Polybenzoxazine/Epoxy Copolymer Reinforced with Phosphorylated Microcrystalline Cellulose: Curing Behavior, Thermal, and Flame Retardancy Properties

Abstract

This study aims to explore new flame-retardant composites based on a phosphorus-functionalized cellulose derivative and epoxy/benzoxazine thermosetting resins in order to broaden the use of natural fibers in advanced applications. The study involved the phosphorylation of microcrystalline cellulose followed by its characterization through employing various analytical methods to corroborate the accomplishment of its functionalization. The curing behavior of composites based on the polybenzoxazine/epoxy copolymer reinforced with (1 and 5 wt.%) modified microcrystalline cellulose was hereafter considered. The thermal behavior of these composites was correspondingly investigated using thermogravimetric analysis, where improved thermal stability and the limiting oxygen index were stressed. Flame retardancy tests using the vertical burning test UL 94 and heat of combustion analysis utilizing an oxygen bomb calorimeter were also carried out to deeply examine the possible flame retardancy ability of the considered composites.

1. Introduction

Nowadays, the design and development of new materials that consider environmental issues are major concerns of the industry, where the integration of ecological materials is an unavoidable prerequisite. In fact, synthetic fibers make up the majority of mass-market products. Indeed, glass and carbon fibers are currently the most commonly used materials in industry. However, their use creates many environmental problems, mainly high production energy and harmful compound emissions, in addition to the fact that there is no viable and economical solution for recycling conventional fibers without polluting the environment [1,2]. Furthermore, it has been recently observed that the use of carbon and glass fibers has additional shortcomings as they can cause upper respiratory tract, skin, and eye deterioration, and long-term exposure to these compounds may also lead to lung scarring (pulmonary fibrosis) and cancer [3].

Natural fibers are being increasingly used in composites due to cost, functionality, and environmental concerns. During the last 20 years, green and sustainable materials have received significant attention for a variety of uses; in fact, the use of natural fibers in industries dates back to the manufacturing of seats in the early 20th century [4,5]. A significant aspect of this area of interest involves natural fiber-based composites, which have the potential to become a compelling substitute for conventional composites, including plastics. They are expected to be used in high-performance applications such as in the aerospace, construction, and automotive industries, to cite a few [6,7].

The most interesting natural fibers available are cellulose-based fibers. They are a ubiquitous and abundant resource on earth and can be found in many different species; they are becoming increasingly crucial renewable resources for replacing petroleum feedstocks [3,8]. Interest in microcrystalline cellulose (MCC) is primarily due to its excellent mechanical properties, remarkable reinforcing capability, high surface area, and environmental benefits. This renewable nanomaterial has been experiencing exponential growth in interest among scientists and researchers for the development of green polymer composites. Moreover, MCC demonstrates significant potential for grafting or customization with various chemicals to enhance the morphology, thermal stability, and mechanical properties of these composites [3,9].

Epoxy resins are thermosetting polymers that grant a large range of properties such as chemical resistance, strength, stiffness, and dimensional stability [10]. Epoxy-based composites are a very interesting research field, especially for flame retardancy applications, where enhanced flame resistance has been noted [11,12]. Also, polybenzoxazines are a new type of thermosetting polymer that combines the properties of epoxy and phenolic resins [13,14]. Copolymerizing the polybenzoxazine with an epoxy can help increase the crosslink density thanks to the reaction of benzoxazine phenols with epoxy at high temperatures [15]. Additionally, these copolymers show improved mechanical properties, such as up to double the strain-at-breakage and stronger flexural strength [15,16]. With the growing need for new and multifunctional materials, these copolymers can be tailored to meet the thermal and mechanical demands of fireproof materials designed to resist burning or prevent fires from spreading. Indeed, flame retardancy (FR) is essential in applications where the risk of fire is high, such as in building construction, transportation, and electronics [17].

