Selasa, 03 Februari 2009

Strength Recovery of Fired Concrete

Strength Recovery of Fired Concrete

Amir Partowiyatmo and Sudarmadi
UPT – Laboratoria Uji Konstruksi - BPPT
Kawasan Puspiptek Serpong – Tangerang 15314
Telp. : 021-7560930 / Fax : 021-7560903
E-mails : amir@luk.or.id

Abstract – Because of the influence of elevated temperature concrete convert into cement again, hence there is possibility to recover the deteriorated internal structure and decreasing strength of concrete by rehydration process. In this research, some experiments was performed by preparing different type of specimens. Fire accident on the concrete structure is simulated by heating the specimens in the furnace. After cooling all heated-specimens taken out from the furnace and then recured under moisture or water. For monitoring the strength recovery development the following measurements on the specimens are conducted. The mortar specimens tested by using of flexure and compressive mode. In the other side strength recovery development on cylinder, coloumn and beam specimens periodically investigated by mean of combining rebound hammer and ultrasonic pulse velocity method. Especially for cylinder specimens also mechanicaly tested in order to evaluate the gaining of actual compressive strength. This study include also microstructure analysis using apparatus of scanning electron microscope.
Keywords – concrete, fire, strength reduction, deterioration, rehydration, strength recovery, restoration.


1. Introduction

Concrete is basically made from the mixture of cement, water and aggregates. During the hydration process, both two essential chemical compounds in cement grains C3S and C2S react with H2O produce CSH and Ca(OH)2. CSH act as strength carrier the hardened cement paste as well as the concrete and Ca(OH)2 as corrosion protection for the reinforced bar. Due to the hydration products bond firmly to and always cover the unreacted cement grains, so the rate of hydration will decrease countinously. Consequently in the internal structure of hardened cement paste as well as concrete there are remains an appreciable amount of unhydrated cement even after a long year.

Because concrete possess the physical properties of noncombustible as well as high fire-resistance material, hence concrete is most commonly used as building materials. Nevertheless, by the influence of elevated temperature during fire accident in concrete will occur a number of complex physical and chemical reactions. They occur more particularly in the hardened cement paste, but also in the aggregates.

The reactions initiated during heating of concrete can be observed with the aid of differential thermal analysis (DTA). In the hardened cement paste until the temperature of 300 o C water from the large and the finer pores is liberated, Ca(OH)2 decomposed into CaO and H2O at temperature between 450 o C and 550 o C. In the temperature range of 400 to 600 o C the pore system desiccated completely and followed by the destruction of CSH gels. Decomposition of CSH generally occur at 600 to 700 o C to form b-C2S. Whereas in the aggregate - depend on the type of stone concerned - at 573 o C a ® b inversion of quartz (SiO2) take place and in the range of 600 - 900 o C the limestone (CaCO3) begins undergo decarbonation. When during fire accident the temperature of CSH decomposition reached, it means concrete convert into cement again.

The residual strength for dense concrete after cooling varies depending on the maximum temperature attained, mix proportions and conditions of loading during fire. For temperature up to 300 o C, the residual strength of concrete is usually not severely reduced. Temperature greater than 500 o C can reduce the compressive strength of concrete to only a small fraction of its original value, and such concrete is unlikely to possess any useful structural strength.
In an actual fire, spatial temperature can reach greater than 900 o C. However, because of heat transfer characteristic in a member, only the temperature of the outside layers is drastically increased, and so the temperature of internal concrete may be comparatively low. So that, related to the temperature gradient which depend on the severity of a fire, the strength reduction due to the high temperature effects will occur in the outer 3 – 5 cm of the concrete member. The general qualitative assessment method gives a simple way to indicate the degree of damage and likely repair of individual members or overall structure.

