Mineral. Deposita 28, 240-252 (1993)
O ep'one. ralium slta
© Springer-Verlag 1993
The Ribeira fluorite district, southern Brazil Geological and geochemical (REE, Sm-Nd isotopes) characteristics L.H. Ronchi 1, J.-C. Touray 1, A. Michard 2, and M.A. Dardenne 3 1 Universit6 d'Orl6ans, ESEM, GdR 0969 et URA n°1366 du CNRS, Rue Leonard de Vinci, F-45072 Orleans Cedex 2, France 2 Universit6 d'Aix-Marseille III, Facult6 des Sciences et Techniques de St-J6r6me, Case 431, F-13397 Marseille Cedex 13, France 3 Universidade de Brasilia, Instituto de Geoci~ncias, 70.910 Brasilia, DF, Brasil Received: May 8, 1992/Accepted: April 1, 1993
Abstract. The origin and evolution of different ore deposits grouped in the same district are often complex and may involve inheritance from crustal or mantle geochemical anomalies, remobilization of former ore deposits and a polyphase hydrothermal history. Localized in a Proterozoic basement in the Parana state, the Ribeira fluorite district is such an example composed of three deposit types with distinct geological and geochemical characters. Emplaced at different periods from the late Proterozoic to the Cretaceous, they are roughly aligned along a belt nearly 10 km in width and 50 km in length, the southern boundary of which is a transcurrent fault. Two main ore facies are present: (1) microcrystalline ore ( < 0.1 mm grains) and (2) macrocrystalline ore (with a grain size of several millimetres). The former results from the replacement of metalimestones or internal karstic sediments and the latter from microcrystalline ore dissolution and pore precipitation or recrystallization. At least two different groups of source rocks can be proposed for the trapped REE in CaF2: (1) fluorite samples associated with the Mato Preto carbonatitic rocks display a slightly negative eNd compatible with a mantle source and a REE pattern with the higher ZREE and La/Yb ratio in the district; (2) other fluorites have a strongly negative sNd ( - 14 to - 20) which indicates a crustal source. That fluorine and REE have the same source is possible in strata-bound and fracture-filling deposits, but is doubtful at Mato Preto, the only economic fluorite deposit associated with carbonatite rocks in Brazil. This occurrence within a Precambrian fluorite belt suggests that remobilization of a former strata-bound deposit was a more significant metallogenic process than magmatic differentiation.
systems in Santa Catarina State, Brazil (Bastos Neto et al. 1991), is about 100 km in length. All the deposits belong to the low temperature fluorite + quartz vein type and were formed during a relatively short time interval, related to the pre-Southern Atlantic rifting (Bastos Neto 1990), and apparently from the same granite source-rock. More complex explanations are required when ore deposits grouped in the same belt are of different types and different ages (heterotypic and heterochrone, Routhier 1980). In this case, inheritance from crustal or mantle geochemical anomalies or remobilization of former ore deposits may be suspected. Most of the studies concerning mineral deposit belts have been based upon geological arguments. In the present article, a geochemical discussion, dealing with REE patterns and Sm-Nd systematics, is presented for fluorite deposits found in a linear distribution extending about 50 km in southern Brazil. In the Ribeira River Valley, six fluorite deposits are roughly aligned along a narrow E N E belt of less than 10km width delimited on its southern boundary by a transcurrent fault, the Ribeira lineament (Fig. 1). This 'fluorite trend' is composed of three economic or subeconomic strata-bound deposits (Volta Grande, Mato Dentro and Sete Barras) hosted by Proterozoic metalimestones, a major deposit related to an alkaline complex with carbonatites (Mato Preto), a non-economic occurrence of disseminated fluorite in carbonatites and metalimestones (Barra do Itapirapu~), and finally a fracture-filling non-economic deposit (Braz). Only the Mato Preto and the Volta Grande deposits are mined at present.
Regional geology The geographical distribution of ore deposits in continental crust is often along narrow belts or trends. The origin of these trends is sometimes dearly dependent on the geodynamical framework. For example, the 'Canela Grande' lineament, grouping five economic fluorite vein
The most prominent structural features in the region, as observed from LANDSAT 5 image n ° WRS 220/077C TM5-02080-S005 (23 MAY 86 21:11), are the Langinha and Morro Agudo transcurrent faults (Fig. 1), which limit three great tectonic blocks. The Ribeira lineament is considered a branch of the Morro Agudo fault (Fig. 1, Fiori
241 + ÷ + + ÷@@÷ +÷@@ +@÷@ + ÷ ÷ ÷ @+÷÷ ÷ + ÷ ÷
Brazil
÷ ÷ ÷ + + @ + + ÷ + ÷ + ÷ ÷ @ ÷ ÷ ÷ ÷ @ ÷ ÷ ÷ ÷ + ÷ ÷ ÷ + + @@÷+÷@ ÷ = * = ÷
• Apiai
÷ ÷ + + + ÷
4"4.+ •~ '1, ÷ + @ +'1' + + ~ ++4'+ @4.@@@ @+@@4. @ ,I..I. @ ,I.
+
::::: +@@@ @ @ @ @ ÷
@
@@÷@÷ ÷@@÷÷
+*+*+1~@
.....
X~
,@ ÷ @ @ @
+ + + ÷ @ + ~÷@+@+ @+@+÷÷ @@÷~+@
.............
::@
÷ ÷ + ÷ @ + + ÷ @ @ ÷ @ ÷ ~
.~ ~ ÷ + + ÷ ÷ ÷ + ÷ ÷ + + + + @ + ! i @ ÷ ÷ @ 0 ÷ ~ @ ÷ @ ~ ÷ ÷ ÷ ÷ + +
,I, @ @ @ ,J, ÷ ÷ @ @~ @ @ @ ÷
÷ @ ,~ ÷ ÷ @÷'@@@~ ÷@@@+ @ @ @ ÷+@÷@@@
@ ÷ + ÷ @ ÷ ÷ ÷ + ÷ @ + @ @ ÷ ~ + @ + ÷ ÷ + ÷ ÷ ÷ + ÷ ÷ @ @ @ ÷ @ ÷ ÷ + + + ÷ ÷ ÷ + ÷ @ + + ÷
PCa
~Rfb6ira
•
/ X ~**'*'X X XX ? X X .X .X . X. .X . X X :¢
IDfj i
X X X X X X X X X X X X X .X .X . X . X. x. .X .X X *:¢* X X
X X X X X X X X X X X X X X X X X X X X X X :,¢"X X X X X X X
g:::12
X X X X X
,1. @ @ + . l . @ ÷ ~ p p ~ + ÷ ÷ ÷ @@4. @ @ ,1. @ . l . ÷ o +÷.l.÷+~
2
.
@ @ 4 . , I ,
@ @ @ @@,1,4.4.@4.@@ @ o ,t, •@ @ ,1~ 0 @ 4 . @
PCa
BR
MP k ,+ +
+
+
.
