all, of these
ha~
also high
nSrfl6Sr ratios (say, > - 0.710); the
strontium isotopic composition is in fact
controlled by the Rh/Sr ratios of the source:
material and the difference between the age
of this material and the one of the partial
melting event.
One of the first examples of high.l8()
magmas reponed in the literature is the PUoPleistoa=ne, anatectic Tuscan Province, northcentral haly, which consists of a number of
rhyolitic volcanic centers and epizonal granitic
intrusions with 618 0= 11 [0 16 (TAYLOR and
TURf, 1976 and unpublished data). The
highest-ISO magmas yet known have been
observed at Talfa, the southermost district of
the province (6 180 of K-feldspar = 15.1-16.1).
More recently, several other occurrences of
high- 180 granitoids have been observed
throughout the world. Examples are the Late
Proterozoic to Mesozoic transformation series
(analogous to the cS-type» grmitoids) of South
China, with 618() = 9-14 and nSrf&6Sr >
0.710 (D. ZHANG et aL, 1984; 1. ZIIANG et
aL, 1984); the peraluminous Cenozoic
adamellite and granites of the North
Himalaya, High Himalaya and Paleozoic
cLesS4=:r Himalaya..., with 618() .. 9.2 to 14
(mostly> 11; BLAITNER et al., 1983; VIDAL
et aI., 1984; DEBON et al., 1986) and
nSrf&6Sr mostly > 0.720; the Hercynian
aluminopotassic granitoids of southwest and
central Europe, with 6180 values mostly
between 10.0 and 13.5 (VIDAL et al., 1984);
the «S-type» (pelitk) granitoids of SE
Australia (O'NElL and CHAPPELL, 1977;
O'NEIL et aI., 1977), with 0180=:10-12.
Closing statement
Stable isotope geochemistry is mostly
concerned with the natural variations of the
isotope ratios of 6 elements of low atomic
number <H, C, 0, N, S, and Si); among these,
oxygen is by far the most important because
it is a major constituent of ordinary rocks as
well as of many naturally occurring fluids.
Stable isotope analyses of minerals and rocks
are more and more being used to provide
information on a variety of petrological
problems
regarding
sedimentary,
metamorphic, and igneous rocks; they also
proved to be very useful in understanding the
origin of magmas, especially when combined
with analyses of the radiogenic isotopes of
heavy elements.
180/ 160
(and D/H) analyses are of
fundamental importan~ to elucidate the role
of fluids in geological processes; among other
things, this approach gave a significant
contribution to our knowledge of the
metasomatic processes occurring in the upper
mantle and demonstrated that high· and low.
temperature interactions between meteoric
waters and rocks are widespread and
significant in the Earth's crust.
R E FER E N C E S
BETIIKE P.M., RVE R.O. (1979) - Environmml of 0"
d~sitio" in l~ C~ mining districJ, S#n Jwm
Mounlilins, Colorado - PII" N, Souru of/!u;ch from
oxygl'7l, hydrofpt IIlfd cllrbon now~ stuJi~. Econ.
GeoL, 74, 1832·18:51.
M.G. (1947) - Ukuutt01l of
BIGEl.£1SEN J., MAYU
~uilibrium conflilnlf
fOT isotopu t!JC4hfJ,nv ructions.
Joom. Chcm. Phys., U, 261-267.
BLArrND P., DIETlUOI V., GANSSD A. (l9S3)
Contrasting "0 mrichmcdJ .nd oriVIff of High
Hima/4y4n and Trrurshi_1Iyan intrusiPes. Earth Planet.
Sci. un., 6', 276-286.
BoAiU G. (1960 -l~/ru£ti0n4tion f1rOUJSI:J in MJurt.
Summtt COUl"$e on Nuclear Geology, Varenna, 1960;
Pisa, Laboratorio di Geologia Nuclean-, 129·149.
CASADEVAI..L T., OllOMOTO H. (1977) . Sun",.siJ~ min~,
Eurtka mining districl. San Juan Country, Colorado.
G~DChrotistry of gold uti IMs~·m~liIl ore dqosilion in
a volcanic el/vironmmt. Econ. Geol., 72, 12S,·1320.
CLAYTON R.N. (1976) - A classification ofmeteoritn bDwd
on oxygel/ iSOlopt'S. Earlh Planet. Sci. Leu., 30, 10·lS.
CRAIG H. (1961) - Isolopic varialions in meteoric walm.
Science, D3, 1702·1703.
CRISS R.E., TAYI.OIlI-I.P. JR. (l9S3)· An 1I0f~lInd
D/H sludy of Tertillry hydrothermlll syStn!lS in tIN
souliKm half of tIN IJoho &tholith. Bun. Geol. Soc.
