A kaolin deposit at the Hiraki mine, situated approximately 50 km northwest of Osaka, SW Japan, occurs in the rhyolitic non-welded tuff sandwiched between underlying rhyolite and overlying rhyolitic welded tuff of Late Cretaceous. Volcanostratigraphy of the mine area is designated as follows: (in ascending order) the rhyolite, the non-welded tuff (with three members) and the Hiraki rhyolitic vitritic welded tuff. The boundary between the rhyolite and the non-welded tuff is transitional, whereas that between the non-welded tuff and the Hiraki welded tuff is defined as unconformity. Chemically, these volcanic and pyroclastic rocks are all of low alkali and peraluminous. Variation diagrams for incompatible element pairs, both major and trace elements, suggest that the deposit was altered from a single precursor, the rhyolitic non-welded tuff. The deposit was formed by the interaction between the rhyolitic non-welded tuff and acidic fluids which passed mainly through the fracture zones formed after the emplacement of the overlying welded tuff. Mineralogically, it is composed exclusively of kaolin group with a few chlorite, illite, montmorillonite, mixed-layer clay minerals, sericite and pyrophyllite, confined to fracture zones, and ubiquitous occurrence of quartz throughout the deposit in varying degrees of abundance. Essentially, it is a kaolinite monomineralic deposit, showing no distinct zonal distribution of alteration mineral assemblages.
K-Ar ages were measured on the whole-rock samples of felsic volcanics and the clays from the Hiraki mine. The results are: (1) 70.0 ± 1.5 Ma for the underlying Kamogawa rhyolite; (2) 68.9 ± 1.6 Ma and 69.1 ± 1.6 Ma for the least-altered and altered rhyolitic non-welded tuff (= ore horizon), respectively; (3) 67.6 ± 1.5 Ma for the unconformably overlying rhyolitic Hiraki welded tuff; and (4) 63.8 ± 1.5 Ma for clays from the fracture zone crosscutting the whole volcanic sequence. These ages are fairly consistent with the volcanostratigraphy in the mine area, indicating that various volcanisms and mineralization took place within a time span of a few million years in the latest Cretaceous to the earliest Tertiary Periods. The unconformity recognized, therefore, means a minor time-gap, not a long cessation of volcanism, at least in the Hiraki mine area.
Immobility of element during the hydrothermal alteration was tested for Al2O3, TiO2, Nb and Zr. Among them Zr and Nb show highest immobility throughout the deposit. Mass-transfer calculation, based on Zr as an immobile element monitor, and petrographic evidences reveal the followings: (1) dissolution of CaO, Na2O, K2O, Fe2O3 and MnO in the kaolinite zone; (2) slight enrichment of K2O and Fe2O3 in the sericite "zone"** and chlorite "zone", respectively; (3) dissolution of silica at the initial stage and later precipitated as quartz from the fluid introduced to the system during alteration process; (4) immobility of Zr and Nb, as well as less immobility of TiO2 and Al2O3 in the whole deposit; and (5) Al2O3 content in the non-welded tuff might vary within the range of 15 to 24 wt% .
Hydrogen and oxygen isotopic studies were carried out on the ore specimens and whole-rock samples to obtain information on the type of water responsible for alteration process and nature of water-rock interaction at the Hiraki mine. δD and δ18O analyses, expressed as ‰ relative to SMOW, of the volcanic host rocks and ores (kaolinite+quartz) reveal that all the volcanic sequences display their magmatic δ18O nature but with D-depleted nature except for the overlying rhyolitic welded tuff, while ores, exclusively localized in the underlying non-welded tuff, are much depleted in both D and δ18O compared to a protolith, the non-welded tuff. Hydrothermal quartz shows δ18O values of around 8‰. Kaolinite has a fairly consistent isotope values at δD of -94 ‰ and δ18O of 0.9 ‰. Among alteration minerals, chlorite shows isotopically lowest value in δD of -138 ‰, probably due to its Fe-rich composition.
Equilibrium oxygen isotopic fractionation between quartz and kaolinite in ore is indicative of the formation temperature of about 150℃ , which is consistent with the preservation of marcasite throughout the ore deposit. Pressure prevailing in the alteration process is estimated to be vapor pressure. The fluid-rock interaction is likely to have been isothermal and isobaric process. Isotope-shift and the presence of marcasite in the deposit indicate that the kaolin mineralization occurred under acidic condition (pH < 5).
Fluid flux, which passed through the system during the hydrothermal alteration, is calculated based on the material-balance of K+ between fluid and precursor. The result reveals that the amount of fluid would be as much as 15000 times (in volume) of the host rock.
Infiltration metasomatism is a possible mechanism of the hydrothermal alteration at the Hiraki kaolin deposit. The fluid incurred the non-welded tuff at the inlet of metasomatic column was equilibrium with kaolinite and would have a concentration of log mK+ = -4.29 and log mAl3+ = -11.39 at assumed concentration of log mSiO2 = -2.8 and pH = 4.79. Most of time in the course of alteration, hydrothermal fluid would be slightly undersaturated with respect to muscovite. Non-stoichiometric clays were formed in place of muscovite. It is pointed out that the hydrothermal alteration process could be primarily controlled by composition, especially of a H+ (pH), of the fluid which reacted with the host rock.
Nature of the isotope shift between the host-rock and ores suggests that the water, i. e. ore fluid, reacted with the non-welded tuff would be a mixture of acidic meteoric water with isotope values of δD = -120‰ and δ18O = -16‰, and magmatic water. Possible existence of meteoric water with such extremely light isotope values in the Hiraki mine area suggests that the Japanese island would be situated at much further north of the present location during the latest Cretaceous to the earliest Tertiary.
During the formation of kaolin deposit, the Hiraki welded tuff served as a cap-rock. Possible sources of magmatic water and heat would be the latest Cretaceous granite intrusions, though not exposed in the mine area, but crop out somewhere in the Hyogo Prefecture. A possible model of formation of the Hiraki kaolin deposit is presented.
** The word "zone" is used as an informal term for the mineral assemblages with indistinctive boundary in spatial distribution.