In the presented thesis we study the changes in distribution of the size of vacancy clusters in metals processed by severe plastic deformation, so called ul- tra fine grained materials. We use a modern non-destructive method of positron annihilation spectroscopy, which is one of the few methods that allow us to inves- tigate point defects like vacancies with sizes of a few ˚A. The obtained spectra of positrons annihilating in the samples enable us to determine changes of vacancy cluster sizes depending on temperature or severity of the deformation applied on the samples. 1

This diploma thesis focused on manufacturing and characterization of Ti-15Mo metastable beta-Ti alloy prepared by cryogenic milling and spark plasma sintering. Initial powder was prepared by gas atomization and consequently deformed by cryogenic milling (milled powder). Both initial and milled powders were compacted by spark plasma sintering (SPS) at temperatures from 750 řC to 850 řC. Dependence of microstructure and mechanical properties on the parameters of preparation was studied. During cryo-milling, powder particles significantly changed shape from ball-shaped to disc-shaped. Particles were not refined by milling, but severely plastically deformed. SEM observations showed that all prepared samples contain duplex alpha + beta structure. Volume fraction of alpha phase is significantly higher in the sintered milled powder due to increased beta- transus temperature caused by contamination by oxygen and also due to easier alpha phase precipitation caused by refined microstructure. Maximum microhardness of 350 HV was achieved for both types of sintered powders. High microhardness of sintered initial powder can be attributed to formation of omega phase during cooling, while sintered milled powder is strengthened by refined microstructure and small alpha phase precipitates. Cryogenic milling prior to...

Untersuchungen zur Dauerhaftigkeit von ARGlasbewehrung im Textilbeton haben gezeigt, dass durch die Alkalität des Betons in Verbindung mit Feuchtigkeit eine Glaskorrosion hervorgerufen wird, die im Laufe der Zeit zu Festigkeitsverlusten des Glases führt. Eine Möglichkeit, die durch die Glaskorrosion verursachten Festigkeitsverluste zu reduzieren, stellt die Polymermodifikation des Betons dar. Durch die Polymerzugabe wird die Wasseraufnahme der Feinbetonmatrix reduziert, dadurch sinkt der Gehalt an gelösten Alkalien im Bereich der Bewehrung. Um den Einfluss verschiedener Feinbetonmatrices auf die Dauerhaftigkeit von Textilbeton beurteilen zu können, sind u. a. zeit- und tiefenabhängige Informationen zur Feuchteverteilung erforderlich, die durch den Einsatz der NMR-Technik gewonnen werden. Der nachfolgende Artikel beschreibt den Feuchtetransport in einer speziell für den Textilbeton entwickelten Feinbetonmatrix sowie den Einfluss verschiedener Modifikationsstoffe auf das Wasseraufnahmeverhalten des Betons. / Durability tests of textile reinforced concrete revealed a loss of strength of the AR-glass reinforcement due to glass corrosion effected by the alkalinity and moisture content of the concrete. In order to reduce this strength loss of AR-glass in cementitious matrices, polymers can be used for concrete modification. The aim of the polymer addition is to reduce the amount of capillary water absorption of the matrix, which reduces the amount of free alkalies closed to the reinforcement. In order to evaluate the effect of the concrete matrix on the durability of TRC, it is necessary to determine the moisture content as functions of time and depth. This data can be obtained by the use of nuclear magnetic resonance (NMR) technique. This paper deals with the moisture transport in a finegrained concrete matrix especially developed for the use in TRC as well as the influence of polymer addition on the water absorption properties of the concrete matrix.

Die Einhaltung von Brandschutzanforderungen ist ein wichtiger Aspekt für sichere Baukonstruktionen. Beim innovativen Werkstoff textilbewehrter Beton, der einen Verbundwerkstoff aus einer Feinbetonmatrix und textiler Bewehrung darstellt, ist das Brandverhalten bisher nur unzureichend erforscht worden. Insbesondere das Tragverhalten der einzelnen Komponenten unter hohen Temperaturen stellt noch eine Wissenslücke in der heutigen Forschung dar. Dieser Artikel befasst sich mit den experimentellen Untersuchungen an einer Feinbetonmatrix, die ein Größtkorndurchmesser von 0,6 mm aufweist, sowie an AR-Glas- und Carbongarnen. Basierend auf instationären Versuchen werden das Spannungs- und Dehnungsverhalten unter hohen Temperaturen abgeleitet und Ansätze zur rechnerischen Beschreibung des Hochtemperaturverhaltens vorgeschlagen. / The design of structural members under fire attack is an important aspect for safe constructions. For the innovative material textile reinforced concrete (TRC), which is a composite material made of fine-grained concrete and textile reinforcement, the fire behavior has not been investigated insufficiently yet. Especially the load-bearing behavior under high-temperatures of the single components marks a gap in the state-of-the-art of science and technology today. This article deals with experimental investigations on a fine-grained concrete matrix, which has maximum grain size of only 0.6 mm, as well as yarns made of AR-glass and carbon. On the basis of transient tests the stress and strain behavior under high temperatures is derived. Finally, a calculative approach for the hightemperature behavior is presented.