Am., 94, 640·663.
CRISS R.E., TAYLOR H.P. JR. (1986) . /Il~leoric·
b)·drothmnaIJYJJems. In: VALLEY lW., TAYLOR H.P.
Ja., O'NEIL J.R. (Eds.), Stable Isotopes in High
Tempenlllure Geological Processes; Reviews in
Mineralogy 16, 37)-424.
CIlSS R.E., EoDI E.B., HAJl.DYl\.1AN R.F. (1984) - C4JJo
ring :zonr.. 4,'00 knrl fossil bydrothnmlll syJJem in the
Chaf/is IIObnu fi~/4. cmtral Iti4ho. Geology, 12,
HI-334.
STABLE ISOIOPES IN PETROLOGY, A BRIEF SURVEY
DEIlON F., 1.£ Fon P., SHEPP.u.D S.M.F., SoNo J.
(1986)· The four plutonic bdu of the Transhimalaya.
Himala)'ll: a chemiClll, mincralogical, isotopic and
chronological synthesis along I Tibet·Nepal section.
Jour. Petrol., 27, 219-250.
FERRARA G., LAURENZl M.A., TAYLOR H.P. JR.,
TONARlNt S., TURI B. (1985) . Oxyg~ and strontium
isotope studid of K·rich uokanic rodtJ from the A/ban
Hills, Italy. Earth Planet. S<:i. Leu., n, 13·28.
FURAIA G., Pa£rrE M.u.TINEZ M., TAYWk H.P. JR.,
TONAAlNl S., TUJlt B. (1986) - Ellidmu JOT crustal
assimiliztion, mixing of magmas, nd a '15T·ridJ 1Ipp"
mtlntk: An 0It'fV1I fUrS JIroI/tillm iJou>~ lhuJ, of t«
VII/siniJm DiJtrict, Untrtll Italy. Comrib. Mineral.
Petrol., 92, 269-280.
FIELD C.W., Rn: R.D., DYMONO }.R., WHELAN JF.,
SENECHAt. R.G. (1983) - M~ta/liftrOus sdimmn of the
&st Paci/K. In; SHANKS W.C., III (Ed.), Cameron
Volume on Unconventional Minentl Deposits, Society
of Mining Engineers, New York, 1)}·156.
FIElD e.W., FIFAREK R.H. (I 98') - Light stable.isotope
systematics in the epithermaJ environment. In: BEll.Gr:Jt.
B.R., BETIlKE P.M. (Eds.), Geology and
Geochemistry of Epithermanl Systems, Reviews in
Economic Geology, 99-128.
FoaESTEJ. R.W., TAYWR. H.P. JR. (1976)· ISO<kpImJ
ign~IIS rot:h from tiN Tt'TtitlT'f complec of tiN isk of
Mlll/, Scotland. Earth Planet. sa. Lett., 32, 11·17.
FORESTEJ. R.W., TAYLOR H.P. JR.. (1977) • l'Off(),
DIH tmd lJC/'lC J/Ullin of t« Tntiary ign~UJ comples
of S/ryt, Scotland. Am«. Jour. Sd., 277, 136-177.
FRIEDMAN 1., O'NEILJ.R. (l977) • Compilation ofuabk
isotopt fmctionation factors ofgtoehtmical in/~t. In:
FLEISCltER M. (Ed.), Data of Geochemimy, 6th Ed.,
U.S.Gov. Priming Office, Washington, D.e.
G.4.RUCK G.D., EP$TF.lN S. (1967) . Oxygen iJoto~ rtltios
in coaisting mintnlfs of ~/Jy mtt4morphosd meh.
Gcochim. Cosmochim. Acta, 31, 181·214.
GREGOKY R.T., TAYLOa H.P. Jt. (1986) _ NontquiJibrillm, mttasomatic lJ()ff() qftcts in IIppn
ItUUItk mintnlla_hlaus. Contrib. Mineral.. Pctrol.,
93, 124·U'.
KN.4.UJ1-{ L.P., EPS'mN S. (1976) . HyJrogt1l and 0X'ffP'
is%pt filtiOS in nodlllar tlnd b«1tkd chen. Gcochim.
Cosmochim. Actl, 40, 1095·1108.
KOLODNY Y., E!'STEIN S. (1976) • Stabk isoto~
gtoehemistry oftktp sta chem. Geochim. Cosmochim.
Acta, 40, 119'·1209.
KYSER T.K., O'NEILJ.R., CARMICllAEL I.S.E. (1981)
. Oxygtn isotopt thtrmom~try ofbasic /alIaS and mantk
nodules. Contrib. Mineral. Petrol., 77, 11-23.
