Canadian MineralogistVol. 17, pp. 33-38 (1979)

HETEROGENEOUS FELDSPARS IN THE MID.TRIASSIC VOLCANIG ROGKSOF THE DOLOMITES

P. L. ROSSII, R. RINALDI2 eno G. SIMBOLI'llstituto di Mineralogia e Petografia, Universitd d.i Bologna,

Piazza Porta S. Donato 1, Bologna, Italy 40127

2lstituto di Mineralogia e Petrologia, Universitd di Modena,Via S. Eufemia 19, Modena, Italy 41100

ABSTRAcT

Feldspar phenocrysts from effusive rocks ofthe Dolomites (Southern Alps) show coarse, con-cenftic zonation of plagioclase and K-feldspar.Electron-microprobe analyses of the various zoneswithin any one phenocryst indicate atypical com-positions for the coexisting phases. The plagio-clases range from labradorite to andesine (An 45-70Vo); the K-feldspars have very high K contents(Or 95-96Vo) and correspondingly very low Abcontents (34Vo). A growth model for such pheno-crysts is difficult to establish. The proposed mechan-ism is that of a primary sequential crystallizationfollowed by successive modifications occurring dur-ing the complex postmagmatic history of the rock.

Sorvtrvr.lns

Les ph6nocristaux de feldspath des roches 6rup-tives des Dolomites (Alpes M6ridionales) montrentuno zonation concentrique grossidre de plagioclaseet de feldspath potassique. Les analyses i la micro-sonde des zones d'un m€me ph6nocristal r6vdlentdes compositions atypiques des phases coexistantes.Irs plagioclases varient du labrador i I'and6sine(An 45-70), tandis que les zones de feldspathpotassique ont une teneur tr0s 61ev6e en potassium(Or 95-96) et par suite trEs basse en sodium(Ab 3-4). Un schdma de croissance qui explique-rait de tels ph6nocristaux n'est pas facile i 6tablir.Nous proposons comme m6canisme une cristallisa-tion primaire en stades successifs, suivie de modifi-cations postmagmatiques qui survinrent durant l'6-volution complexe de la roche.

Clraduit par la R6daction)

INrnonucrtoN

During the interval between the UpperLadinian and Lower Carnian of the Mid-Triassic, extensive magmatic activity occurredin the southern Alps, particularly in the areaof the Dolomites, leading to the emplacementof large volcanic bodies, mostly underwater;

these filled the basins between the Anisianand the Ladinian carbonate formations. Theemplacement of two intrusive complexes asring dykes of monzonitic and granitic rocksaround a volcanic system at Predazzoo and ofpyroxenitic through monzonitic rocks at Mon-zoni, is also related to this activity.

The composition of these igneous rocks inmany respects suggests a shoshonitic series(Rossi et a/. 1976, Castellarh et aL. L977,Calanchi et al. 1978). Their petrochemistry,however, is rather complex, owing both tothe lack of differentiation in the volcanic units(important only in small intrusive bodies), andto the common and often remarkable postmag-

TABLE 1. IIIHOLE-ROCK l'lET CHEI{ICAL ANALYSES 0F X45 AND 15COMPARED TO A NORE TYPICAL COMPOSITION (X)

si 02

Ti 02

At 203Fe^0^

F C U

x45 15 X

51 .57 47 .58 49 .46

0 . 8 7 0 . 9 9 I . 1 3

21 .35 15 .57 17 . 39

Na20

K2o

P^0.t )

Hztru n +"2-cozTotal s

4 . 1 9

0 . 1 7

n ? q

0 . 94

2 .44

3 . 0 8 3 . 6 1

3.49

4 .00

n ? 7 0 . 3 7

0 . 9 4 1 . 5 8

2 .79 2 .71

6 .96 5 .39

MnO 0.06

HSO 3'37

Cao 3.74

6 .32 4 .95

6 .83 8 .55

2 .00 2 .66

4 .85 4 .57 2 .30

- . 59

100.48 99.79 99.92

X45: Pi l low lava f rom "Monte Migogn".t5 : Lava f lorr , l f ron "Alpe di Siusi" .X : Shoshonitic basalt; a representative sanple of the

liliddle Triassic volcanisn of Southern Alps.

