The Diavik Waste Rock Project: Design,construction, and instrumentation of field-scale experimental waste-rock piles

The physicochemical processes that affect acid mine drainage (AMD) in unsaturated waste rock piles and the capabilities of small-scale laboratory experiments to predict AMD from waste rock are not well understood. An integrated laboratory and field study to measure and compare low sulfide waste rock and drainage characteristics at various scales has been initiated. This paper describes the design, construction and instrumentation of three field-scale experimental waste rock piles (test piles), and six active zone lysimeters at the Diavik diamond mine in the Northwest Territories, Canada. The test piles are comprised of granitic and sulfide-bearing metasedimentary waste rock excavated during open pit mining operations. One test pile contains waste rock with a target S content of <0.04 wt.% S; the second test pile contains waste rock with a target S content of >0.08 wt.% S; and the third test pile contains the higher sulfide waste rock (>0.08 wt.% S) and was re-sloped and capped with a low permeability till layer and a low sulfide waste rock cover. The first two test piles are approximately 15 m high with bases of 50 m by 60 m, and the re-sloped test pile has a larger base of 80 m by 125 m. Instrumentation was selected to measure matrix flow, geochemistry of pore water and drainage, gas-phase O₂ concentration, temperature evolution, microbiological populations, waste rock permeability to air, and thermal conductivity, as well as to resolve mass and flow balances. Instrument locations were selected to characterize coupled physicochemical processes at multiple scales and the evolution of those processes over time. Instruments were installed at a density such that the number of instruments that survived construction (40% to >80% by instrument type) was sufficient to allow adequate characterization of the physicochemical processes occurring at various scales in the test piles.     

The Diavik Waste Rock Project: Geochemical and microbiological characterization of low sulfide content large-scale waste rock test piles

Two experimental waste-rock piles (test piles), each 15 m in height × 60 m × 50 m, were constructed at the Diavik diamond mine in Northern Canada to study the behavior of low-sulfide content waste rock, with a similarly low acid-neutralization potential, in a continuous permafrost region. One test pile with an average of 0.035 wt.% S (<50 mm fraction; referred to as Type I) and a second test pile with an average of 0.053 wt.% S (<50 mm fraction; referred to as Type III) were constructed in 2006. The average carbon content in the <50 mm fraction of waste rock in the Type I test pile was 0.031 wt.% as C and in the Type III test pile was 0.030 wt.% as C. The NP:AP ratio, based on the arithmetic mean of particle-size weighted NP and AP values, for the Type I test pile was 12.2, suggesting this test pile was non-acid generating and for the Type III test pile was 2.2, suggesting an uncertain acid-generating potential. The Type I test pile maintained near-neutral pH for the 4-year duration of the study. Sulfate and dissolved metal concentrations were low, with the exception of Ni, Zn, Cd, and Co in the fourth year following construction. The pore water in the Type III test pile contained higher concentrations of SO₄²⁻ and dissolved metals, with a decrease in pH to <4.7 and an annual depletion of alkalinity. Maximum concentrations of dissolved metals (20 mg L⁻¹ Ni, 2.3 mg L⁻¹ Cu, 3.7 mg L⁻¹ Zn, 35 μg L⁻¹ Cd, and 3.8 mg L⁻¹ Co) corresponded to decreases in flow rate, which were observed at the end of each field season when the contribution of the total outflow from the central portion of the test pile was greatest. Bacteria were present each year in spite of annual freeze/thaw cycles. The microbial community within the Type I test pile included a population of neutrophilic S-oxidizing bacteria. Each year, changes in the water quality of the Type III test-pile effluent were accompanied by changes in the microbial populations. Populations of acidophilic S-oxidizing bacteria and Fe-oxidizing bacteria became more abundant as the pH decreased and internal test pile temperatures increased. Irrespective of the cold-climate conditions and low S content of the waste rock, the geochemical and microbiological results of this study are consistent with other acid mine drainage studies; indicating that a series of mineral dissolution–precipitation reactions controls pH and metal mobility, and transport is controlled by matrix-dominated flow and internal temperatures.     

The Diavik waste rock project: Initial geochemical response from a low sulfide waste rock pile

Three large-scale experimental waste rock piles (test piles) were constructed and instrumented at the Diavik Diamond Mine in the Northwest Territories, Canada, as part of an integrated field and laboratory study to measure and compare physical and geochemical characteristics of experimental, low sulfide waste rock piles at various scales. This paper describes the geochemical response during the first season from a test pile containing 0.053 wt.% S. Bulk drainage chemistry was measured at two sampling points for pH, Eh, alkalinity, dissolved cations and anions, and nutrients. The geochemical equilibrium model MINTEQA2 was used to interpret potential mineral solubility controls on water chemistry. The geochemical response characterizes the initial flushing response of blasting residues and oxidation products derived from sulfides in waste rock exposed to the atmosphere for less than 1 year. Sulfate concentrations reached 2000 mg L⁻¹ when ambient temperatures were >10 °C, and decreased as ambient temperatures declined to <0 °C. The pH decreased to <5, concomitant with an alkalinity minimum of <1 mg L⁻¹ (as total CaCO₃), suggesting all available alkalinity is consumed by acid-neutralizing reactions. Concentrations of Al and Fe were <0.36 and <0.11 mg L⁻¹, respectively. Trends of pH and alkalinity and the calculated saturation indices for Al and Fe (oxy)hydroxides suggest that dissolution of Al and Fe (oxy)hydroxide phases buffers the pH. The effluent water showed increased concentrations of dissolved Mn (<13 mg L⁻¹), Ni (<7.0 mg L⁻¹), Co (<1.5 mg L⁻¹), Zn (<0.5 mg L⁻¹), Cd (<0.008 mg L⁻¹) and Cu (<0.05 mg L⁻¹) as ambient temperatures increased. Manganese is released by aluminosilicate weathering, Ni and Co by pyrrhotite [Fe_1−xS] oxidation, Zn and Cd by sphalerite oxidation, and Cu by chalcopyrite [CuFeS₂] oxidation. No dissolved metals appear to have discrete secondary mineral controls. Changes in SO₄, pH and metal concentrations indicate sulfide oxidation is occurring and effluent concentrations are influenced by ambient temperatures and, possibly, increasing flow path lengths that transport reaction products from previously unflushed waste rock.     

