International Journal of Civil, Structural, Environmental and Infrastructure Engineering Research and Development (IJCSEIERD) ISSN 2249-6866 Vol. 3, Issue 3, Aug 2013, 105-112 © TJPRC Pvt. Ltd.
EFFECT OF ELEVATED TEMPERATURES ON PERFORMANCE OF RCC BEAMS ASHOK R. MUNDHADA1 & ARUN D. POPHALE2 1
Professor, Department of Civil Engineering, Prof. Ram Meghe Institute of Technology & Research, Badnera, Amravati, Maharashtra, India
Professor, Department of Civil Engineering, Visvesvaraya National Institute of Technology, Nagpur, Maharashtra, India
ABSTRACT The present work was aimed at assessing the response of reinforced concrete beams to elevated temperatures. Effect of fire on RCC is a relatively less explored area because of the lesser use of RCC structures in Europe/USA as compared to steel structures & because of the inherent fire resistance of RCC structures. Forty five RCC beam samples were cast with identical cross-sectional areas, length and grade of concrete and clear cover to reinforcement. Six Specimens were tested for the Flexural strength using UTM before heating at room temperature and the results were tabulated. Twelve specimens (6 specimens of 25mm clear cover & 6 specimens of 30mm clear cover) each were heated in an electrical furnace at 550°C for 1 hour and 2 hour respectively without any disturbance. Same procedure was repeated for 750°C and 950°C. After heating, these specimens were allowed to cool at room temperature & then tested for flexural strength on an UTM. Change in appearance & weight loss was also studied. Results revealed fairly robust performance up to 550°C. The drop in flexural strength & other parameters was noticeable but not alarming up to 750°C. Around 950°C, the rcc members lost their fidelity on all counts.
KEYWORDS: RCC, Fire, Beam Flexure Strength, Thermogravimetric Analysis Notations: C25 - Beam with clear cover 25mm (Nominal cover 20mm), C30 - Beam with clear cover (Nominal cover 25 mm)
INTRODUCTION Importance & Necessity Damage that fire can cause in terms of loss of life, homes and livelihoods is too well known. A study of 16 industrialized nations (13 in Europe plus the USA, Canada and Japan) found that, in a typical year, the number of people killed by fires was 1 to 2 per 100,000 inhabitants and the total cost of fire damage amounted to 0.2% to 0.3% of GNP. UK statistics suggest that of the half a million fires per annum attended by firefighters, about one third occur in occupied buildings and these result in around 600 fatalities (almost all of which happen in dwellings). The loss of business resulting from fires in commercial and office buildings runs into millions of pounds each year. The extent of such damage depends on a number of factors such as building design and use, structural performance, fire extinguishing devices and evacuation procedures. Although fire safety standards are written with this express purpose, it is understandably the safety of people that assumes the greater importance. Appropriate design and choice of materials is crucial in ensuring fire safe construction. Codes and regulations on fire safety are updated continually, usually as a result of research and development. Because of the extensive use of steel building frames along with combustible materials like carpets & plywood wall paneling, fire has always been an area of concern with the consultants & civic authorities in Europe & the U.S.A. That has not been the case with India where established practices like use of RCC structures & limited use of combustible materials like carpets & wooden flooring have withstood the test of time till now. But rising income levels is changing all
Ashok R. Mundhada & Arun D. Pophale
that. One of the advantages of concrete over other building materials is its inherent fire-resistive properties. However, concrete structures must still be designed for fire effects. Structural components must still be able to withstand dead & imposed loads without collapse even though the rise in temperature causes a decrease in the strength & modulus of elasticity of concrete & steel reinforcement. Fire is accidental & causes extensive damage & hardship. Like earthquake it is a calamity that comes unannounced. In fact many of the earthquakes result in fire that at times causes greater damage than its cause. Whereas, the earthquakes in India during last couple of decades have resulted in a lot of awareness & concern amongst individuals, media & authorities, the same can‘t be said about fire. Ignorance on part of the consultants & civic authorities has resulted in structures that are sub-standard from fire resistance view point. There is an urgent need to gather additional information about performance of R.C.C. under fire in order to create a general awareness & improve the existing practices & Code provisions. Research Significance Most of the engineers are not bothering about fire resistivity of structures. Not many structures under construction incorporate fire resistant design principles & measures. But the rising per capita income world-wide is making it even more essential to search for more scientific engineering solutions & upgrade our knowledge base. Rather than Active measures, design for fire safety should address passive measures. Active measures include