Search Documents

This method determines tellurium at concentrations down to 0·01 p.p.m., with geochemical accuracy. Tellurium is volatilized from the rock at 800°C in the presence of sulphur from decomposing pyrite, condensed, dissolved, and finally determined by atomic absorption in a long tube burner. The distribution of Te is shown near some gold lodes at Kalgoorlie. Lastly, a method of synthesizing Te-free pyrite from sulphur and ferrous sulphate is described.

Jan 1, 1969

By Donald R. Williamson

Statistical studies at Cyprus Pima Mine indicate that samples from 22 blastholes are necessary for determining average copper grades, within a desired range of precision, for mine-model blocks by arithmetic averaging. The average number of holes in such blocks is 14. Many considerations show that more holes will not be drilled for sampling. A way to determine block grades more reliably than by" arithmetic averaging was sought. Thus far arithmetic averaging has proven to be the best procedure for the particular operation characteristics despite the small number of samples. It naturally follows that expenditure of great effort in sampling blasthole cuttings is unwise.

Jan 1, 1977

By F. E. Senftle

Introduction The usual laboratory methods of uranium analysis are either tedious, time-consuming chemical procedures or elaborate physical methods requiring sensitive equipment. There is thus an urgent need for a method of estimating the uranium content of ore samples by approximate analysis in the field. When a new property is being developed, it is desirable to know, with the least possible delay, the uranium content of channel samples in order to guide the diamond drilling and digging operations. Prospectors, too, require approximate analyses of their hand specimens to evaluate their discoveries. To satisfy the demand for these types of analyses, several field procedures have been tried. Two general plans can be followed. One is fast but will yield only a rough estimate. The other is slightly longer but, if done carefully, will give an assay probably within 10 per cent of the true value. For field operation it is necessary to strip the more accurate laboratory methods of analysis of all refinements because of the limited equipment and skilled techniques at hand. Most of the corrections normally applied, with the exception of that for equilibrium, will not alter the gross analysis to any appreciable extent. When uranium atoms in pitchblende disintegrate, they break down into other radioactive elements, and eventually the number of elements being formed is equal to the number breaking down. The ore is then said to be in equilibrium. Certain geological processes, however, such as leaching, may selectively remove some of these elements and hence reduce the radioactivity of the ore sample. The ore is then said to be out of equilibrium. The field analysis may then be in error by as much as 50 per cent since the apparatus is calibrated with standards that are in equilibrium. However, such a large error due to loss of equilibrium is the exception, and in most cases uncorrected field types of analyr.is are justified.

Jan 1, 1949

By K. D. Dill, S. D. Rosenthal

"This project analyzed the efficiency of incorporating the use of directional drilling technology into coal overburden blasting. Directional drilling is currently in use in the petroleum industry and it is believed that it will be a valuable asset in the mining industry. This project has shown that directional drilling can be a viable technology for use in the coal overburden removal process resulting in increased drill utilization and potential for cost savings. Future work regarding blasting and geotechnical evaluation should be performed to solidify the concept. INTRODUCTION Determine the Potential Drill Utilization Improvements and Rock Fragmentation Requirements Using Directional Drilling in a Coal Mining Overburden Highwall Application was a project funded by Atlas Copco Drilling Solutions (Atlas Copco) based in Garland, Texas. The project was developed and researched by Montana Tech of the University of Montana. This project was intended to show how the use of directional drilling can increase drill utilization, while decreasing costs, without sacrificing overburden blasting results. This paper discusses a review of the literature, the methods performed, conclusions and future work Background Atlas Copco theorized that an alternative approach to drilling a cast blast pattern would be beneficial to coal mining operations by increasing drill utilization and therefore, decreasing overall costs. Increasing drill utilization results in an increase in the amount of time the drill is being productive, more drill bit time in ground. Atlas Copco’s theory is that by incorporating directional drilling in the design of a cast blast pattern, a drill rig could drill multiple directional holes, which are the length of the cast blast, from a single drill set up. A drill goes through a series of steps to complete a blast pattern, these steps include: tramming to the hole location, leveling the drill, drilling the hole, cleaning the hole, and retracting the drill steel. This process is repeated hundreds of times until the whole pattern is completed. The most time intensive part of this process is drilling out the blast pattern. Integrating directional drilling into this process could decrease the overall time it takes to drill out a blast pattern by minimizing the amount of time spent tramming and setting up for each subsequent hole. Tramming and drill set up are two aspects of the drilling process where the drill tool is not engaged with the ground and, therefore, not being utilized. A cast blast pattern designed to utilize directional drilling will allow the drill tool to be “in the hole” longer and will improve drill utilization. This change in drill utilization was analyzed. Drilling and blasting is an integral part of the mining process, without it mining would be a slower process and less material could be mined. Drilling provides holes for explosives to be loaded and then blasted. Blasting has been found to be the most cost effective method of fracturing rock in mining for material removal. Blasting is also used as a low-cost method to move the rock in open cast mines, commonly referred to as cast blasting. A cast blast is a large blast designed to allow for the maximum amount of overburden to be removed using explosives in a single blast thereby uncovering more coal at a lower cost. Not all the material is blasted to cast-to-final, which requires that material to be moved with large mining machinery. Material that is considered cast-to-final is material that has been thrown far enough to never be moved again in the coal extraction process, therefore it is in its final resting place. A typical cast profile and cast-to-final area is shown in Figure 1."