In this study, to increase inference speed of convolutional neural networks, the optimization technique low-rank tensor decomposition has been implemented and applied to AlexNet which had been trained to classify dog breeds. Due to a small training set, transfer learning was used in order to be able to classify dog breeds. The purpose of the study is to investigate how effective low-rank tensor decomposition is when the training set is limited. The results obtained from this study, compared to a previous study, indicate that there is a strong relationship between the effects of the tensor decomposition and how much available training data exists. A significant speed up can be obtained in the different convolutional layers using tensor decomposition. However, since there is a need to retrain the network after the decomposition and due to the limited dataset there is a slight decrease in accuracy. / För att öka inferenshastigheten hos faltningssnätverk, har i denna studie optimeringstekniken low-rank tensor decomposition implementerats och applicerats på AlexNet, som har tränats för att klassificera hundraser. På grund av en begränsad mängd träningsdata användes transfer learning för uppgiften. Syftet med studien är att undersöka hur effektiv low-rank tensor decomposition är när träningsdatan är begränsad. Jämfört med resultaten från en tidigare studie visar resultaten från denna studie att det finns ett starkt samband mellan effekterna av low-rank tensor decomposition och hur mycket tillgänglig träningsdata som finns. En signifikant hastighetsökning kan uppnås i de olika faltningslagren med hjälp av low-rank tensor decomposition. Eftersom det finns ett behov av att träna om nätverket efter dekompositionen och på grund av den begränsade mängden data så uppnås hastighetsökningen dock på bekostnad av en viss minskning i precisionen för modellen.

No description available.

No description available.

The industrial shift from single core processors to multi-core ones introduced many challenges. Among them, a program cannot get a free performance boost by just upgrading to a new hardware because new chips include more processing units but at the same (or comparable) clock speed as the previous generation. In order to effectively exploit the new available hardware and thus gain performance, a program should maximize parallelism. Unfortunately, parallel programming poses several challenges, especially when synchronization is involved because parallel threads need to access the same shared data. Locks are the standard synchronization mechanism but gaining performance using locks is difficult for a non-expert programmers and without deeply knowing the application logic. A new, easier, synchronization abstraction is therefore required and Transactional Memory (TM) is the concrete candidate. TM is a new programming paradigm that simplifies the implementation of synchronization. The programmer just defines atomic parts of the code and the underlying TM system handles the required synchronization, optimistically. In the past decade, TM researchers worked extensively to improve TM-based systems. Most of the work has been dedicated to Software TM (or STM) as it does not requires special transactional hardware supports. Very recently (in the past two years), those hardware supports have become commercially available as commodity processors, thus a large number of customers can finally take advantage of them. Hardware TM (or HTM) provides the potential to obtain the best performance of any TM-based systems, but current HTM systems are best-effort, thus transactions are not guaranteed to commit in any case. In fact, HTM transactions are limited in size and time as well as prone to livelock at high contention levels. Another challenge posed by the current multi-core hardware platforms is their internal architecture used for interfacing with the main memory. Specifically, when the common computer deployment changed from having a single processor to having multiple multi-core processors, the architects redesigned also the hardware subsystem that manages the memory access from the one providing a Uniform Memory Access (UMA), where the latency needed to fetch a memory location is the same independently from the specific core where the thread executes on, to the current one with a Non-Uniform Memory Access (NUMA), where such a latency differs according to the core used and the memory socket accessed. This switch in technology has an implication on the performance of concurrent applications. In fact, the building blocks commonly used for designing concurrent algorithms under the assumptions of UMA (e.g., relying on centralized meta-data) may not provide the same high performance and scalability when deployed on NUMA-based architectures. In this dissertation, we tackle the performance and scalability challenges of multi-core architectures by providing three solutions for increasing performance using HTM (i.e., Part-HTM, Octonauts, and Precise-TM), and one solution for solving the scalability issues provided by NUMA-architectures (i.e., Nemo). • Part-HTM is the first hybrid transactional memory protocol that solves the problem of transactions aborted due to the resource limitations (space/time) of current best-effort HTM. The basic idea of Part-HTM is to