Search Documents

By M. Yavuz

As a branch of Operations Research, decision making is a widely used application in both science and social sciences. There are many different methods developed to solve decision making problems in the literature. In this study, Fuzzy Analytic Hierarchy Process (AHP) method is performed for the selection of reclamation project selection in an open pit mine and the optimal solution of the problem is investigated. Different reclamation project alternatives are prepared for an open pit mine, which continues to its production activities in magnesite sector in Turkey. The data obtained from Magnesite Cooperation is used in the application of the method. A decision-making process is prepared as considering the project alternatives for the purpose of selecting the most suitable reclamation project. The entire criteria that affect the decision-making process are identified and transformed into a hierarchical structure. The problem is solved by using Fuzzy AHP method developed according to this structure. It is found that Fuzzy AHP is a method that may help decision-makers in solving different decision-making problems in mining.

Aug 1, 2013

By L. Bichler

Lightweight magnesium alloys are gaining in popularity for applications in the automotive and aerospace industries, Their use, however, remains limited due to their relatively low strength and creep properties. Lost Foam Casting (LFC) is a new casting process and the strength of foam cast magnesium alloys has not been investigated in great detail thus far, In this research, magnesium alloy AZ91 E and three new alloys were cast by the LFC process to study the effect of varying aluminum and manganese content on the alloy's tensile properties, The study reveals, that the tensile strength of the magnesium alloy is closely related to the relative amount of aluminum in the system. Formation of ß-Mg17Al12 phase was seen to significantly affect the yield strength and ductility of the alloy, Furthermore, formation of manganese intermetallic compounds was seen to effectively increase yield strength of the base alloy, Individual alloy morphologies were analyzed via optical and SEM techniques.

Jan 1, 2004

By Yoshimasa Funakawa, Kazuhiro Seto, Akio Kobayashi

A new hot rolled high strength steel consisting of ferrite matrix and nanometer¬sized precipitates was developed. The ultra fine precipitates were identified as TiMoC2, and the strengthening by the precipitates was evaluated more than 300MPa in case of TS/800MPa grade steel. Due to the strengthening effect of the ultra fine precipitates, alloy elements for solid solution hardening can be reduced. It is notable that the steel shows very high yield strength (730MPa in case of TS/800MPa) and very high local elongation leading to excellent stretch flange formability at the same time. The mechanical properties of the steel are attractive for chassis parts and structural parts of auto body. The steel has been produced commercially, and applied to the parts successfully.

Jan 1, 2004

By M. Souley, G. Armand, M. Ghoreychi, D. Seyedi, J. -B. Kazmierczak

"An anisotropic constitutive model is proposed in this paper accounting for both structural anisotropy and induced anisotropic plasticity. It is assumed that the rock is composed of a matrix and of potential planes of weakness. The matrix is assumed to be linear, transversely isotropic and the plasticity is described by a non linear yield function where the parameters are deduced from the nonlinear Hoek-Brown envelopes in pre- and post-peak, and derived from the laboratory characterization. A non-associated flow rule is used with a distinction between compression and extensional stress paths, as well as the absence of volumetric strain beyond large plastic distortion.The planes of weakness are considered as known a priori or assumed to be oriented perpendicular to the direction of the current minor principal stress. An elastic-perfectly plastic behavior according to the Mohr-Coulomb criterion is assumed in the planes of weakness; while the elastic part is considered as linear and transversely isotropic.Finally, the proposed model was implemented in FLAC3D and used to simulate triaxial compressions to provide a verification of the implementation. The applicability of the implemented model to reproduce damage (pre-peak) and/or failure developments around a circular opening is checked. The GCS drift, one of the mine-by experiments set up at the main level of the Meuse/Haute-Marne Underground Research Laboratory, is selected for this first application."

