Fire Flame Influence on the Behavior of reinforced Concrete Beams Affected by Repeated Load

The influence and hazard of fire flame are one of the most important parameters that affecting the durability and strength of structural members. This research studied the influence of fire flame on the behavior of reinforced concrete beams affected by repeated load. Nine selfcompacted reinforced concrete beams were castellated, all have the same geometric layout (0.15x0.15x1.00) m, reinforcement details and compressive strength (50 Mpa). To estimate the effect of fire flame disaster, four temperatures were adopted (200, 300, 400 and 500) o C and two method of cooling were used (graduated and sudden). In the first cooling method, graduated, the tested beams were leaved to cool in air while in the second method, sudden, water splash was used to reduce the temperature. Eight of the tested beams were divided in to four groups, each were burned to one of the adopted temperature for about half an hour and cooled by the adopted cooling methods (one by sudden cooling and the other by graduated cooling). After burning and cooling the beams were tested under the effect of repeated load (loading – unloading) for five cycle and then up to failure. As a compared with the nonburned beam, the results indicated that the ultimate load capacity of the tested beams were reduced by (16, 23, 54 and 71)% after being burned to (200, 300, 400 and 500) o C , respectively, for a case of sudden cooling and by (8, 14, 36 and 64)% , respectively, for a case of graduated cooling. It was also found that the effect of sudden cooling was greater than that in a case of graduated cooling. Regarding the failure mode, there was a different between the non-burred beam and the other ones even that all of them had the same geometric layout, compressive strength and reinforcement details. The failure mode for all burned beams was combined shearflexure failure which was belong to the reduction in the compressive strength of the concrete due to the effect of the temperature rising , while the failure mode of the non-burned beam was flexure failure which was compatible with the preliminary design. It was also detected that the residual deflection proportion directly with the temperature, as the temperature increase to (200, 300, 400 and 500) o C the residual deflection compared with the non-burned beam increased by (32, 48, 326 and 358)% for a case of sudden cooling and by (13, 29, 303 and 332)% for a case of graduated cooling. Another effect was appear represented by the method of cooling, the results showed that the sudden cooling had more effect on the residual deflection than the graduated cooling by (15-6)% approximately. To vanish the residual deflection, numbers of cycle (loading-unloading) were required. It was found that this number increase as the temperature of burning increased and it’s also larger in a case of sudden cooling.


INTRODUCTION
Exposure to high temperatures resulting from the fires is one of the common things in concrete and steel buildings.In such case, concrete composition will suffer from self-deterioration due to the difference in the thermal expansion of its components.Fletcher et al 2007, stated that, the free water evaporates when concrete heated, and above 100 ᴏ C, approximately, there will be a releasing of water that chemically bonds in the hydrated calcium silicate.In some cases, the surface layer of concrete specimen is not able to resist the pressure of the water and steam, and spalling occurs.Shrinkage of the hydrated cement paste will accurse due to the released water if the concrete dose not spall, while both the reinforcing bars and the coarse aggregate will subject to thermal expansion.Consequently, stresses will develop in the composite material and form micro cracks through the matrix.Above approximately 400 ᴏ C the crystals calcium hydroxide begin decomposing into calcium oxide and water process reaching its highest intensity at above 535 ᴏ C. Venkatesh, K. 2014, indicated that the compressive strength of the concrete decrease slightly up to 400 ᴏ C, then it decreased rapidly when it reached about 600 ᴏ C then it began to diminish continuously as temperature increased more and more till approximately disappeared at 1000 ᴏ C.Many theoretical researches or finite element models have been conducted to study the behavior of different structural elements exposed to high temperature, Obaidat and Haddad 2016, Lakhani et.al. 2014, Neno et.al. 2013, but a very little experimental works were carried out to investigate the behavior of burred beams under the effect of repeated load, therefore an experimental program was performed to find the behavior, ultimate load and the residual deflection of self-compacted concrete beams subjected to fire flame under the effect of repeated load.

