Carbonate rocks contain most of the world&rsquo / s proven hydrocarbon reserves. It is essential to predict flow properties and understand flow mechanisms in carbonates for estimating hydrocarbon recovery accurately. Pore network modeling is an effective tool in determination of flow properties and investigation of flow mechanisms. Topologically equivalent pore network models yield accurate results for flow properties. Due to their simple pore structure, sandstones are generally considered in pore scale studies and studies involving carbonates are limited. In this study, in order to understand flow mechanisms and wettability effects in heterogeneous carbonate rocks, a novel pore network model was developed for simulating two-phase flow. The constructed model was composed of matrix, fissure and vug sub domains and the sequence of fluid displacements was simulated typical by primary drainage followed by water flooding. Main mechanisms of imbibition, snap-off, piston like advance and pore body filling, were also considered. All the physically possible fluid configurations in the pores, vugs and fissures for all wettability types were examined. For configurations with a fluid layer sandwiched between other phases, the range of capillary pressures for the existence of such a layer was also evaluated. Then, results of the proposed model were compared with data available in literature. Finally, effects of wettability and pore structure on flow properties were examined by assigning different wettability conditions and porosity features. It was concluded that the proposed pore network model successfully represented two phase flow in fissured and vuggy carbonate rocks.

The Germencik Omerbeyli geothermal field is considered to be one of the most important geothermal fields in Turkey. A numerical modeling study was carried out to simulate the response of the field to different production/injection scenarios. The reservoir performance evaluation was based on the numerical simulation of the reservoir behavior using the simulation code TOUGH2. The numerical simulation model includes a total area of 85.8 km2 and extends from the surface at +330 m msl (mean sea level) to a depth of -4581 m msl. Through a trial and error process, the natural state model was satisfactorily matched with the initial temperature and pressure data measured at the wells. The natural state model was further calibrated using the long term flow test (LTFT) data conducted in 2006, including OB-6 and OB-9 as flowing wells and OB-8 as an injection well. The model was then used to predict reservoir performance under different production/injection scenarios over the next 30 years. Forecast runs showed that the pressure declines almost equally in all areas, consistent with the high permeability and connectivity of the reservoir, which had been established from the LTFT.

Asphaltite is petroleum originated substance and formed by metamorphism. Turkey has 82 million tons of asphaltite reserves in Sirnak and Silopi region of Southeastern part of Anatolia. The proximate analysis of Sirnak Karatepe asphaltite sample indicated that the ash and sulfur content were 46.86 % and 5.56 %, respectively. In this study, Sirnak Karatepe asphaltite sample was concentrated by gravity separation and flotation methods. The aim of this research was to decrease the ash and sulfur content below 20% and 2%, respectively, which are the requirements of Sirnak Municipality. Gravity based cleaning equipments such as multi gravity separator, shaking table and Falcon concentrator gave no satisfactory results in terms of ash and sulfur removal. The products with lowest ash content were obtained with Falcon concentrator after two stage cleaning of -100 micrometer feed. Flotation parameters of Karatepe asphaltite were also examined during the study. As a result of multi-stage flotation with stage addition of Collector Accoal 9630 and depressant Na2SiO3 , ash content of asphaltite was decreased to 17.59 % with 15.31 % combustible recovery. Even though the ash content specification met by flotation, the sulfur content of cleaned asphaltite (6.68 % S) was more than the sulfur limit of Sirnak Municipality.

The objective of this thesis study is to recover nickel and cobalt at maximum efficiency from column leach liquor of lateritic nickel ores existing in G&ouml / rdes region of Manisa by performing various hydrometallurgical methods under the optimum conditions. This column leach solution of nontronite type lateritic nickel ore was initially neutralized and purified from its basic impurities by a two stage iron removal process under the optimum conditions determined experimentally. Then, nickel and cobalt were precipitated in the form of mixed hydroxide precipitate from the purified leach solution by a two stage precipitation method called &ldquo / MHP&rdquo / and a manganese removal process was carried out also under the optimum conditions determined. By decreasing Mn concentration with this process to an acceptable level yielding at most 10% Mn in hydroxide precipitate, it was possible to produce a qualified MHP product suitable to the current marketing and standard conditions. As a result of this thesis study, the experiments conducted showed that by recycle leaching with sulfuric acid about 81% of Ni and 63% of Co in the lateritic nickel ore (9.72 kg Ni / ton of ore and 0.28 kg Co / ton of ore) could be extracted as mixed hydroxide precipitate by MHP method. The MHP product contains 41.9% Ni, 1.0% Co, 2.3% Mn, 0.06% Al, 1.5% Mg, 0.02% Fe, 0.01% Cr, 0.25% Zn, 0.03% Cu and 4.73% S.

