Fundamentals of gas containment in unlined rock caverns

Summary The criterion for gas entry into idealized, uniform, water-saturated fractures is examined for one-dimensional, elliptical and cylindrical cavern geometries. This limiting criterion is a necessary condition for gas escape from unlined, pressurized gas storages.Gas-entry phenomena were examined using 450-mm-square glass plates in Hele Shaw configuration with nominal apertures of 50 and 200 microns. The fundamentals of gas entry, as first enunciated by Åberg (1977), are confirmed by the experiments; the critical gas-entry pressure is a function of hydraulic flow regime, fracture-mouth capillarity and fracture-mouth profile. Finite-element simulation based on these fundamentals is substantiated by experimental observations and is therefore used for generalization.The gas-entry criterion is found to be sensitive to details of fracture geometry such as aperture, connection of the fractures to the groundwater regime, geometry of the intersection of the fractures with the gas cavern, disturbance of the rock mass as a consequence of excavation, and cavern shape. In view of the inability to predict or even measure these factors, an alternative criterion which ensures no gas migration away from the gas storage is proposed. This simple criterion insists that gas escape will not occur as long as the water pressure along all possible escape paths increases for some small distance in the direction of potential gas escape. This criterion is a generalization of Åberg's (1977) proposal that the vertical hydraulic gradient should exceed one for no gas escape.     

Ore lead isotope variations in the Proterozoic massive pyrite deposit at Näsliden,Skellefte district,Sweden

The ore lead isotope ratios from the Näsliden deposit are mantle-like in terms of the idealized plumbotectonics model. They suggest that the Näsliden deposit was formed in the Proterozoic equivalent of a Phanerozoic primitive island arc. The mantle-like nature of the lead and its homogeneity indicate that crustal lead does not constitute the major component of the ore lead. Crustal lead contamination is suggested to be the cause of divergences in ore lead isotopic compositions associated with later veins in the deposit.     

The millennium eruption of Hekla in February 2000

The 18th historic eruption of Hekla started on 26 February, 2000. It was a short-lived but intense event, emitting basaltic andesitic (55.5 wt% SiO₂) pyroclastic fragments and lava. During the course of the eruption, monitoring was done by both instruments and direct observations, together providing unique insight into the current activity of Hekla. During the 12-day eruption, a total of 0.189 km³ DRE of magma was emitted. The eruptive fissure split into five segments. The segments at the highest altitude were active during the first hours, while the segments at lower altitude continued throughout the eruption. The eruption started in a highly explosive manner giving rise to a Subplinian eruptive column and consequent basaltic pyroclastic flows fed by column collapses. After the explosive phase reached its maximum, the eruption went through three more phases, namely fire-fountaining, Strombolian bursts and lava effusion. In this paper, we describe the course of events of the eruption of Hekla and the origin of its magma, and then show that the discharge rate can be linked to different style of eruptive activity, which are controlled by fissure geometry. We also show that the eruption phases observed at Hekla can be linked with inferred magma chamber overpressure prior to the eruption.     

Realistic Solar Convection Simulations

We report on realistic simulations of solar surface convection that are essentially parameter-free, but include detailed physics in the equation of state and radiative energy exchange. The simulation results are compared quantitatively with observations. Excellent agreement is obtained for the distribution of the emergent continuum intensity, the profiles of weak photospheric lines, the p-mode frequencies, the asymmetrical shape of the mode velocity and intensity spectra, the p-mode excitation rate, and the depth of the convection zone. We describe how solar convection is non-local. It is driven from a thin surface thermal boundary layer where radiative cooling produces low entropy gas which forms the cores of the downdrafts in which most of the buoyancy work occurs. Turbulence and vorticity are mostly confined to the intergranular lanes and underlying downdrafts. Finally, we present some preliminary results on magneto-convection.     