In general, the damaged concrete may be repaired by breaking away deteriorated layer and replacing them properly by new concrete or shotcrete. It is essential that a repair must restore any loss of strength, maintain durability and fire protection. Consequently repairability is the ability to do this, with respect to technical and economical constraints.
Concerning above consideration, scope of the research activity is governed to find out the prospective restoration method of fired concrete instead of chipped out technique. So that restoration method by the autogenously healing is cost constraint in consideration. It is expected that strength recovery of fired concrete method by rehydration process in the future is possible to be applied and therefore should be developed.

2. Test Procedure

The following investigation methods were employed for the observation the strength recovery of fired concrete by autogenously healing. To simplify the test method, in the first step of experiment using a series of mortar prism specimens. After that, in order the concept of the strength recovery is technically applicable and cost constraint in consideration, so in the next experiment needed other series of real concrete specimens with the different geometry and dimension. The description of all specimens is listed in Table 1 and 2.

Table 1: Test parameters of mortar prism specimen
Dimension : 40 x 40 x 160 mm
Cement : Portland cement Type I
W/C- ratio : 0.5
Sand : 0/2 mm
Curing : In closed chamber, 100 % r.h., 64 days
Heating simulation : At temperature of 400, 500, and 700 o C in the electrical furnace
Rehydration process : up to 240 days under water

Table 2: Mix design of concrete

* Portland cement type I
Geometry and dimension of concrete specimen :
Cylinder : 15 x 30 cm
Column : 30 x 40 x 1500 cm
Beam : 30 x 40 x 3500 cm (only for w/c = 0.5)
Curing : 28 days in wet condition, after that all specimens cover by plastic as along as 180 days until beginning for heating. For recording the temperature-time development during heating in the column and beam specimens inserted thermocouples in the different depth of 2.5; 7.5; 10; and 15 cm.

The mortar prism specimens are divided in three series. Each series contains 24 specimens heated homogenously in electrical furnace at the maximum temperature of 400; 500; and 700 o C respectively. After cooling the specimens taken out from the furnace, then placed under water until determined days of 3, 7, 14, 28, 90, 180 and 240, in order to the rehydration process will take place.

The presence of strength recovery development on the mortar prisms monitored by the sequence of flexure and compressive test. The strength recovery development is represented by increasing value of flexure and compression strength relative to the value of their normal and heated specimens. For understanding and to be convinced of occuring the strength recovery on fired concrete, in the research activity include also microstructure analysis using apparatus of scanning electron microscope (SEM).

In the next experiment in order to simulate the actual fire, the concrete specimen of cylinder and column heated together in the burner furnace for the time of 90 minutes, whereas the beam specimen heated separately as long as for 50 minutes. They expected that in the both heating sequence the temperature of the surface of specimens can reach 900 o C. The sequence of heating is performed according to the standard fire curve of ISO 834. Furthermore, after cooling all heated specimens taken out from the furnace and then recured by water–sprayed or under water until determined day.

Strength recovery development of concrete specimens periodically monitored by mean of combining rebound hammer test and ultrasonic pulse velocity method. Finally, especially for cylinder specimens are mechanically tested in order to evaluate the gaining of actual compressive strength.

3. Result of Experiments and Discussion

The original flexural and compressive strength of mortar are 7.6 N/mm2 and 51.7 N/mm2. At the temperature of 400 o C, 500 o C, and 700 o C the residual flexural strength of mortar are 81.4%, 73.2 %, and 24.9 % respectively of the original value. Whereas its compressive strength at same temperatures are 89.3 %, 72.5 %, and 35.9 % of the original value.

Figure 1,2 and 3 show the graphically strength recovery development on the mortar specimens that firstly heated at the temperatures of 400 o C, 500 o C, and 700 o C and then recured under water. After heating at the temperature of 400 o C followed by recurring (Fig.1), the strength recovery of mortar is no significant increasing, because of at those temperature no presence an appreciable of additional hydration of unhydrated cement grains and besides rehydration of b-C2S.