Jan 1, 2017

By Brian T. Burch, Paul N. Worsey

When fired, submerged Linear Shaped Charges (LSCs) lose all effectiveness in the cutting of steel. Users in underwater applications have reported having to substantially increase the charge size to the point where brute explosive force shears rather than cuts the target.

Jan 1, 2015

By Azin Zangooi, Cesar Gomez, James Finch

Frother concentration is a variable which significantly affects the flotation performance by decreasing bubble size and helping froth stabilization. Variations in frother concentration, which mostly go unnoticed in industrial operations, may affect these functions. Some factors that can cause variations include the fact that frothers are normally dosed based on feed tonnage and not solution concentration, unstable frother delivery, incomplete dissolution for low solubility frothers, addition of frother-free wash water, remnant frother in recycle waters and frother partitioning to the froth. A colorimetric technique developed for determining frother concentration on site makes it possible to detect the occurrence of these problems and to establish their magnitude, allowing conditions to be adjusted to reduce frother concentration variations. By sampling and analysis, frother distribution throughout a circuit can be established, referred to as ‘frother distribution mapping’. Experiences at several concentrators processing sulphide ores are used to illustrate time restrictions for sample analysis, and to demonstrate temporal and spatial frother concentration changes as a consequence of variations in the grinding stage, recycled waters containing frother, and frother partitioning.

Jan 1, 2014

By J. R. CAIN

SAMUEL L. HOYT, ? Nela Park, Cleveland, Ohio (written discussion?). The work that Dr. Cain is doing on gases in steel should have a highly important bearing on investigational work in connection with steel-making practice. We found, for example, in our work in connection with the use of manganese in open-hearth practice that we needed expressions for the manganese efficiency and the "condition" of the steel. To get these figures we assumed that we would need to know the amount of FeO at different stages of the heat and the amounts of the gases present. This required simultaneous analyses of our samples for FeO and the gases and Dr. Cain seems to be the only investigator who has considered the analytical work from this point of view. That he has done so is very encouraging and it is to be hoped that his work will solve the mys-teries now surrounding this neglected phase of steel metallurgy. Those interested in the Ledebur method for use in connection with the efficiency of manganese as a "deoxidizing " agent will he disappointed to learn that more recent work at the Bureau indicates that Ledebur determinations are not reliable. The principal source of error seems to be due to the partial reduction of the ferrous oxide by the carbon present in the steel. However, it seems like an unnecessary criticism of the Ledebur method to state (p. 1319) that it proved of little value because it showed little differences in oxygen for a series of steels treated with ,various reagents and showing variations in physical properties and free-

Jan 12, 1919

Sampling protocols and metallurgical flow sheets should be designed to suit the style of mineralisation in question. In support, the a priori need for evaluation programs is the characterisation of the mineralisation type(s) and domain definition. The proportion and size of gold particles has a direct impact on the effectiveness of sampling protocols and metallurgical processes employed. Within the coarse- to-fi ne gold spectrum, the samplability of a deposit ranges from relatively simple for fine-grained disseminated particles, through to difficult for coarse gold mineralisation. Conversely, metallurgical processing may be relatively simple for coarse gold ore and more difficult for very fi ne gold ores. Most deposits contain a proportion of both particle size types. The nature of mineralisation, mineralogy and associated gold particle sizing can only be determined by preliminary sampling to facilitate testing. Mineralogical studies, focussing on ore mineralogy and gold particle deportment, size and distribution are required as part of wider paragenetic and geometallurgical investigations. This paper discusses key issues and presents a methodology for the determination of gold particle size in coarse gravity-gold ores to support sampling optimisation and which also contributes to metallurgical characterisation.