Cellulose-based composites have experienced exponential growth in recent decades [18], yet modifications have to be made to overcome the compatibility problem of hydrophilic cellulose with hydrophobic matrices, especially when it comes to flame retardancy applications since cellulose is considered a flammable material [19]. This kind of modification not only enhances compatibility problems but also helps improve the mechanical and thermal features of the prepared composites. Phosphorous-based flame retardants are gaining attention due to their reduced toxicity and their effectiveness in the condensed phase with the production of shields through char formation and in the gaseous phase through chain reactions [20]. In this case, modifying cellulose through phosphorylation leads to the dehydration of cellulose, thus enhancing char formation and improving its flammability properties [21].

This research was motivated by the urgent need to substitute synthetic fibers with sustainable and biosourced alternatives while retaining all the possibilities for use in various advanced applications. A novel flame retardant composite system designed to enhance fire safety across various applications using natural fibers as reinforcement would contribute to the development of safer and more sustainable materials.

In this paper, we intend to prepare the microcrystalline cellulose/benzoxazine/epoxy composite as a fire-retardant material. The MCC was first modified by phosphorylation (MCC-P), and various structural, thermal, crystalline, and morphological characterizations were achieved to validate the success of the modification. Next, MCC-P/epoxy/benzoxazine composites with different loadings of MCC-P were prepared, and the curing behavior, thermal stability, and flame retardancy features were comprehensively studied.

2. Materials and Methods

2.1. Materials

Commercial microcrystalline cellulose powder was supplied from Avicel MERCK Darmstadt, Germany; analytical grade chemical products such as urea (NH₂CONH₂ 99.5%), sodium phosphate monobasic (H₂NaO₄P > 99%), and diglycidyl ether of bisphenol A (Ep) were supplied by Sigma-Aldrich, Burlington, MA, USA. The bisphenol A and aniline chemicals used were provided by Aladdin reagents Co., Ltd. (Shanghai, China), and the paraformaldehyde reagent was supplied from Sigma-Aldrich. The bisphenol A aniline-based benzoxazine resin (Bz) was synthesized according to the largely used solventless technique [22] in a molar ratio of 1:4:2 (bisphenol A, formaldehyde, and aniline). Adequate amounts of bisphenol A and aniline were placed in a clean and dry boiling flask at 135 °C until a clear solution was formed, paraformaldehyde was then gradually added while stirring continuously to prevent foaming for 2 h, and bisphenol A aniline-based benzoxazine monomer was then obtained.

2.2. Preparation

2.2.1. Phosphorylation Procedure

The phosphorylation of microcrystalline cellulose was carried out according to Ghanadpour et al. [23]. Briefly, 9.1 g of urea and 4.44 g of sodium phosphate monobasic were dissolved in water, then 5 g of microcrystalline cellulose powder was added and stirred for 30 min. After mixing, the product was placed in a Petri dish and dried at 105 °C; the hardening step was held at 150 °C for one hour in an oven. The obtained product was washed with boiling water and water under filtration, and phosphorylated MCC (MCC-P) was then obtained.

2.2.2. Composite Preparation

The copolymer samples were prepared following a specific procedure: 55% of Bz and 45% of Ep (in weight) were mixed at125 °C for 15 min, and then poured in a steel mold and cured at 150 °C/1 h, 165 °C/2 h, and finally 180 °C/3 h. The Bz-Ep copolymer was then obtained. Composites were also obtained following the same procedure and adding the proportions of 1 and 5% in weight of MCC-P to the copolymers in order to obtain Bz-Ep-MCC-P 1% and Bz-Ep-MCC-P 5%, respectively.

2.3. Characterization Methods

2.3.1. Fourier Transform Infra-Red Spectroscopy (FTIR)

The FTIR analysis was accomplished using a Perkin Elmer, SpectrumTwo (Waltham, MA, USA) FTIR spectrometer. The spectra measurement range for wavenumbers was from 400 cm⁻¹ to 4000 cm⁻¹, with a spectral resolution of 2 cm⁻¹ and a scan number of 32.