external appurtenances like fire alarms & sprinkler systems. On the other hand, passive measures contain designing & constructing a sound fire resistant structure. However, simulating the real environment during an actual fire is a difficult & expensive job. Worldwide to this date only a couple of full sized structures have been subjected to real time fire for experimentation purpose. The most notable is the famous Cardington fire test performed in the U.K. more than a decade back. The experimentation involves expensive equipments like furnace & consumes lot of energy. Most of the experiments can be carried out on individual elements of an R.C.C. frame like beams, slabs & columns. In reality, these elements do not act in isolation. They are part of a “whole”. Restrained & seldom fully stressed! Present State of Art Across the globe, structural design for fire safety is mostly based on “Prescriptive Approach.” The prescriptive approach involves fire resistance rating of structures & was developed almost 100 years back. The data of course is being modified with new findings but is still conservative. Recommendations made by IS 1642:1989 (7) & Table 16 A in IS 456: 2000(8) subscribe to the time tested prescriptive approach. Later half of nineties brought in a paradigm shift in the fire safety engineering with the onset of the performance based design approach. George Faller (3) of Arup Fire, has advocated the performance based approach. His paper presents a performance-based method for calculating fire resistance requirements, based on the time equivalent concept. The tequivalent calculation is a function of fire load, compartment linings and ventilation conditions. The fire resistance period is calculated and then adjusted to take account of the probability of occurrence of a fire, the consequences of structural failure and the effects of an automatic suppression system. David N. Bilow et al. (4) provide structural engineers with a summary of the complex behavior of structures in fire and the simplified techniques which have been used successfully for many years to design concrete structures to resist the effects of severe fires. After the 9-11 attack on the World Trade Center, interest in the design of structures for fire greatly increased. Some engineers have promoted the use of advanced analytical models to determine fire growth within a compartment and have used finite element models of structural components to determine temperatures within a component by heat transfer analysis. Following the calculation of temperatures, the mechanical properties at various times during the
Effect of Elevated Temperatures on Performance of RCC Beams
period of the fire must be determined. S. C. Chakrabarti et al. (5) conducted an extensive test program for assessing the residual strength of concrete after fire. As per the authors, concrete doesn’t lose much of its strength up to 500°C & in fact regains 90% of lost strength up to this temperature after about a year. (The theory of fire affected concrete regaining some of its strength with time is not an established one) Concrete cubes heated beyond 750°C for 4 hours started crumbling after 2-3 days. In concluding remark, the researchers suggest that one way of assessing the residual compressive strength is by estimating the temperature of exposure (time of exposure being mostly known) & then using the tables & graphs presented by the paper. Samir Shihada et al. (32) examined the impact of polypropylene fibers on fire resistance of steel reinforced concrete beams. In order to achieve this, concrete mixtures were prepared by using different contents of polypropylene; 0, 0.45 and 0.67 kg/m3. RCC beam samples were heated in an electric furnace to a temperature of 400° for exposure times of 2.5 & 4.5 hours and tested under a central static point load on a universal loading frame. Based on the results of the study, it was concluded that the ultimate residual strengths of RC beams containing polypropylene fibers were higher than those without polypropylene fibres. Addition of fibers resulting in reduced spalling has been widely reported(2). But fibers aiding & abetting the flexural performance at elevated temperatures is a relatively new finding that needs to be further substantiated.
EXPERIMENTAL WORK Moulds & Cover Blocks In assembling the mould for use, the joints between the sections of mould were thinly coated with oil and a similar coating of oil was applied between the contact surfaces at the bottom of the mould and the base plate in order to ensure zero leakage during the filling. The interior surfaces of the assembled moulds were thinly coated with mould oil to prevent adhesion of concrete. An important factor to be considered in complying with fire-resistivity requirements is the minimum thickness of concrete cover for reinforcement. To achieve required durability, proper cover is specified in design codes as in IS 456-2000(8) & IS 1642-1989(7) but how this cover is obtained is ignored. To use any available materials like broken marble or tile pieces or pieces of aggregates as is being done at construction sites is crude and amateurish and will affect strength and durability. It is essential to provide systematic cover blocks, cast in moulds, having dimensional fidelity.
