An investigation into waterborne intumescent coating with different fillers for steel application

K. Md Nasir (Department of Civil Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia)
N.H. Ramli Sulong (Department of Civil Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia)
M.R. Johan (Department of Mechanical Engineering, Centre of Advanced Materials, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia)
A.M. Afifi (Department of Mechanical Engineering, Centre of Advanced Materials, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia)

Pigment & Resin Technology

ISSN: 0369-9420

Publication date: 5 March 2018

Abstract

Purpose

This study aims to discuss the modification and/or improvement of intumescent coating system by incorporating waterborne resin with an appropriate combination of flame-retardant additives and four different fillers, namely, TiO2, Al(OH)3, Mg(OH)2 and CaCO3.

Design/methodology/approach

Coating mixtures are characterized using the Bunsen burner, thermogravimetric analysis, limiting oxygen index, scanning electron microscope, static immersion bath, Fourier transform infrared and adhesion tester.

Findings

Results show that the combination of coating with CaCO3 filler significantly improved fire protection performance because of its thick char layer and the equilibrium temperature being 264°C. Char layer showed a uniform dense foam structure on micrograph and this formulation had adhesion strength of 2.13 MPa, which indicates effectiveness of the interface adhesion on substrate. Conversely, the combination of coating with Al(OH)3 exhibited highest oxygen index of 35 per cent, which resulted in excellent flammability resistance.

Research limitations/implications

This paper discusses only the effect of mineral fillers on properties of intumescent coatings.

Practical implications

In the modern design of building infrastructure, fire safety is significant for the protection of human life and assets. The application of intumescent coating in buildings is currently practiced because of its effect on material flammability during a fire.

Originality/value

The analysis method to evaluate the performance of water-borne resin with different fillers is formulated, and it could be applied in all kinds of coatings and mixtures to be used as an effective fire protection system for steel constructions.

Keywords

Citation

Md Nasir, K., Ramli Sulong, N., Johan, M. and Afifi, A. (2018), "An investigation into waterborne intumescent coating with different fillers for steel application", Pigment & Resin Technology, Vol. 47 No. 2, pp. 142-153. https://doi.org/10.1108/PRT-09-2016-0089

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


Introduction

Intumescent coatings are insulating systems designed to protect and maintain substrate integrity during a fire, thereby preventing structural collapse. This specialized coating chemically reacts to protect buildings and structures. This coating has wide applications in steel structures as this construction material starts to lose its mechanical properties at temperatures above 400°C (Jimenez et al., 2006a). The coating compositions will react in a fire at temperatures between 200°C and 250°C to form a thermally insulating carbonaceous char that protects the underlying substrate and hence prolongs its load-bearing capacity. The coating comprises three active elements, usually an acid source (ammonium polyphosphate, APP), a carbon source (pentaerythritol, PER) and a blowing agent (melamine, MEL) (Jimenez et al., 2006a). The three main flame-retardant additives are bound together with a binder, which then reacts together at higher temperatures. The result of the reaction initiates a series of chemical and physical processes, leading to an expanded multicellular layer, which acts as a thermal barrier that effectively protects the structural elements against a rapid increase in temperature, thereby preventing the building from collapsing under severe fire conditions (Otáhal et al., 2011).

The binder acts as a “vehicle” film-forming component, which has a good expanding effect and char structure that could control the migration of flame-retardant additives and access of corrosive substances (Weil and Levchik, 2004). Binders can be solvent-borne or water-borne resin, depending on the application. The use of synthetic solvent-borne resins can have an adverse effect because intumescent coatings contain synthetic polymers that release acidic degradation products, which can cause water pollution if not manned properly. Considering the ecological issues, the use of solvent-free or water-borne coatings will likely increase in demand, and the production of intrinsically fire-retardant coatings is therefore highly promising. Solvent-based coatings emit volatile organic compounds (VOCs) during application and trap solvents during the drying process for weeks afterward (Guo et al., 1999). In solvent-based coatings, the lacquer thinner is usually used to thin and reduce the viscosity of coatings, including for cleaning process. VOCs represent a serious hazard. As such, VOCs should be used with care and should always be disposed of properly. Strict worldwide limits on VOCs in paints are regularly tightened. In any application, alternative non-solvent products pose less risk to the environment, the public and the users themselves. Furthermore, the utilization of water as the reaction medium is also another derivation toward greener chemical reactions (Ravichandran et al., 2011).

In this study, a water-borne vinyl acetate copolymer (VAC) emulsion is used to minimize smoke and toxic fume emission without compromising the quality and effectiveness of the intumescent coating in fire protection. The influence of the binder in water-borne coatings was studied by Wang and Yang (2010). The investigation of the fire protection and the physical and mechanical properties of intumescent coatings is among the tests that will be used to optimize the combinations of water-borne resin, flame-retardant additives and inorganic flame-retardant fillers.

The incorporation of inorganic flame-retardant filler can play a significant role in modifying the final characteristics of intumescent coatings, including its resistance to ignition and the extent and nature of smoke and toxic gas emissions. Among the 21 most important fillers, calcium carbonate, CaCO3, holds the largest market volume. CaCO3 is mainly used in plastics. Using flame-retardant fillers such as aluminum hydroxide, Al(OH)3 and magnesium hydroxide, Mg(OH)2 presented better insulation properties and further decreased the heat release rate because of their low flame-retarding efficiency (Huang et al., 2006; Yeh et al., 1995). The performance of the intumescent coating system depends on the appropriate formulation and compatibility of the flame-retardant filler with the binder matrix (Cross et al., 2003; Tai and Li, 2001).