Abundance, Sources and Speciation of Trace Elements in Humus-Rich Streams Affected by Acid Sulphate Soils

The behaviour of trace elements (Al, As, Cd, Co, Cr,Cu, Fe, Mn, Ni, V, Zn) was studied in five humus-richstreams (dissolved organic carbon = 14–40 mg/L)impacted by acid sulphate soils developed in marinesulphide-bearing fine-grained sediments. During heavyrainfalls in autumn, on which the study focusses, themetals Al, Cd, Co, Cu, Mn, Ni and Zn are extensivelyleached from these acidic soils (pH = 2.5–4.5), whileAs, Cr, Fe and V are not leached more strongly fromthis soil type than from areas of till and peat. Aspeciation experiment, based on anion and cationexchange of the stream waters in the field, showedthat (1) the metals Al, Cd, Co, Mn, Ni and Zn aretransported in the streams mainly as inorganiccations, (2) Cu exists mainly in cationic form but isalso to a significant extent associated with dissolvedhumic substances, (3) Fe occurs mainly in the anionicfraction explained by organic coating on colloidal Feoxyhydoxides and (4) the hydrochemistry of As, Cr andV is complex as these elements may exist in severalunquantified anionic fractions and to a minor extentin cationic species/forms. Whereas the proportion ofacid sulphate soils in the catchments had a largeimpact on concentrations levels of several elements inthe stream waters, these soils did not have a largeaffect on the speciation of elements in water.     

Dating mafic–ultramafic intrusions by ion-microprobing contact-melt zircon: examples from SW Sweden

Zircons from anatectic melts of the country rocks of three Proterozoic mafic–ultramafic intrusions from the Sveconorwegian Province in SW Sweden were microanalyzed for U–Th–Pb and rare earth elements. Melting and interaction of the wall rocks with the intrusions gave rise to new magmas that crystallized zircon as new grains and overgrowths on xenocrysts. The ages of the intrusions can be determined by dating this newly crystallized zircon. The method is applied to three intrusions that present different degrees of complexity, related to age differences between intrusion and country rocks, and the effects of post-intrusive metamorphism. By careful study of cathodoluminescent images and selection of ion probe spots in zircon grains, we show that this approach is a powerful tool for obtaining accurate and precise ages. In the contact melts around the 916?±?11?Ma Hakefjorden Complex, Pb-loss occurred in some U-rich parts of xenocrystic zircon due to the heat from the intrusion. In back-veins of the 1624?±?6?Ma Olstorp intrusion we succeeded in geochemically distinguishing new magmatic from xenocrystic zircon despite small age differences. At Borås the mafic intrusion mixed with country rock granite to form a tonalite in which new zircon grew at 1674?±?8?Ma. Reworking of zircon occurred during 930+33/–34?Ma upper amphibolite facies Sveconorwegian metamorphism. Pb-loss was the result of re-equilibration with metamorphic fluids. REE-profiles show consistent differences between xenocrystic, magmatic, and metamorphic zircon in all cases. They typically differ in Lu/La_N, Ce/Ce*, and Eu/Eu*, and igneous zircon with marked positive Ce/Ce* and negative Eu/Eu* lost its anomalies during metamorphism.     

The western border of the Archaean province of the Baltic shield: evidence from northern Sweden

Restricted areas of acid to intermediate gneisses in northernmost Sweden are known to be Archaean and several other gneissic granodioritic to tonalitic massifs have been suggested to be of this age. To delimit the western border of the Archaean province of the Baltic Shield, and to characterize the Swedish Archaean, we have studied the 2.83 Ga old Soppero gneiss and 8 of the proposed Archaean massifs in northern Sweden by Sm-Nd isotopic analyses, U-Pb zircon dating (in two cases), and geochemistry.Of the Archaean gneisses proposed thus far, only those from the Soppero and Kukkola areas have Archaean ages (2.83 and 2.67 Ga, respectively). These gneisses have geochemical characteristics such as LREE-enrichment, low U-contents, and low K/Na-ratios, which are typical of Archaean TTG (tonalite-trondhjemite-granodiorite) rocks worldwide. The REE results indicate that the Soppero and the Kukkola gneisses were generated by partial melting of basic rocks, presumably amphibolites. According to the Sm-Nd isotopic data for the Soppero gneiss, this process occurred only a short time after initial segregation from the mantle ( _Nd values between +0.9 and +3.5). The Kukkola gneiss, however, has longer crustal residence time as indicated by negative _Nd values (–1.0 and –2.5) at the U-Pb zircon age of 2.67 Ga. Evidence in support of an important event at that time is given by secondary overgrowth on zircon cores in the 2.83 Ga old Soppero gneiss.The Sm-Nd isotope results show that the Proterozoic granitoids in northern Sweden have a decreasing involvement of Archaean source material with increasing distance from the Archaean areas. Before the Proterozoic reworking, the rather small Archaean areas in northern Sweden presumably belonged to a large continuous Archaean craton. The western palaeoboundary of this craton in Sweden probably follows approximately a line extrapolated from the western limit of the Archaean of central Finland, passing west of Kukkola, and then continues to the Lofoten area in northern Norway.     

