AN INTERSTRATIFIED MIXTURE of MICA CLAY MINERALS Susuuu Snrnooeanp Tosnro Suoo,Tokyo Uniaersityof Etlucation,T Okyo,J Apan

1960 THE AMERICAN MINERALOGIST, VOL 45, SEPTE\'IBER-OCTOBER'

AN INTERSTRATIFIED MIXTURE OF MICA CLAY MINERALS Susuuu SnrnooeaNp Tosnro Suoo,Tokyo Uniaersityof Etlucation,T okYo,J aPan. Assrn.{ct

INrtouucuoN

found almost free from impurities. Thus such mineralsare not so Iare as ap- we have hitherto anticipated, and the present mineral is considered propriate for the description of its mineralogical and crystallographic properties.

MopB ol OccuRRENCE 1O7O S. SHIMODA AND 7.. SUDO

complexmineral yonago associations.The mine now being consideredis one of the important pyrophyllite-diaspore depositsin Japan. The area near the ore deposit consists of fine grained sandstoneand shaleintruded by porphyrite and andesitedykes. A[ of theserocks are covered by a lava flow. The ore deposit occursalong the boundary zone between the porphyrite and andesitedykes repracingboth of theserocks. The ore bodies consist of diaspore,cray minerals, quartz and pyrite. Diaspore occurs as compact or powdery massesmostly cemented by pyrophyllite. About eighty specimenswere coilectedin the area inciuding the ore deposit, and the amounts of the minerals in each specimenwere esti- mated by r-ray analysesusing an internal standard of fluorite. standard curves were made using pure clay mineralsfrom the other locaiities.The confirmed mineralsare diaspore,pyrophyllite, quartz,sericite, kaoiinite, halloysite, montmorillonite, and pyrite. By plotting the result of the qualitative estimationsof these minerals except pyrite, zonar distributions were confirmed particularly with regard to ttre distributions of diaspore,pyrophyllite, and kaolin minerals.The center of the zonal dis- tribution was confirmed in two difierent places; in both of which the zonal distributions were found to be (diaspore and pyrophyllite)_ (kaolin minerals)-( quaftz), going progressively u*uy fro- ih" ."rrr".. The mineral nameswritten in parenthesesare thoseof the principal minerals in each zone. The boundariesbetween these zonesu.. roi sharp, but gradually gradeinto one another. Diasporeis rather more dominant in the central area than pyrophyilite, but in most casesboth of these minerals occur mixed with each other in various proportions. pyrite crystals are dominant in the kaolin mineral zone.Sericite occurs in small amounts throughout thesezones, but tendsto be dominant in the silicious zone. Montmorillonite occurslocaily in the siliciousand kaolin mineral zones. clay minerals showinglarge spacingsin their powder diffraction data are frequently noted: Type I (specimens 11 and 63): This mineral showsa 26 A spacingand will be discussedin detail in this report. rt occurswidely in the ore area, particularly dominantly in the pyrophyllite-diaspor",on",frequently as vein-shapedmasses in the pyrophyllite bodies. Type (specimen II 1): This mineral alsoshows a spacingof about 26 A and occurs only in the kaorinitezone. Because of difficulty.-inobtaining a pure sample,we could not ascertainits detailedmineralogical properties, but it seemsto agree with rectorite (Bradley, 1950), or the 26 A.tay mineral found from the Honami pyrophyllite deposit,Nagano prefecture (Kodama, 1958). Type rrl (specimen 38): This is found rarely in the siliciouszone as INTERSTRATIFIEDMIXTURE OF MICA CLAY MINERALS 1O7I thin crusts covering siliciousrock fragments. The siliciousarea usually shows brecciated structures consistingof siliciousrock fragments. Be- causeof difficulty in obtaining a pure sample,we could not determineits detailed mineralogicaland properties, but it seemsto agree with the aluminian mixedJayer mineral recently found from the Kamikita mine, Aomori Prefecture(Sudo and Kodama,1957).

26 A Cuv MrNnn.q,r(Tvro I) Macroscopicproperties: This mineral occursas white compact masses rvith greasyluster resemblingthe pyrophyilite masses,and is associated with the pyrophyllite-diasporezone. We could not determineits detailed mode of occurrence,but someof it surelyoccurs as vein-shapedmasses in pyrophyllite bodies. By r-ray testing,we found specimensalmost free from impurities, and obtained the following data from the crude specimens.Their lc-rayre- flections consistof the clear and sharp 26 A reflection and its higher orders except the weak 4.27h reflection which is probably due to a small amount of impurity. On close examination oI the 26 A reflection under various experimentalconditions as studiedwith a slow scanningspeed, or using the internal standardof a-stearicacid, the spacingappears to vary slightly from specimento specimen.Basal spacingsobtained from nine referencespecimens and also the mean valuesof eachset of basal reflections are given in Table 1. From eachof thesemean value the 001 spacings werecalculated, and 26.1* 1 A was obtainedas the final meanvalue.