33

THE CANADIAN MINBRALOGIST

matic mineral transformations that the min-erals have undergone, most likely as a resultof the subaqueous environrnent of their em-placement. Such transformations have affectedmostly the plagioclases, which may be com-pletely transformed into albite, whereas thepyroxenes, of the augite--salite type, are almostalways unaffected. The albitization of theserocks proceeds as stated by Vallance (1969),with a general enrichment in NagO and lossof CaO relative to the unaltered types. How-ever, it is not our aim here to discnss thecomplex problem of spilitization or albitization.

In a few samples, an abnormal enrichmentin IGO has been observed which cannot beaccounted for by the normal process of dif-ferentiation. This paper reports the study oftwo such K-rich specimens; under the rnicro-scope, both show very interesting textural rela-

tionships involving the feldspars. Our aim inthe present work is primarily to report analy-tical data bearing on these phases in the twosamples. A more complete study of the genesis,chemistry and structural relationships is plannedon a larger number of samples with analogousfeatures, possibly including other occurrencesof the same phenomenon.

The two specimens considered are labeledX45 and t5. Whole-rock chemical analyses arereported in Table 1 together with one analysisof the most common rock-type (X).

PETROGRAPHY

The two specimens selected represent separateflows. One (X45) comes from an exposureof a pillow lava near Mt. "Migogn" in the

Fra. l. A) K-feldspar zoning in plagioclase phenocryst, crossed polarizers, sample 15; B) K-feldspatcore and inner rim within phenocryst, crossed polarizers, sample 15; C) K-feldspar (dark) andplagioclase (light) alternations, crossed polarizers, sarnple X45; D) microphenocrysts of K-feldspatwith thin plagioclase outer rim, plaue polarized light, sample X45.

upper Cordevole valley lBelluno); stratigra-phically, it occurs above the Marmolada con-glomerate and is therefore associated with alate volcanic event (Lower Carnian). Thesecond occurrence ()5) has the appearance of asubaerial lava flow; it is located west of the Alpedi Tires (Bolzano) and is probably also relatedto volcanic activity of the lower Carnian.Microscopic examination of both roaks revealsa porphyritic texture. Sample X45 containsalmost exclusively plagioclase as phenocrysts,embedded in a groundmass of feldspathic rnicro-lites, whereas 15 contains phenocrysts of plagio-clase, clinopyroxene and olivine (partly replacedby calcite, iron oxides and hydroxides) em-bedded in a feldspathic groundmass.

The plagioclase phenocrysts in both sampleshave average dimensions of approximately 2-3mrn, with slight alteration to sericitel in somecases, the phenocrysts also contain small grainsof opaque minerals. The main sharacteristicof these phenocrysts is the presence of K'feldspar distributed within them in various pat-terns. The most common variety consists of afew regular and nearly continuous concentriczones of K-feldspar 10 to 20 ;r,m wide thatdevelop within a phenocryst parallel to itsmargins (Fig. 1A, B, C, D). There are uPto four K-feldspar zones in the same crystal,always with a common optical orientation. Ifthe plagioclase is twinned, so is the K-feldspar.In other cases the K-feldspar constitutes thecore of phenocrysts; the core zone, up to 60

TABLE 2. AVERAGE OXIDE COMPOSITIONS AND ATOMIC RATIOS OF THE.ITREE

REPRESENTATIVE FELDSPAR ZONES FOR EACH ROCK

HETEROGENEOUS FELDSPARS

Inner outer K-feldspar lnner 0uter K-feldsparPlag. Plag. Plag' Plag'

t m across, is associated with small alternatingiims of plagioclase towards the border. Com-plete variation between the two extreme caseshas been observed.