Influence of freeze–thaw dynamics on internal geochemical evolution of low sulfide waste rock

Continuous monitoring of a 15 m high heavily instrumented experimental waste rock pile (0.053 wt.% S) since 2006 at the Diavik diamond mine in northern Canada provided a unique opportunity to study the evolution of fresh run-of-mine waste rock as it evolved over annual freeze–thaw cycles. Samples were collected from soil water solution samplers to measure pore water properties, from twelve 4 to 16 m² basal collection lysimeters to measure basal leachate properties in the region underlying the crest of the pile (the core), and from basal drains to measure aggregate total pile leachate properties. By 2012, monitoring of pore water geochemistry within the core structure of the test pile revealed an apparent steady state with respect to weathering geochemistry, represented by (i) a flush of pre-existing blasting residuals and applied tracers, (ii) declining pH, (iii) a stepwise progression and subsequent equilibrium with acid-neutralizing phases (depletion of available carbonates; equilibrium with respect to aluminum hydroxide phases and subsequent iron (III) hydroxide phases), and (iv) concordant release of SO₄, major cations (Ca, Mg, K, Na, Si), and trace metals (Al, Fe, Ni, Co, Cu, Zn). Distinct, high concentration ‘spring flushes’, characteristic of drainage in northern environments and primarily explained by a combination of fluid residence time and the build-up of oxidation products over the winter, were released from core drainage each season. Following the initial flush, the concentration of all dissolved constituents steadily declined, with distinct minimums prior to freeze-up. The opposite trend was observed in the cumulative pile drainage, in which early season leachate dominated by snowmelt and batter flow had low concentrations and late season leachate dominated by contributions from the core of the pile (indicated by season end merging of core and cumulative drainage geochemistry) had higher concentrations. Northern waste rock pile drainage geochemistry is strongly influenced by freeze–thaw cycling and varying core and batter subsystem contributions to total drainage. A comprehensive understanding of thermal cycling in waste rock piles is an important component of temporal predictions of drainage water composition based on up-scaling or reactive transport modeling.     

The Diavik waste rock project: Water flow through mine waste rock in a permafrost terrain

A field experiment is being carried out at the Diavik diamond mine in northern Canada to investigate the influence of unsaturated flow behavior on the quality of drainage from mine waste rock piles in a region of continuous permafrost. This paper is part of a series describing processes affecting the weathering of waste rock and transport of reaction products at this site; here the focus is on unsaturated water flow and its role in mass loading. Two 15 m-high instrumented test piles have been built on 60 m by 50 m collection systems, each consisting of lysimeters and a large impermeable high-density polyethylene (HDPE) liner. Collection lysimeters are installed nearby to investigate infiltration in the upper 2 m of the waste rock. Porosity, water retention curves, and hydraulic conductivity functions are estimated from field measurements and for samples ranging in size from 200 cm³ to 16 m³. Net infiltration in 2007 is estimated to have been 37% of the rainfall for mean annual rainfall conditions. Early-season infiltration freezes and is remobilized as the waste rock thaws. Wetting fronts migrate at rates of 0.2–0.4 m d⁻¹ in response to common rainfall events and up to 5 m d⁻¹ in response to intense rainfall. Pore water and non-reactive solutes travel at rates of <10⁻² to 3 × 10⁻² m d⁻¹ in response to common rainfall events and up to 0.7 m d⁻¹ in response to intense rainfall. Time-varying SO₄ mass loading from the base of the test piles is dictated primarily by the flow behavior, rather than by changes in solute concentrations.     

Numerical modeling of contaminated neutral drainage from a waste-rock field test cell

Contaminated drainage related to the leaching of soluble metals under near-neutral conditions, known as contaminated neutral drainage (CND), may arise when metal species are soluble at neutral pH. Such a phenomenon has been sporadically observed in effluent from the Tio mine waste-rock pile in Quebec, Canada, particularly from older sections of the pile, where Ni concentrations are increasing with time. It has been postulated that Ni is retained within the fresh waste rock as sorbed species, but as the rock ages, sorption sites become saturated and more Ni is released to the effluent. A field test program was initiated to evaluate the geochemical behavior of the waste rock. This paper presents a numerical analysis of CND generation from waste-rock field test cells including water flow and multi-component transport with geochemical reactions (e.g. sulfide oxidation, pH neutralization, and Ni sorption), using the code MIN3P. The model was able to represent the delay before Ni is seen in the effluent, as caused by sorption of Ni onto Fe-oxide particles. Once the sorption sites are saturated, the model allows Ni release into the effluent by millerite dissolution, expressed by the shrinking core model. A sensitivity analysis indicated that sorption parameters significantly affected the simulated results, so their selection should be based on sound independent field or experimental data.     