In this paper, the fire protection performance and the physical and mechanical properties of flame-retardant fillers were evaluated by identifying the appropriate potential source of fillers and creating an inter cross-linking network between water-borne intumescent coatings and flame-retardant fillers.

Experimental work

Materials

In this research, VAC with a viscosity (Brookfield RVT 3/10) of 4,000 centipoises was used as a binder. This vinyl acetate acrylic binder is soft with a medium to coarse particle size and possesses good stability and pigment-binding properties. VAC appears as a milky-white emulsion with 55 per cent solid content. A particle size (median) of 0.3 µm was used. Flame-retardant additives were supplied by International Chemical Ltd, China. Titanium dioxide, TiO2 (R706); aluminum hydroxide, Al(OH)3; magnesium hydroxide, Mg(OH)2; and calcium carbonate, CaCO3 were supplied by Scientific Group Sdn. Bhd., Malaysia.

Methods

The flame-retardant additives contain APP/MEL/PER blend at a weight ratio of 2:1:1 and VAC with content of 53 per cent by mass were prepared using a high-speed disperse mixer. The mixture of flame-retardant water-borne binder (FRWB) was then added to 10 per cent of four different fillers, namely, FRWB/TiO2, FRWB/Al(OH)3, FRWB/Mg(OH)2 and FRWB/CaCO3. The coatings were dispersed using a high-speed disperse mixer. The prepared intumescent coating mixtures were coated on the steel plate. Then, the coatings were left to dry and were then stored in desiccators to prevent contamination. A gauge Elcometer model A456 was used to measure the thickness of the intumescent coating within the range of 1.5 ± 0.2 mm. A pH meter (Hanna Instrument pH 211) was used to obtain the pH of coatings at the room temperature of 25°C.

Static immersion test

The static immersion test is considered a standard method that evaluates the water resistance (i.e. the degree of water absorption) of thin-film intumescent coatings using the gravimetric method. Thin-film intumescent coatings (dimensions: 20 mm × 10 mm × 0.5 mm ± 0.1 mm) were immersed in distilled water at 25°C. At specific time intervals, the samples were removed and were blotted with a piece of paper towel to absorb excess water. Weight change was calculated using equation (1) and expressed as a function of time:

(1) ESW=(WEWO)WO×100%
where ESW is the water uptake ratio of the film, WE denotes the weight of the film at different times and WO is the dry weight of the sample.

Scanning electron microscope observation

Microscopic analyses were conducted using a tabletop scanning electron microscope (SEM; Phenom ProX desktop) to examine the surface morphology of the intumescent coatings and the efficiency of the char layer formations. For SEM observation, a low beam accelerating voltage of 10-15 kV was operated at a magnification of 6,000 to reduce the possibility of thermal damage to the char layers. The phosphorus and other element content were tested by the energy dispersive spectroscopy (EDS) spectra and maps, and the electron beam energy was 15 kV. The EDS measurement was performed at four different positions for each sample, and the average phosphorus composition value was used as the result.

Adhesion strength test

The adhesion strength process is important in ensuring a strong bond between the steel substrate and intumescent coating layer interface. The adhesion strength of the coated sample was determined using a pull-off adhesion tester (PosiTest-AT-A Automatic, DeFelsko). The coatings were each sprayed on one side of 50 mm × 50 mm × 2.6 mm steel plates with a film thickness of 1.5 ± 0.05 mm. During operation, the flat face of a pull stub (14 mm dolly) was adhered to the coating using the epoxy glue (thickness of 0.5 ± 0.05 mm) for evaluation. The forces of the peeled-off areas of the samples were calculated and classified according to the ASTM D4541 standard classifications. Force was calculated on the basis of the following equation (2):

(2) F=fbπd2 4
where F is the actual force (N or kgm/s) applied to the test surface, fb is the largest mean pull-off stress applied during a pass/fail test or the pull-off strength achieved at failure (MPa) and d is the equivalent diameter of the original surface area stressed in meters, which is usually equal to the diameter of the loading fixture.

Fire protection test

This test was used to characterize the formation of the char thickness and the reaction of the intumescent coating during and after the burning process. The fire protection test was conducted using the equipment in Figure 1. The prepared coating was applied onto grit blasted steel plates (100 mm × 100 mm × 3 mm) and allowed to dry at room temperature. This process was repeated 6-8 times until a 1.5 ± 0.2-mm dry film thickness was obtained. The gas consumption of the Bunsen burner was 160 g/h. The steel plate coated with the intumescent formulation was exposed to gas consumption and fire for 100 min. A high-temperature flame (approximately 1,000°C) was applied with a Bunsen burner to a bare steel plate and to a vertically mounted coated plate. The measurement of the temperature profile of the backside of steel plate during fire exposure was recorded using a digital handheld thermometer. In this experimental work, 400°C was chosen as the critical temperature for steel (Yew and Ramli Sulong, 2011).

Thermogravimetric analysis

Thermogravimetric (TG) analysis was conducted using a PerkinElmer TGA 4000 model to study the thermal degradation and determine the residual weight of the intumescent coatings. A total of 5-10 mg of intumescent coating was placed in a ceramic crucible and heated from 30°C to 900°C at a ramp rate of 20°C/min in an airflow rate of 20 mL/min.

