On the origin of agate: A 300-year-old enigma

Terry Moxon (UK)

Quartz has been estimated to occupy around 12% of the earth’s crust and can be found in many forms, ranging from the massive, clear crystals of quartz and amethyst to the microcrystalline quartz that is to be found in jasper, agate, chalcedony, chert and flint. World-wide, the distribution of agate is not equal, but it can be found in every continent and probably exists in every country. However, only three countries extract enough agate for world export: Botswana, Brazil and Mexico.

Agate is most frequently found in fine-grained, igneous rocks filling gas cavities (Fig. 1), but it can also be found in sedimentary limestone hosts (Fig. 2) and fossil wood (Fig. 3). The most common agates are the wall-lining and horizontally banded types (Figs. 4 and 5 respectively). Rapid identification of agate in the field relies on the natural translucency of a fractured sample, but final confirmation is supplied by examining a thin section under a polarising microscope (Fig.6). Agate and chalcedony show a fibrous structure, whereas the quartz in flint, chert and jasper is generally granular. Granular quartz can demonstrate regular banding, but this is not agate (Fig. 7). Nevertheless, it is the colours and rhythmic banding that makes agate the most recognisable of all gemstones.

The distinctive repetitive banding found in agate inevitably provokes questions about its origin, which have now been asked for over 300 years. German scientists were particularly vocal and no doubt this was prompted by the importance of the agate manufacturing and mining industries based around Idar-Oberstein. Discussions on agate genesis are to be found in the German scientific journals of the mid to late nineteenth and early twentieth centuries. The most famous of these German scientists was Raphael E Liesegang (1869-1947), who made major contributions to studies of rhythmic growth. Research on natural rhythmic patterns are to be found in the scientific literature from 1850, but it was Liesegang’s 30 scientific papers on rhythmic banding that lead to the phenomenon becoming known as Liesegang rings.

Large Liesegang rings readily develop in sediments (Fig.8) and can be frequently observed outside the geological environment. One such example is the beautiful banding shown in Fig. 9 where bands have been etched into a steel-plate backing for a battery powered clock. However, it is in agate where rhythmic banding is at its most striking and, in 1915, Liesegang published his book, Die Achate, in which he argued that the major patterns could be explained by a rhythmic diffusion mechanism in silica gel. In this, he claimed that agate genesis depends upon two essential prerequisites: a gas cavity is first filled with silica gel and the banding is then caused when a solution from the surrounding host rock permeates through the gel. There is no doubt that agate-type banding can be simply created in gels (silica or agar) and these patterns do show a resemblance to the wall-lining or horizontal banding found in agate (Moxon, 2001).

However, there are a number of objections to the Liesegang hypothesis, for example, the need for a pre-existing gel to be the first cavity infill. Furthermore, the laboratory-produced Liesegang rings are only very approximate and do not show the subtleties that are observed in agate. However, Liesegang’s major omission was his failure to address the problem shown by the dehydration of a water-rich silica gel leaving behind a powder and destroyed banding. Surprisingly, some texts on geochemistry still treat Liesegang’s hypothesis as a solution for the problem of agate genesis. However, the scientific journals have not been so kind and there have been few supporters since the publication of Die Achate in 1915.

Until the 1950s, discussions on agate genesis were based mainly on speculation after an examination of the hand specimen or a thin section. Over the last 60 years, major technological advances in electron microscopy, cathodoluminescence, infrared spectroscopy, electron microprobe analysis and powder X-ray diffraction have lead to much progress in silica research. These same techniques have also resulted in a better understanding of the real agate structure. Unfortunately, advances in instrumentation have not greatly improved our knowledge of the origin of agate. An identification of the silica source together with the manner of agate crystallisation still requires answers before a final solution to the genesis problem can be reached.

It could well be that, while agate has a common general appearance, conditions for formation in igneous and sedimentary environments might be different. In this short article, discussion is limited to wall-lining agate in an igneous setting. Some of the problems are set out, together with a few of the solutions.

What is the silica source for agate?

Although agate has a glassy look, any consideration of a contemporary formation of agate and host has to be discounted. The 97% silica content in agate requires a melting point of about 1,700^oC and this is some 600^oC higher than the molten magma temperature on the earth’s crust. Likely silica sources are hydrothermal solutions, host rock leaching, volcanic ash and hot springs. All these contain silica, but, given the present state of knowledge, hot springs seem the most likely suppliers of silica. These could transport solutions of silicic acid (Si(OH)₄) to the gas cavities and eventually deposit amorphous silica.

