Queensland Precious Opal Deposits and Associated Critical Mineral Enrichment within Silicified Palaeochannels

Abstract

Queensland boulder opal host rocks were investigated and geochemical markers identified that are indicative of the presence of precious opal. Artesian groundwater migrating and accumulating in sandstone palaeochannels was subject to changes in pH, leading to opalisation within structural and stratigraphic permeability barriers. This process was accompanied by accumulation of several rare earth elements, elevated base metals, scandium, vanadium and possible occurrences of silver, palladium and gold. Alkaline groundwater moving surfacewards by capillary or fracture porosity and permeability from aquifers in the Eromanga Basin sequence, scavenged metals from ‘basement’ and from enclosing rocks that contain disseminated pyrite. During deep chemical weathering in the late Oligocene and early Miocene (Canaway profile), mixing of alkaline aquifer and meteoric acid groundwater led to complex pH changes. Colloidal silica and other substances were precipitated within former organic-rich, sedimentary rock layers within the palaeochannels.

Introduction

This paper deals with new and broad ranging discoveries with regard to the genesis of precious opal and chemical elements that have accompanied the deposition of this material. It is intended to be a starting point for future investigations and may help to change the way precious opal formation has been formerly viewed. It is generally acknowledged by opal miners that there is an urgent need for more effective methods of finding and recovering opal. There is also an understanding that there is likely to be large amounts of opal still present in the known opal fields and within extensive tracts of geologically suitable land which separate these deposits. However, exploration methods have not advanced in more than one hundred years while increasing economic, social and environmental pressures are making opal mining increasingly marginal. This paper deals with entirely new geochemical concepts which by necessity cannot be completely explained in their entirety and are presented as a broad overview best suited for the gemmologist and opal mining audience. Full details of the analyses of opal host rocks are available to fellow researchers on request.

Although the chemistry of opal (McOrist and Smallwood, 1995) and the microstructure of precious opal has been studied in considerable detail by several authors, including Sanders (1964), Jones et al. (1964) and Stewart et al. (2010), very little attention has been given to the enclosing host rocks and possible mineralogical changes that may have occurred within them during opalisation. The so-called Queensland ‘boulder opal’ deposits are concretionary ironstones with contained opal veins and septaria, and are found in specific structural and stratigraphic entrapment sites within sandstone palaeochannels (Senior, 1977; Senior and Chadderton, 2007; Figure 1).

These palaeochannels comprise kaolinised, feldspatholithic and lithofeldspathic, former volcanogenic sedimentary rocks and occur sporadically within a deep weathering profile which was named the Canaway profile by Senior and Mabbutt (1979). This profile is up to 35m in thickness and has a silica, alumina and iron-oxide enriched indurated capping that forms a distinctive, scarp-bounded margin to flat-surfaced erosional landforms within the region (Figure 2). This photo is reproduced from a seminal work that describes the Canaway profile in more detail (Senior, 1977) and is a very rare example where natural erosion has exposed the width of a former, very rich, opalised palaeochannel.

Open cut mining shows that the palaeochannels are sinuous and generally elongated, forming anastomosing bodies that in large opal mines extend laterally for tens or even hundreds of metres. The sandstones in productive mines are floored by impermeable rocks and the main palaeochannel may encompass other interbedded clayey lutite layers, formed due to incision of secondary palaeochannels and these may also contain additional opal deposits. Some mines, such as the Elusive had two superimposed palaeochannels (known locally as ‘levels’) and the Rocky Mine had three, which were economically productive.

It seems likely that artesian water, saturated with colloidal silica (Deveson, 2004), percolated through these sandstones, on a time scale of millions of years, leading to the deposition of dissolved metals and silica. The presence of porosity and permeability barriers, especially in concave depressions along the floor of these channels and in the vicinity in intra-profile faults (Wise, 1993; Senior and Chadderton, 2007) form entrapment sites for opal formation. These sedimentary and structural features are essential prerequisites for siliceous groundwater to accumulate forming a gel that hardened into precious opal. This process probably happened quite quickly due to interaction between alkaline and acidic groundwater and it is probable that hydraulic forces were involved, squeezing the gel into veins and spaces within the host rocks (Figure 3). These forces led to radial striations on the upper surfaces of most ironstone concretions and internal brecciation in many others (Pecover, 2005) and were probably caused by interaction between swelling smectitic and non-swelling kaolinitic clay species (Figure 4). The so called ‘blows’ in the Lightning Ridge opal fields (NSW) and an example recorded at the Gorge Mine by one of the authors (BRS) in Queensland, are small, intra-profile, diapiric, piercement structures, that in many instances have brecciated overlying rocks and dragged clay with opal fragments to the surface.