Aug 1, 2010

By Vivek Samu, Murthy Guddati

A new method called Effective Dispersion Analysis of Reflections (EDAR) was recently developed to effectively analyze the dispersive waves generated from lateral impact to piles. This paper presents the results of nondestructive length estimation of existing piles through EDAR. EDAR provides a new visualization method for analyzing wave reflections by transforming the response into the frequency domain thus eliminating the need to perform difficult peak picking in the time domain. In addition to being able to analyze dispersive waves, EDAR also incorporates the effects of radiation damping due to the surrounding soil leading to more accurate length estimates. Concrete filled steel tubes of different embedment depths were tested both in laboratory and field conditions. Length estimates obtained were within 10% error margins after careful analysis accounting for the presence of soil effects and effects from pile top reflection.

By Jerry Tien

Mine regulators are normally used for proper air distribution in underground mines. They are deliberately introduced resistance in the regulated airway, and by altering sizes, they can distribute specific air quantities to regulated splits and free splits (4, 5). The enactment of the Federal Coal Mine Health and Safety Act of 1969, the Federal Mine Safety and Health Amendments Act of 1977, and the regulations adopted by the Mining Enforcement and Safety Administration (now called the Mining Safety and Health ministration) prompted greater discussions of optimal mine regulator locations because of the Act's stringent ventilation requirements (2, 3). For example, by requiring the separation of the air used in the belt haulage entries from the intake and return airways, the regulations required the creation of a third type of airway---the neutral airway, as opposed to the intake and return airways, adding to the complexity of a mine's ventilation network. It is the intent of this report to show that panel regulators should be located in intake airways at the neck of a panel for a blowing fan system and in return airways at the neck of a panel for an exhausting fan system in order to achieve optimal air control. A computer simulation model is used to determine optimal mine regulator location.

Jan 1, 1983

By Tapiwa Chenje, Victor Olivas, Peter Radziszewski

"Ore breakage and grinding media wear have always been considered and determined separately in laboratory tests. However the two processes are linked with the ore broken down in any comminution device being responsible for the majority of the steel wear experienced. The Comminution Dynamics Laboratory (CDLab) successfully developed a test methodology that can more accurately predict grinding media wear in tumbling mills than the Bond Abrasion test. This paper looks at the possible application of the abrasive wear methodology to determine breakage parameters of the same ore used in the abrasion test. Results indicate that breakage parameters comparable to t10 can be determined using the Steel Wheel Abrasion Test (SWAT), while using significantly less ore than standard breakage tests.BACKGROUNDThe Rubber Wheel Abrasion Test (RWAT) is an internationally accepted and widely used method for determining the abrasion wear behavior of materials. The abrasion wheel test was developed to primarily utilize a steel wheel (Brinell, 1921). Haworth (1949) criticized using of the steel wheel by arguing that increasing the contact area size in the steel wheel test causes a change in the wear rate with time. He suggested that the use of a rubber wheel instead eliminates this problem, leading to the adoption of the RWAT which was standardized in ASTM G65 (2004).Radziszewski (2002), however, found that the use of the Steel Wheel Abrasion Test (SWAT) was more applicable for wear testing in the mining industry for two critical reasons. Firstly, the major cause of wear in a ball mill, the most frequently used comminution device, is ball-ore-ball contact and also ball-ore-liner which both are metal-ore-metal contact. Therefore, using the steel wheel has the advantage of providing a similar abrasive environment to the one that takes place in a real ball mill. Secondly, the SWAT creates the potential of running experiments at much higher force regimes that are closer to what happens in actual mills, unlike the RWAT which is a low stress test."