Individual plagioclase phenocrysts cgntain-ing such K-feldspar zones have decreasing Ancontents from core to rim despite the presenceof intervening K-feldspar layers (Table 2).Typical compositions range from apout QS-V.an for the

-inner plagio-lase to about 55/e

An for the outer plagioclase. The outermostrim of the phenocrysts is always very freshplagioclase, never K-feldsPar.'

i.-f.ldrput can also bJ found distributed insmall paiches within the plagioclase pheao-crysts, apparently without preferred orientation'In such cases the K-feldspar may constitute upto half the volume of the phenocrysts. The sameassociations aro present in the microlites ofthe groundmass; there, one most commonly ob-servis u K-feldspar core with only one rirnof plagioclase (Fig. 1D); occasisa,6lly,- onefindi a plagioclase core rimmed by K-feldspar'

ExPsnuvrsNrer,

Electron microprobe work was carried outin the wavelength-dispersive mode on a fullyautomated ARL-SEMQ instrument operated at15 kV and 0.15-0.20 pA beam current, usinga focused beam of approximately 1 g,m dia-meter. Quantitative analyte* were performed

at several points for several representatlvefeldspar cryitals in thin sections of the tworock types, with counting times of. /, 20- Vnd2 seconds for high background, peak and lowbackgtound, respectively. On-line gulu -I19y1-tion ivas based on the Ziebold & Ogilvie (19'64)

metlod by the use of Bence & Albee (1968)

and Albei & Bny (1970) correction. tactoS'Svnthetic plagioclase glass of composition An160, natur;l albite, microcline and olivine weretr"i u* standards. Scanning X-ray micrographswere obtained with very long exposure timesfor maximum resolution and detail.

35

)t u^

. J

feu^

u9o<rn

Ba0

Ca0

Nazo 4.4

KzO 0.6

Total 100.2

53 .6 56 .6 65 .4 49 .3 51 .3 64 .1

28.5 26.3 17.4 30.6 29.6 18.8

0 .6 0 .9 o . l 0 . 9 1 .0 0 .3

o . l o . l 0 . 0 0 . 2 0 . 2 0 . 3

0 .3 N .D . 0 .0 N .D . N .D . N .D .

o .o N .D . 0 .0 N .D . N .D . N 'D .' l ? , 1 9 .5 0 .0 14 .7 13 .4 0 .1

0 . 3 2 . 9 3 . 4 0 . 45 . 1

siAt

Na

K

2 . 4

1 . 5

0 . 6

2 . 6

0 .5

? n

1 . 0

0 . 0

0 . 0

1 . 0

2 . 3't.7

0 .7

' 1 .63 . 0' 1 .0

0 .0

0 .0n o

0 .4 0 .4

0 . 0 0 . 1

\ . 7 1 6 . 4 0 . 5 0 . 6 1 5 . 9

100 .2 99 .6 99 .1 99 .5 99 .9

Nmber of Cations (0 = 8)

DrscussroN oF THE REsuLTs

Typical compositions of the feldspar zones

are ?ported in faUte 2 as averages of four

to nini point-analyses for each feldspar com-position.

^ The composition of each type of

?eldspar ranges within :l 5% of the .per-

""tttug"t of the major elements and within

* Z-O% for the elements present in co^ncentra-tions below lO% in weight percent oxide' The

0 . 3 0 . 3

0 .0 0 .0

x Tota l I ron ca lcu la ted as Fe0.

36 THE CANADIAN MINBRALOGIST

Ftc.2. A), B) and C): X-ray Ka maps of K, Ca and Na radiations,respectively, in sample X45; D), E) and F): analogous X-ray maps,samplo !5.

elemental distribution of K, Ca andthe phenocrysts is shown in the X-raygraphs of Figure 2.

