American Mineralogist, Volume 68, pages 972-980, I9E3

A review of the todorokitebuserite problem: implications to themineralogy of marine manganese nodules

Rocen G. Bunns, VrRcrNre Mee BunNs AND Henr-eN W. SrocrN,rnNrl

Department of Earth and Planetary SciencesMassachusetts Institute of Technology

Cambridge Massachusetts 02 139

Abstract

The names todonokite and buserite are two competing terminologies for the abundantmanganese oxide mineral occurring in deep-sea manganese nodule deposits. Proponents ofthe buserite phase claim it to be the parent of todorokite which is considered to be amixture of buserite and its breakdown products birnessite and manganite (rMnOOH). Therecent experimental evidence demonstrating the integrity of todorokite (a tekto-manganatewith multidimensional tunnels formed by walls of edge-shared [MnOe] octahedra) has led toan assessment of the proposed relationships between synthetic buserite (a phyllo-manganate containing layers of edge-shared [MnOo] octahedra) and its transformationproducts. Inconsistencies are found to exist in published electron diffraction data underly-ing the proposed topotactic transformation mechanism of synthetic birnessite platelets toacicular "7MnOOH" crystallites suggested to constitute natural todorokites. The crystalchemistry of todorokite structure'types is discussed. We suggest that three types of atomicsubstitution occur in the tunnel structures: first, substitution of Mn2* by other divalentcations (e.g., Mg2*, Niz+, Cu2*, Zn2*1 in the "walls" formed by chains of edge-shared[MnOe] octahedra, characteristically three octahedra wide; second, replacement of Mna*by similar sized cations (e.g., low.spin Co3+) in the "ceilings" or "floors" typically three,but as many as seven or more, [MnOe] octahedra wide; and third, in the tunnel interiorsadjacent to Mna+ vacancies in the "ceilings." Here, a variety of large cations (K+, Ba2+,Ag*, Pb2*, Ca2+, Na*;, H2O molecules, and hydrated transition metal cations could beaccommodated. We recommend that the name todorokite be universally adopted for thepredominant manganese oxide mineral accommodating divalent cations of nickel andcopper in marine manganiferous concretions.

Introduction

Ono of the most common manganese oxide phasesdetected in marine ferromanganese nodules, crusts andrnetalliferous sediments is that with characteristic X-ravdifraction lines at 9.5-9.8A and 4.8-4.9A. Chemical anielectroR microprobe analyses of this nodule l0A pfiasephase indicate that it is a hydrated Mn-Mg-Ca-Na-Ni-Cu oxide. Its ability to concentrate Ni and Cu to severalweight percent (Burns and Burns, 1978, 1979b) makesdeep-sea manganese nodules the focus of considerablescientific and economic interest (Glasby, 1977; Crortan,l9ro).

X-ray diffraction patterns similar to that of the nodulel0A phase are shown by two other species: todorokite, aMn-Mg-Ca-Na-K oxide originally reported in a non-maririe environment (Yoshimura, 1934); and derivatives

I Present address: Sandia National Laborataries, Albuquer-que, New Mexico 87185.

0003-(M)vE3/09 l 0-0972$02.00

of a synthetic sodium manganese oxide hydrate (Feitk-necht and Marti, 1945; Wadsley, 1950a, b) called "l0Amanganite." As a result, the phase occurring in manga-nese nodules was identified as either todorokite (Straczeket al., 19ffi; Hewitt et al., 1963), or synthetic "l0Amanganite" (Buser, 1959). This dual nomenclature haspervaded the literature on marine manganese deposits(Burns and Burns, 1977a, 1979b), d-espite the complaint(Arrhenius, 1963) that the term "l0A manganite" causesconfusion with the mineral manganite (rMnIllOOH).

Giovanoli et al. (1971) proposed that the nodule l0Aphase be called buserite, in honor of W. Buser, andsuggested that buserite has the same crystal structure assynthetic "l0A manganite." Buserite was accepted as amineral name by the Commission on New Minerals by asmall majority (M. Fleischer, personal communication,1974), and although it is listed in Hey and Embrey (1974),the name is absent from Fleischer's (1980) "Glossary ofMineral Species." Giovanoli and coworkers (1971, 1975)further contended from X-ray diffraction and electron

972

microscopy measurements of synthetic manganese ox-ides that natural todorokites are a mixture of buserite andits decomposition products birnessite and manganite (rMnOOH). The Swiss group also recommended that to-dorokite should be discredited as a valid mineral (Giovan-oli and Brutsch, 1979a, b; Giovanoli, 1980). However,recent experimental evidence verifying the integrity oftodorokite includes results derived from infrared spec-troscopy (Potter and Rossman,1979), electron diffraction(Chukhrov et al., 1978, 1979, l98l; Siegel, l98l), hiehresolution transmission electron microscopy (snreu)(Turner and Buseck, 1979, 1981, Turner et al., 1982), andextended X-ray absorption fine structure (exnrs) mea-surements (Crane, 1981) of manganese (IV) oxide miner-als.

The conflicting viewpoints about todorokite have led usto critically evaluate experimental results and literatureon buserite and todorokite. We summarize here evidencedemonstrating that buserite and todorokite are distinctphases, describe recent observations of todorokite inmarine manganese nodules, and discuss possible crystalstructures and site occupancies of cations in the todoro-kite polymorphs. Finally, we recommend that todorokitetakes precedence over buserite for the predominant l0Aphase in manganese nodules.