LOI test

The LOI is widely used in the determination of the relative flammability of polymeric materials, the most well known of the standard fire tests. The LOI tests were conducted in accordance with the standard method of BS 2782:141 and the BS EN ISO 4589-2:1999 standards.

In the LOI test, a flat steel bar (dimensions: 130 mm × 10 mm × 3.0 mm) coated with intumescent coating (thickness of 2.0 ± 0.5 mm) was supported in a vertical glass column and a slow stream of oxygen/nitrogen mix was fed into the glass column. The coated flat steel bar was ignited by a flame that burns downward, where steady burning conditions can be replicated on a small scale. The oxygen/nitrogen ratio can be varied, and the test recorded the minimum concentration of oxygen (as a percentage) that supports combustion. The result is usually expressed as equation (3):

(3) n=O2O2+N2×100%
where n is the oxygen index percentage, O2 denotes the volume of oxygen gas and N2 is the volume of nitrogen gas.

The results from the test will vary considerably, are repeatable and are used for quality assurance and for indicating the potential flammability of the specific polymer being tested. A higher oxygen index value indicates better flame-retardant properties of the coating.

Fourier transform infrared

The Fourier transform infrared (FTIR) spectrum was used to analyze the molecular structure changes that could occur from mixing the flame-retardant additives, binders and different fillers after fire exposure. The wavenumber spectra in the region from 600 cm−1 to 4,000 cm−1 with 30 resolutions of 4 cm−1 were investigated for identifying the presence of specific functional groups in the intumescent coatings and char layer formations. The analyses were conducted using a Nicolet iS10 spectrophotometer with a PerkinElmer Spectrum 400 spectrometer.

Results and discussion

Static immersion test

The static immersion test was used to evaluate the water resistance of the thin-film intumescent coatings. The differences among the weight change rate curves of the thin-film intumescent coatings are shown in Figure 2. The weight change rate of all intumescent coatings increased dramatically from the first day until the fourth day of immersion. The percentage declined slightly from the fourth day until the seventh day of immersion. FRWB/CaCO3 showed the slowest rate of water penetration, which was because of the low solubility of CaCO3 in water, which is 0.013 g/L (at 25°C) (Tegethoff et al., 2001). After 24 h of exposure, coatings FRWB/TiO2, FRWB/Al(OH)3 and FRWB/Mg(OH)2 showed the weakest resistance to water penetration with a weight gain of more than 0.4 per cent compared with that of FRWB/CaCO3. After the fourth day, coating FRWB/CaCO3 showed the slowest rate of water penetration compared with the other samples. The 10 per cent CaCO3 filler composition had a slower water penetration rate by decreasing the formation of cracks and crevices.

When the specimens were soaked in water for seven days, two major processes (permeation and migration) simultaneously occurred. However, for this test, only the permeation process occurred for all specimens, which may be attributed to the fact that the water permeation process exceeded the migration process of the fire-retardant ingredients (Yew et al., 2013). When intumescent coating with the composition of the water-borne binder was immersed in water, its film-forming material and weight gain were caused by water permeation at the initial immersion stage. Therefore, the intumescent coating with the water-borne foundation easily made the small molecules and ions permeate into the pores of the water-borne intumescent coating, resulting in a weight increase. The weight of the coating continually increased to a maximum value after 96 h of immersion. Weight gain of intumescent coating evidently weakened after 96 h of water immersion because the system reached a physical and chemical equilibrium. Furthermore, water easily destroyed the flame-retardant additives after the equilibrium was achieved (Wang et al., 2006). When the filler was mixed with intumescent coating, both flame-retardant additives and VAC formed a cross-linking network to modify the water resistance of the intumescent coating. The CaCO3 filler efficiently interpenetrated the cross-linking network to prevent the water penetration rate change in the water-borne intumescent coating.

The results of the pH value of coatings are shown in Table I. The intumescent coating with TiO2 filler is more alkaline and absorb more water after seven days of immersion.

Scanning electron microscope and EDS analyses

The SEM micrograph images and EDS of all samples with different fillers added before and after being exposed to the Bunsen burner test are illustrated in Figures 3 and 4, respectively. Coating FRWB/TiO2 showed a slightly larger amount of holes and voids compared with other samples. Small cracks were also present on the surface structure of coatings FRWB/TiO2 and FRWB/Mg(OH)2 before the Bunsen burner exposure. All micrograph images from the charred foam structure showed that the coating did exhibit swelling and experienced slight shrinking that causes the structure to appear dehydrated. The holes were formed from the formation of trapped gas by the blowing agent when the coating was subjected to fire. Heat may get transferred to the steel substrate through cracks in the foam structure, which could lead to a decline in the fire protection of the intumescent coating and increase the efficiency of heat transfer. The size and and shape of holes or “cavities” on the fracture surface have been reported caused by the exfoliation of the APP particles. According to the data in Figure 3, the formation of coating FRWB/CaCO3 was uniform without any cracks. This rigid formation has improved the mechanical and fire protection properties of the coating systems.

Choosing the appropriate filler improves the efficiency of intumescent coatings. As it can be seen, the char layer of sample FRWB/CaCO3 after the Bunsen burner test had a more uniform and denser foam structure, which could isolate the steel substrate from fire and subsequently contribute to a better intumescent effect. Some of the charred residues of the burned FRWB/CaCO3 also appeared to be intumescent-like bubbles, which believed can slowly impede the heat and mass transfer between the gas and condensed phases and protect the underlying polymeric substrate from accelerate spread the heat flux in a flame, resulting in the good filler properties (Malucelli et al., 2014). As reported by Yew et al. (2013), the samples that contained a chicken eggshell bio-filler produced a denser and more uniform foam structure. Chicken eggshell waste consists of ±95 per cent calcium carbonate in the form of calcite and ±5 per cent organic materials. The addition of the CaCO3 filler into the intumescent coating led to a more even distribution, a homogeneous and smoother finish and a better interfacial bonding of the filler/binder compared with the other fillers.

