Research on Fireproof and Anti-Corrosion Integrated Coatings for Modular Integrated Buildings

Abstract

With the development of prefabricated buildings, the challenge of integrating fireproofing and anti-corrosion in steel structures has become increasingly prominent. Based on epoxy resin, we developed a multifunctional coating with high-flame retardant efficiency and corrosion resistance, which can be employed in the key parts of modular integrated construction (MiC), thereby enhancing the safety of the prefabricated buildings. Experimental data showed that the fire resistance limitation reached 124 min, the salt spray resistance 2540 h, and the adhesion grade 1. The limiting oxygen index (LOI) of the cured coating was 45%, corresponding to the V-0 classification in the vertical burning test from Underwriters Laboratories Inc. (Northbrook, IL, USA) (UL 94). Compared with the latest studies, the integrated formulation exhibits simultaneous gains in fire and corrosion protection, offering a promising single-layer solution for MiC.

1. Introduction

Against the backdrop of global industrialization and carbon neutrality goals, energy conservation, cost reduction, and green development have become core challenges in the construction industry. Prefabricated buildings, especially steel-structured MiC, have emerged as a key direction for industrial transformation due to their high assembly rate (>90%), low carbon emissions, and rapid construction. According to the Global Prefabricated Building Market Report 2024, the global prefabricated building market size exceeded USD 1.2 trillion in 2023, with MiC accounting for over 15% and growing at a CAGR of 8.5%. In regions such as Singapore and Japan, MiC has been widely applied in public housing and infrastructure projects, but its promotion is limited by the lack of efficient fireproof and anti-corrosion solutions for steel structures.

According to the China Steel Structure Industry Development Report (2021), for instance, Chinese steel structure output has exceeded 1.3 billion tons, accounting for over 50% of global production. The Chinese government requires “accelerating new-type building industrialization, vigorously developing prefabricated buildings, and promoting steel structure housing”, with especial emphasis on “actively advancing high-quality steel structure residential construction.” As a representative of steel structure prefabricated buildings, MiC has become a main direction for construction industrialization transformation due to its high assembly rate (over 90%), low-carbon features, and rapid construction advantages. However, MiC has significant challenges in coordinating fire resistance with anti-corrosion in practical applications. Traditional protection processes require layered coating of primer, intermediate paint, a fireproof layer, and a topcoat (4–5 working procedures). This involves conflicts between the prolonged construction cycles and material performance, the flammable corrosion-resistant resins, and acidic components, such as ammonium polyphosphate, which can accelerate steel corrosion. Additionally, the module gaps of MiC are prone to water infiltration and can serve as channels for fire spread during fire hazards. Statistical data showed that in the year 2021, steel structure buildings accounted for 27% of fire incidents caused by construction defects in China, with direct economic losses beyond 5 billion yuan.

Current MiC steel structure protection mainly relies on a layered coating of anti-corrosion and fireproof paints, respectively [1,2,3,4]. Anti-corrosion mechanisms based on galvanic cell principles typically use metal or metal oxide powders such as sacrificial anodes. However, corrosion rates multiply when coatings are locally damaged. As far as we know, fireproof coatings are normally categorized into intumescent and non-intumescent types. Intumescent coatings form carbonized insulation layers through thermal expansion but exhibit poor weather resistance and contain excessive inorganic fillers, such as aluminum hydroxide, which can reduce the coating adhesion. Non-intumescent coatings, such as cement-based thick coatings, require thick applications (≥15 mm), occupying space and easily delaminating, in spite of high fire resistance.