Survey of the crystal chemistry of todorokite andbuserite

Todorokite

Todorokite occurs in a variety of terrestrial deposits.At the type locality in Japan, it is formed as an alterationproduct of inesite (Yoshimura, 1934). The economicallyimportant todorokite deposits of Cuba, such as CharcoRedondo in Oriente Province, were formed syngenetical-ly during the Eocene in foraminiferal oozes on the sea-floor near fumerolic hot springs accompanying volcanicactivity. Subsequent to lithification, some todorokite inthe weathering zone has altered slightly to manganite,with more intense weathering yielding pyrolusite (Simonsand Straczek, 1958; J. A. Straczek, personal communica-tion, 1978). In other terrestrial localities, todorokite isreported to be of secondary origin (Frondel et al.,1960;Larson, 1962). X-ray difraction data show that consider-able variations exist between relative intensities of com-parable lines for different todorokite samples. Faulring(1962) attributed these large intensity variations, as wellas the diffuseness of certain reflections, to preferredorientation of the fibrous crystallites of todorokite studiedby her from Charco Redondo, Cuba. Thus, the mostintense lines at 9.65A and 4.82A observed when X-raysare difracted parallel to the fiber axis were weak orundetected for X-rays diffracted perpendicular to thecrystallite axis. The most intense lines in the latterorientation occur at 2.424 and 1.424. Faulring ( 1962) alsosuggested that manganite was topotactically intergrown

973

with the Cuban todorokite, either forming simultaneouslywith todorokite, or resulting from its alteration. The latterobservation was later cited as evidence by Giovanoli andcoworkers (1971, 1975) for rejecting todorokite as a validmineral.

Chemical analyses of todorokites in continental depos-its (Straczek et al.,1960; Frondel et aI., 1960; Nambu elal., 1964) show that manganese is present in two oxida-tion states, and that MnII/MnIv ratios fall in the range0.15-0.23. Significant amounts of Mg are also present,suggesting that relatively small divalent Mg2* and Mn2*ions are essential constituents oftodorokite. The analysesshow that Ca2*, Na* and to lesser extents K+, Ba2*,Ag* lRadtke et aL, 1967) and Zn2* (Larson, 1962), arealso common constituents. Several chemical formulaehave been proposed for todorokite, including (Ca,Na,K,Ba,Ae) (Mg,Mnz*,Zn) Mnl* O12 ' 3H2O (Frondel er a/.,1960).

Most todorokites in hand specimen consist of fibrousaggregates of small acicular crystals, although plateymorphologies have been observed. Electron micrographs(Straczek et al., 1960; Hariya, 1961; Finkelmann et al.,1974, Chukhrov et al.,1978; Burns and Burns, 1979a)reveal that the crystals consist of narrow lathes or bladeselongated along one axis (parallel to b) and frequentlyshow two perfect cleavages parallel to (001) and (100).

A structural model for todorokite was inferred from itscrystal morphology, cleavage properties and electrondiffraction data (Burns and Burns, 1977b). It was notedthat minerals of the hollandite-cryptomelane and psilom-elane (romandchite) groups also have fibrous or acicularhabits and two perfect cleavages parallel to the fibre axis,as do a variety of synthetic Tilv oxides. Since thesephases possess tunnel structures consisting ofdouble andtreble chains of edge-shared [MnOo] octahedra,2 Burnsand Burns (1977b) proposed that todorokite might alsohave a tunnel structure based on chains of multiple widthedge-shared [MnOo] octahedra extending along its b axis.The correlation is borne out when comparisons are madebetween the unit cell parameters of todorokite, psilome-Iane, and hollandite-group minerals. Furthermore, todor-okile and psilomelane both contain essential Mn2* ions,and i! psilomelane these divalent cations are located inspecific positions in the triple chains of edge-sharedoctahedra. Burns and Burns suggested that Mn2*, Mg2*,Ni2+, etc. might also be located in analogous sites intodorokite. The larger Ca2*, Na*, K+, Ba2+, erc. ionsand H2O molecules in todorokite were envisaged tooccupy large tunnels in a psilomelanelike tektomangan-ate structure (Burns and Burns, 1979a). The site occupan-cies of cations in todorokite are discussed later.

Recent electron diffraction and high resolution trans-

2 HRTEM measurements show [2x2] tunnels in hollandite-cryptomelane and [2x3] tunnels in psilomelane (romandchite)(Turner and Buseck, 1979).

BURNS ET AL.: REVIEW OF THE TODOROKITE-BUSERITE PROBLEM

974 BURNS ET AL.: REVIEW OF THE TODOROKITE-BUSERITE PROBLEM

mission electron microscopy measurements of naturaltodorokites have confirmed the structural model for to-dorokite proposed by Burns and Burns (1977b,1979a).Selected area electron diffractio-n (SAD) patterns of theCuban todorokite with a :9.754 (Chukhrov et al..1978.1979) revealed three less intense reflections (d : 9.754)corresponding to four sub-cells along the a* axis betweenthe strong reflections (d : 2.444) forming a pseudohexa-gonal net (Straczek et al., 19ffi1' V. M. Burns, unpub-lished data). This is in contrast to SAD patterns ofsynthetic buserite discussed later which contain onlythree sub-cells along the ax axis (Giovanoli, 1980; V. M.Burns, unpublished data). Hnrev images of the Cubantodorokite (Turner and Buseck, l98l; Chukhrov et al.,1978) revealed it to have a tunnel structure, in which[3x3] tunnels with dimensions 9.754 x 9.594 predomi-nate (Turner and Buseck, l98l: Turner et al., 1982).Chukhrov et al. (1978, 1979, l98l) studied todorokitesfrom other localities, including a Pacific Ocean manga-nese nodule, and found SAD patterns having b (2.844)and c (9.59A) parameters identical to the Cuban todoro-kite, but larger a parameters (14.64, 19Szland 24.384).Trilling intergrowths of these todorokites produce plateymorphologies, while SAD patterns of the platelets revealnets of hexagonal, triangular, and rhombic cells with fouror five weaker reflections (d : 14.64 or 24.38A) betweenthe strong reflections (d : 2A4h along the a* axes(Chukhrov et al., 1978;1979; Burns and Burns, 1979a).Chukhrov et a/. proposed that a family of todorokitespecies might exist having a unit cell parameters whichare integral multiples of 4.88A. Lattice image photo-graphs of todorokite published by Chukhrov et al. (1978)also show evidence of structural disorder. In addition tothe structure-types suggested by the variability of the aparameter, Chukhrov et al. fonnd variations in the peri-odicity of the (100) planes. Thus, in addition to the mainsuper-periodicity oftodorokites producing a = 9.75A and14.6A, defect layers wrtn I.SZA and 12.4A were ob-served. These layers are multiples of 2.444, the height ofthe base (i.e., dimensions of a [1ll] axis) of a MnOcoctahedron.