During this study the geochemistry of different rock types within the palaeochannels was studied. This research was aided by obtaining grab samples from the SJ Mine, from the surface to a depth of about 18m during an open cut mining operation located near Opalton in western Queensland. In addition, the host rocks surrounding precious, semi-precious and non-precious or ‘potch’ opal specimens were studied from a variety of widely spaced locations, and were obtained from active, or formerly active, major opal mines. The main objectives were to find out if there are any geochemical markers that would assist in the exploration for precious opal.

Geochemistry of Palaeochannel Sedimentary Rocks

Methods and equipment

Rock samples representing each individual rock type were collected from two open cuts shown in the aerial drone photograph of the SJ Mine (Figure 5) in MC 300036. These samples were sent to our laboratory in Canberra for preliminary analyses using a new Niton XL5-plus, XRF analyser. This instrument analyses forty-five elements with the results recorded as parts per million. It was quickly realized that most samples were not homogenous and gave varied results especially when the unit was used to compare different layers within ironstone concretions. Accordingly, the samples were oven dried and pulverised in a stainless-steel laboratory ball-mill, and the powders compressed within plastic cups covered with a window of transparent X-ray film. It is realised that iron contamination could occur from the mill but as the specimens have an intrinsic high iron content such contamination would not make any significant difference. The stainless steel cup and ball bearings were regularly cleaned and a pure quartz sample was milled following each sample to further reduce contamination. Each sample was X-rayed for three minutes and the results recorded in a database. This method provided comparable results in most cases.

In addition to the mine samples, opal specimens from several former or currently productive mines were analysed using the XRF and three of these are illustrated (Figures 3, 4 and 6).

In these instances, because of the intrinsic value of the contained precious opal, these were not ball-milled, and each sample was X-rayed at least twice and the results were averaged. Care was taken to focus only on the host rock thus avoiding the precious opal veins. Most of the specimens had been polished for display purposes and to avoid contamination from any residual polishing agents, the reverse side was lightly sanded with silicon abrasive paper prior to the Niton XRF analyses. Each specimen was X-rayed for three minutes as per the mine samples.

For purposes of comparison and to verify the Niton XRF results, twelve duplicate samples were sent to the Australian Laboratory Services (ALS), Stafford, Brisbane for more detailed analyses and to identify elements that are not available in the Niton XRF analyser. In the main, these unavailable elements obviously include gaseous and several light elements in the periodic table. Other elements that are not in the XRF analytical spectra include most of the lanthanoids (rare earth elements) with atomic numbers between 61 and 71. ALS data was also sought to verify the presence of palladium, scandium and silver, and the possible presence of gold.

The analytical procedures undertaken by ALS were:

• 35 Element Aqua Regia using their ICP-AES instrument
  (ALS code ME-ICP41)
• Lithium Borate Fusion ICP-MS using their ICP-MS instrument (ALS code ME-MS81) and
• Pt, Pd, Au 30g FA ICP using their ICP-AES instrument
  (ALS code PGM-ICP23).

In addition to the above, several samples were examined ‘in house’ using a Diffuse Reflectance Infra-red Fourier Transform Spectrometer (DRIFTS) equipment and this was used to see if the structure of kaolinite was altered or ‘pillared’ in the palaeochannels or in the vicinity of opal deposits.

Results

The palaeochannel rocks show no obvious visible mineralization apart from iron oxide enrichment of the concretions (boulders), varicolored beds and mottling due to diffuse iron oxide staining. Dense ironstone concretions and ferruginised layers commonly occur and are replacing former calcareous concretions and beds of impure limestone that are present throughout the unweathered Winton Formation (Senior, 1977). Some beds are weakly to strongly indurated with glassy amorphous or opaline silica. Apart from cryptocrystalline silica, there are no visible gangue minerals which would draw attention to mineralization. It was therefore surprising to see several elements that are elevated many times above their average crustal abundances and, in some instances, reaching possible economic grade (Table 1). These include the light rare earth elements (LREEs) lanthanum, cerium, praseodymium, and neodymium.