partition those transactions into multiple sub-transactions, which can likely be committed in hardware. Due to the eager nature of HTM, we designed a low-overhead software framework to preserve transaction's correctness (with and without opacity) and isolation. Part-HTM is efficient: our evaluation study confirms that its performance is the best in all tested cases, except for those where HTM cannot be outperformed. However, in such a workload, Part-HTM still performs better than all other software and hybrid competitors. • Octonauts tackles the live-lock problem of HTM at high contention level. HTM lacks of advanced contention management (CM) policies. Octonauts is an HTM-aware scheduler that orchestrates conflicting transactions. It uses a priori knowledge of transactions' working-set to prevent the activation of conflicting transactions, simultaneously. Octonauts also accommodates both HTM and STM with minimal overhead by exploiting adaptivity. Based on the transaction's size, time, and irrevocable calls (e.g., system call) Octonauts selects the best path among HTM, STM, or global locking. Results show a performance improvement up to 60% when Octonauts is deployed in comparison with pure HTM with falling back to global locking. • Precise-TM is a unique approach to solve the granularity of the software fallback path of best-efforts HTM. It provide an efficient and precise technique for HTM-STM communication such that HTM is not interfered by concurrent STM transactions. In addition, the added overhead is marginal in terms of space or execution time. Precise-TM uses address-embedded locks (pointers bit-stealing) for a precise communication between STM and HTM. Results show that our precise fine-grained locking pays off as it allows more concurrency between hardware and software transactions. Specifically, it gains up to 5x over the default HTM implementation with a single global lock as fallback path. • Nemo is a new STM algorithm that ensures high and scalable performance when an application workload with a data locality property is deployed. Existing STM algorithms rely on centralized shared meta-data (e.g., a global timestamp) to synchronize concurrent accesses, but in such a workload, this scheme may hamper the achievement of scalable performance given the high latency introduced by NUMA architectures for updating those centralized meta-data. Nemo overcomes these limitations by allowing only those transactions that actually conflict with each other to perform inter-socket communication. As a result, if two transactions are non-conflicting, they cannot interact with each other through any meta-data. Such a policy does not apply for application threads running in the same socket. In fact, they are allowed to share any meta-data even if they execute non-conflicting operations because, supported by our evaluation study, we found that the local processing happening inside one socket does not interfere with the work done by parallel threads executing on other sockets. Nemo's evaluation study shows improvement over state-of-the-art TM algorithms by as much as 65%. / Ph. D.

Sentimentanalys är en teknik som syftar till att automatiskt identifiera den känslomässiga tonen i text. Vanligtvis klassificeras texten som positiv, neutral eller negativ. Nackdelen med denna indelning är att nyanser går förlorade när texten endast klassificeras i tre kategorier. En vidareutveckling av denna klassificering är att inkludera ytterligare två kategorier: mycket positiv och mycket negativ. Utmaningen med denna femklassificering är att det blir svårare att uppnå hög träffsäkerhet på grund av det ökade antalet kategorier. Detta har lett till behovet av att utforska olika metoder för att lösa problemet. Syftet med studien är därför att utvärdera olika klassificerare, såsom MLP, CNN och Bi-GRU i kombination med word2vec för att klassificera sentiment i text i fem kategorier. Studien syftar också till att utforska vilken förbehandling som ger högre träffsäkerhet för word2vec.   Utvecklingen av modellerna gjordes med hjälp av SST-datasetet, som är en känd dataset inom finfördelad sentimentanalys. För att avgöra vilken förbehandling som ger högre träffsäkerhet för word2vec, förbehandlades datasetet på fyra olika sätt. Dessa innefattar enkel förbehandling (EF), samt kombinationer av vanliga förbehandlingar som att ta bort stoppord (EF+Utan Stoppord) och lemmatisering (EF+Lemmatisering), samt en kombination av båda (EF+Utan Stoppord/Lemmatisering). Dropout användes för att hjälpa modellerna att generalisera bättre, och träningen reglerades med early stopp-teknik. För att utvärdera vilken klassificerare som ger högre träffsäkerhet, användes förbehandlingsmetoden som hade högst träffsäkerhet som identifierades, och de optimala hyperparametrarna utforskades. Måtten som användes i studien för att utvärdera träffsäkerheten är noggrannhet och F1-score.   