The SEM–EDS analysis was used to assess the content of phosphorus and filler elements on the surfaces of intumescent coating before and after its burned residue, after the fire protection test, and the corresponding data are shown in Figure 5. As shown in Figure 5(a), (b) and (c), the phosphorus content in intumescent coating decreased after the fire protection test. Meanwhile, Figure 5(d) shows that the linear relationship of phosphorus content increased indicating that the phosphorus content contributes to good flame retardancy when mixed with calcium carbonate filler. Adverse reaction as presented in Figure 5(b) shows that the aluminium content of the residue collected from the fire protection test increased, meanwhile the other filler content of the residue on other specimens decreased. This finding suggests that aluminium hydroxide filler is involved in the formation of char layer, which hinders the transfer of heat flow and combustible gas and thus provides good flame retardancy. Therefore, it may be concluded from the above results that the condensed-phase mechanism is applicable to the flame-retardant water-borne intumescent coating with calcium carbonate and aluminium hydroxide fillers.

Adhesion strength

The pull-off adhesion tester, following the ASTM D4541 international standards by using a self-aligning aluminum dolly, was applied to the intumescent coating samples for performing the test for adhesion. The measurements of the average adhesion strengths and crack charges of coatings FRWB/TiO2, FRWB/Al(OH)3, FRWB/Mg(OH)2 and FRWB/CaCO3 are tabulated in Figure 6. All samples showed high adhesion strengths between 1.84 MPa and 2.13 MPa. The adhesion strength might have increased because the carbonyl group between the VAC and the surface oxide/hydroxide of the steel plate forms a hydrogen bond in the mixture of the coating. The cross-linking between these filler molecules enhanced the interaction between the coating mixture and the substrate as well as between themselves (Bhattacharya et al., 2009). Hydrogen bonding plays an important role in the adhesive properties of coatings. The hydrogen bonding is formed between a strongly electronegative atom, usually oxygen, nitrogen or fluorine, and a hydrogen atom. This phenomenon for the O electronegative atom is shown in Figure 7.

The incorporation of a filler improved the adhesion of the intumescent coating to the steel plate substrate. FRWB with the CaCO3 filler exhibited the highest adhesion strength and cracked charge with 2.13 MPa and 33.4 × 103 N, respectively, compared with the other coating formulations. Mechanical stress can promote the formation of voids in the matrix because of the loss of the adhesion between the coating and the filler (Fredj et al., 2008). The improvement in the mechanical properties is attributed to a more efficient CaCO3 particle size, which provides better reinforcement properties (Yew et al., 2013). The intumescent coating with CaCO3 particles was better embedded before and after the Bunsen burner as shown in the SEM micrographs in Figure 3. The improvement in adhesion strength is because of the improved reinforcement properties between the CaCO3 filler and the VAC binder. Coating that contained 10 per cent of CaCO3 in combination with the water-borne binder was sufficiently efficient to improve the adhesion strength without the expense of the fire-resistant properties.

Fire protection test

All samples after the Bunsen burner test are presented in Figure 8, with recorded maximum thickness of the char layer formation.

During the test, the temperature of the backside of the single-side-coated steel plate was measured using a digital thermometer connected to a type-K thermocouple wire. The test was plotted as a function of time, as shown in Figure 9. The influences of the content of filler on the equilibrium temperature of the intumescent coatings are shown in Table II.

The initial 12 min showed that similar temperatures below 200°C were recorded for all samples. After that, the temperature rate for FRWB/CaCO3 began to slow down, whereas the rate for FRWB/Mg(OH)2 continued to rapidly increase until reaching 370°C within 30 min of fire exposure. The temperature rate of both coatings FRWB/TiO2 and FRWB/Al(OH)3 began to slow down after approximately 20 min of fire exposure. After 40 min of the test, the temperature reached an equilibrium value and almost remained unchanged until the last stage of the fire protection test. The equilibrium temperature of FRWB/CaCO3 was significantly lower than the other coatings. FRWB/CaCO3 displayed the best result in fire protection performance. The equilibrium temperature was just 264°C. The char combined the positive effects of the flame-retardant additives, binder and CaCO3, promoting adhesion to the steel and producing an extremely hard char, while ensuring the formation of a uniform foam structure (Yew and Ramli Sulong, 2012). Physical and chemical interactions occurred, resulting in the superior mechanical resistance of the char as well as its adhesion to the steel plate. FRWB/Mg(OH)2 had the highest equilibrium temperature, that is 390°C, almost reaching the critical temperature after 100 min of fire exposure. The equilibrium temperature in FRWB/Mg(OH)2 could be explained by the loss of adherence of the coating to the steel plate because of its lightness and porosity, or by a loss of cohesion of the char. Furthermore, coating with 10 per cent of Mg(OH)2 filler had a lower thickness of char formation compared with other coatings, as shown in Figure 8.