Response of 1 × 2 pile group under eccentric lateral loading

A large scale model test of a 1 × 2 pile group was conducted in silts to investigate its behavior under eccentric lateral loading. The model pile group consisted of two well instrumented steel piles and was installed in a large soil tank with a close spacing of three-pile diameters on centers. The test results revealed that the eccentricity of lateral loads had limited effect on the overall performances of the 1 × 2 pile group, but significantly contributed to the unevenness of internal forces of the individual piles. The coupling effect between the lateral deflection and torque gave rise to the substantial increase in the torsional resistance of individual piles within the group, comparing to that of a torsionally loaded single pile. The contribution provided by the torsional resistances of individual piles in resisting the external torque continually decreased when the applied lateral load increased. In addition, a three-dimensional finite-element analysis for the pile group was performed and the simulated response was found to be in good agreement with the measured test results. Based on the same model, more cases with different loading conditions were further analyzed. It could be concluded from the analyses that the layout of individual piles within the group obviously affected the behavior of the 1 × 2 pile group under eccentric lateral loads.     

The Diavik Waste Rock Project: Measurement of the thermal regime of a waste-rock test pile in a permafrost environment

The interior thermal regime of a field-scale experimental waste rock pile in the Northwest Territories, Canada, was studied. Test pile construction was completed in the summer 2006, and temperature data was collected continuously since that time to February 2009. The temperature data indicates the test pile cooled over the study period, with an average heat energy release of −2.5 × 10⁴ and −2.6 × 10⁴ MJ in 2007 and 2008, respectively. The mean annual air temperature (MAAT) at the site was −8.9 °C during the period between 2006 and 2009, with a permafrost table at a depth of 4 m in bedrock away from the pile. Because of this cold environment, the upward movement rate of the 0 °C isotherm into the test pile at its base was approximately 1.5 m a⁻¹ during 2007 and 2008. Thermistor strings installed immediately below the base of the test pile showed the test-pile basal temperatures remained near and below 0 °C during the study period. Furthermore, due to low rates of sulfide mineral oxidation, elevated temperatures in the interior of the test pile were not observed. The average air velocity in the pore space in July 2007 and 2008 was about one third of that during January of each year based on temperature distributions. Therefore, due to higher air velocity during the winter, it is expected that heat transfer is greater during winter.     

Estimation of potential pollution of waste mining dumps at Peña del Hierro (Pyrite Belt,SW Spain) as a base for future mitigation actions

Peña del Hierro is an abandoned mine site located in the catchment area of the Tinto river (Pyrite Belt, SW Spain). As leaching from the spoils affect the quality of the stream water, the waste dumps have been characterized for mineralogy, geochemistry and granulometry to obtain an estimate of the potential pollution. Waste rock dumps in Peña del Hierro are very heterogeneous and are mainly composed of acid volcanic tuffs > gossan > shales > roasted pyrite ashes > floated pyrite. The volcanic tuffs, the gossan and the shales coexist in the same piles. The roasted pyrite ashes and the floated pyrite form more homogeneous dumps. The dissolution of pyrite concentrated in pyrite ashes and floated pyrite units can generate acid mine drainage. Nevertheless, acid volcanic tuffs, which are rich in pyrite and have no neutralizing minerals, are the main source of these acidic effluents. Only muscovite might partially neutralize the acidity, but the dissolution of this mineral is too slow to compensate for acidity. The occurrence of jarosite in the <2 mm fraction indicates that extreme acid mine drainage occurs. The gossan and roasted pyrite ashes have high contents of trace elements. According to their concentration, As (46–1710 ppm), Pb (113–3455 ppm) and Hg (0–53) are some of the most important toxic trace elements in these wastes. In dumps mainly composed of volcanic tuffs most of the trace elements derive from the gossan mixed in the piles. Gossan is stable in an oxidizing environment, but acidic effluents (pH < 2) can dissolve Fe oxyhydroxides from them and release high amounts of trace elements to the stream water. This research contributes to estimating the production of acid mine drainage and the actual contamination risk of potentially toxic elements in soils and waters of this area, and could be the base for possible future mitigation actions in other areas affected by mining wastes.     

Metal and arsenic distribution in soil particle sizes relevant to soil ingestion by children