Jan 1, 2011

By Robert J. Timko, R. Lincoln Derick

The Bureau of Mines evaluated the integrity of escapeways in a three entry underground coal mine. A center conveyor belt entry (secondary escapeway), was flanked by an intake (primary escapeway) and a return entry. Diesel equipment was used throughout the mine for material and personnel haulage. A carbon monoxide monitoring system, tied to a data recording computer on the surface, was deployed throughout the belt entry and at selected sites in the intake and return entries. Previously, the company had a fire of unknown origin in another mine, which prompted an interest in the performance of their escapeways. An initial, qua1itative evaluation was performed by burning approximately 14 kg of coal at the mouth of the intake entry. The first test was followed by a similar burn at the mouth of the belt entry. These initial trials showed a high quantity of leakage between entries. In several follow-up evaluations, simulated fires, using a tracer gas, were set in the intake and belt entries. Tracer gas concentrations were obtained in all entries at four different underground locations. The results of these tests showed that permanent stoppings could not guarantee isolated, contaminant-free escapeways.

Jan 1, 1989

By M Zhu, J Safarian

Understanding the slag composition-structure-property relationship remains challenging due to the amorphous nature of slag melts and the difficulty of high-temperature experimental investigations. Since the CaO-Al2O3-SiO2 slag system is maybe the most important slag system for various hightemperature metallurgical process, the present research employed high-throughput molecular dynamics (MD) simulations on the entire composition region of the CaO-Al2O3-SiO2 slag system to determine physical properties like molar volume, density and diffusivity, structural properties such as distribution of different oxygen types for each simulated slag melt. The obtained slag properties result from the simulations displayed good consistency with experimental data, underscoring the reliability of the MD simulation as a novel tool to predict reasonable slag properties and to study slag behaviour from atomistic insight. The simulated results demonstrate a near-linear relationship between molar volume changes and composition, highlighting the predominant role of oxygen atoms, which is attributed to the significantly larger size of oxygen compared to other cations in the slag melts. This disparity in size leads to oxygen ions occupying most of the space in slags, thereby playing a major role in determining the overall molar volume. The simulation results of slag density revealed a decreasing trend with increased SiO2 content and identified a local minimum at constant SiO2 when Al2O3 content increases, also consistent with established slag density models. Furthermore, the simulated oxygen self-diffusion coefficients reveal a strong composition-structureproperty relationship, where an increase in CaO content enhances slag diffusivity, while a minimum in diffusivity is observed near a CaO/Al2O3 ratio of unity. The diffusivity minimum also corresponding to the local viscosity maxima, indicating effective simulation of charge compensation effects and dynamic characteristics of slag melts. The simulated distribution of oxygen types in slag — bridging oxygen (BOs), non-bridging oxygen (NBOs), free oxygen (FOs) and tricluster oxygen (TOs) — reveals a composition-dependent pattern: BOs peak near CaO/Al2O3 = 1, NBOs and FOs are abundant in CaO-rich areas with NBOs peaking near the Ca3SiO5 composition, FOs in the CaO corner, and TOs predominantly in the Al2O3 corner, highlighting the complex interplay of network formers and modifiers for the charge compensation effects on the slag’s structural properties. The successful application of MD simulations in this study paves the way for a powerful approach to explore the complex relationships in multicomponent slag systems and bridges the gap between mathematics modelling and experimental observations in high-temperature metallurgical processes.

Aug 21, 2024

Project developent proceeds by increasingly detailed studies addressing overall project returns through an analysis of the substance, risks and costs associated with the following elements: ò Reserves ò Mining ò Processing ò Infrastructure ò Environmental Impact ò Product Market Computer-based expertise has a potential to ò Capture perishable expertise, making this available for shared access and so reducing the cost of human experience ò Consolidate know-how for optimising process operations, bringing consistency of judgement and advice to decision- making by making the rules, facts and preferences explicit ò Allow assignment of the basis and authority of decisions, so experts can pass routine expertise down the line, freeing their time to generate more. ò Force a review of expertise when creating the knowledge base.