Global research on flame-retardant and corrosion-resistant synergistic coatings primarily focuses on filler modification and formulation optimization, with limited publications addressing integrated fire–corrosion protection through the coordination of self-flame-retardant resins and fillers [5,6]. Most intumescent fireproof coatings for steel structures employed the phosphorus–nitrogen intumescent flame retardants that reached high-efficiency fire protection by releasing inert gases in the gas phase and forming the carbonization layer in the condensed phase. However, these fireproof coatings suffered from insufficient corrosion resistance [7,8,9,10,11]. Recent advancements introduced nanomaterials, such as vermiculite and mica, into both intumescent and non-intumescent fireproof coating systems, enhancing their specific surface area and shielding effects. For instance, nano-vermiculite (50–100 nm) with lamellar structures can prolong corrosion medium penetration paths and enhance the salt spray resistance, while mica powder can improve mechanical stability by filling microcracks [12,13,14,15]. Anti-corrosion coatings typically incorporated zinc powder, zinc phosphate, or iron oxide additives into resins. Commercially available anti-corrosion coatings generally exhibit low oxygen index and lack fire resistance, being flammable during fires [16,17]. Current research on dual-functional fireproof and anti-corrosion coatings is largely limited to the laboratory scale, lacking engineering validation for MiC’s special structures, such as corrosion protection of connection nodes and the fireproofing of module gaps.

Recent international efforts have focused on inherently flame-retardant epoxy matrices. As far as we know, epoxy resin has high strength and rigidity, can withstand large pressure and impact, has a high surface hardness and good wear resistance, and has excellent corrosion resistance to acids, alkalis, salts, and other chemical substances. In addition, good thermal stability and excellent insulation features endow its use as various electronic components. More importantly, this kind of resin is also a recyclable material with low environmental pollution. As a result, the fireproofing of epoxy resin and the corresponding application have attracted great attention [18,19,20,21,22]. For instance, Rybyan et al. cured DER-331 with arylaminocyclotriphosphazenes, achieving a V-0 rating with only 6 wt % P [23]. A hyperbranched flame retardant SD containing P, N, and Si was successfully synthesized for the manufacture of high-performance, flame-retardant EP systems by Wang et al. [24]. However, in most studies on fireproof coatings, it is hard to find reports about resistance to neutral salt spray exceeding 2000 h. It can be seen that the design of synergistic resins/fillers that coordinate high barrier and carbon formation functions is highly necessary.

Current technologies present three major limitations: (1) single functionality failing to meet MiC’s complex operational requirements; (2) cumbersome construction processes with poor quality control; and (3) absence of systematic solutions for critical components, especially in connection nodes. To address these challenges, this study proposes a multifunctional coating technology based on modified epoxy resin, achieving integrated improvement in both fireproofing and anti-corrosion by virtue of molecular modification and multiscale synergistic design.

The present work departs from previous studies by (i) employing a self-crosslinking P-N epoxy that functions simultaneously as corrosion barrier and char former, (ii) replacing conventional multilayer practice (4–5 coats) with a single-layer direct-to-metal system, and (iii) validating performance on MiC-specific joints and module gaps rather than flat steel panels only. This innovation not only simplifies construction processes (60% procedure reduction) but also significantly reduces lifecycle maintenance costs, providing a groundbreaking solution for efficient protection of prefabricated buildings.

2. Materials and Methods

2.1. Main Raw Materials

Matrix material: self-made resin modified from bisphenol A epoxy resin (EPON 828, HexionSpecialty Chemicals, Columbus, OH, USA), weight per epoxide 200–240 g/eq; viscosity at 25 °C 8000–12,000 mPas; n values for the epoxy oligomer were 4–8; exhibiting strong adhesion (≥4 MPa) and low curing shrinkage (<2%). Flame retardant: modified phosphorus–nitrogen flame retardant (self-developed). Nano-additives: nano-vermiculite (50–100 nm, Zhejiang Fenghong New Materials Co., Ltd., Huzhou, China), mica powder (1–5 μm, China National Nuclear Corporation, Shaoguan, China).