Figure 1: Cover Blocks Being Cast in 1:2 Mortar
Figure 2: Skeletons before Placing in Mould
Test Specimens The specimens for testing were individual RCC beams. Forty eight RCC beams were cast with similar crosssectional details, length and grade of concrete and identical reinforcement. All the beam specimens were 150*150*700mm size. M: 20 grade design mix concrete was used. Steel, cement & aggregates from the same batch were used for all the specimens. But testing was carried out over a span of 450mm only (@ 6*Depth) to exclude the solid cover blocks placed at
Ashok R. Mundhada & Arun D. Pophale
ends that were affecting the homogeneity of concrete. The beam section was a typical under reinforced section used for lintels, conforming to IS 456:2000(8). It contained 0.45% main tensile steel. Normal singly reinforced beams & slabs in residential buildings & apartments contain anywhere between 0.3 & 0.6% tensile steel. Greater tensile steel would have required a greater point load at centre which could have made shear an equally important player along with flexure. The aim was to simulate normal RCC beams that get subjected to both shear & flexure but fail due to flexure. All the beams tested failed due to flexure. Only the flexural crack at centre at bottom could be seen & shear cracks were conspicuous by their absence. Cement used was OPC conforming to IS 8112-1989(9). Fine aggregates consisted of natural river sand conforming to Zone III of IS383-1970. The coarse aggregates consisted of crushed hard blue granite passing through 200 mm sieve & retained on 4.75mm sieve. Potable water was used for mixing & curing. HYSD bars confirming to IS 1786:1985 (10) were used as main steel along with 5 Φ mild steel stirrups.
Figure 3: C/S of Beam (150mm X 150mm) Electric Furnace Test furnaces are the most common ones used to evaluate the fire resistance of structural elements. The furnaces’ chamber is heated either electrically or by burning liquid fuel. An electric furnace was used to heat the specimens. The maximum attainable temperature in this furnace was 1000°C. The inner depth of the furnace was 1000mm. Initially the furnace was heated to the required temperature and when the required temperature was attained the specimens were put inside with the door closing tightly so that no air could enter.
Figure 4: Electric Furnace
Figure 5: Hot Beam Being Taken out of Furnace
Testing It has to be mentioned here that to obtain pure flexural failure, two point load test is generally resorted to which is very theoretical & hardly resembles the actual site conditions where pure bending is almost non-existent. All the beam specimens were cured in a curing tank for 28 days & then tested on the 30th day. Six specimens
Effect of Elevated Temperatures on Performance of RCC Beams
were tested for the flexural strength using UTM at normal temperature, and the results were tabulated. Twelve specimens (6 specimen having 20mm nominal cover/25 mm clear cover & 6 specimens having 25mm nominal cover/30 mm clear cover) each were heated in the electrical furnace at 550°C for 1 hour and 2 hour respectively without any disturbance. Same procedure was repeated for 12 specimens each at 750°C and 950°C. After heating, specimens were kept aside for normal cooling at atmospheric temperature. Three more beams having clear cover 25mm (nominal cover 20 mm as per cl. 26.4.1 of IS: 456:2000(9) ) were heated for 2 hours at 750°C but were quenched with water after 2 hours & rapidly cooled. A set of 3 beam specimens were kept at temperatures of 550°C, 750°C & 950°C respectively and the same procedure was repeated for 1 & 2 hours time duration. Also, the clear cover to main steel was varied between 25mm & 30mm. Since 5mm ø stirrups were used, the nominal cover for the two cases was 20mm & 25mm. The beams were then placed on a Universal Testing Machine as simply supported (c/c span 450mm) & applied with a central point load. The load was increased gradually till the flexural failure occurred in a typical beam fashion. Since all other parameters were kept constant, bending stress was not worked out & performance with respect to failure load only was studied. Percentage difference between bending stress at various stages in any case would have remained identical to the % difference in failure load as all other parameters remained constant.