Subsequent HRTEM images of todorokites from terres-trial deposits (Turner and Buseck, l98l) and from manga-nese nodules (Turner et al.,1982; Turner and Buseck,1982) revealed them to be intergrowths of tunnels ofdifferent widths. Although [3x3] dimensional tunnelspredominate in todorokite, occasional tunnels with [3x2],[3 x 4], [3 x 5] , [3 x 8] and higher dimensions were observedin crystals from both terrestrial and manganese noduledeposits.

Buserite

Buserite, which is assumed (Giovanoli et al.,l97l) tobe equivalent to synthetic sodium manganese (II, In)manganate (IV) hydrate or " l0A manganite", is preparedby oxidation of fresh Mn(OH)z suspensions in cold aque-ous NaOH solutions by molecular oxygen (Giovanoli er

al., 1970a). Synthetic buserite consists of elongated plate-lets which give an electron diffraction pattern with pseu-dohexagonal symmetry and weak superstructure reflec-tions (d : 7.38A) with a pgriodicity of three betweenmajor reflections (d : 2.464) in the basal (001) planes(Giovanoli, 1980). X-ray difraction measurements (Gio-vanoli et al., 1975; Giovanoli, 1980) reveal that thediagnostic intense lines representing basal plane separa-tions occur in sodium buserite at l0.l-10.2 and 5.0-5.1A;these distances are significantly larger than those mea-sured in samples of marine manganese oxide deposits andin todorokite. A variety of buserite derivatives can besynthesized by cation exchange reactions (Wadsley,1950a, b; Giovanoli et al.,1975; Giovanoli and Arrhenius,1983), and the Mg2+, Cu2*, Ni2*, erc. derivatives havesmaller basal plan spacings than the sodium-bearing par-ent buserite (Giovanoli, 1980). The transition metal deriv-atives appear to have greater thermal stabilities and to bemore resistant to dehydration than the parent Na buserite(Giovanoli et al. 1975; Giovanoli and Arrhenius, 1983).

Sodium buserite readily decomposes when exposed toair. Drying over PaOls in vacuo leads to the formation ofsodium birnessite, NtuMnr+Ozz ' 9H2O, which also has aplatey morphology (Giovanoli et al., 1970a; Giovanoli,1980). The strongest basal plane X-ray difraction linesthen occur at 7.1 and 3.55A. The electron diffractionpatterns of the platelets again show pseudohexagonalsymmetry and superstructure with only triple periodicity(Giovanoli et al., 1970a; Giovanoli, 1980). The lattersuperstructure forms the basis for postulating a vacancy-ordered layered structure for birnessite (Giovanoli et al.,1970a) based on the chalcophanite structure (Wadsley,1955). The structure of chalcophanite, Zn2MnoOra '

6H2O, consists of layers of edge-shared [MnOe] octahe-dra and single sheets of water molecules between whichthe Znz+ ions are located. The octahedral layer is only 6/7filled by Mna+; that is, one out of every seven octahedralsites is vacant. The Znz* atoms occur above and belowthe vacancies in the octahedral layers, and the stackingsequence along the c axis is -O-Mn-O-ZwH2O-Zn-O-Mn-O-, so that consecutive [MnOe] Iayers are about7.164 apart. In the structure of sodium birnessite pro-posed by Giovanoli et al. (1970a), one out of every sixoctahedra is vacant, and Mn2* and Mn3* are suggested tolie above and below vacancies in the octahedral layer.The positions of the Na* ions are uncertain. Buserite isconsidered to have a similar layered structure, but addi-tional OH- ions and HzO molecules are present so thatthe v-acancy-ordered layers of [MnOo] octahedra are9.6-10.1A apart (Giovanoli, 1980). Recent intercalation stud-ies of synthetic burserite (Paterson, 1981; Giovanoli andArrhenius, 1983) correlate with its proposed layer struc-ture.

When NaaMn ro,Ozt ' 9HzO is refluxed with dilute HNOrat 40'C, a sodium-free birnessite, Mn7O13 ' 5H2O, isobtained (Giovanoli et al., 1970b) which again has aplatey habit. The electron diffraction patterns of (001)

BURNS ET AL.: REVIEW OF THE TODOROKITE-BIISERITE PROBLEM 975

platelets show hexagonal symmetry, but streaks in thesub-cell are interpreted to be indicative of vacancy disor-der ing in the IMnOe] octahedral layers. WhenMn7O13 . 5H2O is subjected to careful reduction withcinnamyl alcohol, the platelets break down to extremelythin needles (Giovanoli et al.,l97l). These needles failedto give an X-ray diffraction pattern, but were believed tobe IMnOOH (manganite). Giovanoli et al. (1971) pro-posed a mechanism for the topotactic transformation ofMn7O13 . 5H2O to "yMnOOH". They believed that the(fi)l) layers of linked lMnlvoul octahedra in birnessitecorrespond to chains of [MnrII(OH,O)o] octahedra in the(010) plane of "7-MnOOH," and that the [100] directionof birnessite became the needle axis (c axis) of the ,.yMnOOH" formed topotactically. The structural correla-tions suggested by Giovanoli et al. (1971) are shown inFigure l, and are discussed later.

These observations for buserite and its dehydrationproducts formed the basis for Giovanoli and his cowork-ers (1971; 1975:- l979a,b; 1980) to propose that world-widetodorokite ores actually consist of buserite partly dehy-drated to birnessite and partly reduced to needles ofmanganite (Giovanoli and Biirki, 1975). According to thisview, the small amount of buserite in the ores accountsfor the characteristic X-ray diffraction pattern, while alarge amount of fibrous, nearly amorphous manganiteaccounts for the morphology. Apparently this "y-MnOOH," which could not be detected by X-ray diffrac-tion (due to its small crystallite size), is recognized on thebasis of electron microscope observations of morpholo-gy. Thus, the argument against the integrity of todorokiteas a mineral depends on the analogy with the topotacticdecomposition of synthetic Mn7O13 . 5H2O (birnessite) towhat could be unambiguously identified as y-MnOOH onthe basis of morphology and electron diffraction. Appar-ently, the morphology argument was considered strongerthan the diffraction evidence, as the electron diffractionpatterns observed for the hypothetical todorokite decayproducts do not match those of the experimental birnes-site decay product.

