Small Furnace Research Articles

Slag remains from the Na Slupi site (Prague, Czech Republic): evidence for early medieval non-ferrous metal smelting

Archaeological investigation of the Na Slupi site in the central part of Prague (the capital of the Czech Republic) and located in the vicinity of Vyšehrad Castle has demonstrated the presence of early medieval (12th century) large-scale smelting of non-ferrous metals. This study is focused on the investigation of smelting remains (Pb-rich red and black slags) using a combination of geochemical and mineralogical methods in order to understand the technologies and processes used at this site. Glassy red slags are composed of high-temperature Ca-, K- and Pb-rich silicates (wollastonite, larnite, kalsilite, melilite) and Pb-bearing glass, and are associated with less abundant litharge fragments/red-slag-impregnated furnace walls composed of litharge/massicot (PbO) and Cu2PbO2. They indicate that the cupellation process in small furnaces, rather than in crucibles (open-vessels) was used for Ag refining at the studied site. The black slags, found in much lower quantity at this site, are composed of maghemite, fayalite, clinopyroxene, melilite and glass and are S-poor. We believe that they correspond to waste products from the first stage of Ag purification (prior to final cupellation), where slagging additives such as Fe-rich material, silica sand or ashes could be added to remove impurities. The Pb isotopic composition of both types of slags is similar to the isotopic signature of Pb–Zn ores from the vicinity of Cracow (Poland) and supports the hypothesis that Pb was probably traded between Prague and Poland in the 12th century and used for refining precious metals in metallurgical workshops.

In Operando Observation of (La, Sr)CoO3 Perovskite Oxide Solid/Gas Interface by X-ray Absorption Spectroscopy

1. Introduction Lowering of the cathodic overvoltage is one of the paramount strategies to improve the performance of solid oxide fuel cells (SOFCs). La1-xSrxCoO3-δ (LSC) is considered a potential cathode material for SOFCs since it shows the high electronic and ionic mixed conductivity. The cathodic reaction contains many elementary reaction processes such as gas diffusion, surface reaction, bulk diffusion and ion transfer. The previous studies suggested that the surface reaction is the rate-determining step of LSC electrode reaction [1]. However, the processes occurring at the electrode surface is poorly understood. In this study, we investigate the chemical structure of LSC surface during SOFC operating condition by means of in operandoX-ray absorption spectroscopy, and then surface modification is examined to improve the surface reaction kinetics. Some researchers indicate that LSC is exposed to Sr segregation under SOFC operating condition [2, 3]. Because this reaction is occurred at nanometer scale of the electrode surface, the true phenomena have not been clearly understood. In order to analysis surface chemistry during SOFC operation condition, in operandototal-reflection fluorescence XAS was applied. This method is useful to observe the surface reaction at the nanometer scales[4]. Surface modification of LaSrCoO3- δ (LSC214) on La0.6Sr0.4CoO3- δ (LSC113) is interesting strategy to improve surface reaction [5]. To understand the effect of surface modification, we examined uncoated LSC113 and LSC214 coated LSC113 electrodes with electrochemical impedance spectroscopy, in operandoXAS and fluorescence X-ray analysis (XRF). 2. Experiment Model thin film electrodes were prepared by plasma laser deposition (PLD) method. Substrate was YSZ (110) single crystal. First we deposited Gd0.2Ce0.8O1.9 buffer layer about 5 nm and then La0. 6Sr0. 4CoO3 - d was deposited for about 70 nm thickness. For the surface modification, LaSrCoO3- δwas coated with various thicknesses.We characterized crystal structure by XRD, thickness by a transmission electric microscope, surface roughness by atomic force microscope. Electrochemical impedance spectroscopy was measured range of 1 M-0.1 Hz.For in operandoXAS measurement, homemade cells were used. Counter and reference electrode were porous Pt. We made these 3-electrode cells 5 mm-square as samples for XRF and XAS, and placed into small electric furnace. Co K-edge XAS was measured at BL01B1, SPring-8 in Japan. The fluorescence mode was used and incident X-ray angle was determined as a half of the count value at the critical angle in total reflection measurements. The penetration depth is estimated approximately 5 nm. 3. Results Figure 1 shows Nyquist plots from electrochemical impedance spectroscopy for uncoated LSC113 and LSC214 coated electrodes. The LSC214 coating significantly decreases the electrode resistance. Especially LSC214 coating for 3 nm on LSC113 cathode shows highest reaction kinetics.Figure 2 shows absorption edge energy shift from Co K-edge XAS as a function of applying voltage which is converted to effective oxygen partial pressure [1]. For the uncoated LSC113 electrode, significant energy shift is observed in cathodic condition. Combined with the result of fluorescence spectra, Sr segregation is proofed for the LSC113 electrode. On the other hand, the energy shift is prohibited by LSC 214 coating. As shown in Fig. 2, LSC214 coating suppresses the reduction of Co ion in LSC113 and Sr segregation on surface.Fig.1. Nyquist plot of electrochemical impedance spectroscopy for the uncoated LSC113 and the LSC214 coated LSC113 at 1%-O2atmosphere.Fig.2. Absorption edge shift of XAS as a function of effective partial pressures of oxygen (left). Models of Sr segregation for the uncoated LSC113 and the LSC214 coated LSC113 (right). Reference [1] T. Kawada et. al., J. Electrochem. Soc. 149E252 (2002)[2] E. Crumlin et. al., Energy Environ. Sci., 5, 6081(2011)[3] E. Mutoro et al., J. Phys. Chem. Lett., 3, 40 (2012).[4] D.Takamatsu et a l., Angew. Chem. Int. Ed., 51 ,11597 (2012).[5] E Crumlin et. al., J. Phys. Chem. Lett., 1, 3149 (2010) Acknowledgement This research was partially supported by the Ministry of Education, Science, Sports and Culture, Grant-in-Aid for Grant-in-Aid for Young Scientists (B) (25810134）

