Evaluation of the Flame-Retardant Performance and Fire Risk of Cellulose Building Finishing Material Due to the Particle Size of Expandable Graphite

Abstract

The burning of building finishing materials containing chemical substances can lead to the spread of fire with a large number of harmful substances. To prevent this, cellulose, an eco-friendly material that minimizes the generation of harmful substances, was chosen as the main material for building finishing materials. Wastepaper was recycled to extract cellulose, and a finishing material was manufactured by mixing in expandable graphite and magnesium hydroxide. The flame-retardant performance of the finishing material was evaluated by measuring the total heat emission rate using the ISO 5660-1 cone calorimeter, with the particle size of the expanded graphite as a variable. The measured physical properties value was used as the FDS parameter to evaluate the risk of fire. Results show that the cellulose-based building finishing material meets the life safety standard of NFSC 203 during the simulation time and has a FED value that does not exceed 0.001 at maximum. This confirms its effectiveness in providing safe egress time for residents.

1. Introduction

The cost of fire losses worldwide is estimated to be approximately 1% of global GDP per year, and enacting measures to protect citizens and properties from fire is becoming increasingly important [1]. Fire is a potential risk that can threaten life and property. However, as the construction industry grows rapidly due to the development of technology, fire accidents are also occurring frequently, and interest in this phenomenon is gradually increasing due to the seriousness of fires in buildings [2,3]. In contrast, according to the distribution of fires by type in 24 countries in 2019, fires in all structures, including 22.4% in residential facilities, accounted for 31.6% of the total number of fires [4].

In addition, an average of 41,738 fires occurred in South Korea from 2016 to 2020, resulting in 2322 casualties (334 deaths, 1988 injuries) and KRW 589.2 billion in property damage. Among them, the spread of combustion caused by the rapid combustion of combustible materials accounts for 17.34% of the total fires, and the fatality rate is 34.88% [5].

The main causes of casualties caused by fires in indoor buildings include heat, smoke, and toxic gas generated by combustible materials, especially in the case of building finishing materials comprising chemicals and numerous harmful substances during combustion [6]. Furthermore, the smoke generated by combustion contains soot particles and steam; death accidents caused by smoke occur frequently due to the high toxic property [7]. In addition, the analysis of the rise in temperature over time is necessary to understand fire ignition and development [8], which can be predicted through the heat release rate (HRR) and total heat release (THR) rate values [9]. Therefore, the components and flame-retardant performance of building finishing materials are crucial factors in measures to reduce human casualties.

However, the risk to those staying in the room also increases proportionally with the amount of fire and combustion products, as well as with the characteristics of the building finishing materials currently in use [10]. In particular, polyurethane foam is a material with a wide range of applications, such as insulation, buffering, etc., and is often used as a finishing material for building insulation. However, this material has the disadvantages of low thermal stability and mechanical strength [11,12]. Research on flame-retardant materials is continuously being conducted to reduce this risk.

In an existing study, Rie et al. announced that fire-retardant materials absorbed in paper honeycomb can be used as a building interior material, as well as can improve flame-retardant performance by conducting a cone calorimeter flammability test [13].

Vojta et al. aggregated statistics on the types of flame retardants and analyzed the current status of flame retardants in use by checking the types and ingredients of flame retardants used as building materials. This has shown that the content of flame retardants in most building materials is lower than the level that guarantees the flame-retardant function [14].

Almirón et al. confirmed that natural zeolites collected from volcanic ash affect the synergistic action of flame-retardant systems comprising ammonium polyphosphate, pentaerythritol, and polypropylene [15].

Han et al. melted and mixed polylactic acid (PLA) with flame retardants manufactured on the basis of polyacrylamide (PAM), sodium alginate, and coated ammonium polyphosphate (APP) and developed a polyacrylic acid (PLA) composite material with excellent tensile force and flame-retardant performance [16].