Resultaten från studien visade att EF-metoden presterade bäst i jämförelse med de andra förbehandlingsmetoderna som utforskades. Den modell som hade högst noggrannhet och F1-score i studien var Bi-GRU. / Sentiment analysis is a technique aimed at automatically identifying the emotional tone in text. Typically, text is classified as positive, neutral, or negative. The downside of this classification is that nuances are lost when text is categorized into only three categories. An advancement of this classification is to include two additional categories: very positive and very negative. The challenge with this five-class classification is that achieving high performance becomes more difficult due to the increased number of categories. This has led to the need to explore different methods to solve the problem. Therefore, the purpose of the study is to evaluate various classifiers, such as MLP, CNN, and Bi-GRU in combination with word2vec, to classify sentiment in text into five categories. The study also aims to explore which preprocessing method yields higher performance for word2vec.   The development of the models was done using the SST dataset, which is a well-known dataset in fine-grained sentiment analysis. To determine which preprocessing method yields higher performance for word2vec, the dataset was preprocessed in four different ways. These include simple preprocessing (EF), as well as combinations of common preprocessing techniques such as removing stop words (EF+Without Stopwords) and lemmatization (EF+Lemmatization), as well as a combination of both (EF+Without Stopwords/Lemmatization). Dropout was used to help the models generalize better, and training was regulated with early stopping technique. To evaluate which classifier yields higher performance, the preprocessing method with the highest performance was used, and the optimal hyperparameters were explored. The metrics used in the study to evaluate performance are accuracy and F1-score.   The results of the study showed that the EF method performed best compared to the other preprocessing methods explored. The model with the highest accuracy and F1-score in the study was Bi-GRU.

The aim of the thesis has been to evaluate and understand the effect of caustic alkali solution of varying composition on the behaviour of expansive soil containing mixed layered minerals. Mixed layered minerals are formed of two or more kinds of inter grown layers, not physical mixtures. Illite - smectite is the most abundant and wide spread of the mixed layered clay minerals in sedimentary rocks and soils and also more common than either discrete illite or smectite. In geotechnical engineering much attention has not been paid to the behaviour of soils containing mixed layered minerals. Much less is known about the behaviour of these soils in polluted environment. Mixed layered minerals are more susceptible to environmental changes as the structural linkages between the layer minerals are weak compared to normal layered phyllosilicates. One important pollutant that can have considerable effect on the behaviour of soils is the caustic alkali contamination released from various industries. Recent studies have shown that the behaviour of even stable minerals is affected by alkali contamination. However, the effect of caustic alkali contamination on the behaviour of soils containing mixed layered minerals is not known and has been chosen for detailed study. Also to understand the mechanism of their interaction with alkali, it is necessary to study the effect of alkali solutions on the constituent clay minerals viz., montmorillonite and illite under similar conditions. To elucidate the mechanism of soil alkali interaction limited tests were conducted with simple electrolyte solution, as the alkali solution also acts as electrolyte apart from being alkaline. To confirm the mechanism of interaction, tests are also conducted on these soils with industrial spent liquor containing high caustic alkali and suspended alumina obtained from an alumina extraction plant treating bauxite with high alkali solutions at high temperatures. The results obtained in the laboratory are compared with the soil samples contaminated with leaking industrial Bayer's liquid in the field. Studies are also conducted to suggest remedial measures to control the adverse effects of alkali solutions on soil containing mixed layer minerals. The content of the thesis is broadly divide into 8 Chapters - viz., Introduction, Background and overview, Experimental program and procedures, Behaviour of soils containing mixed layer mineral illite - smectite (BCSI), Behaviour of montmorillonite and illite, Influence of Bayer's liquor and study on the field contaminated soils, Measures to control the influence of alkali contamination on BCSI and Summary and conclusions. The broad outline of these chapters is given in Chapter 1. A review of literature on the behaviour of soils containing different types of clay minerals with emphasis on mixed layer minerals has been presented in Chapter 2. The influence of different inorganic contaminants on the properties of soils in terms of their physical and chemical characteristics as well as their concentration has been summarized. The importance of changes in surface characteristics of soil particles and the changes in the thickness of diffuse double layer in altering the property of soils at low concentration of contaminants and changes in the mineralogy with high concentrated contaminants such as acids and alkalis has been highlighted. This forms the background information necessary to bring out the scope of the study. Four soils having different mineralogy have been used in this study. These soils are, black