Ingestion of soil is a common behaviour in young children as a means of exploring their surroundings. Much attention has been given to remediation of point-source polluted sites with regard to potential health risks for children. However, because of diffuse pollution and long-range atmospheric deposition, soil contaminant levels are generally increased in urban areas compared to their rural counterparts, even in areas located away from any point sources of pollution. Intake of urban soil can thereby result in significant amounts of the child’s daily metal intake. In the present study, soil samples were collected from 25 playgrounds around urban Uppsala, Sweden and analysed for contents of Al, As, Fe, Cr, Cu, Cd, Hg, Mn, Ni, Pb, W and Zn. Prior to aqua regia digestion, the samples were wet-sieved in order to separate soil particle fractions representing deliberate (<4 mm) and involuntary (<50 μm) soil ingestion by children, as well as a third size fraction of 50–100 μm representing soil that is easily transported by suspension. While the metal and As contents in the 50–100 μm fraction were similar to those of the <4 mm fraction, the <50 μm fraction had metal and As contents on average one and a half times higher than those of the <4 mm fraction. The metal and As contents correlated negatively with the sand content in both particle size fractions <4 mm and 50–100 μm, suggesting a general decrease in metal and As content with increasing sand content. However, a positive correlation was found between sand content and the metal and As contents of the finest fraction (<50 μm), suggesting that when the sand content is high, the bulk of the sorbed elements are on the finest particles. The difference between metal and As contents in the different size fractions was greater in the soil sample with the highest sand content than in the sample with the lowest sand content. This implies that texture is a significant factor in metal and As distribution in soils with moderate metal and As contents, when the number of binding sites associated with small particles is low. Tolerable daily intake (TDI) values for Pb and As were exceeded at all sites, and at two sites for Cd, for children with pica behaviour. A high ingestion rate of mainly small particles could also result in the TDI value for Pb being exceeded at 10 sites and that for As at one site. This study also found that soil analysis by the procedure recommended by Swedish authorities accurately represents the metal intake from deliberate soil ingestion, whereas involuntary soil ingestion of mainly small particles could result in metal intakes which are up to twice as high.     

Sulfidic and non-sulfidic indium mineralization of the epithermal Au–Cu–Zn–Pb–Ag deposit San Roque (Provincia Rio Negro,SE Argentina) — with special reference to the “indium window” in zinc sulfide

At San Roque in Patagonia's Rio Negro Province, Argentina, an In–Au–Cu–Zn–Pb–Ag mineralization (< 0.24 wt.% In, < 7 ppm Au, < 0.45 wt.% Cu, < 14.1 wt.% Zn, < 0.55 wt.% Pb, < 60 ppm Ag) is bound to lava, and volcaniclastics of Triassic through Jurassic age. The polymetallic sulfidic and non-sulfidic indium mineralization is attributed to the low-sulfidation (LS) to intermediate sulfidation (IS) epithermal type of mineralization. Its vein-type and stockwork mineralization developed at 39.2 bars under hydrostatic conditions, corresponding to a depth of 400 m below the water level of the paleoaquifer. In the redox-controlled hypogene mineralization, the temperature increased from 130 °C up to as much as 250 °C at depth, while the pH regime changed from slightly acidic near surface to more alkaline conditions around pH 8 at a depth of approximately 150 m. The monophase mineral associations composed of sphalerite, Ag–Bi-enriched and inclusion-free galena (< 1.7 wt.% Ag, < 3.7 wt.% Bi), chalcopyrite, pyrite, gold, silver, digenite, various In–Cu- and Pb–Zn–Ag “intermediate products”, wittichenite, roquesite, sakuraiite, dzhalindite, brochantite, antlerite, cerussite, and “manganomelane” in a quartz and muscovite-rich gangue have been subdivided into three different stages: (1) Stockwork mineralization of LS to IS epithermal type (hypogene), (2) quartz vein mineralization (hypogene), and (3) salar mineralization (supergene–hypogene).Salt–mud flats controlled the youngest mineralization with Mn, Li, Ca, Mg, V, Sr, Cu, Ag and In bound to oxides, hydroxides, sulfates and subordinate carbonates. The quartz vein mineralization is made up of oxides, hydroxides prevailing over sulfides and containing W, Fe, Au, As, Pb, In, and Cu. It formed at the passage from the vadose into the phreatic zones under oxidizing to slightly reducing conditions. The level marks the boiling level of the hydrothermal solutions involved in the mineralizing process. The hypogene stockwork mineralization is exclusively made up of sulfides containing Zn, Pb, Cu, In, Ag and Bi in the phreatic zones. It developed under reducing conditions. Indium is present at all levels within the volcanic rocks and has been derived from sphalerite rich in Cd (< 1.6 wt.% Cd), In (< 7.3 wt.% In) and Cu (< 7.2 wt.% Cu) while the Fe contents are moderate in sphalerite (< 6.8 wt.% Fe). Indium reached economic grade only through the segregation of a Cu–In–S phase in the “indium window” which is defined by a Cd content of sphalerite in the range 0.2–0.6 wt.% Cd. This concentration of In is controlled by the crystal morphology and the lattice parameters of the minerals involved. It is described as a two-stage process with interdiffusion processes in an Fe-enriched system (stage I) and zoned replacement in an Fe-poor system enriched in indium (stage II). Cu-bearing sphalerite decomposed into sphalerite poor in trace elements and into Cu–In-bearing sphalerite. Further indium concentration took place, when roquesite and sakuraiite decomposed along with an increase in oxygen pressure under hypogene and supergene conditions into dzhalindite. The physical–chemical conditions of the mineralogy and chemical changes in the system In–Cu–Zn–Cd observed in nature have been approximated based upon the results obtained during laboratory studies in material sciences that were focused on solar energy.     