The significance of the strength recovery showed by the mortar specimens which heated at 500 o C and 700 o C (Fig. 2 and 3). After recuring under water for 30 days increasing both the flexural and compressive strength revert to the original values. By further recuring the development of strength recovery is relative constant, due to the decreasing of rate of hydration.
At 400 o C and 500 o C the curve of the flexural and compressive strength development are closed to the other. But for 700 o C both curve are spread out, the value of recovery of flexural strength is slightly lower than the compressive strength. The presence of significantly strength recovery caused by the additional hydration of unhydrated cement grains as well as the rehydration of b-C2S include the reformation of Ca(OH)2 as result the reaction of CaO and H2O. By those temperatures the envelop of CSH gels destroyed and convert to b-C2S, so that the unhydrated cement cores unclosed tightly.
By microstructure analysis using SEM, the presence of autogenous healing process in the 700 o C - heated specimen is revealed by the rebuilding of deteriorated internal structure (Fig. 4 and 5), forward to the sound microstructure of normal mortar (Fig. 6).


______ Flexural Strength
--------- Compressive Strength
Figure 1. Curve of Strength Recovery Development of 400 o C- heated mortar.








_____ Flexural Strength
------- Compressive Strength
Figure 2. Curve of Strength Recovery Development of 500 o C- heated mortar.
_____ Flexural Strength
------- Compressive Strength
Figure 3. Curve of Strength Recovery Development of 700 o C- heated mortar.

The normal 28-days age compressive strength of concrete measured from cylinders for w/c ratios of 0.3, 0.4, 0.5, 0.6, and 0.7 are 508, 444, 440, 318, and 243 kg/cm2 respectively. (see Fig. 13). At the time of heating of column specimens and the furnace room temperature reached 1000oC, temperature on the concrete surface was 835oC, however in the depths of 7.5 cm and 10 cm the temperatures were only 150oC and 100oC respectively. In the depth of 2.5 cm, that was the position of reinforcement, the temperature was quite high, i.e. 650oC. (See Fig. 7).

Figure 4. Deteriorated Microstructure of 700 o C- heated mortar.


Figure 5. Rebuilding of Microstructure of 700 o C-heated mortar after autogenous healing


Figure 6. Sound microstructure of normal mortar

Figure 7. Temperature – time curve for column
The range of ultrasonic pulse velocity in concrete column specimens at normal condition was 4 – 4.3 km/sec. After heating it decreased to 2 – 2.6 km/sec. (55% of normal) and after recuring as long as 60 days it increased again to the range of 3.2 – 3.8 km/sec. (80% of normal) (See Fig. 8). The result of measurement by rebound hammer test can be seen in Fig. 9. It is shown that the concrete residual strength after heating is about 60% from normal condition, namely decreasing from the range of 450 – 600 kg/cm2 to the range of 250 – 400 kg/cm2. Then, it can be seen that the increase of strength after recuring is different among the w/c ratios. Concrete with higher w/c ratio show greater recovery of strength (90% of normal) compared to concrete with lower w/c ratio (60% of normal).

Figure 8. Change of ultrasonic pulse velocity in column specimens.
Figure 9. Change of compressive strength for column specimens from rebound hammer test.

In Fig. 10 it is depicted the change of ultrasonic pulse velocity of beam specimen of normal condition, after heating (i.e. about 75% of normal), and after recuring (i.e. approximately 100%). The same trend is shown in the result of rebound hammer test. (See Fig. 11). The numbers in Fig. 10 and 11 show the location of measurement, i.e. 1: left side, 2: center, and 3: right side.
The rebound hammer test results always show the occurrence of concrete strength increase. It means that the increase of ultrasonic pulse velocity show the increase of concrete strength, meaning the concrete experiences rehydration.

Figure 10. Change of ultrasonic pulse velocity in beam specimen


Figure 11. Change of compressive strength for beam specimen from rebound hammer test.