Guo produced a flame retardant mixed with magnesium hydroxide and sodium dodecylbenzene sulfonate and developed a flame-retardant PS composite plate mixed with polystyrene [17]. Some previous studies on flame retardant are summarized in

Table 1. Previous studies on flame retardant.

Authors	Content	Reference No.
Rie et al. (2012)	A Study on the Improvement of Flame-Retardant Performance for the Use of Construction Finishing Materials in Paper Beehive	[13]
Vojta et al. (2017)	Statistical analysis of the type and content of flame retardants used	[14]
Almirón et al. (2022)	Analysis of the effect of zeolites of volcanic ash on the synergistic effects of flame-retardant systems	[15]
Han et al. (2022)	Development and verification of PLA composite materials with improved flame retardants through the mixing of polylactic acid (PLA) and flame retardant	[16]
Guo (2022)	Development of flame-retardant polystyrene composite panels using magnesium hydroxide flame retardant	[17]

.

The recycling ratio of paper waste generated worldwide is only up to 65%, and efforts are being made to improve recycling rates [18]. This study reviews the use of wastepaper as a building finishing material with flame-retardant properties, incorporating a recycling method for wastepaper. The building finishing material was produced by mixing expanded graphite and magnesium hydroxide with cellulose extracted from waste-corrugated cardboard, and the material’s fire characteristics were verified. The variable for flame-retardant performance is the particle size of expanded graphite, and the verification procedure was conducted in an experiment using the ISO 5660-1 [19] cone calorimeter. In addition, the values of the physical properties obtained from the experiment results were applied to the FDS, which is a fire simulation, and a quantitative evaluation of the effect of materials manufactured on indoor persons during combustion was conducted through a fire hazard assessment. The fire hazard assessment was analyzed by measuring the values of FED, Temperature, Visibility, O₂, CO, and CO₂. Figure 1 represents the flow chart of this study.

2. Experimental

2.1. Materials and Characterization

2.1.1. Flame Retardant

Flame retardants are materials added to prevent or delay further progression of ignition by various physical and chemical methods in the event of a fire, and various types of flame retardants, such as bromide, chlorine, transfer, and inorganic flame retardants, are available. However, in the case of some flame retardants, such as brominated flame retardants, their use is regulated due to toxic substances generated during combustion [20,21].

Expandable graphite is a mineral flame retardant, a graphite laminar compound widely used to optimize flame-retardant properties and has characteristics of high flame-retardant efficiency compared to low cost [22]. The expandable graphite works by expanding at a temperature of 180 degrees or higher and producing a “worm-like” type of flake that creates a dense carbon layer which helps to suppress heat transfer [23,24]. Furthermore, the larger the particle size of the expandable graphite, the more effective it is to improve the flame-retardant characteristic by a synergistic effect with APP, which is the phosphorus-based flame retardant [25].

Magnesium hydroxide is an inorganic flame retardant, and water and magnesium oxide are produced through a pyrolysis reaction, thereby suppressing fire due to the endothermic reaction, cooling fire extinguishing effects, and demonstrating its nontoxic property. However, low flame-retardant efficiency has disadvantages [26].

2.1.2. Interior Finishing Material Specimen Making

A building finishing material is produced in this study by mixing cellulose with expandable graphite and magnesium hydroxide, which are mineral and inorganic flame retardants, respectively. The components of the produced sample have the same content ratio of 50 w% cellulose, 30 w% expandable graphite, and 20 w% magnesium hydroxide. A variable for flame-retardant performance is also set to the particle size of expandable graphite.

In the process of manufacturing the building finishing material, wastepaper is ground with a grinder and cellulose is extracted. The cellulose is then mixed with water using a mixer, and crushed, and expandable graphite and magnesium hydroxide are added to the mixture. A 3D printer is used [27] to produce a sample with consistent physical properties, which is molded into a size of 10 cm × 10 cm × 2 cm and then dried in a dryer for 48 h. Afterward, the sample is stored in a constant temperature and humidity machine at 25 °C ± 2 °C and 50% humidity for approximately two days. Figure 2 illustrates the process of producing the building finishing material specimen, and

Table 2. Expandable graphite particle size types and manufactured specimen.