Numerical analysis of pile response to open face tunnelling in stiff clay

Three-dimensional (3D) finite element analyses have been performed to study the behaviour of a single pile and 3 × 3 and 5 × 5 pile groups during open face tunnelling in stiff clay. Several governing factors, such as tunnelling-induced ground and pile settlement, axial pile force changes and shear transfer mechanism at the pile–soil interface, have been studied in detail. Tunnelling resulted in the development of pile head settlement larger than the free-field soil surface settlement. In addition, axial force distributions along the pile change substantially due to changes in the shear transfer between the pile and the soil next to the pile, which triggers tunnelling-induced tensile forces in the piles with tunnel advancement. It was found that the relative displacements and the normal stresses at the pile–soil interface drastically affected shear transfer. The extent of slip length along a pile increased as the tunnelling proceeded. The apparent allowable pile capacity was reduced by up to approximately 42% due to the development of tunnelling-induced pile head settlement. Shear stress on the pile was increased for most of the pile depth with tunnel advancement, which was associated with changes in soil stresses and ground deformation, and hence, the axial pile force was gradually reduced with tunnel advancement, indicating the development of tunnelling-induced tensile pile force. The maximum tunnelling-induced tensile force on the pile was approximately 0.33P_a, where P_a is the allowable pile capacity applied to the pile head prior to tunnel excavation. The range affected by tunnelling in the longitudinal direction may be identified as approximately −2D ∼ +(1.5–2.0D), where D is the tunnel diameter, from the pile centre (behind and ahead of the pile axis), in terms of pile settlement and axial pile force changes based on the analysis conditions assumed in the current study. Larger pile head settlements and smaller changes in axial pile forces were computed for piles that were part of groups. It has been found that the serviceability of piles experiencing adjacent tunnelling is more affected by pile settlement than by axial pile force changes, in particular for piles inside groups. The magnitude of the tunnelling-induced excess pore pressure was small and may not substantially affect pile behaviour.     

Hanford tank residual waste – Contaminant source terms and release models

Residual waste is expected to be left in 177 underground storage tanks after closure at the US Department of Energy’s Hanford Site in Washington State, USA. In the long term, the residual wastes may represent a potential source of contamination to the subsurface environment. Residual materials that cannot be completely removed during the tank closure process are being studied to identify and characterize the solid phases and estimate the release of contaminants from these solids to water that might enter the closed tanks in the future. As of the end of 2009, residual waste from five tanks has been evaluated. Residual wastes from adjacent tanks C-202 and C-203 have high U concentrations of 24 and 59 wt.%, respectively, while residual wastes from nearby tanks C-103 and C-106 have low U concentrations of 0.4 and 0.03 wt.%, respectively. Aluminum concentrations are high (8.2–29.1 wt.%) in some tanks (C-103, C-106, and S-112) and relatively low (<1.5 wt.%) in other tanks (C-202 and C-203). Gibbsite is a common mineral in tanks with high Al concentrations, while non-crystalline U–Na–C–O–P ± H phases are common in the U-rich residual wastes from tanks C-202 and C-203. Iron oxides/hydroxides have been identified in all residual waste samples studied to date. Contaminant release from the residual wastes was studied by conducting batch leach tests using distilled deionized water, a Ca(OH)₂-saturated solution, or a CaCO₃-saturated water. Uranium release concentrations are highly dependent on waste and leachant compositions with dissolved U concentrations one or two orders of magnitude higher in the tests with high U residual wastes, and also higher when leached with the CaCO₃-saturated solution than with the Ca(OH)₂-saturated solution. Technetium leachability is not as strongly dependent on the concentration of Tc in the waste, and it appears to be slightly more leachable by the Ca(OH)₂-saturated solution than by the CaCO₃-saturated solution. In general, Tc is much less leachable (<10 wt.% of the available mass in the waste) than previously predicted. This may be due to the coprecipitation of trace concentrations of Tc in relatively insoluble phases such as Fe oxide/hydroxide solids.     

Pile and pile group response to tunnelling using a large diameter slurry shield – Case study in Shanghai

The study investigates the disturbance to piles and pile groups caused by multiple nearby drives of a large diameter slurry shield-driven tunnelling machine in Shanghai. The minimum distance between the slurry shield tunnel (with diameter D = 15.43 m) and the adjacent pile groups of Metro Line 3 and Yixian Elevated Road is 1 m. The nonlinear finite element (FE) software program ABAQUS was used to analyse the movement of the pile groups caused by the process of shield tunnelling. A field investigation was conducted before the multiple crossings to study the impact (movement and excess pore water pressure) on the surrounding soil and piles caused by the tunnelling process. The field investigation is divided into two sections: (i) free-field tunnelling, and (ii) tunnelling close to trial piles pre-installed in the section. For the full-scale test, tunnelling variable, including slurry pressure and grouting pressure, are adjusted during construction to reduce the disturbance during tunnelling close to pile groups supporting two elevated bridges. The FE simulation of the multiple crossings includes two steps: (a) shield tunnelling along the south bound tunnel approaching the working shaft at the western bank of the Huangpu River, and passing separately between two adjacent pile groups of each bridge; (b) return tunnel passing again between the pile groups of the two bridges along the north bound tunnel. Three different FE models are generated: (1) free-field tunnelling process, (2) tunnelling close to trial piles, and (3) multiple tunnel passes adjacent to pile groups supporting the two elevated bridges. Most of the relevant factors in tunnelling are taken into consideration in the FE models including (a) slurry pressure, (b) grouting pressure, (c) grouting material hardening, and (d) soil-pile interaction.     