What is the temperature of agate formation?

Important work was carried out by Fallick and co-workers in 1985 when they investigated the oxygen and hydrogen isotope content of water found in agate. Their experiments identified the formation temperature of agate from the Midland Valley, Scotland and the Isles of Mull and Rum. Later work by other workers has now resulted in a consensus that agate genesis probably occurs at about 50^oC and certainly less than 100^oC.

What is the nature of the initial silica deposit?

There are only two viable alternatives: either agate is formed directly from a hydrous solution or silica is deposited as a gel or powder that later evolves into agate. Gelatinous silica has been observed around fissures of the Golovnin volcano in Russia. This gel has naturally produced various forms of silica, including quartz (Naboko and Silnichenko, 1957). After ageing, amorphous silica found around hot springs also leads to the formation of quartz.

Laboratory experiments have produced quartz crystals from super-saturated silica solutions (Morey et. al, 1962) and minor quartz spherulites have been obtained after heating silica gel at 300^oC and 3kbar pressure (Oehler, 1976). Therefore, although the nature of initial silica deposit has not yet been established for agate, field and laboratory experiments show that there is no immediate objection to genesis starting from silica gel or an amorphous powder.

How does the final crystallisation process occur?

It is unlikely that agate forms as a result of the direct precipitation of chalcedony (a microcrystalline variety of quartz). Scientists, who study the process of ‘sintering’ (that is, a method for making objects from powder, by heating the material in a sintering furnace below its melting point until its particles adhere to each other), tend not to differentiate between fibrous chalcedony and granular quartz, but chalcedony has been identified forming as a later product in the sinter. Given the present state of knowledge, it seems more likely that agate forms from an initially amorphous deposit and, over time, the silica eventually transforms into agate.

Agate has a very simple formula (SiO₂), but is created by a complex development process. Agate can be found in hosts as old as 3.5 billion years, but agates of all ages show more or less the same general morphology and chemical composition (Moxon et al. 2006). On ageing, there is a continuous development that is slowly being understood. These include age-related changes, when individual silica building units (crystallites) coarsen with time while the moganite content (a form of silica found in microcrystalline quartz) decreases (Moxon and Carpenter, 2009). Continued studies on the actual structure of agate might eventually lead to a solution of this fascinating problem of agate genesis.

Acknowledgements

Historically, the full characterisation of agate by scientists has used samples from around the world. Donated material has been used in the present article and my thanks are due to Roger Clark (Fig. 2); Brad Cross (Figs. 4 and 6); Dick Dayvault (Fig. 3); Brian Leith (Fig. 7) and John Raeburn (Fig. 9).

Terry Moxon has been a research visitor at the Dept of Earth Sciences, Cambridge University, England for the past ten years. The work has involved examining agates from around the world and studying agate growth in the laboratory. These investigations have been described in a recently published book, reviewed opposite and entitled Studies on Agate. Further details are available from http://www.agateworld.co.uk.

References

Fallick, A.E., Jocelyn, J., Donelly, T., Guy, M. and Behan, C., 1985. Origin of agates in the volcanic rocks of Scotland. Nature, 313, 672-674.

Liesegang, R.E., 1915. Die Achate. Dresden Leipzig.123p.

Morey, G.W., Fournier, R.G. and Rowe, J.J., 1962. The solubility of quartz in water in the temperature interval of 25^oC to 300^oC. Geochimica et Cosmochimica Acta, 26, 1029-1043.

Moxon, T., 2001. Liesegang rings. Education in Chemistry. 38, 105-107

Moxon, T. Nelson, D.R. and Zhang, M., 2006. Agate recrystallisation: evidence from samples found in Archaean and Proterozoic host rocks, Western Australia. Australian Journal of Earth Sciences. 53, 235-248.

Moxon, T. and Carpenter. M.A., 2009. Crystallite growth in nanocrystalline quartz (agate and chalcedony). Mineralogical Magazine, 73, 551-568.

Naboko, S.I. and Silnichenko, V.G., 1957. Formation of silica gel on solfataras of the Golovnin Volcano on the Kunashir Island. Geochemistry. 3, 253-256. (in Russian with English Abstract)

Oehler, J.H., 1976. Hydrothermal crystallization of silica gel. Geological Society of America Bulletin, 87, 1143-1152.