Thermal analysis

The viscoelastic behavior of the binder significantly affects the fire protection of intumescent coatings (Wang and Yang, 2010). Thus, high-temperature rheological measurements of the binders were performed to identify the changes in the rheological properties of the binders under thermal stability conditions. Results of TG analysis are shown in Figure 10 for pure VAC as binder and in Figure 11 for intumescent coatings FRWB/TiO2, FRWB/Al(OH)3, FRWB/Mg(OH)2 and FRWB/CaCO3.

The thermogram of the binder shows that pure VAC undergoes three degradation steps. The first mass loss stage below 200°C is because of the evaporation of moisture (water content). The second mass loss is below 400°C, which accounts for the highest percentage change in mass because of the thermal degradation VAC. The third mass loss stage between 380°C and 900°C is because of the thermal degradation of VAC residue. Finally, the binder is completely decomposed at 900°C.

For the thermograms of the intumescent coatings, the curves were almost similar from 100°C to 320°C except for FRWB/CaCO3, which was the degradation temperature slightly lower than the other coatings. Between 320°C and 420°C, the TGA curves were almost similar except for FRWB/Mg(OH)2, which has slightly higher degradation temperature. Above 420°C, the TGA curves of all coatings began to differ from each other. For all coatings, the highest mass loss were at temperature level between 320°C and 420°C because of the decomposition of the VAC main chain. The residue mass loss at temperatures from 420°C to 900°C was because of the decomposition of fillers main chain. The TGA curves for intumescent coatings with TiO2, Mg(OH)2, Al(OH)3 and CaCO3 fillers had residue mass losses of 8.68, 6.28, 9.08 and 13.98 per cent, respectively, at temperature 900°C. On the other hand, the study of the evolution of mass decomposition had shown that pure fillers decompose between 400°C and 500°C for Mg(OH)2 (L’vov et al., 1998), between 600°C and 700°C for Al(OH)3 (Hu et al., 2016), between 700°C and 800°C for TiO2 (Xing et al., 2016) and between 800°C and 900°C for CaCO3 (Li et al., 2013). The higher residue mass loss indicated that the appropriate combination of filler with flame-retardant additives and binder resulted in the enhancement of the antioxidation, thermal stability and fire protection performance of the coating. Intumescent coating containing CaCO3 filler can increase the cohesion of the residue and form a barrier layer, which can provide a better fire protection performance evolution.

Flammability test

The sample coatings after the LOI test are shown in Figure 12. The results of the LOI value of coatings FRWB/TiO2, FRWB/Al(OH)3, FRWB/Mg(OH)2 and FRWB/CaCO3 are shown in Table III.

All sample coatings had an oxygen index value in the range of 29 to 35 and are unlikely to support burning in atmospheric air. In the category of individual filler, sample with aluminium hydroxide filler exhibited the highest oxygen index value of 35, which resulted in excellent flammability resistance. On the other hand, sample with TiO2 filler had an oxygen index value of 29, which exhibited the lowest flammability in the category of individual filler. Study had shown that the presence of TiO2 in particular coatings deposited on polyamide does not seem to improve the fire-retardant properties and lead to lowest flammability (Apaydin et al., 2014).

Study had shown that the largest group of mineral fire retardants most widely used in industry are metal hydroxides (e.g. aluminium hydroxide and magnesium hydroxide). Metal hydroxides which act as fire retardants when used as a filler will dilute the combustible polymer decomposition products, by releasing water vapor through the endothermic decomposition leaving a thermally stable inorganic residue, and the heat capacity will increase the amount of heat needed to vaporize the same amount of fuel (Hollingbery and Hull, 2010). The presence of gas-phase flame diluents (water or carbon dioxide) will also tend to swell the flame and reduce its temperature, hence reducing the proportion of heat transferred back to the sample. Therefore, different behaviors of fire retardants will contribute toward the increase in oxygen index. The detailed reaction of the endothermic decomposition of the fillers is explained according to these reactions in equations (4), (5), (6) and (7):

(4) TiO2(s)  Ti(s) + O2(g)
(5) 2Al(OH)3(s)  Al2O3(s)+3H2O(g)
(6) Mg(OH)2(s)  MgO(s)+H2O(g)
(7) CaCO3(s)  CaO(s) + CO2(g)

The heat capacity in terms of J · mol−1 · K−1 (Table IV) of aluminium oxide is slightly higher than that of calcium carbonate. Aluminium hydroxide produces water as the decomposition product, whereas calcium carbonate produces carbon dioxide. The heat capacities of water vapor and carbon dioxide are slightly similar. Although, it is well known that calcium carbonate is more stable than calcium oxide at the temperature range of stage (Wu et al., 2016), aluminium hydroxide leaves an aluminium oxide residue with the highest heat capacity, compared to calcium carbonate which leaves a calcium oxide residue with lower heat capacity. Such factor may cause the aluminium hydroxide to decompose over a wider temperature range than calcium carbonate, allowing it to continue diluting the flame and cooling the solid phase over a longer period and recording the highest oxygen index compared to other fillers. Study had shown that hybrid fillers (mixtures of two or more fillers) can lead to higher oxygen index (Abdel‐Mohsen and Emira, 2007).

Analyses of char samples

The elemental functional groups analyses of the intumescent coatings of all samples are shown in Figure 13. Table V lists the references of peak wavenumbers and band positions with the vibration modes in the spectrum of intumescent coatings. The infrared spectra of the intumescent coating filler show their fingerprints and changes in the peak positions for each filler.