The.intimate association of feldspar phasesmay be explained by one or more of the fol-lowing mechanisms (Smith 1974): (1) sub-solidus exsolution, (2) sequential growth, (3)replacement. In our case, the regularity ofthe K-feldspar zones and their remarkable com-positional homogeneity suggest that ttre firstprocess is rather unlikely. The second mechan-ism proposed, the sequential growth of thefeldspar phases, would seem, from tlte distribu-tion of the phases, to offer a more likely ex-planation. Ilowever, the composition of theK-feldspar zones differs greatly from what onemight expect in phenocrys8 of effusive rocks.The Or contents of the K-feldspars, almostconstant at 95-96/s, are distinctly different

TABLE 3. COI,IPOSITIONS OF THE POINTS RTPORTED ItI FIGURE 3

P o l n t s I 2 3 4 5 6 7 8 9 l 0 1 l 1 2

s i02 53 .3 65 .0 65 .3 64 .8 63 .8 68 .3 68 .7 55 .1 54 .6 55 .9 55 .9 58 .0

A l^0- 29 .0 17 .9 17 .7 17 .9 17 .9 18 .7 t8 .4 26 .1 26 .5 27 .0 26 .1 25 .5

F e o ' . 0 . 0 0 . 1 0 . 2 0 . 2 0 . 3 0 . 4 0 . 2 0 . 7 0 . 7 0 . 9 1 . 0 0 . 8

M s o 0 . 1 0 . 0 0 . 0 0 . 0 0 . 0 0 . 0 0 . 0 0 . 1 0 . 1 0 . 1 0 . 1 0 . 1

C a o 1 2 . 7 0 . 0 0 . 0 0 . 0 0 . 1 0 . 1 0 . 0 9 . 8 l 1 . 0 1 0 . 3 8 . 5 8 . 9

N a 2 0 4 . 0 0 . 4 0 . 1 0 . 1 0 . 8 7 . 1 1 . 3 4 . 4 4 . 7 4 . 9 4 . 9 5 . 6

( 2 0 0 . 6 1 5 . 9 1 6 . 2 1 6 . 4 1 4 . 5 6 . 0 t 0 . 4 1 . 2 t . 0 1 . 7 2 . 4 1 . 8

Tota ls 99 .7 99 .3 99 .5 99 ,4 97 .4 1006 99 .0 97 .4 98 .6 t008 99 .2 1007

t. Total Iron calculated as Feo.

Note: All point-analyses taken at f ixed spatJal lntervals.Conposltions that are not ln full agreerent with Figure 3,are prcbably due to alteratlon, graln borders not parallelto the beam, surface topography, etc.

from those noted by Lofgren & Gooley (1977)for synthetic phases. They found compositions

Na inmicro-

HBTEROGENEOUS FELDSPARS 37

ferent compositions of the K-feldspars (Table4) with approximately Or 60, Ab 35, An 5 inthe microphenocrysts and microlites of thegroundmass. Our degree of confidence associatedwith the analyses of Table 3 and 4 is 'muchless than that of the analyses of Table 2 (seenotes at bottom of tables); for this reason,atomic ratios are not reported. Nevertheless,these analyses represent quantitative measure'ments of the spatial-chemical relationship be-tween phases. Such compositions could rep-resent the product of primary crystallizationand/or initial phases of a transformation thatwould have involved the phenocrysts more thanthe microlites of the groundmass. The com-position of the plagioclase is nearly the samefor both the phenocrysts and the microlitesof the groundmass.

The sharp separation between the phases inqontact and the continuity and shape of thezones, together with their reasonable homo-geneity, favor a sequential crystallization, clearlydominated by a K-feldspar composition forsome period, and then by a plagioclase. Thismechanism could develop in response to thevariations in water pressure inside the magmachamber as a result of the complex dynamicsof the effusive activity. Two phases could thusbe generated: a plagioclase such as the oneobserved and a potassic feldspar that musthave been richer in Na than that presentlyfound.

The third hypothesis, that of a metasomaticorigin, could explain the K-feldspar composi-tions, although the regular, continuous andrepetitive concentric zones, commonly in op-tical continuity with the plagioclase zone$, arequite difficult to explain in this way. With thishypothesis, there exists one additional problern:the constant ,presence, in the outermost border,of plagioclase, never K-feldsPar.

In conclusion, we believe that no singlemechanism can be firmly supported. However,we prefer a hypothesis of primary sequentialcrystallization of the K-feldspar and plagio-clase, with local modifications due to the pos-sible ion exchange by solid-state reactions dur-ing the complex subsolidus history of the rocks.