Several samples (Table 2) also contain elevated heavy rare earth elements, and in SJ-7 include samarium 141ppm, europium 30ppm, gadolinium 73ppm and dysprosium 29ppm. Vanadium (up to 692ppm) and nickel (up to 445ppm) are elevated in several of the samples analysed. Other elements that are sporadically elevated include cobalt, chromium, lead and copper. In some instances, the XRF results were confirmed through more precise, duplicate analyses undertaken by ALS (shown in blue in Table 1).

The rare and valuable critical mineral element scandium was indicated in all but two XRF samples, less than the limit of detection. The presence of scandium was confirmed by ALS in amounts averaging about 26ppm. However, an organic rich layer at a depth of about 3m in the SJ Mine (SJ-7) has an impressive 224ppm scandium. Similarly, the fossil Diprotodon bone embedded in late Pleistocene sediments within an artesian spring area, located south of the Yowah opal field, has 268ppm scandium. These two analyses show that the presence of organic materials in these sedimentary rocks probably scavenged and enriched this element from artesian water.

The Niton XRF identified silver in nearly every sample varying between 3 and 20ppm. Palladium is indicated in nine samples, including 5ppm Pd (SJ-6), 6ppm Pd (SJ-11A), 5ppm Pd (SJ-14 and SJ-15), 4ppm Pd (SJ-17), 8ppm Pd (SJ-22) and 4ppm Pd (SJ-23, SJ-27 and SJ-29). Quite high levels of gold occur, between 2 and 9ppm, but are only listed in the less than ‘limit of detection’ category in the XRF data. Unfortunately, ALS failed to confirm the presence of either silver, palladium, or gold in their analyses, possibly because these elements are locked within cryptocrystalline silica and were not released during aqua regia acid leaching. These precious metals are probably present in sub-microscopic particles that might only be revealed by fire assays, or by analyses following total rock sample digestion methods. In view of the high economic value of palladium, gold and silver, further analytical work is being undertaken to see if these elements occur and to ascertain if they could be commercially extracted.

From where did the elevated elements originate?

There are three possible sources of the elevated elements measured in these palaeochannel deposits that occur within the Canaway profile of the Winton Formation. These are:

• Aquifer sourced and transported colloidal silica containing dissolved elements derived from pressure aquifers in the Great Artesian Basin (GAB)
• Leaching of metals from disseminated iron pyrite and opaque minerals which are found throughout the Winton Formation, and
• Natural beneficiation of pre-existing elements due to leaching of soluble minerals causing a reduction of rock volume and absolute enrichment within deeply weathered profiles.

Aquifer sourced elements

Habermehl (1980) showed from his study of several hundreds of water wells in the GAB that the most widely exploited aquifers in the Jurassic and Lower Cretaceous contain between about 500 and 1000mg/l total dissolved solids, mainly sodium bicarbonate, with some chloride and minor sulphate. The quality of the water generally increased downwards. The water is alkaline and because of the high sodium and high residual alkalinity it cannot be used for large scale irrigation.

Deveson (2004 and 2005) reported on the presence of trace elements in the Lightning Ridge town artesian water bore. He evaporated 20 litres of fresh bore water in acid washed Pyrex beakers and measured the residue using a portable XRF unit. The detection limits of the XRF are quite high and several elements within the sample were not detected because of this. It is noteworthy that anomalously high silver (20ppm), mercury (10ppm) and copper (9ppm) determined in the XRF sample, were later confirmed by laboratory trace metal analyses. Deveson also found that alkaline artesian water is saturated or nearly saturated with silica and if the pH is lowered and mixed with more acid water, then silica polymers form. If the pH drops further the silica will quite quickly coagulate forming a silica gel, the precursor of precious opal.

