Sedat Akkurt

The use of GA-ANNs in the modelling of compressive strength of cement mortar

In this paper, results of a project aimed at modelling the compressive strength of cement mortar under standard curing conditions are reported. Plant data were collected for 6 months for the chemical and physical properties of the cement that were used in model construction and testing. The training and testing data were separated from the complete original data set by the use of genetic algorithms (GAs). A GA– artificial neural network (ANN) model based on the training data of the cement strength was created. Testing of the model was also done within low average error levels (2.24%). The model was subjected to sensitivity analysis to predict the response of the system to different values of the factors affecting the strength. The plots obtained after sensitivity analysis indicated that increasing the amount of C3S, SO3 and surface area led to increased strength within the limits of the model. C2S decreased the strength whereas C3A decreased or increased the strength depending on the SO3 level. Because of the limited data range used for training, the prediction results were good only within the same range. The utility of the model is in the potential ability to control processing parameters to yield the desired strength levels and in providing information regarding the most favourable experimental conditions to obtain maximum compressive strength. D 2003 Elsevier Science Ltd. All rights reserved.

Ettringite formation in historic bath brick–lime plasters

Abstract Two types of historic hydraulic brick–lime plasters from a selected Ottoman bath are examined to characterize their technology and the appropriateness of their use in bath. The first type of plaster is original and structurally sound, while the historic repair plaster is the second type and is found to have deteriorated despite being exposed to the same environment. This difference is investigated by comparing their raw material compositions and the pozzolanic activities of the brick powders used in the plasters. Although these results showed no significant differences, ettringite crystals are detected in the historic repair ones by XRD, FTIR, and SEM-EDS analysis. The repair plaster may have lost its integrity owing to the expansion generated by the growth of ettringite crystals in the plaster. In this study, the ettringite formation is discussed in relation to hydration reaction products of lime–brick plasters, possible sources of gypsum, and the climatic conditions of historic bath building.

Low Temperature Synthesis of Spinel Powders by Mechanical Grinding

Proceedings of the 8th Conference and Exhibition of the European Ceramic Society; Istanbul; Turkey; 29 June 2003 through 3 July 2003

Investigation of the Pozzolanic Properties of Bricks Used in Horasan Mortars and Plasters in Historic Buildings

Crushed brick used in historic brick-lime mortars and plasters (Horasan) were characterized for understanding their properties to develop intervention mortars for conservation. Brick powders were separated from original mortars and plasters, and analyzed by XRD, SEMEDS, TGA and pozzolanic activity (PA) measurements. The results showed that the microstructures of brick powders were not completely vitrified, and their XRD analysis failed to show the peaks for high temperature (T>900C) phases. PA measurements of all brick powders showed an ample amount of artificial pozzolan property. Consequently, brick powders aimed for conservation of historic mortars must have high clay contents in addition to being low-fired (T<900C) in order to produce maximum conversion into amorphous phase for ideal PA values.

Genetic algorithm-artificial neural network model for the prediction of germanium recovery from zinc plant residues

Abstract A multi-layer, feed-forward, back-propagation learning algorithm was used as an artificial neural network (ANN) tool to predict the extraction of germanium from zinc plant residues by sulphuric acid leaching. A genetic algorithm (GA) was used for the selection of training and testing data and a GA-ANN model of the germanium leaching system was created on the basis of the training data. Testing of the model yielded good error levels (r2 = 0.95). The model was employed to predict the response of the system to different values of the factors that affect the recovery of germanium and the results facilitate selection of the experimental conditions in which the optimum recovery will be achieved.

Prediction of the Slag Corrosion of MgO-C Ladle Refractories by the Use of Artificial Neural Networks

A multilayer feed-forward back-propagation learning algorithm was employed as an artificial neural network (ANN) tool to create a model to predict the corrosion of MgO-C ladle refractory bricks based on laboratory slag corrosion test data. The corrosion process occurred by immersion of the rectangular refractory specimens in molten slag-steel bath. An ANN model to predict the amount of corrosion was created by using the training data. The model was also tested with experimentally measured data and relatively low error levels were achieved. This model was then used to predict the response of the slag-corrosion system to different values of the factors affecting the corrosion of bricks at high temperatures. Exposure time, exposure temperature of slagbrick contact and CaO/SiO2 ratio of the slag were the factors used for modelling. Model results provided the potential for selection of the best conditions for avoiding the factor combinations that may accelerate corrosion.

Microstructural Characterization of Industrial Chromite and Spinel Cement Kiln Refractories with Emphasis on the Iron-Rich Rims

Magnesia-chromite (MgO + MgO⋅Cr2O3) and magnesia-spinel (MgO + MgO⋅Al2O3) refractory bricks that are used in the high temperature zones of rotary cement kilns are investigated for their microstructural characteristics. Their microstructures are important because the size, shape and distribution of periclase grains, chromites and the quality of their bonding phases significantly affect their service performances. The purpose of this study was to characterize the microstructures of industrial brick samples to develop a protocol to compare different products e.g. for evaluation as replacement bricks. In some of the chromite containing bricks iron-rich rims were observed, while a domestic brick with similar chemistry had no such feature. These iron-rich rims were examined using SEM-EDS. It was found that the counter-diffusion of Fe +3 and Cr +3 were responsible for their formation. Exsolved chrome-spinel was widely observed in the microstructures of chromite bricks. Magnesia-spinel bricks were found to contain low melting calcium aluminates as bond phases in the microstructure, posing a threat to service performance. Portmortem microanalysis of industrially used bricks revealed alkali attack in addition to creep as main destruction mechanisms for brick. Traces of elements like Mo, S and alkalies were observed and thought to originate from the use of waste derived fuels. Introduction Refractory materials like magnesia-chromite (MgO + MgO⋅Cr2O3), magnesia-spinel (MgO + MgO⋅Al2O3), dolomite (MgO + CaO) and bricks with zirconia (ZrO2) additions are used for lining the hot zones of cement kilns [1]. Recently, due to environmental pressure, the traditional magnesiachromites are replaced by the more environment friendly magnesia-spinel and dolomite bricks in the transition and sintering zones of the kiln [2,3]. This transtion is partially complete in the developing world but totally accomplished in the industrialized countries. To compare different products for evaluation as replacement bricks and to minimize refractory wear, an understanding of the underlying chemical mechanisms must be developed. The way in which the different phases are assembled in the microstructure significantly determines the resulting corrosion behavior [4]. Refractories are generally composed of large filler grains of high melting point and a smaller grained second phase to impart high temperature bonding and packing [5]. High-melting bond phases between the refractory grains are favorable while low melting phases with a high degree of wetting are avoided when possible. Fortunately in industrial practice the hot faces of cement kiln refractories are coated with a partially solidifed cement clinker melt that acts as a protective coating like an in-situ refractory. This is a favorable condition that reduces total wear. The corrosion of these refractories were investigated in a number of studies [6-9]. The iron rich rim formation and the resulting microstructures were, however, not much reported except the works of Key Engineering Materials Online: 2004-05-15 ISSN: 1662-9795, Vols. 264-268, pp 1763-1766 doi:10.4028/www.scientific.net/KEM.264-268.1763