Jan 1, 2015

By J. E. Gill

"The Wasa Lake property is a block of about 1,640 acres in Beauchastel township, Quebec. Gold was discovered there first in 1936 in what is now known as the 'Southern area ( Fig. 1). Stripping and trenching were followed by 9,218 feet of diamond drilling in twenty holes. A number of good intersections were obtained, but their interpretation in terms of ore was difficult. A shaft was then sunk to 220 feet vertical depth. Thirteen hundred feet of lateral work and 543 feet of diamond drilling were completed at the 200-foot level before work was discontinued in 1938. Results up to this point had been just fair.Early in 1944, surface diamond drilling was resumed. The first hole cut material of ore grade in a strong shear zone 1,000 feet north of the old workings, and holes to the east and west showed this to be an extensive gold-bearing zone.In August 1945, a main working shaft was started. This has been sunk to 550 feet slope depth. A cross-cut at 390 feet vertical depth was driven to the ore zone and, by drifting and drilling at regular intervals, a length of 725 feet had been tested to December 31st, 1946 ( Figs. 1 and 2). The results check those from surface drilling quite closely.In the main zone, scoping widths from 8 to 40 feet and averaging about 25 feet are indicated. The grade after dilution should be about 0.15 oz. gold a ton and operation will probably start at 1,000 tons a day."

Jan 1, 1954

By E. M. Hillary, D. F. Sangster

"Abstract - As a contribution to the World Minerals Geoscience Database Project, an index-level database comprising information on 142 SEDEX Pb-Zn deposits and occurrences was constructed in Microsoft Access 95 format. Although consisting of only basic information, the compilation, with full referencing of all data, is possibly the largest non-corporate SEDEX Pb-Zn database in the world. It thus provides an opportunity to make geoscience queries of SEDEX Pb-Zn deposits on a scale not previously available. This report provides a summary of the results of a few such queries both as a contribution to a better understanding of SEDEX lead-zinc deposits and as a demonstration of what even such a simple database can offer.During the compilation process, it became apparent that SEDEX Pb-Zn deposits could be assigned to one of the following three sub-types: Type 1 (53.5% of total) in which sulfide minerals are the only exhalative components; Type 2 (33.1%) in which the exhalative components are sulfides plus chert, siderite/ankerite, and/or barite; Type 3 (13.4%) which contains any components of Types 1 and 2 plus abundant iron-oxide minerals (commonly magnetite) and/or is associated with iron oxide-rich sedimentary rocks.Highlights of systematic queries of the database reveal that:1. a majority (84%) of the world’s SEDEX Pb-Zn deposits are not underlain by a discordant footwall feeder zone or vent complex;2. 63.6% of deposits occur in Proterozoic strata followed by a second peak in the Devonian (16.8%). Of those in the Proterozoic, 83% of Type 3 deposits occur in the Gondwanaland supercontinent, whereas only 42% of Type 2 and 39% of Type 1 are located in this large paleo-landmass;3. median (Pb+Zn) grades in Types 1 and 2 fall between 9% and 10% with a majority clustering within this range. Type 3 grades, however, range from ~1% to ~20% (Pb+Zn) with a median of 5%; 4. median size of Type 3 deposits is ~33 Mt, whereas Type 1 and 2 median sizes are ~7 Mt and ~8 Mt, respectively.The report concludes with a brief discussion of how an index-level database such as this can be used in conjunction with other, similar databases to reveal interesting and instructive relationships for guiding both research and exploration."

Jan 1, 1998

By N. J. Contini, R. W. Knopp

"In March 1988, old Denver flotation cells were removed from Dickenson's Arthur w. White Mill, and a new flotation circuit, consisting of four Prochem rougher cells and a Hydrochem agitated cleaner column, was commissioned. The refractory portion of the gold in the ore is associated with arsenopyrite, therefore the flotation circuit is set up to produce a saleable sulphide concentrate. Tailings from the leach circuit are conditioned prior to being pumped as feed to the flotation circuit. The prototype Hydrochem agitated column has been in operation as a single cleaner stage for rougher concentrate. The effect of flotation circuit changes on gold recovery data, as well as cleaner concentrate production, and development work to achieve the target results, have been investigated. Flowsheet modifications and the resulting circuit performance data, along with test results to date, will be presented.INTRODUCTIONThe mill is capable of treating 1000 stpd of ore averaging approximately 0.30 oz/ton Au. The ore is crushed and ground to 75% - 200 mesh. Gravity concentration, cyanidation, and Merrill - Crowe precipitation yields 84% gold recovery. Detailed descriptions of the comminution, gravity separation, and leaching circuits, as well as background information have been presented elsewhere (Suttill, Van Weert et al). The target cleaner concentrate grade is 1.0 oz/ton Au for custom smelting and refining. Historical gold recovery values of the old circuit was approximately 20% of the gold present in the float feed at the target grade. The recovery value objective set for the new circuit during the planning stages was 50% of the gold present in the flotation feed at 1.0 oz/ton Au in cleaner concentrate."