Meteoric water entering aquifers that crop out along the western margin of the Great Dividing Range takes more than one million years to reach natural leakage areas along its western margin. Mahara et al. (2009) used naturally occurring isotopes ⁴He and ³⁶Cl to calculate the transit times. During groundwater movement the water initially flows over pre Jurassic ‘basement’ and through a wide variety of different aquifer rock types of different ages, wherein various metals are likely to occur. This system has operated since deposition ceased in the late Cretaceous (Cenomanian) and for most of this time the potentiometric surface has been above the natural ground surface. According to Habermehl (1980) aquifer leakage also involves vertical movement through semi-pervious confining beds towards the surface via faults and joints and by capillary movement. Considerable volumes of water have moved and continue to move surfacewards by these means. However, due to high evaporation rates and deep phreatic water (generally tens of metres below the ground surface), these processes are disguised. Under these circumstances colloidal silica and contained metals could be precipitated in receptive near-surface rocks.

Leaching of elements from host sedimentary rocks

Measured sections and cores from drill holes in the Winton Formation contain disseminated iron pyrite and carbonaceous material (Senior, 1979), and according to Exon and Senior (1976) this immature sequence was derived from an andesitic volcanic province located along the Queensland coast where the Great Barrier Reef now lies. Pyrite can host a wide variety of metals which could be leached, transported and deposited within aquifers. Sedimentary rocks containing organic materials (black shales) and elements such as manganese, would have actively scavenged these metals in situations where pH changes, due to groundwater mixing and interaction with meteoric near-surface groundwater.

Beneficiation within deep chemically weathered profiles

Two major deep chemical weathering events have led to profile development in the Winton Formation. The Morney profile was palaeomagnetically dated by Idnurm and Senior (1977) to have developed in the late Cretaceous and Paleocene and the Canaway profile dated to have formed in the interval late Oligocene to early Miocene. The development of the Morney profile, which is up to 100m in thickness, straddles in time, the Cretaceous/Tertiary boundary. It is tempting to equate this event with the extra-terrestrial impact and resulting acid rain which led to the mass extinction of megafaunas at that time. Acid rain would easily break down the feldspar and lithic sand grains in the pyrite-rich Winton Formation sedimentary rocks, and would add sulphuric acid aiding the deep leaching and profile forming process.

The Morney profile was gently folded and partly eroded prior to the second phase of deep weathering which was superimposed in it, forming the Canaway profile. At the Yowah opal field the palaeomagnetic age of the Canaway profile was further refined by Schmidt and Dickson (2017) and estimated at 35Ma ± 7Ma, which places the development of the Canaway profile and indicates that the maximum age of the opalised ironstone nodules, in the interval from late Eocene to early Oligocene.

These weathering processes reduced the volume of the weathered rocks through leaching and removal of soluble elements; this process led to beneficiation of residual elements and those formerly introduced, along with colloidal silica in groundwater. It is likely that all three of the above mechanisms were involved in the elevation of elements within the palaeochannels.

Geochemical Indicators of Precious Opal Formation

This study was initially aimed at finding geochemical markers that could assist in locating precious opal. With the availability of portable analytical equipment such as the Niton XRF, RS-125 Gamma-ray Spectrometer and DRIFTS spectrometer, it is now possible to obtain rapid analyses in real time which can be used to guide an exploration or mining operation.

Potassium and rubidium ratios

One of the most promising geochemical markers is the potassium to rubidium ratio. These ratios were calculated and plotted in graphical form (Figure 7). Most of the rocks within the SJ Mine contain rubidium, however, it can be seen in the graph that the ratios are high in barren rock and decrease where the non-precious and semi-precious samples are positioned. For most of the precious opal samples rubidium is reduced to <2ppm, or is entirely absent. Unfortunately, no precious opal was found during the excavation of the two 18m deep pits at the SJ Mine and so the ratios in a precious opal environment at this site were unavailable. The K:Rb ratios in these pits indicate that the potential opalised areas are several tens of metres distant and are probably located near the central, previously mined-out area. However, from the wide range of barren rock and opal containing rocks that were analysed from other mines, it seems likely that a low or zero K:Rb ratio is indicative of the nearby presence of precious opal.