+
,
+
÷XK +
+
.b}
C6rr
4.@/
PCa
::7
Tc
:f
x
Cretaceous intrusive rocks (kk)
I
0
x x
Post-tectonic granites ('~, 7 1) Syn-tectonic granites ('y rc) I Undifferentiated Proterozoic Agungui and Setuva Groups (PCa)
lOkm
~77~] River ~;t ~:]
Transcurrent fault Geologic contact
Fig. 1. Regional geologic setting. Fluorite deposits: VG Volta Grande; M D Mato Dentro; SB Sete Barras; M P Mato Preto; BR Braz; BI Barra do Itapirapu~. 1 Morro Agudo transcurrent fault; 2 Ribeira lineament; 3 Lan~inha transcurrent fault; 4 C~rro Azul fault
et al. 1984). Another roughly E-W fault, called C~rro Azul, is also known (Felipe and Oliveira 1986), and at least four fluorite deposits are directly associated with either the Ribeira or the C~rro Azul lineaments. The present regional geology results from three main known tectonic events (Fiori 1991) of Proterozoic age. Earlier work that ignored these tectonic features led to at least 21 different stratigraphic columns. It is now possible to recognize, from the oldest to the youngest, the following lithologic groups: 1. Precambrian rocks: mainly migmatites, schists and metasediments (Costeiro Complex, Setuva Group and Aqungui Group). 2. Syn-tectonic granitoid batholiths (Tr~s C6rregos Complex, dated between 510-640 Ma, Issler and Freire 1985). Post-tectonic granitic stocks including alaskitic facies and
alaskitic dykes in the age range of 500-600 Ma (for example Itadca granite, aged 500 _ 15 Ma: Cordani and Bittencourt 1967; Issler and Freire 1985; Hasui et al. 1984). Field observations (alaskitic dykes cutting Tr~s Cdrregos granitoid) strongly suggest that the alaskitic dykes are younger than the syn-tectonic granites (Ronchi 1986). 3. Mesozoic magmatic intrusive rocks (Jurassic dolerite dykes, carbonatite and other alkaline Cretaceous rocks). Phonolitic dykes that cut the carbonatite complex of Mato Preto were dated by Cordani and Hasui (1968) and showed minimum ages of 65.2-t-3.3Ma and 67.0 + 3.4 Ma using the K-Ar method. The exact age of carbonatites is hence not known at Mato Preto. With respect to dating of intrusives from the alkaline province in Sho Paulo, Rio de Janeiro and Minas Gerais State, a Cretaceous (65-135 Ma, Cordani and Hasui 1968) age may be inferred.
242 Recent K-Ar data on illite + kaolinite + smectite (Dos Santos and Bonhomme 1991 and oral communication) indicate at Mato Dentro apparent ages of 402-397 Ma and younger ages at Sete Barras (255-268 Ma). Although clays are clearly younger than the fluorite ores, these data confirm the pre-Cretaceous age of these strata-bound fluorite deposits. The Mato Preto deposit is directly related to Cretaceous intrusive alkaline rocks and hence of Cretaceous age. No geochronological data are available at Volta Grande and Braz.
Characterization of the main fluorite deposits A summary of the principal characteristics of the Vale do Ribeira fluorite deposits is available in Table 1 and a brief description of the essential geological data is presented below.
layering (So- 1) of the metalimestone. Pontes (1980) observed that fluorite veins have a variable width from millimetre to metre-scale. He has also described, in metalimestones, replacement fluorite and disseminated fluorite associated with recrystallized calcite. Some fluorite veinlets locally contain tourmaline. These veinlets are crosscut by millimetre-scale muscovitefilled fractures. Alaskitic dykes, ranging up to several metres in width, cut the ore, forming breccias with fluorite-rich fragments in a granitic matrix. The fluorite + tourmaline association is related to a hydrothermal process, probably later than the emplacement of the syn-tectonic granites, the presence of tourmaline suggesting an early post-granitic hydrothermal event. On the other hand, this paragenesis is older than the alaskitic granite dykes. The available K-Ar dates (see above) suggest ages in the range 500-600 Ma for the different granites. Hence it is possible to suggest the same approximate age for the fluorite occurrence at Braz (550 + 50 Ma).
Mato Preto
The hydrothermal fluorite mineralization is associated with an alkaline igneous complex which was situated at the intersection between a northeastern (N60-70E) lineament (C~rro Azul fault) and the N10E Morro Agudo transcurrent fault, regionally forming the contact between the Tr~s C6rregos granitoid and the A~ungui Group (Jenkins II 1987). The Mato Preto carbonatite complex is composed of four main circular structures of about 1 km in diameter. Three of them consist largely of fenitized syenites with minor proportions of calcitic carbonatite, phonolite and ultramafic rocks. The fourth (and possibly youngest) structure, comprises mainly phonolites with xenoliths and explosive breccias (Jenkins II 1987). The mineralization occurs in three principal ore bodies; however numerous minor occurrences are also known. The largest body is the Clugger deposit (2.65 MT, 60% CaF2), consisting of four main subparallel and coalescing lenses. According to Santos and Dardenne (1988) silicification and argillization are the main hydrothermal alterations related to mineralization. They determined the following successive fluorite generations: (1) disseminated, veinlet and massive concentrations of black and violet fluorite; (2) massive colourless to brown fluorite; and (3) late remobilized massive fluorite (white, blue-violet and yellow-colourless). The main ore is made up of the massive microcrystalline and macrocrystalline colourless fluorite, replacing calcite within carbonatites. Remobilized fluorite ore usually fills up pores of the latter. The microthermometric data for these ore types, as described by Santos (1988), are summarized in Table 2. Braz
The Braz occurrence is located on the N60-70E Ribeira lineament and hosted by a metalimestone belonging to the Aqungui Group. Violet and green fluorite crystals are found filling parallel fractures which are vertical with a N65W strike almost perpendicular to the N30E, vertical
Barra do Itapirapu~
A carbonatite body intrusive in the Tr~s C6rregos granitoid is exposed at Barra do Itapirapu~ and near the contact with the Setuva Group metasediments. The fluorite is found, in an uneconomic disseminated form, similar to the disseminated ore from Mato Preto. At least four minor fluorite occurrences, replacing Setuva metasediments are present in the vicinity. These occurrences are probably of pre-alaskitic granite age, and similar to the ore types described at Volta Grande.
Sete Barras
Fluorite occurs as discontinuous flat lenses hosted by metalimestones belonging to the Setuva Group. This irregular strata-bound ore body (2.5 MT, 50-60% CaF2) has an overall length of 2 km, and is 15-20 m in width and nearly 150 m in depth, with a N60-70E direction, concordant with the So-1 foliation. Folds have been described in the fluorite ore (Fagundes et al. 1984) but it is not clear whether they result from a syn-tectonic episode or are related to local disharmonic deformations related to the later granite intrusion. The host metasedimentary sequence and the ore are cut by the porphyritic posttectonic Ita6ca granite. This contact is locally masked by the Ribeira fault, but it is revealed by hornfels outcrops. In the ore zone, fluorite and country rock xenoliths have been described within the granite (Fagundes et al. 1984). Two main ore types are present at Sete Barras: 1. The microcrystalline black ore is composed of finegrained fluorite (less than 0.1 mm) and quartz in variable proportions (40-90% CaF2). Accessory phases include muscovite, carbonates, iron oxides and other opaques. This ore displays replacement textures sometimes with remarkable layered structures (Fagundes et al. 1984) suggesting karstic sediments. Apparently, both the host metalimestone and calcitic 'internal' sediments have been
243
~.~
..
-
+o
F, ~.-=
0
o~
~
g :=
o
~
-~
• o
e~ 0
~,.o
. ~= m
~ + "~'
0
~
'x::l
".~
e~ o
~+~+o
ao~
e
~
._,9.
~+ ~.~ +~
r~ .~.~a"
~ o
~
o o
~:~ ~+
N 0 e~
0
0
Z
o "~ ..,-
=+++
,-~
-~
.~-
0 0
~.
.~
..
=
o
o
a
=oo
r~
o~
~2
.=, ¢.2 o'~
"~
o
.~ a
~ ~ .~ ~ ~ ~ ' ~~
~
~
~
.~
n~
~
o
0
0
-9,
e~
_o =
o=
...-, e',
. . - o ~ .~
E +++++
.,..,
~
o
-
~
~
~
0. N
eq
~2 ell k.)
,~...-+
0
•~ = ~
..
