American Mineralogist, Volume 67, pages 3%-398, 1982

Nomenclaturefor regular interstratifications

S. W. BnILr,vr Department of Geology and Geophysics University of Wisconsin Madison, Wisconsin 53706

It is now generally accepted that speciesnames even and odd subordershave closely similar diffrac- can be given to regularly interstratified clay miner- tion breadths. If any odd 00/ subordersare absent, als, in accordancewith the recommendationof the calculations must be given to show that their inten- International Association for the Study of Clays sities are too small to be observed. The coefficient (AIPEA) Nomenclature Committee (Brindley and of variation of the d(00/) values should be less than Pedro, 1970).In Part A ofthis report we suggest(1) 0.75 to demonstrate adequateregularity of alterna- criteria for defining the degree of regularity of tion. The coefficient of variation is defined as CV : alternation of different layer types that should be 100s/X, where the standard deviation for a small required to merit a name, (2) data that should be sample is provided for documentation of a regular interstrati- -x)'1"' fication, and (3) some examples of interstratifica- ':l f>(x, tions that do not merit names. In Part B of the '-l .1 report we apply the suggestedcriteria to the analy- sis of regular interstratifications previously report- Xs is an individual observed / x d(00/) value, X is ed in the literature, and make recommendations the mean of the X; values, and n is the number of regarding the usage of existing names. observed X1 values. 3. Names should not be used for less regular Part A interstratifications, for specimensthat deviate from The Nomenclature Committee makes the follow- the ideal mixing ratio or chemistry, or for less well ing recommendationsregarding the usageof species documented specimens.Instead, such specimens names for regular interstratifications. , should be characterized according to the informa- 1. Names should be restricted to regular interstra- tion available, e.g. -smectite irregular inter- tifications where the kinds of layers, their relative stratification or [1: I dioctahedral mica-dioctahedral proportions, chemical compositions, and regularity smectite]p1a.B):0.7,etc. of interstratification have been well documented.In 4. Interstratifications incorporating imperfect determining the kinds of layers present, it is impor- types of layers, which could not qualify for names tant that the swelling-shrinking behavior of each by themselves in the non-interstratified state (e.9. layer type be demonstratedrelative to water, organ- "swelling chlorite" and ",labile chlorite") do not ic solvates.and heat merit names. Interstratifications with considerable 2. In order to merit a name, an interstratification inhomogeneity of layer charge do not merit names. of two layer types A and B should have sufficient Interstratifications with only a single summation regularity of alternation to give a well defined series peak d(001) : dr * ds and no other odd suborders of at least ten 00/ summation spacings das : d6 I do not merit names, because a single peak could ds, for which the suborders are integral and the result from a short-rangeassociation of layers. 5. It is not certain that singlelayers of smectiteor rChairman of the AIPEA Nomenclature Committee. Other of vermiculite sandwiched between layers of a members A. Alietti (Italy), G. W. Brindley (USA), M. L. L. different type in an interstratification react in the Formosa (Brazil), K. Jasmund (FRG), J. Konta (Czechoslova- same way as the non-interstratified minerals would kia), R. C. Mackenzie (UK), K. Nagasawa (Japan), R. A. react to solvation, hydration, and dehydration tests. 'Rausell-Colom (Spain), (USSR). This report, and B. B. Zvyagin (1980) stated that glycerol originally prepared by the Nomenclature Committee of The Clay MacEwan and Wilson Minerals Society (U.S.), was approved by the AIPEA Nomen- solvation of a Mg-saturated clay gives the best clature Committee at its meeting of September9, 19E1. differentiation between nonswellingvermiculite and m03--004)o82l0304-03%$02.00 f94 BAILEY : REGULAR INTERSTRATIFICATION S