2.2. Preparation of Modified Resin

Ordinary bisphenol A epoxy resin lacks prominent flame-retardant properties and is typically used in anti-corrosion coatings with limited performance, heavily relying on the fillers for corrosion resistance. In this study, bisphenol A epoxy resin was chemically modified using phosphorus- and nitrogen-containing molecules to synthesize a hybrid polymeric resin, incorporating phosphorus and nitrogen elements in the modified resin. The phosphorus–nitrogen polymer inherently exhibited excellent flame retardancy, while the cured resin showed great adhesion to metal substrates. Figure 1 illustrated the synthetic design and reaction equations of the self-made modified resin.

2.3. Experimental Procedure

Under nitrogen protection, 100 g (0.25–0.26 mol) of bisphenol A epoxy resin (EPON 828), 0.5 mol% DBU catalyst (relative to epoxy resin, 0.12 g), 100 g xylene, and 100 g chloroform were added to a 500 mL three-neck flask and stirred at 300 rpm for 30 min. A mixed solution of 32.4 g (0.15 mol, 0.6 equivalent relative to epoxy groups) of diethyl aminomethyl phosphonate and 50 g chloroform was slowly added dropwise (2 h). After the addition, the mixture was heated to 60 °C and reacted for 8 h. The product was purified by vacuum distillation (temperature 60–80 °C, pressure 0.08–0.1 MPa) to remove solvents, yielding a viscous hybrid liquid (self-made modified resin) with a yield of 85%–90%. Upon reaction completion, vacuum distillation yielded a viscous hybrid liquid, designated as the self-made resin. The reaction pathway was shown in Figure 1. During the reaction, the amino group in diethyl aminomethyl phosphonate attacked the epoxide groups in the bisphenol A epoxy resin. By controlling the dosage of diethyl aminomethyl phosphonate to 0.6 equivalents, a product of epoxide ring-opening derivatives was obtained. The GPC figure was listed in Figure 2. The mean Mw of the self-made resin is about 4600. From the ³¹P NMR, it was found that the chemical shift was moved from 27.53 to 27.75 ppm. From the ¹H NMR, it was found that diethyl aminomethyl phosphonate was bonded to the epoxy resin to give the phosphorous-containing resin, as shown in Figure 3 and Figure 4.

Also, the IR spectra indicated that there was a broad peak at 3375 cm⁻¹ related to the stretching vibration of -OH. Compared to the original IR (Figure 5), the OH group was increased, implying that some of the epoxy groups were opened to form OH group. The peak around 2983–2893 cm⁻¹ should be the stretching vibration absorption of -CH₂-, and the absorptions at 1019 cm⁻¹ can be ascribed to the P-O-C bending vibration. A characteristic peak of P=O stretching vibration should be 1215 cm⁻¹, as shown in Figure 6. In terms of these data analysis, the self-made resin should contain the phosphorous functional groups.

2.4. Paint Preparation

Specific preparation steps:

The scale of coating preparation is 1 kg. The coating formulation was listed in Table 1. The dispersion disk with an outer diameter of 100 mm was used, and the entire preparation process is carried out at 25 °C.

(1)

At room temperature, we added self-made resin and the corresponding solvent to the paint mixing tank. We started stirring at a speed of 800 rpm until a vortex appeared at the center of the liquid surface;

(2)

Dispersants, defoamers, wetting agents, and anti-settling agents were slowly added to a flask, while leveling agents and other additives were added into the preparation tank from step (1). Then, the modified phosphorus nitrogen flame retardant, carbonization agent, and foaming agent were added into the preparation tank. After the addition of these reagents, the reaction mixture was dispersed at a stirring speed of 1800 rpm for 35 min to ensure thorough mixing. Afterwards, the pigment filler was added to the mixing tank and dispersed at a stirring speed of 2200 rpm for 45 min. Finally, the mixture was further stirred and dispersed at a speed of 1800 rpm for 65 min.

