Shuichi HATTORI Director, 1st Construction Division, Japan Railway Construction, Transport and Technology Agency

Transcription

1 PAPER Evaluation of Rock Characteristics for Acid Water Drainage from Rock Muck Takehiro OHTA, Dr.. Sci. Senior Researcher, Geology Laboratory, Disaster Prevention Technology Division Hideo KIYA, Dr.. Eng. Deputy Director, Administration Division Shuichi HATTORI Director, st Construction Division, Japan Railway Construction, Transport and Technology Agency Surface/ground water pollution caused by the inflow of acid water enriched with heavy metals poses a serious environmental problem around excavation sites where sulfide minerals chemically decompose. To establish a method to evaluate acid water drainage from rock, we examined the percolation and geochemical characteristics of rock. It became clear that it was possible to evaluate acid water drainage by using the following indices: () the ph value one hour after a batch-leaching test; () the rock's sulfur content; and () in the case of mudstone only, the S/Ca mole ratio of the rock. Keywords: rock muck, pyrite, acid water, sulfur content, S/Ca mole ratio. Introduction Surface/ground water pollution caused by the inflow of acid water enriched with heavy metals has become a serious environmental problem in the areas around mines [,]. This results from the sulfate discharge generated by the decomposition of pyrite contained in the altered rock in mining areas []. The same problem has arisen at engineering work excavation sites in recent years []. Therefore, if a tunnel is to be constructed in the vicinity of a mine area, the evaluation and prediction of the possibility of acid water drainage from rock muck and rock masses will be important environmental geology aspects. The,m-long Hakkoda tunnel is currently under construction between Hachinohe and Shin-Aomori Stations on the Tohoku Shinkansen network in the northern part of Honshu, Japan. Because there are a number of mines around this tunnel, it is feared that the peripheral environment would be polluted by acid water and heavy metals percolating from the rock muck released by the tunnel's excavation. Therefore, we investigated the geochemical features of rock distributed around the Hakkoda tunnel to establish a method to evaluate the acid water drainage. This paper clarifies the geochemical characteristics of rock that discharges acid water. We also propose a method to evaluate acid water drainage from rock that was used in the construction of the Hakkoda tunnel.. Drainage mechanism of acid water from rocks: dependence on water-rock interaction In water-rock interactions, acid water drainage is caused by the generation of sulfate and hydrogen ions due to the oxidation of pyrite in the rock (Equation ). The pyrite reacts with the oxygen contained in groundand/or rainwater. FeS + /O + H O = Fe + + SO - + H + () On the other hand, the rock generally contains calcite and feldspar that can neutralize acid water. When these minerals contact acid water they restrain ph loss in the water because they consume the sulfate ions and change to clay minerals. Therefore, the potential for acid water drainage from rock can be determined by comparing the ratio of the amount of sulfate ion generated from rock to that consumed by buffer mineral reactions.. Samples and investigation method. Geological setting and samples The Tertiary rocks of the Kanegasawa, Yotsuzawa, Wadagawa and Ichinowatari Formations are distributed around the Hakkoda tunnel and are intruded by Tertiary igneous rocks. Samples of these Tertiary rocks were collected from the face of the excavations and pilot drilling cores of the Hakkoda tunnel. We examined ore and/ or clay vein samples,, igneous rock samples,, tuffaceous rock samples and sedimentary rock samples.. Major,, trace and heavy metal element content In order to understand the chemical features of rock, we investigated major element (SiO, TiO, Al O, total iron as Fe O, MnO, MgO, CaO, Na O, K O, P O ) and trace and heavy metal element (S, Cu, Pb, Zn, As, Cd, Se, Cr, Ni, Ba) content in the samples by X-ray fluorescence (XRF, Rigaku ZSXe). The analysis procedure was as follows. ) Rock samples were dried completely in an electric oven. ) Powder samples less than micrometers in size were created by crushing from the dry samples. ) Pressing the powder samples produced tablet samples for QR of RTRI, Vol., No., Sep.

