J. Cent. South Univ. Technol. (2011) 18: 2185-2191

DOI: 10.1007/s11771-011-0961-6

Heat resistance and water protection effectiveness for

large single-pane fireproof glass

SHAO Quan(邵荃)¹, LI Fang(李芳)¹, CHEN Tao(陈涛)², SUN Zhan-hui(孙占辉)²

1. Civil Aviation College, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;

2. Center for Public Safety Research, Department of Engineering Physics,Tsinghua University, Beijing 100084, China

? Central South University Press and Springer-Verlag Berlin Heidelberg 2011

Abstract: The effects and heat transfer mode of water film and sprinkler system on the heat-resistant property associated with the insulation of a fireproof glass were investigated. In the experiments, fireproof glass with a size of 3 300 mm (height) × 2 200 mm (width) × 12 mm (thickness) was exposed to an oil pool fire with a power of approximately 1.4 MW. The experimental results show that the application of the water film or sprinkler system on the glass can effectively resist the intensive heat from the fire in the test due to the absorption of latent heat. The permitted period of integrity and insulation with a water film and a sprinkler system could be extended to 60 min. It should be noted that the temperature of the glass surface can be kept under 60 °C in a 60 min test. The experimental results suggest that it is feasible to substitute fireproof glass with water film for a fireproof door as long as the water film or sprinkler system can work stably and water can cover the whole surface of the fireproof glass.

Key words: water film; sprinkler system; heat-resistance; fireproof glass; fire test

1 Introduction

The demands for glazing assemblies are increasing in modern buildings for aesthetic elegance and economic reasons. However, thermal stresses will cause ordinary glass to crack when it is heated to exceed its critical temperature, resulting in the flames and smoke to spread to other parts of the building. Therefore, a special type of glass called fireproof glass is being used more and more in buildings. In many buildings, such as shopping malls and some special architecture as high atriums, fireproof glass is often installed in the entrance doors or display windows. Large panes of fireproof glass are also used in high-technology factory buildings, and the sizes of fireproof glass become larger and larger.

Single-pane and compounding fireproof glass are widely used. Single-pane fireproof glass is favored by more people and used more widely as it has the advantages of more light transmission and lighter weight compared with the compounding one. The single-pane fireproof glass can directly and efficiently prevent flame passage and smoke spreading. However, its small heat resistance may result in high temperatures on the protected side and excessive thermal radiation by the glass, which may lead to the ignition of items on the unexposed side of the glass. The refractory and water-resistant silicone begins to deform and melt in the real fire due to the spread of fire and smoke. Usually, fireproof glass is quite close to evacuation passages. As a result, the heat insulation of the glass and silicone is indispensable, especially for life safety during a fire with intensive heat radiation or heat convection. These shortcomings prevent the single-pane fireproof glass from being used in practice. In China, new codes for fire protection design of tall buildings [1] have been announced, in which the fireproof glass cannot be used as a separation for a fire subarea.

The heat insulation of fireproof glass can be improved by water film with the characteristic of large specific heat capacity resulting in cooling of the glass. THOMAS and SUNDERLAND [2] and AIHARA et al [3] used the spray method to form a water film on solid surfaces to take the heat away. The results indicated that the water film had good cooling effects. YAN et al [4-5] and WU et al [6-7] investigated the effect of a water film on insulating the fireproof glass. The evaporative latent heat for the water film was used to protect the fireproof glass from melting or breaking, and the water film could effectively resist the intensive thermal radiation in the fire test. However, the water flow of the water film was less than 0.67 L/(s·m) in the tests performed by WU et al [6-7]. As we know, the water film protection is more complex and more expensive than a sprinkler system. In the above references [2-7], the scholars focused on how to use a little water flow to protect the glass in the heat radiation conditions, and the experimental environments were not real fire environments and the thickness of the water film is too thin. This is the reason that those authors think the total heat transferred to the water film surface in these experiments was mainly caused by evaporation. However, the physical mechanism is different if the object is directly exposed to a real fire environment; the amount of water flow can affect the heat transfer mode, too. Some scholars [8-13] have launched some research work in protecting ordinary glass and fireproof glass with sprinklers and have got many useful scientific conclusions. These research works indicated that the sprinkler system could protect the fireproof glass in real fire, but just one glass in the framework was used in the tests and the glass used was for windows which were smaller than the applications of interest. The performance of refractory silicone which was filled in the gap between the glass and framework had not been the concern of authors. Available research on the heat transfer mode of sprinkler systems for protecting fireproof glass from real fire environments is scarce.

