Fire Resistance and Mechanical Properties of Intumescent Coating Using Novel Bioash for Steel

Article Fire Resistance and Mechanical Properties of Intumescent Coating Using Novel BioAsh for Steel

Jing Han Beh 1,*, Ming Chian Yew 2,* , Lip Huat Saw 2 and Ming Kun Yew 3

1 Department of Architecture and Sustainable Design, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Cheras, Kajang 43000, Malaysia 2 Department of Mechanical and Material Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Cheras, Kajang 43000, Malaysia; [email protected] 3 Department of Civil Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Cheras, Kajang 43000, Malaysia; [email protected] * Correspondence: [email protected] (J.H.B.); [email protected] (M.C.Y.); Tel.: +60-3-0986-0288 (J.H.B. & M.C.Y.) Received: 6 October 2020; Accepted: 3 November 2020; Published: 20 November 2020

Abstract: Recent developments of intumescent fire-protective coatings used in steel buildings are important to ensure the structural integrity and safe evacuation of occupants during fire accidents. Flame-retardant intumescent coating applied to structural steel could delay the spread of fire and heat propagation across spaces and structures in minimizing fire risks. This research focuses on formulating a green intumescent coating utilized the BioAsh, a by-product derived from natural rubberwood (hardwood) biomass combustion as the natural substitute of mineral fillers in the intumescent coating. Fire resistance, chemical, physical and mechanical properties of all samples were examined via Bunsen burner, thermogravimetric analysis (TGA), carbolite furnace, scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX), Fourier transform infrared (FTIR), freeze–thaw cycle, static immersion and Instron pull-off adhesion test. Sample BioAsh intumescent coating (BAIC) 4-7 incorporated with 3.5 wt.% BioAsh exhibited the best performances in terms of fire resistance (112.5 ◦C for an hour under the Bunsen burner test), thermal stability (residual weight of 29.48 wt.% at 1000 ◦C in TGA test), adhesion strength (1.73 MPa under Instron pull-off adhesion test), water resistance (water absorption rate of 8.72%) and freeze–thaw durability (no crack, blister and color change) as compared to other samples. These results reveal that an appropriate amount of renewable BioAsh incorporated as natural mineral fillers into the intumescent coating could lead to better fire resistance and mechanical properties for the steel structures.

Keywords: intumescent coating; flame-retardant; fire resistance; rubberwood ash; mineral fillers; mechanical properties

1. Introduction The fire resistance of materials is one of the key fire safety measures in building design and the construction industry—it must comply with BS 476. Fire safety rules and regulations in the building are important to reduce deaths and serious injuries [1–4]. Intumescent coating (IC) has been used as a passive fire protection design measure in the building industry especially to protect steel structures from fire damage [5,6]. Flame-retardant IC can ensure structural integrity by providing a protective barrier to building materials from reaching a critical temperature when exposed to fire. IC can delay the propagation of fire and maintain enough time for the safe evacuation of occupants during a fire incident [7]. The fire-protective mechanism of IC is working via the formation of uniform multicellular char layer, which proves the ability to delay the speed of heat transfer to the building structure. In this

Coatings 2020, 10, 1117; doi:10.3390/coatings10111117 www.mdpi.com/journal/coatings Coatings 2020, 10, 1117 2 of 21 research, vinyl acetate-ethylene copolymer (VAC) emulsion was chosen as the binder for BioAsh IC samples due to its more eco-friendly, less flammable and has lower volatile organic contents as compared to the solvent-based binder. It is also less hazardous and harmful to the environment and is able to offer good adhesion strength and humidity resistance [8]. However, direct and continuous exposure of VAC binder to moisture will affect its chemical bonding and weakens the adhesion strength [9]. The binder used in the IC is crucial because it assists in the expansion of char and ensures the formation of a uniform char foam structure, which acts as an insulating layer to protect the substrate [10,11]. Hydrophilic fire-retardant additives such as ammonium polyphosphate (APP) and pentaerythritol (PER) in the IC are sensitive to corrosive components like water, acid, and alkali. These additives could easily migrate to the coating’s surface when exposed to a corrosive environment. These phenomena will significantly suppress the performance of IC. VAC copolymer binder used in the IC can limit the migration of flame-retardant additives from accessing the corrosive substances [12–14]. IC consists of three main halogen-free flame-retardant additives: ammonium polyphosphate (APP), an acid source, melamine (MEL), a blowing agent, and pentaerythritol (PER), a charring agent [15]. (1) An acid source such as inorganic acid, acid salt or other acids that elevate the dehydration of carbonizing agent; (2) a carbonizing agent which is a carbohydrate that will be dehydrated by the acid source to become a char; (3) a blowing agent that will be decomposed to release gas resulting in the increase in polymer’s volume and the formation of a swollen multi-cellular layer to protect the steel underneath from fire [16]. This will tremendously reduce the structural degradation of buildings during fire events. Industrial mineral fillers such as titanium dioxide, magnesium hydroxide and aluminium hydroxide are vital flame-retardant ingredients added into IC to enhance the fire-resistance performance. There is no adverse effect reported on toxic smoke leading to death or injury in fire using flame-retardant mineral fillers. The decomposition temperature of flame-retardant fillers to polymer is important to optimize the use of mineral fillers. Factors such as the coherency of the residual layer and its tendency to reconduct thermal emission are also needed to be considered [17]. In this research, BioAsh was incorporated as green mineral filler to partially replace the use of industrial flame-retardant fillers in the IC. Incorporation of novel BioAsh into IC samples was emphasized in this research. BioAsh was derived from rubberwood biomass combustion obtained from a fuel-burning factory located in Sitiawan, Perak, Malaysia. BioAsh was the residual of the burning activity of rubberwood biomass. Rubberwood biomass was left over from rubber tree logging activities. Hevea brasiliensis (rubberwood) is a type of hardwood tree species commercially planted in Malaysia for latex cultivation. Malaysia is the fifth largest rubber producer and exporter in the world [18], with rubber plantations occupied at about 1.07 million hectares land areas in the country [19]. The logging of rubberwood when they reach the end of latex secretion cycle left a large amount of biomass. These rubberwood biomasses are readily available to use as bioenergy or biofuel for factories. Combustions of biomass produce a relatively large amount of BioAsh. BioAsh is derived from the biomass combustion of natural rubber wood in this research. Most of the BioAsh generated by biomass combustion will be considered as a type of industrial waste and will be disposed to landfill at a certain cost [20]. Environment hazards would be minimized by utilizing renewable BioAsh in the green IC as initiated by this research due to its eco-compatible filler. In general, wood-based ash was identified with several minerals, such as carbon, calcium, potassium, magnesium, phosphorus, and sodium [21]. Wood ash is also identified with compounds such as silicon dioxide, aluminium hydroxide, ferric oxide, calcium oxide, magnesium oxide, titanium dioxide, potassium oxide and sulfur trioxide. Most of the decomposed elements possess high thermal degradation after the pyrolysis. Hemicellulose, cellulose and lignin were reported in plant-based materials and thermal degradation varied [22]. Wood ash that is rich in pozzolan is also used as partial cement replacement for building materials in the construction industry. A pozzolan is a material rich in silica and alumina that will react with calcium hydroxide Ca(OH)2, forming a cementitious-like compound with the presence of moisture [23]. The physical and chemical properties Coatings 2020, 10, 1117 3 of 21 of wood ash are depending on factors such as wood species, types of wood, combustion method, combustion temperature and soil condition. Combustion of hardwood tree species will obtain more ash than softwood species and the yield of wood ash will decrease when burning temperature raising. Many types of research have been done to examine different mix proportion of binder, flame-retardant additives, and fillers to enhance the fire resistance performance of IC [24–53]. Conventional IC reported an average equilibrium temperature of 160–190 ◦C (epoxy-based) and 220–290 ◦C (water-based) when the steel substrate was exposed to 1000 ◦C heat in the fire test [37,48,52,53]. Advanced studies have been conducted to explore the potential of various aviculture waste (eggshell), agriculture waste (rice husk ash), aquaculture waste (clam shell) and crop-based biomass (ginger powder and coffee husk) to produce eco-compatible IC formulations with enhanced fire-resistant performance [48,51–53]. These studies of using natural-based fillers in the IC reported various improvements in the fire-resistant properties. Equilibrium temperature of water-based IC using eggshell (160–260 ◦C), rice husk ash (180–220 ◦C), clam shell (136 ◦C) was reported [43,52,53]. Epoxy resin IC formulated with crop-based carbon source can block the transfer of fire heat to steel and maintain the steel temperature at 130 to 150 ◦C[28,29]. Calcium carbonate and lignin were the main components found in eggshell/clamshell and ginger/coffee powder, respectively [29,48]. These components had reported an enhancement in the thermal stability and char strength of the IC. It is noteworthy to investigate the potentials of BioAsh (rubberwood ash) that was sourced locally to be renewed as green mineral filler in the development of environmentally friendly water-based IC. Currently, no data can be found in the literature pool concerning the use of rubberwood ash in water-based IC formulation. In this study, performances of fire resistance, thermal, physical, chemical and mechanical properties of green intumescent flame-retarded coating incorporated with BioAsh were evaluated and examined.

2. Materials and Methods

2.1. Raw Materials and Intumescent Coating Samples Preparation Water-based vinyl acetate copolymer (VAC) was used as the binding agent for BioAsh intumescent coating (IC) in this research. VAC was a milky white emulsion with mean particle size approximate 0.12 µm supplied by Afza Maju Trading, Terrengganu, Malaysia. Three main halogen-free fire-retardant additives: APP, MEL and PER with a mean particle size of 18, <40 and <40 µm, respectively, were supplied by Synertec Enterprise Sdn Bhd, Kuala Lumpur, Malaysia. Titanium dioxide (TiO2) with a particle size < 10 µm was used as an industrial flame-retardant white pigment in the formulation. TiO2 was supplied by Chemolab Supplies Sdn Bhd., Selangor, Malaysia. BioAsh was used as the natural substitute of flame-retardant mineral fillers in the IC formulations due to it consisting of a group of high thermal stability elements [37]. Sieve analysis was used to identify the particle size of BioAsh and the most abundant particle size of 300 µm was obtained. BioAsh with weight percentage of 1.0, 3.5, 5.0 and 10.0 wt.%, respectively, were incorporated to the IC samples as shown in Table1.

Table 1. Formulation of BioAsh intumescent coating samples.

Flame-Retardant Additives: Mineral Fillers (wt.%) Sample VAC (wt.%) APP:PER:MEL (wt.%) Industrial: TiO2 Green: BioAsh BAIC 1-0 50 20:10:10 9.0 1.0 BAIC 3-5 50 20:10:10 6.5 3.5 BAIC 5-0 50 20:10:10 5.0 5.0 BAIC 10-0 50 20:10:10 - 10.0

BioAsh intumescent coating (BAIC) samples were formulated in the laboratory. Respective raw ingredients were mixed using a high-speed disperser at 1500 rpm for 60 min to achieve a homogenous Coatings 2020, 10, x FOR PEER REVIEW 4 of 21

