Eur. J. Mineral. 1989,1,479-487 The genesis of zeolites GLAUCoGOTTARDIt* Istituto di Mineralogia e Petrologia, Università di Modena, via S. Eufemia 19,1-41100 Modena, Italy Abstract: The equilibrium diagrams of zeolites and the different possibilities of synthesizing zeolites starting from chemicals, minerals, and natural glasses are reviewed so to have a general picture of the conditions of crystallization of these minerals. Subsequently, a description and interpretation is given of the geological environments where zeolites crystallize in nature. Key-words: zeolite, diagenesis, very-low-grade metamorphism, hydrothermalism, volcanic glass. 1. Introduction and heulandite generally contain some M+ ca­ tions, which are almost absent in laumontite, yu­ This topic has been the subject of so many publi­ gawaralite and wairakite, so the alkali metal con­ cations (e.g. Hay, 1978, 1986; Iijima, 1978, 1980; centration in the system may influence the given Kastner & Stonecipher, 1978; Surdam & Shep- boundaries. Additional diagrams on these zeolites pard, 1978) over the last ten years, that one may can be found in the literature, but none is known wonder "Why another one?". As a matter of fact, to the author for zeolites other than those men­ all these previous studies give detailed informa­ tioned here. Field and laboratory evidence suggests tion on rock-forming zeolites, generally crystal­ that some other alkali zeolites may have a stabil­ lized from natural glasses during diagenesis, but ity field; this is certainly true for clinoptilolite, they omit any consideration of zeolites in veins the siliceous alkali-rich variant of heulandite, and and vugs of massive rocks. The author also aims is also probably true for natrolite and mordenite. to give information on this latter occurrence, al­ though it is impossible to fill the gap entirely, because of the paucity of detailed research. 3. Synthesis of zeolites The literature on the synthesis of zeolites is very 2. Equilibrium diagrams involving zeolites extensive; a comprehensive review is given in Barrer (1982). The synthesis of a zeolite is com­ Only a few zeolite equilibrium diagrams have monly accomplished from a recipe: changing the been published so far: data are known only for ingredients and the way of cooking may change analcime, laumontite, wairakite, stilbite, heulan- the final result. This fact is certainly in some way dite, and yugawaralite. For analcimes with Al/Si connected with the nucleation difficulties encoun­ ranging from 1:3 to 1:1.7, diagrams at P(H2O) of tered, especially with some zeolite species that are 2 and 5.15 kbar may be found in Kim & Burley (1980). Cho et al (1987) show a P(H2O)/T dia­ gram where stability fields for stilbite, heulandite, * The author who was a member of the editorial yugawaralite, laumontite, wairakite and lawsonite board of this journal passed away shortly after having are indicated. It is important to note that stilbite submitted this manuscript. 0935-1221/89/0001-0479 $ 2.25 © 1989 E. Schweizerbart'sche Verlagsbuchhandlung, D-7000 Stuttgart 1 Downloaded from http://pubs.geoscienceworld.org/eurjmin/article-pdf/1/4/479/4001658/479_gseurjmin_1_4_479_488_gottardi.pdf by guest on 01 October 2021 480 G. Gottardi very common in nature. For instance, laumontite the amount of active silica the more nuclei are has never been synthesized, although it is one of generated, so larger crystals can be obtained by the few zeolites whose stability diagram is known; diluting the active silica source; higher tem­ these studies have been performed only using peratures favour the formation of larger crystals natural crystals. In other words, laumontite crys­ (Barrer, 1982, p. 177 and 178, and cited references). tals may be easily grown but not nucleated. Near­ Raising the pH has the same effect as raising the ly the same is true for natrolite, also a common temperature, but a too high pH (> 13) favours the natural zeolite; its nucleation is possible only in reaction of the liquid with the first formed crystals the presence of nuclei of well crystallized silicates to give a new crystal phase. (nepheline) but not from gels alone (Senderov & The Si/Al-ratio is normally different in the Khitarov, 1971). There is a "magnesium mys­ liquid phase and in the crystallized solid; it is usu­ tery": if Mg is added to any mixture intended for ally higher in the liquid, but may be higher in the zeolite synthesis, only sheet silicates (clay miner­ solid if the ratio is as low as 2 in the liquid. als) are obtained, but Mg is present in many natu­ The alkali metal cations may be in solution as ral zeolites (faujasite, mazzite, erionite, offretite, bases, but also in part as salts (chlorides, sul­ and ferrierite). Phases equivalent to these Mg- phates, etc.), which may or may not be present in bearing zeolites may be synthesized in the pres­ the synthesized crystals. The crystals enclosing the ence of an organic base such as tetramethyl am­ anions may correspond to well-known natural monium (TMA), which is unlikely to occur in phases (sodalite, cancrinite, etc.), which bear an­ natural enviroments. ions in their framework cavities, but may also be Three kinds of ingredients will be considered a synthetic phase (e.g. ZK-5) without a natural here: chemicals, silicate minerals, and natural counterpart. glasses. 