Information from infrare d sp e ctros copy

Some of the problems of identification and structuralcorrelations for todorokite and buserite have been ad-dressed by infrared spectroscopy (Potter and Rossman,1979). In a study of numerous natural and syntheticmanganese oxides, Potter and Rossman (1979) demon-strated that it is possible to distinguish between different[MnOo] structural linkages by their spectral profiles in themid-infrared region. They have shown, for example, thatthe proposed layer structure of birnessite is supported byits infrared spectrum, and have confirmed the identity ofnatural birnessite with its synthetic analogues. The infra-red spectra also show that synthetic buserite and birnes-site have analogous structures, the shift from the l0A to7A basal plane spacing in X-ray diffraction patterns being

mangan i t e b i rness i t e

Fig. l. Schematic structural diagrams relating the syntheticbirnessite MnTO;3 . 5H2O phase (right) to its proposed topotactictransformation product manganite yMnOOH (left) [fromGiovanoli et al.,1971, Fig. l l l .

the result of water loss alone rather than to a structuralrearrangement of the [MnOe] framework.

The infrared spectra of natural todorokites (Potter andRossman, 1979) indicate that this mineral is not analogousto any synthetic phases, including buserite and its cation-exchanged derivatives, or to any decomposition productsof buserite such as birnessite and manganite. Potter andRossman suggested that the todorokite spectra are con-sistent with either a layer structure of linked [MnOe]octahedra containing vacancies, or a highly polymerizedchain or tunnel structure with quadruple chains. Theirhypothesis conforms with observations of the multidi-mensional tunnels found in subsequent studies ofnaturalterrestrial and marine todorokites (Turner and Buseck,1981, 1982; Turner et al.,1982).

Discussion

The information derived from infrared spectroscopy,electron diffraction, and Hnreu measurements appear todemonstrate conclusively that todorokite and buserite aredistinct phases having different linkages of edge-shared[MnOo] octahedra. Nevertheless, Giovanoli and cowork-ers (1971; 1975; 1979a, b; 1980; 1983) maintain thatbuserite is the primary phase and that todorokite is anassemblage of buserite and its breakdown productsbirnessite and manganite. However, close examination ofthe evidence presented by the Swiss group reveals appar-ent inconsistencies in interpretations of their data orquestionable relevance of the results to geochemicalprocesses.

First ofall, there are the inferences from the laboratorystudies that are not correlatable with sedimentary mineralformation. For example, the decomposition studies ofsynthetic Mn7O13 . 5H2O to the presumed manganite (yMnOOH) phase reported by the Swiss group were notperformed on buserite, but on a sodium-free birnessiteunder conditions (reduction with cinnamyl alcohol) notfound in nature. Also, the initial dehydration of Nabuserite by drying over PaO16 in vacuo and conversion of

976

Na birnessite to Mn7O13 . 5H2O by refluxing with diluteHNO3 at 40'C are unlikely geochemical processes. Fur-thermore, the implication that todorokite in deep-seamanganese nodules represents buserite which has dehy-drated, in situ to birnessite is untenable. Second, theSwiss group did not prove that crystal morphology ismore definitive than diffraction evidence for identifyingmanganite, but instead assumed that the fibrous pseudo-morphs of decomposed Mn7O13 . 5H2O must be manga-nite, despite diffraction evidence to the contrary (seebelow). Third, although some todorokite ores in theweathering zone in Cuba have been slightly weathered tomanganite and pyrolusite, there is no reason to believethat the fibrous habit of this material is due to a largecomponent of non-diffracting manganite since fibrosity isan inherent property of all todorokites from other local-ities. Gram for gram, many todorokite ores give charac-teristic X-ray powder diffraction patterns (nor containingmanganite d-values) as intense or more intense than thoseof synthetic buserites. Such results are inexplicable if theores consist of a mixture of non-difracting manganite andundecomposed buserite.

A more rigorous examination of the evidence for theproposed birnessite-+manganite topotactic relationship(Giovanoli and Stiihli, 1970; Giovanoli, et al., 1970a,b,l97l) reveals that the structural interpretation is incor-rect. In the schematic structure of manganite (e.g., Figurell in Giovanoli et a\.,1971), reproduced here in Figure I,the authors inadvertently switched the a and b transla-tions (with respect to the structure). In order to have theelectron diffraction measurements actually agree with thetrue manganite structure (Buerger, 1936; Dachs, 1963),the idealized structure depicted in Figure I must berotated 90'about the c axis. In this orientation, the planartriangular [111] surfaces of the [Mnur(OH,O)6] octahedrain manganite no longer coincide with similar surfaces ofthe layers of [MnIvOu] octahedra in birnessite, so that thepostulated topotaxy is much less evident. The structurecorrelation between manganite and birnessite suggestedby Giovanoli et al. in Figure I has another inconsistencyin that each row of [MnO6] octahedra in the birnessitestructure is drawn to align with two rows of octahedra inthe manganite structure one of which is devoid of Mna+ions, resulting in octahedra with unusual proportions. Inaddition, the structures sketched in Figure I reveal anoth-er problem (S. Turner, pers. comm. , 1982) in that thespacing between the layers of birnessite is incorrect(relative to the^manganite structure). They show-a spac-ing of about 5A instead of the true value of 7 .27 A.

One of the consequences of mislabelling the axes ofmanganite in Figure I is that the indexing of electrondiffraction patterns of the break down product ofMn7O13' 5H2O purported to represent the a*-c* zerolevel of manganite (Giovanoli et al.,1971, Filures 7 and12; Giovanoli, 1980, Figure 15) does not sarisfy criteriafor the space group B21ld deterrnined for the manganitestructure (Buerger, 1936). The difraction pattern of the

BURNS ET AL.: REVIEW OF THE TODOROKITE-BUSERITE PROBLEM

altered Mn7O13 '5H2O phase (Giovanoli et al., l97l;Giovanoli, 1980) does, however, roughly fit that expectedfor a (100) orientation for manganite (S. Turner, pers.comm., 1982). Apparently, previous experience with syn-thetic manganite (Giovanoli and Leuenberger, 1969) ledto recognition of a manganite diffraction pattern butincorrect indexing of it.