(3)

The curing agent was added according to the actual required ratio, followed by thorough stirring and curing. Then, this mixture was filtered with a fine mesh strainer. The shelling-out can be carried out with a brush coating or an airless spray coating on the surface of the substrate. Before coating, the surface needed to be thoroughly polished to remove dust and rust. The treated surface should be kept clean and coated within 2 h.

Table 1. Coating formulation (phr, parts per hundred resin by mass).

Material	Mass Fraction (Phr)	Material	Mass Fraction (Phr)
Self-made resin	100	Flame retardant	40–65
Dispersant	0.5–1.5	Charring agent	22–35
Defoamer	0.5–1.5	Blowing agent	38–58
Wetting agent	0.5–1.5	Color fillers	22–40
Anti-settling agent	0.5–2.5	Solvent	50–100
Leveling agent	0.5–2.5	Curing agent	10–40

Note: Color fillers are a mixture of titanium dioxide and zinc phosphate (mass ratio 1:1); solvent is xylene or xylene/n-butanol (mass ratio 3:1–3:2); and curing agent is amine (diethylene triamine), polyamide (polyamide 650), or polyether amine (D230).

Table 1. Coating formulation (phr, parts per hundred resin by mass).

Material	Mass Fraction (Phr)	Material	Mass Fraction (Phr)
Self-made resin	100	Flame retardant	40–65
Dispersant	0.5–1.5	Charring agent	22–35
Defoamer	0.5–1.5	Blowing agent	38–58
Wetting agent	0.5–1.5	Color fillers	22–40
Anti-settling agent	0.5–2.5	Solvent	50–100
Leveling agent	0.5–2.5	Curing agent	10–40

Note: Color fillers are a mixture of titanium dioxide and zinc phosphate (mass ratio 1:1); solvent is xylene or xylene/n-butanol (mass ratio 3:1–3:2); and curing agent is amine (diethylene triamine), polyamide (polyamide 650), or polyether amine (D230).

According to the preparation steps, we adjusted the ratio of each component in the formula to give the various formula PT1, PT2, PT3, PT4, and PT5 and the corresponding coatings separately, as shown in

Table 2. The control experimental groups with the variation in the components.

Project	PT1	PT2	PT3	PT4	PT5
Self-made resin	100	100	100	100	100
Flame retardant	55 (P-N Flame retardant)	55 ((NH4)3PO4)	55 (APP)	55 (C2H10N6·H3PO4:NH4)3PO4:APP = 1:2:3	55 (C2H10N6·H3PO4)
Charring agent	26 (Dipentaerythritol)	26 (Starch)	26 (Cyclodextrin/Pentaerythritol = 1:2)	26 (Starch)	Cyclodextrin/Glucose/Pentaerythritol = 1:3:4
Blowing agent	38 (Melamine)	38 (Dicyandiamide)	38 (Melamine cyanurate)	38 (Dicyandiamide)	38 (Dicyandiamide)
Dispersant	1.2	1.2	1.2	1.2	1.2
Defoamer	1.2	1.2	1.2	1.2	1.2
Wetting agent	1.2	1.2	1.2	1.2	1.2
Anti-settling agent	2.3 (Hydrogenated Castor Oil)	2 (polyamide wax)	2 (Bentonite)	2.2 (Hydrogenated Castor Oil)	2.2 (Hydrogenated Castor Oil)
Color fillers	30 (Titanium dioxide + zinc phosphate)	30 (Titanium dioxide + zinc phosphate)	30 (Titanium dioxide + zinc phosphate)	30 (Titanium dioxide + zinc phosphate)	30 (Titanium dioxide + zinc phosphate)
Leveling agent	2	2	2	2	2
Solvent	65 (Xylene)	70 (Xylene: n-butanol = 3:1)	65 (Xylene/n-butanol = 3:2)	70 (Xylene)	70 (Xylene)
Curing agent	25 (Amine-curing agent)	25 (Polyamide-curing agent)	25 (Polyether amine-curing agent)	25 (Amine-curing agent)	25 (Amine-curing agent)

Note: phr (parts per hundred resin) was employed. The unit for each component was mass fraction.