-~,'n ~
.-.~
,. ~ ~,.~
==
~o=o
~ "= o= = o .~..~ ~
.-.= ,. ~',= ~,[-.
o ~, o
ech
F.,
801 0
+++
~
~ - ~ d ~
~
0.
0 0
o
~
~
"-= ~
"~ ell
J
[..,
244 Table 2. Microthermometric data of the fluid inclusions of principal fluorite ores. aNo detectable fluid inclusions; bMetastable positive melting temperatures ( + 0.6 to + 4.0°C) Ore type
Deposit TH
°C
Sal% wt eq NaC1
Min Max Min Microcrystalline ore VG MD SB Macrocrystalline ore VG MD SB
90 a 70 85 115 80
Max
140
0
3.5
100 168 150 120
b 0 0
4.0 4.0
Mato Preto (data from Santos 1988) Colourless fluorite Disseminated fluorite Violet remobilized Colourless remobilized
Only decrepited inclusions 80 120 0 7.5 220 80
260 120
0 12
fluid circulation of unknown age. Aqueous two-phase fluid inclusions from macrocrystalline fluorite give homogenization temperatures in the 110-150°C range and salinities of about 2 - 4 % wt eq NaC1 in spite of some metastable positive ice melting temperatures (Table 2).
11.0 15.0
replaced by fluorite. Fluid inclusions display homogenization temperatures in the 80-90°C range and metastable (positive) ice melting temperatures (Table 1). 2. The macrocrystalline ore is made of millimetre-scale to centimetre-scale crystals of an almost pure yellow fluorite. No observable fluid inclusions have been detected. A priori this ore could have formed by the recrystallization of microcrystalline fluorite in the granite contact, or by a dissolution/reprecipitation process possibly caused by fluid circulation associated with the Ita6ca granite emplacement. Alternatively, this remobilization and even some addition of fluorite, could be attributed to fluid circulation associated with the Cretaceous magmatic event (Silva et al. 1981), as suggested in this article.
Mato Dentro
This strata-bound fluorite deposit, located 10 km north of the Ribeira lineament, shows little deformation and outcrops nearly 200 m away from the Ita6ca granite contact. Accordingly it is the least modified of the Vale do Ribeira strata-bound deposits. Two ore bodies with reserves of about 1.5 M T CaF2 have been recognized from drillings. They lie concordantly within a metalimestone horizon of the Aqungui Group. As observed in drill cores this metalimestone displays a folded S o - 1 foliation with local crenulation and oblique $2. The metalimestone replacement by fluorite is post-deformational. A microcrystalline ore with similar textures to those observed at Sete Barras is the predominant type. It is composed of fluorite and quartz in various proportions associated with subordinate muscovite, calcite dolomite, pyrite, iron oxides and other opaques. There are no fluid inclusions appropriate for microthermometric studies in this ore type. Later macrocrystalline fluorite, karstic breccias and calcitic veinlets with remobilized fluorite and pyrite were formed after fracturing and hydrothermal
Volta Grande
This strata-bound deposit is composed of three main ore bodies and seven known minor occurrences. Body I (0.6 MT tonnes, 35% CaF2), and body II (0.4 MT, 40% CaFz), are enclosed in irregular roof pendants of silicified metalimestones and schists located within the late Proterozoic syntectonic Tr~s C6rregos granitoid. These bodies, in addition, are intruded by an alaskitic granitic vertical ramified dyke. The smaller body III (0.1 M T 38% CaFz), occurs concordantly in a larger metalimestone roof pendant over the Tr~s C6rregos granitoid and similar to the one observed at Mato Dentro and Sete Barras. Satellite images and field studies show that body I is located at the intersection of two faults. One of these, the C~rro Azul fault (Fig. 1), can be extended to the Mato Preto deposit. This lineament meets the Morro Agudo fault at Mato Preto and corresponds exactly to a radioactive Th-rich anomaly detected by Nuclebras (unpublished report). The yellow microcrystalline ore of body I is roughly homogeneous, usually silicified, and fractured. It is sometimes tectonically brecciated with a cement of macrocrystalline violet and colourless fluorite and quartz (Ronchi et al. 1987). The primary aqueous two-phase fluid inclusions found in these two types of fluorite display the same homogenization temperatures of 100-150°C, and similar salinities in the 0 - 5 % wt eq NaC1 range with frequent metastable ice melting temperatures (Ronchi 1986). These low salinities preclude the possible role of basinal brines in the genesis of the ore (Touray 1989). A late Ba-rich adularia + adularia + quartz + fluorite + barite paragenesis is present. The barite also occurs as late fracture fillings cutting the microcrystalline ore and as minor isolated occurrences associated with the C~rro Azul fault. Lying near and on the south side of the principal (N70E) C~rro Azul fault, body II shows no evidence of strong tectonic brecciation. A fluoritized karstic sediment with oblique stratification, irregular angular fragments and lateral facies variations is dominant. Discordant recent karstic breccias with microcrystalline or barite cement were observed. Fluorite aqueous two-phase fluid inclusions gave the same results as described in body I. Recrystallization textures of the fluorite in contact with the Tr~s C6rregos granitoid and a gradual replacement of the enclosing dolomitic metalimestone by fluorite are visible in drill core samples from body III.
Conclusion
Carbonate replacement and open space filling textures have been observed in all the fluorite ores including those related to magmatism at Mato Preto. From grain-size differences it is possible to distinguish microcrystalline
245 (lower than 0.1 mm) from macrocrystalline (crystals several millimetres in size) fluorite ores in s t r a t a - b o u n d deposits. This distinction is due to differences in ore genesis: microcrystalline fluorite is formed after replacement of former calcite-rich internal karstic sediments with distinct layering and of h o m o g e n e o u s metalimestones in the strata-bound deposits. Macrocrystalline fluorite results from dissolution and reprecipitation in open spaces or recrystallization of former microcrystalline ores. F r o m the above geological data it is possible to distinguish three basic types of fluorite deposit in the Vale do Ribeira region: (1) strata-bound, (2) carbonatite associated, and (3) fracture filling. The strata-bound and fracture-filling deposits intruded by the alaskite granite dykes of a b o u t 500 M a age, are late Proterozoic or slightly older in age while the carbonatite-associated deposits are sync h r o n o u s with or later than their Cretaceous host rocks.
Geochemical
studies
Previously published R E E analysis by I N A A (Commissariat /t l'Energie A t o m i q u e - Orsay, France) or I C P ( G E O S O L - Belo Horizonte, Brazil, Ronchi and Dardenne 1987; Santos 1988) have been used in conjunction with new I C P data ( C R P G - Nancy, France and B R G M - Orlrans, France) for M a t o Dentro, to characterize the different fluorite ores. To m a k e possible c o m p a r i s o n between I N A A and I C P analyses, reference to total R E E (Y~'REE) contents means La + Ce + Sm + Eu + Yb + Lu, because only these elements are determined in c o m m o n by the two methods. S m / N d isotopic analyses have been performed at the C R P G , Nancy. The R E E analytical results (ICP and I N A A ) for the fluorite ores and c o u n t r y rocks are shown in Tables 3, 4 and 5. With respect to analytical accuracy we think that R E E contents below 0.1 p p m m a y be in error. Accordingly, the normalized '0.1 p p m curve' has been drawn Table
1 3 3 1 1 1 1 2 2 2 2 2 2 1 1 1 2 2 1 2 1
when necessary to d r o p out data points localized below this curve. In this respect, we note that the Yb negative a n o m a l y sometimes present is p r o b a b l y erroneous.