2 XRF spectroscopy. ) The tablet samples were analyzed by XRF.. X-ray diffraction experiments Representative samples were examined by X-ray powder diffraction (XRD) in order to grasp the influence of the mineral assemblages on acid water drainage. We examined six ore vein samples, igneous samples, tuff samples and mudstone samples.. Batch-leaching test A simplified batch-leaching test [] was carried out for all collected samples to examine the chemical features of the rock leachate. The procedure for this test was as follows. ) The samples were crushed to grains smaller than mm in size after drying. ) g portions of crushed samples were mixed with ml of distilled water and the mixture samples shaken for three minutes to produce a leachate. ) The ph and electric conductivity of the leachate were measured after minutes, hour, hours, days, days and days. ) After days, the concentrations of metal elements (Cu, Pb, Zn, Fe, As, Cd, Cr, Mn, Se) and cations (Na +, K +, Ca +, Mg + ) in the leachate were determined by ICP (Inductively coupled plasma) emission spectrometry (Shimazu ICPS-). The SO - and Cl - content in the leachate were determined by ion chromatography, and the HCO - content measured by the sulfate titration method. Leachate ph after days Leachate ph after days SiO wt% Na O wt%. Rock drainage characteristics. Drainage features and mechanism in mudstone The relationship between leachate ph levels after days and the contents of representative elements in mudstone samples is illustrated in Fig.. The mudstone samples, which become acid in leachate, are siliceous sediments containing SiO more than wt % that are rich in sulfur (>. wt %) and poor in CaO (<. wt %). Table shows the mineral assemblages of the representative samples examined by XRD. The mudstone contained quartz, pyrite and feldspar as major minerals, and included calcite, mica group, kaolinite, chlorite, smectites and gypsum as trace minerals. There was only a very small amount of buffer minerals such as calcite and feldspar in the samples that gave low in ph values in leachate. As the leachate changed from alkaline to acid, SO -, Ca + and Mg + content increased, and that of Na + and HCO - decreased (Fig. ). The concentration of Ca + in leachate correlated closely not only with the ph value but also with SO - content. As the examination results explain above and in keeping with the water-rock interaction theory mentioned in the Chapter, the minerals that dissolved during the mudstone drainage process were pyrite, calcite, feldspar and the mica group. The leachate became acid due to pyrite decomposition. Calcite, feldspar and the mica group acted as buffer minerals. Ohta et al. [] produced the same hypothesis based on the chemical change of mudstone samples in a batch-leaching test. Leachate ph after days Leachate ph after days CaO wt% S wt% Fig. Relationship between leachate ph and contents of representative elements in mudstone QR of RTRI, Vol., No., Sep.

3 Ion content in leachate (mg/l) SO -, Ca +, Na +. Leachate ph after days in batch-leaching test Fig. Ion content against ph in mudstone leachate Leachate SO - content (mg/l)... Mudstone sulfur content (wt %) Fig. Leachate sulfate content against mudstone sulfur content The SO - content in the leachate increased as the mudstone sulfur content increased (Fig. ). This suggests that SO - is released due to the dissolution of pyrite in the mudstone and that the concentration of this ion depends on mudstone sulfur content. Therefore, it is thought that the mudstone discharges sulfate into the leachate due to the dissolution of pyrite in the early drainage stages and that the amount of sulfate is in proportion to mudstone pyrite content. If the leachate becomes acidic, Ca + and Mg +, etc. are discharged into the leachate because calcite, feldspar and the mica group are decomposed by the reaction with sulfate. This process is inferred from the good correlation between the contents made up of these ions and the ph, SO - content in the leachate (cf. Fig. ). The concentration of Ca + in the leachate does not correlate with the CaO content of mudstone. The relationship between Mg + and MgO is similar to the relation between Ca + and CaO. Therefore, it is presumed that the volume of dissolved buffer minerals matches the amount of sulfate in the leachate. When the buffer minerals can react with the leachate because these minerals are present in the mudstone, ph loss will be restrained in spite of the sulfate discharge. CaO content in tuffaceous rock (wt %) Na O content in tuffaceous rock (wt %) SiO content in tuffaceous rock (wt %) Sulfur content in tuffaceous rock (wt %) Fig. The relationship between leachate ph and representative element contents in tuffaceous faceous rock Leachate ph after days Leachate ph after days Leachate ph after days Leachate ph after days QR of RTRI, Vol., No., Sep.