Properties	A	B	C
Particle size	74 μm	173 μm	279 μm
Mesh Size	200	80	50
Expandable graphite
Specimen

provides information about the particle type of the expandable graphite used as a variable and the resulting manufactured specimen.

2.2. Cone Calorimeter Method

The cone calorimeter test method begins with the following basic principle: the consumption of 1 kg of oxygen generates 13.1 MJ/kg of heat because the amount of pure combustion heat is proportional to that of the oxygen required for combustion [19]. An expression for determining the heat emission rate by utilizing the amount of oxygen consumption in a combustion process is calculated by Equations (1) and (2) based on the oxygen consumption principle [19].

$${\dot{\mathrm{q}}}^{″}\left(\mathrm{t}\right)=\dot{\mathrm{q}}\left(\mathrm{t}\right)/\mathrm{A}\mathrm{s}$$

(1)

$$\dot{\mathrm{q}}\left(\mathrm{t}\right)=(∆\mathrm{h}\mathrm{c}/{\mathrm{r}}_0)(1.10)\mathrm{C}\sqrt{\frac{∆\mathrm{P}}{\mathrm{T}\mathrm{e}}}\frac{{{\mathrm{X}}^0}_{{\mathrm{O}}_2}-{\mathrm{X}}_{{\mathrm{O}}_2}\left(\mathrm{t}\right)}{1.105-1.5{\mathrm{X}}_{{\mathrm{O}}_2}\left(\mathrm{t}\right)}$$

(2)

where:

• ${\dot{\mathrm{q}}}^{″}$: Heat release rate per unit area (kW/m²)
• $\dot{\mathrm{q}}$: Heat release rate (kW)
• $\mathrm{A}\mathrm{s}$: The official surface area of the specimen (0.01 m²)
• $∆\mathrm{h}\mathrm{c}/{\mathrm{r}}_0$ = 13.1 × 10³ kJ/kg
• $\mathrm{C}$: Oxygen consumption correction constant (m^1/2 kg^1/2 K^1/2)
• $∆\mathrm{P}$: Orifice pressure difference
• $\mathrm{T}\mathrm{e}$: The absolute temperature of a gas in an orifice
• ${{\mathrm{X}}^0}_{{\mathrm{O}}_2}$: Initial value of oxygen analyzer scale
• ${\mathrm{X}}_{{\mathrm{O}}_2}$: oxygen analyzer scale reading of mole fraction of oxygen

Consequently, the HRR, ignition time, oxygen consumption, CO and CO₂ production amount, and the flow rate of the burned gas are all measured. Figure 3 shows the ISO 5660-1 cone calorimeter (FESTEC Co., Seoul, Korea) used in the experiment, and

Table 3. Mechanical Properties and Test Condition of ISO 5660-1 Test.

Mechanical Properties
Size (W × D × H)	1750 mm × 600 mm × 2180 mm
Power	220 V AC, 60 Hz, 30 A
Weight	450 kg
Exhaust	50 L/s
Utilities	Methane, CO (0.8%)/CO2 (8%) Gas, Nitrogen, Dust collector, Compressed Air
Test Condition
Temperature	25 ± 2 °C
Humidity	50%
Flow rate	24 L/s
Radiant Heat	50 kW/m2
Separation distance	22.5 mm

shows the specifications and experimental conditions of the cone calorimeter equipment used in this study.

In addition, building finishing materials should meet the certification criteria for flame-retardant performance specified in ISO 5660-1. The criteria for the flame-retardant performance certification of ISO 5660-1 are as follows [28,29]:

-

Total radiant heat of 5 min after heating is 8 MJ/m²;

-

Within 5 min, the maximum heat radiant rate should not exceed 200 kW/m² for longer than 10 consecutive seconds;

-

No crack should penetrate the sample, hole, or melting (for mixed content materials, including melting and dissipating of all core materials) after heating for 10 min.