R E E patterns All the samples from the s t r a t a - b o u n d (VG, SB, M D ) deposits displayed almost the same fiat R E E pattern for the two main ore types (microcrystalline and macrocrystalline), with limited variations (Fig. 2). F o r example the h o m o g e n e o u s microcrystalline and macrocrystalline ore of Volta G r a n d e and Sere Barras are slightly H R E E enriched and the Volta G r a n d e stratoid, M a t o D e n t r o microcrystaUine and macrocrystalline ores are slightly L R E E enriched. E ' R E E contents for all the strata-bound ore scatter in the range from 5 to 60 ppm. The Barra do Itapirapu~ stratoid microcrystalline ore and the Braz fracture-filling ore display the same flat pattern and global R E E contents as observed in M a t o Preto remobilized ores and the same flat pattern that characterizes the s t r a t a - b o u n d ores, but with significantly higher total E ' R E E contents (Fig. 3). The granite and metalimestone patterns are roughly parallel, as shown in Fig. 4, the E ' R E E granite being higher than for metalimestones. The Tr~s C6rregos syntectonic granitoid and the I t a 6 c a post-tectonic granite exhibit superposing patterns with a slightly negative Eu anomaly. The post-tectonic alaskitic granite also displays a parallel pattern characterized by a slightly positive Eu a n o m a l y and lower global R E E contents. The fenitized ( M P G r F ) and h y d r o t h e r m a l l y altered (VG 96ii) Tr~s C6rregos granitoid, locally present as host rocks, are slightly H R E E enriched (Fig. 4). F o r strata-bound ores and remobilized calcite for M a t o Dentro, R E E patterns are oblique to the granitoid and metalimestone patterns. The Yb, and possibly the Eu anomalies of the metalimestone, are p r o b a b l y analytical errors because they are below the
3. REE analytical data (neutron activation analyses) (1) microcrystalline ore; (2) macrocrystalline ore; (3) metalimestone replacement ore
Braz B1105 B1107 VG23 VG 51 VG 53 VG 54 SBX1 SB X2 SBX3 SB X4 SB X41 SB X44 SB 79 SB 80 SB 81 SB 82 SB83 SB 85 SB 88 SB 89
La
Ce
Nd
Sm
Eu
Tb
Dy
Yb
Lu
124.28 20.13 73.20 15.00 2.80 3.80 5.80 7.00 12.50 2.66 0.90 5.50 11.20 11.50 10.40 8.40 7.40 11.60 40.70 4.80 19.50
50.48 25.50 110.30 32.80 12.80 9.40 20.60 14.20 27.20 5.10 2.30 12.00 23.70 25.70 15.30 18.10 25.00 24.50 61.80 10.10 25.50
17.10 5.90 3.70 13.00 -
41.29 5.15 7.15 3.51 2.17 1.17 4.58 1.16 3.22 0.90 2.80 1.20 2.60 3.48 2.51 1.88 3.34 3.27 7.20 1.66 3.80
27.58 1.37 1.62 1.47 0.92 0.43 2.50 0.03 0.99 0.20 0.05 0.46 0.96 1.27 0.75 0.55 1.17 1.15 1.89 0.44 1.12
1.63 1.30 1.40 1.15 1.00 0.21 2.18 0.17 0.80 0.10 0.03 0.30 0.50 1.00 0.36 0.30 1.10 0.79 1.17 0.42 0.80
7.49 7.91 -
53.30 4.13 4.65 12.20 9.20 4.80 16.70 0.88 5.70 0.54 0.17 1.19 2.50 5.40 2.13 1.27 12.40 5.30 3.90 1.93 3.80
61.83 0.53 0.62 2.10 1.72 0.87 2.80 0.17 1.02 0.10 0.03 0.20 0.44 0.90 0.36 0.24 2.13 0.92 0.66 0.37 0.66
246 Table 4. REE analytical data (ICP) ii = CRPG (France) analyses, MPLHIa, b BRGM (France) analyses, others = GEOSOL (Brazil) analyses. ~Samples also analyzed for Sm-Nd isotopes. (1) MicrocrystaUine ore; (2) Macrocrystalline ore; (3) Colourless bulk sample; (4) Isolated fluorite; (5) Violet disseminated ore; (6) Remobilized ore La 1 1 1 2 2 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 2 2 1 1 1 3 4 5 5 5 5 5 5 3 3 3 3 6 6 6 6
MD 1 MD4b MD 4b ii MD 4a MD4a ii MD 17" MD 18 MD 18a ii MD 27 MD41 a SB66 i? SB 10 SB 15 SB 24 SB 30 ii SB 34a i? SB 34b i? VG 10 VG 17 VG27 VG 34 VG 38* VG 52 VG66 VG 87 VG94 VG 119 MPLHla a MPLHlb a MP 1 MP 2 Mp 3 MP4 MP 5 MP6 MP7 MP 8 MP 9 MPLH92 MP 10 MP 11 MP 12 MP 13
Ce
Nd
2.04 4.04 2.02 11.80 13.10 6.10 5.37 17.57 6.96 13.20 16.70 7.70 4.63 18.12 6.66 7.65 7.30 9.08 2.65 4.36 1.95 5.23 25.04 5.35 2.50 3.08 1.00 6.11 10.01 5.17 11.72 31.29 13.40 3.39 6.24 3.26 1.93 3.42 1.80 4.21 5.42 1.31 5.42 19.66 5.54 5.51 18.34 5.94 4.51 14.21 5.05 7.32 12.50 7.22 9.33 16.80 10.27 4.96 8.80 6.08 9.68 16.85 8.65 11.31 19.35 10.17 6.56 13.44 9.08 4.20 11.10 10.50 17.70 31.50 19.80 18.60 18.10 17.40 5.91 11.31 9.12 492.40 1 0 4 0 . 0 0 448.60 23.50 43.00 23.60 1 5 1 1 3 . 0 0 88869.00 4 4 1 6 . 0 0 2881.00 3 0 1 7 . 0 0 550.20 699.00 1 1 7 8 . 0 0 411.80 167.10 2 1 2 1 . 0 0 670.40 761.80 1 2 4 1 . 0 0 407.60 126.30 183.10 65.40 126.30 183.10 65.40 2122.00 2 5 2 2 . 0 0 421.00 501.30 841.50 439.30 928.70 1 5 8 4 . 0 0 539.30 55.10 78.80 32.40 35.10 21.90 50.10 15.80 28.10 26.10 13.90 25.20 14.10
Sm
Eu
Gd
Dy
0.60 0.69 1.92 1.60 1.95 1.55 0.40 1.10 0.21 1.12 3.93 0.55 0.35 0.22 1.22 1.61 1.75 1.79 2.51 1.68 2.18 2.77 3.06 3.60 5.00 3.70 3.45 69.30 5.30 453.10 59.30 67.50 96.90 69.10 10.20 10.20 35.80 50.20 93.10 6.80 9.50 7.90 4.30
0.10 0.23 0.55 0.29 0.57 0.28 0.086 0.46 0.038 0.20 1.05 0.14 0.08 0.04 0.29 0.37 0.34 0.14 0.50 0.36 0.56 0.83 0.98 1.10 0.96 0.61 1.14 18.10 1.90 94.30 14.40 17.30 24.60 21.30 2.70 2.70 6.90 14.20 17.80 2.50 3.00 3.30 2.10
0.58 0.95 1.80 1.30 2.35 1.059 0.37 1.45 0.18 0.62 4.41 0.42 0.29 0.20 1.06 1.43 1.76 1.43 1.75 1.48 2.09 3.38 3.94 4.40 3.00 1.70 4.93 41.60 6.90 241.40 49.80 43.80 66.60 71.60 7.20 7.20 24.60 31.50 45.90 9.10 8.50 11.70 8.60
0.65 0.76 2.05 0.80 2.02 0.62 0.28 0.91 0.20 0.42 6.54 0.39 0.20 0.19 0.77 2.26 3.33 2.11 1.76 2.01 3.23 5.87 6.20 7.30 3.60 1.70 8.89 31.60 11.10 81.50 38.20 27.00 29.60 47.20 6.40 6.40 8.90 14.10 24.10 13.00 8.90 25.50 11.90