4 . Drainage features in tuffaceous rocks Figure shows the relationship between leachate ph after days and the contents of representative elements in tuffaceous rock samples. The samples that become acid in leachate are rich in sulfur (>. wt %) and poor in CaO. The relationships between the mineral assemblages and leachate ph after days in tuffaceous rock samples are as follows (cf. Table ). ) The samples that became acidic in leachate contained pyrophyllite and kaolinite. ) The leachate of the tuffaceous samples that included calcite, feldspar and smectite remained neutral or alkaline for days. ) The samples that showed a high leachate ph value contained a zeolite group and magnetite. Mg + and Ca + content increased and Na + content decreased, as the leachate ph was lost in the same way as the mudstone samples. However, SO - (Fig. ) and HCO - content showed no correlation with ph. There was a good correlation between the SO - content and the Ca + content in leachate (Fig. ). There was a close correlation between SO - content in the leachate and sulfur content in tuffaceous rock samples. However, there was no corre- Sample Number Rock type Table Mineral assemblages of representative samples Mineral assemblage S wt% ph of leachate after days Quartz Feldspar mudstone mudstone mudstone mudstone ± + + mudstone ± ± ± ± mudstone ± mudstone ± ± ± ± mudstone ± ± + fine tuff ± + coarse tuff ± + ± fine tuff ± + + coarse tuff fine tuff ± fine tuff ± ± ± + ± + coarse tuff ± ++ tuff breccia coarse tuff ± lapilli tuff.. ± ± ± ± lapilli tuff.. ± ± ± volcanic breccia andesite dacite ± + dacite ± andesite ± + ± basalt.. ± ± +++ andesite ± ± ± + ± basalt ± dacite ± ± andesite ± basalt ± + ± andesite ± + ± ++ ± ± andesite ± + ± dacite ± ± ++ dacite ± + ++ andesite clay/sphalerite vein.. + ± ++ clay/sphalerite vein ± + ± clay/pyrite vein ± + ± : Very abundant, ++ : Abundant, + : Medium, ± : Poor Pyrite Magnetite Mica group Chlorite Smectite Kaolinite Pyrophyllite Zeolite group Calcite QR of RTRI, Vol., No., Sep.

5 Leachate ph after days Leachate SO- content (mg/l). Intrusive rocks Kanegawawa F.. - Fig. Sulfate content against leachate ph in tuf faceous tuffaceous rock Intrusive rocks Leachate SO- content (mmol/l) Leachate calcium ion content against the sulfate faceous rock tuffaceous content in tuf lation between leachate Ca+ and CaO content in tuffaceous rock samples.. Drainage features in igneous rocks Most of the igneous samples that became acid in leachate were rich in sulfur (>. wt %, Fig. ), and poor in CaO (<. wt %) and NaO (<. wt %). The samples that showed a low ph value in leachate included pyrite and kaolinite and were poor in feldspar, calcite and smectite (Table ). Figure shows that the leachate of some samples, which contained less than. wt % of sulfur, became acid. These samples consisted of Si, Al, Fe and S only because of acid alteration and contained pyrite, pyrophyllite and kaolinite. There was a good correlation between leachate SO and Ca+ content (Fig. ), although there was no correlation between the ion content and ph levels (ex. Fig.). The SO- content increased with sulfur content in rock samples (Fig. ). QR of RTRI, Vol., No., Sep. Leachate SO- content (mmol/l) Fig. Leachate calcium ion content against sulfate content in igneous rock Leachate SO- content (mg/l) Fig. - Fig. Leachate ph against sulfur content in igneous rock Leachate Ca+ content (mmol/l) Leachate Ca+ content (mmol/l) Sulfur content in igneous rock (wt %) - Leachate ph after days Intrusive rocks Kanegawawa F... Leachate ph after days Fig. Leachate sulfate content against ph in igneous rock