The cone calorimeter test measures the specimen three times for the same conditions. The FDS simulation was performed by inputting the measured amount of CO and CO₂ production and the average physical property value of the flow rate of the burned gas. Polyurethane data were measured using data provided to the FDS, and the data on MDF and cellulose-based building finishing materials were measured through a cone calorimeter experiment.

Table 4. Reaction parameter.

Properties	Control Group		Manufactured Specimen
	MDF	Polyurethane Foam	A	B	C
Density (kg/m3)	536.1	40.0	461.0	464.0	477.0
Heat of combustion (kJ/kg)	19,702.5	30,000.0	5193.3	1246.7	1896.7
CO yield (kg/kg)	0.0668	0.0310	0.0234	0.0284	0.0276
Soot yield (kg/kg)	0.009515	0.0310	1.08 × 10−5	1.02 × 10−5	1.00 × 10−6

shows the FDS parameter measured through the cone calorimeter test, and Figure 4 shows the same additional condition, ISO 5660-1 experiments, in triplicate of the fire performance for the THR values of the specimens.

2.3. Fire Analysis

2.3.1. Fire Modeling

In the case of religious facilities, an unspecified number of people are often concentrated; therefore, the scale of damage to those who are present may vary depending on the degree of fire spread inside. Therefore, a fire scenario in religious facilities was set up in this study to conduct a simulation for the evaluation of the spread of fire by building finishing materials. Figure 5 shows the simulation model implemented in FDS.

In this study, the fire was assumed to be caused by the leakage of electrical outlets installed on the wall and the simulation was performed on fire hazard assessments, such as flame diffusion through changes in the type of indoor finishing material attached to the wall. Wooden chairs were arranged inside the religious facilities to understand the spread of fire to combustible materials in the room, and the material of the seats and backrests was set to upholstery. MDF and polyurethane foam were selected and compared as controls to understand the fire performance of the manufactured interior finishing materials. Figure 6 shows a graph of the fire characteristics applied to the simulation and

Table 5. Boundary conditions for fire simulation.

Properties	Condition
Room Size	12 m × 24 m × 5 m
Grid	(80 × 15 × 50) + (120 × 225 × 50)
Heat Release Rate (HRR)	200 kW
Air Temperature	20 °C
Simulation Time	400 s
Thickness of Interior Finishing Materials	2 cm

represents the calculation conditions.

2.3.2. Mesh Resolution

The calculation accuracy strongly depends on the FDS mesh size. The size of the analysis grid can be determined by the characteristic fire diameter using Equation (3) in the fire plume analysis:

$${\mathrm{D}}^{\mathrm{*}}={\left(\frac{\dot{\mathrm{Q}}}{{\mathsf{\rho }}_{\mathrm{\infty }}{\mathrm{C}}_{\mathrm{p}}{\mathrm{T}}_{\mathrm{\infty }}\sqrt{\mathrm{g}}}\right)}^{2/5}$$

(3)

where ${\mathrm{D}}^{\mathrm{*}}$ is the characteristic fire diameter, $\dot{\mathrm{Q}}$ is the THR rate (kW), ${\mathsf{\rho }}_{\mathrm{\infty }}$ is the outside air density (kg/m³), ${\mathrm{C}}_{\mathrm{p}}$ is the specific heat (kJ/kg·K), ${\mathrm{T}}_{\mathrm{\infty }}$ is the outside air temperature (K), $\mathrm{g}$ is the acceleration of gravity (m/s²), and ${\mathsf{\delta }}_{\mathsf{\chi }}$ is the nominal size of a mesh cell.