cotton soil containing predominantly mixed layer mineral illite - smectite mineral called rectorite, illite, montmorillonite (common smectite) and black cotton soil containing predominantly montmorillonite. The properties of the soils used are described in Chapter 3. Caustic alkali solutions of 1N, 4N concentration prepared in the laboratory and industrial alkali-spent liquor are used as contaminants. The spent Bayer's liquor had about 4N alkali concentration and 10% alumina in suspension. To simulate the effect of suspended alumina, two more caustic alkali solutions of 1N and 4N solutions containing 10% alumina by weight of solutions are also prepared. To isolate the effect of electrolyte solutions from that of alkali solution, two electrolyte solutions of 1N and 4N sodium chloride solutions are also used. Test procedures for conducting various tests such as pH, water adsorption characteristics, X-ray diffraction studies, SEM studies, thermal characteristics and geotechnical properties such as Atterberg limits, Oedometer tests and Shear Strength are given in this chapter. The test procedures are modified, wherever necessary, to bring out the effect of contaminants, particularly the effect of duration of interaction on the properties of soils. The source and properties of black cotton soil are presented in Chapter 4. Detailed x-diffraction studies have confirmed the presence of inter layered illite-smectite mineral viz., rectorite, which is uncommon in Indian expansive soils, and is classified as CH (Clay of high compressibility) as per ASTM soil classification. Effect of alkali and salt solutions of 1N and 4N concentration on all physico chemical and geotechnical properties are studied in this chapter. As it is known that presence of certain elements such as aluminium influence the soil alkali interaction, the effect of suspended alumina along with alkali solution has also been investigated. The effect of contaminating fluids such as 1N NaOH, 4N NaOH with and without alumina, 1N NaCl and 4N NaCl on the geotechnical properties of the soil has been studied. Mineralogical changes were observed by XRD and thermal studies in the soil treated with 4N NaOH solution and 4N NaOH + 10% alumina. The interlayer potassium of illite is released and potassium hydroxide is formed in soil treated with 4N NaOH. Swelling compounds such as sodium aluminium silicate hydroxide hydrate (SASH) has formed due to attack of 4N NaOH + 10% alumina on silica rather than on rectorite. Thus the studies clearly bring out that the rectorite present in the soil is dissociated only in the presence of strong alkali solutions of concentration of about 4N. The liquid limit of soil decreased with increase in the electrolyte concentration in the case of NaCl solutions. With 1N NaOH, the liquid limit of soil increased due to increase in the thickness of diffuse double layer due to increased pH. However, Proctor's maximum dry density increased and optimum moisture content decreased with 1N NaOH. With increase in the concentration of alkali solution to 4N, the rectorite dissociates into constituent minerals with the formation potassium hydroxide. The liquid limit of soil decreased probably due to the dominating influence of electrolyte nature of hydroxide solution over the effect of increased negative charge on clay particles due increase in the pH on the constituent minerals. Proctor's maximum dry density decreased and optimum moisture content increased with 4N NaOH. Sediment volume and oedometer free swell at seating/nominal surcharge load of 6.25 kPa of soil increased in 1N and 4N caustic alkali solutions, though by different mechanisms. The increase with 1N solution is essentially due to increased negative charges on clay mineral surface. However, the increase in swelling with 4N solution is associated with the dissociation of rectorite mineral and occurs in two distinct phases unlike in the case of 1N solution. While the first phase can be attributed to the effect of alkaline nature of the solution after reduction in its concentration due to reaction with rectorite and the consequent reduction in its electrolyte nature. The second phase is due to the swelling of the separated constituent minerals in the presence of excess of alkali and occurs after much delay. Consolidation behaviour of rectorite in 1N and 4N alkali solutions has been studied in two ways: 1). Loading without waiting for the second stage of swelling to occur, as in standard consolidation procedure and 2). Loading after completion of second stage of swelling which is occurring after considerable delay as explained earlier. Normally one would initiate loading after equilibrium is reached at the end of first stage of swelling and second stage of swelling is not suspected. As there is no second stage of swelling with 1N solutions, these two types of consolidation tests produced the same results. Abnormal rebound is observed during unloading with 4N solution in which loading cycle is initiated without waiting for second stage of swelling to complete. It is interesting to note that while the liquid limit of soil decreased with increase in the concentration of alkali solution, the swelling increased. The testing procedure and period of interaction as well as the concentration of alkali solution during