2.3.2. X-ray Diffraction (XRD) Analysis

The analytical conditions for the two samples MCC and MCC-P were set using a continuous Gonio scan axis and were positioned [°2θ] from 5 to 135 with a step size [°2θ] of 0.0030. A PANalytical X’Pert PRO instrument (Amsterdam, The Netherlands) was used with generator settings of 15 mA and 40 kV.

2.3.3. Scanning Electron Microscopy (SEM)

A JEOL JSM-6360 LV Scanning Electron Microscope (Peabody, MA, USA) operated at 3 kV was used to examine the effect of the phosphorylation of MCC. The samples were prepared by placing them on the sample holder. The coating was ensured using the arc discharge method, and a secondary electron mode was used for morphology scrutinizing.

2.3.4. Thermogravimetric Analysis (TGA)

The TGA analytical conditions were set using a heating rate of 10 °C/min with a temperature range from 50 to 800 °C. The TA-TGA-Q500 instrument (New Castle, DE, USA) was used with a nitrogen flow rate of 10 mL/min and samples weighing approximately 6–7 mg.

2.3.5. Differential Scanning Calorimetry

The curing behavior study was carried out in a non-isothermal mode using a differential scanning calorimeter (DSC) from a Perkin Elmer DSC 8000 analyzer (Waltham, MA, USA). Samples of 5−6 mg were sealed in aluminum pans. The heat ranged from 20 °C to 350 °C with a heating speed of 15 °C/min under a nitrogen gas atmosphere of 30 mL/min.

2.3.6. Calorimetry ISO 1716 Test

The calorimetry ISO 1716 tests [24,25] were performed using a PARR Oxygen bomb calorimeter 1341 for samples with a weight of approximately 500 mg. An oxygen atmospheric gas was used at a pressure of 30 bar.

2.3.7. Vertical Burning Test

The experimental procedure for vertical burning test UL 94 ASTM D 3801 involves rectangular pieces with dimensions of 125 mm × 13 mm × 3 mm in a standard atmosphere of 23 ± 2 °C and 50 ± 5% relative humidity.

3. Results and Discussions

3.1. Characterization of MCC and MCC-P

The FTIR analysis is an effective method for evaluating structural changes that occur during different treatment stages. The obtained spectra for the cellulosic samples are depicted in Figure 1A. MCC-P and MCC showed pronounced spectra similarities that are attributed to cellulose characteristic FTIR bands, i.e., the hydroxyl groups –OH stretching at around 3400 cm⁻¹, the C-H of cellulose CH₂ groups stretching band around 2900 cm⁻¹, C-O bonds of glucosidic linkage stretching between 1100 and 1160 cm⁻¹, and glycosidic linkage between glucose units C-O-C stretching around 1000 cm⁻¹ [26]. However, new absorption bands appeared, confirming the presence of phosphorous groups grafted onto the MCC. These include vibration bands of P-OH at 920 cm⁻¹, the P-O-C aliphatic bond at 830 cm⁻¹, and a band at 1225 cm⁻¹ corresponding to a P=O bonding [27,28].

Thermogravimetric analysis (TGA) is a characterization method used to evaluate the thermal stability of both phosphorylated and non-phosphorylated microcrystalline cellulose samples. The results of thermal stability using TGA and its derivative (DTG) curves are reported in Figure 1B. From the TGA and DTG curves, one main loss is noted for both curves of commercial and phosphorylated MCC, which proves that the phosphorylation process has not influenced the main structure of the pristine MCC. The onset temperatures were determined to be similar at 230 °C, whereas the peak temperatures of maximum degradation (T_max) were at 328.4 °C versus 320.3 °C for MCC and MCC-P, respectively. These results indicate clearly that MCC-P exhibited slightly lower thermal stability compared to pure MCC, owing to the phosphate groups that act as a catalyst for promoting early dehydration and char formation, resulting in a decrease in the cellulose decomposition temperature, which is the main reason for the use of phosphorus as a flame retardant [28,29]. This explains the obtained results for char at 500 °C for MCC-P, which was 15% compared to only 5% for raw MCC.