Kinetics of selenium release in mine waste from the Meade Peak Phosphatic Shale,Phosphoria Formation,Wooley Valley,Idaho, USA

Phosphorite from the Meade Peak Phosphatic Shale member of the Permian Phosphoria Formation has been mined in southeastern Idaho since 1906. Dumps of waste rock from mining operations contain high concentrations of Se which readily leach into nearby streams and wetlands. While the most common mineralogical residence of Se in the phosphatic shale is elemental Se, Se(0), Se is also an integral component of sulfide phases (pyrite, sphalerite and vaesite–pyrite_ss) in the waste rock. It may also be present as adsorbed selenate and/or selenite, and FeSe₂ and organo-selenides.Se release from the waste rock has been observed in field and laboratory experiments. Release rates calculated from waste rock dump and column leachate solutions describe the net, overall Se release from all of the possible sources of Se listed above. In field studies, Se concentration in seepage water (pH 7.4–7.8) from the Wooley Valley Unit 4 dump ranges from 3600 µg/L in May to 10 µg/L by Sept. Surface water flow, Q, from the seep also declines over the summer, from 2 L/s in May to 0.03 L/s in Sept. Se flux ([Se] ? Q) reaches a steady-state of < 150 mg/day in 1–4 months, depending upon the volume of Q. Se release (mg/L) follows a first order reaction with a rate constant, k, = 1.35 – 6.35e?3 h^? 1 (11.8–55.6 yr^? 1).Laboratory experiments were performed with the waste shale in packed bed reactors; residence time varied from 0.09 to 400 h and outlet pH ～ 7.5. Here, Se concentration increased with increasing residence time and release was modeled with a first order reaction with k = 2.19e?3 h^? 1 (19.2 yr^? 1).Rate constants reported here fall within an order of magnitude of reported rate constants for oxidation of Se(0) formed by bacterial precipitation. This similarity among rate constants from both field and laboratory studies combined with the direct observation of Se(0) in waste shales of the Phosphoria Formation suggests that oxidation of Se(0) may control steady-state Se concentration in water draining the Wooley Valley waste dump.     

Geology and genesis of the post-collisional porphyry–skarn deposit at Bangpu,Tibet

Bangpu deposit in Tibet is a large but poorly studied Mo-rich (~ 0.089 wt.%), and Cu-poor (~ 0.32 wt.%) porphyry deposit that formed in a post-collisional tectonic setting. The deposit is located in the Gangdese porphyry copper belt (GPCB), and formed at the same time (~ 15.32 Ma) as other deposits within the belt (12 ~ 18 Ma), although it is located further to the north and has a different ore assemblage (Mo–Pb–Zn–Cu) compared to other porphyry deposits (Cu–Mo) in this belt. Two distinct mineralization events have been identified in the Bangpu deposit which are porphyry Mo–(Cu) and skarn Pb–Zn mineralization. Porphyry Mo–(Cu) mineralization in the deposit is generally associated with a mid-Miocene porphyritic monzogranite rock, whereas skarn Pb–Zn mineralization is hosted by lower Permian limestone–clastic sequences. Coprecipitated pyrite and sphalerite from the Bangpu skarn yield a Rb–Sr isochron age of 13.9 ± 0.9 Ma. In addition, the account of garnet decreases and the account of both calcite and other carbonate minerals increases with distance from the porphyritic monzogranite, suggesting that the two distinct phases of mineralization in this deposit are part of the same metallogenic event.Four main magmatic units are associated with the Bangpu deposit, namely a Paleogene biotite monzogranite, and Miocene porphyritic monzogranite, diabase, and fine-grained diorite units. These units have zircon U–Pb ages of 62.24 ± 0.32, 14.63 ± 0.25, 14.46 ± 0.38, and 13.24 ± 0.04 Ma, respectively. Zircons from porphyritic monzogranite yield ε_Hf(t) values of 2.2–8.7, with an average of 5.4, whereas the associated diabase has a similar ε_Hf(t) value averaging at 4.7. The geochemistry of the Miocene intrusions at Bangpu suggests that they were derived from different sources. The porphyritic monzogranite has relatively higher heavy rare earth element (HREE) concentrations than do other ore-bearing porphyries in the GPCB and plots closer to the amphibolite lithofacies field in Y–Zr/Sm and Y–Sm/Yb diagrams. The Bangpu diabase contains high contents of MgO (> 7.92 wt.%), FeOt (> 8.03 wt.%) but low K₂O (< 0.22 wt.%) contents and with little fractionation of the rare earth elements (REEs), yielding shallow slopes on chondrite-normalized variation diagrams. These data indicate that the mineralized porphyritic monzogranite was generated by partial melting of a thickened ancient lower crust with some mantle components, whereas the diabase intrusion was directly derived from melting of upwelling asthenospheric mantle. An ancient lower crustal source for ore-forming porphyritic monzogranite explains why the Bangpu deposit is Mo-rich and Cu-poor rather than the Cu–Mo association in other porphyry deposits in the GPCB because Mo is dominantly from the ancient crust.The Bangpu deposit has alteration zonation, ranging from an inner zone of biotite alteration through silicified and phyllic alteration zones to an outer propylitic alteration zone, similar to typical porphyry deposits. Some distinct differences are also present, for example, K-feldspar alteration at Bangpu is so dispersed that a distinct zone of K-feldspar alteration has not been identified. Hypogene mineralization at Bangpu is characterized by the early-stage precipitation of chalcopyrite during biotite alteration and the late-stage deposition of molybdenite during silicification. Fluid inclusion microthermometry indicates a change in ore-forming fluids from high-temperature (320 °C–550 °C) and high-salinity (17 wt.%–67.2 wt.%) fluids to low-temperature (213 °C–450 °C) and low-salinity (7.3 wt.%–11.6 wt.%) fluids. The deposit has lower δD_V-SMOW (− 107.1‰ to − 185.8‰) values compared with other porphyry deposits in the GPCB, suggesting that the Bangpu deposit formed in a shallower setting and is associated with a more open system than is the case for other deposits in this belt. Sulfides at Bangpu yield δ³⁴S_V-CDT values of − 2.3‰ to 0.3‰, indicative of mantle-derived S implying that coeval mantle-derived mafic magma (e.g., diabase) simultaneously supplied S and Cu to the porphyry system at Bangpu. In comparison, the Pb isotopic compositions (²⁰⁶Pb/²⁰⁴Pb = 18.79–19.28, ²⁰⁷Pb/²⁰⁴Pb = 15.64–15.93, ²⁰⁸Pb/²⁰⁴Pb = 39.16–40.45) of sulfides show that other metals (e.g., Mo, Pb, Zn) were likely derived mainly from an ancient crustal source. Therefore, the formation of the Bangpu deposit can be explained by a two-stage model involving (1) the partial melting of an ancient lower crust triggered by invasion of asthenospheric mantle-derived mafic melts that provide heat and metal Cu and (2) the formation of the Bangpu porphyry Mo–Cu system, formed by magmatic differentiation in the overriding crust in a post-collisional setting.     