BioAsh intumescent coating (BAIC) samples were formulated in the laboratory. Respective raw ingredientsCoatings 2020, 10were, 1117 mixed using a high-speed disperser at 1500 rpm for 60 min to achieve a homogenous4 of 21 status. BAIC samples, namely, BAIC 1-0, BAIC 3-5, BAIC 5-0 and BAIC 10-0, were synthesized for investigation. status. BAIC samples, namely, BAIC 1-0, BAIC 3-5, BAIC 5-0 and BAIC 10-0, were synthesized for investigation. 2.2. Fire Resistance Test 2.2. FireThe Resistancefire-protective Test performance of BAIC sample was investigated via 60 min of burning test. This experimentalThe fire-protective setup performancetest was conducted of BAIC in a sample fume hood was investigatedwithin a confine via 60zone min in ofthe burning laboratory. test. TheThis formulated experimental BAIC setup was test applied was conducted using a hand in abrus fumeh onto hood a withincarbon asteel confine plate zone with in the the dimensions laboratory. ofThe 100 formulated mm (length) BAIC × 100 was mm applied (width) using × 2.0 a hand mm brush(thick) onto to cure a carbon in an steelair-dried plate condition with the dimensionsfor 7 days. Thisof 100 procedure mm (length) was continued100 mm for (width) 5 to 7 times2.0 mm until (thick) an end to thickness cure in an of air-dried 1.5 ± 0.2 conditionmm dry film for IC 7 days.was × × acquired.This procedure The dry was film continued IC was measured for 5 to 7 times using until a digital an end Vernier thickness caliper. of 1.5 An auto0.2 mm ignition dry film blowtorch IC was ± gun-acquired. IDEALGAS The dry Laser film− IC3000 was from measured Idealgas, using Casier, a digital Italy Vernierinstalled caliper. with EN417 An auto butane ignition gas blowtorch cartridge fromgun- Providus, IDEALGAS Volpiano, Laser 3000 Italy from was Idealgas,used for Casier,the burning Italy installedtest. Maximum with EN417 gas consumption butane gas cartridgewas 150 − g/h.from The Providus, coated Volpiano,steel plate Italy was was held used in a forvertical the burning position test. using Maximum a three-fing gas consumptioner clamp and was exposed 150 g to/h. aThe 1000 coated °C high-temperature steel plate was flame held infrom a vertical a blowtorch position gun. using The valve a three-finger of the blowtorch clamp gun and was exposed adjusted to a to 60% of its maximum to stabilize the pressure of the gas tank for 60 min. The distance between the 1000 ◦C high-temperature flame from a blowtorch gun. The valve of the blowtorch gun was adjusted BAIC-sample-coatedto 60% of its maximum steel to plate stabilize and thefire pressurefrom the of blowtorch the gas tank was for 70 60mm. min. TheThe temperature distance between of the BAIC-coatedthe BAIC-sample-coated steel plate steelwas platerecorded and firein fromdegree the Celsius, blowtorch measured was 70 mm.by aThe digital temperature thermometer of the connectedBAIC-coated to steela thermocouple plate was recorded attached in at degree the backside Celsius, of measured the carbon by asteel digital plate. thermometer Fire resistance connected tests wereto a thermocouple conducted triplicate attached for at theeach backside sample of to the obtain carbon the steel mean plate. temperature. Fire resistance The tests fire were resistance conducted test wastriplicate set up for as shown each sample in Figure to obtain1. the mean temperature. The fire resistance test was set up as shown in Figure1.

Figure 1. Figure 1. FireFire resistance resistance test. test. 2.3. Carbolite Furnace Test 2.3. Carbolite Furnace Test BAIC sample exposed to a temperature at 600 C was examined [50]. 1.5 0.2 mm thin film BAIC ◦ ± wasBAIC hand brushedsample exposed onto a steelto a temperature plate size 50 at mm 600 (length)°C was examined50 mm (width)[50]. 1.5 ± 0.220 mm thin (thick) film and BAIC air × × wasdried hand at room brushed temperature onto a steel for 7plate days. size The 50 thicknessmm (length) of the × 50 dry mm film (width) was measured × 20 mm (thick) using aand digital air dried at room temperature for 7 days. The thickness of the dry film was measured using a digital Vernier caliper. Carbolite furnace model RHF15/8 was used to heat the BAIC sample at a rate of 50 ◦C Vernierper min caliper. to determine Carbolite and furnace evaluate model the char RHF15/8 thickness was formation used to heat and the fire BAIC resistance sample of BAICat a rate samples of 50 °C per min to determine and evaluate the char thickness formation and fire resistance of BAIC after heat treating at 600 ◦C for a duration of 10 min. samples after heat treating at 600 °C for a duration of 10 min.

Coatings 2020, 10, 1117 5 of 21

2.4. Thermogravimetric Analysis (TGA) Thermogravimetric analyzer from Perkin Elmer model STA8000 was used in analyzing the thermal degradation of all samples. The dried BAIC sample weighed between 5 and 10 mg was positioned in a crucible and heated at an elevating temperature from 30 to 1000 ◦C in a nitrogen gas ﬂow condition. The heating rate was 20 ◦C/min. Thermal degradation of the sample was examined by measuring the residual weight of the sample upon being exposed to constant heating up to 1000 ◦C. The mass change of the BAIC sample was measured and calculated to analyze its thermal stability.

2.5. Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray Spectroscopy (EDX) Surface morphology of BAIC char layer was analyzed through the projection of a focused beam of electrons in the microscope. Char surface topography, cell size and structural strength of the sample at 2.00 k magniﬁcation was studied. A Scanning Electron Microscope of Hitachi EDAX S3400-N × from Hitachi, Tokyo, Japan was used. An ultra-thin gold (Au) ﬁlm with 2–20 nm thickness was sputter coated onto the sample to enhance the thermal conduction and secondary electron emission. A low beam energy of 1000 V was operated to lower the thermal damage on the char layer of the sample. EDX analysis which accommodated by the SEM device was used to investigate the elemental composition on original BioAsh, wet-dried BioAsh and BAIC samples.

2.6. Fourier Transform Infra-Red Spectroscopy (FTIR) Infrared spectrum absorption and emission of the BAIC sample was analyzed. The presence of functional groups and molecular changes were identiﬁed. Nicolet iS10 spectrometer from Thermo 1 Fisher Scientiﬁc was used. An infrared spectra wavenumber between 1000 and 4000 cm− was studied.

2.7. Freeze–Thaw Cycle Test The stability of the BAIC sample towards changes of various physical weathering conditions was tested. 1.5 0.2 mm thin ﬁlm sample was hand brushed onto a steel plate of 50 mm (length) 50 mm ± × (width) 1 mm (thick) and air dried for 24 h. The sample was settled in air temperature at 27 C for × ◦ 2 h and quickly frozen in a freezer at 20 C for 2 h. This was followed by heating the sample in a − ◦ drying oven at 50 ◦C for 2 h. A total of 70 cyclic freeze–thaws were conducted. Any color changed, ﬁssure, blister, and coagulum on the BAIC samples were observed and recorded.

2.8. Static Immersion Test Water resistance of the BAIC thin ﬁlm sample was evaluated. The sample was immersed in distilled water in an enclosed container at room temperature. For every 1-h interval, the sample was withdrawn and blotted dry with a paper towel, any water on the surface was eliminated. Weight change of the sample was measured using the gravimetric device and calculation was made following Equation (1):

Esw = [(We Wo)/Wo] 100% (1) − × where Esw indicates the water uptake ratio of the sample, We indicates the wet weight of the sample at various time, and Wo indicates the initial dry weight of the sample. The water absorption of the BAIC sample was examined every one-hour interval with a total duration of up to 5 h.

2.9. Instron Pull-off Adhesion Test Interfacial bonding of BAIC sample with the steel plate was investigated via Instron Universal Testing machine model 3400 from Illinois Tool Works, IL, USA. The 1.5 0.2 mm thin film BAIC sample ± was applied onto the surface of a cylindrical steel rod with 8.03 10 4 mm2 cross-sectional area and × − cured at room temperature. The steel rod surface coated with BAIC sample was then adhered to the plain surface of another steel rod using epoxy glue and air-dried for 24 h. The tensile pull-off adhesion Coatings 2020, 10, x FOR PEER REVIEW 6 of 21 pull-off adhesion test was conducted by applying a perpendicular tensile force to detach the BAIC sample from the steel rod surface with a consistent speed rate at 1 mm/min. Adhesion strength (fb) of the sample was analyzed in compliance with American Society for Testing and Materials (ASTM) D4541Coatings standard2020, 10, 1117 based on Equation (2): 6 of 21

fb = F/A (2) test was conducted by applying a perpendicular tensile force to detach the BAIC sample from the where fb indicates the adhesion strength (MPa), F indicates the maximum fracture load (N) and A = πsteel(d2)/4 rod indicates surface withthe surface a consistent area (mm speed2) of rate coated at 1 mmcylindrical/min. Adhesion steel rod. strength (fb) of the sample was analyzed in compliance with American Society for Testing and Materials (ASTM) D4541 standard 3.based Results on Equation (2): fb = F/A (2) 3.1. Fire Resistance where fb indicates the adhesion strength (MPa), F indicates the maximum fracture load (N) and A = πThe(d2 )fire/4 indicates resistance the performances surface area (mm of all2) ofBAIC coated samples cylindrical were steelinvestigated rod. via a Bunsen burner device in a confined fume hood. The end temperatures of all samples after 60 min of burning were tabulated3. Results in Table 2. The temperature profiles and standard deviations of all samples are presented in Figure 2. 3.1. Fire Resistance

The fire resistanceTable 2. performances End temperatures of all of BAIC BioAsh samples intumescent were investigatedcoating (BAIC) via samples. a Bunsen burner device in a confined fume hood. The end temperatures of all samples after 60 min of burning were tabulated in Table2. The temperature profilesSample and standard deviations End Temperature of all samples (°C) are presented in Figure2. Unprotected Steel 796.0 Table 2. End temperatures of BioAsh intumescent coating (BAIC) samples. BAIC 1-0 136.3 Sample End Temperature (◦C) BAICUnprotected 3-5 Steel 796.0112.5 BAIC 1-0 136.3 BAICBAIC 5-0 3-5 112.5159.0 BAIC 5-0 159.0 BAICBAIC 10-0 10-0 175.7175.7

Figure 2. Temperature profiles and standard deviations of unprotected and protected steels with BAIC Figure 2. Temperature profiles and standard deviations of unprotected and protected steels with samples after a one-hour fire resistance test. BAIC samples after a one-hour fire resistance test. BAIC 3-5 had the lowest end temperature and standard deviation (σ) after exposing to a 60-min Bunsen burner test as compared to unprotected steel (σ = 171.19) and other BAIC coating sample protected steels. BAIC 3-5 exhibited the best fire-resistant outcome at only 112.5 ◦C and the lowest σ of 16.42 after one hour of fire resistance testing. However, samples BAIC 10-0, BAIC 5-0 and BAIC 1-0 Coatings 2020, 10, x FOR PEER REVIEW 7 of 21

BAIC 3-5 had the lowest end temperature and standard deviation (σ) after exposing to a 60-min Bunsen burner test as compared to unprotected steel (σ = 171.19) and other BAIC coating sample Coatings 2020, 10, 1117 7 of 21 protected steels. BAIC 3-5 exhibited the best fire-resistant outcome at only 112.5 °C and the lowest σ of 16.42 after one hour of fire resistance testing. However, samples BAIC 10-0, BAIC 5-0 and BAIC 1- 0revealed revealed the theσ σvalues values of of 33.11, 33.11, 30.70 30.70 and and 29.98, 29.98, respectively. respectively. The The elevations elevations of BAIC of BAIC samples samples after after one onehour hour of ﬁre of resistancefire resistance testing testing are shownare shown in Figure in Figure3. 3.

(a) (b) (c) (d)

FigureFigure 3. AfterAfter one-hour one-hour fire ﬁre resistance resistance test: ( (aa)) BAIC BAIC 1-0, 1-0, ( (bb)) BAIC BAIC 3-5, 3-5, ( (cc)) BAIC BAIC 5-0, 5-0, ( (dd)) BAIC BAIC 10-0. 10-0.