Composition Dependences of Property and Conductivity in (Sr,Ca)3Nb2O7- δ

INTRODUCTION Superconductors which shows complete conductivity at low temperature is expected to be applied for many electronic technologies, such as power transmission line and an electricity accumulation system like the flywheel, and operation tests for these technologies have been in progress nowadays. Most of superconductors with a high superconduction transition temperature which can be worthy of practical use include copper as a constituent element. However, the high temperature superconduction mechanism has not been understood completely yet. In order to get new knowledge for the high temperature superconduction mechanism, we have tried to discover new superconduction oxide withough copper. In our previous works, we performed systematic investigation on electrical conductivities of alkaline-earth niobium oxides, especially Sr1-xMxNbO3-δ (M = Nd, Ce), Sr2-xMxNbO4-δ (M = Ce, La), Sr3-xMxNb2-yM'yO7-δ (M = La, Ce; M’ = V), CaNbO3-δ, Ca2NbO4-δ and Ca3Nb2O7-δ, but these oxides did not exhit superconductivity.From such background, we paid attention to Sr3-xCaxNb2O7-δnewly in this study, and prepared the samples with various Ca-substitution amounts. For these samples, we examined properties and composition dependence of the conductivity. EXPERIMENTAL We prepared NbO2, which was a starting material for Sr3-xCaxNb2O7-δ synthesis, by firing mixed powder of Nb and Nb2O5 with a molar ratio of 1.1:2 at 1080 oC for 36 h under 10-4 Pa using a small metal vacuum furnace. In addition, we heat-treated SrCO3 and CaCO3 at 1000 oC, for 24 h in air and then made SrO and CaO. We mixed NbO2, SrO and CaO in Ar atmosphere with an excess amount of Nb compoud in order to controll Nb valence to around 4. After pellet molding, we sintered the pellet at 1080 oC for 36 h under 10-4 Pa, and then obtained Sr3-xCaxNb2O7-δ(0≤x≤1.5). We put a Nb small piece as a getter at the sintering.For these samples, we performed phase identification by powder X-ray diffraction (XRD), metal composition analysis by the ICP emission spectrometry analysis, and calculation of the mean valence of the Nb by TG-DTA. Their electrical conductivities were also estimated by the resistivity measurement by 4 probe direct current method at the temperature range from room temperature to 4.2 K. In order to clarify the crystal structures, the Rietveld analysis (Rietan-FP) using synchrotron X-ray diffraction data (BL02B2, SPring-8) was carried out. RESULTS AND DISCUSSION XRD measurements confirmed that a main phase of Nb oxide used as a starting material for Sr3-xCaxNb2O7-δ synthesis was NbO2, and the mean Nb valence was evaluated as 4.11 from a weight increase due to Nb oxidation measured by TG-DTA. We used this NbO2 as a raw material and prepared Sr3-xCaxNb2O7-δ. In this process, 7.5 % excess Nb compound was added. In order to identify the phase of the obtained sample, the XRD pattern was collected. Because a crystal structure of Sr3Nb2O7 is unknown, we assigned the diffraction peaks to the same crystal structure as Sr3Mo2O7 1 ) where Mo was tetravalent as well as Nb in Sr3Nb2O7 and Mo4+ has a close ionic radius to Nb4+. As a result, we confirmed that Sr3-xCaxNb2O7 could be obtained as a main phase of the products regardless of the Ca content. Because the peak intensity ratio was dependent on the Ca-substituted quantity, the crystal structure was supposed to be changed by the partial substitution of Ca for Sr. In addition, from the analytical metal composition estimated by ICP, it was confirmed that compositions were controlled successfully. TG-DTA measurement demonstrated that the mean Nb valence was controlled to near 4 in Sr3-xCaxNb2O7-δ (0≤x≤1.5).As for these samples, we measured resistivity from room temperature to 4.2 K (Fig. 1). From this figure, it was found that the resistivity decreased with increasing the Ca-substituted quantity up to x=0.5. However, the sudden drop of the resistance which is one of the characteristics of a superconductor was not observed, and thus the superconductivity could not be achieved in Sr3-xCaxNb2O7-δ. In addition, the resistivity behaviors against temperature were changed below 150 K. This result suggests a change of the crystal structure with decreasing temperature. REFERENCES 1) Shinji Kouno, Naoki Shirakawa, Physica B, 403, 1029 (2008).