La Josefina Au–Ag deposit (Patagonia,Argentina): A Jurassic epithermal deposit formed in a hot spring environment

The La Josefina Jurassic epithermal Au–Ag deposit located in Patagonia, Argentina, developed in an extensional setting of a back-arc environment, associated with a widespread Middle–Late Jurassic calc-alkaline volcanism. Block faulting has juxtaposed shallow level features evidenced by hot spring manifestations, hydrothermal eruption breccias and Au-rich veins, which suggest that mineralization in these veins, could extend far below the depths already tested by core drilling. Veins are filled by quartz, chalcedony, opal and minor adularia and barite with massive, comb, cockade, colloform–crustiform bandings and lattice-bladed textures. Ore minerals include electrum, Ag-rich sulfosalts (freibergite), pyrite, galena, sphalerite, chalcopyrite and specular hematite with minor arsenopyrite, marcasite, tetrahedrite and bornite. Four mineralizing stages have been identified, the first two (S1 and S2) are Au and Ag-rich, with temperatures ranging from 225 to 290° and salinities from up to 15 wt.% in S1 decreasing to ~ 1 wt.% NaCl in S2. The third stage (S3) displays higher base metal contents at lower temperatures (~ 200 °C). Finally, the last stage (S4) is barren with temperatures lower than 100 °C. Veins are surrounded by a proximal alteration halo of quartz + pyrite ± adularia ± illite followed outwards by illite/smectite interstratified clays and smectites (with less chlorite) to a propylitic zone. Stable isotope values calculated for the fluids show a mostly meteoric origin for mineralization fluids. Such distinct features place the La Josefina deposits in a hot spring environment with evidences of being formed at a proximal position of the Jurassic paleosurface and paleowater level.     

Biogenic components of the Baltic Sea sediments

The contents of biogenic components in 1511 samples of the Baltic Sea sediments (depth range 0–5 cm) are studied, and maps of their distribution are compiled. The sediments contain < 13.03% C_org, < 1.33% N, < 9.0% SiO_2am, < 5.0% CaCO₃, and < 1.45% P. The maximum and elevated contents of components are found in the mud of the sea deeps. The more fraction < 0.01 mm the sediments contain, the higher are the contents of components. Four facies types of carbonaceous mud, precursors of shales, have been recognized: (1) shallow-water (lagoon) lime sapropel, (2) carbonaceous mud of the shallow-water Gulf of Finland, (3) carbonaceous mud of the middle-depth Baltic Sea, and (4) laminated carbonaceous metal-bearing mud. The latter type of mud is strongly enriched in manganese and ore-forming trace elements, which points to its formation in the stagnant environment. In composition the Baltic Sea mud is similar to petroliferous mudstones of the Bazhenov Formation in West Siberia and to ancient black shales.     

Change in the viscosity of kimberlite and basaltic magmas during their origin and evolution (prediction)

Based on the analysis of experimental data on the viscosity of mafic to ultramafic magmatic melts with the use of our structure-chemical model for the calculation and prediction of the viscosity of magmas, we have first predicted that diamond-carryihg kimberlite magma must ascend from mantle to crust with considerable acceleration. The viscosity of kimberlite magma decreases by more than three times during its genesis, evolution, and ascent from mantle to crust despite the significant decrease in the temperature of the ascending kimberlite magma (~ 150 °C) and its partial crystallization and degassing. In the case of partial melting (< 1 wt.%) of carbonated peridotite in the mantle at depths of 250-350 km, high-viscosity (~ 35 Pas) kimberlite melts can be generated at ~ 8.5 GPa and ~ 1350 °C, the water content in the melt being up to ~ 8 wt.%, C(OH^-) = 0-2 wt.%, and C(H₂O) = 0-6 wt.%. On the other hand, during the formation of kimberlite pipes, dikes, and sills, the viscosity of near-surface kimberlite melts is much lower (~ 10 Pa s) at ~ 50 MPa and 1200 °C, the volume contents of crystals (V_cr) and the fluid phase (bubbles) (Vfl) are 35 and 5 vol.%, respectively, and the water content in magma, C(OH^-), is 0.5 wt.%. On the contrary, the viscosity of basaltic magmas increases by more than two orders of magnitude during their ascent from mantle to crust. The basaltic magmas which can be generated in the asthenosphere at depths of ~ 100 km have the minimum viscosity (up to ~ 2.3 Pas) at ~ 4.0 GPa, 1350 °C, C(OH^-) - 3 wt.%, and C(H₂O) - 5 wt.%. However, at the final stage of evolution (e.g., during volcanic eruptions), the viscosity of basaltic magma is considerably higher (600 Pa s) at ~ 10 MPa, 1180 °C, V_cr - 30 vol.%, Vf - 15 vol.%, and C(OH^-) - 0.5 wt.%.     