Comparing the pair of figures, Fig. 8, 9 and Fig. 10, 11 it seems that there is an inconsistency of characteristic of concrete. In Fig. 10 and 11 both ultrasonic pulse and rebound hammer test show the approximately full recovery, but different trend is shown in Fig. 8 and 9. To understand the background of this phenomenon, an investigation on microstructure of concrete by using non-conventional methods of georadar and CT scan is being performed.
The results of monitoring on the change of ultrasonic pulse velocity in cylinder specimens are graphed in Fig. 12. The ultrasonic pulse velocity drops drastically after heating, then increases significantly in the first week, but after that the increase is relatively low. After submerging in water for 78 days, the ultrasonic pulse velocity comes to normal again. The results of compression test of cylinder specimens is shown in Fig. 13. The residual strength of concrete with w/c ratios of 0.3 – 0.6 after heating is about 44% of normal condition. For concrete with w/c ratio of 0.7 is about 66% of normal. After water submerging, the concrete with w/c ratios of 0.6 and 0.7 experience the strength recovery to about 80% of normal, while the concrete with w/c ratios of 0.3 – 0.5 experience the strength recovery to 56 – 66% of normal. The trend is the same as in column. It seems that in concrete with lower w/c the damage due to heating is more severe than in concrete with higher w/c. As a result, to recover the damage in concrete with lower w/c is more difficult.


Figure 12. Development of ultrasonic pulse velocity for cylinder specimens.

Figure 13. Strength of cylinder specimens from compressive strength test at different conditions.

From Fig. 12 and 13 it can be seen that the ultrasonic pulse velocity in cylinder specimen goes back to the value as normal condition, but the recovery of concrete strength does not come back to initial values. This is because, as shown by Fig. 5, there remains microcracks that in the case of submerged cylinder can be water-filled, which do not have effect on strength but enough to pass the ultrasonic pulse.

As addition of information, it is shown the shape of furnace for concrete testing and the activity of measurement in Fig. 14 and 15.


Figure 14. Furnace for fire simulation test of concrete specimens. (owned by Puslit Pemukiman Bandung)

Figure 15. Activity of ultrasonic pulse velocity measurement on column specimen.

4. Conclusion

1. After heating the samples up to temperature of 400 o C and followed by recurring is no presence of strength recovery.
2. On the other hand, significantly strength recovery revealed by heating of mortar samples at the temperature higher than 500 o C.
3. SEM investigations show the rebuilding of deteriorated microstructures of heated mortar, which the change forwards into the sound microstructure of normal mortar.
4. The degree of degradation of concrete owing to fire and the gain of strength recovery is dependent on w/c of the mix of concrete.
5. There is a possibility to repair fired concrete by recovery method and more research in this issue should be conducted.


References :

[ 1 ] V. Lach : Uber die Rehydratation von Portlandzement, Zement-Kalk-Gips, Heft 2 Februar, pp 57 –61, 1970
[ 2 ] J. Piasta, Z. Sawicz, L. Rudzinski : Changes in the structure of hardened cement paste due to high temperature, Materiaux et Contruction, Vol. 17 No. 100, pp 291 –296, 1984
[ 3 ] A.M. Neville : Properties of concrete. ELBS, Pitman Publishing, pp 10 – 22, 1978
[ 4 ] W. Richartz : Uber die Gefuge- und Festigkaeitsentwicklung des Zementsteins, Beton Technische Bericht, pp 67 – 83, (1969)
[ 5 ] W.M. Lin; T.D. Lin; L.J. Powers-Couche : Microstructure of Fire-Damaged Concrete, ACI Materials Journal,V. 93. No. 3, May-June 1996
[ 6 ] K.O. Kjellsen, R.J. Detwiler, O.E. Gjorv : Development of Microstructures in Plain Cement Pastes Hydrated at Different Temperatures, Cement and Concrete Research, Vol. 21, pp 179-189, 1991
[ 7 ] A. Partowiyatmo : Ermittlung des Schaedigungsgrades von verbundwerkstoffen nach Waermeeinwirkung , Disertation – Univ. Bremen, 1996
[ 8 ] Laporan Akhir : Evaluasi degradasi struktur beton dengan teknik georadar, Kegiatan DIP- BPPT, 2001