Tesr-B 1. Beslr- SpecrNcs oF TIrE 26A CI,Iv-MTNERAL lRoM rnr YoNlco MrNn

o 001 I d I

001 zasA 01s 44m zo air 60s 002 128 52vs 47s 125 50s 005 512 25s 27 vs 506 34s 008 324 J0s 40 vs 325 34m 00.10 257 26s ,12vs 2.57 35s

8 Mean

d I d

001 zt-oL 50m 27 6A zt.aL zt oL oo2 1,29 25.8 005 510 25.5 008 3.25 260 00.10 2.57 25.7

Mean 26.111A IO72 S, SHIMODA AND T. SUDO

Worthy of notice is that the 26 A reflectionis usually accompaniedby a tailed reflectionon its small angleside, but we could not find reflections attributable to superstructuresof the 26 A ore. Effects of heat (Table 2): The specimenanalysed by r-ray diffraction washeated at eachof the temperatures300o C.,500o C., 700oC., g00oC., and 1000oc. for one hour and sealedinto a glasstube immediately after cooling,and kept in the sealedglass tube until it was studied by r-rays. rn the specimenheated at 300oc. for one hour, the 26 A reflectionwas just visible, and its higher order reflectionswere somewhat weakened. The 13 A reflectionshifted to a 10.3A reflection,and the 5.10 A reflec-

TasLE 2. X-Rny Drnln,qcrron Der,l ron rnn Arn-Dnvro 26 A Cr,lv MtNnnar (CuKa, 1.5418A) r- Room temD .300'c 700'c 900'c. 1000"c

I dlr dlI

zo.z,&]+o'" z+.sAI s"t rS,+A 130 33 vs rr sA 5b rr.eA 7b 12b r09 5D 104 8b 103 103A 9.9 OD 10.0 10b 93 .JD 510 lo s 8m 507 5D 504 9s 507 10s s09 3m 4.48 19s 20 vs 453 4 51 38 vs 451 42 vs 4.50 8m 4.27 10 vs 21 vs 4 31, 3s 4.27 7vs 387 4vb 3.92 6b 380 6b 3vb 355 JD

340 8vb 335 10m 27 vs 3.35 30s 3.29 8s 7vs 327 9vs 325 310 4b 3 .03 2vb 3.08 4b 284 2vb 288 JO 2.?0 2b 2.56 15s ?

179 4b .tD 172 .t vD JD 1.69 3b 3vb 1.67 3vb JD JD 1.65 155 5s 1.50 9m 4b

w: very sharp. s: sharp. m; medium b: broad. vb: very broad. IN'T'ERSTRAT'IFIEDMIXT'TJIIE OF MICA CLAY ['IINERAL'S 1073

rion was replacetlby a 5.01A reflet.tio..fhe 3.27 A reflectionbccame broader, and its spacing was replaced by a slightly larger spacing of 3.29 A. In the specimenheated at 500oC., the 26 A reflectionwas still just visible.Its higher order reflectionsbecame weaker than in the caseof the specimenheated at 300oC. The 3.27 h reflectionwas replacedby a broad reflectionwith a slightiy larger spacingof 3.35A. Worthy of special attention is that the 10 A reflectionacquired a tail toward its.low angle side.A very broad reflectionat about 11.5 A could be recognizedon its tailed reflection. Such behaviour is closely similar to that of the tailed reflection accompanying the 10 A reflection in the crude specimen of the Kamikita material. After heating to 700oC., the weak and broad 2.45A

1000 5000 10000c

Frc. 1. D.'l'.A. curves of the 26 A clay mineral from the Yonago mine Above: specimen in natural state. Below: specimen treated with piperidine.

reflectionappeared; the intensitiesof the 10 A and 5 A reflectionbecame somewhalmore intense, but the 3.34A and 4.51A reflectionsbccame re- markably intense.The tailed reflectionwas still confirmed.After heating to 900oC., the intensitiesof the 10 A, 5 A, 4.5A and 3.35A are similar to thosein the specimenheated at 700oC., but the 26 A reflectionas well as the tailed reflectioncompletely disappeared, and the very weak 19 A reflection appeared.In the specim-enheated at 1000oC., the reflections were weakened,except the 4.50 A and 3.40 A reflections.The result ob- l-ainedabove suggests that the 26A spacingseems to be related to hydrous statesof the 10A structure. Differential thermal analysis curve: The difierential thermal analysis curve of the specimenanalysed by r-rays showsthe broad endothermic peaks at about 600oC., and also between700' C. and 800oC., excepta clearendothermic peak between100' C. and 200oC. (Fig. 1, above).The specimenwas soakedin piperidinefor about 10 minutes, and dried on a water-bath. The specimenthus pretreatedwas analysedthermally (Fig' TO74 S. SI]IMODAAND T. SUDO