KYSER T.K., O'Nm J.R., C....MIOlAf.L I.S.E. (1982)
. G~til: trlatimu IInfOtrg /Nuic lavas and IIlmt11l4/U
IlOdIlIcs: wiJnJct from o:rt)'lt'I'I iJou>pt comptniti01U.
Contrib. Mineral Petrol., 81, 88·102.
KYSER. T.K. (I986)· St4bk iJoIopt IC'iatiom in t« mantk.
In; VAllLY J.W., TAYLOR H.P. Jt., O'NEIL J.R.
(Eds.), Stable hotopes in High Temperature
Geological Processes. Reviews in Mineralogy, 16,
141·164.
!.ARSON P.B., TAYLOR H.P. JR. (1986) . An oxygen isotopt
study ofbydrothmnal aluration in t«!.Mc City (4kkra,
$tin Juan MOllntains, Colorado. Joor. Volcano!.
93
Geothcrm. Res., }(), 47-82.
!.AWREHCEJR., TAYLOIl. H.P. JR.. (1971) - Dnltnium IUII1
OX)'gtn-IB ~rtr/a.tiOll: Cl4y mi_ls and /rydTuxitks in
Q!u1tcmt11'1 soils comptnaJ to mNoric 1lKlJm. Gc.ochim.
Cosmochim. Acta, 3', 99J.100J.
MUEIlLENB.4.CIIS K., ANDERSON A.T., SIGVALDASON G.E.
(1974) . Low_ IiO basalts from Iceland. Geochim.
Cosmochim. ACla, J8, ,77·,8g.
NORTON D., TAYLOR H.P. JR. (1979) - Quantitatillt!
simulation of tM bydrotMrmtll systems of crystal/hing
magmas on tM basiJ of tTtIlfsport theory tmd OX)'gtn
iJotopt daJ4: till tI1IIllpU 0/tiN ~ inlnlsion.]our.
Petrol., 20, 421·486.
O'NElLJ.R. (1977) - Stabk iJOto~;" minmtIor1. ?hys.
Chem. Minerals, 2, IO,-12}.
O'NElLJ.R., S~1AN M.L., FABB.I B.P., CUESTDMAN
c.W. (19H) - Stahk isotopt and ckmical rtlations
during mi1lml/ization in tht Bodit Mining District, MOll0
County, California. Econ. Geol., 68, 76'·784.
O'NEIL J.R., StLBERMAN M.L. (1974)· Stabk isoto~
rtlatiOI'lJ in tpithtrmal Au.Ag deposits. Econ. Gent., 69,
902·909.
O'NEtt J.R., CUAPPFll B.W. (1977) . Oxygen and
hyJrogm isoto~ trliltionJ in tht Btlrridak Nlholilh.
Jour. GeoI. Soc. London, In, 559-571.
O'NElLJ.R., SllAW S.E., FLOOD R.H. (1971) - Oxygen
IUII1 bytbop iJoto~ composiJiom as inJialIors oftJrlltik
gtntUS in tiN Ntw England btltholith, AIIJIJ'tlu..
Comrib. Minttal. Petrol., 62, 313-J28.
SAVIN S.M., EPSTEIN S. (1970)· TM r.«ygtI'I tmd bydrofpi
iwtopt gtoekmistry of clay mintrals. Geochim.
Cosmochim. Acta, 34, 2'·42.
T AYLOR H.P. JR. (1968) - TM oxygen isotopt w:ochtmistry
of igntous melts. Contrib. Mineral. Petrol., 19 1·71.
TA YLOR H.P. JR. (1974) - TM application of oxygtn and
bydrofPI isotopt studin to pmbkms of hydrolhtrma/
altcation and Otr tkposition. Econ GeoI.., 69, 843-88J.
TAYLOR. H.P. Jt. (1977) - Watn/rock intmlCtion.nd IM
origin of Hp in granitic !MJholiths. Jour. GeoI. Soc.
London, 133, '09-558.
TAYLOR H.P. Jt. (1978) - Oxyr and bydrognt isoto~
studie of plMtonic gnrnitic roch. Eanh Planet. Sd.
Leu., 38, 177-210.
TAYLOR H.P. JR. (1979) - Oxyr and bydrovn isoto~
rtlationJhips in hydrotMmut/ mintrtll dtposit!. In:
B.4.RNES H.L. {Ed-l, Geochemistry of Hydrothermal
Ore: DCJXlsits 2nd Ed.,J. WHey & Sons, New York,
236·277.