Heat integration of isobutane dehydrogenation and hydrogenation of nitrobenzene to aniline in radial flow moving bed reactors: Modelling and operability investigation

In the process intensification subject, heat integration of endothermic and exothermic reactions in a shell and tube reactor is one of the most important approaches to reduce plant size and decreasing operational cost compared to conventional processes. This paper focuses on modelling and simulation of thermally coupled radial flow moving bed reactors to produce isobutene and aniline. A steady state heterogeneous two-dimensional model is developed to predict the performance and operability of the proposed configuration against conventional process. The catalytic nitrobenzene hydrogenation takes place in the exothermic side and supplies a part of necessary heat for isobutane dehydrogenation in the endothermic side. To prove the accuracy of the considered mathematical model and assumptions, the simulation results of the conventional process are compared with the plant data at the same process condition. Isobutane conversion is enhanced about 5.3% in the thermally coupled process compared to the conventional process. In addition, since the nitrobenzene hydrogenation reaction supplies a part of required heat in the isobutane dehydrogenation side, the smaller furnaces can be used in the commercial Oleflex process.

Slag Investigation from Iron Smelting and Iron Processing Sites in Austria – From Hallstatt and Medieval Period and the 19<sup>th</sup> Century

Slags from three iron smelting and processing sites were investigated. The composition of the slags is connected to the metallurgical processes and the investigation should provide an indication of the process conditions.A slag from a small bloomery furnace at the archaeological site Waschenberg (Hallstatt period) was investigated. Slags from the medieval period were found near Eisenerz and at this time iron was produced in developed bloomery furnaces (Stuckofen). These slags are termed fayalitic slags and they consist of wustite (FeO), fayalite (Fe2SiO4) and glass-phase (amorphous silicates) in various concentrations. Three solidification-types were described and correlated with the microstructures and the FeO-SiO2-CaO2 phase diagram.During the 19th century steel was produced in puddling furnaces by remelting iron products from Eisenerz. Such slags, from this process, found in the surroundings of Palfau, were investigated and contain FeO and Ca rich glass-phase, but no fayalite.Due to the different compositions of the slags the solidification and the microstructures are different.

Archaeometallurgical Simulations of the Processes in Bloomery Furnaces from the Hallstatt and Medieval Period

Bloomery furnaces were the first units for iron smelting. In the Hallstatt period small bowl-type furnaces were used and until the medieval period the size of such furnaces was increasing continuously. Experimental archaeologists reconstruct bloomery furnaces to study the processes of bloom production. In a small bowl-type furnace (Hallstatt period) at Asparn and in a larger shaft-type furnace (medieval period) at Ybbsitz smelting experiments were performed, The samples contained metallic iron and slag. Various amounts of iron in different stages of conglomeration up to larger iron pieces were found. The slag belongs to a fayalitic-type, consisting of wustite (FeO), fayalite (Fe2SiO4) and glass-phase (amorphous Ca-, Al-silicates) in various concentrations. The yield of metallic iron was highly different for the various experiments. In general, more metallic iron was formed in the larger shaft-type furnaces. A large bloom was not obtained.