0.1 p p m reference curve, b e n e a t h which analytical errors m a y be highly significant. At M a t o Preto, as described by Santos (1988) three types of ore spectra were identified (see 1, 2 a n d 3 in Fig. 5). 1. The most L R E E - r i c h sample of the c a r b o n a t i t e disseminated violet-black ore ( M P l v ) was collected at a different occurrence (called M a t o Preto I), distinct from the principal Clugger deposit, a n d is characterized by strong radioactive a n o m a l i e s (the sample shows 1560 p p m of Th). The positive Ce a n o m a l y m a y be related to some specific Ce-rich phase (e.g. R E E minerals). 2. The violet disseminated a n d massive ore a n d the colourless to b r o w n c a r b o n a t i t e replacement ore display the same general spectra with a L R E E - e n r i c h e d a n d H R E E relatively depleted p a t t e r n similar to the R E E patterns of
Ho 0.14 0.14 0.15 0.12 0.06 0.041 0.18 0.09 0.05 0.04 0.31 0.26 0.58 0.68 1.28 1.30 1.60 0.61 0.29 1.93 5.70 2.50 17.40 6.80 4.80 5.10 6.90 1.30 1.30 1.90 2.60 4.30 2.70 1.90 6.50 2.00
Er
Yb
Lu
0.43 0.32 1.19 0.33 1.09 0.26 0.14 0.58 0.12 0.40 4.50 0.25 0.13 0.09 0.59 1.53 2.37 1.19 0.94 1.49 2.17 4.03 3.70 4.80 1.70 0.91 5.38 13.30 7.90 38.70 16.60 12.30 11.90 13.70 3.90 3.90 4.70 5.40 9.30 7.80 5.70 24.20 5.50
0.27 0.17 1.09 0.21 0.93 0.17 0.10 0.50 0.012 0.10 4.39 0.21 0.02 0.02 0.50 1.65 2.34 1.65 0.87 1.37 2.89 4.63 4.12 4.80 1.50 0.93 5.67 9.60 9.90 15.40 12.00 9.60 7.30 6.60 3.50 3.50 2.50 2.40 4.00 7.90 5.30 22.70 5.00
0.045 0.07 0.20 0.10 0.07 0.047 0.02 0.01 0.015 0.05 0.63 0.04 0.02 0.02 0.11 0.27 0.36 0.23 0.11 0.19 0.43 0.66 0.58 0.62 0.16 0.16 0.76 1.20 1.40 1.70 1.50 1.20 0.85 0.70 0.40 0.40 0.40 0.30 0.43 1.00 0.70 2.70 0.70
the c a r b o n a t i t e s (Figs. 5 a n d 6). O n e m a y guess that most of the R E E c o n t e n t is n o t c o n t a i n e d by fluorite b u t by the c a r b o n a t i t i c gangue. This hypothesis has been checked by the following analysis: for one ore sample, R E E concentrations in the b u l k ore ( M P L H l a ) a n d in the m e c h a n ically separated fluorite ( M P L H l b ) have been compared. The former displays a similar R E E p a t t e r n to those of the violet a n d colourless ore (Fig. 5), while fluorite has a spect r u m n e a r the h o r i z o n t a l p a t t e r n of the M a t o Preto remobilized ore, Braz, Barra do I t a p i r a p u ~ a n d stratab o u n d deposits (Fig. 2). Accordingly, m o s t of the R E E c o n t e n t in the colourless fluorite ore is c o n c e n t r a t e d in R E E minerals such as monazite, bastnaesite a n d r h a b d o p h a n e (Jenkins II 1987). 3. The remobilized ores p a t t e r n present large variations when c o m p a r e d with the isolated fluorite spectrum
247 Table 5. Analytical data of the host rocks: (1) Metalimestones; (2) Unaltered Tr6s C6rregos granitoid; (2a) Fenitized Tr~s C6rregos granitoid; (2b) Hydrothermally altered Tr~s C6rregos granitoid; (3) Ita6ca granite; (4) Isolated fluorite; (5) Spathic clacite associated with macrocrystalline ore in Mato Dentro; (6) MP carbonatite; (7) MP phonolite, (8) MP syenite
1 1 1 1 2 2a 2b 3 3 4 5 6 6 6 7 8
MD23 MD39 MD49 SB26 MD 89 TC MP TC f VG96TCh MD 83 IT SB 57IT VG 29 A MD 12 MP carb 1 MPcarb2 MP carb 3 MPfon MPSyef
La
Ce
Nd
Sm
Eu
Gd
Dy
3.350 3.610 3.820 3.200 51.81 73.30 72.38 71.00 63.04 17.07 0.800 187.00 935.30 333.50 101.50 154.70
4.440 5.210 5.840 3.910 86.58 141.30 118.43 100.67 102.52 32.91 1.670 278.00 392.00 544.90 161.80 229.50
1.530 1.530 1.900 1.090 35.75 41.60 47.85 48.18 39.37 10.78 1.710 107.60 52.30 218.90 52.30 75.10
0.300 0.360 0.450 0.240 6.50 7.30 9.33 8.14 7.31 1.86 0.400 15.90 79.60 39.80 7.20 13.20
0.051 0.054 0.064 0.038 1.35 1.90 2.02 1.93 1.35 0.69 0.069 3.80 20.20 11.30 2.00 4.10
0.270 0.310 0.250 0.160 4.85 5.80 6.07 4.85 5.56 1.32 0.320 10.10 51.10 30.50 5.50 11.80
0.130 0.250 0.130 0.200 2.20 3.40 7.13 2.06 2.43 0.69 0.290 8.10 20.40 21.80 3.60 7.40
1000
E
=
Ho 0.050 0.050 0.057 0.040 0.48 0.07 1.80 3.30 4.00 0.65 1.30
Er
Yb
Lu
0.130 0.110 0.110 0.130 1.01 1.60 3.64 1.06 1.09 0.38 0.18 4.80 6.40 9.40 2.00 3.70
0.017 0.013 0.015 0.011 0.78 1.10 3.34 0.76 0.90 0.47 0.11 2.70 2.00 4.00 1.95 1.90
0.014 0.014 0.011 0.016 0.140 0.140 0.560 0.390 0.110 0.110 0.031 0.300 0.250 0.400 0.300 0.210
Reference curve (0.1 ppm)
100
Jt
MD 4aii (macro ore)
Jt
MD 4bii (micro ore)
A
MD 83ii (Ita6ca granite)
I.immlOmmm
MD 39 (metalimestone)
10
SB 34aii (macro ore) SB 34bii (micro ore) VG 17e (micro stratoid ore) E
VG 38 (micro homogeneous ore)
r~
,1
,01
..... ~o=o,
i
I
I
La Ce
I
I
Nd
I
I
I
I
SmEu Gd
I
I
I
Dy Ho Er
I
~0
VG 52 (macro ore)
I
Lu
ETR Fig. 2. REE pattern of the microcrystalline and macrocrystalline ores from strata-bound deposits compared to the 0.1 ppm chondrite normalized curve and metalimestone spectrum
( M P L H l b ) a n d the s t r a t a - b o u n d deposits ore, b u t their shape is r o u g h l y horizontal, with a H R E E e n r i c h m e n t a n d a Ce negative a n o m a l y (Fig. 5). It seems clear t h a t R E E rich-minerals are a b s e n t from these ore types. S a n t o s 1988, following S t r o n g et al. 1984, explains the Mato Preto LREE depletion and HREE enrichment by a d i s s o l u t i o n - r e p r e c i p i t a t i o n process with a decrease of fluorine activity in the fluids. The c h a n g i n g of R E E from
d i f l u o r o to m o n o f l u o r o complexes resulted in a decreasing a b i l i t y of the h y d r o t h e r m a l solutions to t r a n s p o r t L R E E with respect to H R E E (higher stability of H R E E m o n o fluoro c o m p l e x e s - S t r o n g et al. 1984). The o b s e r v e d H R E E e n r i c h m e n t also described in V o l t a G r a n d e s t r a t o i d ore, a n d M a t o D e n t r o ores is due to a similar process linked to new fluid c i r c u l a t i o n p r o b a b l y c o n t r o l led b y the R i b e i r a - C ~ r r o Azul l i n e a m e n t a n d it is easily u n d e r s t o o d because the M a t o D e n t r o d e p o s i t a n d the
248
10000
"" E
lll|eO|lll
MPLH1 a (total colourless ore)
II|||~lll|l
MPLHlb (colourless fluorine)
0""0 ........ "0 ......... ~)"O"E)
1000
&
0,. 0,.