swelling smectite, as defined according to the pres- Recommendations: The name aliettite is justified ent boundary between these types of according to our criteria. We suggestthat trioctahe- 0.6 layer charge per formula unit. We suggest,for dral smectite be substituted for in the interstratifications where the distinction between definition. vermiculite and smectite may be critical, that both glycerol and ethylene glycol solvation be used in Corrensite connectionwith severalsaturating cations. Corrensitewas definedby Lippmann (1954)as a 6. Becauseof the difficulty in determination, the regular 1:1 interstratification of trioctahedral chlo- speciesof smectiteinvolved shouldnot be specified rite and trioctahedral swelling chlorite. The swelling in the definition of the speciesof a regular interstra- chlorite component was characterizedas expanding tification, beyond its dioctahedral or trioctahedral from 14A in thicknessto about 18A upon solvation nature. Names so defined are apt to be of more with glycerol but resistant to collapse to 10A upon widespread and practical usage than would be the heating. Becauseswelling chlorite is believed to be case for a more restrictive definition. Where feasi- a defect form of chlorite that would not merit a ble, however, the smectite speciesmay be noted by species name if it occurred by itself, we do not the investigator and used to characterizea particu- consider this definition of corrensite as valid. In lar occurrence in more detail. subsequent articles, Lippmann (1956, 1960) has referred to the swelling component in corrensite as vermiculite (swelling chlorite) and as montmorillon- Part B ite. Because different specimenswere involved in We have applied our criteria to examples of these latter cases, it is not certain that the same regular interstratifications cited in the literature. swelling componentwas involved in each. We conclude that the criteria are not unduly restric- Our analysis of over 30 literature citations of tive, and we use our evaluationsto make recom- corrensite, selectedfrom a comprehensivebibliog- mendationsregarding acceptanceor modification of raphy and data compilation by Hauff, Hayhurst, existing speciesnames. and Hanatani (in preparation) for the U.S. Geologi- cal Survey, predominant Aliettite shows that the current usage of the name corrensite is for a 1:l regular The name aliettite was given by Veniale and van interstratification of trioctahedral chlorite arid trioc- der Marel (1969) to a 1:1 regularly interstratified tahedral smectite. Documentation in many casesis -saponite. Three occunences have been report- poor, and the swelling componenthas been various- ed: (1) in serpentinizedrocks from Monte Chiaro in ly identified as smectite, , saponite, the Taro Valley of Italy (Alietti and Mejsner, 1980, and vermiculite. Severalof the sampleshave coeffi- and earlier reports by Alietti), (2) in weathered cients of variation below 0.75 with the even and odd serpentinefrom Ferriere in the Nure Valley of Italy (X)/orders of equal diffraction breadth. The data of (Veniale and van der Marel, 1968,1969), and (3) in Schlenker (1971) for corrensite from the Middle Precambrian dolomites in the Congo (Guenot, Keuper shales and marls in Germany are the best r970). encounteredin our survey for documentationof the The study of Alietti and Mejsner (1980)for the smectitic component. It was shown that the mineral Monte Chiaro sample is the most rigorous. It is expands to 3l-324 not only with ethylene glycol shown that 10 orders of basal reflections occur in but also with l00Vo relative humidity. Most other the untreated samplewith equal diffraction breadths studies used only ethylene glycol solvation or used for the even and odd values. Although three of the glycerol without Mg-saturation. The Middle