RESULTS & DISCUSSIONS Thermogravimetric Analysis (TGA) Table 1: Percentage Decrease in Weight at Elevated Temperature Temp (0C) room temp room temp room temp room temp 550 550 550 550 750 750 750 750 950 950 950 950
Specimen Identification C25 1hr C30 1hr C25 2hr C30 2hr C25 1hr C30 1hr C25 2hr C30 2hr C25 1hr C30 1hr C25 2hr C30 2hr C25 1hr C30 1hr C25 2hr C30 2hr
Wt. before Heating (Kg) 41 41.13 41 41.13 41 41.13 41 41.13 41 41.13 41 41.13 41 41.13 41 41.13
Wt. after Heating (Kg) 41 41.13 41 41.13 38.56 39.06 38.53 38.76 38.5 38.67 38.4 38.6 37.4 37.53 36 36.33
Wt. as % of Original 100 100 100 100 94.05 94.97 93.97 94.24 93.90 94.02 93.66 93.85 91.22 91.25 87.80 88.33
Thermogravimetric analysis consists of finding change in weight of concrete with increasing temperature. The plot is called a Thermogram. The loss of weight indicates evaporation of the chemically bound water & subsequent decomposition. A cursory glance at figure 6 reveals that the loss of weight up to 550°C was minor, at 750°C it increased & went beyond 5% but was still not alarming. At 950°C, % weight loss went beyond 10% & became alarming. The outer skin of the member became totally dehydrated & there were clear signs of delamination. The beam specimen remained intact during bending test after 24 hours, but kept decomposing after testing for 2-3 days. They had become fragile & friable. It was no coincidence that even loss of flexural strength was pronounced after 750°. Flexural Strength Table 2 below gives the values of failure load in beam test. Figure 6 below gives the values of failure load in
Ashok R. Mundhada & Arun D. Pophale
beam test as a percentage of the original value at room temperature. The decreasing trend is evident. Concrete was able to hold its fort at 550°C. But at 750°C & beyond the drop in strength was alarming. It was @ 50% at 750°C & @ 2/3rd loss at 950°C. As mentioned above, specimen heated to 950°C lost their weight by @ 10-12 % & looked fragile & friable. It could be attributed to heat reaching the core rather than getting restricted to outer 20-30 mm. It should be noted that laboratory samples had to be necessarily thin to accommodate them in a furnace. Thicker & bigger samples with same cover & same heating could be expected to perform better. In case of fire, size does matter!
Figure 6: Percentage Decrease in Failure Load Due to Increase in Temperature Table 2: Decrease in Failure Load with Rise in Temperature Temperature (°C) room temp 550 750 950
1 Hour C25 Failure Load (KN) 69.66 45.33 28.33 25
1 Hour C30 Failure Load (KN) 73.66 48 39.5 28
2 Hour C25 Failure Load (KN) 69.66 39 27.5 24.5
2 Hour C30 Failure Load (KN) 73.66 44 36.91 27.08
Effect of Water Quenching Only one batch of three beams, heated up to 750°C for two hours was cooled by water quenching after taking out from furnace. The quenching was done after about two hours of taking out the samples. The quenched samples gave better results & took 10% greater load to fail. However, one batch of three samples, heated at 950°C was allowed to cool at room temperature for three days on the trot. After 24 hours, it started degenerating rapidly & by third day, the samples had greatly decomposed. In fact, all beam samples heated up to 950°C looked pale yellow/buff colored after 24 hours but continued to degenerate after testing, became fragile & friable after a couple of days & shed their skin (cover) completely. It meant at that high temperature, heat could penetrate to the core. During subsequent cooling cycle, the reverse flow of heat towards surface resulted in extensive degeneration. This observation established beyond doubt the need to extinguish fire at the earliest & then cool the structure rapidly to minimize damage. However, most of the real life structural members are bigger, bulkier & restrained & could be expected to perform better during real time fire.