6 Leachate SO - content (mg/l) - Fig. Intrusive rocks Sulfur content in igneous rock (wt %) Leachate sulfate content against sulfur content in igneous rock. Drainage mechanism in tuffaceous and igneous rock From the relationship between the mineral assemblage and ph in leachate, it is inferred that the quality of leachate is determined by the dissolution of several minerals: ) Pyrite, pyrophyllite and kaolinite will take part in acid leachate generation. ) Calcite, feldspar, smectite and the zeolite group will neutralize acid leachate. Leachate cannot become acidic from the decomposition of pyrophyllite and/or kaolinite, because these clay minerals do not contain sulfur and remain stable at atmospheric pressure and at normal temperatures. As the sulfur content of the rock samples increased, the leachate SO - content increased. This suggests that SO - is generated by pyrite dissolution with the content of this ion depending on rock sulfur content. Thus it is thought that tuffaceous and igneous rocks discharge sulfate into the leachate due to pyrite dissolution and that the amount of sulfate is in proportion to pyrite content in rocks like mudstone. Because there is high correlation between SO - and Ca + content in both tuffaceous and igneous rock leachate, it is inferred that Ca + is released by the decomposition of calcite and/or feldspar with leachate acidification. However, the ion contents have no relation with leachate ph. This suggests that leachate quality is not determined by the relation between the amount of sulfate released by the pyrite decomposition and the amount of sulfate consumed by the dissolution of Ca-containing minerals in tuffaceous and igneous rock. Therefore, it is expected that smectite and the zeolite group have a bearing on leachate quality. Unfortunately, however, we cannot clarify the reaction mechanisms of these minerals with water under atmospheric conditions in this paper.. Evaluation method for acid water drainage from rock In this chapter, we discuss a method for evaluating acid water drainage from rock muck based on the previously mentioned mechanism of acid water drainage from rock. This method was adopted for the construction of the Hakkoda tunnel.. Method for tuffaceous and igneous rock As mentioned above, we can estimate the acidification mechanism of the leachate from the tuffaceous and igneous rock caused by pyrite dissolution, but the neutralization mechanism of the leachate is not clear. Therefore, we adopted a method that evaluates only the ability of the leachate to acidify by pyrite decomposition. Ore veins distributed around the Hakkoda tunnel can be distinguished by eye observation because of the high content of pyrite. When these veins exist in the working face of a tunnel, there is judged to be a possibility that the acid water is made from the rock muck from the face. If sulfur is contained in the tuffaceous and igneous rock samples at levels higher than. wt %, many leachate samples of these rocks become acidic (Figs. and ). Therefore, we regard rock muck that includes rock containing more than. wt % of sulfur as being able to acidify the leachate. A batch-leaching test can assess directly the acidification of leachate from rock. There is no doubt that the final leachate will become acidic if the leachate indicates a ph value lower than. after one hour (Fig. ). Therefore the batch-leaching test was adopted as the evaluation method, with a ph of. as the critical value for leachate acidification. Leachate ph after days Igneous rock Tuffaceous rock Mudstone Leachate ph after hour Fig. ph value after days against ph after one hour. Method for mudstones Igarashi et al. [] explained that the mole ratio of carbonate carbon to sulfur in mudstone represents the possibility of acid water drainage from mudstone. In this study, we proved that not only calcite but also feldspar and the mica group are buffer minerals. Because these buffer minerals are the main calcium-containing minerals in mudstone distributed around the Hakkoda tunnel, the Ca content of mudstone is an index that indicates buffer mineral concentrations. Therefore, it is thought that the mole ratio of sulfur content to calcium content represents the ratio of the amount of pyrite to that of the QR of RTRI, Vol., No., Sep.