X-ray diffraction (XRD) is a technique used to analyze the crystal structure of a material. X-rays are directed at the sample, and the interaction of the X-rays with the atoms in the sample causes the X-rays to diffract. The diffracted X-rays are then collected and analyzed to determine the crystal structure of the sample. Figure 1C exhibits the XRD spectra of the studied samples. The phosphorylated MCC preserved the essential spectrum features displayed by native MCC. A typical cellulose I lattice is established with signals appearing at around 15, 16, 22.6, and 34° relating to (1−10), (110), (200), and (004) lattice planes, individually [30]. This suggested that phosphorylation essentially occurred on the surfaces of the crystalline or within the amorphous regions without affecting any substantial change in their crystalline structures. The crystallinity degree is a measure of the degree to which a material’s molecules are ordered into a crystalline structure. In the case of microcrystalline cellulose, this measure refers to the degree to which the cellulose molecules are arranged in a crystalline structure. The crystallinity index (CrI) of the samples was calculated using the Segal formula in Equation (1) [31]:

$$Crl\left(\%\right)={\displaystyle \frac{I_{200}-I_{AM}}{I_{200}}}\times 100$$

(1)

where I₂₀₀ symbolizes the maximum intensity at about 2θ = 22.6° and I_AM denotes the amorphous region at about 2θ = 18°. The obtained results are shown in

Table 1. Crystallinity index of unmodified and modified MCC.

Sample	Crystallinity %
MCC	82.85
MCC P	78.53

.

The CrI results indicate that after phosphorylation, the crystallinity degree of the microcrystalline cellulose decreased from 82.85% to 78.55%. This suggests that the phosphorylation process may have slightly disrupted or altered the crystalline structure of the cellulose due to the effects of temperature, mechanical mixing, and drying, resulting in a less ordered arrangement of the cellulose molecules. The introduced phosphate groups decrease intermolecular and intramolecular hydrogen bonding, which results in a decrease in crystallinity [32].

From the SEM micrographs in Figure 1D, it can be observed that unmodified and modified MCC samples show short and rod-shaped fibers, and rough surfaces are also visible. It can also be noted that the average dimensions of MCC are conserved after modification. Those observations reveal that the modification process did not alter the microstructure of the microcrystalline cellulose. From the above-mentioned results, it can be concluded that the phosphorylation of MCC was successful.

3.2. Curing Behavior Investigation

DSC studied the curing process of the Bz-Ep copolymer and its MCC-P-reinforced composites, and the resulting thermograms are depicted in Figure 2. The non-isothermal runs were performed at a rate of 15 °C/min for the prepared mixtures before curing.

The thermograms show that curing is achieved through an exothermic reaction taking place within a temperature range from 180 to 280 °C, with a unique peak. In fact, the benzoxazine/epoxy copolymer demonstrated the appearance of a single or two separate peaks, each corresponding to each monomer, depending on the ratio and the nature of the resins used [10,33,34]. The Bz resin is known for its unique and dominant autocatalytic process, resulting from the oxazine ring opening and the subsequent formation of a phenolic Mannich bridge to build its three-dimensional network [35] within a temperature range centered at 220 °C [36]. The Bz resin is thus expected to have two reactions for this binary system, where the first one corresponds to the benzoxazine monomer ring opening and appears as an early peak and the second reaction corresponds to the second peak, occurring at higher temperatures, and is related to the reaction between the Ep monomers and the resulting Bz phenolic groups [10,37]. The unique peak observed in this study may be due to the high amount of Bz monomers that helped produce phenolic moieties, resulting from the Bz ring opening, which caused the reaction between epoxy monomer oxirane rings and phenolic groups. This means that the two reactions happened simultaneously and resulted in a single DSC peak [38]. In addition, the presence of epoxy has shifted the exothermic peak towards higher temperatures, which is also explained by the overlapping of the two reactions. This delay was already stipulated in many previous research works [39,40,41,42].

The exothermic information about curing is extracted and summarized in

Table 2. The DSC parameters of the Bz-Ep copolymer and its Bz-Ep-MCC-P composites.