Manganese

At the Lightning Ridge opal fields, the mineral hollandite is associated with precious opal, and according to Senior and Chadderton (2007), is an opal indicator mineral. However, according to the local miners, it is rarely found with precious opal but is found in the general vicinity, within approximately 10 to 20m. Although hollandite has not been observed in the SJ Mine, manganese is elevated in some of the rocks. Accordingly, the distribution of manganese was studied to see if this element could be used as an opal indicator in the Queensland opal fields.

The analytical abundances of manganese are shown in the graph (Figure 8), and colour-coded as to the types of precious opal in the samples and with those devoid of opal. The XRF has a high detection limit and generally samples with less than 500ppm manganese fall within the ‘less than the limit of detection’ category. These form a cluster in the lower left corner of the graph and comprise a mix of barren, potch and semi-precious opal types. There appears to be a tendency for the higher quality of precious opal to be associated with lower values of manganese between 1000ppm and approximately 3000ppm, indicating that as the colloidal silica approaches the opal ‘kitchen’, manganese is precipitated followed by precious opal in the depleted manganese environment. Unfortunately, because mining did not lead directly to a precious opal discovery the manganese geochemical precipitation pathway was unable to be fully evaluated. However, this method has promise as an indicator to the proximity of precious opal and will be further investigated.

Radioactive elements

Opalised layers can be identified by measuring the natural radioactivity. Senior and Chadderton (2007) used this method to discover an opal deposit at Lighting Ridge, NSW, and to find extensions to opal ‘levels’ at the Hayricks and Adams Mines in Queensland. As mining progressed at the SJ Mine, suspected opal ‘levels’ were routinely surveyed using a RS-125 Gamma Ray Spectrometer but the results were negative with background counts averaging only 79cps and no abnormal, or above background levels of radioactivity were detected.

A few surface ironstone concretions, located outside of the mining claim were found to be radioactive, with levels averaging 141cps and an RS-125 assay recorded potassium at 0.2%, uranium at 3.7ppm and thorium at 14.8ppm. These ironstones are possibly associated with precious opal at depth but could not be legally excavated under existing mining legislation. This study, which included analyses of several precious opal specimens from around the region, shows that in addition to uranium and thorium, radioactive daughter decay products of radium, probably captured in barium minerals, contributed to natural radioactivity emanating from precious opal deposits.

Natural radioactivity measured in opal prospects and operating mines by Senior and Chadderton (2007), strongly indicates that the very best quality precious opal gemstones (Figures 10 and 11) most probably were formed in zones of elevated radioactivity exceeding 175cps.

DRIFTS structure of kaolinite

DRIFTS spectroscopy and chemical analyses indicate that kaolinite in proximity to precious opal appears to have a unique structure and is ’silica pillared’ wherein the two structural sheets are separated and the two sets of hydroxyl/hydrogen bonds are sterically hindered by the silica pillars. Samples from Lightning Ridge and some from the SJ Mine show that the near opal samples have distinct ‘saddles’ in each of the complex hydroxyl/hydrogen absorption bands at about 1.4nm and 2.2nm, whilst those remote from opal do not show this feature.

The instrument used to measure the infra-red absorption is a ’Diffuse Reflectance Infra-red Fourier Transform Spectrometer’ (DRIFTS) and is an Ocean Insight Nano-Quest with a 1,300 to 2,600nm recording range and equipped with a 45^° diffuse reflectance probe. The light source is tungsten halogen 360-2,400nm. It takes only a few moments to obtain a DRIFTS spectral profile to see if the kaolinite is ’pillared’ and therefore located close to opal. Typical spectral profiles of ‘pillared’ (Russel’s Mine) and normal kaolinite (Beefwood prospect, Lightning Ridge) are shown in Figure 9. The presence of ‘pillaring’ in kaolinite indicates an environment where precious opal could occur. However, this information is not necessarily a specific opal locator because all of the numerous SJ Mine rocks that were measured comprise ‘pillared’ kaolinite.

The formation of doublets in both the 1.4nm and 2.2nm hydroxyl/hydrogen bond spectral lines of kaolinite has been previously only ascribed to a high degree of crystallinity. However, this investigation of kaolinite associated with precious opal deposits point to silica ‘pillaring’ as being developed within a near-surface, colloidal silica-enriched, aqueous, palaeochannel environment.