. The experimental results were listed as T1-T5 in Table 2. DT1, DT2, and DT3 were the control experiments. DT1 was a commercially available epoxy based fireproof coating, DT2 was a commercially available epoxy anti-corrosion paint, and DT3 was a clean self-made resin, i.e., the prepared resin without any flame retardants, carbonization agents, or foaming agents, while the other components were the same as those in T1.

Table 3. Test standards and performance indicators.

Test Project	Chinese Standard	Corresponding ISO Standard	Test Indicator
Drying time (25 °C)	GB/T 1728-2020	ISO 9117-1:2013	Surface drying time, actual drying time
Fineness	GB 12441-2018	ISO 1524:2020	≤50 μm
Adhesion	GB 12441-2018	ISO 2409:2020	Grade 1 (cross-cut method)
Limiting oxygen Index (LOI)	GB/T 2406.2-2009	ISO 4589-2:2017	≥32% (V0 class)
Flame endurance time	GB 14907-2018	ISO 834-2:2019	Time to reach critical steel temperature (550 °C)
Neutral salt spray resistance	GB 10125-2012	ISO 9227:2021	No corrosion after 2000 h
Impact resistance	GB/T 1732-2020	ISO 6272-1:2018	50 cm·kg no damage
Flexibility	GB/T 1731-2020	ISO 1519:2002	1 mm bending no crack

presented the test standards and performance indicators, referring to Chinese standards and the corresponding ISO standards. The experimental results were shown in

Table 4. The control experimental results according to the various standards.

Test Project	T1	T2	T3	T4	T5	DT1 (Commercial Fireproof)	DT2 (Commercial Anti-Corrosion)	DT3 (Pure Resin)
Surface drying time (h)	2.8	3.4	3.2	4.0	4.0	5.0	5.2	2.0
Actual drying time (h)	12	14	13	18	17	22	24	12
Fineness (μm)	≤50	≤50	≤50	≤50	≤50	≤50	≤50	≤50
Adhesion (Grade)	1	1	1	1	1	2	1	1
LOI (%)	45	39	41	40	42	20	17	28
Flame endurance time (min)	124	112	109	101	114	100	14	45
Salt spray resistance (h)	2540	2170	2200	2080	2370	168	1060	1690
Impact resistance (cm·kg)	50	50	45	45	48	30	50	50
Flexibility (mm)	1	1	1	1	1	2	1	1

Note: Flame endurance time (min). (1) The target thickness of the dry film of the coating is 4.5 ± 0.2 mm. The test samples are layered with a 0.3–0.5 mm coat according to the coating thickness until the target thickness is reached. The expansion ratio of the carbonized layer is between 20 and 30. (2) The HN400 × 200 hot-rolled H-beam (section coefficient 161 m⁻¹), as specified in GB/T11263-2017, and the 36b hot-rolled I-beam (section coefficient 126 m⁻¹), as specified in GB/T 706-2016, were used as the test base materials. The test piece thermocouple shall be set up in accordance with the relevant provisions of GB/T 9978.6. All tests were conducted in accordance with the corresponding standards, including the surface treatment of the samples, film thickness, number of coatings, drying time between coatings, ambient temperature and humidity, and total curing time, etc.

.

Also, the carbonized layers of T1 and DT3 after combustion were detected by SEM, as shown in Figure 7.

3. Results and Discussion

As shown in Table 2, we initially screened flame retardants, charring agents, and foaming agents, focusing on the limiting oxygen index (LOI) of the coatings. The corresponding results were listed in Table 4. The cured film of DT-3, which was the self-made resin without flame retardants, charring agents, or foaming agents, presented a higher oxygen index of 28. Its flame resistance and neutral salt spray resistance were superior to those of the commercially available anti-corrosion paint DT-2. This indicated that the self-made resin inherently had flame-retardant capability. Moreover, this resin demonstrated excellent compatibility with various flame retardants, charring agents, and foaming agents, allowing various flame-retardant additives to be incorporated effectively. We assumed that the addition of flame retardants could further enhance the overall fire resistance of DT-3.