Ke-uper ten orders were unobserved,calculation of theoreti- corrensite has a mean d(001) value of 29.14 air cal structure amplitudes shows that they are too dried with CV : 0.57and a meand(001) of 31.054 weak to be observed. The coefficient of variation glycolated^withCV : 0.61. Corrensitecollapses to for untreated material is 0.04. The bulk chemical 23.5-24.0L upon heatingat 550"C. compositions and the d valuesfor natural, solvated, Regular 1:I interstratifications of trioctahedral and heated specimens are consistent with the au- chlorite and trioctahedral vermiculite also are thors' interpretation. The diagnostictreatments and known to exist. Well documented examples have d values are: ^d(001)untreated 24.804, glycerol been provided by Johnson (1964) from a soil in solvated27.104, heatedat 600'C 9344. Pennsylvania and by Gradusov (1969)from an as- BAILEY : REGULAR I NTERSTRATIFICATION S bestos deposit in the southern Ural Mountains. The trioctahedral chlorite and talc, but further docu- material from Pennsylvania gives 13 observed or- mentation of the material's behavior with respect to ders of a 28.524 repeat in the air dried statewith all organic solvates and heating is desirable. orders equally sharp and a coefficientof variation of Rectorite 0.23. The Mg-saturatedmaterial does not expand From chemical analyses, cation exchange mea- further upon ethylene glycol solvation, the K-satu- surements, X-ray examination, electron microsco- rated material air-dried shrinks to 24.364, and py, and IR absorption spectra, Brown and Weir collapses to 24.064 upon heating at 475'C. Mg- (1963)established, in general agreementwith previ- saturated material from the Ural Mountains does ous work by Bradley (1950),Cailldre et al. (1950), not expandbeyond 28.54 upon solvationwith glyc- Brindley (1956), and others, that rectorite and alle- erol or ethylene glycol. A calculated Fourier trans- vardite are the same mineral. The name rectorite form and a one-dimensionalelectron density projec- has priority over allevardite. The structure consists tion both indicate regularity of alternation of the of pairs of dioctahedral2:l layers; alternateinter- two components for the Ural's specimen. The dif- layers are micaJike and montmorillonite-like. The fraction patterns of both chlorite-smectiteand chlo- non-swelling, mica-like interlayers contain about rite-vermiculite are characterizedby very low in- 0.E5univalent cations per mica formula unit and the tensities for all odd orders of 00/ except the first swellinginterlayers about 0.35 univalent cations per order for untreated material. Johnson (1964) has smectite formula unit. calculated that the intensities of these odd orders X-ray basal reflections up to 0,0,26 for air-dry should be very small. for material heated to 500" and Recommendations:The definition of corrensite material, 0,0,22 and 0, 0, 22 for ethyleneglycol should be changed to a I : I regular interstratifica- 900"C for 24 hrs., were given by Brindley (1956) tion of trioctahedral chlorite with either trioctahe- saturated material electron density pro- dral smectite or trioctahedral vermiculite, the for- together with one-dimensional jections. for the mineral from Alle- mer constituting low-charge corrensite and the Basal spacings various conditions were as latter high-charge corrensite. Glycerol solvation of vard, France, under water 28.364, air dry natural Mg-saturated material and other methods of deter- follows: immersed in 450'C for 24 hrs. 19.09A, after mining layer charge can be used for differentiation state 24.634, after 19.40A, after ethylene glycol of the two types. A different name for high-charge 900'C for 24 hrs. even and odd 001 orders corrensite may be appropriate if further research treatment 26.414. The breadths, and the coefficient establishes the feasibility of reliable distinction of have similar diffraction the air dry material is 0.50. More the smectite and vermiculite components. of variation for detailed