EXPERIMENTAL PROGRAM:
Three stages were included in the program of the experimental work.In the first stage, nine beams were castled and cured using self-compacted concrete of (50 Mpa) compressive strength and mix proportion as illustrated in Table 1, the properties of the consuming materials were illustrated in Table 2 up to Table 10.All the tested beams had the same geometric layout and reinforcement details (0.15x0.15) m as cross section and (1.00) m total length, Fig. 1.Burning the beams was the second stage of the experimental program, eight of the tested beams were divided into four groups each were burned to one of the adopted temperature (200, 300, 400 and 500) o C using a steel furnace that manufactured by 3mm thick plate bents like two L-shape with a capacity of two specimens, Fig. 2. The clear space around the beam was 500mm height by 400mm width and 2600mm length.These dimensions provide enough space around the beam to reach the fire flame from the sources and to ensure that the flame are not concentrated on a limited area but distributed on a wide area of the beam bottom and sides.Fire sources were designed as a network of methane burners nozzles, the nozzles were allocated, four in each side of the furnace.Two thermocouples were used, one for each beam, to monitoring the temperature.The rate of temperature increasing was adopted to be the same for all burning possess (3-5 o C /min), approximately, and after reaching the adopted temperature of 200, 300, 400 and 500 ᴏ C, the beams kept at the same temperature for half an hour.During the burning process time-deflection was measured by a dial gauge of 0.01 mm sensitivity placed at the mid-top point, Fig. 2.Then after the burned beams of a cetin group were cooled by the adopted methods ( one gradually by leaving at lab temperature and the other suddenly by using water splash).The last stage was the repeated loading test.Each beams was tested under the effect of repeated load for five cycle then after up to failure.The adopted peak load of each cycle was (2500 kN), approximately.

RESULTS AND DISCUSSION
Results discussion considered two main items.The first focus on the output data of the burning stage.In this stage, cracks were generated on all beam surfaces with an intensity increased with the increasing of burning temperature, Fig. 4 up to Fig. 11.For an individual beam the generated cracks were more distributed in the bottom surface (tension zone) due to the deflected shape produced by increasing the temperature and these cracks were mostly extended towered the side surfaces of the beam Fig. 4 up to Fig. 11.There was also an increasing in the maximum crack width with the increasing of the fire temperature to reach 0.45 mm in a case of 500 o C and sudden cooling, Table 11.The results of this stage also improve the effect of the cooling method on the intensity of the generated cracks.Sudden cooling had more effect to generate cracks due to the variation in the rate of temperature rising and rate of temperature reducing (sudden cooling) which had a damage effect on the bond between concrete composite materials.Another phenomena was appeared in this stage this is the spalling of the concrete surfaces, all the beams that burned up to 400 o C and greater in addition to the beam that burned up to 300 o C with sudden cooling were spalled, Regarding the residual deflection, there was a variation in the behavior of the tested beams during the repeated-load.It was detected that the increasing in the burning temperature cause an increasing in the residual deflection by (32, 48, 326 and 358)% as the temperature increase to (200, 300, 400 and 500) o C for a case of sudden cooling and by (13, 29, 303 and 332)% for a case of graduated cooling and number of cycle that required to vanish the residual deflection.