In summary, although the topotactic relationship por-trayed in Figure I has many flaws, the overall interpreta-tion of the experimental work of the Swiss group maypossibly be correct, and synthetic platey birnessite mightbreak down to acicular manganite crystallites. However,it is difficult to follow the logic presented by Giovanoliand Biirki (1975) that fibrous todorokites naturally occur-ing in marine manganese nodules and in non-marinemanganese ore deposits are mostly manganite. It be-comes clear, though, that a critical step in their identifica-tion of manganite based on crystal morphology andelectron diffraction evidence is ambiguous. Again, it isdifficult to understand how the proposed buserite parentin marine manganese nodule deposits could have dehy-drated to birnessite and decomposed to manganite in situon the deep seafloor.

Finally, although past X-ray powder diffraction mea-surements of marine ferromanganese nodules and crustshave led to ambiguous identifications of todorokite orbuserite (" 10A manganite"), it is significant that the morerecent electron microscopy techniques have identifiedonly todorokite in these seafloor deposits (Chukhrov eral., 1978, 1979, l98l; Siegel, l98l; Turner and Buseck,1982; Turner et al.,1982). Buserite has yet to be positive-ly identified in the natural environment. Such findingsundermine the recommendation (Giovanoli et al., l97l)that the "l0A phase" in manganese nodules be namedbuserite.

Crystal structure correlations

The recent HRTEM and infrared spectral measurementsdescribed earlier indicate that todorokites have tunnelstructures analogous to hollandite and romandchite (psi-lomelane). Turner and Buseck (1981) proposed a provi-sional nomenclature scheme for manganese (IV) oxidestructures in which families are designated by the symbol-ism T(m,n), where T denotes a tunnel structure and m, nare the widths of infinite chains of edge-shared [MnOe]octahedra forming the walls of the tunnels. Thus, m : Idefines the nsutite family in which T(l,l) and T(1,2)symbolize pyrolusite and ramsdellite, respectively. Inter-growths of these fundamental units, perhaps with higherdimensional tunnels T(1,3), T(1,4), erc. (Burns andBurns, 1980), characterize synthetic yMnO2 and natural-ly-occurring nsutites (Turner and Buseck, 1983). Similar-ly, T(2,n) includes the cryptomelane-hollandite, T (2,2),and romandchite (psilomelane), T(2,3), groups, togetherwith the observed coherent intergrowths of T(2,4),-T(2,7) multidimensional tunnels found in fibrous manga-nese oxide minerals (Turner and Buseck, 1979). Turner

BURNS ET AL,: REVIEW OF THE TODOROKITE_BTJSERITE PROBLEM 977

a Mn4- , Mn2 ' , e tc , a t y = o

c Mn4. , Mn2 ' ,e tc , a t y= t /2

O o x y g e n a t y = o

Q oxygen a t y=1rz

@ c a 2 - , N a ' , K ' , H 2 o , e t c , i n t u n n e l s

. m a n g a n e s e a t z = O

O o x y g e n

+ / n n 4 * v a c a n c i e s a t z = O

( a f s o i v t 2 a 1 2 = 1 1 4 1

and Buseck (1981) suggested that the 7A phyllomangan-ate birnessite phases represent the end-member T(2,o)tunnel structure.

The todorokite family in the Turner and Buseck fl981)classification is represented by T(3,n) with T(3,3) themost common structure-type. The various coherent inter-growths in todorokites recognized by them (Turner andBuseck 1981, 1982; Turner et al., 1982) and other workers(Chukhrov et al.,1978,1979, l98l) are designated T(3,2),T(3,3), T(3,4), T(3,5), T(3,6), T(3,7), etc., whlte the end-member T(3,o) is suggested to represent the phylloman-ganate buserite phases.

Although this nomenclature summarizes the dimen-sions of the edge-shared [MnOe] octahedral chains intodorokites, it does not define the structural details insidethe tunnels. Some insight into the structure and crystalchemistry oftodorokite can be deduced from correlationswith the known crystal structure of psilomelane (Wads-ley, 1953). The "floors" and "ceil ings" of the [2x3]tunnels of psilomelane are formed by double chains ofedge-shared [MnOo] octahedra containing only Mn4+ ionsin the M3 positions. The "walls", however, consist oftriple chains of edge-shared octahedra (defining the psi-lomelane unit cell parameter a : 9.564), in which Mna+ions in the central Ml position are flanked by largerdivalent cations in the M2 positions. In the todorokite[3x3] structure-type portrayed in Figure 2, the "walls"are suggested to be similar to those of psilomelane withdivalent cations occupying outer M2 positions in thetriple chains of edge-shared octahedra and Mna+ ionslocated in the inner Ml positions (defining the unit cellparameter c = 9.59A common to all structure-types oftodorokite). The "floors" and "ceilings" in the mostcommon todorokite [3x3] structure-type contain Mna+ions in M3 and M4 positions of triple chains of edge-shared [MnOo] octahedra, giving rise to the unit cellparameter a = 9.754 (Fig. 2). A (001) projection of thestructure shown in Figure 2b provides another way ofviewing the "floors" or "ceilings" of the todorokitestructure. The triple chains of linked [MnO6] run parallelto the b axis and are separated from adjacent triple chainsby furrows of Mna+ ion vacancies at the z : 0 level in the(001) plane. In the (001) projecrion (Figure 2b) divalentcations in M2 positions lie above and below the furrows atapproximate levels z : ll4 and 314.

The observations of different todorokite structure-types (Chukhrov et al., 1978, 1979, lg9l) and inrer-growths of variable tunnel widths in todorokites (Turnerand Buseck, 1981, 1982; Turner et al., l98Z) are depictedschematically in Figure 3. This figure shows an inter-growth of [3x3] and [3x5] tqnnels, corresponding to theunit cell parameters a : 9.75A and a : 14.64, respective-ly. Again, furrows of Mna+ ion vacancies at the z = 0level define the widths of the "ceilings" (or ..floors") ofthe tunnels, as portrayed in Figure 3b. It becomes appar-ent from the (001) projections shown in Figures 2b and 3bthat very wide tunnels attaining dimensions of [3x8] and

Fig. 2. Proposed crystal structure of todorokite. The diagramof Turner et al. (1982, Fig. l) has been redrawn to resemble thepsilomelane structure-type illustrated in Burns and Burns 1977b;Fig.4) so as to highlight the linkages of [3x3] tunnels intodorokite (a) a (010) projection of the structure viewed down the[3x3] tunnels. (b) a (001) projection showing the triple chains ofedge-shared [MnOe] octahedra forming the "ceilings" or"floors" of the [3x3] tunnels. A (001) projection showing thetriple chains constituting the "walls" is comparable.

c = 9 5 9 A

I ' "=n ruo

a--9 754

eO

Oe

O @

O O

\t tt

O O

a = z . a q A

, l . l . l l . / .1 .1.1.a

a

a a

a

.1 . / .1 l . l . l . l . l .