The ${\mathrm{D}}^{\mathrm{*}}/{\mathsf{\delta }}_{\mathsf{\chi }}$ parameter would only be valid if the value is between 4 and 16 [30]. The ${\mathrm{D}}^{\mathrm{*}}$ applied to this analysis is 0.510 m, and ${\mathrm{D}}^{\mathrm{*}}/{\mathsf{\delta }}_{\mathsf{\chi }}$ has a value of 5.10 in this paper. Therefore, the mesh size applied to the studies satisfies the convergence condition.

3. Results and Discussion

3.1. Analysis Results

3.1.1. Analysis Results of Evacuation Criteria

The fire hazard assessment is based on a comparison of available safe egress time (ASET), where the ASET in a performance-based fire safety design defines temperature, visibility, and O₂, CO, and CO₂ values at a height of 1.8 m from the floor [31].

Table 6. Performance-based design safety criteria [29].

Properties	Condition
Temperature	60 °C
Visibility	10 m
O2	15%
CO	1400 ppm
CO2	5%

shows the life safety standards for each variable.

Temperature is the standard of thermal damage caused by the breathing of the occupant, and the safety limit is specified below 60 °C in the performance-based fire safety design. Figure 7 shows the amount of temperature change for each control group, and

Table 7. Temperature safety criteria arrival time by interior finishing materials.

Properties	Control Group		Manufactured Specimen
	MDF	Polyurethane Foam	A	B	C
Time (s)	37.6	26.8	>400.0	>400.0	>400.0

shows the time when each variable reaches 60 °C.

In the simulation, the distance between the fire position and the measurement point was about 13 m, and the door was set to be open around the measurement point to allow external air to enter. In the case of MDF and polyurethane foam, the fire spread to combustible sheets, resulting in a steep increase in temperature that exceeded the safety criterion temperature of 60 °C at 37.6 s and 26.8 s, respectively. However, during the simulation operation time, the cellulose-based building finishing materials effectively suppressed the spread of fire in the area around the measurement point. As a result, the temperature at the measurement point did not exceed the safety criterion. This phenomenon proves the effect of preventing diffusion of combustion and suppressing internal temperature rise by the flame-retardant performance of the cellulose-based building finishing material.

Visibility stipulates that the movement of the occupant to the exit due to the spread of smoke caused by a fire is a time constraint, and such a movement is safe only when it is secured at least 10 m in the performance-based fire safety design. Figure 8 shows the amount of change in visibility by time zone, and

Table 8. Visibility safety criteria arrival time by interior finishing materials.

Properties	Control Group		Manufactured Specimen
	MDF	Polyurethane Foam	A	B	C
Time (s)	42.0	28.4	>400.0	>400.0	>400.0

shows the visibility 10 m arrival time for each variable.

The simulation found that visibility was reduced when the polyurethane foam and MDF exceeded 28.4 and 42.0 s, respectively, while cellulose-based finishing materials could secure visibility inside fire chambers for 400 s. This finding proves that the low amount of smoke generation, which is a characteristic of the building finishing material, is effective in securing visibility.

Combustion of substances by fire causes a decrease in O₂ and an increase in CO and CO₂ in the room, which raises ASET due to a lack of oxygen and toxic gas poisoning. Therefore, O₂, CO, and CO₂ are dangerous when they are 15% or less, 1400 ppm or more, and 5% or more, respectively.

Figure 9, Figure 10 and Figure 11 show the occurrence amount by the time zone of O₂, CO, and CO₂, and

Table 9. Safety criteria arrival time by interior finishing materials—O₂, CO, and CO₂.

Properties	Control Group		Manufactured Specimen
	MDF	Polyurethane Foam	A	B	C
O2	43.6 s	30.4 s	>400.0 s	>400.0 s	>400.0 s
CO	41.2 s	30.0 s	>400.0 s	>400.0 s	>400.0 s
CO2	44.0 s	30.8 s	>400.0 s	>400.0 s	>400.0 s

shows the arrival time of safety standards of O₂, CO, and CO₂ by finishing materials.