Heating rate	Composites	Tonset (°C)	Tpic (°C)	$∆\mathit{H}$ (J/g)
β = 15 °C/min	Bz-Ep	209.3	241.6	−337.6
	Bz-Ep-MCC-P-1%	188.3	234.8	−316.5
	Bz-Ep-MCC-P-5%	177.9	221.2	−312.4

. The heat of polymerization is estimated as the area under the recorded peak; approximately similar enthalpies are noted for all systems of about 337 J/g, 316 J/g, and 312 J/g for Ep-Bz, Ep-Bz-MCC-P 1%, and Ep-Bz-MCC-P 5%, respectively. It is clear that the peak temperatures shifted towards lower values when MCC-P was added. This means that adding MCC-P reduces copolymerization temperatures and acts as a catalyst. According to previous studies [43,44,45], the phosphorous element may enter in a Bz ring-opening reaction by receiving the oxazine ring oxygen atom doublets, thus weakening the -CH₂-O bond of benzoxazine and resulting in phenolic groups that further boost the Bz ring opening. This later reacts with epoxy monomers, which results in low curing temperatures.

3.3. Thermal Decomposition

To investigate the effect of MCC-P on the thermal stability of the Ep-Bz copolymer, thermogravimetric analysis was performed up to 800 °C. The reported thermal behavior is depicted in Figure 3 and the reported parameters are shown in

Table 3. The TGA and DTG results of pure epoxy resin and epoxy/benzoxazine-based composites at various loadings.

Sample	Degradation Temperature (°C)			Pics of Degradation DTG	Residual Weight TGA % at 800 °C	LOI
	Tonset	T10%	T50%	T max
Bz-Ep	300	360	434	399	24.1	27.14
Bz-Ep-MCC-P 1%	280	353	449	387	30.8	29.82
Bz-Ep-MCC-P-5%	278	349	443	397	26.5	28.12

. The TGA results exhibited thermal stability over 250 °C for the copolymer and its MCC-P-based composites. The thermal decomposition can be divided into stages where overlapped steps can be noted at a temperature range from 250 °C to 350 °C. An earlier decomposition is witnessed in the DTG of the composites, which is attributed to MCC-P decomposition, as explained in the MCC characterization section. The significant decomposition observed in the temperature range from 350 to 550 °C is related to both resins.

From the TGA curves, the composites show enhanced thermal behavior within the temperature range from 300 °C to 800 °C compared to the neat Bz-Ep baseline. This enhancement is attributed to the residual char yield of composites that is higher than that of a neat copolymer. This increase is explained by the fact that the formed char deposit behaves as an insulator to shield the matrix from decomposition [46]. It is also noteworthy to mention that at 5% loading of MCC-P, the composite showed a slightly lower thermal stability, which might be related to the high amount of MCC that decomposes earlier, thus reducing the char yield.

3.4. Flame Retardancy Properties

The limiting oxygen index (LOI) can be estimated from the TGA residual char following the Van Krevelen and Hfytzer equation [47], using the char yield (CY) at 800 °C under a nitrogen atmosphere:

$$LOI=17.5+0.4\left(CY\right)$$

(2)

The results of LOI determination are presented in Table 3. If a self-extinguishing material has an LOI value greater than 21, it can then be qualified for flame retardancy applications [48]. It can be noted that if the studied samples exhibited LOI values above 21, adding MCC-P increased the LOI to higher values, indicating the efficiency of the phosphorylation in enhancing the FR of the Bz-Ep copolymer. The obtained results are similar to many other works about composites based on cellulose and polylactic acid incorporated with resorcinol bis (diphenyl phosphate) [49] and polybenzoxazine resins with cellulose phosphide [43].

The vertical burning test UL 94 ASTM D 3801 is a widely used method for assessing the flame retardancy of plastics and other polymeric materials. It involves subjecting a sample of the composite to a vertical flame for a specified period of time and observing its behavior, such as the rate of burning, dripping, or afterglow. The vertical burning test UL 94 results are illustrated in the Supplementary Materials Figures S1–S3, while the obtained outcomes are depicted in

Table 4. Results of the vertical burning test UL 94.