Genesis of the Proterozoic Mangabeira tin–indium mineralization,Central Brazil: Evidence from geology,petrology, fluid inclusion and stable isotope data

The Mangabeira deposit is the only known Brazilian tin mineralization with indium. It is hosted in the Paleo- to Mesoproterozoic Mangabeira within-plate granitic massif, which has geochemical characteristics of NYF fertile granites. The granitic massif is hosted in Archean to Paleoproterozoic metasedimentary rocks (Ticunzal formation), Paleoproterozoic peraluminous granites (Aurumina suite) and a granite–gneiss complex. The mineralized area comprises evolved Li-siderophyllite granite, topaz–albite granite, Li–F-rich mica greisens and a quartz–topaz rock, similar to topazite. Two types of greisens are recognized in the mineralized area: zinnwaldite greisen and Li-rich muscovite greisen, formed by metasomatism of topaz–albite granite and Li-siderophyllite granite, respectively. Cassiterite occurs in the quartz–topaz rock and in the greisens. Indium minerals, such as roquesite (CuInS₂), yanomamite (InAsO₄·2H₂O) and dzhalindite (In(OH₃)), and In-rich cassiterite, sphalerite, stannite group minerals and scorodite are more abundant in the quartz–topaz rock, and are also recognized in albitized biotite granite and in Li-rich muscovite greisen. The host rocks and mineralized zones were subsequently overprinted by the Brasiliano orogenic event.Primary widespread two-phase aqueous and rare coeval aqueous-carbonic fluid inclusions are preserved in quartz from the topaz–albite granite, in quartz and topaz from the quartz–topaz rock and in cassiterite from the Li-rich muscovite greisen. Eutectic temperatures are − 25 °C to − 23 °C, allowing modeling of the aqueous fluids in the system H₂O–NaCl(–KCl). Rare three-phase H₂O–NaCl fluid inclusions (45–50 wt.% NaCl equiv.) are restricted to the topaz–albite granite. Salinities and homogenization temperatures of the aqueous and aqueous-carbonic fluid inclusions decrease from the topaz–albite granite (15–20 wt.% NaCl equiv.; 400 °C–450 °C), to the quartz–topaz rock (10–15 wt.% NaCl equiv.; 250 °C–350 °C) and to the greisen (0–5 wt.% NaCl equiv.; 200 °C–250 °C). Secondary fluid inclusions have the same range of salinities as the primary fluid inclusions, and homogenize between 150 and 210 °C.The estimated equilibrium temperatures based on δ¹⁸O of quartz–mica pairs are 610–680 °C for the topaz–albite granite and 285–370 °C for the Li-rich muscovite greisens. These data are coherent with measured fluid inclusion homogenization temperatures. Temperatures estimated using arsenopyrite geothermometry yield crystallization temperatures of 490–530 °C for the quartz–topaz rock and 415–505 °C for the zinnwaldite greisens. The fluids in equilibrium with the topaz–albite granite have calculated δ¹⁸O and δD values of 5.6–7.5‰ and − 67 to − 58‰, respectively. Estimated δ¹⁸O and δD values are mainly 4.8–7.9‰ and − 60 to − 30‰, respectively, for the fluids in equilibrium with the quartz–topaz rock and zinnwaldite greisen; and 3.4–3.9‰ and − 25 to − 17‰, respectively, for the Li-rich muscovite greisen fluid. δ³⁴S data on arsenopyrite from the quartz–topaz rock vary from − 1.74 to − 0.74‰, consistent with a magmatic origin for the sulfur. The integration of fluid inclusion with oxygen isotopic data allows for estimation of the minimum crystallization pressure at ca. 770 bar for the host topaz–albite granite, which is consistent with its evolved signature.Based on petrological, fluid inclusion and isotope data it is proposed that the greisens and related Mangabeira Sn–In mineralization had a similar hydrothermal genesis, which involved exsolution of F-rich, Sn–In-bearing magmatic fluids from the topaz–albite granite, early formation of the quartz–topaz rock and zinnwaldite greisen, progressive cooling and Li-rich muscovite greisen formation due to interaction with meteoric water. The quartz–topaz rock is considered to have formed in the magmatic-hydrothermal transition. The mineralizing saline and CO₂-bearing fluids are interpreted to be of magmatic origin, based on the isotopic data and paragenesis, which has been documented as characteristic of the tin mineralization genetically related to Proterozoic within-plate granitic magmatism in the Goias Tin Province, Central Brazil.