In this study, an experimental apparatus was designed to investigate the heat transfer mode of a sprinkler system and water film with a flow rate more than 2.67 L/(s·m) for protecting fireproof glass, and the performance of refractory silicone between two panes of glass in the real fire. The flow rate was nearly equal to the flow of a sprinkler system on the heat-resistance property associated with the insulation of a large fireproof glass in a real fire environment. The environment of experiment was mostly similar to the real fire environment.

2 Experimental apparatus and methods

In this study, an experimental apparatus including a fire experimental room, water film and water sprinkler system, temperature and radiation flux measurement system, and water recycling measurement system was designed to perform the experiment. This was similar to a real fire scenario for a room with glass inside a modern building, such as dining room, and retail clothing store.

2.1 Experimental environment

Figure 1 shows the sketch of the fire test building with a length of 70 m, a width of 24 m, and a height of 12 m. The floor and the walls of the building were constructed of reinforced concrete structures. The fire experiment room was in the center of the building, whose length is 5.5 m, width is 4.5 m, and height is 3.5 m. The details of the fire experimental room are shown in Fig.2.

The fireproof glass was set on one of the walls of the fire experimental room, which had the U-shape glass frame. A ventilation opening of 2 000 mm (height) ×   2 000 mm (width) was incorporated into the wall opposite to the fireproof glass. The exhaust hood of    2 000 mm (length) × 2 000 mm (width) was close to the opening. The design exhaust volume of the exhaust system was about 15 000-20 000 m³/h. The other walls of the room were constructed by firebricks. Figure 3 shows the sketch of the fireproof glass used for the fire experiment. There were two panels of fireproof glass set in the glass frame with each size of 3 300 mm (height) × 2 200 mm (width) × 12 mm (thickness). The blank between the glass and the frame was simply filled with refractory and water-resistant silicone (whose fire resistance duration is more than 4 h) to prevent the leakage of smoke and water. Thirty thermocouples, as shown in Fig.3, were set on both surfaces of the fireproof glass and one radiation flux measurement was set on the outer surface of the fireproof glass. There were ten thermocouples in the fire experimental room to measure the smoke gas temperature, as shown by the interior view of the fire experimental room in Fig.4.

Fig.1 Sketch of experimental environment

2.2 Water supply system

The water supply system in this study included a water tank, a variable-frequency hydraulic pump, sprinkler system, water film system, and pipes. The sketch of the water system is shown in Fig.5. According to the requirements of the experiments, the capacity of the water tank was 30 m³, and the variable-frequency hydraulic pump was used for water supply to guarantee that the pressure of the water in the pipes was steady. There were flow meters and manometers that were set on the pipes to measure the water mass flow rate.

Fig.2 Sketch of fire experimental room

Fig.3 Sketch of fireproof glass

Fig.4 Sketch of thermocouples in fire experimental room: (a) Side view; (b) Vertical view

Fig.5 Sketch of water system

The water film system was installed at the top of the fireproof glass. The diameter of the pipeline for the water film system was 0.050 8 m with a length of 4.5 m. The pipeline was of a double casing type. The inner pipe is a PVC pipe connected to the water supply pipe, which is drilled to form hole. From the hole, water can flow to the outer pipe of the pipeline. A triangle flume facing the fireproof glass in outer pipe was used to make water flow along the fireproof glass. The design was made to make water flow evenly and form a stable water film.

Based on tests of various spray types using different nozzles, the fast responding sprinkler was found to be more suitable [14]. Thus, a type of fast responding sprinkler (model ZSTBZ-15) was selected to perform the experiment. Some experiments were also conducted for different numbers of sprinklers and various installation positions. Through these experiments, the best installation layout was determined to be with three sprinklers equally spaced along the length of the glass, 200 mm below the ceiling, and 150 mm from the glass. The pitch between nozzles was 1 100 mm, and the middle nozzle was set opposite to the middle of glass curtain wall.