Effect of Elevated Temperatures on Performance of RCC Beams
Figure 7: Tested Sample Showing Flexural Crack
Figure 8: Tested Samples
Other Observations Sundry observations (Ref Table 3) too revealed minor damage up to 550°C & major damage only above 750°C. S. C. Chakrabarti et al.(5) too have reported the phenomenon of concrete cubes becoming fragile & friable at & above 800°C. In an actual fire affected structure, the changed color could be utilized to assess the temperature attained by the member during fire. Table 3: Comparison of Damage to Beams at Different Temperatures Observation Cracks Color Spalling Distortion Scaling
Room Temperature No Normal Grey None None No
No No change Minor No No
Moderate Blackish Grey Localized to corners Slight but insignificant No
Major Buff Corners & edges spalled Slight Yes, Delamination
CONCLUSIONS To sum up, up to 550°C, the weight loss was negligible & the flexural strength got reduced by 1/3rd. No cracking, spalling or scaling was observed up to this stage. It could be concluded that up to 550°C the fire affected structure was only mildly affected & would require rapid cooling & minor repairs. At @ 750°C, there was further drop in weight & flexural strength, cracks did appear but there was hardly any spalling or scaling. The fire affected members at this point would require rapid cooling & retrofitting after proper evaluation. Factor of safety would come down but the structure would still be serviceable. Beyond this stage, all the parameters dropped alarmingly. Weight loss at 950°C exceeded 10%, flexural strength came down by 2/3rd, major cracking, spalling & scaling could be observed. Concrete became friable during cooling. The fire affected portion at this stage would require major retrofitting or might need replacement after detailed assessment. Based on the experimental work carried out in the laboratory, the following conclusions could be drawn:
Up to 550°C the fire affected rcc member would hold its forte & would require only minor repairs.
At higher temperatures the rcc beam would lose its fidelity rapidly and at or around 750°C it would require rapid cooling & retrofitting after proper evaluation. Factor of safety would come down but the structural member would be serviceable.
At and around 950°C the fire affected member would require major retrofitting or might need replacement.
Ashok R. Mundhada & Arun D. Pophale
In a nut shell, during most short duration fires (up to a few hours) the skin (cover) would disintegrate but a visibly shaken core would remain largely unharmed. Greater cover & faster cooling through might be beneficial.
ACKNOWLEDGEMENTS The authors are grateful to Principal, Prof. Ram Meghe Institute of Technology & Research, Badnera, Amravati; & the Director, Visvesvaraya National institute of Technology, Nagpur (India) for making the library & laboratory facilities available for this work.
Phan Long T. , Carino N. J., (2000), “Fire performance of high strength concrete: Research need”, Proceedings of ASCE/SEI Structures Congress 2000, Philadelphia, USA
Kodur V. K. R., (2000), “Spalling in High Strength Concrete Exposed to Fire — Concerns, Causes, Critical Parameters and Cures", Proceedings of ASCE/SEI Structures Congress 2000, Philadelphia, USA, pp. 1-8
Faller George, (2001), “Fire Resistance Requirements for Buildings: A Performance Based Approach”, Proceedings of ASCE Structures Congress 2001, Structures — A Structural Engineering Odyssey , Section: 37, Chapter:, pp. 1-12
Bilow David N., Kamara M. E., (2008), “Fire and Concrete Structures”, Part of ASCE Structures Congress 2008: Crossing Borders, pp. 1-10
Chakrabarti S. C. et al (1994), “Residual strength in concrete after exposure to elevated temperature”, The Indian Concrete Journal, pp. 713-717
Shihada Samir& Mohammed Arafa , (2012), “Mechanical Properties of RC Beams with Polypropylene Fibers under High Temperature”, International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-1, Issue-3, pp. 194-199
BS 8110-1:1997-Structural Use of Concrete, “Part 1: Code of practice for design & construction”, British Standards, Great Britain
IS 1642: 1989, “Fire safety of buildings (General): Details of construction”, Bureau of Indian Standards, India
IS 456: 2000, “Code of practice for Plain and reinforced concrete” Bureau of Indian Standards, India
10. IS 8112: 1989, “Specifications for 43 grade OPC” Bureau of Indian Standards, India 11. IS 1786: 1985, “Specifications for HYSD steel bars & wires for concrete reinforcement” Bureau of Indian Standards, India