ACKNOWLEDGEMENTS

Thanks are due to Professor R. B. Fergusonfor critical reading of the manuscript. Electronmicroprobe work at the "Istituto di Mineralogiae Petrologia, Universit6 di Modena", was madepossible by the financial suPport of the "Con-siglio Nazionale delle Ricerche" of ltaly.

Frc. 3. Scheme of compositional zoning and pointsanalyzed as reported in Table 3. G: groundmass;F: plagioclase zones; K: K-feldspar zones; A:patches with anorthoclase composition. Numbersrefer to analyses in Table 3, dots mark beamposition and approximate spot-size for each pointanalysis; sample X45.

between Or 75 and Or 55. Therefore, withthe compositions we found, .the hypothesis ofsequential growth cannot be taken into con-sideration unless one admits subsequent phe-nomena of unmixing and removal of sodiumfrom alkali feldspar to give K-rich sanidine.

A notable contribution to such a hypothesisis given, in some cases, by the occurrence ofvery small and irregular patches of "anortho-clase" composition (Ab-63, 0r-36) (Table 3,Fig. 3, point 6) within the pure K-feldsparzones and the fairly cornmon existence of dif-

TABLE 4. AVERAGE SELICTED COHPOSITIONS OCCURRING lIIftIIN I}IEI{I CROL ITES SHOI{I NO K-FELDSPARS O I FFERENT FROM THOSEOF ]}IE PHENOCRYSTS."

K-Feldspar K-Feldspar Inner Plag. outer Plag'

s i02

A1 203Fe0

Hgo

Ca0

Na20

Kzo

Totals

64.4

?0 .7

0 . 4

0 . 0

1 . 3

' 10 .0

65 .4

19.2

0 .4

0 . 1

0 , 8

5 . t

1 0 . 3

50 .6

30 .9

1 . 0

0 , 1

12.81 1

0 . 6

5 l , 0

3 1 . 7

0 . 9

0 . ?

1 3 . 6

J . )

0 . 6

1 00 ,7 99 .9

ti Analyses of very snall grains (20-30 urnZ).

3 8 THE CANADIAN MINERALOGIST

RnnnnsNcEs

ALBET, A. L. & Rey, L. (1970): Correction fac-tors for electron probe microanalysis of silicates,oxides, carbonates, phosphates and sulphates.Anal. Chem, 42, 1408-L414.

BEucr, A. E. & Annr, A. L. (1968): Empiricalcorrection factors for the electron microanalysisof silicates and oxides. J. Geol.76, 382-403.

CArANcHr, N., LuccnrNr, F. & Rossr, P. L. (1978):The volcanic rocks from the Mount Agnelloarea (Fiemme Valley, Italy): a contributionto th3 knowledge of the Mid-Triassic mag-matism of the Southern Alps. Tschennaks Min-eral. Petog. Mitt. 25, l3l-143.

CASTELLARIN, A., Rosst, P. L., Sruaou, G,, SoM-MAvTLLA, E. & Dr, Luce, A. (1977): Datigeologici e petrografici sul gruppo del Buffaure.Mineral, Petrog. Acta 21, 165-187.

LorcnrN, G. E. & Goorpv, R. (1977): Simulta-neous crystallization of feldspar intergrowthsfrom the melt. Amer. Itlineral. 62, 217-228.

Rossr, P. L., Srunorr, G. & Vrrr., G. (1976):Significato paleogeografico e magmatico-tetto-nico della serie vulcanica, vulcano-clastica ladi-nica superiore del Monte Civetta. Soc. Geol.Ital. Boll. 95, 433-458.

SnatrH, J. V. (1974): Feldspar Mitrerals, 2. Chem-ical and Textural Properties, Springer-Verlag,Berlin.

VettaNcE, T. G. (1969): Spilites again: some cor-sequences of the degradation of basalts. Proc.Linn. Soc. N.S. l lales 94, 8-51.

Ztnyoto, T. O. & OcILvIE, R. E. (1964): An em-pirical method for electron microanalysis. lncl.Chem. 36, 322-327.

Received. lanuary 1978; revised manuscript ac-cepted November 1978.