Jan 1, 1989

By M Araos, I Onederra

A novel explosive formulation that substitutes the use of ammonium nitrate (AN) with hydrogen peroxide (HP) as the main oxidising agent has been developed and tested as part of an Australian Coal Industry Research Program (ACARP) project. The main objective was to provide a step-change solution that has the potential to completely eliminate NOx fumes in blasting. The development of this HP/fuel-based explosive started with a comprehensive detonation characterisation program, consisting of unconfined velocity of detonation (VOD) measurements of samples at different diameters, densities and void sensitisation techniques. VOD results showed that the HP/fuel-based explosive mixtures behaved similarly to non-ideal commercial explosives; in that way mixtures could be tailored to achieve specific detonation performance targets. The next stage of testing in this development involved preliminary confined tests to evaluate the rock breakage potential of the HP/fuel-based explosive mixtures. Fully instrumented single-hole tests were conducted in a limestone quarry. Results from these preliminary trials confirmed the rock breakage performance of HP/fuel-based explosive mixtures. The confined VOD measured in one of the tests was 5100 m/s in a 102 mm diameter blasthole at a density of 0.96 g/ml. The corresponding borehole pressure measurement of this HP charge was 2.61 GPa with a detonation temperature of 2921°C. Based on near-field accelerations, both HP and ANFO charges displayed similar attenuation characteristics in the test conditions. Post-blast observations indicated that the breakage performance of the HP mixtures clearly reflected the in-hole detonation performance measurements obtained, confirming the ability of these explosive mixtures to both fragment and effectively displace the rock mass under evaluation. Further research work continues with the future commissioning of a prototype mixing and delivery unit, the successful completion and analysis of data from multiple-hole trials, and further characterisation work associated with product stability, reliability and performance under different geotechnical conditions.CITATION:Onederra, I and Araos, M, 2015. Detonation and breakage performance of a hydrogen peroxide-based explosive formulation, in Proceedings 11th International Symposium on Rock Fragmentation by Blasting, pp 565–574 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 24, 2015

By R. Aubertin, G. Panneton, P. Donovan

The Detour Lake gold project is located in northeastern Ontario, approximately 300 km northeast of Timmins, within the Abitibi greenstone belt in the northwestern portion of the Archean Superior Province. Historically this greenstone belt has produced approximately 180 million ounces of gold, largely from narrow vein underground mining. The Detour Lake project is a site of extensive past exploration since 1974, as well as underground and limited open pit mining activity from 1983 to 1999. During this period, it produced 1.8 million ounces of gold. Underground mining focused on narrow vein conventional shrinkage and mechanized longhole stoping. After acquiring the property in January 2007, Detour Gold Corporation initiated a comprehensive drill campaign to define the potential for a large, open pitable, low-grade gold deposit. From 2007 to present, Detour Gold has completed over 500,000 metres of diamond drilling. A technical report completed in March 2011 defined an open pit mineral reserve of 14.9 million ounces of gold, which is Canada's largest undeveloped gold reserve. The Detour Lake project is currently in development and infrastructure construction of the mine facilities started at year-end 2010, after all the necessary Provincial regulatory permits were received. Gold production is expected to commence in early 2013. Gold mineralization at Detour Lake is associated with a large scale deformation zone (Sunday Lake Deformation Zone; "SLDZ") that occurs at the contact between the 2.72 Ga Deloro Group mafic metavolcanics and the 2.69 Ga Porcupine Assemblage metasedimentary and metavolcanic rocks. The bulk of the gold mineralization is found in the hanging wall of the SLDZ, which is associated with a stockwork of auriferous quartz veins that splay from a flexure along the deformation corridor. The mineralization forms a 200-350 metres wide auriferous corridor defined over 3 kilometres along the strike of the mafic metavolcanic assemblage, from surface to at least 1,000 metres depth. The richest gold mineralized zones are commonly associated with quartz veins with a strong biotite alteration and local zones of pyrrhotite-chalcopyrite-pyrite-gold enrichment formed during the deformation of the quartz vein system.

Jan 1, 2011

By A. Griffiths

The? use of hydrogen peroxide to detoxify gold mill waste waters containing cyanide and other toxic substances is already an established technology. Recent advances have extended the range of application of hydrogen peroxide treatment to include the removal of complex iron cyanides, the simultaneous removal of cyanide and toxic metals down to very low levels and treatment of barren bleeds, CIP' pulps and tailings pond discharges. Three treatment plants are described here. Each was installed at a gold mine in North America in 1987 and each illustrates a different application of hydrogen peroxide. The need for individual design of the detoxification plant to fit particular requirements is shown by the examples chosen.