2.3 Recycled water measurement system

The recycled water measurement system is important for validation of the effects of the water protection to the fireproof glass. With the recycled water measurement system, the water amount for cooling the fireproof glass was collected in the trough. The dimension of the recycled water trough was 5 000 mm (length) × 400 mm (width) × 400 mm (height). Two drain pipelines were connected at the bottom of the trough with each drain pipeline connected to a high precision flow meter and a pump. The outer surface of the water trough was covered with fireproof cotton. Some high-sensitivity thermal resistances were set on different positions of the water trough for measuring the temperature of water in the trough. A number of tests with the water trough were done before this study to ensure that the water flowing down the surface of the glass could be fully collected in the trough and the water in the trough could be drained by pumps quickly.

3 Results and discussion

A diesel oil tank with the area of 1.0 m² is used as the fire source. The fire power of the diesel oil tank is 1.3-1.7 MW at steady-state burning when the area of the oil tank is 1 m² according to Ref.[15]. In our experiments, the air temperature of environment is 24.3 °C. In the fire tests for fireproof glass without water, the fire develops rapidly after ignition and reaches a stable state in 30 min when the temperatures of the thermocouples are steady. Figure 6 shows the temperature on the outer surface of fireproof glass in the experiments without water. The temperature of the center of outer surface represents the average value of all the measured temperatures in the center of outer surface, and the temperature of the upper of outer surface represents the average value of all the measured temperatures in the upper of outer surface, as shown in Fig.6.

The height of the smoke layer above floor is 1.9 m, and the height of the flame is 2.5-3.0 m. The average temperature of the smoke layer is 580 °C. The average temperature of lower layer near the opening is 227 °C. The average temperature of lower layer between the fire source and the fireproof glass is 413 °C. The temperature of the upper outer surface of the fireproof glass is 280 °C, which is higher than that of the other zones of the outer surface due to the heat transfer from the hot smoke layer. The maximum of radiation flux at the center of outer surface of fireproof glass is 4.26 kW/m², as shown in Fig.7. The results in the experiment show that the mean and standard deviation of radiation flux at the center of the outer surface is (3.01±0.62) kW/m².

Fig.6 Temperature on outer surface of fireproof glass in experiments without water

Fig.7 Radiation flux at center of outer surface of fireproof glass in experiments without water

The entire experiment continues for 2 h, and the total consumption of diesel oil in this test is 400 L. By using Eq.(1), the heat release rate  is 1.37 MW, when calorific value of diesel ΔH is 42 000 kJ/kg and the proportion of diesel is 0.838 kg/L if the combustion efficiency  is set to be 0.7:

                                                             (1)

From the observation of the test, it is found that in the whole process the glass surface seems to remain the same, and the silicone between fireproof glass and the glass frame begins to deform and melt after 45 min, due to the spread of smoke, as shown in Fig.8.

According to the experiment parameters, the heat rate of the whole fireproof glass reception was calculated to be about 135.4 kW by using FDS.

Fig.8 Status of fireproof silicone in fire test without water

In the fire tests for fireproof glass with the water film, the temperature of the water before the fire test is 14.8 °C and the pressure of the supply water is 0.12 MPa. During experiments, the water flow for the water film is opened as soon as ignition occurs. The fire develops rapidly after ignition and reaches a stable state in 8 min when the temperature of the thermocouples is steady. Figure 9 shows the average temperatures on the outer surface of fireproof glass and average water temperatures in the recycled water trough during the experiments with water film. The temperatures of the glass will remain stable as long as the water film is applied to the surface.