study of the cornponentlayers of rectorite Kulkeite from Baluchistan by Kodama (1966) led to the consists of paragonite- Abraham et al. (1980)defined kulkeite as a I : 1 conclusion that the mineral layers having beidellitic regular interstratification of trioctahedral chlorite like layers and expansible compositions. Similar regular and talc. The mineral occurs as single crystals in a and montmorillonitic in which K is the dominant cat- metamorphosedevaporite sequencein Algeria. The interstratifications have been reported by ideal formula is listed as MgsAl(Si7Al)O20(OH)r0, ion in the mica component (1980) in bentonites from British but a imall substitution of NaAl + Si (about 0.40 Pevear et al. in which Ca is the dominant cation atoms) takes place in the actual mineral. The NaAl Columbia, and (1981)from the Sanomine, Japan' is assumgd to be in the talc cgmponent, similar to by Matsuda et al. The name rectorite is justi- analyzeddiscrete talc in the samerock. The authors Recommendations: regular interstratification of a diocta- say the chlorite-talc 1:1 interstratification also can fied for a 1:1 and dloctahedral smectite. The kind of be pictured as a 2:1 regular alternation of talc and hedral mica not be specifiedin the definition. A brucite units. smectite should prefix be used to specify the The mean d(001)value for ten basal reflections is Na-, K-, or Ca- can cation in the mica component. 23.7154,with CV :0.12. No solvationor heating dominant interlayer tests are mentioned. Optical properties and cell dimensionsare listed. Tarasovite Recommendations:Thename kulkeite appears to The name tarasovite was given by Lazarenko and be justified for a 1:l regular interstratification of Korolev (1970)to a2:1 regularinterstratification of BAILEY : REGULAR INTERSTRATIFICATIONS

mica layers and rectorite, which also can be de- of di,dioctahedral chlorite (donbassite)with either scribed as a 3:1 alternation of mica and smectite dioctahedral or trioctahedral smectiteand of diocta- layers MMMS. The mean d(001)repeat is 43.814 hedral smectite with either di.trioctahedral chlorite for l1 observed orders in the original state, with CV (sudoite or cookeite) or tri,dioctahedral chlorite : 1.73, and 46.104 glycolated for 20 observed (unknown to date). (In di,trioctahedral, etc., the orders, with CV : 1.08(omitting 0,0,18as a print- first designator refers to the octahedral sheet in the ing error). The materialcollapses to 19.54with loss 2:l layer of chlorite and the second designator to of the higher periodicity on heating at 600.C, but the interlayer sheet.) rehydrates with time to 424. The authors presented chemical analyses, a Fourier transform, a one- References dimensional electron density projection, and calcu- Abraham, K., Schreyer,W., Medenbach,O., and Gebert, W. lation of 00/ structure amplitudes to propose that (1980)Kulkeit, ein geordnetesl: I Mixed-Layer-Mineral zwis- (abstr.). the two mica layers contain in the inter- chen Klinochlor und Talk Fortschritte der Mineralo- h.oNao.+ gie, Beiheftl,5E,4-5. layer regions and have the 2Ml stacking Alietti, A. and Mejsner, J. (1980) Structure of a talc/saponite sequence; the mica layer in the rectorite portion mixedJayer mineral. Clays and Clay Minerals, 2E, 3EE-390. contains Na6.5(H3O)6.5,with Na and Ca as ex- Bradley, W. F. (1950)The alternating layer sequenceof rector- changeable cations in the smectitic layer of the ite. American Mineralogist, 35, 590-595. rectorite. Brindley, G. W. (1956)Allevardite, a swelling double-layermica mineral. American Mineralogist, 41, Recommendations: 9l-103. Although the sample has Brindley, G. W. and Pedro, G. (1970) Report of the AIPEA been documented well, the degree of regularity is Nomenclature Committee. AIPEA Newsletter No. 4. 