Jan 1, 1988

By Andrew Griffiths

The emphasis in the title of this paper should be placed on the world "Total", since the detoxification of cyanide in its free and weakly-complexed form with H202 has already found application in many areas, including the treatment of mining effluents. One of the most common forms of cyanide found in gold-mill effluents is hexacyanoferrate (usually ferrocyanide), which is included in measurements of total cyanide but is too stable to be oxidized by H202 or any other conventional oxidizing agent. This has led to the view that H202 cannot be used to detoxify effluents containing hexacyanoferrates, when a limit has been set on total cyanide. A process is described in this paper which does meet this requirement, and which has already been demonstated in two Canadian gold mills. Before describing the new process, the terms "free", "total" and "weakly-complexed" cyanide will be defined as they are used in this paper.

Jan 1, 1998

By R. L. Templin

The increasing use of aluminum and its alloys in commercial fields tias demanded a better understanding of their machining properties. This fact is exemplified by problems that have arisen in the automotive and airplane industries, but many in other fields might be cited. As pure alunlinum and its alloys in their various commercial conditions show appreciable differences in their machining properties, it is not surprising that quite divergent solutions have been offered for the machining problems encountered. However, if the fundamental requirements of the most suitable cutting tools for these metals are understood, these machining problems lend themselves more readily to satisfactory solutions. Since the machining of free cutting brass and mild steel is understood by most persons accustomed to working these metals, it may serve our purpose better to first make a general comparison of the tools more commonly used in machining these metals with the tools most suitable for machining aluminum, then proceed to a more specific discussion of the individual tools. Comparison oF Cutting Tools Cutting tools commonly used for machining free cutting brass usually have little, if any, top and side rake; they are ground on a medium to coarse abrasive wheel and used without any cutting compound or with a cutting compound that has a paraffin base. Those ordinarily used for steel have some top and side rake, are usually ground on a medium to fine abrasive wheel, and are often used with soluble-oil cutting compounds. The proper tools for aluminum and its alloys should have appreciably more side and top rake than the tools for cutting steel; should have very keen edges obtained by grinding with fine or very fine abrasive wheels supplemented in many cases by hand stoning with an oil-stone; and should be used with suitable cutting compounds whenever possible. In many cases, tools suitable for machining aluminum and its alloys are not appreciably different from tools commonly used for cutting hardwoods. The front clearance of a tool most suitable for machining aluminum and its alloys should be about 6°, the top rake from 30" to 50°, making the total angle of the cutting edge of the tool from 35' to 55". A side

By M. J. Voigt, R. A. Govender, A. Mainza, L. Bbosa

"Mineral texture is a critical factor which controls ore variability and is an important attribute in geometallurgy. In relation to downstream processes, it affects the fracture pattern during breakage, where rock strength is inherently a function of mineral texture. Because of the subjective nature of mineral texture, it has not been easy to quantify, especially in the context of a measurement suitable for use in geometallurgical programmes. The aim of this paper is to present the first steps in developing a 3D mineral texture quantification method for drill core and to assess its sensitivity to differences in rock strength using a case study. The methodology includes classifying the textural information using the 3D grey level co-occurrence matrices (GLCM) and X-ray computed tomography (XCT) coupled method. Rock strength tests were performed using the split Hopkinson pressure bar (SHPB). The case study investigates a heterogeneous polymetallic sulphide deposit and a homogeneous shale subdivided into three ’mineral textural types’. The variability is largely captured by the GLCM matrices, and preliminary trends can be observed where the shale is finer grained and has a higher yield strength in comparison with the coarser grained polymetallic sulphide ore. IntroductionModern mining requires the exploitation of lower grade and more complex, heterogeneous orebodies to address the growing industry demands. Historically, processing plants were designed and built based on average ore characteristics and therefore are relatively inflexible when facing the full spectrum of ore variability in heterogeneous orebodies (Lamberg, 2011; Lotter, 2011; Schouwstra et al., 2013). The practice of geometallurgy represents an opportunity to manage ore variability in the mining industry (Powell 2013; Baum, 2014; Yildirim et al., 2014; Nguyen, 2013). It focuses on quantifying the relationships between ore characteristics and their downstream mining and processing responses, often starting with small-scale laboratory testing. The purpose of geometallurgy is linked to maximizing the net present value (NPV) of an orebody by minimizing technical and operational risk – where ‘ore variability’ is considered one of the most significant technical risks."

Apr 1, 2019