978 BURNS ET AL,: REVIEW OF THE TODOROKITE-BUSERITE PROBLEM

e a = 9 . 7 5 A * a = 1 4 6 A +

lent cations are bonded in postions above and below thevacancies in the layered structures (Crane, 1981). Wenoted when discussing Figure 2b earlier that divalentcation-bearing M2 sites are located immediately aboveand below the furrows of Mna* ion vacancies in the (001)planes of todorokite. Similar Mna* ion vacancies mayalso be present within the "ceilings" of todorokite,particularly where multiple-width edge-shared [MnOe]domains exist. Examples are sketched in Figure 3b.These cation vacancies not only may nucleate faults,kinks, and twinning observed in nntElvt micrographs oftodorokite fibers (Turner et al., 1982), but they alsowould influence the crystal chemistry and site occupan-cies of the tunnel interiors. As a result, three types ofatomic substitution might contribute to the crystal chem-istry of todorokite. First, substitution of Mna+ cations bycations of similar ionic radii (0.53-0.55A) in the "ceil-ings", such as low-spin Co3* ions (Burns, 1976). Second,substitution of divalent Mn2* ions in the "walls" byMg2*, Ni2*, Ctr2* , Zrf* and other cations having ionicradii in the range 0.65-0.804, and third, constituents ofthe tunnels. The latter would consist of a variety of largecations (K+, Ba2+, Ag+, Na+, Ca2*,Pb2*), H2O mole-cules, and smaller hyrated cations adjacent to Mna+vacancies in the "ceilings."

The cation site occupancies of todorokites and buser-ites proposed here also account for relative stabilitiestoward oxidation. Todorokite and synthetic buserite areboth destabilized by the presence of substantial Mn2+ions in the structures, which are vulnerable to oxidation.Thus, Mn2+-todorokites are oxidized to vernadite(Chukhrov et al., 1978, 1979), while synthetic buseritesare oxidized (accompanied by dehydration) to birnes-sites. However, replacement of Mn2* by Mg2* , Zn2* ,Ni2*, and Cu2*, which are not susceptible to oxidation,stabilize both todorokite and buserite.

Although detailed crystal structure refinements ofthese cryptocrystalline manganese oxides are urgentlyrequired to establish the precise similarities and differ-ences between todorokites and buserites, it is apparentthat currently available evidence demonstrates that theyare distinct phases.

Conclusions and recommendations

Critical examination of the published literature ontodorokites and buserites leads to the following conclu-slons.

(1) Naturally occurring todorokites and synthetic bu-serites are crystallographically distinct phases.

(2) Stability relationships linking the mineral todorokiteto synthetic buserite and its suggested break down prod-ucts birnessite and TMnOOH were deduced from phasesprepared and degraded in the laboratory under conditionsnot encountered in geochemical processes.

(3) The validity of todorokite as a distinct mineral grouphas been demonstrated and a family of todorokite struc-ture-types exists possessing multi-dimensional tunnels

t

c = 9 .59A

J

+

+

+ a

a

a

a

a

a +a

. +

a +

. +

. +a

. +a

. +

. +

. +

. +

. +

. +

a

a

a

a

+a

+

+

+

+a

+

+

+a

+a

. +

a +

. +

a l

a a

. +

. +

a +

a +

a +

a +

. +

+ .

+ .

+

+

+

o

+a

+

+

+

+

+

a

a

a

+a

+

+

+a

+

+a

a

* a a a * a a a r a

* a a a * a

a a a a *

F a = 9 . 7 5 A - l - a = 1 4 . 6 A _ l

Fig. 3. Schematic intergrowth of [3x3] and [3x5] tunnels intodorokite. (a) a (010) projection viewed down the tunnels. (b) a(001) projection showing locations of Mna* cations in the"ceilings" of the tunnels. Some sporadic Mn4* vacancies areshown to exist within the "ceilings" on the right of the diagram.

higher (Turner and Beseck, 1982) tend towards the [3 x o]

Iayered stmcture postulated (Turner and Buseck, 1981)for buserite. Their predominance would provide groundsfor expecting a buserite-like phase to occur naturally inmarine manganese oxide deposits.

An important factor to be considered in the "ceilings"of the todorokite tunnels are cation vacancies within thebands of edge-shared [Mnlvou] octahedra which proba-bly become more prevalent the wider the "ceiling"dimension. As noted earlier, Mna' cation vacancies areessential features of chalcophanite (Wadsley, 1955), aswel l as numerous d ivalent cat ion (R2*)-bear ingR2+2Mn3Os phases (e.g., Ostwald and Wampetich, 1967;Riou and Lecerf, 1975, 1977) and possibly syntheticbirnessite (Giovanoli et al.,1970a). In these phylloman-ganates, the Mna* vacancies in sheets of edge-sharedoctahedra dictate their crystal chemistries because diva-

a

+

+

a a

a

a

a

symbolized by T(3,n), where n, the dimension of the"ceilings," is most commonly 3 (triple chains of edge-shared [MnO6] octahedra). However, n may range from 2to 7 or higher. With this terminology, n + @ mayrepresent the buserite group.

(4) Todorokite, not " l0A-manganite" or buserite, is theabundant Mn(IV) oxide l0A phase occurring in mangani-ferous concretions and crusts including seafloor ferro-manganese nodule deposits.

AcknowledgmentsWe are indebted to several people for helpful discussions,

including Drs. G. Arrhenius, C. Bowser, p. Buseck, S. Crane, G.Faulring, M. Fleischer, C. Frondel, M. Hey, J. Hunter, H. Jaffe,R. McKenzie, U. Marvin, G. Mellors, P. Moore, J. Murray, R.Potter, G. Rossman, J. Straczek, and W. Zwicker. We thank Dr.G. Arrhenius for sending us a preprint of his paper with R.Giovanoli. Dr. S. Turner provided an extremely constructivereview of the manuscript. Research on the mineralogy of manga-nese nodules was supported by the National Science Foundation(grant number OCE-7821495) and the I.C. Sample Office (U.S.Electrochemical Society, Cleveland Section).