3-4. not sufficient to warrant a species name (coeffi- Brown, G., Bourguignon, P. and Thorez, J (1974) A lithium- cients of variationsof 1.73and 1.08relative to the bearing aluminian regular mixed layer montomorillonite-chlo- desired 0.75). The name can be reservedpending rite from Huy, Belgium. Clay Minerals, 10, 135-144. Brown, G. and Weir, A. H. (1963)The identity of rectorite and discovery of a more regular interstratification of the allevardite. Proceedingsof the International Clay Conference, same type. Stockholm, l, 27-i5 and 2, 87-90. Cailldre, S., Mathieu-Sicaud, A., and H6nin, S. (1950)Nouvel Tosudite essai d'identification du mindral de la Table prds Allevard, I'allevardite. Bulletin de Soci€td Francaisede Min€ralogie et The name tosudite was given by Frank-Kamen- des Cristallographie, 73, 193-201. Frank-Kamenetskii, etskii er al. (1965) to a 1:1 regularly interstratified V. A., Logvinenko, N. V., and Drits, V A. (19ti5) Tosudite-a new mineral, forming the mixed-layer dioctahedral chlorite-smectite. At least 14 occur- phase in alushtite. Proceedings of the International Clay renceshave beenreported, and six ofthese samples Conference,Stockholm, 2, l8l-1E6. give 10 or more orders of basalreflections. Only the Gradusov, B. P. (1969) An ordered trioctahedral mixedJayer two samplesfrom Tooho, Japan(Nishiyama et al., form with chlorite and vermiculite packets. Doklady Akade- miya Nauk, SSSR, lE6, l4O-142(English 1975),and Huy, Belgium(Brown et al.,1974), meet translation). Guenot, B. (1970)Etude d'un mindral argileux du type interstra- the requirements that the coefficient of variation tifi6 talc-saponite trouv6 dans le Prdcambriendu Congo Kin- should be less than 0.75 (0.33and 0.55, respective- shasa.Bulletin du Groupe Francais des Argiles, 22,97-104. ly). Both samplesgive d(001)values of 31.14 on Johnson, L. J. (1964) Occunence of regularly interstratified glycolation and d(001)of 23.3A on heatingat 500- chlorite-vermiculite as a weathering product of chlorite in a soil. American 650'C, although the mean d(001) of the Tooho Mineralogist, 49, 556-572. Kodama, H. (1966) The nature of the component layers of specimen under air-dry conditions is 0.65A hrger rectorite. American Mineralogist, 5 I, 1035-1055. that that of Huy (29.474and 28.82A,respectively). Lazarenko, E. K. and Korolev, Yu. M. (1970)Tarasovite, a new With glycerol solvation the Huy sampleexpands to dioctahedral ordered interlayered mineral. Zapiski Vsesoyuz- 32.24. The d(060)values are 1A924 for the Tooho hogo MineralogicheckogoObshchestra, 99, 214-224. Lippmann, (1954) specimenand 1.5064for the Huy specimen. F. Uber einen Keuperton von Zaiserweiher bei Maulbronn. Heidelberger Beitrdge zur Mineralogie und Recommendations:Thename tosudite is valid for Petrographie,4, 130-134. a 1: I regular interstratification of chlorite and smec- Lippmann, F. (1956)Clay minerals from the R

intercalation complexes of clay minerals. In G. W. Brindley Schlenker, B. (1971)Petrographische Untersuchungen am Gips- and G. Brown, Eds., Crystal Structures of Clay Minerals and keuper und Lettenkeuper von Stuttgart. OberrheinischeGeo- their X-ray Identification, chapter 3, Monograph 5 of Mineral- logische Abhandlungen, 20, 69-102. ogical Society, London. Veniale, F. and van der Marel, H. W. (1968) A regular talc- Matsuda,T., Nagasawa,K., Tsuzuki,Y., and Henmi, I(. (1981) saponite mixed-layer mineral from Ferriere, Nure Valley Regularly interstratified dioctahedral mica-smectitefrom Ro- (Piacenza Province, Italy). Contributions to Mineralogy and seki deposits in Japan. Clay Minerals, 16,91-102. Petrology, 17, 237-254. Nishiyama,T., Shimoda,S., Shimosaka,K., and Kanaoka,S. Veniale, F. and van der Marel, H. W. (1969)Identification of (1975)Lithium-bearing tosudite. Clays and Clay Minerals, 23, some l:l regular interstratified trioctahedral clay minerals. 337-342. Proceedingsof the International Clay Conference, Tokyo I, Pevear, D. R., Williams, V. E., and Mustoe, G. E. (1980) 233-244. , smectite, and K-rectorite in bentonites: relation to coal rank at Tulameen, British Columbia: Clays and Clay Manuscript received, October 15, l98I; Minerals, 28,241-254. acceptedfor publication, December 4, 1981.