ThisThis result reveals that 3.5 wt.% of BioAsh added into BAIC 3-5 induced an optimum physical andand chemical chemical reaction reaction with with 6.5 6.5 wt.% wt.% TiO TiO2, and2, andAPP-PER-MER/VAC APP-PER-MER/VAC in the in formula, the formula, resulting resulting in good- in qualitygood-quality char with char the with best the fire-resistance best fire-resistance outcome. outcome. The secondary The secondary rise was rise noticed was noticed in BACI in BACI3-5 after 3-5 15after min 15 and min returned and returned to equilibrium to equilibrium at 45 min. at 45 After min. 15 After min, 15TiO min,2 accumulated TiO2 accumulated at the surface, at the creating surface, acreating low emissive a low emissivematerial. material.It changed It the changed boundary the boundary condition condition and thus increased and thus increased the heat flux the received heat flux byreceived the material. by the material.Fire resistance Fire resistance efficacy was effi cacyfollowed was followedby BAIC 1-0 by BAICand BAIC 1-0 and 5-0 BAICat 136.3 5-0 and at 136.3159.0 and°C, respectively.159.0 ◦C, respectively. BAIC 5-0 BAICstarted 5-0 with started a lower with temperature a lower temperature than BAIC than 1-0 BAICin the 1-0first in 15 the min. first After 15 min. 15 min,After the 15 temperature min, the temperature of BAIC 5-0 of BAICremained 5-0 remainedequilibrium equilibrium throughout, throughout, whereas BAIC whereas 1-0 was BAIC declining 1-0 was indeclining temperature. in temperature. The decline The in burning decline intemperature burning temperature meaning a meaningmore stable a more char stablestarted char to take started place to intake BAIC place 1-0 in only BAIC after 1-0 15 only min. after This 15 could min. be This due could to a slower be due char to a slowerformation char process formation in BAIC process 1-0 as in comparedBAIC 1-0 asto comparedothers. 9.0 towt.% others. of TiO 9.02 added wt.% ofcould TiO 2loweradded the could melt lowerflow rate the (MFR) melt flow of BAIC rate (MFR)1-0. Low of MFRBAIC will 1-0. inhibit Low MFR gas released will inhibit from gas APP-PER-MEL/ released from APP-PER-MELVAC decomposition/VAC decompositionto expand the carbonaceous to expand the charcarbonaceous [30]. BAIC char 10-0 [30 displayed]. BAIC 10-0 the displayedworst fire theresist worstance fire with resistance the highest with end the temperature highest end temperatureat 175.7 °C. Thisat 175.7 was◦ C.probably This was attributable probably attributable to the elemental to the composition elemental composition present in present10.0 wt.% in 10.0BioAsh wt.% that BioAsh was chemicallythat was chemically inert, unable inert, to unablegenerate to good-quality generate good-quality fire-protective fire-protective char. This char.implied This single implied 10.0 singlewt.% BioAsh10.0 wt.% in BioAshBAIC 10-0 in BAICwas unideal 10-0 was to activate unideal toreaction activate with reaction APP-PER-MER/VAC with APP-PER-MER formula/VAC producing formula limitedproducing poor limited char to poor deter char heat to transmittance. deter heat transmittance. Nonetheless, Nonetheless, TiO2 shall not TiO be2 shallused notalone be as used thealone only mineralas the only filler mineral in IC formula filler in as IC this formula led to aspoor this ad ledhesion to poor strength adhesion of the strength coating ofto the steel, coating exposed to the thesteel, steel exposed to fire theresulted steel toin firebadresulted fire resistance in bad [31]. fire resistanceThe combination [31]. The of combinationmineral fillers of (hybrids) mineral fillersin IC formulation(hybrids) in ICwas formulation required and was crucial required to ensuring and crucial a promising to ensuring fire-resistance a promising outcome fire-resistance and adhesion outcome strengthand adhesion to the strength steel [31]. to theAll steelsteel [ 31plates]. All coated steel plates with coatedBAIC coating with BAIC samples coating displayed samples a displayed better fire a resistancebetter fire resistanceefficiencye ffi(belowciency 200 (below °C) 200than◦C) unpr thanotected unprotected steel steelplate plate that that raised raised to tothe the maximum maximum temperaturetemperature ofof 796796 ◦°CC whenwhenexposed exposed to to 60 60 min min of of fire fire [13 [13,31].,31]. All All BAIC BAIC samples samples did did not not detach detach from from the thesteel steel plate. plate. This This indicated indicated an acceptable an acceptable surface surface adhesion adhesion strength strength of BAIC of samplesBAIC samples to the steelto the surface. steel surface.This result reveals that elemental compositions present in 3.5 wt.% BioAsh were able to produce the bestThis synergy result reveals effect withthat elemental 6.5 wt.% TiOcompositions2 in APP-PER-MER present in/VAC 3.5 wt.% formula BioAsh forming were theable best-quality to produce themulticellular best synergy char effect to constrain with 6.5 wt.% direct TiO heat2 in transfer APP-PER-MER/VAC to the steel plate. formula It can forming be concluded the best-quality that the multicellularhybridization char of BioAsh to constrain (green direct flame-retardant heat transfer mineral to the fillers) steel andplate. TiO It 2canwas be appropriate concluded tothat use the to hybridizationformulate the environmentallyof BioAsh (green friendlyflame-retardant IC against mineral fire. fillers) and TiO2 was appropriate to use to formulate the environmentally friendly IC against fire. 3.2. Char Thickness

3.2. CharCarbolite Thickness furnace was used to examine all BAIC coating samples at a temperature of 600 ◦C. The charCarbolite thickness furnace of each was sampleused to isexamine shown all in FigureBAIC coating4. samples at a temperature of 600 °C. The char thickness of each sample is shown in Figure 4.

Coatings 2020, 10, 1117 8 of 21 Coatings 2020, 10, x FOR PEER REVIEW 8 of 21

14 13

10 9.5

8 7.5 6.5 6

4 Char thickness (mm) Char thickness

0 BAIC 1-0 BAIC 3-5 BAIC 5-0 BAIC 10-0

Figure 4. Char thickness (mm) of BAIC samples. Figure 4. Char thickness (mm) of BAIC samples. Various ingredients in IC formulation reacted and swelled when exposed to elevated heat. Various ingredients in IC formulation reacted and swelled when exposed to elevated heat. The The property and structure of IC changed differently from low to high temperature. 600 ◦C is propertydetermined and as structure the temperature of IC changed where mostdifferently of the from reactions low takento high place temperature. in IC were 600 finished °C is determined and an inert as the temperature where most of the reactions taken place in IC were finished and an inert char was char was formed. At a temperature below 600 ◦C, ingredients in IC were still reacting and complex formed. At a temperature below 600 °C, ingredients in IC were still reacting and complex chemical chemical reactions were in progress. At a temperature above 600 ◦C, certain ingredients will start reactionsachieving were their in limiting progress. thermal At a resistance,temperature decompose above 600 and °C, turncertain into ingredients ash instead will of inert start char achieving that is theirmeasurable. limiting BAICthermal 3-5 resistance, with 3.5 wt.% decompose BioAsh and formed turn theinto thickest ash instead char of layer inert of char 13 mm that amongis measurable. the rest. BAICBAIC 3-5 1-0, with BAIC 3.5 5-0 wt.% and BioAsh BAIC 10-0 formed was the 9.5, thickest 7.5, and char 6.5 mmlayer thick, of 13 respectively. mm among Thisthe rest. indicated BAIC that1-0, BAICBAIC 5-0 3-5 formulationand BAIC 10-0 under was the 9.5, condition 7.5, and of6.5 3.5 mm wt.% thick, BioAsh respectively. had achieved This theindicated optimal that reaction BAIC with 3-5 formulation under the condition of 3.5 wt.% BioAsh had achieved the optimal reaction with 6.5 wt.% 6.5 wt.% TiO2 in APP-PER-MER/VAC formula triggered the densest char layer when exposed to heat. TiOThis2 couldin APP-PER-MER/VAC be largely contributed formula by the triggered mineral the elements dense andst char their layer appropriate when exposed composition to heat. present This couldin 3.5 wt.%be largely of BioAsh contributed that promote by the themineral greatest elemen charts thickness. and their Full appropriate replacement composition of industrial present mineral in 3.5fillers wt.% using of BioAsh 10.0 wt.% that of BioAshpromote in the BAIC greatest 10-0 led char to thethickness. thinnest Full char. replacemen Char thicknesst of industrial was not correlated mineral fillerswith theusing increasing 10.0 wt.% amount of BioAsh of BioAsh. in BAIC The 10-0 least led char to expansionthe thinnest in char. BAIC Char 1-0 couldthickness be attributed was not correlatedto the excessive with the composition increasing of amount mineral of elements BioAsh. presentThe least in char 10.0 wt.%expansion BioAsh in thatBAIC suppressed 1-0 could thebe attributedexpansion ofto desirablethe excessive char thickness.composition This of research minera determinedl elements anpresent optimum in 10.0 ratio wt.% of 3.5 BioAsh wt.% BioAsh that suppressed the expansion of desirable char thickness. This research determined an optimum ratio of and 6.5 wt.% TiO2 in the green IC formulation which could ensure the yield of an ideal char thickness 3.5to protectwt.% BioAsh the steel and from 6.5 fire.wt.% TiO2 in the green IC formulation which could ensure the yield of an ideal char thickness to protect the steel from fire. 3.3. Thermogravimetric Analysis 3.3. Thermogravimetric Analysis Fire protection performance has correlated with the thermal properties of coating [32]. In this research,Fire protection thermal degradationperformance ofhas all correlated BAIC samples with th weree thermal examined properties via TGA.of coating Residual [32]. weightIn this research,vs. temperature thermal of degradation all BAIC samples of all BAIC as plotted sample insFigure were 5examined. The curves via TGA. of all samplesResidual seemweight quite vs. temperature of all BAIC samples as plotted in Figure 5. The curves of all samples seem quite similar similar at the first 100 ◦C. Obvious thermal decomposition and mass change of all BAIC samples at the first 100 °C. Obvious thermal decomposition and mass change of all BAIC samples started started between 200 and 500 ◦C indicated by constant drop in curves. The mass loss of BAIC 3-5, between 200 and 500 °C indicated by constant drop in curves. The mass loss of BAIC 3-5, BAIC 1-0, BAIC 1-0, BAIC 5-0 and BAIC 10-0 at 500 ◦C was 60.34%, 61.12%, 63.70% and 66.67%, respectively. BAICDerivative 5-0 and thermogravimetric BAIC 10-0 at 500 analysis°C was 60.34%, (DTG) is61.12%, shown 63.70% in Figure and6 .66.67%, The peak respectively. thermal degradation Derivative thermogravimetric analysis (DTG) is shown in Figure 6. The peak thermal degradation of BAIC 3-5, BAIC 1-0, BAIC 5-0 and BAIC 10-0 occurred at a peak temperature of 495, 480, 425 and 400 °C, respectively.

Coatings 2020, 10, 1117 9 of 21

Coatings 2020, 10, x FOR PEER REVIEW 9 of 21 ofCoatings BAIC 2020 3-5,, 10 BAIC, x FOR 1-0, PEER BAIC REVIEW 5-0 and BAIC 10-0 occurred at a peak temperature of 495, 480, 4259 andof 21 400 ◦C, respectively.

Figure 5. Thermogravimetric analysis (TGA) curves of BAIC samples. Figure 5. Thermogravimetric analysis (TGA) curves of BAIC samples. Figure 5. Thermogravimetric analysis (TGA) curves of BAIC samples.

Figure 6. DerivativeDerivative thermogravimetric analysis (DTG) of BAIC samples. The potentialFigure thermal 6. Derivative decompositions thermogravimetric of compounds analysis that (DTG) might of take BAIC place samples. caused a mass loss of The potential thermal decompositions of compounds that might take place caused a mass loss samples, as shown in Table3[ 32–36]. These mineral compounds were predicted from EDX and FTIR of samples, as shown in Table 3 [32–36]. These mineral compounds were predicted from EDX and analysisThe of potential BioAsh ofthermal those most decompositions likely to undergo of compound thermal decompositions that might take in coatingplace caused when heat-treateda mass loss FTIR analysis of BioAsh of those most likely to undergo thermal decomposition in coating when heat- inof TGA.samples, as shown in Table 3 [32–36]. These mineral compounds were predicted from EDX and treated in TGA. FTIRThe analysis dominant of BioAsh decomposition of those most was likely mainly to contributed undergo thermal to by the decomposition reactive phosphates in coating that when contained heat- treated in TGA. 3 inorganicTable (PO4) 3. Potential− ion compound potentially present found in BAIC in the samples, compound. decompositionThe phosphates reaction and were temperature. very reactive; hence, when hydrated and heated, they can react with other elements, creating new compounds transformedTableCompound 3. into Potential other compound phosphate Decomposition present forms in that BAIC Reaction were samples, highly decomposition soluble, Decompose ﬂammable, reaction Temperature andand volatile.temperature. (°C) BAIC 3-5 → had the highestCompoundCaCO residual3 weight DecompositionCaCO of 29.483 CaO wt.%, Reaction + CO followed2 by Decompose BAIC 1-0 ofTemperature700 27.99 wt.% (°C) and BAIC 5-0 Mg(OH)2 Mg(OH)2 → MgO + H2O 360 of 25.93 wt.%CaCO at 10003 ◦C. BAICCaCO 10-0 displayed3 → CaO + theCO2 lowest residual weight700 of 24.27 wt.%. BAIC 10-0 Al(OH)3 Al(OH)3 → Al2O3 + H2O 180 supplementedMg(OH) with 10.02 wt.%Mg(OH) BioAsh2 degraded → MgO + the H2 most.O BAIC 1-0 had a higher360 residual weight than 2H3PO4 2H3PO4 → H4P2O7 + H2O 200 BAIC 5-0 at theAl(OH) end of3 the test.Al(OH) BAIC 1-03 → withAl2O ﬁller3 + H2 combinationO of 1.0 wt.%180 BioAsh and 9.0 wt.% TiO2 4AlPO4 4AlPO4 → 2Al2O3 + P4O10 100 2H3PO4 2H3PO4 → H4P2O7 + H2O 200 Mg(H2PO4)2 Mg(H2PO4)2 → MgH2P2O7 + H2O 450 4AlPO4 4AlPO4 → 2Al2O3 + P4O10 100 –C2H5– –C2H5– → –C2H4– + H+ 95 Mg(H2PO4)2 Mg(H2PO4)2 → MgH2P2O7 + H2O 450 –C2H5– –C2H5– → –C2H4– + H+ 95