Fig.9 Temperature on outer surface of fireproof glass and in recycled water trough in experiments with water film

The height of the smoke layer and floor is 1.8 m. The height of the flame is 2.0-3.0 m. The average temperature of the smoke layer is 578 °C. The average temperature of lower layer near the opening is 215 °C. The average temperature of lower layer between the fire source and the fireproof glass is 407 °C. The temperature of the upper outer surface of fireproof glass is 38 °C. The maximum radiation heat flux at the center of the outer surface is 0.035 kW/m². The mean and standard deviation of the heat flux is (0.021±0.009) kW/m². The flow rate of the water film is 4.944 L/s, and the flow rate of the recycled water is 4.942 L/s. This shows that nearly all the water from the water film enters in the recycled water trough and is drained out. Assuming that the loss of water evaporated is 0.002 L/s and the gas temperature is the maximum room temperature, the heat removed by water evaporation is 4.52 kW by Eq.(2):

                                                                     (2)

When the evaporation latent heat of water, e_w, is         2 257.2 kJ/kg, water evaporation rate is 0.002 L/s. The average temperature of the water is 20.3 °C when the water just enters the trough. Using the water temperature rise, the heat rate taken from the fireproof glass is calculated to be 114.2 kW by Eq.(3):

                                                      (3)

where T_r is expressed as temperature of the collected water (K); T_o is the initial temperature (K); c_w is the specific heat of water; m_r is the water mass flow rate (kg/s); Q_a is the absorption heat flow rate (kW). When T_o is 287.8 K and T_r is 293.3 K, the specific heat of water c_w is 4.2 kJ/(kg·°C), and water mass flow rate m_r is     4.942 L/s. The entire experiment with water film continues for 1 h. The total consumption of diesel oil is 204 L, and the heat release rate is 1.40 MW. From the observation of the experiment, it is found that in the whole process the glass surface seems to remain the same, the silicone between fireproof glass and the glass frame is not burnt, and the leakage of smoke and water is not seen.

In the fire tests for fireproof glass with the sprinkler system, the temperature of the water before the fire test is 15.1 °C and the pressure of the sprinkler system is   0.12 MPa. During the experiments, the three fast responding sprinklers open at 2 min by the breaking of the glass bubble on the sprinkler. The fire develops rapidly after ignition and reaches a stable state in 14 min when the temperatures of the thermocouples are steady. Figure 10 shows the average temperatures on the outer surface of the fireproof glass and the average water temperature in the collected water trough during the experiments with a sprinkler system. The temperatures of the glass remain stable with the sprinkler system.

The distance between the smoke layer and floor is 1.68 m, and the height of the flame is 2.0-3.0 m. The average temperature of the smoke layer is 547 °C, and the temperature of upper outer surface of fireproof glass is 53 °C. The maximum radiation flux at the center of the outer surface of fireproof glass is 0.062 kW/m². The mean and standard deviation of the heat flux is (0.034±0.017) kW/m². The flow rate of the water in the sprinkler system is 4.382 L/s, and the flow rate of the collected water is 2.811 L/s. This shows that nearly two-third of the water amount from the sprinkler system enters the recycled water trough and is drained out. As water splashes, the other one-third of the water is not used directly to protect the glass but decreases the temperature of smoke in the experimental room. The average temperature of the water is 23.1 °C when the water just enters the trough. Using the recycled water, the heat rate taken from the fireproof glass is calculated to be 94.5 kW by Eq.(3). When T_o is 288.1 K and T_r is 296.1 K, the specific heat of water c_w is 4.2 kJ/(kg·°C), and the water mass flow rate m_r is 2.811 L/s. The entire experiment with sprinkler system continues for 1 h. The total consumption of diesel oil is 205 L, and the heat release rate is 1.40 MW. From the observation of the test, it is found that in the whole process the glass surface seems to remain the same, the silicone between fireproof glass and the glass frame is not burnt, and the leakage of smoke and water is not seen, as shown in Fig.11.

Fig.10 Temperature on outer surface of fireproof glass and in recycled water trough in experiments with sprinkler system

Fig.11 Status of fireproof silicone in fire test with sprinkler system

Table 1 gives the comparisons among the experiments without water, with a water film, and with a sprinkler system. It is shown from the upper outer surface temperature that in the fire experiments, water can effectively resist the intensive heat from a fire in the test mainly due to the absorption of latent heat which takes away about 100 kW from the fireproof glass. The effect of the water film and the sprinkler system for protecting the glass in this experiment is different from other experiments [2-6] because the water flow rate is larger and the water film is thicker. In addition, the experiment environments are different including air temperature, humidity, flow rate of the smoke, etc. Water film protects the fireproof glass more effectively compared with the sprinkler system due to the water flowing directly along the fireproof glass. This is demonstrated by comparing the temperature data on the upper outer surface, the maximum radiation flux at the center of the outer surface, and the heat rate taken away by water since the heat release rate of fire source is similar. From the experiments, the temperature and the maximum radiation flux of the outer surface of the fireproof glass are below 60 °C and 0.1 kW/m², respectively. It is suggested that fireproof glass with water film or sprinkler system could be used as a substitute for a fireproof door. In this application, the water film or sprinkler system must operate continuously, and the water must cover the entire surface of the fireproof glass.