Destruction of polychlorinated naphthalenes by a high-temperature melting treatment (GeoMelt process)

A series of treatment experiments were carried out to evaluate the applicability of a high-temperature melting treatment (GeoMelt process) to the destruction of polychlorinated naphthalene (PCN) formulation. We started with 10-kg-scale experiments in which a small melting furnace was used and then scaled up to a 1-t-scale experiment in which a melting furnace that resembled an actual treatment system was used. These runs were evaluated whether destruction efficiency (DE) of total PCNs was more than 99.999% and whether concentrations of PCNs and polychlorinated dibenzo-p-dioxins/dibenzofurans (PCDDs/DFs) in vitrified materials, emission gas, and scrubber water were below the target levels. Because DE values and the target levels of PCNs and PCDDs/DFs in these runs were satisfactory, then we carried out a demonstrative experiment using the actual treatment system and confirmed destruction of PCNs. Based on good results of the demonstrative experiment, stock of PCN formulation was successfully treated continuously.

Modeling Based on T-S Model for Thermal Process in Metallurgical Performance Testing

Most of metallurgical performance testing devices use small high-temperature furnace to simulate physical environment for the sample testing. Since the controlled object has the dynamic characteristics of nonlinear, time-varying, large delay and large inertia during heating process, it is difficult to establish an accurate models to control thermal processes and optimize. This paper presents an adaptive neural fuzzy modeling approach based on T-S model for the heating process. Using the fuzzy system structure identification and parameter identification, the more accurate nonlinear model can be obtained. Duo to the fuzzy neural network has the capability of autonomous, quickly and effectively converging to the required relations of the input and output, the modeling accuracy has been improved. The simulation results demonstrate the effectiveness of the proposed algorithm, and the method can provide a reference for obtaining accurate nonlinear model.

Holzfeuerungen: eine bedeutende Quelle von Feinstaub in der Schweiz

Wood combustion: a substantial source of airborne particulate matter in Switzerland Wood is a renewable energy source. Wood combustion for heating purposes therefore helps in reducing CO2 emissions. However, it often results in high emissions of particulate matter (PM) which includes both black carbon (BC) and organic carbon (OC). PM has adverse health effects and should therefore be minimized. This paper reports on the latest methods to quantify the contribution of wood combustion to PM load and gives values for PM, BC, and OC from wood combustion at a number of different sites in Switzerland. State of the art methods to characterize emissions are presented and examples are given. It is shown that a major fraction of the emissions stems from small wood stoves, where the emissions are especially high during the starting phase. In addition, these small furnaces emit large amounts of gases which are rapidly oxidized and form secondary aerosols in the atmosphere. Improvements in the emissions of small wood stoves can be achieved by an increased deployment of pellet ovens, by the development and application of new technology for exhaust gas cleaning, as well as through other ways of wood usage.

Temperature assessment on fire resistance experiments of partially encased beams

Twelve fire resistance tests, grouped in four series, were developed using partially encased beams (PEB) without concrete slab, for different load level and shear connection. PEB were built with standard hot rolled IPE100 profile and reinforced concrete between flanges. Fire resistance was determined for standard ISO834 nominal heating curve, using small fire resistance furnace and portal frame. Temperature was measured in three different sections along PEB length, for three different materials (steel, concrete and reinforcement). Average temperature in each section and material was compared to the average over length temperature. Each series presented similar results, with good reproducibility. Special focus was given to critical temperature. The maximum temperature difference between sections S1, S2, S3 and the average element length temperature is smaller than 3.2% for test series 1. For test series 2, 3 and 4 the maximum temperature difference is smaller than 5.1%, 6.3% and 11.2%, respectively. Temperature is not uniform in cross-section. After the initial heating stage, temperature revealed a constant difference of approximately 150oC between temperatures measured inside and outside, defining two main temperature evolutions. Temperatures measured outside revealed always higher temperature level.