Coatings 2020, 10, 1117 10 of 21

performed better than BAIC 5-0 with 5.0 wt.% BioAsh and 5.0 wt.% TiO2. Variables in the residual Coatingsweight 2020 of,BAIC 10, x FOR samples PEER REVIEW against high heat had been subjected to different BioAsh: TiO2 ratio10 given of 21 that APP-PER-MEL/VAC was the fixed parameter. The previous study revealed the use of a suitable amountThe ofdominant eggshell (ES)decomposition as filler enhanced was mainly thermal contributed properties and to charby the thickness reactive by phosphates reducing the that heat containedrelease rate inorganic (HRR), time(PO4) to3− ignitionion potentially (TTI) and found total in heat the release compound. (THR) The in the phosphates IC [37]. The were primary very reactive; hence, when hydrated and heated, they can react with other elements, creating new compound found in ES was calcium carbonate (CaCO3). A suitable amount of Al(OH)3, TiO2 and ES compoundsadded to IC transformed formulation assistedinto other inlowering phosphate the forms index valuethat were of fire highly propagation soluble, [32 flammable,]. and volatile. BAIC 3-5 had the highest residual weight of 29.48 wt.%, followed by BAIC 1-0 of 27.99 wt.% and BAICTable 5-0 3. ofPotential 25.93 wt.% compound at 1000 present °C. BAIC in BAIC 10-0 samples,displayed decomposition the lowest residual reaction andweight temperature. of 24.27 wt.%. BAIC 10-0 supplemented with 10.0 wt.% BioAsh degraded the most. BAIC 1-0 had a higher residual weight thanCompound BAIC 5-0 at the end of Decomposition the test. BAIC Reaction 1-0 with filler combination Decompose Temperatureof 1.0 wt.% (BioAsh◦C) and CaCO CaCO CaO + CO 700 9.0 wt.% TiO2 performed3 better than BAIC3 → 5-0 with 5.02 wt.% BioAsh and 5.0 wt.% TiO2. Variables in Mg(OH)2 Mg(OH)2 MgO + H2O 360 the residual weight of BAIC samples against→ high heat had been subjected to different BioAsh: TiO2 Al(OH) Al(OH) Al O + H O 180 3 3 → 2 3 2 ratio given that2H APP-PER-MEL/VACPO 2H PO was theH P fixedO + paH rameter.O The previous 200study revealed the use 3 4 3 4 → 4 2 7 2 of a suitable 4AlPOamount of eggshell4AlPO (ES) as filler2Al O enhanced+ P O thermal properties 100and char thickness by 4 4 → 2 3 4 10 reducing theMg(H heatPO release) rateMg(H (HRR),PO time) toMgH ignitionP O (TTI)+ H Oand total heat release 450 (THR) in the IC [37]. 2 4 2 2 4 2 → 2 2 7 2 –C H – –C H – –C H – + H+ 95 The primary compound2 5 found in ES2 was5 → calcium2 4 carbonate (CaCO3). A suitable amount of Al(OH)3, TiO2 and ES added to IC formulation assisted in lowering the index value of fire propagation [32].

ThisThis research research predicted predicted the the proper proper combination combination of of CaCO CaCO33 andand Al(OH) Al(OH)3 3presencepresence in in 3.5 3.5 wt.% wt.% BioAshBioAsh and and 6.5 6.5 wt.% wt.% TiO TiO2 could2 could enhance enhance thethe thermal thermal stability stability of BAIC of BAIC 3-5 as 3-5 compared as compared to the to rest. the Thisrest. research This research concluded concluded that the that synergistic the synergistic reaction reaction of 3.5 wt.% of 3.5 BioAsh wt.% BioAsh with 6.5 with wt.% 6.5 TiO wt.%2 in TiOthe2 formulationin the formulation was able was to ablelower to the lower thermal the thermal degradation degradation of BAIC of 4-7 BAIC from 4-7 more from heat more in heatTGA. in HRR, TGA. TTI,HRR, THR TTI, of THR BAIC of samples BAIC samples in accordance in accordance with BS476 with Part BS476 6: Fire Part propagation 6: Fire propagation test will test subject will to subject future to investigation.future investigation.

3.4.3.4. Surface Surface Morphology Morphology CharChar surface surface morphology morphology of of all all BAIC BAIC samples samples afte afterr Bunsen Bunsen burner burner test test were were investigated investigated under under highhigh magnification magniﬁcation of of ×2.002.00 k k in in SEM. SEM. Surface Surface micrographs micrographs of of all all samples samples as as shown shown in in Figure Figure 7.7. × (a) (b)

FigureFigure 7. 7. CharChar morphology: morphology: (a (a) )BAIC BAIC 1-0, 1-0, (b (b) )BAIC BAIC 3-5, 3-5, (c (c) )BAIC BAIC 5-0 5-0 and and (d (d) )BAIC BAIC 10-0. 10-0.

Coatings 2020, 10, 1117 11 of 21

Char layer quality of samples was analyzed. BAIC 3-5 showed a more compact and solid char structure among all samples. Char of BAIC 3-5 displayed a multicellular rosette-like crystalline structure that was rigid with minimal micropores [38]. Complex physical and chemical characteristics of wood ash when hydrated will expand and result in a rosette-like structure. This rosette-like form will remain after being air dried [38]. The multicellular rosette-like crystalline char structure indicated that BAIC 3-5 added with 3.5 wt.% BioAsh was able to form a more solid and uniform carbonaceous matrix resulting in a better fire-protective barrier to resist heat transfer. This rosette-like multicellular matrix was occupied with abundant fine void channels (air gaps) that prevented the speedy spread of heat. The rosette-like structure did not present in other samples. In contrast, BAIC 10-0 showed a soft, fluff and filamentous-like char cell that was fragile and with many obvious irregular macro pores. This soft and spongy fibrous-like char structure could be attributed to the high dosage of plant fibers that existed in 10.0 wt.% BioAsh. Non-multicellular and weak char structure happened in BAIC 10-0 accelerated the heat transfer and declined the fire-retardant efficacy. This resulted in the worst fire resistance in BAIC 10-0. Mineral components that present in 10.0 wt.% BioAsh constrained the chemical reactions with TiO2 in APP-PER-MER/VAC formula. The char structure of BAIC 1-0 displayed fewer macropores and resulted in a better fire-resistant outcome than BAIC 5-0. This research suggests that a suitable mix ratio of BioAsh at 3.5 wt.% and TiO2 at 6.5 wt.% in the APP-PER-MER/VAC formulation could promote uniform multicellular rosette-like char that was excellent in the fire resistance. An in-depth study into the synergistic and rheological behavior about the formation of unique rosette-like multicellular compacted char using hybridization ratio of BioAsh 3.5 wt.%:TiO2 6.5 wt.% will subject to future research.

3.5. Elemental Composition Analysis of BioAsh and Coating Samples Condition of original BioAsh and BioAsh after being wet-dried were analyzed using EDX. Primary mineral contents and changes in wt.% are as tabulated in Table4. Changes in mineral contents in the original BioAsh and BioAsh after being wet-dried were identiﬁed in Figure8.

Table 4. Primary elemental composition change of original BioAsh and BioAsh after being wet-dried.

Element (wt.%) Sample Si Ca Mg Al P C Zn S Ru O Ga BioAsh (original) 1.26 11.59 3.10 - 5.22 41.13 1.08 1.66 0.94 34.03 - BioAsh (wet-dried) 1.57 22.87 5.10 0.65 8.44 21.37 --- 38.31 1.68

Silica (Si) and Calcium (Ca) at 1.26 wt.% and 11.59% were found in the original BioAsh. It was noticed that amount of Si was slightly increased to 1.57 wt.% and the amount of Ca was doubled up to 22.87 wt.% in BioAsh after wet and dried. Minerals such as magnesium (Mg), alumina (Al) and phosphorus (P) in BioAsh were also found increased in amount to 5.10, 0.65 and 8.44 wt.% in a wet-dried condition. Carbon (C) was diminished from 41.13 wt.% to 21.37 wt.% in wet-dried BioAsh. Ca was determined as the primary constituent in rubber tree biomass [38]. Compounds such as CaO, CaCO3, Ca(OH)2, Ca2SiO4 were present in wood ash [38]. Various crop-based plant biomasses were examined and chemical compounds such as CaO, SiO2, MgO, Al2O3,P2O5 were found [40]. EDX ﬁndings in this research discovered only the chemical elements in BioAsh. This research assumed that major mineral compounds such as CaO, CaCO3 and minor mineral compounds such as SiO2, MgO, P2O5, Al2O3 were potentially present in the original BioAsh. Primary elemental composition changes, where most elements noticed in EDX were increased in volume (except carbon) in hydrated BioAsh involved a series of complicated chemical and physical reactions. Compounds dissolved, decomposed, and forming new compounds caused variation in chemical and physical changes. Potential compounds and their series of complex reactions that might happen in the wet-dried BioAsh were tabulated in Table5. Coatings 2020, 10, x FOR PEER REVIEW 11 of 21 will remain after being air dried [38]. The multicellular rosette-like crystalline char structure indicated that BAIC 3-5 added with 3.5 wt.% BioAsh was able to form a more solid and uniform carbonaceous matrix resulting in a better fire-protective barrier to resist heat transfer. This rosette-like multicellular matrix was occupied with abundant fine void channels (air gaps) that prevented the speedy spread of heat. The rosette-like structure did not present in other samples. In contrast, BAIC 10-0 showed a soft, fluff and filamentous-like char cell that was fragile and with many obvious irregular macro pores. This soft and spongy fibrous-like char structure could be attributed to the high dosage of plant fibers that existed in 10.0 wt.% BioAsh. Non-multicellular and weak char structure happened in BAIC 10-0 accelerated the heat transfer and declined the fire-retardant efficacy. This resulted in the worst fire resistance in BAIC 10-0. Mineral components that present in 10.0 wt.% BioAsh constrained the chemical reactions with TiO2 in APP-PER-MER/VAC formula. The char structure of BAIC 1-0 displayed fewer macropores and resulted in a better fire-resistant outcome than BAIC 5-0. This research suggests that a suitable mix ratio of BioAsh at 3.5 wt.% and TiO2 at 6.5 wt.% in the APP-PER- MER/VAC formulation could promote uniform multicellular rosette-like char that was excellent in the fire resistance. An in-depth study into the synergistic and rheological behavior about the formation of unique rosette-like multicellular compacted char using hybridization ratio of BioAsh 3.5 wt.%: TiO2 6.5 wt.% will subject to future research.

3.5. Elemental Composition Analysis of BioAsh and Coating Samples Condition of original BioAsh and BioAsh after being wet-dried were analyzed using EDX. Primary mineral contents and changes in wt.% are as tabulated in Table 4. Changes in mineral contents in the original BioAsh and BioAsh after being wet-dried were identified in Figure 8.

Table 4. Primary elemental composition change of original BioAsh and BioAsh after being wet-dried.