the test in these two tests are different. The effects of alkali solution are more severe in case of liquid limit because of thorough mixing and consequent effective reaction during testing. Similarly, the volume changes in soil that has already reacted with 4N alkali solution when exposed to further to alkali contamination are considerably less compared to uncontaminated soil exposed to fresh contamination. The shear strength of soil treated with 4N-alkali solution has increased particularly after long period of interaction. This indicates that the soil after mineralogical changes posses good strength. Chapter 5 presents the effect of alkali and salt solutions on the physico chemical and geotechnical properties of component minerals of mixed layered illite/smectite. For this study, commercially obtained montmorillonite (bentonite), naturally occurring black soil containing montmorillonite and commercially pure illite are used. It was observed that montmorillonite alkali reactions would not produce significant mineralogical changes where as illite is dissociated into smectite with the formation of potassium silicate by the interaction of released potassium with soluble silica. This confirms that the ultimate products of rectorite with alkali solutions would be smectite and compounds of potassium. In the absence of mineralogical alterations the liquid limit of montmorillonite decreases due to suppression of diffuse double layer thickness due to dominating influence of alkali solutions on this highly active clay. However a small increase in liquid limit is observed in illite with alkali solutions. Thus the net effect of alkali on rectorite is to decrease the liquid limit with increase in alkali concentration. While the free swell and oedometer swelling of montmorillonite generally decreases with increase in the alkali concentration, they increase in illite. However, in both the minerals the swelling occurs only in one phase. Thus the second phase of swelling that has been observed in rectorite can be attributed to delayed swelling of montmorillonite that has been released by the attack of alkali on rectorite. The behaviour of black soil containing mixed layer mineral contaminated in the field and laboratory by leaking Bayer's spent liquor in an alumina extraction plant has been studied in Chapter 6. The Atterberg limits of the samples treated with liquor are reduced and sediment volume increased. Similarly the swelling at seating load in consolidation test is higher in sample compacted with water and inundated with liquor. X-ray diffraction studies showed that the mineralogical changes are similar to those occurred with 4N caustic alkali solution. The mineralogical and micro structural changes in the soil samples that are contaminated by leaked spent liquor in the field are relatively more marked. Also the behavior of highly montmorillonite clay, bentonite, has been studied contaminated with liquor in the laboratory. The study on the effect of high concentrated alkali solutions on montmorillonite can be useful to study the effect of interaction on the dissociated montmorillonite. These studies are helpful to suggest some possible remedial measures to control the adverse effect of alkali on soils. Possible Remedial schemes that can be adopted before and after contamination of the soil to control the adverse effect of alkali solutions on the black cotton soil containing mixed layered mineral are listed and their effectiveness examined in Chapter 7. The suggested remedial measures include flushing with water to dilute the effect of alkali, neutralisation with dilute hydrochloric acid, stabilisation of soil with lime and calcium chloride and use of impervious membrane to separate the foundation soil from alkali solution. The effectiveness of different measures as well as the method of their application has been described. Efforts are made to understand the mechanism of remedial action. Consolidation tests conducted on soil contaminated with 4N alkali solution and inundated with water showed increased swelling due to dilution of the alkali concentration. Though the swelling of contaminated soil can be controlled by passing dilute hydrochloric acid (1N), the method is not advocated as it can lead to ground water contamination. Mixing the soil with solutions containing up to 5% by weight of calcium compound in water could not prevent the alkali induced heave in the long run when inundated with 4N alkali solution. This was due to dissolution of silica by the strong alkali solutions and formation of swelling compounds such as sodium aluminium silicate hydroxide hydrate (SASH). The formation of sodium aluminates occurred only when the alkali solution contained alumina or soil contained calcium compounds. There are no significant variations in the effects of calcium chloride or calcium hydroxide on contaminated soil. Replacing the foundation soil with soil thoroughly contaminated with 4N alkali solutions and controlling the migration of contaminants into the foundation soil using high-density polyethylene (HDPE) geosynthetic membranes can be an effective measure to control the heaving in alkali contaminated foundation soil containing interstratified illite – smectite. Summary and the major conclusions of the thesis are presented in Chapter 8.