7 Leachate ph after days S < wt % S >= wt % Fig. Leachate ph against mudstone S/Ca mole ratio buffer minerals in mudstone. Figure shows the relationship between the mudstone S/Ca mole ratio and leachate ph after days. As the S/Ca mole ratio of mudstone increases, the ph value of the leachate declines. The ph value becomes lower than. if the S/Ca mole ratio is higher than.. Hence, it seems that the S/Ca mole ratio can be used to evaluate the possibility of acid water drainage from mudstone.. Rock evaluation flow chart for acid water drainage Figure shows the acid water drainage evaluation flow chart for rock that was adopted in the construction Eye observation Ore veins Batch-leaching test ph value after hour =<. Chemical analysis Mudstone S>=. Mudstone S/Ca mole ratio S>=. or S/Ca>=. Controlled muck Controlled muck Controlled muck Uncontrolled muck Controlled muck Uncontrolled muck Fig. Rock evaluation flow chart of rocks for acid water drainage of the Hakkoda tunnel. This flow chart was designed based on the analytical time of each evaluation item in consideration of the operational excavation procedures at the tunnel.. Conclusions We examined the drainage features and geochemical characteristics of the rock distributed in the Hakkoda tunnel. From the results of our examinations, it is thought that sulfuric acid is released by the resolution of sulfide minerals and that the calcium-containing minerals will neutralize the sulfuric acidic water. Therefore, it is possible to evaluate the drainage ability of acid water from rocks distributed in mine areas by using the following indices: () leachate ph value one hour after batch-leaching test, () the sulfur content of the rock and () in the case of mudstone, the S/Ca mole ratio. Acknowledgements The authors would like to thank all the members of staff at the Hakkoda tunnel testing laboratory for their technical assistance and the members of the Hakkoda tunnel committee for their advice on the issues treated in this paper. The authors are also indebted to all the engineers engaged in the construction of the Hakkoda tunnel. References ) Taylor, B. E., Wheeler, M. C. and Nordstrom, D. K., "Stable isotope geochemistry of acid mine drainage: Experimental oxidation of pyrite," Geochimica et Cosmochimica Acta, Vol., pp. -,. ) Kurosawa, K., "The acid mine drainage of the Nissan- Toi mine, Kameda Peninsula, Southwest Hokkaido," Report of the Geological Survey of Hokkaido, Vol., pp. -, (in Japanese with English abstract). ) Singer, P. C. and Stumm, W., "Acidic mine drainage: the rate determining step," Science, Vol., pp. -,. ) Nosaka, T., Isahai, H., Kawagoe, T., Katayama, M. and Ishihama, S., "Several examples of environmental geological problems on some construction works," presented at the JSEG Annual Meeting, Kyoto, Japan, October - November,, pp. - (in Japanese). ) Hattori, S, Ohta, T. and Kiya, H., "Engineering geological study on exudation of acid water from rock mucks. - Evaluation method of rocks at the Hakkouda tunnel near mine area -," Jour. Japan Soc. Eng. Geol., Vol., pp. -, (in Japanese with English abstract). ) Ohta, T., Kiya, H., Hattori, S. and Asakura, T., "Elution characteristics of fresh mudstone from the underground opening," presented at the st Kyoto International Symposium in Underground Environment, Kyoto, Japan, March -,, Environmental QR of RTRI, Vol., No., Sep.

8 Rock Engineering UE-, pp.-. ) Igarashi, T., Oyama, T. and Saito, N., "Experimental study on acidification potential of leachate from sedimentary rocks containing pyrite," Jour. Japan Soc. Eng. Geol., Vol., pp. -, (in Japanese with English abstract). QR of RTRI, Vol., No., Sep.