Influence of pulling rate on multicrystalline silicon ingots' properties

In this study, the influence of pulling rate on lifetime and impurity distribution in multicrystalline silicon has been investigated. Ingots have been pulled at different rates in a small scale Bridgman type furnace and were intentionally polluted with 50ppma copper and 0.5ppma of boron. The low lifetime regions are related to different defects, depending on the pulling rate. During the growth, the impurities interact with structural defects, such as grain boundaries or dislocations, leading to inhomogeneous distributions. The relative impacts of solid state diffusion and limited rejection at structural defects during the growth are investigated for each pulling rate. It is argued that solid state diffusion is the main mechanism for impurity inhomogeneity of fast diffusing elements like copper, while more slowly migrating elements, e.g. boron, show a more uniform distribution.

Characterization on The Mode of The Surface Combustion of Organic Building Materials in Madison-type Small Tunnel Furnace (II). Factors Dominating the Flame-Spread Patterns and Evolution of the Smoke for the Organic Building Materials

Characterization on The Mode of The Surface Combustion of Organic Building Materials in Madison-type Small Tunnel Furnace (II). Factors Dominating the Flame-Spread Patterns and Evolution of the Smoke for the Organic Building Materials

Optical emission from a small scale model electric arc furnace in 250–600 nm region

Optical emission spectroscopy has been for long proposed for monitoring and studying industrial steel making processes. Whereas the radiative decay of thermal excitations is always taking place in high temperatures needed in steel production, one of the most promising environment for such studies are electric arc furnaces, creating plasma in excited electronic states that relax with intense characteristic emission in the optical regime. Unfortunately, large industrial scale electric arc furnaces also present a challenging environment for optical emission studies and application of the method is not straightforward. To study the usability of optical emission spectroscopy in real electric arc furnaces, we have developed a laboratory scale DC electric arc furnace presented in this paper. With the setup, optical emission spectra of Fe, Cr, Cr2O3, Ni, SiO2, Al2O3, CaO, and MgO were recorded in the wavelength range 250-600 nm and the results were analyzed with the help of reference data. The work demonstrates that using characteristic optical emission, obtaining in situ chemical information from oscillating plasma of electric arc furnaces is indeed possible. In spite of complications, the method could possibly be applied to industrial scale steel making process in order to improve its efficiency.

Impact of growth rate on impurities segregation at grain boundaries in silicon during Bridgman growth

Bi-crystal silicon ingots with low angle grain boundaries have been grown at three different pulling rates, namely 3µm/s, 13µm/s and 40µm/s, in a small scale Bridgman type furnace. The melts have been contaminated with 50ppma of copper or indium. Selected samples were investigated by secondary ion mass spectrometry (SIMS) chemical analysis. The results show an influence of growth rate on copper segregation. Indium segregation has also been investigated. Relative impacts of solid state diffusion and solute microsegregation during solidification on the grain boundary concentration have been investigated for each growth rate.

Global simulation of the Czochralski silicon crystal growth in ANSYS FLUENT

Silicon crystals for high efficiency solar cells are produced mainly by the Czochralski (CZ) crystal growth method. Computer simulations of the CZ process established themselves as a basic tool for optimization of the growth process which allows to reduce production costs keeping high quality of the crystalline material. The author shows the application of the general Computational Fluid Dynamics (CFD) code ANSYS FLUENT to solution of the static two-dimensional (2D) axisymmetric global model of the small industrial furnace for growing of silicon crystals with a diameter of 100mm. The presented numerical model is self-sufficient and incorporates the most important physical phenomena of the CZ growth process including latent heat generation during crystallization, crystal–melt interface deflection, turbulent heat and mass transport, oxygen transport, etc. The demonstrated approach allows to find the heater power for the specified pulling rate of the crystal but the obtained power values are smaller than those found in the literature for the studied furnace. However, the described approach is successfully verified with the respect to the heater power by its application for the numerical simulations of the real CZ pullers by “Bosch Solar Energy AG”.

Highly efficient and stable implementation of the Alexander–Haasen model for numerical analysis of dislocation in crystal growth

To effectively simulate the time evolution of the dislocation density during the crystal growth process under high stress by the Alexander–Haasen model, a fully implicit and fully coupled implementation of the Alexander–Haasen model has been proposed. Numerical tests on low-stress multicrystalline silicon grown in a small furnace and high-stress seed-cast monocrystalline silicon grown in an industrial-scale furnace have been done. Results indicate that the proposed algorithm is highly efficient, strongly stable and applicable to any stress level. This algorithm provides an effective tool to optimize the crystal growth process and reduce dislocation density in industrial-scale furnaces.