Agricolaite, a new mineral of uranium from J��chymov, Czech Republic

The new mineral agricolaite, a potassium uranyl carbonate with ideal formula K₄(UO₂)(CO₃)₃, occurs in vugs of ankerite gangue in gneisses in the abandoned Giftkiesstollen adit at Jáchymov, Czech Republic. The name is after Georgius Agricola (1494?C1555), German scholar and scientist. Agricolaite occurs as isolated equant irregular translucent grains to 0.3?mm with yellow color, pale yellow streak, and vitreous luster. It is brittle with uneven fracture and displays neither cleavage nor parting. Agricolaite is non-fluorescent. Mohs hardness is ~4. It is associated with aragonite, brochantite, posnjakite, malachite, rutherfordine, and ??pseudo-voglite??. Experimental density is higher than 3.3?g.cm^?3, Dcalc is 3.531?g.?cm^?3. The mineral is monoclinic, space group C2/c, with a 10.2380(2), b 9.1930(2), c 12.2110(3) ?, ?? 95.108(2)°, V 1144.71(4) ?³, Z?=?4. The strongest lines in the powder X-ray diffraction pattern are d(I)(hkl): 6.061(55)(002), 5.087(57)(200), 3.740(100)(202), 3.393(43)(113), 2.281(52)(402). Average composition based on ten electron microprobe analyses corresponds to (in wt.%) UO₃ 48.53, K₂O 31.49, CO₂(calc) 22.04 which gives the empirical formula K_3.98(UO₂)_1.01(CO₃)_3.00. The crystal structure was solved from single-crystal X-ray diffraction data and refined to R ₁?=?0.0184 on the basis of the 1,308 unique reflections with F _o?>?4??F _o. The structure of agricolaite is identical to that of synthetic K₄(UO₂)(CO₃)₃ and consists of separate UO₂(CO₃)₃ groups organized into layers parallel to (100) and two crystallographically non-equivalent sites occupied by K⁺ cations. Both the mineral and its name were approved by the IMA-CNMNC.     

The use of the Pareto distribution for fracture transmissivity assessment

For many applications data on the transmissivity distribution of individual fractures are necessary, i.e., discrete fracture network (DFN) modelling of groundwater flow and transport of solutes in fractured rock and design and performance of rock grouting. Using borehole data from the Äspö Hard Rock Laboratory, Sweden, it is shown that evaluated fracture transmissivities from three boreholes at Äspö can be well described by a Pareto or power-law distribution. Evaluated distribution parameters for the three boreholes are similar, which indicates that the Pareto distribution is a robust tool to assess three fracture transmissivity distributions. Using the evaluated distribution parameters random simulations of the original interval test data show that the approximate lognormal distributions of these are reproduced. This strengthens the credibility of the approach. It is shown how the distribution parameters can be assessed from incomplete data using the properties of the distribution. Finally, Pareto distribution transmissivities also imply Pareto distribution apertures of the fractures.