Element (wt.%) Sample Si Ca Mg Al P C Zn S Ru O Ga

BioAsh 1.26 11.59 3.10 - 5.22 41.13 1.08 1.66 0.94 34.03 - (original)

BioAsh (wet- 1.57 22.87 5.10 0.65 8.44 21.37 - - - 38.31 1.68 Coatings 2020, 10, 1117 12 of 21 dried)

(a)

Coatings 2020, 10, x FOR PEER REVIEW 12 of 21

(b)

Figure 8. Elemental composition change of (a) original BioAsh and (b) BioAsh after being wet-dried. Figure 8. Elemental composition change of (a) original BioAsh and (b) BioAsh after being wet-dried. CaO, CaCO3, and SiO2 were three main potential mineral compounds that contributed to the formationSilica (Si) of Calcium and Calcium silicate (Ca) hydrate at 1.26 (C–S–H).wt.% and C–S–H 11.59% is were a gel-like found bindingin the original agent accountableBioAsh. It was to coherenoticed admixturethat amount for of strength Si was sl inightly cementitious increased materials to 1.57 wt.% [51]. and Pozzolan the amount shall consistof Ca was of SiOdoubled2, Al2 Oup3 andto 22.87 ferric wt.% oxide in (FeBioAsh2O3) asafter prescribed wet and indried. ASTM Minerals C618. Pozzolanicsuch as magnesium activities (Mg), might alumina occur in (Al) BioAsh and whenphosphorus wetting, (P) converting in BioAsh potentialwere also pozzolans found increased SiO2/CaO in /amountCaCO3 intoto 5.10, a very 0.65 minor and 8.44 quantity wt.% ofin C–H–S a wet- withdried the condition. cementitious Carbon property. (C) was Combination diminished from of flame-retardant 41.13 wt.% to mineral21.37 wt.% fillers in suchwet-dried as Mg(OH) BioAsh.2 and Ca Al(OH)was determined3 with an appropriateas the primary amount constituent can lower in therubber fire propagationtree biomass index [38]. value,Compounds heat release such rateas CaO, and formCaCO a3, more Ca(OH) uniform2, Ca2 andSiO4thicker were present char layer in wood [37]. Namely, ash [38]. Mg(OH) Various2 crop-basedand Al(OH) plant3 that biomasses exist in 3.5 were wt.% BioAshexamined might and contribute chemical tocompounds the best outcomes such as of CaO, BAIC SiO 3-5.2, QuantitativeMgO, Al2O3,analysis P2O5 were on thefound actual [40]. amount EDX offindings Mg(OH) in2 thisand research Al(OH)3 discoveredin 3.5 wt.% only BioAsh the ischem necessaryical elements in future in research. BioAsh. This research assumed that major mineral compounds such as CaO, CaCO3 and minor mineral compounds such as SiO2, MgO, P2O5, Al2O3 were potentially present in the original BioAsh. Primary elemental composition changes, where most elements noticed in EDX were increased in volume (except carbon) in hydrated BioAsh involved a series of complicated chemical and physical reactions. Compounds dissolved, decomposed, and forming new compounds caused variation in chemical and physical changes. Potential compounds and their series of complex reactions that might happen in the wet-dried BioAsh were tabulated in Table 5.

Table 5. Potential compounds and their reactions occurred during the hydration of BioAsh.

Compound Chemical Formula Reaction (Equation) Carbon C C + O2 → CO2 CaO CaO + H2O + CO2 → Ca(OH)2 Calcium oxide CaO 3 CaO + SiO2 → Ca3SiO5 Calcium carbonate CaCO3 CaCO3 + H2O → Ca2+ + CO2 Silicon dioxide SiO2 SiO2 + H2O → Si(OH)4 Ca(OH)2 Ca(OH)2 + Si(OH)4 → C–S–H Calcium hydroxide Ca(OH)2 Ca(OH)2 + CO2 → CaCO3 Tricalcium silicate Ca3SiO5/C3S Ca3SiO5 + H2O → C–S–H Magnesium oxide MgO MgO + H2O → Mg(OH)2 Phosphorus pentoxide P2O5 P2O5 + H2O → H3PO4 Aluminum oxide Al2O3 Al2O3 + 3 H2O → Al(OH)3

Coatings 2020, 10, 1117 13 of 21

Table 5. Potential compounds and their reactions occurred during the hydration of BioAsh.

Compound Chemical Formula Reaction (Equation) Carbon C C + O CO 2 → 2 Coatings 2020, 10, x FOR PEER REVIEW CaO CaO + H O + CO Ca(OH) 13 of 21 Calcium oxide 2 2 → 2 CaO 3 CaO + SiO Ca SiO 2 → 3 5 CaO, CaCO3, and SiO2 were three main potential mineral compounds 2+that contributed to the Calcium carbonate CaCO3 CaCO3 + H2O Ca + CO2 formation of Calcium silicate hydrate (C–S–H). C–S–H is a gel-like binding→ agent accountable to Silicon dioxide SiO2 SiO2 + H2O Si(OH)4 cohere admixture for strength in cementitious materials [51]. Pozzolan→ shall consist of SiO2, Al2O3 and Ca(OH) Ca(OH) + Si(OH) C–S–H ferric oxide (FeCalcium2O3) as hydroxideprescribed in ASTM C618.2 Pozzolanic activities2 might4 → occur in BioAsh when Ca(OH)2 Ca(OH)2 + CO2 CaCO3 wetting, converting potential pozzolans SiO2/CaO/CaCO3 into a very minor→ quantity of C–H–S with Tricalcium silicate Ca SiO /C S Ca SiO + H O C–S–H the cementitious property. Combination3 of5 flame-retardant3 3 mineral5 2 fillers→ such as Mg(OH)2 and Al(OH)3 with Magnesiuman appropriate oxide amount MgO can lower the fire propagation MgO + H O indexMg(OH) value, heat release rate 2 → 2 2 3 and form a morePhosphorus uniform pentoxide and thicker P char2O5 layer [37]. Namely,P2O5 Mg(OH)+ H2O andH3PO Al(OH)4 that exist in 3.5 wt.% BioAsh might contribute to the best outcomes of BAIC 3-5. Quantitative→ analysis on the actual Aluminum oxide Al2O3 Al2O3 + 3 H2O Al(OH)3 amount of Mg(OH)2 and Al(OH)3 in 3.5 wt.% BioAsh is necessary in future→ research. This study reveals that potential elemental compounds and their reactions that took place in 3.5 This study reveals that potential elemental compounds and their reactions that took place in wt.% BioAsh produced the most positive synergic effect with 6.5 wt.% TiO2 in APP-PER-MEL/VAC formula3.5 wt.% in BioAsh enhancing produced the fire-resistance the most positive properties. synergic eﬀect with 6.5 wt.% TiO2 in APP-PER-MEL/VAC formulaOxygen in enhancing and carbon the ﬁre-resistancecomposition of properties. all BAIC coating samples were examined via EDX. The oxygenOxygen to carbon and carbonratio is compositionshown in Figure of all 9. BAIC coating samples were examined via EDX. The oxygen to carbon ratio is shown in Figure9.

44 1.5

33 1.13

22 0.75 O/C Ratio

11 0.38 O and C Composition wt.% Composition C O and 0 0 BAIC 1-0 BAIC 3-5 BAIC 5-0 BAIC 10-0 Oxygen, O (%) Carbon, C (%) O/C

Figure 9. Oxygen and carbon composition, and O/C ratio of BAIC samples. Figure 9. Oxygen and carbon composition, and O/C ratio of BAIC samples. Oxygen to carbon ratio was crucial in investigating the antioxidation property of the char layer [41]. BAICOxygen 3-5 reinforced to carbon with ratio 3.5 was wt.% crucial BioAsh in hadinvestigatin achievedg the lowestantioxidation O/C ratio property at only of 0.69, the whichchar layer was [41].able BAIC to slow 3-5 down reinforced the fire with and 3.5 heat wt.% propagation BioAsh had toward achieved the steel the lowest underneath O/C ratio it. BAIC at only 3-5 0.69, manifested which wasthe bestable fire-resistant,to slow down furnace the fire and and TGA heat outcomes. propagation In contrast,toward the BAIC steel 10-0 underneath sample with it. 10.0BAIC wt.% 3-5 manifestedBioAsh displayed the best the fire-resistant, highest O/C furnace ratio at and 1.50 TG demonstratedA outcomes. the In weakestcontrast, performance BAIC 10-0 sample to resist with fire. 10.0Char wt.% layer BioAsh of BAIC displayed 10-0 oxidized the highest tremendously O/C ratio causedat 1.50 the demonstrated loss of electrons the damaged weakest theperformance molecules to in resist fire. Char layer of BAIC 10-0 oxidized tremendously caused the loss of electrons damaged the char cells. SiO2 and Al2O3 predicted present in the BioAsh were known for their antioxidant properties. molecules in char cells. SiO2 and Al2O3 predicted present in the BioAsh were known for their SiO2 and Al2O3 in BAIC 3-5 might be assisted in the best synergy with TiO2-APP-PER-MEL/VAC that antioxidantenhanced the properties. antioxidant SiO quality2 and ofAl the2O3 charin BAIC layer 3-5 formed. might Thisbe assisted research in suggested, the best synergy 3.5 wt.% with of BioAsh TiO2- APP-PER-MEL/VACin BAIC 3-5 was appropriate that enhanced to offer the an excellentantioxidant antioxidant quality of value the char to enhance layer formed. the charred This quality research to suggested,better suppress 3.5 wt.% the heatof BioAsh transfer in toBAIC steel 3-5 [52 was]. appropriate to offer an excellent antioxidant value to enhance the charred quality to better suppress the heat transfer to steel [52].

3.6. FTIR Analysis FTIR was used to characterize the cross linkage of molecular functional groups in all BAIC samples. The relationship of functional groups with adhesion strength then can be diagnosed [42]. Analysis of elemental functional groups with the peak wavenumber and band position of all BAIC samples is shown in Figure 10. Coatings 2020, 10, 1117 14 of 21

Figure 10. Infrared spectra of BAIC samples. BAIC samples incorporated with a diﬀerent weight percentage of BioAsh exhibited variation BAIC samples incorporated with a different weight percentage of BioAsh exhibited variation in in peak position. Peak ranging from 2915.7 to 2916.6 cm 1 indicated the symmetrical stretching peak position. Peak ranging from 2915.7 to 2916.6 cm−1 indicated− the symmetrical stretching of C–H of C–H of alkyl chain in the methyl group [42]. Stretching of C=O (carbonyl group) in lignin and of alkyl chain in the methyl group [42]. Stretching of C=O (carbonyl group) in lignin and hemicellulose that originated in BioAsh (rubberwood biomass) displayed at peak 1727.2 to 1727.9 cm 1. hemicellulose that originated in BioAsh (rubberwood biomass) displayed at peak 1727.2 to 1727.9− C O (ester bond) and C=O (ester carbonyl) that were sensitive to form a hydrogen bond can −cm−1. −C−O (ester bond) and− −C=O (ester carbonyl) that were sensitive to form a hydrogen bond can be found in vinyl acetate copolymer. The transmittance of –C–O and –C=O showed peaks ranging be found in vinyl acetate copolymer. The transmittance of –C–O and –C=O showed peaks ranging from 1246.0 to 1246.7 cm 1 and 1727.2 to 1727.9 cm 1, respectively, in all BAIC samples proving the from 1246.0 to 1246.7 cm−−1 and 1727.2 to 1727.9 cm−1, respectively, in all BAIC samples proving the formation of hydrogen bonds [39]. The peak from 1246.0 to 1246.7 cm 1 proved the deformation of formation of hydrogen bonds [39]. The peak from 1246.0 to 1246.7 cm−−1 proved the deformation of the carbonyl group (C H) present in cellulose, hemicellulose, and lignin [29]. Aromatic ring in lignin the carbonyl group (C−−H) present in cellulose, hemicellulose, and lignin [29]. Aromatic ring in lignin was identified at the peak from 1600 to 1585 cm 1 and 1500 to 1400 cm 1 [44]. All BAIC samples was identified at the peak from 1600 to 1585 cm−1 and− 1500 to 1400 cm−1 [44].− All BAIC samples showed showed a peak between 1451.61 and 1451.96 cm 1 that manifested the stretching of C=C (aromatic ring) a peak between 1451.61 and 1451.96 cm−1 that− manifested the stretching of C=C (aromatic ring) in in lignin present in BioAsh. Lignin reported a decomposition temperature of 150–900 C in the lignin present in BioAsh. Lignin reported a decomposition temperature of 150–900 °C in◦ the TGA TGA [22,54,55]. This wide range of decomposition temperatures of lignin further predicted its present [22,54,55]. This wide range of decomposition temperatures of lignin further predicted its present in in the BAIC samples. Polymer compounds associated with aromatic rings such as lignin are reported to the BAIC samples. Polymer compounds associated with aromatic rings such as lignin are reported to be advantageous for thermal stability and char barrier formation [45]. Wide band position from 3100 to be advantageous for thermal stability and char barrier formation [45]. Wide band position from 3100 3600 cm 1 was exhibited in all samples. This could be attributed to the stretching of the hydroxyl group to 3600 −cm−1 was exhibited in all samples. This could be attributed to the stretching of the hydroxyl (O–H) [46] that derived from intense intermolecular and intramolecular of hydrogen bonds. This O–H group (O–H) [46] that derived from intense intermolecular and intramolecular of hydrogen bonds. functional group could contribute to the interfacial hydrogen bonding between BAIC samples and the This O–H functional group could contribute to the interfacial hydrogen bonding between BAIC steel surface [47]. samples and the steel surface [47]. 3.7. Freeze–Thaw Cycle 3.7. Freeze–Thaw Cycle Any cracking, color change, blister, and coagulum on all BAIC coating samples were visually Any cracking, color change, blister, and coagulum on all BAIC coating samples were visually assessed. The appearance of BAIC 1-0, BAIC 3-5, BAIC 5-0, BAIC 10-0 after 70 freeze–thaw cyclic tests assessed. The appearance of BAIC 1-0, BAIC 3-5, BAIC 5-0, BAIC 10-0 after 70 freeze–thaw cyclic tests is shown in Figure 11. is shown in Figure 11.