Next, we optimized dispersants, defoamers, wetting agents, anti-settling agents, and leveling agents. The experimental results indicated that the coatings exhibited sedimentation, sagging, excessive bubbles, and incomplete curing without these additives. After addition of these additives, all of these issues can be solved, which showed the necessity of additive incorporation with the modified resin. At the same time, the resin presented good compatibility with these additives. The additives listed in Table 2 were conventional additives, which were tailored for compatibility with those specific solid components in the coating.

Also, the morphologies of the carbonized layers of T1 and DT3 after combustion were detected by SEM. From the SEM pictures, it was found that the film of T1, which was the self-made resin with different optimized additives, such as flame retardants, charring agents, and foaming agents and so on, can give the integral carbonized layer to resist the heat propagation, as shown Figure 7a. The layer was constituted from small sheets (Figure 7b). In contrast, without the additives, the film of the pure self-made resin could not form integral carbonized layers and instead produced separated carbonized particles, as shown in Figure 7c,d, which consisted of different porous particles. This particle layer could not resist heat propagation.

Finally, the pigments, solvents, and curing agents were screened. After optimization, titanium dioxide and zinc phosphate were chosen as representative pigments across all test groups. Using amine- or polyamide-based curing agents, all coatings successfully cured into films and demonstrated outstanding flame retardancy and anti-corrosion performance. For instance, as shown in T1 in Table 3, the performances of the film showed an oxygen index 45, flame resistance time 124 min, adhesion grade 1, and the neutral salt spray resistance 2450 h.

All of these performance experiments revealed the technical advantages of the self-made resin as below.

• Excellent solubility and dispersibility: The resin kept stable in various polar solvents, such as xylene and alcohols. Moreover, the resin can be cured into a film with high integrity through simple processing.
• Controllable crosslinking: Retained epoxy groups in the resin enable the controlled crosslinking reactions with diverse curing agents, affording larger and more complex polymer networks that enhance water and salt spray resistance.
• Superior interfacial adhesion: The film demonstrated Grade 1 adhesion with cross-cut tests. The salt spray tests indicated that corrosion resistance can exceed that of traditional epoxy zinc-rich anti-corrosion coatings.
• Synergistic flame-retardant effects: Phosphorus–nitrogen structures in the resin backbone, combined with various flame retardants, produce significant synergy, meeting the requirements of the GB14907-2018 national standard.

Figure 8 showed a sample of a fireproof and anti-corrosion integrated coating, which was used for prefabricated building, tested in a salt spray machine after undergoing 2540 h neutral salt spray resistance tests in terms of GB 10125-2012. From the figure, we observed that the coating is basically intact, without obvious damage, such as bulging, cracking, or peeling, which indicated that the T1 film had great anti-corrosion capabilities.

As for the mechanism of fire resistance, we assumed that the phosphorous-containing resin can give various phosphorous radicals at high temperatures. These radicals can catch the carbon radicles to form the stable compounds, resulting in the radicle propagation and extinguishing the flame, as shown Scheme 1.

4. Conclusions

A fireproofing and anti-corrosion integrated coating was developed for prefabricated buildings, which has an oxygen index of 45, a fire resistance limit of 124 min, and a salt spray resistance of 2540 h. The comprehensive performance of this kind of coating is significantly better than that of most of current coatings. The coating has a wide application in MiC structural construction, which can effectively reduce the bearing capacity of connection nodes and the water leakage of module gaps, overcoming the shortcomings in MiC construction. In the future, we will focus on researching the weather resistance of coatings, such as UV aging and freeze–thaw cycles, to promote the development of industry standards and meet the advanced requirements for high-quality prefabricated buildings.