1). Its thermal curve sho'r,san additional exothermic peak between 600' C. and 700oC. as in the caseof montmorillonite or vermiculite. Chemicalcomposition: The specimenanalysed thermally and by n-rays was also analysed chemically. rts chemical composition is as follows; SiO, 43.177a;AlzOs, fi.5a/6; TiO2,O.5I7o; Fe2O3,0.267o; FeO,0.13/6; CaO,0.52/6; MgO, 0.657o;K2O,2.8470; NazO, 0.38/6; prOs,tr.; HrO (+) 10.4870;HrO (-) 7.757a,Total, 100.2370.Itis worthy of notice that the mineral almost agrees with allevardite (caillBre, Mathieu- Sicaud,and H6nin 1950;Brindley, 1956),except that the amount of po- tassiumis slightly larger and the amount of sodiumis smallerin the present material than in the caseof allevardite.The structural formula was constructed on the basis of the number of ions in the octahedral and tetrahedral positionsa"s 12 lor the sakeof comparisonwith that of alre- vardite.

(Na6.16KsszCao.oa)(Als.szFef,'oLFei-:orl4go.oE)(Si6 rrAlr sr)Orz os(OH)z nn

Treatment with water or ethyleneglycol gave no unique results; they seemto vary slightly even from portion to portion in a hand specimen. Treatment with water always caused the 26 A reflectionto be repraced by the 29.2+ | A reflection,but treatment with ethyleneglycol causedno expansionin somespecimens, yet clearexpansion to about z8 A in others. Treatment with ammonium nitrate solution (1 N) by boiling for 10 minutes causedno remarkable changeexcept the 26 A and 13 A reflections becamesomewhat broad. X-ray analysis: fn generai, the powder reflections are somewhat broad' and we could obtain only five basal reflections for crystallographic considerationseven with a slide of well-oriented aggregates. Thereforewe could not obtain reliable data for Fourier synthesisof the electron density or the Fourier transform (MacEwan 19.56).But we could suggestthe probability of occurrencesof spacingsof about 10 A, 13 A, and 16 A.

DrscussroN The present material has mineralogical properties almost agreeing with those of allevardite, but the following slight differencesexist between them, which must be consideredin drawing a conclusion: (1) For allevardite,the largestspacing obtained was 24.12A in air-dry unheated material, and very many basal reflectionswere found. rn the present casethe r-ray powder reflectionsare generallybroad, and only five basal reflections are available. The 26 A spacing varies slightry from specimento specimen,and even from portion to portion in a hand specimen, and is accompaniedby a tailed reflectionon its low angle side.But INTERSTRATIFIDD MIXTLIRE OF MICA CLAY MINERALS 1075

we could not find reflectionsdue to super-structuresby the small-angle scattering technique. (2) The intensities of the ;r-ray powder reflectionsof the specimen heated at 900o C. almost agree with the calculatedintensities for the regularpiling of the 10 A mica clay-mineralstructure and its silicatelayer. But the 10 A reflectionis stronqer than its calculatedvalue. which im- pliesthat the 10 A structurernuyl. alsoproduced in the firingproduct. On this basis,it may be inferredthat the 26 A spacingis not necessarily attributed to a unique regular piling of the 10 A structure and its more hydrated one, but to a more complexmixed-layer mineral; it is strongly

'l'.tsr-n 3. OssBnvro Valurs on Spacrncs eNo IxmNsrrrns or soME Basal Rrrr,ncuoxs Cowanro wrrrr rrrErR Clrcullrro Vlr-urs OsrerNpn lnou e RaNoou Mrxoo-Levrn Srnucrurr or Af C lNo B*C Srnucrunrs