(a) (b)

Coatings 2020, 10, x FOR PEER REVIEW 14 of 21

Figure 10. Infrared spectra of BAIC samples.

BAIC samples incorporated with a different weight percentage of BioAsh exhibited variation in peak position. Peak ranging from 2915.7 to 2916.6 cm−1 indicated the symmetrical stretching of C–H of alkyl chain in the methyl group [42]. Stretching of C=O (carbonyl group) in lignin and hemicellulose that originated in BioAsh (rubberwood biomass) displayed at peak 1727.2 to 1727.9 cm−1. −C−O (ester bond) and −C=O (ester carbonyl) that were sensitive to form a hydrogen bond can be found in vinyl acetate copolymer. The transmittance of –C–O and –C=O showed peaks ranging from 1246.0 to 1246.7 cm−1 and 1727.2 to 1727.9 cm−1, respectively, in all BAIC samples proving the formation of hydrogen bonds [39]. The peak from 1246.0 to 1246.7 cm−1 proved the deformation of the carbonyl group (C−H) present in cellulose, hemicellulose, and lignin [29]. Aromatic ring in lignin was identified at the peak from 1600 to 1585 cm−1 and 1500 to 1400 cm−1 [44]. All BAIC samples showed a peak between 1451.61 and 1451.96 cm−1 that manifested the stretching of C=C (aromatic ring) in lignin present in BioAsh. Lignin reported a decomposition temperature of 150–900 °C in the TGA [22,54,55]. This wide range of decomposition temperatures of lignin further predicted its present in the BAIC samples. Polymer compounds associated with aromatic rings such as lignin are reported to be advantageous for thermal stability and char barrier formation [45]. Wide band position from 3100 to 3600 cm−1 was exhibited in all samples. This could be attributed to the stretching of the hydroxyl group (O–H) [46] that derived from intense intermolecular and intramolecular of hydrogen bonds. This O–H functional group could contribute to the interfacial hydrogen bonding between BAIC samples and the steel surface [47].

3.7. Freeze–Thaw Cycle Any cracking, color change, blister, and coagulum on all BAIC coating samples were visually assessed. The appearance of BAIC 1-0, BAIC 3-5, BAIC 5-0, BAIC 10-0 after 70 freeze–thaw cyclic tests Coatingsis shown2020 in, 10 Figure, 1117 11. 15 of 21

(a) (b)

Coatings 2020, 10, x FOR PEER REVIEW 15 of 21

Figure 11. After 70 cyclic tests: (a) BAIC 1-0, (b) BAIC 3-5, (c) BAIC 5-0 and (d) BAIC 10-0. Figure 11. After 70 cyclic tests: (a) BAIC 1-0, (b) BAIC 3-5, (c) BAIC 5-0 and (d) BAIC 10-0. Obvious coagula were accumulated in BAIC 5-0 and BAIC 1-0. BAIC 3-5 had shown much lesser Obvious coagula were accumulated in BAIC 5-0 and BAIC 1-0. BAIC 3-5 had shown much lesser and finer coagulum in size compared to BAIC 5-0 and BAIC 1-0. These coagula were noticed when the and finer coagulum in size compared to BAIC 5-0 and BAIC 1-0. These coagula were noticed when samples were in an air-dried condition before the freeze–thaw cycle test. After the test, coagula in the samples were in an air-dried condition before the freeze–thaw cycle test. After the test, coagula three samples remained unchanged. No coagulum was found in BAIC 10-0. The coagula condition in three samples remained unchanged. No coagulum was found in BAIC 10-0. The coagula condition was due to the difference in mix proportion of BioAsh and TiO2. Coagula was greatly relevant to the was due to the difference in mix proportion of BioAsh and TiO2. Coagula was greatly relevant to the combination of lower BioAsh and higher TiO2 composition in the formula. All coating samples were combination of lower BioAsh and higher TiO2 composition in the formula. All coating samples were attached firmly to the steel plate. No crack, blister and color change were found in all coating samples attached firmly to the steel plate. No crack, blister and color change were found in all coating samples after the test. This study revealed BioAsh had positive synergy effect with TiO2-APP-PER-MEL/VAC on after the test. This study revealed BioAsh had positive synergy effect with TiO2-APP-PER-MEL/VAC weather resistance of BAIC samples. Other factors could be subjected to very minor C–S–H compound on weather resistance of BAIC samples. Other factors could be subjected to very minor C–S–H potentially formed from the reactions of SiO2, CaO, CaCO3 that assisted in enhancing the coating’s compound potentially formed from the reactions of SiO2, CaO, CaCO3 that assisted in enhancing the strength against weathering test. coating’s strength against weathering test. 3.8. Static Immersion 3.8. Static Immersion Water resistance of all BAIC samples were investigated using static immersion. The weight change of samplesWater duringresistance the testof all was BAIC measured samples every were one-hour investigated interval using up to static 5 h. Waterimmersion. absorption The weight rate of allchange samples, of samples as shown during in Figure the test 12 .was measured every one-hour interval up to 5 h. Water absorption rate ofBAIC all samples, 3-5 with as 3.5 shown wt.% BioAshin Figure had 12. the lowest water absorption rate at 8.72% with the smallest standard deviation20 of 2.57 as compared to other samples. The standard deviations of samples BAIC 1-0, BAIC 5-0 and BAIC 10-0 were 5.99, 2.74 and 3.44, respectively. In contrast, BAIC 1-0 exhibited the highest water absorption rate at 19.09% after 5 h. Water permeation in BAIC 1-0 added with 1.0 wt.% BioAsh and 9.015 wt.% TiO2 was the highest among all. BAIC 10-0 added with 10.0 wt.% of BioAsh had a lower water absorption rate at 10.89% compared to BAIC 1-0. This demonstrated the higher water absorption rate was found to be correlated with the increase in TiO2 composition.BAIC 1-0 TiO2 with 10 stronger hydrophilic properties than BioAsh reduced the water resistance by allowingBAIC 3-5 more water particles to penetrate the sample. Inter-surface structure of BAIC samples varied due to different BAIC 5-0 BioAsh: TiO2 5ratio in APP-PER-MEL/VAC formula. BioAsh with a particle size of 300 µm had the largest particles among all ingredients. BAIC 10-0 with the highest portion of 10.0 wt.%BAIC BioAsh 10-0 added forming moreWater absoprtion rate (%) pores in the surface structure. Porosity in BAIC 10-0 allowed more water particles (0.275 nm) to penetrate0 BAIC 10-0, causing decrement in water resistance as compared to BAIC 3-5 and BAIC 5-0. An012345 appropriate ratio of 3.5 wt.% BioAsh and 6.5 wt.% TiO2 in BAIC 3-5 was able to Time (hr)

Figure 12. Water absorption rate of BAIC samples.

BAIC 3-5 with 3.5 wt.% BioAsh had the lowest water absorption rate at 8.72% with the smallest standard deviation of 2.57 as compared to other samples. The standard deviations of samples BAIC 1-0, BAIC 5-0 and BAIC 10-0 were 5.99, 2.74 and 3.44, respectively. In contrast, BAIC 1-0 exhibited the highest water absorption rate at 19.09% after 5 h. Water permeation in BAIC 1-0 added with 1.0 wt.% BioAsh and 9.0 wt.% TiO2 was the highest among all. BAIC 10-0 added with 10.0 wt.% of BioAsh had a lower water absorption rate at 10.89% compared to BAIC 1-0. This demonstrated the higher water absorption rate was found to be correlated with the increase in TiO2 composition. TiO2 with stronger

Coatings 2020, 10, x FOR PEER REVIEW 15 of 21

Figure 11. After 70 cyclic tests: (a) BAIC 1-0, (b) BAIC 3-5, (c) BAIC 5-0 and (d) BAIC 10-0.

Obvious coagula were accumulated in BAIC 5-0 and BAIC 1-0. BAIC 3-5 had shown much lesser and finer coagulum in size compared to BAIC 5-0 and BAIC 1-0. These coagula were noticed when the samples were in an air-dried condition before the freeze–thaw cycle test. After the test, coagula in three samples remained unchanged. No coagulum was found in BAIC 10-0. The coagula condition was due to the difference in mix proportion of BioAsh and TiO2. Coagula was greatly relevant to the combination of lower BioAsh and higher TiO2 composition in the formula. All coating samples were attached firmly to the steel plate. No crack, blister and color change were found in all coating samples Coatings 2020, 10, 1117 16 of 21 after the test. This study revealed BioAsh had positive synergy effect with TiO2-APP-PER-MEL/VAC on weather resistance of BAIC samples. Other factors could be subjected to very minor C–S–H compoundslow down potentially the water formed permeation from the rate reactions due to the of formation SiO2, CaO, of CaCO a better3 that particles assisted distribution in enhancing in the the coating’sinter-surface strength matrix. against Incorporation weathering of test. the poorly soluble Al(OH)3 in IC can help to enhance the water resistance [44]. Hence, it is assumed minor Al(OH)3 that potentially present in 3.5 wt.% may assist in 3.8. Staticpromoting Immersion a better water resistance in BAIC 3-5 without compromising the fire-resistance outcomes. However, hydroxyl groups (O–H), as identified in FTIR can readily form hydrogen bonds to increase Waterthe water resistance solubility of of theall compound.BAIC samples The various were numberinvestigated of O–H using groups static existed immersion. in each BAIC The coating weight changesample of samples resulted during in the the diff erenttest was water-resistance measured ever performance.y one-hour BAIC interval 3-5 wasup to predicted 5 h. Water to have absorption the rate ofleast all O–Hsamples, groups as dueshown to its in best Figure water 12. resistance among all others.

BAIC 1-0 10 BAIC 3-5 BAIC 5-0 5 BAIC 10-0 Water absoprtion rate (%) 0 012345 Time (hr) Figure 12. Water absorption rate of BAIC samples. Figure 12. Water absorption rate of BAIC samples. It can be suggested that the best water resistance of BAIC 3-5 manifested an appropriate amount BAICof 3.5 3-5 wt.% with BioAsh 3.5 wt.% was BioAsh more likely had tothe produce lowest awater better absorption inter-surface rate matrix at 8.72% and with bonding the smallest with standardTiO2 -APP-PER-MELdeviation of 2.57/VAC as compared to minimize to the other water samples. absorption The rate. standard Water deviations contact angle of (wettability)samples BAIC 1-0, BAICof diff 5-0erent and BAIC BAIC samples, 10-0 were BioAsh, 5.99, TiO 2.742 could and 3.44, influence respectively. the water-resistance In contrast, properties BAIC 1-0 and exhibited further the investigation is required. highest water absorption rate at 19.09% after 5 h. Water permeation in BAIC 1-0 added with 1.0 wt.% BioAsh3.9. and Pull-o 9.0ff wt.%Adhesion TiO Strength2 was the highest among all. BAIC 10-0 added with 10.0 wt.% of BioAsh had a lower water absorption rate at 10.89% compared to BAIC 1-0. This demonstrated the higher water Adhesion strength was referred to the physical and chemical bonding of IC samples applied on absorption rate was found to be correlated with the increase in TiO2 composition. TiO2 with stronger the steel surface, and the strength required to entirely detach the coating samples from the steel surface. The crack charge value and adhesion strength of all samples are shown in Figure 13.