It is noteworthy that spectral data equivalent to the DRIFTS spectra could be obtained by some satellite hyperspectral systems such as have been used by NASA to map kaolinite and opal on Mars, or by airborne hyperspectral systems. This line of research is in its infancy and has not been previously used in opal fields and further work is required to indicate its efficacy in locating opalised zones. It is also of note that ‘pillared’ kaolinite has industrial applications.

Conclusions

Silica-laden alkaline groundwater moving along a palaeochannel begins to deposit silica when the pH changes due to the water encountering more acid conditions. The first elements to be deposited in this ‘roll front’ as it approaches the opal ‘kitchen’ are manganese, potassium and rubidium and these form consistent geochemical indicators demarking the peripheral zones of opal deposits. Other elements are elevated in the rocks enclosing the opal and these include several rare earth elements, scandium, barium, cobalt, nickel, vanadium, silver and possibly palladium and gold. Radioactivity increases towards precious opal depositional sites with background radiation of about 80cps increasing to between 140cps and 170cps in the proximity of precious opal. Gamma-ray logging of drill holes or monitoring the radioactivity of ironstone-enriched ‘levels’ using a gamma-ray spectrometer, provides very effective indicators as to the proximity of concealed deposits.

Although there are conflicting results between the XRF and ALS analytical data it seems likely that opalisation within sandstone palaeochannels was accompanied with enrichment of several elements that would not normally be present in sedimentary rocks. Undoubtably, the process of element enrichment is very complex and it is likely that most metals and all of the precious metals are locked within cryptocrystalline silica. This problem might be resolved through use of different analytical techniques that would unlock the silica and release contained precious metals.

As this is the first study of this type, it is possible that other palaeochannels in the Canaway profile are similarly, or even more strongly enriched, than the SJ Mine example. Analyses of the host rocks in samples containing precious opal appear to confirm elevated enriched palaeochannels at other locations in western Queensland. It appears likely from the analytical results of samples obtained from several of the largest precious opal mines, that the process of opalisation was accomplished in a colloidal silica-charged environment that was also enriched in metals. The precipitation of precious opal was probably facilitated and may have taken place concurrently with these metals. However, the geochemistry of this association is poorly understood and worthy of further research.

The elevated rare earth elements (REEs) are particularly interesting, as these occur in near-surface, flat-lying, mostly soft, kaolinitic rocks (like the world class Chinese deposits?; Zhi et al. 2014) and indicate the possible presence of large, low-grade tonnages, amenable to simpler extraction methods for recovery of their contained critical elements.

Former organic beds such as black shales within the palaeochannels are likely to have strongly elevated levels of the precious element scandium. If this element, plus the REEs and the precious metals could be extracted from these palaeochannels, then there is a potential for economic resources with the added attraction of discovering scattered pockets of precious opal.

In order to continue this research, four sub-blocks encompassing the Opalton SJ Mine are under application by DBM Exploration Pty Ltd (EPM 28846). When granted, this will allow further geochemical sampling of this paleochannel and perhaps the discovery of others in this highly prospective region. Similarly, this company has, under application, seventeen sub-blocks (EPM 28651) in the vicinity of the Doreen’s Glory Mine. This mine is reputed to have produced many millions of dollars of world-renowned opal gemstones (Figures 10 and 11) and the sample analysed from this location (SJ-22) has the highest praseodymium of all the samples analysed and other elevated metals. The presence of Hooray Sandstone artesian aquifers in this area, that directly rest on metal-rich Proterozoic rocks of the Mt Isa Inlier, might be the primary source of enriched elements within this palaeochannel. Similar deposits may occur in the northwest part of the Canaway profile’s distribution. Other areas of potentially enriched palaeochannels, perhaps overlying metal-rich basement can be identified in the analyses. We are hopeful that this paper will stimulate further research and exploration in Canaway profile rocks for other economic minerals, that were formerly only recognised as having potential for economic precious opal deposits.