MD 4bii (micro ore) SB 34bii (micro ore)
= ,,,=
1 O0
0.o,
~0~*m, •,. "::.... ......
"+ .....
---U--
...... 0 .....
:;"-'.".-:".'""
' VG 17 (micro stratoid ore)
- --o--
VG 38 (micro homogeneous ore)
..... ÷ ....
IT F105 (replacement ore)
. . . . N---
Braz (microore)
....
CI.
1
,1
I
I
I
I
La C e
I
hid
I
I
I
I
Sm Eu C-d
I
I
I
I
Dy Ho Er
I
I
Yb Lu
ETR Fig. 3. REE pattern of the strata-bound deposits ore compared to fracture-filling and carbonatite-associated ores
1000 Reference curve (0.1 ppm) w
E
100
SB 26 Metalimestone
.........•~,,°° .....
MD 89ii Tr~s C6rregos granitoid
-----~----
MD 12 Calcite
e,i 0.
"l0. -
o
i,-.
O
10
o
MD 39 Metalimestone
...,,.
&
E
" _m
MD 49 Metalimestone
1
SB 57ii Ita6ca granite ° " ' ° ' ° " i .......
E
(/)
MD 23 Metalimestone
MD 83ii lta6ca granite VG 96ii T. C6rregos granitoid
,1
MP GrF T. C6rregos granitoid VG 29ii Alaskitic granite
,01
I
I
La Ce
I
I
Nd
I
I
I
I
Sm Eu Gd
I
I
I
I
Dy Ho Er
I
I
I
Yb ku
ETR Fig. 4. REE pattern of the strata-bound deposits host rocks stratoid samples of Volta G r a n d e are spatially distant from this lineament. In conclusion different groups m a y be distinguished suggesting that at least two R E E sources have been involved: the first one with elevated L R E E contents is
related to carbonatites; the other one with lower L R E E enrichment and Y,'REE can be related to crustal materials. However, the R E E patterns alone are not sufficient to characterize these sources and therefore isotopic investigations have been performed.
249
1000000 1
100000
E
=-
Q
10000
"O O ,,==
1000
--
MP 1 bv
A
MP 2v
¢,
MP3v
o
MP4v
rn
MP5
!
MP6v
v
. . . . •e - - - -
MP 8cl
.... -e----
M P 9 cl
...........! ........
MP 10rm
............ • .........
MP
........... N
1 lrm
MP 12rm
........
......................
MP
13rm
M P L H I b cl MPLHla
100 .... r']---
10
MPLH
cl 9 2 cl
La Ce Pr NdPmSm Eu Gd Tb Dy Ho ErTm Yb Lu
ETR Fig. 5. The three spectra of R E E ore of M a t o Preto. by, v = violet disseminated ore; cl = colourless ore; rm = remobilized ore
1000000 M P 1 bv
100000 10000 -~
1000
0
100 e~
10 e~
E
,1
La Ce
Nd
Sm Eu Gd
Dy Ho Er
Yb Lu ETR
Fig. 6. A comparison of the M a t o Preto R E E ore spectra with the carbonatite R E E pattern
¢-
MP4v
,t
MP 10rm
•
MPLHla
~-
MPLHlb
..........[ ] .......
M P Carb 1
..........o ........
M P Carb 2
..........Q ........
M P Carb 3
250
Sm-Nd isotope studies Two or three representative samples were analyzed from each previously described deposit. Two microcrystalline ores were selected from M a t o Dentro, the former showing remaining carbonates of the original replaced karstsediment (MD 41) and the latter being carbonate free (MD 17). F r o m Volta Grande typical samples of microcrystalline ore, have been selected in body I (VG 38 unlayered ore) and in a minor occurrence near body III (VG 87 layered ore). In both samples fluorite is in association with later adularia, Ba-rich adularia and quartz. F r o m the Sete Barras deposit a microcrystalline ore (SB 66) was selected resulting from the replacement of a former karstic sediment. On the other hand in sample SB 34, 34a has a macrocrystalline texture, while 34b is microcrystalline and homogeneous. The carbonatite-associated ore from M a t o Preto ( M P L H l a and b) is a unique sample, where M P L H l a is the bulk ore and M P L H l b is macrocrystalline colourless pure fluorite. In the Braz samples (BR 2 and BR 5) fluorite and tourmaline are associated. F o r dissolution these fluorite samples were mixed with ultra-pure silica and Li-metaborate before melting in a platinum crucible (G Mevelle, personal communication). After addition of 1475m and 15°Nd spikes and gentle drying, samples were processed as described by Boher et al. (1992). The results are reported in Table 6. Total blank procedure including the addition of 90 mg of optically pure natural quartz, is around 1 ng Sm, 2 ng Nd, which amounts to less than one hundredth of the least concentrated sample. The definition of eNdo is as follows:
e
N
d
F(i43Nd/14*Nd)sample ] x104 o = L ~ - i
with (143Nd/i44Nd)crlUR = 0.512638. Model ages Tmu ) were calculated using the depleted mantle evolution of Ben O t h m a n et al. (1984). Based on eNdo data or model ages calculated relative to the depleted mantle two groups may be distinguished (Table 6).