During the simulation, smoke flow due to thermal buoyancy carried CO, CO₂, and O₂, causing them to accumulate and descend from the ceiling. For MDF and polyurethane foam, which experienced rapid temperature increases, the concentrations of CO, CO₂, and O₂ exceeded the life safety standards at an average time of 42.3 s and 30.4 s, respectively. However, it was confirmed that the cellulose-based building finishing material manufactured in this study did not exceed the life safety standards during the simulation time, providing occupants with sufficient time to escape.

The simulation results show that the cellulose building finishing material reduces carbon products, such as CO and CO₂, as well as flame-retardant performance, depending on the comparison results for each variable in the human safety standard.

3.1.2. Fractional Effective Dose Analysis Results

Fractional effective dose (FED) is a behavioral inability assessment criterion generated by hazardous gas concentrations in the respiratory region of the occupant. The calculation of FED is determined by CO, CO₂, and O₂ and computed by a time-based integration method. Equations (4)–(7) show the FED formula of the FDS User’s guide [32].

$${\mathrm{F}\mathrm{E}\mathrm{D}}_{\mathrm{t}\mathrm{o}\mathrm{t}}={\mathrm{F}\mathrm{E}\mathrm{D}}_{\mathrm{C}\mathrm{O}}\times {\mathrm{H}\mathrm{V}}_{{\mathrm{C}\mathrm{O}}_2}+{\mathrm{F}\mathrm{E}\mathrm{D}}_{{\mathrm{O}}_2}$$

(4)

$${\mathrm{F}\mathrm{E}\mathrm{D}}_{\mathrm{C}\mathrm{O}}=4.607\times {10}^{-7}{\left({\mathrm{C}}_{\mathrm{C}\mathrm{O}}\right)}^{1.036}\mathrm{t}$$

(5)

$${\mathrm{F}\mathrm{E}\mathrm{D}}_{{\mathrm{O}}_2}=\frac{\mathrm{t}}{60\mathrm{e}\mathrm{x}\mathrm{p}\left[8.13-0.54\left(20.9-{\mathrm{C}}_{{\mathrm{O}}_2}\right)\right]}$$

(6)

$${\mathrm{H}\mathrm{V}}_{{\mathrm{C}\mathrm{O}}_2}=\frac{\mathrm{e}\mathrm{x}\mathrm{p}\left(0.1930{\mathrm{C}}_{{\mathrm{C}\mathrm{O}}_2}+2.0004\right)}{7.1}$$

(7)

where:

• t = unit time,
• ${\mathrm{C}}_{\mathrm{C}\mathrm{O}}$ = CO concentration(ppm),
• ${\mathrm{C}}_{{\mathrm{O}}_2}$ = O₂ concentration (%),
• ${\mathrm{C}}_{{\mathrm{C}\mathrm{O}}_2}$ = CO₂ concentration (%).

The FED value of 0.3 is generally presented as the human body incapacitation index [33]. Figure 12 shows the amount of FED change by each control group, and

Table 10. FED safety criteria arrival time and maximum by interior finishing materials.

Properties	Control Group		Manufactured Specimen
	MDF	Polyurethane Foam	A	B	C
Time (s)	46.4	32.8	>400.0 s	>400.0 s	>400.0 s
Maximum value	120.73	88.78	0.00004	0.00051	0.00026

shows the time of the FED value reaching 0.3 and the maximum value.

The FED value exceeded the standard values of the MDF (46.4 s) and polyurethane foam (32.8 s). However, the manufactured building finishing material did not exceed the standard values during the simulation operation time, and the maximum value was less than 0.001, which was effective in securing ASET.

3.2. Fire Location Change

Predicting the location of a fire that occurs indoors is impossible; thus, checking the spread of fire due to changes in combustible materials is necessary. Therefore, an additional simulation was performed in this study, assuming that a fire broke out in the seating area of the chair.