3.2. Synthesis with silicate minerals 3.1. Synthesis with chemicals There are many publications (see Barrer, 1982, The literature is vast and commonly in patents. Typ­ p. 216 ff.) on how zeolites may be obtained from ical ingredients are SiO2 (as silica-gel or soluble clays plus alkali; this is important not only from silicate) + Al2O3 (as soluble aluminate or alumi­ the point of view of industry, because of the lower nium hydroxide gel) + a pure or mixed strong price of these raw materials, but also from a geo­ base + H2O with hydrothermal treatment from logical point of view, because some zeolite occur­ 80 to 350°C; the strong base may be an inorganic rences in soils may have formed by this kind of one (alkali metal or alkaline earth) or an organic process. one (e.g. tetramethylammonium hydrate). Zeo­ The already mentioned synthesis of natrolite lites are always obtained from alkaline media, the from nepheline + NaOH (Senderov & Khitarov, pH ranging from 14 down to 8.5. At a lower pH, 1971) is also an example of this kind. synthesis is still possible, but longer times are A particular example of zeolite synthesis is giv­ needed to form a detectable amount of zeolite. en by Balagna et al. (1977), studying the alteration The mechanism of nucleation and growth of of a biotite granodiorite in contact initially with zeolites is certainly complex and not easy to un­ distilled water at 300°C and 0.33 kbar for several derstand. Some reactants (SiO2, AI2O3) have low months. The overgrown phases included vermicu- solubilities in the pH range of interest, but they lite (a clay mineral), thomsonite (a zeolite), and form gels and colloidal particles which dissolve ashcroftine (a zeolite-like mineral). rapidly as soon as possible; so the silica available for synthesis in the liquid phase depends on the rate at which the "solid" phase dissolves and re­ 3.3. Synthesis with natural glasses places the ions subtracted from the solution by the process of nucleation and growth of crystals. There is much literature also on this type of syn­ The first nuclei of the newly formed alumino-sili- thesis, which is particularly important for Earth cate are very poorly crystalline and unstable (Aiel- scientists. Obviously, "natural glass" means almost lo et al., 1971) and are subsequently transformed only volcanic glass in geology, although the rare into better crystallized material; also, when the impact glasses are also natural and may alter to reaction is started from clear solutions. The larger zeolites. For example the work of Colella et al. Downloaded from http://pubs.geoscienceworld.org/eurjmin/article-pdf/1/4/479/4001658/479_gseurjmin_1_4_479_488_gottardi.pdf by guest on 01 October 2021 The genesis of zeolites 481 (1977) employed rhyolitic pumices from Lipari generally thought of as a low-temperature pro­ with added alkali: chabazite, merlinoite, phillip- cess, where "low" means "less than 200°C". site, and analcime were crystallized, depending on Comprehensive reviews on the diagenesis of zeo­ temperature and NaOH/KOH-ratio. lites are given in the references quoted at the be­ More akin to geological conditions are syn­ ginning of this paper. theses with glass plus water; unfortunately, publi­ Two kinds of diagenesis must be distinguished: cations of this kind are few. Holler & Wirsching above the water table, and below the water table. (1978) obtained mordenite from a rhyolitic glass, Diagenesis of zeolites above the water table may chabazite, phillipsite, and analcime from a phono- occur in soils, in hydrologically open systems, in litic glass (final pH up to 9.5) and also from a hydrologically closed systems, and in geoauto- basaltic glass. Note that the phonolitic glass is claves. Diagenesis below the water table may oc­ altered to the same zeolites in nature. Similar re­ cur in marine sediments and as burial diagenesis sults were obtained by De Gennaro et ah (1988) (the latter has a continuous transition to very- with a trachytic glass at 200° C. Phillipsite crystal­ low-grade metamorphism). lized, and the final pH in the solution was 10. 3.4. General remarks on synthesis 4.1.1. Diagenesis in soils The synthesis of zeolites requires high pH, but It is well-known that soils of arid and semi-arid very high values are not necessary. If in the lab­ regions may be very rich in sodium carbonate (or oratory a pH of9.5 is sufficient, in a natural envi­ bicarbonate); even crystallization of the corres­ ronment a lower crystallization rate is possible ponding mineral (trona) is possible.