BAIC 3-5 achieved the highest adhesion strength at 1.73 MPa, followed by BAIC 1-0 at 1.37 MPa and BAIC 5-0 at 1.51 MPa. BAIC 10-0 displayed the lowest adhesion strength value at 1.07 MPa. BAIC 10-0 with 10.0 wt.% BioAsh exhibited the weakest surface bonding to steel. An increment in BioAsh composition was not correlated to an increment in adhesion strength. BAIC 1-0 with 1.0 wt.% BioAsh showed a better adhesion strength to steel surface as compared to BAIC 10-0. This could be attributed to better interfacial bonding between 1.0 wt.% BioAsh, polymer binder and steel surface. The adhesion strength of BAIC 3-5 and BAIC 5-0 was better than BAIC 1-0. A higher amount of TiO2 at 9.0 wt.% in the formula reduced the adhesion strength of BAIC 1-0 to steel [48]. Single use of TiO2 as the only mineral filler in the IC formulation was unideal as this will lead to a tremendous loss in adhesion strength of the coating to the steel plate [31]. Hybridization of mineral fillers is required. This study revealed BAIC 3-5 with 3.5 wt.% BioAsh and 6.5 wt.% TiO2 were able to achieve the best synergic effect of interfacial bonding to the steel surface resulting in the strongest adhesion strength. An appropriate amount of Mg(OH)2 and CaCO3 help to improve the adhesion strength of the IC [49]. A stronger bonding between the metal surface and Mg/Ca interface allowed for a better Coatings 2020, 10, x FOR PEER REVIEW 16 of 21 hydrophilic properties than BioAsh reduced the water resistance by allowing more water particles to penetrate the sample. Inter-surface structure of BAIC samples varied due to different BioAsh: TiO2 ratio in APP-PER-MEL/VAC formula. BioAsh with a particle size of 300 μm had the largest particles among all ingredients. BAIC 10-0 with the highest portion of 10.0 wt.% BioAsh added forming more pores in the surface structure. Porosity in BAIC 10-0 allowed more water particles (0.275 nm) to penetrate BAIC 10-0, causing decrement in water resistance as compared to BAIC 3-5 and BAIC 5-0. An appropriate ratio of 3.5 wt.% BioAsh and 6.5 wt.% TiO2 in BAIC 3-5 was able to slow down the water permeation rate due to the formation of a better particles distribution in the inter-surface matrix. Incorporation of the poorly soluble Al(OH)3 in IC can help to enhance the water resistance [44]. Hence, it is assumed minor Al(OH)3 that potentially present in 3.5 wt.% may assist in promoting a better water resistance in BAIC 3-5 without compromising the fire-resistance outcomes. However, hydroxyl groups (O–H), as identified in FTIR can readily form hydrogen bonds to increase the water solubility of the compound. The various number of O–H groups existed in each BAIC coating sample resulted in the different water-resistance performance. BAIC 3-5 was predicted to have the least O– H groups due to its best water resistance among all others. It can be suggested that the best water resistance of BAIC 3-5 manifested an appropriate amount of 3.5 wt.% BioAsh was more likely to produce a better inter-surface matrix and bonding with TiO2- APP-PER-MEL/VAC to minimize the water absorption rate. Water contact angle (wettability) of different BAIC samples, BioAsh, TiO2 could influence the water-resistance properties and further investigation is required.

3.9. Pull-offCoatings Adhesion2020, 10, 1117Strength 17 of 21 Adhesion strength was referred to the physical and chemical bonding of IC samples applied on the steelstress surface, transfer and [53 the]. Magnesium strength and required calcium compositionto entirely thatdetach existed the in 3.5coating wt.% BioAsh samples were from highly the steel surface. Therelevant crack to thecharge best interfacial value and bonding adhesion in BAIC strength 3-5. of all samples are shown in Figure 13.

Figure 13. Crack charge and adhesion strength of BAIC c samples. Figure 13. Crack charge and adhesion strength of BAIC c samples. It can be suggested that an appropriate amount of Mg and Ca elemental composition present in 3.5 wt.% BioAsh were vital to the adhesion strength of the coating to the steel surface. BAIC 3-5 achieved the highest adhesion strength at 1.73 MPa, followed by BAIC 1-0 at 1.37 MPa However, Al(OH)3 in BioAsh will compromise the adhesion strength of the samples. Quantitative and BAICanalysis 5-0 at of1.51 3.5 MPa. wt.% BioAsh BAIC was10-0 essential displayed in future the lowest research. adhesion Factors such strength as variation value in theat 1.07 interface MPa. BAIC 10-0 withregion, 10.0 wt.% the structure BioAsh of exhi atomicbited bonding, the weakest toughness surface of fracture, bonding purity, to and steel. thickness An increment will also have in BioAsh compositionimpacts was on not the adhesioncorrelated strength. to an increment in adhesion strength. BAIC 1-0 with 1.0 wt.% BioAsh showed a4. better Conclusions adhesion strength to steel surface as compared to BAIC 10-0. This could be attributed to better interfacial bonding between 1.0 wt.% BioAsh, polymer binder and steel surface. The In this research project, all samples BAIC were conducted via the Bunsen burner test, adhesioncarbolite strength furnace of BAIC test, TGA 3-5 test,and FTIR BAIC test, 5-0 SEM was/EDX better test, Instron than pull-oBAICff adhesion1-0. A higher test, water amount resistance of TiO2 at 9.0 wt.% testin the and freeze–thawformula reduced resistance the test adhesion to evaluate strength the performances of BAIC of fire1-0 resistance, to steel [48]. mechanical, Single thermal, use of TiO2 as the only chemicalmineral and filler physical in the properties. IC formulation Research findingswas un wereideal concluded as this aswill follows: lead to a tremendous loss in

1. BAIC 3-5 sample with 3.5 wt.% BioAsh revealed the best results in ﬁre resistance performance, thermal, physical, and mechanical properties. It had the lowest temperature in ﬁre resistance test at only 112.5 ◦C, highest residual weight in TGA at 29.48 wt.%, thickest char layer of 13 mm in carbolite furnace test, more uniform and denser multicellular rosette-like char structure present in SEM, lowest water absorption rate and strongest adhesion strength at 1.73 MPa.

2. BAIC 3-5 showed the highest residual weight of 29.48% at 1000 ◦C in TGA. The synergistic reactions of CaCO3 and Al(OH)3 present in 3.5 wt.% BioAsh with 6.5 wt.% TiO2-APP-PER-MEL/VAC were able to minimize the thermal degradation from a high temperature. 3. 3.5 wt.% BioAsh in BAIC 3-5 stimulated the carbon composition to 42.96 wt.% and lessened the oxygen composition to 29.93 wt.%, resulting in the lowest O/C ratio against oxidation. 1 4. All BAIC samples displayed a wide band position at 3100 to 3600 cm− , indicating the presence of the O—H functional group for interfacial hydrogen bonding to the steel surface. The peak at 1 1451.61 to 1451.96 cm− showed in all samples demonstrated the existence of an aromatic ring of lignin present in BioAsh (rubber hardwood ash) accountable for the thermal stability and the formation of good-quality char. Coatings 2020, 10, 1117 18 of 21

5. The inclusion of BioAsh enabled all BAIC samples to withstand freeze–thaw cycles, which could be attributed to the C–S–H compound formed.

6. The static immersion test showed an appropriate amount of Al(OH)3 present in BioAsh (3.5 wt.%), hybridized with TiO2 (6.5 wt.%) in APP-PER-MEL/VAC in BAIC 3-5 enhanced the water resistance. BAIC 3-5 was predicted with the least O-H groups that water solubility was minimized.

7. The signiﬁcant synergistic eﬀects between 3.5 wt.% BioAsh and 6.5 wt.% TiO2 in BAIC 3-5 were most stimulated by a series of complex reactions derived from the combination Si, Ca, Mg, Al, P, C and lignin found in the BioAsh. The performance of BAIC 3-5 was the most prominent among all.

Fire protection performance of BAIC 3-5 with lowest equilibrium temperature at 112.5 ◦C had outstripped the conventional coatings of water-based (220–290 ◦C) found in the literatures. Moreover, there was 24 ◦C lowest equilibrium temperature difference between BAIC 3-5 (112.5 ◦C) and the water-based IC using aquaculture waste (136 ◦C) reported in the literature. BioAsh contained several elements—Ca, Al, Mg, Si, P, C and lignin (aromatic ring)—at different ratios and significantly enhanced the fire resistance, thermal stability and mechanical strength of BAIC 3-5. These elements that existed in the BioAsh were not commonly reported in the IC formulas using natural-based substances such as clam shell, eggshell and vegetable compounds in the literature. It can be summarized that 3.5 wt.% BioAsh has great potential to be renewed as novel green mineral filler in the IC formulation. The utilization of BioAsh as a natural substitute in the IC could minimize the rely on exotic industrial fillers and contribute to a more sustainable environment.

Author Contributions: Interpretation, J.H.B. and M.C.Y.; methodology, J.H.B.; validation, J.H.B., M.C.Y., M.K.Y. and L.H.S.; formal analysis, J.H.B.; investigation, J.H.B. and M.C.Y.; resources, J.H.B.; data collection, J.H.B.; writing—original draft preparation, J.H.B.; writing—review and editing, M.C.Y.; visualization, M.K.Y. and L.H.S.; supervision, M.C.Y.; project administration, M.C.Y.; funding acquisition, J.H.B. All authors have read and agreed to the published version of the manuscript. Funding: This project was funded by the Universiti Tunku Abdul Rahman under the Universiti Tunku Abdul Rahman Research Fund (UTARRF). Acknowledgments: The authors would like to thank Universiti Tunku Abdul Rahman for the Universiti Tunku Abdul Rahman Research Fund (UTARRF), grant number IPSR/RMC/UTARRF/2019-C2/B01 support in this research. Conﬂicts of Interest: There is no conﬂict of interest declared by authors.