Acknowledgements

Alexander Kent (Mine Manager) is acknowledged for permission to collect samples from the SJ Mine concurrently with the mining operation that took place in 2022. He is also thanked for the drone photograph of the mining claim area, photographs of precious opals from his archives, and provision of opal specimens from other mines used for XRF analyses. Mr R. Newsham, is kindly thanked for collecting  samples of fossil bone and mud from nearby active mound springs, and opalised rocks from his mine located in the vicinity of the Hayricks Mine. Similarly, Sue Cooper and Ashley Theuerkauf kindly provided hospitality, access and samples from the Russel’s Mine. Danièle Senior is thanked for MapInfo construction of the regional geological map and graphs displaying the results of opal analyses and for editing the draft manuscript. Exploration was supported by DBM Exploration Pty Ltd and the CEO, Michael McCann is thanked for his enthusiastic support, funding for the ALS anayses and unlimited use of the Niton XRF. Thanks are also given to an anonymous reviewer who provided valuable comments which led to improvements in the manuscript.

References

Deveson, B., 2004. The origin of precious opal: A new model.
The Australian Gemmologist, 22(4), pp.50–58.

Deveson, B., 2005. The role of artesian water in the formation of precious opal. Abstract in the 4th National Opal Symposium, July, Lightning Ridge, NSW.

Exon, N., and Senior, B., 1979. The Cretaceous of the Eromanga and Surat Basins. Bureau of Mineral Resources Journal of Australian Geology and Geophysics, 1, pp.33-50.

Habermehl, M., 1980. The Great Artesian Basin. Bureau of Mineral Resources Journal of Australian Geology and Geophysics, 5, pp.9-38.

Idnurm, M., and Senior, B., 1978. Palaeomagnetic age of weathered profiles in the Eromanga Basin, Queensland. Palaeogeography, Palaeoclimatology, Palaeoecology, 24, pp.263-277.

Jones, B., Sanders, J. and Segnit, E., 1964. Structure of precious opal. Nature, 204, pp.990-991.

Mahara, Y., Habermehl, M., Hasegawa, T., Nakata, K., Ransley, T., Hatano, T., Mizuochi, Y., Kobayashi, H., Nimomia, A., Senior, B., Yasuda, H. and Ohata, T., 2009. Groundwater dating by estimating of groundwater flow velocity and dissolved ⁴He accumulation rate calibrated by ³⁶Cl in the Great Artesian Basin, Australia. Earth and Planetary Science Letters, 287(1-2), pp.43-56. DOI.org/10.1016/j.epsl.2009.07.034

McOrist, G. and Smallwood, A., 1995. Trace elements in coloured opals using neutron activation analysis. Journal of Radioanalytical Nuclear Chemistry, 185, p.193.

Pecover, S., 2005. Breccia pipes – fluid pathways of the opal gods. Abstract in the 4th National Opal Symposium, Lightning Ridge, NSW.

Sanders, J., 1964. Colour of precious opal. Nature, 204, p.1151.

Schmidt, P. and Dickson, B., 2017. Palaeomagnetic dating of ironstone nodules (‘’nuts’’) from the Yowah opal field, central southern Queensland. Australian Journal of Earth Sciences, 64(6), pp.743-742.

Senior, B., 1977. Landform development, weathered profiles and Cainozoic tectonics in southwest Queensland. PhD thesis, University of New South Wales, Sydney.

Senior, B., 1979. Mineralogy and chemistry of weathered and parent sedimentary rocks in southwest Queensland. Bureau of Mineral Resources Journal of Australian Geology and Geophysics, 4,
pp.111-124.

Senior, B. and Chadderton, L., 2007. Natural gamma radioactivity and exploration for precious opal in Australia. The Australian Gemmologist, 23(4), pp.160-176.

Senior, B. and Mabbutt., J., 1979. A proposed method for defining deeply weathered rock units based on regional geological mapping in southwest Queensland. Journal of the Geological Society of Australia, 26, pp.237-254.

Stewart, A.M., Chadderton, L.T. and Senior, B.R., 2010. Self-assembly in the growth of precious opal. Journal of Crystal Growth, 312(3), pp.391-396.

Wise, R., 1993. Queensland boulder opal. Gems and Gemology.
29, pp.1-15.

Zhi, L., and Yang, X., 2014. China’s rare earth ore deposits and beneficiation techniques. 1st European Rare Earth Resources Conference, Milos, pp.26-36.