T.4.YIOR H.P. JR. (1980) - Stabk IJOtopt studit'J ofsprtading
CtntnJ and thtir MJring on tht origin of gnrnophym and
plagiognvlitn. CoIJoques internationaux du CNRS nO
272 - Associations mafiques et ulrramafiques clans les
orogenes, 149-165.
TAYLQa H.P. JJl., EPSTEIN S. (1962) - Rn.tionship
brtw«n Il()fWO nUios ;" wt:ICisting minmJ/s 01 ifp«1IIs
and m~taf1lOfPhic rot:h. BuiL GeoL Soc. Am., H,
675-694.
TAYLOR H.P. JR., FORESTER. R.W. (1971) . Low-r'O
ign~us rocks from tht intrusivt compl= ofS/cyt, Mul/
and Ardnamurclxll'l, Wtstem Scotland. Jour. Petrol., 12,
465·497.
T.4.YLOR H.P., FORESTER R.W. (1979) -An oxygtn and
hydrogtn
is%~
study 0/ tM SkMrgtUlrrl intnlsion and
94
8. TURI
in country rot:ks: " dnuiption of a 55 M. Y. oM fOUlI
hydrotMmta/ rystnn. Jour. Petrol., 20, J~5-419.
TAYLOR H.P. )11.., MAGARITZ M. (1978)· Oxygen and
hydrogen isotopr studin ofthe Co,dillnan /Mtholiths of
wf!Slern Norlh America. In: ROBINSQN B.W. (Ed.),
Stable Isotopes in Ihe Earth Sciences, DSIR Bull.,
220, IB·17l.
TAYL.Oa H.P. JR., GIANNETTl B., TUaJ B. (1979) - c.>xnm
ooto~ g«K:bnnisJry 0/ dH pol4sJk ign«JIIS rot:ks from
~ ROCUInOIfji".llOIuno. Rommr Comagmatit: &,;on,
ItII/y. EUlh Planet. sa. I...eu., 46, 81·Hl6.
TAYI.OR H,P. JR., SIIEPPARD S.M..F. (986) - 19ntollS
rocb: I. hoc6St'S ofisotopic fractiollation and ;SOlOpr
ryltemafU;J. In: VALLEY ).W., TAYl.OR H.P. )11..,
O'NEIL ).R. (Eds.), Slable Isotopes in High
Temperature Geological Processes. Reviews in
Mineralogy, 16, 227-271.
TAYl.OR H,P. )Il., TURl B. (1976) . High.I'D ipNJUJ
rodes from tk TllJefl1l
/Ifllgmlltit;
i"rovinu, fury.
Contrib. Mineral. PetrOl., 55, })-54.
T"l'l.O& H.P.)Il.• TUl{1 B., CUNDAIU A. (1984) -il()f'O
and cht:mkal rrlationships in K-rich 1IO/aJ"k rochfrom
AU$tTlIlia, Eim Africa, Ant4rctica and Sa" Vtnanto·
Cuflt"l/o, lt4ly. Earth Aanet. Sci. un., 69, 263-27;.
TUIlI B., TAYLOR. H.P. JR. (1976) . Oxygt:n iwto~ Jtud~
ofpot4$$k uokanic rocks oftM Roman Privinu, eenfml
Italy. Contrib. Mineral. Petrol., ", I·H.
VElZER J., HoEFS}. (976) . The naturr of "0/,60 and
uqrlC s«ular rrrnds in sNJmmt4ry car!xmaU rockJ.
Geochim. Cosmochim. Acta, 40, 1387·1395.
VIDAL P., BEJlN.UI).GJJF'FTTltsJ., COCIIERIE A., LE FOltT
P., PEUCAT }.}., SIlEPP.....O S.M.F. (984) .
Gcocht:miul CO"'/Jllrison ~ Him4Lzy"" and
H~"Uin ln1cognz"ius. Phys. EllI'th Planet. Inter. 35,
179·190.
ZIlANG 0 .. HU"'NG F., ZIIENG S. (1984) . Tht oxygen,
hydrogt" and carbon iSOlOpt studitJ oftlmgstm.btaring
granitoidJ. In: Xu T., Tu G. (Eds.), Geol08Y of
Granites and Iheir Metal10gentic Relations. Science
Press, Beijing, China, 875·890.
ZIIA.'!G L., Guo Y., Qu P. (1984)· A JJuJ, 0" tht 0X1fpI
isotopn of JO",t Mnowk granitoidJ in southast""
China. In: Xu T., Tu G. (Eds.), Geology of Granites
and their Metallogenelic Rdations. Science Press,
Beijing, China, 90'·919.