Jan 1, 1989

By Shuan-cheng Gu

Roof bed separation often occurred due to the deformation incoordination of surrounding rock following the excavation of mine roadway. Bed separation has effect on the bolts, and the estimation of the influence will directly affect the safety of support systems. In this paper, the hyperbolic function model to analyze the load transfer mechanism on the bolts was simplified as the exponential function form for easy calculation. An elastic-plastic analysis model of the bolt loading occurred by the bed separation was established. And the development low that the bed separation affected to the bolt loading was discussed. According to the parameters analysis of different separation values and locations, the influence law of the bolt loading caused by roof bed separation was summarized. Finally, the effects of bed separation were considered for the bolting design, it was able to provide a reference to the anchorage design of stratified rock.

Aug 1, 2013

By Norman Orr, J. W. Gregg, Jack Wahl

The Pembina Cardium pool is a low permeability, stratified, solution-gas drive reservoir. Water-flooding has proven to be an effective means of pressure maintenance for all areas of the pool in which an economic rate of water injection can be maintained. The majority of water injection wells in the pool are converted oil well~ with capacities greater than 100 millidarcy-feet. If the average injection rate per unit of capacity can be in-creased the lower capacity areas of the pool can be economically flooded. Theory, laboratory work, and two . early miscible-flooding projects with-in the pool suggested that average water injection rates for converted wells could be increased two to three-fold by the initial injection of natural gas liquids and natural gas. How-ever, these two early projects received volumes of natural gas liquid!:' and high-pressure natural gas which would be economically prohibitive for routine stimulation. Area D, one of the Five-

Jan 1, 1962

By G. Knight

"Results of a program of dust research carried out at the Mines Branch Laboratory, Elliot Lake, indicate that dust-concentration estimates vary greatly, depending on the sampling instrument, the assessment technique and the properties of the dust cloud. The wide range of the ratio between the estimates by any pair of instruments shows that the correlation between the health hazard and dust-concentration estimates by most existing sampling instruments must be questionable. Important studies are being carried out to permit the definition of mine environments in more precise technical terms. The influence of size selection, the physical property measured, composition and sampling period on the assessment of the health hazard are discussed."

Jan 1, 1969

By Gwenaële Petit, Ulan Massimkhanov, Gilles Joubert

For its Zuuvch Ovoo uranium deposit in Mongolia, Badrakh Energy (partly owned by Orano Mining) is working on the prefeasibility study for exploitation by ISR (in situ recovery). As part of this project, Orano built a 2-cell (each cell is constituted of one producer well located in the center and surrounded by six injection wells in an hexagonal pattern) and a processing pilot plant that will be used to demonstrate the operability and economic viability of an ISR process. Another main objective of this pilot trial is to demonstrate the ability of Orano’s modeling tools to predict the ore behavior under given leaching conditions. To model uranium production by ISR, Orano uses HYTEC, a reactive transport computational software, which was adapted for its existing operation in Kazakhstan (Katco) to predict uranium recovery and reagent usage based on operating conditions. In order to perform similar simulations for this new deposit, Orano engaged in a comprehensive program including ore sampling, detailed characterization, and metallurgical testwork. The mineralogical conclusions were used to prepare a simplified geochemistry for the ore, in particular carbonate and reactive clay grades, uranium mineralization and penalizing elements such as pyrite. Then, bottle roll leach tests were used to characterize dissolution kinetics for smectite clays. Eventually, column leach test results such as pH, free acid, and uranium recovery were modeled on HYTEC to improve the fit of some of the dissolution kinetic parameters. Parameters fitted using metallurgical test work results will now be used as inputs for predictive model at the pilot scale, in order to anticipate operational results and help with operating decisions for production ramp up.