1. Fluorite samples related to the M a t o Preto carbonatite, with slightly negative eNdo and model ages < 1.1 G a similar to the other Cretaceous carbonatites from Brazil (Walter 1991). 2. Other fluorite, with strongly negative eNdo ( - 14 to - 20) is related to an old crustal source. The crustal residence time of this material is estimated at ca. 2 Ga, taking into account only the samples with typical i47Sm/l**Nd crustal values (0.1 to 0.125). Using the diagram T(DM) versus 1/(i478m/144Nd)DMo --(147Sm/ig4Nd)sampl¢ (Zhao et al. 1992, Fig. 2), the two trends are also clearly identified (Fig. 7). The linear correlation observed for some of the analyzed fluorites (Mato Dentro, Braz, one sample from Sete Barras and Volta Grande) can be interpreted as a similar source and a comm o n fractionation age for these minerals. However, the large errors bars on the calculated T~DM), which increase when the 147Sm/144Nd ratio of the sample tends toward the i47Sm/i44Nd of the depleted mantle, and the few samples analyzed of the M a t o Preto ore exclude a determination of the REE fractionation ages (intersections of the T~DM)axis with the linear trends). The small number of samples analyzed in each deposit precludes any precise dating of the mineralization as performed in British fluorite deposits (Chesley et al. 1991). However, it is possible to draw some reference lines (Fig. 8) and verify that the M a t o Dentro strata-bound ore is roughly near the 1.2 G a reference isochron and that the Braz fracture-filling ore is near the 0.3 G a reference isochron. Both the M a t o Preto carbonatite-associated ore and the Sete Barras strata-bound ore display nearly horizontal lines compatible with a recent Sm-Nd fractionation. This nearly horizontal line at M a t o Preto could be related to the Cretaceous event suggested by geological data. At Sete Barras the discrepancy with the minimum geologic age of 0.5 Ga, is probably indicative of Cretaceous re-equilibration. However there is no evidence of contamination by REE from the carbonatites because i43Nd/i44Nd ratios are lower than at Mato Dentro or Braz (Fig. 8). Accordingly, it can be supposed that during the 'Cretaceous event' circulation of hydrothermal solutions was independent of carbonatite-related fluids. On the other hand, the
Table6. Sm-Nd analytical data. MD - Mato Dentro samples; SB
Sete Barras samples; VG - Volta Grande samples; BR - Braz samples; MP - Mato Preto samples. Analytical error for 143Nd/i44Nd ratios are expressed in 2a. La Jolla standard analyses give, in static mode, an average value of 0.511852 + 10 (60 runs) using a Finnigan Mar 262, 6 collectors, mass spectrometer. 143Nd/14*Nd ratios are normalized to i46Nd/144Nd = 0.7219. Error on the Sm/Nd ratio is estimated at 1% for this type of sample and from replicate analyses Sample
Sm
Nd
i47Sm/a4*Nd
143Nd/i*4Nd
Error
MD41a MD 17 SB 34a SB 34b SB 66 VG 87 VG 38 BR2 BR 5 MPLH la MPLH lb
1.500 1.990 1.600 1.750 3.360 2.840 1.640 0.931 0.829 60.100 1.880
7.43 11.39 6.18 4.95 12.73 13.78 5.99 3.56 3.76 400.00 7.86
0.1220 0.t056 0.1565 0.2137 0.1595 0.1246 0.1665 0.1580 0.1316 0.0908 0.1445
0.511843 0.511709 0.511726 0.511634 0.511641 0.511832 0.511481 0.511910 0.511849 0.512609 0.512578
0.000013 0.000016 0.000014 0.000022 0.000012 0.000016 0.000010 0.000011 0.000013 0.000016 0.000014
eps0 -
15.5 18.1 17.8 19.6 19.4 15.7 22.4 14.2 15.4 0.6 1.2
TDM 2.07 1.94 3.58 3.98 2.15 3.20 2.30 0.61 1.15
251 6 5
4
o
sources are always the same. In this respect it is important to notice that the Mato Preto carbonatite is an exception in Brazil: in other carbonatites fluorite is scarce, if present (JC Gespar, oral communication; Mariano 1989). Our preferred hypothesis is based on geological evidence: since Mato Preto is a carbonatite intrusion localized at the vicinity of older strata-bound occurrences, it is possible to suggest dissolution, then limited transport and redeposition by 'carbonatitic' related hydrothermal fluids, of a former strata-bound fluorite deposit. Another example is given by the fluorite occurrences at Barra do Itapirapu~ where disseminated fluorite is associated with carbonatitic rocks outcropping at less that 1 km from a stratoid fluorite occurrence hosted by Proterozoic metalimestones (Silva and Oliveira 1984; Dardenne and Touray 1988).
MatoDentro
A SeteBarras
I
[]
1
VoltaGrande
-I- Braz
T
v
I-
0
i
i
i
I
i
8
10
12
14
16
18
Conclusions
1 (147S]II/144Nd)DMo - (147Sm/144Nd)sampl e
Synthesizing the various data, a hydrothermal history of the Vale do Ribeira zone may be outlined through the study of fluorite deposits. A Proterozoic stage has been identified by geological data, in the Braz fracture-filling deposit and in the Volta Grande, Sete Barras and Mato Dentro strata-bound deposits, the last one being least affected by later magmatic intrusions, hydrothermal activity and brittle tectonics. In the 0.6-0.5 Ga range granitic intrusions could explain the anomalous Volta Grande isotopic data by introducing source heterogeneities but no other geochemical influence of these intrusions was demonstrated, probably because of insufficient Sm-Nd isotopes analyses. The different K-Ar ages determined on clay minerals respectively from Mato Dentro and Sete Barras (Dos Santos and Bonhomme 1991) may not necessarily reflect two distinct hydrothermal events. More likely,
Fig. 7. T(DM) ¥S l/(lgVSm/144Nd)DMo -(igVsm/lg4Nd)sample plot. The value of (147Sm/144Nd)DMo ratio for the present day depleted mantle is 0.225
strong negative slope observed in Fig. 8 for Volta Grande samples suggests heterogeneities between fluorite sources of body I and the occurrence outcropping near body III. Geological and geochemical data reveal the existence of at least two different groups of source rocks for REE trapped in fluorite in the .Vale do Ribeira region. The Sm-Nd isotopic data are specially important in this regard, suggesting that REE trapped by the fluorite from strata-bound and fracture-filling deposits could originate from old crustal material and reveal significant differences with the carbonatite-associated fluorite. A paramount issue is to know whether REE sources and fluorine 0,5130
o
Mato Dentro
A
Sete Barras
[]
VoltaGrande
0,5124
+
Braz
0,5122
O
Mato Preto
0,5128 0,5126
65 Ma ......
"'''''''''''''''''''''''''''''''''''
......
" ......
O" ........
"O
z ,,¢
""O z
1200 Ma
0 5120
300 Ma
o°~°°~
,,.o....IF,'""'"
0,5118 .4,....
"
......
65 Ma
A
0,5116 []
0,5114 0,51120,5110 0,05
0,10
I
I
0,15
0,20
147Sm/144Nd
Fig. 8. 14aNd/144Nd vs 1478m/144Nd plot, with reference isochrons
0,25
252 o l d e r clays have been m o r e or less r e j u v e n a t e d at the s a m e time, the process being the m o s t effective at Sete Barras. This conclusion is in a g r e e m e n t with the greater r e c r y s t a l l i z a t i o n of the l a t t e r deposit. The r o u g h l y horiz o n t a l S m - N d i s o t o p i c d i s p l a y (Fig. 7) at Sete B a r r a s suggests t h a t the a b o v e - m e n t i o n e d reset of K - A t ages is p r o b a b l y related to the h y d r o t h e r m a l fluid circulation l i n k e d to the C r e t a c e o u s t h e r m a l event. T h e 0.3 G a age is p o s s i b l y less t h a n the geological age of the d e p o s i t ( a b o u t 0.5 G a ) a n d hence c o u l d reflect a p a r t i a l rejuvenation. H o w e v e r , this process m a y be very limited considering the u n c e r t a i n t y of + 0.15 Ga. F i n a l l y the C r e t a c e o u s t h e r m a l event at 65 M a detected b y d a t i n g the r e g i o n a l occurrences of alkaline r o c k s c o u l d e x p l a i n the S m - N d isotopic d a t a collected on fluorite at M a t o P r e t o as well as the i s o t o p i c reset n o t i c e d at Sete Barras. In this respect, these d a t a suggest that S m - N d isochrons from fluorites are sensitive to later t h e r m a l h i s t o r y a n d m a y be helpful in recognizing otherwise p o o r l y d e t e c t a b l e h y d r o t h e r m a l events. Finally, c o n c e r n i n g the 'linear' g e o g r a p h i c a l distribution of the fluorite deposits, there is no o b v i o u s e x p l a n a tion because the M o r r o A g u d o t r a n s c u r r e n t fault (Fig. 1) was clearly active after the i n t r u s i o n of the Tr~s C 6 r r e g o s g r a n i t o i d a n d hence after the f o r m a t i o n of the s t r a t a b o u n d fluorite deposits. U n f o r t u n a t e l y , the value of the l a t e r a l d i s p l a c e m e n t at t h a t time r e m a i n s u n k n o w n .