Table 1 Comparisons among experiments without water, with water film and with sprinkler system

4 Conclusions

1) Water in the experiments with a water film and a sprinkler system can effectively resist the intensive heat from fire in the test mainly due to the absorption of latent heat in the practical application. The temperature and the maximum radiation flux of the outer surface of the fireproof glass are controlled under 60 °C and 0.1 kW/m², for a 60 min test.

2) The water film and sprinkler system can be used to protect the fireproof glass from melting or breaking in the fire.

3) The water film and sprinkler system can be used to protect the fireproof silicone which is filled in the gap between the glass from melting or deforming in the fire.

References

[1] GB 50045—95. Code for fire protection design of tall buildings [S]. Beijing: China Planning Press, 2005. (in Chinese)

[2] Thomas W C, Sunderland J E. Heat transfer between a plane surface and air containing suspended water droplet [J]. Ind Eng Chem Fundam, 1970, 9(3): 369-374.

[3] Aihara T, Taga M, Haraguchi T. Heat transfer from a uniform heat flux wedge in air-water mist flows [J]. Int J Heat Mass Transfer, 1979, 22: 51-60.

[4] Yan W M, Song C Y. Numerical study of liquid film cooling in a turbulent gas stream [J]. Int J Heat Mass Transfer, 1993, 36: 3877- 3886.

[5] Yan W M. Convective heat and mass transfer from a falling film to a laminar gas stream [J]. Warme-und Stoffubertragung, 1993, 29: 79-87.

[6] Wu C W, Lin T H, Lei M Y, Chung T H, Huang C C, Chiang W T. Fire test on a non-heat-resistant fireproof glass with down- flowing water film [J]. Fire Safety Science, 2005, 8: 327-338.

[7] Wu C W, Lin T H, Lei M Y, Chung T H, Huang C C, Chiang W T W T. Fire resistance characteristics of a glass plane with down- flowing water film [J]. Journal of the Chinese Institute of Engineers, 2008, 31(5): 737-744.

[8] Wu C W, Lin T H. Full-scale evaluations on heat resistance of glass panes incorporated with water film or sprinkler in a room fire [J]. Building and Environment, 2007, 42: 3277-3284.

[9] Richardson J K, Oleszkiewicz I. Fire tests on window assemblies protected by automatic sprinklers [J]. Fire Technology, 1987, 23(2): 115-132.

[10] Beason D. Fire endurance of sprinklered glass walls [J]. Fire Journal, 1986, 7: 43-45.

[11] Kim A, Taber B, Lougheed G. Sprinkler protection of exterior glazing [J]. Fire Technology, 1998, 34(2): 116-138.

[12] Kim A, Lougheed G. Fire protection of windows using sprinklers [J]. Construction Technology Update, 1997, 12: 4-10.

[13] Kim A, Lougheed G. The Protection of glazing systems with dedicated sprinklers [J]. Journal of Fire Protection Engineering, 1990, 2(2): 49-59.

[14] NI Zhen-chu. Experimental study on sprinkler protection of single- pane fire-resistant glass [D]. Hefei: University of Science and Technology of China, 2005. (in Chinese)

[15] Shu Xue-ming. Experimental study of single-pane fire-resistant glass performance with sprinkler protection [J]. Engineering Science, 2006, 8(6): 64-68. (in Chinese)

(Edited by YANG Bing)

Foundation item: Project(2006BAK06B02) supported by the National Science and Technology Pillar Program during the Eleven Five-Year Plans of China; Project(70701019) supported by the National Natural Science Foundation of China

Received date: 2010-09-25; Accepted date: 2011-04-29

Corresponding author: SHAO Quan, PhD; Tel: +86-25-84896260; E-mail: shaoquan@nuaa.edu.cn