ABSTRACT HIGH PRESSURE ACID LEACHING OF TURKISH LATERITES Kaya, Serif M.Sc., Department of Metallurgical and Materials Engineering Supervisor: Prof. Dr. Yavuz A. Topkaya January 2011, 91 pages The aim of this thesis study was to investigate and find the most cost effective way of extracting nickel and cobalt into the pregnant leach solution (PLS) from G&ouml / rdes lateritic nickel and cobalt ore by means of sulphuric acid leaching under high temperature and high pressure conditions.The high pressure acid leach (HPAL) experiments were conducted with nontronitic and limonitic types of G&ouml / rdes lateritic nickel ore, respectively. Leaching experiments of nontronite ore have shown that almost all of the nickel and cobalt contained in the nontronitic ore were easily extracted into the (PLS). Therefore, the rest of the experiments were concentrated on difficult to leach limonitic sample when compared with the nontronitic one, and higher nickel and cobalt extractions were aimed to be obtained. By taking economic and technical considerations into account, the basic (HPAL) process parameters for the limonitic sample were optimized as / leaching at 255 &deg / C with a particle size of 100% -850 &mu / with 0.30 sulphuric acid to ore weight ratio in 1 hour of leaching duration. The experiments were conducted with 30% solids ratio and it was found that 87.3% of nickel and 88.8% of cobalt present in the limonitic ore could be extracted into the pregnant leach solution. Nevertheless, these results were found to be below the desired values. Therefore, the possible reasons of this behavior were investigated and the presence of hematite mineral in the limonitic ore was found to be the most probable one. Therefore, in order to dissolve the nickel and cobalt present in the hematite mineral, the additions of HCl, ferrous ions, cuprous ions and sulphur were tried, respectively and they were found to be beneficial in order to increase the degree of nickel and cobalt extractions.

The conventional cement grinding system, the ball mill, has very poor efficiencies in spite of innovative improvements. For this purpose, development of new techniques, which allow proper size reduction and uniform particle size distribution with less specific energy consumptions, have become a necessity. The aim of this study is to make a comparative analysis of the fairly new cement grinding technologies, COMFLEX&reg / Grinding System, Roller Press and HOROMILL&reg / , at the same cement production plant with the same raw materials. In this context, CEM I 42.5 R type cement was produced with a fixed Blaine fineness of 3600 (&plusmn / 100) cm2/g at three different grinding units. The same raw materials, clinker and gypsum, and identical feeding ratios, 95% and 5%, were used to produce cement. Accordingly, these different grinding techniques were inspected with respect to the microstructural properties of cement particles, and the relative chemical, physical and mechanical properties of products. It was found that the main cement grinding parameters, specific surface area and sieve residue, do not show expected relation and change with each grinding system due to differences in the size reduction technique. Moreover, strength and other hardened mortar properties are directly affected by the liberation conditions of reactive grains at grinding stages.High capacity and low specific energy consumption i.e. the breaking and cracking efficiency of the roller press and higher grinding performance of the ball mill promoted the COMFLEX&reg / system. On the other hand, the roller press was clearly advantageous at early strength performances with moderate specific energy usages during grinding. Nonetheless, it also had drawbacks like higher water demand and earlier setting times (which mean higher hydration temperatures). When the wideness and sharpness of classification results were considered, HOROMILL&reg / gave better results with high circulation and efficient air classification design / although there were weaknesses of the system such as lower capacity and higher specific energy consumption rate.