Jan 1, 2020

By Walter K. (Cory) Conrad

Two pumping tests in the Chilson Member of the lower Lakota Formation were completed at nearly the same location in 1979 and 2008. The 1979 test consisted of a pumping well and several surrounding observation wells penetrating the full 55-m thickness of the aquifer. The 2008 test consisted of a pumping well and three surrounding observation wells screened over 10 m in one of the known uranium ore horizons in the lower portion of the Lakota penetrated only about 3 meters of the more than 55-m thick lower Lakota aquifer. The average transmissivity and storativity values determined from the 1979 tests were 17.6 m2/day and 1.8 x 10-4, respectively. Comparing median transmissivity of 13.9 m2/day and storativity of 1.2 x 10-4 determined in the 2008 test to the 1979 test, the new aquifer parameters for the lower Lakota are respectively about 80 and 70 percent of the 1979 results. Because transmissivity and storativity depend on aquifer thickness, comparing the results suggests that there may be some scaling effect between the tests due to the differing lengths of screened intervals. However, the effect cannot be predicted by calculations such as relative screen lengths or open intervals to the formation, or sandstone versus shale thicknesses on geophysical logs. The scaling effect is attributed to partial disconnection between the lower and upper portions of the Lakota Formation aquifer by locally continuous shale interbeds. Suggestions to resolve this scaling issue are presented.

Aug 1, 2010

By B. Friedrich, N. Anastasijevic, S. Stopic

The extraction of nickel from Serbian and Western Australian lateritic ore by sulphuric pressure acid leaching (PAL) has been studied. The following parameters were tested: stirring rate, acid to ore ratio, reaction time, pressure and temperature. These studies were performed with ore samples from the Rudjinci, Serbia and Silcrete, Australia, both belonging to the laterite-silicate group of deposits. An increase of the reaction temperature and the stirring speed at a fixed acid/ore ratio showed a positive influence on nickel extraction. The decrease of the oxidation potential with time correlated with the increase in Fe(II) in solution and nickel extraction as the reaction went to completion for both ores. The target values of 95 % Ni extraction and 60 % Co extraction were attained in 5 min at 250°C for the Serbian silicate ore, at an agitation rate of 1500 rpm, 10 % solids, and acid/ore ratio of 0.40. Under similar conditions only 60 % Ni, but 69 % Co could be extracted from the Australian Silcrcte ore due to different mineralogy. The nickel extraction was shown follow a diffusion-controlled reaction model.

Jan 1, 2004

By L. A. Homeniuk, R. J. McH. Clark

"The Maurice Bay deposit, the Stewart Island showing, the Fond-du-Lac deposit, the Middle Lake showing and the Nisto deposit are located on the North Rim of the Athabasca Basin. They are probably unconformity-related. These occurrences tend to be small. The Fond-du-Lac and Maurice Bay deposits are the largest and contain 450,000 kg and 650,000 kg of U10 8 respectively.Pitchblende is the main uranium mineral of all the occurrences, except at Middle Lake where autunite predominates. A II of the deposits are overlain and partially hosted by the Manitou Falls Formation of the Athabasca Group, except the Nisto deposit which is entirely within basement rocks, and the Maurice Bay deposit which is partially within the basement and overlying the Fair Point Formation of the Athabasca Group.The North Rim deposits are considered to be genetically similar to the other unconformity-type deposits found in northern Saskatchewan. It is suggested that a mechanical concentration of uranium may have occurred in heavy mineral beach sand type placers. The stability of the uranium minerals may have been enhanced by special local conditions. The fossil placers, after hydrothermal alteration and enrichment, would form the apparently ""rootless"" occurrences of uranium found totally within the sandstones above the unconformity.IntroductionAt least five deposits and showings and one train of mineralized glacial erratics of unknown provenance have been located on the North Rim of the Athabasca Basin from Maurice Bay in the west to Black Lake in the east. Distributed over a distance of 250 km, (Fig. l), they occur near the edge of the Basin, and are probably unconformity-related. The following have been identified from west to east (Fig. 2): The Maurice Bay deposit, the Stewart Island showing, the Isle Brochet train, the Fond-du-Lac deposit, the Middle Lake showing, and the Nisto deposit."

Jan 1, 1986

By J. A. Finch

For base metal sulphides, iron rejection starts in mineral processing. This review focuses on the changes in plant practice specifically to improve iron sulphide rejection by controlling contaminant ion effects and discusses selected innovations in mineral processing that can be expected to play a role in iron control. One change to control contaminant ions is size reduction (the use of stage grinding and inert stirred milling) and another is system chemistry (order of reagent addition and exploitation of the flexibility of sulphoxy species). Plant examples illustrate the successes, and indicate that there are several options available. The general innovations consider the new tools for measuring gas dispersion properties and quantifying froth characteristics from images. Examples of their application to improve concentrate quality illustrate the potential. Other innovations adapt geostatistics to model variations in ore flotation response to provide feed forward control and rigorously apply the statistical experimental methodology in plant test work. An example from one concentrator shows that significant improvements to reduce the iron content of concentrates can be achieved by applying these basic principles.