CONCLUSIONS:
1.As a compared with the non-burred specimen, the results indicated that the ultimate load capacity of the tested specimens were reduced by (16, 23, 54 and 71)% after being burned to (200, 300, 400 and 500) o C , respectively, for a case of sudden cooling and by (8, 14, 36 and 64)% , respectively.2. The effect of sudden cooling on the ultimate load capacity was greater than of that in a case of graduated cooling and this variance was reduced as the temperature increased.3.There was a different in the failure mode between the non-burred specimen and the other ones even that all of them had the same geometric layout, compressive strength and reinforcement details.The failure mode for all burred specimens was combined shear-flexural failure which is belong to the reduction in the compressive strength of the concrete due to the effect of the temperature, while the failure mode of the non-burred specimen was flexural failure which was compatible with the design.4. It was detected that the residual deflection proportion directly with the temperature, as the temperature increase to (200, 300, 400 and 500) o C the residual deflection compared with the non-burned beam will be increased by (32, 48, 326 and 358)% for a case of sudden cooling and by (13, 29, 303 and 332)% for a case of graduated cooling. 5. Method of cooling affected the residual deflection.The result showed that the sudden cooling had more effect on the residual deflection than the graduated by (15-6)%, approximately.6.After burning, cracks were performed on the concrete surface of the beams especially in tension zone (bottom surface), and these cracks increase in length, width and depth as the fire flame temperature increase.
7. Method of cooling had an effect on the intensity and width of the generated cracks, the sudden 8.The required number of cycles to vanish the residual deflection proportion directly with the fire temperature and sudden cooling method.

Fig. 6
up to Fig. 11.This is usually belong to the rapid varying in the interior temperature caused by sudden cooling and/or high burning temperature that expand the consuming materials of the concrete (sand and gravel) and steel reinforcement, DeHaan, John D, 2006, NFPA 921, 2004, Lentini, John J.,2006.During this stage, middle deflection of the tested beams was recorded versus the measured temperature.The trend of the deflection-temperature curves, Fig.12, showed the compatibility between the curves in a case of individual temperature (same temperature and different method of cooling) and between cases of different temperature.This improve the control of the temperature rising rate that adopted in the test.Repeated-load test was the second stage to be discussed.The results of this stage improved the effect of both temperature rise and method of cooling on the ultimate load capacity of the burned beams.Regarding the burning temperature, there was indirect proportion with the ultimate load capacity, increasing the temperature to (200, 300, 400 and 500) o C produced a reduction in ultimate load capacity by (16, 23, 54 and 71)% in a case of sudden cooling and by (8, 14, 36 and 64)% in a case of graduate cooling.While the effect of cooling method demonstrate that sudden cooling had more influence on the reduction of ultimate load capacity.Even that the preliminary design of all the tested beams was checked to be flexure failure, the failure mode of all the burned beams was combined shear-flexure mode with different degree of participation between shear and flexure failure.As the burning temperature increase the percentage of shear failure increase and shear cracks began to generate in earlier stage of loading, Fig. 13.This is belong to the decrease in the compressive strength of concrete due to the breakdown of interfacial bond which is caused by the incompatible volume change between the concrete components during heating and cooling , Georgali, B. and Tsakiridis, P. 2005, Koksal, et.al. 2011.

Figure 1 .Figure 3 .
Figure 1.Geometric layout and reinforcement details of the tested beams.

Figure 4 .Figure 5 .Figure 7 .Figure 9 .
Surface of specimen T200S after fire test.Surface of specimen T200G after fire test.Surface of specimen T300G after fire test.Surface of specimen T400G after fire test.

Figure 11 .
Figure 11.Surface of specimen T500S after fire test.

Figure 14 .
Figure 14.Load-central deflection of specimen WOF tested under repeated load

Figure 16 .
Figure 16.Load-central deflection of specimen T200G tested under repeated load

Figure 20 .
Figure 20.Load-central deflection of specimen T400G tested under repeated load

Table 1 .
Details of the adopted mix .

Table 2 .
Chemical composition of cement.*All the test were conducted by the National Center of Laboratories and Researches (Baghdad). *

Table 4 .
Physical properties of the fine aggregate.*No. Physical Properties Test Result Iraqi Specification No. 45 / 1993 1 All the test were conducted by the National Center of Laboratories and Researches (Baghdad). *

Table 5 .
Grading of the fine aggregate.

Table 6 .
Grading of the coarse aggregate.

Table 7 .
Physical properties of the coarse aggregate.*All the test were conducted by the National Center of Laboratories and Researches (Baghdad). *