---

Resumen Los datos sobre la distribución de la transmisividad en fracturas individuales, son necesarios para muchas aplicaciones, Ej. Modelación de redes de fracturamiento discreto (RFD) para flujo de agua subterránea y transporte de solutos en roca fracturada y para el diseño y ejecución de operaciones de cementación en roca. Mediante el uso del Laboratorio de Rocas Duras de Äspö en Suecia, se ha demostrado que las transmisividades por fractura evaluadas en tres perforaciones en Äspö pueden ser descritas bien por una distribución Pareto o Potencial. Los parámetros de distribución evaluados para las tres perforaciones son similares, lo cual indica que la distribución Pareto es una herramienta adecuada para evaluar tres distribuciones de transmisividad por fractura. Mediante el uso de simulaciones al azar, de los parámetros de distribución evaluados, a partir de los datos del intervalo de la prueba original, se muestra que las distribuciones lognormal así aproximadas de aquellos, pueden ser reproducidas. Esto refuerza la credibilidad del método. Se demuestra como los parámetros de distribución pueden ser evaluados a partir de datos incompletos, usando las propiedades de la distribución. Finalmente, las transmisividades con distribución Pareto, también implican aberturas de las fracturas con distribución Pareto.

---

Résumé Pour beaucoup dapplications sont nécessaires des données sur la distribution des transmissivités des fractures individuelles; des exemples sont la modélisation de lécoulement et du transport dans les roches fracturés ainsi que lanalyse de performance de la cimentation des roches. Les données des forages analysées dans le Laboratoire des roches dures dÄsprö-Suède ont montré que la distribution des transmissivités des fractures estimée pour trois forages dÄsprö est en accord avec les lois de Pareto ou de puissance. paramètres pour trois forages ont des valeurs semblables ce quindique que la loi de Pareto peut bien évaluer la distribution des transmissivités des fractures. Les simulations aléatoires réalisés sur lintervalle original des données qui ont utilisé les paramètres estimés des distributions ont bien reproduit la distribution log-normale approximée. Ce résultat a consolidé la confiance dans lapproche faite. On a montré que les paramètres de la distribution peuvent être estimés à partir de données incomplètes si on utilise les propriétés de la distribution. Finalement, la distribution de Pareto pour les transmissivités implique la même loi pour la distribution de lépaisseur des fractures.

     

Characterisation of rock aggregate breakage properties using realistic texture‐based modelling

Realistic texture‐based modelling methods, that is microstructural modelling and micromechanical modelling, are developed to simulate the rock aggregate breakage properties on the basis of the rock actual microstructure obtained using microscopic observations and image analysis. The breakage properties of three types of rocks, that is Avja, LEP and Vandle taken from three quarries in Sweden, in single aggregate breakage tests and in inter‐aggregate breakage tests are then modelled using the proposed methods. The microstructural modelling directly integrates the microscopic observation, image analysis and numerical simulation together and provides a valuable tool to investigate the mechanical properties of rock aggregates on the basis of their microstructure properties. The micromechanical modelling takes the most important microstructure properties of rock aggregates into consideration and can model the major mechanical properties. Throughout this study, it is concluded that in general, the microstructure properties of rock aggregate work together to affect their mechanical properties, and it is difficult to correlate a single microstructure property with the mechanical properties of rock aggregates. In particular, for the three types of rock Avja, LEP and Vandle in this study, crack size distribution, grain size and grain perimeter (i.e. grain shape and spatial arrangement) show good correlations with the mechanical properties. The crack length and the grain size negatively affect the mechanical properties of Avja, LEP and Vandle, but the perimeter positively influences the mechanical properties. Besides, the modelled rock aggregate breakage properties in both single aggregate and inter‐aggregate tests reveal that the aggregate microstructure, aggregate shape and loading conditions influence the breakage process of rock aggregate in service. For the rock aggregate with the same microstructure, the quadratic shape and good packing dramatically improve its mechanical properties. During services, the aggregate is easiest to be fragmented under point‐to‐point loading condition, and then in the sequence of multiple‐point, point‐to‐plane and plane‐to‐plane loading conditions. Copyright © 2011 John Wiley & Sons, Ltd.