References

1. Prabir, B. Biomass Gasification, Pyrolysis and Torrefaction: Practical Design and Theory, 3rd ed.; Academic Press: Cambridge, MA, USA, 2018. 2. Sigmann, S.B. Playing with fire: Chemical safety expertise required. J. Chem. Educ. 2018, 95, 1736–1746. [CrossRef] 3. Harada, T. Nuclear flash burns: A review and consideration. Burn. Open 2018, 2, 1–7. [CrossRef] 4. Altarawneh, M.; Saeed, A.; Al-Harahsheh, M.; Dlugogorski, B.Z. Thermal decomposition of brominated flame retardants (BFRs): Products and mechanisms. Prog. Energy Combust. Sci. 2019, 70, 212–259. [CrossRef] 5. Upadhya, S.C. Intumescent coating in the service of fire protection. Paint India 2000, 50, 45–52. Available online: https://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=831167 (accessed on 8 September 2020). 6. Bourbigot, S.; Le Bras, M.; Duquesne, S.; Rochery, M. Recent advances for intumescent polymers. Macromol. Mater. Eng. 2004, 289, 499–511. [CrossRef] 7. Zia-ul-Mustafa, M.; Ahmad, F.; Ullah, S.; Amir, N.; Gillani, Q.F. Thermal and pyrolysis analysis of minerals reinforced intumescent fire retardant coating. Prog. Org. Coat. 2017, 102, 201–216. [CrossRef] 8. Amir, N.; Ahmad, F.; Megat-yusoff, P.S.M. Study on the fibre reinforced epoxy-based intumescent coating formulations and their char characteristics. J. App. Sci. 2011, 11, 1678–1687. [CrossRef] 9. Duquesne, S.; Magnet, S.; Jama, C.; Delobel, R. Intumescent paints: Fire protection coatings for metallic substrates. Surf. Coat. Technol. 2004, 302, 180–181. [CrossRef] Coatings 2020, 10, 1117 19 of 21

10. Duquesne, S.; Magnet, S.; Jama, C.; Delobel, R. Thermoplastic resins for thin film intumescent coatings-towards a better understanding of their effect on intumescent efficiency. Polym. Degrad. Stab. 2005, 88, 63–69. [CrossRef] 11. Yang, H.; Yu, B.; Song, P.; Maluk, C.; Wang, H. Surface-coating engineering for flame retardant flexible polyurethane foams: A critical review. Compos. Part B Eng. 2019, 176, 107–185. [CrossRef] 12. Wang, G.; Yang, J. Influences of binder on fire protection and anticorrosion properties of intumescent fire resistive coating for steel structure. Surf. Coat. Technol. 2010, 204, 1186–1192. [CrossRef] 13. Yew, M.C.; Ramli Sulong, N.H. Fire-resistive performance of intumescent flame-retardant coatings for steel. Mater. Des. 2012, 34, 719–724. [CrossRef] 14. Yew, M.C.; Ramli Sulong, N.H.; Yew, M.K.; Amalina, M.A.; Johan, M.R. Fire propagation performance of intumescent fire protective coatings using eggshells as a novel biofiller. Sci. World J. 2014, 2014, 805094. [CrossRef][PubMed] 15. Jimenez, M.; Duquesne, S.; Bourbigot, S. Characterization of the performance of an intumescent fire protective coating. Surf. Coat. Technol. 2006, 201, 979–987. [CrossRef] 16. Laoutid, F.; Bonnaud, L.; Alexandre, M.; Lopez-Cuesta, J.M.; Dubois, P. New prospects in flame retardant polymer materials: From fundamentals to nanocomposites. Mater. Sci. Eng. R Rep. 2009, 63, 100–125. [CrossRef] 17. Hull, T.R.; Witkowski, A.; Hollingbery, L. Fire retardant action of mineral fillers. Polm. Degrad. Stab. 2011, 96, 1462–1469. [CrossRef] 18. Natural Rubber Statistic 2018—Malaysia Rubber Board. Available online: http://www.lgm.gov.my/nrstat/ Statistics%20Website%202018%20(Jan-Dec).pdf (accessed on 29 June 2020). 19. Rubber Asia: Malaysia for Model Sustainable Rubber Industry. Available online: https://rubberasia.com/ 2017/11/28/malaysia-model-sustainable-rubber-industry-dr-zairossani-bin-mohd-director-general-mrb (accessed on 29 June 2020). 20. Demeyer, A.; Nkana, J.; Verloo, M. Characteristics of wood ash and influence on soil properties and nutrient uptake: An overview. Bioresource. Tecnol. 2001, 77, 287–295. [CrossRef] 21. Tarun, R.N.; Rudolph, N.K.; Rafat, S. Use of Wood Ash in Cement-Based Materials; A CBU Report, CBU-2003-19 (REP-513); University of Wisconsin-Milwaukee: Milwaukee, WI, USA, 2003. 22. Waters, C.L.; Janupala, R.R.; Mallinson, R.G.; Lobban, L.L. Staged thermal fractionation for segregation of lignin and cellulose pyrolysis products: An experimental study of residence time and temperature effects. J. Anal. Appl. Pyro. 2017, 126, 380–389. [CrossRef] 23. ASTM C 618-05; Standard Specification For Coal Fly Ash and Raw or Calcined Natural Pozzolans for Use in Concrete; American Society of Testing and Materials: West Conshohocken, PA, USA, 2005. 24. Ramazani, S.A.A.; Rahimi, A.; Frounchi, M.; Radman, S. Investigation of flame retardancy and physical-mechanical properties of zinc borate and aluminium hydroxide propylene composites. Mater. Des. 2008, 20, 1051–1056. [CrossRef] 25. Chuang, Y.J.; Chuang, Y.H.; Lin, C.Y. Fire tests to study heat insulation scenario of galvanized rolling shutters sprayed with intumescent coatings. Mater. Des. 2009, 30, 2576–2583. [CrossRef] 26. Suardana, N.P.G.; Ku, M.S.; Lim, J.K. Effects of diammonium phosphate on the flammability and mechanical properties of biocomposites. Mater. Des. 2011, 32, 1990–1999. [CrossRef] 27. Han, Z.D.; Fina, A.; Malucelli, G.; Camino, G. Testing fire protective properties of intumescent coating by in-line temperature measurements on a cone calorimeter. Prog. Org. Coat. 2010, 69, 475–480. [CrossRef] 28. De Souza, M.M.; De Sá, S.C.; Zmozinski, A.V.; Peres, R.S.; Ferreira, C.A. Biomass as the Carbon Source in Intumescent Coatings for Steel Protection against Fire. Ind. Eng. Chem. Res. 2016, 55, 11961–11969. [CrossRef] 29. De Sá, S.C.; De Souza, M.M.; Peres, R.S.; Zmozinski, A.V.; Braga, R.M.; Melo, D.M.D.A.; Ferreira, C.A. Environmentally friendly intumescent coatings formulated with vegetable compounds. Prog. Org. Coat. 2017, 113, 47–59. [CrossRef] 30. Li, H.F.; Hu, Z.W.; Zhang, S.; Gu, X.Y.; Wang, H.J.; Jiang, P.; Zhao, Q. Effects of titanium dioxide on the flammability and char formation of water-based coatings containing intumescent flame retardants. Prog. Org. Coat. 2015, 78, 318–324. [CrossRef] 31. Yew, M.C.; Sulong, N.R.; Yew, M.K.; Amalina, M.A.; Johan, M.R. Influences of flame-retardant fillers on fire protection and mechanical properties of intumescent coatings. Prog. Org. Coat. 2015, 78, 59–66. [CrossRef] Coatings 2020, 10, 1117 20 of 21

32. Anna, P.; Marosi, G.; Csontos, I.; Bourbigot, S.; Le Bras, M.; Delobel, R. Influence of modified rheology on the efficiency of intumescent flame retardant systems. Polym. Degrad. Stab. 2001, 74, 423–426. [CrossRef] 33. Karunadasa, K.S.P.; Manoratne, C.H.; Pitawala, H.M.T.G.A.; Rajapakse, R.M.G. Thermal decomposition of calcium carbonate (calcite polymorph) as examined by in-situ high-temperature X-ray powder diffraction. J. Phys. Chem. Solids. 2019, 134, 21–28. [CrossRef] 34. Conceicao, T.F.d.; Scharnagl, N. Fluoride conversion coatings for magnesium and its alloys for the biological environment. In Surface Modification of Magnesium and Its Alloys for Biomedical Applications, 1st ed.; Sankara-Narayanan, T.S.N., Park, I.-S., Lee, M.-H., Eds.; Woodhead Publishing: Cambridge, UK, 2015; pp. 3–21. [CrossRef] 35. Wypych, G. Fillers—Origin, Chemical Composition, Properties and Morphology: Handbook of Fillers, 4th ed.; ChemTec Publishing: Toronto, ON, Canada, 2016. 36. Phosphoric Acid. Wiley Online Library. Available online: https://onlinelibrary.wiley.com/doi/abs/10.1002/ 0471743984.vse5529 (accessed on 17 June 2020). 37. Beh, J.H.; Yew, M.C.; Yew, M.K.; Saw, L.H. Fire protection performance and thermal behavior of thin film intumescent coatings. Coatings 2019, 9, 483. [CrossRef] 38. Etiegni, L.; Campbell, A.G. Physical and chemical characteristics of wood ash. Bioresour. Technol. 1991, 37, 173–178. [CrossRef] 39. Wang, Y.C.; Zhao, J.P.Facile preparation of slag or fly ash geopolymer composite coatings with flame resistance. Const. Build. Mat. 2019, 203, 655–661. [CrossRef] 40. Hytonen, J.; Nurmi, J.; Kaakkurivaara, N.; Kaakkurivaara, T. Rubber tree (Hevea brasiliensis) Biomass, nutrient content, and heating values in southern Thailand. Forests 2019, 10, 638. [CrossRef] 41. Kraszkiewicz, A.; Kachel-Jakubowska, M.; Niedziolka, I. The chemical composition of ash from the plant biomass in terms of indicators to assess slagging and pollution of surface heating equipment. Fresenius Environ. Bull. 2017, 26, 6383–6389. 42. Aziz, H.; Ahmad, F. Effects from nano-titanium oxide on the thermal resistance of an intumescent fire retardant coating for structural applications. Prog. Org. Coat. 2016, 101, 431–439. [CrossRef] 43. Yew, M.C.; Yew, M.K.; Saw, L.H.; Ng, T.C.; Rajkumar, D.; Beh, J.H. Influence of nano bio-filler on the fire-resistive and mechanical properties of water-based intumescent coatings. Prog. Org. Coat. 2018, 124, 33–40. [CrossRef] 44. Cayla, A.; Rault, F.; Giraud, S.; Salaün, F.; Sonnier, R.; Dumazert, L. Influence of Ammonium Polyphosphate/Lignin Ratio on Thermal and Fire Behavior of Biobased Thermoplastic: The Case of Polyamide 11. Materials 2019, 12, 1146. [CrossRef] 45. Fan, F.Q.; Xia, Z.B.; Li, Q.Y.; Li, Z. Effects of inorganic fillers on the shear viscosity and fire retardant performance of waterborne intumescent coatings. Prog. Org. Coat. 2013, 76, 844–851. [CrossRef] 46. Hong, S.M.; Soleimani, M.; Liu, Y.Q.; Winnik, M.A. Influence of a hydrogen-bonding co-monomer on polymer diffusion in poly(butyl acrylate-co-methyl methacrylate) latex films. Polymers 2010, 51, 3006–3013. [CrossRef] 47. Jessica, J.K.Y.; Yew, M.C.; Yew, M.K. Development of advanced intumescent flame-retardant binder for fire rated timber door. MATEC Web Conf. 2020, 306, 02006. [CrossRef] 48. Yew, M.C.; Sulong, N.H.R.; Yew, M.K.; Amalina, M.A.; Johan, M.R. Eggshells: A novel bio-filler for intumescent flame-retardant coatings. Prog. Org. Coat. 2015, 81, 116–124. [CrossRef] 49. Li, G.Q.; Zhang, C.; Lou, G.B.; Wang, Y.C.; Wang, L.L. Assess the fire resistance of intumescent coatings by equivalent constant thermal resistance. Fire Technol. 2012, 48, 529–546. [CrossRef] 50. Wang, Y.C.; Zhao, J.P. Benign design and the evaluation of pyrolysis kinetics of polyester resin based intumescent system comprising of alkali-activated silica fume. Prog. Org. Coat. 2018, 122, 30–37. [CrossRef]

51. Devendra, K.; Rangaswamy, T. Determination of mechanical properties of Al2O3, Mg(OH)2 and Sic filled E-glass/epoxy composites. Int. J. Eng. Res. Appl. 2012, 2, 2028–2033. 52. Li, Y.Z.; Feng, Y.W.; Xu, Z.S.; Yan, L.; Xie, X.J.; Wang, Z.Y. Synergistic effect of clam shell bio-filler on the fire-resistance and char formation of intumescent fire-retardant coatings. Prog. Org. Coat. 2020, 9, 9479–12030. [CrossRef] 53. Nasir, K.M.; Sulong Ramli, N.H.; Johan, M.R.; Afifi, A.M. Synergistic effect of industrial and bio-fillers waterborne intumescent hybrid coatings on flame retardancy, physical and mechanical properties. Prog. Org. Coat. 2020, 149, 105905. [CrossRef] Coatings 2020, 10, 1117 21 of 21

54. Yang, H.; Yan, R.; Chen, H.; Lee, D.H.; Zheng, C. Characteristics of hemicellulose, cellulose and lignin pyrolysis. Fuel 2007, 86, 1781–1788. [CrossRef] 55. Lv, D.; Xu, M.; Liu, X.; Zhan, Z.; Li, Z.; Yao, H. Eﬀect of cellulose, lignin, alkali and alkaline earth metallic species on biomass pyrolysis and gasiﬁcation. Fuel Process Technol. 2010, 91, 903–909. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).