In recent years, the risks created by the import of energy on the security of energy supply has encouraged countries to utilize local resources to a greater extend, and for many countries including Turkey coal is at the top of the local resources. However, it is not sufficient for countries to have an energy resource itself. Intens competition in today&#039 / s globalized system requires the resources to be produced and utilized in the most economical manner. The supply of the resources to the market in a competitive way is possible by employing an efficient operation, which is then possible only through the business units producing these resources working efficiently. In this study, the efficiencies of the eight establishments of the Turkish Coal Enterprises (TKI) between 2006 to 2009 were analyzed by using Data Envelopment Analysis (DEA), Super Efficiency (SE) and Malmquist Total Factor Productivity Index (MI) methodologies. For the analyses, three output oriented models were constructed and used: Production Efficiency, Revenue Efficiency and Work Safety Efficiency models. In determining the input and output data used in the analyses, it was benefited from similar studies searched in the literature, knowledge of business and economics and a series of brainstorming of the expert panel consisting of ten high level representatives of the public and private lignite mining companies as well as that of the Turkish Ministry of Energy and Natural Resources. As a part of the analyses, first the production and revenue efficiencies of the establishments were analyzed by using Constant Return to Scale (CRS) and Variable Return to Scale (VRS) methods of DEA. Within this context, efficient and inefficient establishments for the years between 2006 to 2009, and benchmarks for inefficient establishments to move to becoming efficient were determined. Furthermore, for the production and revenue efficiency models, the target values and improvement potentials for the inefficient establishments in CRS analyses to becoming more efficient were calculated by using benchmarks tables and 2009 realized values of inputs and outputs. Second, the efficiency rankings of the efficient establishments among themselves between 2006 to 2009 were determined by using SE methodology. Third, to provide the dynamic analysis of the development of the establishments&#039 / s efficiency levels in time, the changes of production, revenue and work safety efficiencies of the establishments between 2006 to 2009 were analyzed using the Malmquist Index (MI) methodology. MI analyses included the analysis of the efficiencies in four efficiency components (Technical, Technological, Pure and Scale efficiencies) as well as the calculation of the Total Factor Productivity Indexes of the establishments.

As high grade iron ore deposits are rapidly depleted, the minerals industry is increasingly obliged to enrich their ores and process lower grade iron ores. Production of large quantities of fine ores as tailings or by-products in mining operations and mineral liberation at fine particle sizes have led to the development of concentration methods employed to fine ores. Selective flocculation is one of the beneficial methods which can be used in recovering of very fine particles It is obvious that process water chemistry has a significant influence on the flocculation efficiency, and the water chemistry is deeply affected by carbon dioxide dissolution. Carbonate content of natural waters regulated by CO2, carbonicacid and solid carbonatesis known as &lsquo / carbonate alkalinity&rsquo / whichmay be an important factor in flocculation especially at alkaline pH. Selective flocculation of iron ores is usually run at around pH 11 where carbonate alkalinity could be rather high. There have been no reports in literature regarding the effect of carbonate alkalinity on the flocculation behavior of iron oxides, mainly hematite. In this study, the flocculation behavior of iron ores with starch under different alkalinities has been investigated. The extent of flocculation has been assessed by settling rate and suspended solid content measurements at different starch doses and pH values. Zeta potential measurements and starch adsorption studies were carried out to explain these effects by carbonate alkalinity. According to the results, flocculation is enhanced by addition of low amount of carbonate alkalinity, less than 2.4 mEq/L. However, adding the higher amount of alkalinity adversely affected the flocculation of hematite. Similar behavior was also observed during the starch adsorption tests, larger amount of starch wasadsorbed by hematite in low alkalinity compared to high alkalinity. Zeta potential measurements indicate that, by increasing the carbonate alkalinity of suspension, zeta potential values of solids will become more negativeleading to an increase in the stability of suspension and then adversely affecting flocculation. Therefore, as carbonate alkalinity has a significant effect on the flocculation of hematite, it should be seriously taken into account to optimize the selective flocculation of hematite ores.