Composites		Time (s)	Drops	Classification
Bz-Ep	1st test	55	No drops	FAILED
	2nd test	0	No drops
Bz-Ep-MCC-P 1%	1st test	28	No drops	V-2
	2nd test	0	No drops
Bz-Ep-MCC-P 5%	1st test	24	No drops	V-2
	2nd test	0	No drops

.

Based on the information provided, it appears that the baseline Bz-Ep composite extinguished during the test after 55 s with no particle drops or a passage of fire to the cotton lid, revealing that this composite satisfies only one condition of the two test conditions for the classification, which means it failed the test. However, the other two composites successfully passed the test, as they extinguished after 28 and 24 s, with no drops or transfer of the flame to the cotton. Subsequently, both composites can be classified according to the UL 94 in the V-2 class. A carbon layer covering the samples was also observed, which acted as a protection for the sample. It is important to mention that bisphenol A-based benzoxazine shows low flame retardancy behavior compared to other benzoxazine resins and burns when analyzed by UL 94 [43,50]; its copolymerization with epoxy enhanced this behavior, and adding MCC-P further boosted its FR properties. A similar rating and burning behavior was obtained for the nanofibrillated cellulose/poly lactic acid/ammonium polyphosphate composite [51].

In order to deepen the flame retardancy investigation of the studied composites, bomb calorimetry ISO 1716, commonly used as a method for measuring the heat of the combustion of a material, is employed [24,25]. This method involves the combustion of a small sample of the composite in a bomb calorimeter and measuring the heat released during the combustion process. The heat of combustion can be used to calculate various fire-related parameters, such as the heat release rate and the total amount of heat released. The calorimetry ISO 1716 test measures the heat of the combustion of a material, expressed in units of megajoules per kilogram (MJ/kg). The test does not classify materials as being fire-resistant or not, but rather provides information about the amount of heat that is released when the material burns. A higher heat of combustion indicates a greater potential for the material to contribute to a fire.

Based on the results depicted in Table S1 of the Supplementary Materials, it can be noted that the Bz-EP composite material has the highest heat of combustion (30.822 MJ/kg). The Bz-EP-MCC-P 1% sample reveals a lower heat (28.552 MJ/kg), whereas the Bz-EP-MCC-P 5% sample shows an intermediate value. This indicates that the Bz-EP composite material has a higher potential for flammability and fire hazards compared to the MCC-P-loaded samples. Indeed, the composites reinforced with MCC-P passed the test successfully with a heat release under 30 MJ/kg, which is the criterium to be classified as medium in the calorimetry ISO 1716 test, hence proving that the phosphorylated cellulose functions very well as a flame retardant. The 5% MCC-P-loaded composite showed a slightly lower behavior compared to the 1% MCC-P composite, which is in accordance with the previous results of the FR test, confirming that 5 wt.% is considered a high loading that may alter the desired FR properties.

4. Conclusions

This study was performed with the main objective of developing a flame-retardant system with thermosetting resins and the phosphorus-functionalized cellulosic derivative for more reliable and eco-friendly material development. The phosphorylation was successfully carried out according to the characterization results through different methods, namely, TGA, FTIR, SEM, and XRD, where the major results confirmed the presence of phosphorus compounds on MCC and the increase in thermal residue.

The thermal behavior of the developed composites was investigated using TGA. Effectively, the results were very significant and the TGA analysis highlighted that adding the MCC-P-enhanced thermal stability and char residue, especially with 1% MCC-P, which had the best result. The LOI was also determined with TGA and revealed an enhancement in its value when MCC-P was incorporated into the copolymer. This study elaborated on the evaluation of the flame retardancy characteristics of the composites with the vertical burning test UL 94 and the bomb calorimetry, which revealed that the Bz-EP-1%MCC composite successfully passed both tests and is promising for future use as a flame retardant.