Note Added

Two recent publications support the conclusionsreached in this paper. First, todorokite is now acknowl-edged by Perseil and Giovanoli fl982) to have severalfeatures in common with the l0A phase in marine manga-nese nodules. This finding originates from XRD, TEM,SEM and EMP measurements of todorokite from theeastern Pyrenees. Second, further evidence for the tunnelstructure of todorokite is described by Arrhenius and Tsai(1981), who suggest that todorokite forms diageneticallyfrom buserite in marine manganiferous deposits.

References

Arrhenius, G. (1963) Pelagic sediments. In M. N. Hill, Ed.. TheSea, p. 655-727, Vol. 3. Interscience Publicarion. New york.

Arrhenius, G. O. and Tsai, A. G. (1981) Structure, phasetransformation, and prebiotic catalysis in marine manganateminerals. Scripps Institution of Oceanography Reference Se-ries 8l-28, l-19.

Buerger, M. J. (1936) The symmetry and crystal structure ofmanganite, Mn(OH)O. Zeitschrift fiir Kristallographie, 95,163-174.

Burns, R. G. and Burns V. M. (1977a) Mineralogy of manganesenodules. In G. P. Glasby, Ed., Marine Manganese Deposits, p.185-148. Elsevier Publishing Company, New york.

Burns, R. G. and Burns, V. M. (1977b) The mineralogy andcrystal chemistry of deep-sea manganese nodules, a polyme-tallic resource of the twenty-first century. In Mineralogy:Towards the 2lst Century (Centenary Volume, MineralogicalSociety, London, p. a9-67) Philosophical Transactions oftheRoyal Society of London, A286, 283-301.

Burns, R. G. and Burns, V. M. (1979a) Manganese oxides. ln R.G. Burns, Ed., Marine Minerals, p. 146. Reviews of Mineral-ogy, Vol. 6. Mineralogical Society of America, Washington,D. C.

Burns, R. G. and Burns, V. M. (1980) Recent structural data formanganese (IV) oxides. In B. Schumm, Jr., H. M. Joseph, and

979

A. Kozawa, Eds., Manganese Dioxide Symposium, Vol. 2, p.97-112. Electrochemical Society Publication, Cleveland.

Burns, V. M. and Burns, R. G. (1978) Post-depositional metal-enrichment processes in manganese nodules from the equato-rial Pacific. Earth and Planetary Science Letters, 39,341-j48.

Burns, V. M. and Burns, R. G. (1979b) Observations of process-es leading to the uptake of transition metals in manganesenodules. In C. Lalou, Ed., La Gendse des Nodules de Manga-ndse, no. 289, p. 305-315. Colloquium International du CentreNational de la Recherche Scientifique.

Buser, W. (1959) The nature of the iron and manganese com-pounds in manganese nodules. In M. Sears, Ed., InternationalOceanography Congress, Abstracts, p. 962-963.

Chukhrov, F. V., Gorshkov, A. I . , and Sivtsov, A. V. (1981) Anew structural variety of todorokite. Izvestia Akademia Nauk,S.S.S.R., Series Geology, 5, 88-91.

Chukhrov, F. V., Gorshkov, A. I., Sivtsov, A. V., and Bere-zovskaya, V. V. (1978) Structural varieties of todorokite.Izvestia Akademia Nauk, S.S.S.R., Series Geology, 12, 86-95 .

Chukhrov, F. V., Gorshkov, A. I., Sivtsov, A. V., and Bere-sovskaya, V. V. (1979) New data on natural todorokites.Nature, 278,631-632.

Crane, S. E. (1981) Structural Chemistry of the Marine Manga-nate Minerals. Ph.D. Thesis, University of California, SanDiego.

Cronan, D. S. (1980) Underwater Minerals. Academic Press,New York.

Dachs, H. (1963) Neutronen-und Rdntgenuntersuchungen amManganite, yMnOOH. Zeitschrift fiir Kristallographie, ll8,303-326.

Faulring, G. M. (1962) A study of Cuban todorokite. In M.Mueller, Ed., Advances in X-ray Analysis, Vol. 5, p. 117-126.

Feitknecht, W., and Marti, W. (1945) Uber die oxidation vonmangan (II) hydroxide mit molekularem sauerstoff. Ubermanganite und ktinstlichen braunstein. Helvetica ChimicaActa. 28. 129-148 and 149-157.

Finkelmann, R. B., Evans, H. T., Jr., and Matzko, J. J. (1974)Manganese minerals in geodes from Chihuahua, Mexico.Mineralogical Magazine, 39, 549-558.

Fleischer, M. (1980) Glossary of Mineral Species 19E0. Mineral-ogical Record, Arizona.

Frondel, C., Marvin, U. B., and Ito, J. (1960) New occurrencesof todorokite. American Mineralogist, 45, 1167-1173.

Giovanoli, R. (1980) On natural and synthetic manganese nod-ules. In I. M. Varentsov, and G. Grasselly, Eds., Geology andGeochemistry of Manganese, vol. l, p. 159-202, E. Schwei-zerbart'sche Verlagsbuchhandlung, Stuttgart.

Giovanoli, R. and Arrhenius, G. (1983) Structural chemistry ofmarine manganese and iron minerals and synthetic modelcompounds. In P. Halbach, Ed., Marine Mineral Deposits:New Research Results and Economic Prospects. Gluckauf,Essen, West Germany.

Giovanoli, R., and Briitsch, R. (1979a) Uber Oxidhydroxide desMn(IV) mit Schichtengitter. 5. Mitteilung: Stochiometrie,Austrauschverhalten und die Rolle bei der Bildung von Tief-see-Mangankonkretionen. Chimia, 33, 372-376.

Giovanoli, R., and Briitsch, R. (1979b) L'dchange des ions detransition par le manganate-l0A et le manganate-7A. In C.Lalou, Ed., La GenEse des Nodules de Mangandse, no. 289,Coloquium International du Centre National de la RechercheScientifique no. 289, 305-315.