Acknowledgements. A first draft of this article was significantly improved by the observations of Mr Julian F Menuge (University College Dublin) and two anonymous referees. The work, performed at the University of Orleans, was supported by a CAPES/PADCT Brazilian grant (CBE/5957/89). The authors wish to thank Mr James Bourne (Universit6 du Qu6bec ~ Montreal) for English improvement. References Bastos Neto, A.C. (1990) Le District ~ Fluorine de Santa Catarina (Br6sil): Mineralisations et alt6rations hydrothermales dans leur cadre geodynamique. Th6se de Doctorat (Nouveau R~gime), Universit6 d'Orl6ans, 366 p. Bastos Neto, A.C., Touray, J.C., Dardenne, M.A., Charvet, J. (1991) Chronologie et 6volution des ftuides hydrothermaux dans le district ~ifluorine de Santa Catarina, Br6sil: donn6es de l'analyse des terres rares et de l'&ude des inclusions fluides. Mineral. Deposita, 26:211-216 Ben Othman, D., Polv6, M., Allegre, C.J. (1984) Nd-Sr isotopic compositions of granulites and constraints on the evolution of the lower continental crust. Nature 307:510-515 Boher, M., Abouchami, W., Michard, A., Albarde, F., Arndt, N.T. (1992) Crustal growth in West Africa at 2.1 Ga. J. Geophys. Res. 97:345-369 Chesley, J.T., Halliday, A.N., Scrivener, R.C. (1991) Samariumneodymium direct dating of fluorite mineralization. Science 252:949-951 Cordani, U.G., Bittencourt, I. (1967) Determina~Ses de idade pot/tssio-arg6nio em rochas do Grupo A~ungui. In: Cong. Brasileiro Geol., 25, Anais., S~o Paulo, SBG, pp. 105-110 Cordani, U.G., Hasui, Y. (1968) Idades K-Ar das rochas alcalinas do Primeiro Planalto do Estado do Paran~i. In: Cong. Brasileiro Geol., Anais, SBG, Ouro Preto, pp. 149-153 Dardenne, M.A., Touray, J.C. (1988) La fluorine au Br6sil: gisements filoniens traditionnels et nouveaux types de mineralisations. Chronique de la Recherche Mini6re 490:35-46 Dos Santos, P.R., Bonhomme, M.G. (1991) K/Ar dating of clays associated with fluorite mineralizations along the Atlantic coast of South America - Relationships with South Atlantic Ocean
opening. In: Pagel, LeroY (eds.) Source, transport and deposition of metals. Balkema, Rotterdam, pp. 381-384 Fagundes, P.R., Dardenne, M.A., Veiga, A.T.C. (1984) Evoluq~o da mineraliza~o de fluorita no dep6sito de Sete Barras - Paranh. In: Cong. Brasileiro Geol., 32, Anais., Rio de Janeiro, SBG, pp. 3755-3769 Felipe, R.S., Oliveira, L.M. (1986) Relat6rio de Etapa: Fluorita - Lageado Grande. Unpublished report, Mineropar, 20 p. Fiori, A.P., Monastier, M.S., Camargo, E.C., Lima, R.G. (1984) Lineamentos tect6nicos e possiveis mineraliza~Ses associadas no pr6-Cambriano paranaense. RelatSrio do conv~nio Mineropar/ Universidade Federal do Paran~i, 261 p. Fiori, A.P. (1991) Tectonica e estratigrafia do grupo A~ungui a norte de Curitiba. Tese de livre-doc~ncia apresentada ao Instituto de Geoci6ncias da Universidade de $5o Paulo, 261 p. Hasui, Y., Fonseca, M.J.G., Ramalho, R. (1984) A parte central da regi~o de dobramentos Sudeste e o maciqo mediano de Guaxupe. In: Schobbenhaus, C., Campos, D.A., Derze, G.R., Asmus, H.E. (eds.) Geologia do Brasil. DNPM, Brasilia, pp. 307-328 Issler, R.S., Freire, F.A. (1985) A diversidade dos granitos da regi~o sul do Brasil. In: Simp. Sul-Brasileiro de Geol., 2, Anais, Florian6polis, SBG, pp. 89-101 Jenkins II, R.E. (1987) Geology of the Clugger fluorite deposit, Mato Preto, Paran~, Brazil. Rev. Brasileira Geoci~n. 17:288-294 Mariano, A.N. (1989) Nature of economic mineralization in carbonatites and related rocks. In: Bell, K. (eds.) Carbonatites: genesis and evolution, pp. 149-176 Pontes, J.B. (1980) Pesquisa preliminar da falha da Ribeira Adrian6polis. Unpublished report, Mineropar, Curitiba, 19 p. Ronchi, L.H. (1986) Evolu~o e Geoquimica do Dep6sito de Fluorita de Volta Grande - Pr. Tese de Mestrado, Universidade de Brasilia, 133 p. Ronchi, L.H., Dardenne, M.A. (1987) Padr~o geoquimico de elementos terras raras dos dep6sitos de Sete Barras e Volta Grande PR. In: 1° Cong. Bras. Geoquim., Atas... Porto Alegre, SBGq 1:17-30 Ronchi, L.H., Felipe, R.S., Dardenne, M.A. (1987) O dep6sito de fluorita de Volta Grande: tipos de min6rio e sucess~o mineral. In: 3° Simp6sio Sul-Brasileiro de Geologia, Atas... Curitiba, SBG. 2:615-622 Routhier, P. (1980) Off sont les m6taux pour l'avenir? M6moire du BRGM n ° 105, Orl6ans, 410 p. Santos, R.V. (1988) Geologia e Geoquimica do Dep6sito de Fluroita do Complexo Alcalina Carbonatitico de Mato Preto - Paran~t. Tese de Mestrado n ° 47, Universidade de Brasilia, Instituto de Geoci~ncias Santos, R.V., Dardenne, M.A. (1988) Fluorita de Mato Preto: um easo particular de mineraliza~o de fluorita associada a complexo alcalina carbonatltico. In: Cong. Brasileiro de Geol., 35, Anais, SBG, Bel6m, pp. 1251-1261 Silva, D.C., Oliveira, L.M. (1984) Projeto Barra do Itapirapu~ - RelatSrio de etapa. Unpublished report, Mineropar, Curitiba, 54 p. Silva, D.C., Felipe, R.S., Pontes, J.P. (1981) Notas sobre as ocorr~ncias de fluorita do vale do Ribeira (PR). In: Simp. Reg. Geol., 3, Anais, SBG, Curitiba, 1:21-35 Strong, D.F., Fryer, B.J., Kerick, R. (1984) Genesis of the St. Lawrence fluorspar deposits as indicated by fluid inclusions, rare earth elements and isotopic data. Econ. Geol. 29:1142-1158 Touray, J.C. (1989) Etudes d'inclusions fiuides et modelisation de la gen6se des gites du type "Misssissipi Valley". Chronique de la Recherche Mini6re n ° 495:21-30 Walter, A.V. (1991) Caracterisation geochimique et mineralogique de l'alt6ration du complexe alcalin de Juquia (Br6sil) - Comportement des terres rares dans les mineraux phosphat6s. Universit6 de Aix-Marseille III, Th6se Zhao, J.X., McCulloch, M.T., Bennett, V.C. (1992) Sm-Nd and U-Pb isotopic constraints on the provenance of sediments from Amadeus Basin, central Australia: evidence for REE fractionation. Geochim. Cosmochim. Acta 56:921-940 Editorial handling: DR