Figure 13 shows that all coatings had similar trends of peak spectroscopies and almost identical intensities. For all coatings, the presence of a broadband from 3,100 cm−1 to 3,500 cm−1 may be due to the O-H stretching because of the strong intramolecular and intermolecular hydrogen bonds (Abd El-Wahab, 2015). This functional group was predicted to promote the hydrogen bonding between the coating and steel substrate interface (Wicks and Jones, 2013).

VAC consists of –C-O (ester bond) and –C = O (ester carbonyl) groups (Scheme 1). These groups are sensitive to forming hydrogen bonds. –C-O forms structurally flexible functional groups because the rotation around the C-O-C bonds has a low barrier. However, –C-O vibration mode can bond to other functional groups such as methyl group (-CH3) compared to the –C = O vibration mode. The –C-O and –C = O transmittances suggest another evidence of hydrogen bonding occurrence. Given hydrogen bonding plays an important role in the adhesive properties of coatings because the formation of hydrogen bonding strengthens the interaction between the ingredients in intumescent coating and reinforces the binder matrix. The FTIR study is the main characteristic transformation spectroscopy that explains the cross-linking between the components in a mixture and can be used to support the hydrogen bonding results obtained from the adhesion strength analysis.

Conclusions

This study demonstrates an interesting correlation between the filler and water-borne binder intumescent coating. The characterization of flame-retardant additives and VAC resin with different fillers was conducted to form a new binder resin for application on steel structures as a substrate. Various characteristic analyses covering the fire protection and the physical and mechanical tests were conducted to study the behavior of intumescent coatings. From the fire protection study, FRWB/CaCO3 had the highest char thickness of 31.2 mm and the lowest equilibrium temperature of 264°C. Sample with aluminium hydroxide filler exhibited the highest oxygen index value of 35 per cent, which resulted in an excellent flammability resistance. From the physical studies, the effect of water penetration after the seventh day had less influence on FRWB/CaCO3 compared with other coatings. The char layer formation from FRWB/CaCO3 also had a more uniform and denser foam microstructure. The SEM–EDS analysis showed that the phosphorus content of the burned fabric was higher than that of the unburned intumescent coating with calcium carbonate filler, meanwhile the aluminium content was increased compared to other filler content of the residue burned. TGA conducted on the coatings showed the three degradation steps and the highest mass loss because of the decomposition of the pure VAC main chain. It is clear that the decomposition mechanism and oxygen index of filler are influenced by a number of factors, including heat capacity and composition of the burning at atmospheric air. FTIR revealed that the existence of hydrogen bonding promoted the interfacial adhesion bonding strength between the coating and substrate. From the mechanical study, FRWB/CaCO3 was able to withstand adhesion strengths and forces up to 2.13 MPa and 33.4 kN, respectively. Thefore, it may be concluded from the above results that both aluminium hydroxide and calcium carbonate fillers reduced the fire propagation index, hence giving the best fire protection and physical and mechanical properties.

Figures

Fire protection test setup: (1) retort stand; (2) Bunsen burner; (3) intumescent coating sample; and (4) digital handheld thermocouple thermometer connected at backside of steel plate

Figure 1

Fire protection test setup: (1) retort stand; (2) Bunsen burner; (3) intumescent coating sample; and (4) digital handheld thermocouple thermometer connected at backside of steel plate

Weight change time curves of water-borne intumescent coating show the relationship between rate of weight change and immersion time for intumescent coatings with different fillers immersed in water

Figure 2

Weight change time curves of water-borne intumescent coating show the relationship between rate of weight change and immersion time for intumescent coatings with different fillers immersed in water

Surface micrographs of the coating before the Bunsen burner test and foam structure after the Bunsen burner test

Figure 3

Surface micrographs of the coating before the Bunsen burner test and foam structure after the Bunsen burner test

EDS of the coating before the Bunsen burner test and foam structure after the Bunsen burner test

Figure 4

EDS of the coating before the Bunsen burner test and foam structure after the Bunsen burner test

The content of phosphorus and fillers in specimen on the surface of intumescent coating before and after its burned residue: (a) FRWC/TiO2; (b) FRWB/Al(OH)3; (c) FRWB/Mg(OH)2; and (d) FRWB/CaCO3

Figure 5

The content of phosphorus and fillers in specimen on the surface of intumescent coating before and after its burned residue: (a) FRWC/TiO2; (b) FRWB/Al(OH)3; (c) FRWB/Mg(OH)2; and (d) FRWB/CaCO3

Adhesion strength and crack charge curves of the intumescent coatings

Figure 6

Adhesion strength and crack charge curves of the intumescent coatings

Hydrogen bonds formed between a monomer of the VAC molecule and the steel plate substrate

Figure 7

Hydrogen bonds formed between a monomer of the VAC molecule and the steel plate substrate

Char layer of samples FRWB/TiO2, FRWB/Al(OH)3, FRWB/Mg(OH)2 and FRWB/CaCO3 after the Bunsen burner test and thickness of the char formation

Figure 8

Char layer of samples FRWB/TiO2, FRWB/Al(OH)3, FRWB/Mg(OH)2 and FRWB/CaCO3 after the Bunsen burner test and thickness of the char formation