The fire location was set to the center of the first chair seat in the right row, and the reaction material was set to polyurethane. The material selected for the fire position was ‘UPHOLSTERY’, the same material used for the sofa provided in the FDS version 5 user guide, and the volume was set to 1.4 × 0.5 × 0.2 m³. Only cellulose-based materials, excluding the control group, were applied to the interior building finishing materials. Figure 13 represents the simulation model implemented in FDS, and

Table 11. Reaction parameters.

Properties	Condition
Reactant of fuel	C = 6.3, H = 7.1, N = 1.0, O = 2.1
Molecular weight of fuel	130.3 g/mol
CO yield (kg/kg)	0.0310
Soot yield (kg/kg)	0.10

shows the reaction parameters.

Analysis Results due to Change in Fire Source

Changes in the fire position assume variations in the ignition sources and combustible materials, facilitating the identification of the fire diffusion inhibition effect of interior finishing materials installed on the wall.

Figure 14 shows the change in the time zone of the safety criteria and the FED in the event of a fire, and

Table 12. Safety criteria and FED arrival time by interior finishing materials.

Properties	A	B	C
Temperature	>400.0 s	>400.0 s	>400.0 s
Visibility	>400.0 s	>400.0 s	>400.0 s
O2	>400.0 s	>400.0 s	>400.0 s
CO	>400.0 s	>400.0 s	>400.0 s
CO2	>400.0 s	>400.0 s	>400.0 s
FED	>400.0 s	>400.0 s	>400.0 s

shows the arrival time.

The spread of indoor fires is caused by direct propagation to nearby combustibles or by the room temperature rising to a high temperature of 500–600 °C, igniting the surface of the combustibles [34]. However, the simulation confirmed that the flame-retardant performance of the cellulose-based finishing material prevented the spread to the surrounding combustibles for a period of 400 s. Therefore, no exceedance of the life safety standard of NFSC 203 was observed at the measurement point near the entrance, such as an increase in temperature or a decrease in visibility. This demonstrates the importance of flame-retardant performance in interior finishing materials and indicates that the manufactured cellulose-based construction finishing materials have an excellent effect on ensuring evacuation safety for occupants.

Research on cellulose as a building insulation material continues for recycled wastepaper and reducing environmental impact. Previous studies have investigated the addition of cellulose to construction materials, such as cement matrix and polymer foam, to confirm the improvement in insulation performance and thermal conductivity. Additionally, it was found that paper cellulose exhibits thermal stability up to 250 °C [35,36,37]. After verifying the flame-retardant performance of cellulose-based building finishing materials, this study confirmed the effect of securing ASET and suppressing fire spread. Based on these results, a new plan for recycling wastepaper into high-value-added products was presented.

3.3. Discussions

The result confirmed that the increase in the size of the expandable graphite particles was affected by the decrease in THR, but no significant difference was observed when the particle was larger than a certain size. A trend line for THR based on the particle size of expandable graphite is presented in Equation (8).

$$\mathrm{y}=0.00027{\mathrm{x}}^2-0.12\mathrm{x}+16.29$$

(8)

Table 13. Cone calorimeter test result.

Properties	No.	THR at 300 s (MJ/m2)	Mean (MJ/m2)	SD (THR)
A	1	8.754	8.795	0.357
	2	8.379
	3	9.251
B	1	3.261	3.384	0.217
	2	3.203
	3	3.689
C	1	3.895	3.462	0.753
	2	2.402
	3	4.088

represents THR, Mean, and SD from the cone calorimeter experiment results, and Figure 15 shows the distribution of THR after heating for 5 min based on the size of expandable graphite particles.

Flame retardant performance improves as the particle size of expandable graphite increases, with improvement when the particle size is greater than 150 μm [25,38].

This study confirmed the effect of the particle size on expanded graphite added to cellulose extracted from wastepaper on the Total Heat Release (THR). The experiment showed that after 300 s, the THR value exceeded 8 MJ/m², which means that it did not meet the flame-retardant standard. However, cases B and C, which had particle sizes exceeding 100 μm, met the flame-retardant standard.