Jan 1, 2006

By E. M. De Souza, J. F. Archibald

This paper examines aspects of Canadian mine backfilling operations that may be implemented to assist in reducing solid waste production, lowering energy costs, and restricting Greenhouse Gas emissions. Many Canadian underground mines use backfill to mitigate environmental effects associated with tailings disposal. Backfill production and disposal are energy intensive processes that require an average energy cost of 37.7 MJ/tonne of backfill to complete. Mine backfills, in engineering applications, also commonly make use of cement as a stabilizing agent. Cement requires an additional energy cost ranging between 4,637 and 6,600 MJ/tonne to produce. The large energy needs associated with mine backfilling and cement production operations create significant CO2 and Greenhouse Gas emissions. Approximately 1 tonne of CO2 and Greenhouse Gas is emitted for each tonne of Normal Portland Cement produced. For single mines producing 3,000 tonnes of backfill per day, considering both backfill and cement emission source terms, total emissions of CO2 ranging between 120.36 tonnes and 169.44 tonnes/day are anticipated. In Ontario, mining operations contribute 700,000 - 840,000 tonnes/year of CO2 and Greenhouse Gas emissions to satisfy backfill cement consumption needs alone. Potential reductions of cement in backfill would significantly reduce emission of these gases and lower their adverse environmental impacts. This paper identifies alternate, equally effective and lower cost binder agent strategies for cement in backfill that may lead to cement consumption reductions. Post-consumer glass represents one such potential alternative backfill binder. Over 950,000 tonnes of glass are disposed of annually in Canada, two-thirds of which exists in the form of easily recoverable glass packaging that is not currently recycled or reused. The partial replacement of cement by waste glass may address goals for reducing non-recyclable glass waste disposal; enhancing mine back filling operations; reducing total backfill energy costs; and ultimately reducing Greenhouse Gas emissions levels currently generated by the Canadian mining industry.

Jan 1, 2004

By D. K. Sarkar, J. Perron, J. D. Brassard

"Nearly perfect spherical shaped water drops rolling-off from the surface of lotus leaf is one of the beauties of nature. This kind of surface is called superhydrophobic, since the water does not like it. Physically, a superhydrophobic surface provides a contact angle (CA) of water over 150°. The combinations of both mico-nanostructure topography as well as low surface energy are essential to have such high CA on a surface. Nano-micro structured superhydrophobic surfaces have been developed in different methods. (i) Monodispersed fluorinated silica nanoparticles have been synthetized via Stöber Sol-Gel process and deposited on various surfaces, such as aluminium or glass, to obtain transparent superhydrophobic thin films. (ii) Similarly, zinc stearate nanocomposite coating has been fabricated by spray process on the same substrates. These thin films are found mechanically robust with drag reduction of coefficient 16% lower than as-received surfaces. (iii) By electrodepositing pure zinc on steel surfaces and further coating it with diluted room-temperature vulcanized silicone rubber, the surface become superhydrophobic. This superhydrophobic surface protects the steel against corrosion by increasing the polarization resistance from 1.5 k? on steel to 44.4 k?. This coating also reduced the ice adhesion on the steel surface 6.3 times as compared to bare steel substrates. Globally, these superhydrophobic surfaces have tremendous applications against corrosion reductions, drag reductions and ice adhesion reductions.INTRODUCTIONSuperhydrophobic coatings are increasingly attractive to the industry and academia due to their unique self-cleaning properties as a result of their water repelling characteristics. Fabrication of superhydrophobic coating is an inspiration from nature as water is seen to repel on many natural surfaces such as those of the lotus leaves, butterfly wings, water striders’ legs, and so on (Gao & Jiang, 2004; J. Huang, Wang & Wang, 2006; Neinhuis & Barthlott, 1997; Zheng, Gao & Jiang, 2007). This non-wetting phenomenon is well understood on lotus leaves, shown on the Figure 1 (a), and it is considered as the basis of the studies of superhydrophobicity. The existence of this binary structure on the surface of the lotus leaves allows a large amount of air entrapment and consequently reduces the contact area of water on it and the presence of low surface energy waxy components with reduced affinity to water together leads to the superhydrophobic properties. The phenomenon is shown on Figure 1 (b). Cassie and Baxter (Cassie & Baxter, 1944) as well as Wenzel (Wenzel, 1936) proposed two mathematical models to explain the wetting phenomena on rough surfaces that leads to superhydrophobic properties. By definition, superhydrophobic surfaces are characterized by a water contact angle value = 150° with the rolling off of water drops from the surfaces (Sarkar & Saleema, 2010). Superhydrophobic surfaces find tremendous importance in the technological world where it can applied in domains such as anti-corrosion, anti-biofouling, electrowetting, icephobic and even in drag reductions (Bushnell & Moore, 1991; Cao, Jones, Sikka, Wu & Gao, 2009; Carlborg & van der Wijngaart, 2010; Carré & Mittal, 2009; Y. Huang, Sarkar & Chen, 2010a; R. Karmouch, Coude, Abel & Ross, 2009). However, the most important challenge faced by industries is their feasibility large scale in an effective an economical manner."