Influence of different filler content on the fire protection

Figure 9

Influence of different filler content on the fire protection

TGA thermogram of pure VAC

Figure 10

TGA thermogram of pure VAC

TGA thermogram of intumescent coatings

Figure 11

TGA thermogram of intumescent coatings

Intumescent coatings after the limiting oxygen index (LOI) test

Figure 12

Intumescent coatings after the limiting oxygen index (LOI) test

Infrared spectra of intumescent coatings

Figure 13

Infrared spectra of intumescent coatings

-C = O and -C-O bonds in a VAC molecule

Scheme 1

-C = O and -C-O bonds in a VAC molecule

pH scale of the intumescent coatings

Coating pH value at 25°C
FRWB/TiO2 9.15
FRWB/Al(OH)3 7.80
FRWB/Mg(OH)2 7.74
FRWB/CaCO3 7.66

Equilibrium temperature of sample coatings FRWB/TiO2, FRWB/Al (OH)3, FRWB/Mg (OH)2 and FRWB/CaCO3

Sample Equilibrium temperature of coating (°C)
FRWB/TiO2 294
FRWB/Al(OH)3 345
FRWB/Mg(OH)2 390
FRWB/CaCO3 264

Results of oxygen index value of intumescent coatings and heat capacity of individual filler

Coating Oxygen index, %
FRWB/TiO2 29
FRWB/Al(OH)3 35
FRWB/Mg(OH)2 33
FRWB/CaCO3 34

Heat capacity at 298 K

Filler Chemical formula Heat capacity, Cp (J · mol−1 · K−1)
Titanium (IV) dioxide TiO2 49.30a
Calcium carbonate CaCO3 82.30a
Aluminium hydroxide Al(OH)3 93.10b
Magnesium hydroxide Mg(OH)2 77.00b
Water vapor (measured at 500 K) H2O 35.22c
Oxygen O2 29.11c
Carbon dioxide CO2 37.22c
Aluminium oxide Al2O3 78.77c
Magnesium oxide MgO 37.01c
Calcium oxide CaO 42.09c
Notes:
a

Figures taken from Perry’s Chemical Engineers’ Handbook, 6th ed. (Perry and Green, 1997);

b

Figures taken from Ashton (2005);

c

Figures taken from NIST Chemistry WebBook, available at: http://webbook.nist.gov/chemistry

The assignments of the vibration modes of intumescent coatings

Assignment Band position (cm−1)
FRWB/TiO2 FRWB/Al(OH)3 FRWB/Mg(OH)2 FRWB/CaCO3
P-O-P stretching mode 860 858 868 840
C-O-C stretching mode 1010 1008 1008 1010
PO2/PO3 stretching mode 1224 1228 1230 1230
P=O stretching mode 1430 1430 1430 1428
CH2 stretching mode 1412 1422 1422
C=O stretching mode 1728 1728 1726 1732
C-H asymmetric stretching mode 2916 2924 2924 2920
N-H stretching mode 3300 3312 3298 3300

Source: Gavgani et al. (2014)

References

Abd El-Wahab, H. (2015), “Synthesis and characterisation of the flame retardant properties and corrosion resistance of Schiff’s base compounds incorporated into organic coating”, Pigment and Resin Technology, Vol. 44 No. 2, pp. 101-108.

Abdel‐Mohsen, F.F. and Emira, H.S. (2007), “A study of the effects of different binders and fillers on the properties of flame retardant paints”, Pigment & Resin Technology, Vol. 36 No. 2, pp. 67-73.

Apaydin, K., Laachachi, A., Ball, V., Jimenez, M., Bourbigot, S., Toniazzo, V. and Ruch, D. (2014), “Intumescent coating of (polyallylamine-polyphosphates) deposited on polyamide fabrics via layer-by-layer technique”, Polym Degrad Stab, Vol. 106, pp. 158-164.

Ashton, H.C. (2005), Fire Retardants, in Functional Fillers for Plastics, Xanthos M. (Ed), Wiley -VCH Verlag GmbH & Co. KGaA, Weinheim, New Jersey.

Bhattacharya, A., Rawlins, J.W. and Ray, P. (2009), Polymer Grafting Crosslinking. Wiley, New York, NY.

Cross, M.S., Cusack, P.A. and Hornsby, P.R. (2003), “Effects of tin additives on the flammability and smoke emission characteristics of halogen-free ethylene-vinyl acetate copolymer”, Polym Degrad Stabil, Vol. 79 No. 2, pp. 309-318.

Fredj, N., Cohendoz, S., Feaugas, X. and Touzain, S. (2008), “Effect of mechanical stress on kinetics of degradation of marine coatings”, Progress in Organic Coatings, Vol. 63 No. 3, pp. 316-322.

Gavgani, J.N., Adelnia, H. and Gudarzi, M.M. (2014), “Intumescent flame retardant polyurethane/reduced graphene oxide composites with improved mechanical, thermal, and barrier properties”, Journal of Materials Science, Vol. 49 No. 1, pp. 243-254.

Guo, Z., Chang, J.C.S., Sparks, L.E. and Fortmann, R.C. (1999), “Estimation of the rate of VOC emissions from solvent-based indoor coating materials based on product formulation”, Atmospheric Environment, Vol. 33 No. 8, pp. 1205-1215.

Hollingbery, L.A. and Hull, T.R. (2010), “The thermal decomposition of huntite and hydromagnesite – a review”, Thermochimic Acta, Vol. 509 Nos 1/2, pp. 1-11.

Hu, S., Chen, F., Li, J.-G., Shen, Q., Huang, Z.-X. and Zhang, L.-M. (2016), “The ceramifying process and mechanical properties of silicone rubber/ammonium polyphosphate/aluminium hydroxide/mica”, Polymer Degradation and Stability, Vol. 126, pp. 196-203.