It was confirmed that the safety of cellulose-based building finishing materials was ensured by meeting the safety criteria and FED values. However, to further understand the difference in particle size on the expandable graphite, the cumulative variation over time considering the FED value by the ignition source was identified as an integral value.

Consequently, the cumulative variation of FED values due to the increase in the size of expandable graphite particles showed different results depending on the ignition source.

Table 14. FED sum and integral value due to fire source change.

Properties		A	B	C
Sum	location wall	0.00547	0.07099	0.03736
	location chair	0.18742	0.18279	0.18258
Integration	location wall	0.00217	0.02819	0.01484
	location chair	0.07460	0.07278	0.07269

shows the value of FED accumulated during simulation time according to ignition source conditions.

Analysis results based on the integrated value of time revealed that the fire did not spread to the wall surface, and no significant difference was observed in the case of the chair equipped with the first ignition source. However, the results were different from the THR distribution when the construction finishing material on the wall was the first cause of ignition. A trend line on the effect of the particle size of expandable graphite on the amount of FED cumulative change is shown in Equation (9).

$$\mathrm{y}=-6.76962\times {10}^{-5}{\mathrm{x}}^2+0.02797\mathrm{x}-4.3626$$

(9)

Figure 16 shows the distribution of FED cumulative changes based on the particle size of expandable graphite.

4. Conclusions

Fire characteristics, such as flame-retardant performance and combustion heat, were derived in this study through a cone calorimeter test on a building finishing material made by adding expandable graphite and magnesium hydroxide based on cellulose extracted from wastepaper. ASET was then analyzed through the performance-based fire safety design evaluation of human safety standards and FED by reflecting the identified physical property values in the FDS. The following results were obtained.

(1) ISO 5660-1 test confirmed that the flame-retardant performance of the building finishing materials made with a mass ratio of cellulose, expandable graphite, and magnesium hydroxide of 50:30:20 is ensured when the particle size is over 100 μm. Furthermore, the results also showed that thermal conductivity and soot yield consistently decreased due to increased particle size. However, the THR value tends to rise slightly when the particle size of expandable graphite exceeds a certain size, as confirmed by the ISO 5660-1 test.

(2) The ASETs of the control groups, MDF and polyurethane, showed an average of 42.5 s and 29.9 s, respectively, during fire simulation. However, it has been confirmed that the ASET of the experimental group’s building finishing material exceeded the simulation time of 400 s, regardless of the ignition source. This clarified that cellulose-based construction finishing materials extracted from wastepaper is effective in suppressing fire spread, including flame retardant performance, and securing the safety of occupants’ evacuation from dangerous factors such as smoke and CO.

(3) The particle size and THR of expandable graphite, a cellulose-based building finishing material, can be represented by a trend equation, such as Equation (8), thus, confirming the correlation between expandable graphite and THR.

(4) The maximum value of FED, a cellulose-based building finishing material, was 0.001 or less, proving that the breath of the occupants was slightly affected. Furthermore, the results confirmed that the CO yield value of the cone calorimeter test affects the amount of FED cumulative change. Consequently, the correlation between the particle size of expandable graphite and the FED cumulative variation can be expressed by a trend equation of Equation (9).

(5) Based on the results of the fire risk evaluation, the possibility of using cellulose-based building finishing materials extracted from waste cardboard was verified. However, in terms of recycling, additional discussions are needed on reducing flame retardant materials, such as expandable graphite and increasing cellulose. In addition, XRD and FTIR characterization-based analysis can be used to accurately understand the physical and chemical composition of the mixture of cellulose and flame-retardant materials and to develop efficient materials.

In addition, this study evaluated the fire performance of building finishing materials based on the performance-based fire safety design. Further research is needed to investigate the ASET according to the height of the ceiling and the area of the fire room, considering the behavior characteristics of smoke that move downwards to the ceiling by thermal buoyancy.