Jan 1, 2016

By C. M. Campbell

IN 1902, a famous Canadian geologist made the following statement : ?'From what we know row of the iron-ore fields of the Province, we may feel that Ontario is assured of a great future as an iron-ore producing country. Judging from the extent of the territory over which the newly discovered ranges have been found, Ontario should in the not far distant future take her place as one of the chief iron producers in the world". In 1905, an official of the Department of Mines said: "In ten years Canada will have b:ecome a great metallurgical country. You will see an iron industry in this Dominion larger than in any country in the world. Canada will yet furnish the world with its iron supply. It will be the same with iron as with wheat. A decade from now Canada will outstrip all other countries in wheat growing. The production of iron at a cheaper rate than it can be made elsewhere will cause Canada to take a similar position as far as iron is concerned". In 1924, a former President of the Institute, in arguing his claim for inconceivable mineral wealth in this country, stated as a "certainty" that iron-ore deposits on the scale of the ones in Minnesota would be found "again and again" in the Canadian Precambrian. In 1930, a Minister of Mines announced that "The country surrounding Churchill holds large deposits of valuable iron and coal", and he stated that, at that place, 'We will have an industry in iron that will compare with Pittsburgh". These statements are definite promises, just as definite and important as the statement "I promise to pay" on a note, and they are not fulfilled by casually stating, as is the habit, that someday we will work our low-grade iron ores. Nor are they fulfilled by stating, as is often done, "Oh, So-and-So did not mean that ' . Due to their importance and to the prominent positions held by the men who made these statements, they were widely circulated throughout Canada. They were taken at their face value, and on the strength of them and many more of the same kind in regard to our other resources we have planned and built, at huge expense, provinces, governments, civil services, railways, and other equipment. The reserves that were to justify these expenditures have not appeared. There has not been a ton of iron ore mined in Canada for many years, and important commercial reserves are unknown. We have built the mill, but adequate ore is lacking.

Jan 1, 1937

By G. R. Parslow

U, Fe, Mn, Ni, Cu, Zn and Co data for the sediment in both Key and Seahorse lakes, which overlie portions of known economic uranium deposits in the region, are presented. With the exception of U, Fe and Mn, the elemental distributions can be considered anomalous, in a statistical sense, in both lakes. The U values are of particular interest in that Key Lake is not anomalous, whereas Seahorse Lake is markedly anomalous. Tentative correlations made with background data from other surveys in an attempt to differentiate between anomalous and background values indicate that deviations from linearity in a distribution, and not absolute mean or maximum values, are indicative of anomalous samples within the distribution.

Jan 1, 1979

By G. P. Gu, B. Fang, R. W. Revie, M. Elboujdaini

Stress-corrosion cracking (SCC) tests were conducted in the near-neutral pH standard solution, NS4, and in an actual soil solution, using four-point bending at high a stress ratio and low frequency conditions very similar to those of operational pipelines. Pitting incubation appeared first and then pitting initiated and grew in both solutions although there were many more pits on the specimen tested in soil purged with 5% CO2 + 95% N2 than in the specimen tested in NS4 solution purged with the same gas. These observations show that samples in soil solution are more susceptible to pitting than those in NS4 solution. When the pit reached a critical size, the increased stress concentration around the pits, resulted in the transition to a crack.

Jan 1, 2005