BURNS ET AL.: REVIEW OF THE TODOROKITE-BUSERITE ZROBLEM

9E0

Giovanoli, R., and Btirki, P. (1975) Comparison of X-ray evi-dence of marine manganese nodules and non-marine manga-nese ore deposits. Chimia, 29, ll4-117.

Giovanoli, R. and Leuenberger, U. (1969) Uber die oxydationvon Manganoxidhydroxid. Helvitica Chimica Acta, 52, 2333-2347.

Giovanoli, R., and StAhli, E. (1970) Oxide und Oxidhydroxidedes dreiund vierwertigen Mangans. Chimia, 24, 49-61.

Giovanoli, R., Biirki, P., Giutrredi, M., and Stumm, W. (1975)Layer structured manganese oxide hydroxides. IV: The buser-ite groups; structure stabilization by transition elements. Chi-mia ,29 ,110-113.

Giovanoli, R., Feitknecht, W., and Fischer, F. (1971) UberOxidhydroxide des vierwertigen Mangans mit Schichtengitter.3. Mitteilung: Reduktion von Mangan (Ill)-manganat (IV) mitZimtalkohol. Helvitica Chimica Acta, 54, 1112-1124.

Giovanoli, R., Stehli, E., and Feitknecht, W. (1970a) UberOxidhydroxide des vierwertigen Mangans mit Schichtengitter.l. Mitteilung: Natrium-mangan (II,ilI) manganat (IV). Helvi-tica Chimica Acta, 53, 2W-220.

Giovanoli, R., Suhli, E., and Feitknecht, W. (1970b) UberOxidhydroxide des vierwertigen Mangans mit Schichtengitter.2. Mitteilung: Mangan (Ill)-manganat (IV). Helvitica ChimicaActa,53, 4534@.

Glasby, G. P. (1977) Marine Manganese Deposits. ElsevierPublishing Company, New York.

Hariya, Y. (1961) Mineralogical studies on todorokite and birnes-site from the Todoroki mine, Hokkaido. Japanese Journal ofthe Association of Mineralogists, Petrologists, and EconomicGeologists, 45, 219-230.

Hewitt, D. F., Fleischer, M., and Conklin, N. (1963) Deposits ofthe manganese oxides: supplement. Economic Geology" 58, l-5 1 .

Hey, M. H. and Embrey, P. G. (1974) Twenty-eighth list of newmineral names. Mineralogical Magazine, 39, %)3-932.

Larson, L. 'l . (1962) Zinc-beaing todorokite from Philipsburg,Montana. American Mineralogist, 47, 59-66.

Levinson, A. A. (l%0) Second.occurrence of todorokite. Ameri-can Mineralogist. 45, 802-807.

Nambu, M., Okada, K., and Tanida, K. (1964) Chemical compo-sition of todorokite. Japanese Journal of the Association ofMineralogists, Petrologists, and Economic Geologists, 5l, 30-38.

Paterson, E. (19E1) Intercalation of synthetic buserite by dode-cyclammonium cbloride. American Mineralogist, 66, 424-427 .

Perseil, E-A. and Giovanoli, R. (1982) Etude comparative de latodorokite d'Ambollas (Pyr€n6es Orientales), des manganatesi l0A recontr6s dans les nodules polymdtalliques des oc€ans

BURNS ET AL.: REVIEW OF THE TODOROKITE-BUSERITE PROBLEM

et des products de synthCse. Compte Rendu Academic Sci-ence de Paris,294, 199-202.

Potter, R. M. and Rossman, G. R. (1979) The tetravalentmanganese oxides: identification, hydration, and structuralrelationships by infrared spectroscopy. American Mineral-ogist,64, 1199-l2lE.

Radtke, A. S., Taylor, C. M., and Hewett, D. R. (1967) Aurorite,argentian todorokite, and hydrous silver-bearing lead manga-nese oxide. Economic Geology, 62, 186-2M.

Siegel, M. D. (l9El) Studies of the Mineralogy, Chemical Com-position, Textures and Distribution of Manganese Nodules at aSite in the North Equatorial Pacific Ocean. Ph.D. Thesis,Harvard University, Cambridge.

Simons, F. S., and Straczek, J. A. (1958) Geology of themanganese deposits of Cuba. U. S. Geological Society Bulle-tin. 1057. l-2E3.

Straczek, J. A., Horen, A., Ross, M., and Warshaw, C. M.(1960) Studies of the manganese oxides. IV. Todorokite.American Mineralogist, 45, ll7 4-1184.

Turner, S. and Buseck, P. R. (1979) Manganese oxide tunnelstructures and their intergrowths, Science, 203, 456-45E.

Turner, S. and Buseck, P. R. (1981) Todorokites: a new family ofnaturally occurring manganese oxides. Science, 212, 1024-1027.

Turner, S. and Buseck, P. R. (19E2) HRTEM and EDS of topsand bottoms of some Ni-bearing manganese nodules. EOS, 63,1010.

Turner, S. and Buseck, P. R. (1983) Defects in nsutite (rMnOJ,a dry-cell battery cathode. Nature, in press.

Turner, S., Siegel, M. D., and Buseck, P. R. (1982) Structuralfeatures of todorokile intergrowths in manganese nodules.Science, 296, 841-842.

Wadsley, A. D. (1950a) A hydrous manganese oxide withexchange properties. Journal of the American Chemical Socie-ty, 72, 1782-1784.

Wadsley, A. D. (1950b) Synthesis of some hydrated manganeseminerals. American Mineralogist, 35, 458499.

Wadsley, A. D. (1953) The crystal structure of psilomelane,(Ba,HzO)zMnsO 1q. Acta Crystallographica, 6, 165, 172.

Wadsley, A. D. (1955) The crystal structure of chalcophanite,ZnMryO1 . 3H2O. Acta Crystallographica, 8, 165-172.

Yoshimura, T. (1934) Todorokite, a new manganese mineralfrom the Todsoroki Mine, Hokkaido, Japan. Journal of theFaculty of Science, Hokkaido University, Sapporo, Series 4,YoL 2,289-297.

Manuscript received, December 2I, l98l;accepted for publication, February 22, 19E3.