I Observed values Calculated values

26.rA 26]\ I 12.9 t2 0.40 5.10 5 0.15 J.t 030 suggestedthat the 26 A spacingis due to a random piling of trvo struc- tureswith spacingsnear to 26A. As already shown by Brindley (1956),the cell heightsof the hydrated forms of the mica clay-mineralstructure are as follows: A: 9.98 A-mica clay-mineralstructure. B: 12.48A-potassium ions are completelyreplaced by a singlelayer of water molecules. C: 15.16A-potassium ions are completelyreplaced by a doublelayer of water molecules. Of course these three models are ideal ones. We can considerinter- mediate states as potassium ions are still associatedwith water layers. But since the amount of potassium associatedwith the water layers is small, the fluctuation of these cell heights may be small, and the principal cell heights can be grouped at about 10 A, 13 A, and 16 A, which were aiready suggestedby Fourier-transformdiagram. Various combinationswere consideredand finally it was shown that the structureof a random piling of 25.1A (,S+C) and.27.6A (f+C) in equal proportions,gave the calculatedintensity valuesnearest to ob- servedones (Table 3). Actually, it may be more appropriate to consider T076 S. SHIMODAAND T. SUDO thal somepotassium is still preservedeven in the B and C structuresbe- cause,as shown above,it seemsthat the firing product at 900oC. is not solely composed of a regular piling of the 10 A mica clay-mineral structure, its silicatelayer, but also of someamounts of the 10 A structure. Worthy of special attention is that a tailed reflection appears associated with the 10 A reflectionon heating.This fact suggeststhat various spacingsslightly different from the 10 A spacingoccur on heating, and also suggeststhat dehydration did not occur uniformly throughout all the layers on heating. Furthermore in relation to this fact, it is also worthy of notice that effects of treatments with ethylene glycol are not unique. This fact implies that the crystallochemicalbehavior of the potassiumions is not unique; someof them are exchangeableand others not.

OnrcrN On the basis of the model, it seemslikely that the mineral now in question is a mica clay mineral partially altered by hydration and leaching, the degreeof which may vary from place to place, as shown by variation of cell heights and organic complexes.It is believedthat with increase of hydration and leaching, potassium ions decrease and the number of layers of water moleculesincreases. Thus, although potassium ions are still preserved in the water layers, the amount can be ex- pected to be lessin the caseof a double layer than in the caseof a single layer.If the mineralis an alterationproduct of the 10 A mica clay mineral it is possiblethat the mineral is not necessarilyrepresented by the ideal model, but small amounts of potassium ions are still preserved inBandCstructures. The above statement can be supportedby the mode of occurrenceof this mineral, namely, in the central area of the deposit closely associated with pyrophyllite-diaspore which may have been formed in a ieaching environment. In the caseof the Kamikita material, the 26 A mineralsare dominant in the central area of the depositassociating with pyrophyllite and pyrite, and the 10 A mica clay mineral is rather dominant on going outward. Recently one of the writers emphasizedthe concept of the intermediate mineral in clay mineralogy (Sudo, 1957).It means a mineral which behaves as mineral "a" in some of its properties but as mineral "b" itr the other properties.These minerals have frequently beenfound between certain principal clay minerals such as montmorillonite and chlorite. It is frequently noticed that someof theseintermediate minerals have a mixed-layer structure with two kinds of mineralsr Thus the present INTERSTRATIFIED ],IIXTURE OF MICA CLAY MINERALS IO77 material may be describedas an intermediate mineral, formed during the hydration of the 10 A mica clay mineral.

AcrxowrroGMENTS The writers are indebted to Mr. Miura and S. Ota of the Chugai- mining Company, and Mr. J. Takahashi, of the Yonago Mine, for help of Geological survey. Thanks are due to the officials of the Depart- ment of Education for their kindnessin granting funds for this study. The writers wish to expresstheir deepappreciation to ProfessorG. W. Brindley of the Pennsylvania State University, for critically reading of this manuscript.

Rrrrnrxcrs Bn.ror.nv, W. F. (1950) The alternating layer sequence of rectorite, Am. M.ineral., 35, 590-595. Bntxorv, G. W. (1956) Allevardite, a swelling double-layer mica mineral. Am. M.i.neral,., 41, 91-103. Carr,rinB S., Mlrmru-srceuo A., aNo H6NrN S. (1950) Nouvel essai d,identification du mineral de Ia Table pres Allevard, I'allevardite. Bull. Soc.Frane Miner. Crist.,73, 193-20t. Kooarre H. (1958) Mineralogical study on some pyrophyllite in Japan. Miner. Jotu,n.,2, 236-2M. MacEwaN, D. M. C. (1956) Itourier transform methods for studying from lamellar systems. 1. A direct method for analyzing interstratified mixtures. Kotrl.Zei.t.,149,9G108. Suno T.^.mo Havassr H. (1955) New types of clay minerals with long spacings at about 30 A found from the altered area developed around certain ore bodies of the Hanaoka mine, Akita Prefecture. Sci,. Rep. Tokyo Kyoiku Dai.goku (Tokyo Uni.o. oJ Education), Sac. C., No. 25, 281-291. Suno, T., Hevasnr H., eNo Yororunn H. (1958) Mineral associationin ore deposits, Cloy Miner. Bul,l.3, No. 19, 258-263. Suoo T. aNo H. Kooeue (1957) An aluminian mixed-iayer mineral of montmorillonite- chlorite. Zeit. Krist.,109, 4-6. Suno T. (1957). The discovery of mixed-layer minerals and its contribution to clay science (in Japanese).Jotan. Miner. Soc.Japan,39,286-299.

Manuscript receited June 19, 1959.