Comprehensive Study of the Influence of Different Aging Scenarios on the Fire Protective Behavior of an Epoxy Based Intumescent Coating

Passive fire protection systems are widely used by the oil, gas, and chemical industries to protect steel against fire. However, there are concerns that the performance of these systems in a fire may be deteriorated because of weathering. Different weathering causes can modify fire performances of coatings, such as UV exposure, moisture, temperature, and immersion in a corrosive environment such as seawater. In this study, an intumescent fire retardant epoxy based coating containing ammonium polyphosphate (APP), melamine, and titanium dioxide was chosen as a model system. Primed steel plates covered with a 1500 μm thickness coating were exposed to different accelerated aging conditions: 80% moisture atmosphere at 70 °C for 2 months and a static immersion bath with and without NaCl (5 g/L) at 20 °C for 1 month. The formulations were then tested under hydrocarbon fire conditions using a small scale furnace test developed in our lab. Immersion in distilled water for 1 month shows a slight decrease of the protective properties: On coating extreme surface, APP turns into polyphosphoric acid, which, combined with the mechanical erosion effect of water, is dissolved/detached in water. The coatings immersed for 1 month in the salted water bath show very poor protection of the steel plate, and no intumescence is observed. Sodium and chloride migrate very rapidly into the matrix. The sodium reacts with the polyphosphates to substitute ammonium species, preventing the release of ammonia during burning and thus the char swelling. As the sodium polyphosphate created is much more soluble than the ammonium polyphosphate, it dissolves rapidly inside the matrix and reduces the quantity of phosphorus. Melamine corrosion is also accelerated by the presence of chlorides. The chemical structure of the char is as a consequence different: NASICON (Na–O–P–Ti species) are created during burning, whereas only titanium pyrophosphate is created when the reference coating is burnt.

Review of Indonesian Standard for Concrete Building Subjected to Fire

Abstract Fire safety engineering design of concrete structures in Indonesia is principally based on individual member from results of isolated beams, columns, and slabs tested in small furnaces. However, researchers identify that the behaviour of individual isolated members is significantly different from behaviour of complete structure connected together when subjected to fire. Although the current guidance is conservative when the entire building performance is considered, it should be noted that the beneficial and detrimental effects in complete structures should be taken into account. The effect of concrete spalling in fire has also not been enclosed sufficiently in the code. It should also be noted that the reduction of steel reinforcement cover could generate premature collapse of the structures. Since designers and engineers need to understand the behaviour of entire structure in fire, developing engineering guidance based on realistic structural behaviour is essential.

Modelling Possibilities of the Medieval Bloomery Process under Laboratory Conditions

Two different models have been developed under laboratory conditions based on the experiences of smelting experiments carried out in bloomery furnaces patterned on some excavated 10-12th century ones. Using Rudabánya iron ore, experiments were conducted in a closed pot and in a small open shaft furnace. During the experiments the air supply, the temperature and the weight of the iron ore and the charcoal were measured. SEM-EDX analyses were performed on bloom pieces obtained from the experiments. The results of the modelling may be correlated with the results of the previous smelting experiments. The model is sufficient for investigating some adequate parameters of the medieval bloomery technology, e.g. the iron yield.

Experimental investigation of the fire resistance of partially encased beams

This research presents the fire resistance of partially encased beams (PEB). Fire resistance was determined by means of twelve experimental tests, according to European standard EN1363-1, for different load levels and shear conditions between stirrups and web steel. Three tests were also performed at room temperature for load bearing resistance. The experimental setup was designed for this purpose, using a small fire resistance furnace and portal reaction frame. Details about the specimens tested, the instruments used and the characteristics of the materials, are also presented. Load-deflection and load-strain results were determined at room temperature. Deflection and temperature measurements were determined under ISO834 fire exposure.Results reveal the dependence of fire resistance on load level. An increase in 37% on load level is responsible for a decrease in 28% on fire resistance. Tests developed for the highest load level present similar fire resistance, independently of the shear condition; however, tests with welded stirrups present higher critical time.The main experimental results are presented and discussed. Particular emphasis was given to changes in temperature for each material, and on the global thermo-mechanical behaviour of this composite steel and concrete element.