The main objective in this study is to compare breakage parameters of narrow size fractions of cement clinker taken from the product end and feed end of industrial-scale high pressure grinding rolls (HPGR) in order to assess whether the breakage parameters of clinker broken in HPGR are improved or not. For this purpose, drop weight tests were applied to six narrow size fractions above 3.35 mm, and batch grinding tests were applied to three narrow size fractions below 3.35 mm. It was found that the breakage probabilities of coarse sizes and breakage rates in fine sizes were higher in the HPGR product. This indicated that clinker broken by HPGR contained weaker particles due to cracks and damage imparted. However, no significant weakening was observed for the -19.0+12.7 mm HPGR product. Although HPGR product was found to be weaker than HPGR feed, fragment size distribution of HPGR product did not seem to be finer than that of the HPGR feed at a given loading condition in either the drop weight test or batch grinding test. Also, drop weight tests on HPGR product and HPGR feed showed that the breakage distribution functions of coarse sizes depended on particle size and impact energy (J). Batch grinding tests showed that the specific breakage rates of HPGR product and HPGR feed were non-linear which could be represented with a fast initial breakage rate and a subsequent slow breakage rate. The fast breakage rates of each size fraction of HPGR product were higher than HPGR feed due to cracks induced in clinker by HPGR. However, subsequent slow breakage rates of HPGR product were close to those of HPGR feed due to elimination of cracks and disappearance of weaker particles. Besides, the variation in breakage rates of HPGR product and HPGR feed with ball size and particle size also showed an abnormal breakage zone where ball sizes were insufficient to effectively fracture the coarse particles. Breakage distribution functions of fine sizes of HPGR product and HPGR feed were non-normalizable and depended on particle size to be ground. However, batch grinding of -2.36+1.7 mm and -1.7+1.18 mm HPGR feed yielded the same breakage pattern.

<p>This master thesis consists of the creation of shock absorber models for percussive rock drills. A model of the rock was also developed. These models were connected to existing models of the rock drill to create a complete simulation of the rock drilling process.</p><p>When the models had been created the rock model was evaluated by comparison with measured values from lab tests. The verified rock model was then used to evaluate the shock absorber models.</p><p>Four new shock absorber concepts have been developed. Three of these have been modelled, while the fourth have only been sketched and described.</p><p>The evaluation of the models showed the following:</p><p>• Simulations of the single shock absorber don’t correlate to reality very well. The movement is too rapid and poorly damped.</p><p>• The simulated behaviour of the tandem shock absorber corresponds to reality, but the simulated hydraulic pressures aren’t reliable and neither is the damping.</p><p>• Simulation of the double shock absorber reflects reality fairly well, although the model of the check valve can be improved.</p><p>• There are problems with fluttering damper pressures in simulation whenever an accumulator is connected directly to the damper volume. The problem doesn’t exist if there is a line or an orifice with enough losses between the volume and the accumulator.</p><p>• The feeder system needs to be modelled.</p> / <p>Examensarbetet gick ut på att ta fram simuleringsmodeller av olika dämpare i en bergborrmaskin. Dessutom behövdes en bergmodell som gav korrekta reflexer från berget vid simulering. Dessa modeller kopplades samman med befintliga modeller för att skapa en simulering av hela bergborrningsprocessen.</p><p>Efter att modeller skapats utvärderades bergmodellen mot mätvärden som uppmätts under prov. Efter att bergmodellen verifierats användes den för att verifiera dämparmodellerna.</p><p>Fyra dämparkoncept har tagits fram. Tre av dessa har modellerats, medan det fjärde endast har skissats och förklarats.</p><p>Verifieringen av modellerna visade följande:</p><p>• Simulering av enkeldämpare stämmer dåligt med verkligheten utan ger för snabba och odämpade rörelser.</p><p>• Simulering av dubbeldämparen stämmer bra med verkligheten men modellen över backventilen behöver ses över något.</p><p>• Simulering av tandemdämpare stämmer beteendemässigt men de simulerade trycken är inte tillförlitliga och inte heller dämpningen.</p><p>• Det finns problem med trycksvängningar vid simulering av modeller där dämpvolymen är direkt ansluten till en ackumulator. Om en strypning eller en ledning med tillräckliga förluster ansluter ackumulator och dämpvolym med varandra märks inte detta problem.</p><p>• Modell över matningen behöver tas fram.</p>