Huang, H., Tian, M., Liu, L., Liang, W. and Zhang, L. (2006), “Effect of particle size on flame retardancy of Mg(OH)2-filled ethylene vinyl acetate copolymer composites”, Journal of Applied Polymer Science, Vol. 100 No. 6, pp. 4461-4469.

Jimenez, M., Duquesne, S. and Bourbigot, S. (2006a), “Intumescent fire protective coating: toward a better understanding of their mechanism of action”, Thermochim Acta, Vol. 449 Nos 1/2, pp. 116-126.

Li, X.-G., Lv, Y., Ma, B.-G., Wang, W.-Q. and Jian, S.-W. (2013), “Decomposition kinetic characteristics of calcium carbonate containing organic acids by TGA”, Arabian Journal of Chemistry, Vol. 10 No. 2, pp. S2534-S2538.

L’vov, B.V., Novichikhin, A.V. and Dyakov, A.O. (1998), “Mechanism of thermal decomposition of magnesium hydroxide”, Thermochim Acta, Vol. 315 No. 2, pp. 135-143.

Malucelli, G., Bosco, F., Alongi, J., Carosio, F., Di Blasio, A., Mollea, C., Cuttica, F. and Casale, A. (2014), “Biomacromolecules as novel green flame retardant systems for textiles: an overview”, Rsc Advances, Vol. 4 No. 86, pp. 46024-46039.

Otáhal, R., Veselý, D., Násadová, J., Zíma, V., Němec, P. and Kalenda, P. (2011), “Intumescent coatings based on an organic-inorganic hybrid resin and the effect of mineral fibres on fire-resistant properties of intumescent coatings”, Pigment & Resin Technology, Vol. 40 No. 4, pp. 247-253.

Perry, R.H. and Green, D.W. (Eds) (1997), Perry’s Chemical Engineers’ Handbook, Wiley, New York, NY.

Ravichandran, R., Bouldin, R.M., Kumar, J. and Nagarajan, R. (2011), “A renewable waste material for the synthesis of a novel non-halogenated flame retardant polymer”, Journal of Cleaner Production, Vol. 19 No. 5, pp. 454-458.

Tai, C.M. and Li, R.K.Y. (2001), “Mechanical properties of flame retardant filled polypropylene composites”, Journal of Applied Polymer Science, Vol. 80 No. 14, pp. 2718-2728.

Tegethoff, F.W., Rohleder, J.W. and Kroker, E. (2001), Calcium Carbonate: From the Cretaceous Period into the 21st Century, Springer, Berlin.

Wang, G. and Yang, J. (2010), “Influences of binder on fire protection and anticorrosion properties of intumescent fire resistive coating for steel structure”, Surface and Coatings Technology, Vol. 204 No. 8, pp. 1186-1192.

Wang, Z., Han, E. and Ke, W. (2006), “An investigation into fire protection and water resistance of intumescent nano-coatings”, Surface and Coatings Technology, Vol. 201 Nos 3/4, pp. 1528-1535.

Weil, E.D. and Levchik, S.V. (2004), “Commercial flame retardancy of polyurethanes”, Journal of Fire Sciences, Vol. 22 No. 3, pp. 183-210.

Wicks, Z.W. and Jones, F.N. (2013), “Kirk-othmer encyclopedia of chemical technology”, Coatings, Wiley, New York, NY, pp. 1-88.

Wu, K., Zhua, K., Kang, C., Wu, B. and Huang, Z. (2016), “An experimental investigation of flame retardant mechanism of hydrated lime in asphalt mastics”, Materials & Design, Vol. 103, pp. 223-229.

Xing, W., Wang, X., Song, L. and Hu, Y. (2016), “Enhanced thermal stability and flame retardancy of polystyrene by incorporating titanium dioxide nanotubes via radical adsorption effect”, Composites Science and Technology, Vol. 133, pp. 15-22.

Yeh, J.T., Yang, H.M. and Huang, S.S. (1995), “Combustion of polyethylene filled with metallic hydroxides and crosslinkable polyethylene”, Polymer Degradation and Stability, Vol. 50 No. 2, pp. 229-234.

Yew, M.C. and Ramli Sulong, N.H. (2011), “Effect of epoxy binder on fire protection and bonding strength of intumescent fire protective coatings for steel”, Advanced Materials Research, Vols.168/170, pp. 1228-1232.

Yew, M.C. and Ramli Sulong, N.H. (2012), “Fire-resistive performance of intumescent flame-retardant coatings for steel”, Materials & Design, Vol. 34, pp. 719-724.

Yew, M.C., Ramli Sulong, N.H., Yew, M.K., Amalina, M.A. and Johan, M.R. (2013), “The formulation and study of the thermal stability and mechanical properties of an acrylic coating using chicken eggshell as a novel bio-filler”, Progress in Organic Coatings, Vol. 76 No. 11, pp. 1549-1555.

Further reading

Jimenez, M., Duquesne, S. and Bourbigot, S. (2006b), “Characterization of the performance of an intumescent fire protective coating”, Surface and Coatings Technology, Vol. 201 Nos 3/4, pp. 979-987.

WebBook. (2015), “National Institute of Standards and Technology Chemistry WebBook”, available at: http://webbook.nist.gov/chemistry (accessed 12 November 2015).

Acknowledgements

The authors would like to acknowledge the support of MyPhD (MoHE, Malaysia) and University of Malaya PPP Grant PG177-2015B.

Corresponding author

N.H. Ramli Sulong can be contacted at: hafizah_ramli@um.edu.my