3.1. Characterization of the cysteine-based FR
The chemical structure of the cysteine-based FR was analyzed by FTIR, 1H NMR, and 31P NMR. In the FTIR spectrum (Fig. 3a), the stretching and deformation vibrations of NH4+(Liu et al. 2018a) are at 3158 cm-1 and 1402 cm-1, respectively. There are signals assignable to P-OH (2360 cm-1), P=O (1208 cm-1), P-O-C (1072 cm-1, 948 cm-1)(Su et al. 2021), C-P (1457 cm-1), C=O bond of -COOH (1717 cm-1)(Chen et al. 2021c) and C=O bond of -COO- (1637 cm-1), indicating the presence of functional groups such as -C=O and O-NH4+ in the FR. The NMR signal in Fig. 3c is 4.75 ppm of D2O; 3.25 ppm of H3, 2.79 ppm of H4, H5, 2.41 ppm of H2, and 2.23 ppm of H1; and the peak area ratio is 2:4:1:1. The phosphorus atom (P) signal of the -P=O(O-NH4+)2 group in Fig. 3d is 1.21 ppm. The overall information is consistent with the expected chemical structure, providing evidence for the successful synthesis of the cysteine-based FR.
3.2. Fourier transform infrared spectroscopy (FTIR) analysis of flame retardant wood powder
The FTIR spectra of pristine wood powder and that treated with 20% cysteine-based FR are shown in Fig. 3b. The absorption peaks of wood powder fibers are at 3395 cm-1 (O-H), 2910 cm-1 (C-H), and 1060 cm-1 (C-O-C). The treated wood powder displays additional peaks at 1243 cm-1 and 1552 cm-1 of P=O and C-N stretching vibrations, and at 1402 cm-1 assignable to C-P absorption. There are also new peaks at 1696 cm-1 and 1160 cm-1 of C=O and C(=O)-O stretching vibrations(Xu et al. 2019c). The NH4+, P-O and C-S stretching vibration absorption peak of cysteine-based FR is at 3200, 886 and 769 cm-1, respectively(Chen et al. 2021c). The formation of P(=O)-O-C and C(=O)-O-C covalent connections between FR and the fiber functional groups is accredited for the generation of these new absorption peaks. These findings show that the FR was successfully grafted onto the wood powder fibers through P(=O)-O-C and C(=O)-O-C covalent bonds.
3.3. X-ray photoelectron spectroscopy (XPS) analysis
The results of XPS analysis are shown in Fig. 4 and Table 1. Compared with pristine wood powder, the FR-treated wood powder sample displays signals of P, N, and S in addition to C and O signals(Su et al. 2021). The results can be attributed to the presence of P, N, and S in the cysteine-based FR. The findings support the structural integrity of FR during the grafting procedure.
Table 1
Elemental composition of wood powder and treated wood powder based on XPS analysis displayed in.
Element
|
Atomic (%)
|
The wood powder
|
The treated wood powder
|
C
|
70.4
|
48.6
|
O
|
29.6
|
34.2
|
N
|
-
|
10.6
|
P
|
-
|
6.40
|
S
|
-
|
0.27
|
3.4. Thermogravimetry analysis
The impact of cysteine-based FR on the thermal cracking of wood powder was investigated by TG analysis. Fig. 5 shows the TG curves of the two samples. For the pristine one, there was a sharp decrease of mass from 250 to 380 °C, and from 380 to 680 °C the decrease was relatively gentle. In the 250–380 °C range, there was major decomposition of wood powder, and the cracking produced a massive amount of carbonyl and aromatic compounds and tiny solid substances. Above 380 ℃, the residue of wood powder continued to dehydrate and decarboxylate, releasing H2O and CO2. In the stage of 680–720 ℃, there was largely the decomposition of lignin and residual carbon species that produced CO2, and by 800 ℃ there was 32.8% of solid residue left. As for the wood powder sample treated with 20% FR, the mass decreased slowly below 160 ℃ but sharply in the 160–310 ℃ range, which was the main decomposition stage of the FR-treated wood powder. Compared with the TG curve of pristine wood powder, it is deduced that the decomposition of cysteine-based FR provides phosphite and/or orthophosphoric acid, which promote the dehydration of wood powder fibers as well as the formation of carbon. Finally, by 800 °C, the amount of residue was 51.0%, higher than that of pristine wood powder, indicating that the cysteine-based FR could promote the charring of wood powder to isolate oxygen and producing a large quantity of NH3 to reduce the decomposition of wood powder, hence hindering the generation of flammable gases. In other words, the FR provides the wood powder with good thermal stability and excellent fire-resistance properties.
3.5. Thermogravimetric test coupled with FTIR (TG-FTIR)
The gas composition during the combustion of poplar wood powder and that of 20% FR-treated wood powder was studied using the TG-FTIR technique (Fig. 6). For the untreated wood powder, there are two fast breakdown stages starting at ca. 260 °C and 680 °C (Fig. 6a). The gases produced were identified at these temperatures. The signals in the range of 3500–4000 cm-1 are assignable to the stretching vibration of -OH, which are brought about by water. The absorption peaks of CO2 and CO stretching vibrations are located near 2380 cm-1 and 2181 cm-1, respectively(Wang et al. 2020). The C=O absorption of carbonyl compounds is responsible for the prominent peak at 1776 cm-1(Liu et al. 2020b). The peak close to 1265 cm-1 of medium intensity is attributable to the C-O absorption of phenolic compounds, whereas the peak near 1620 cm-1 to C=C absorption of aromatic compounds. Above 680 °C, there is significant increase of CO2 absorption intensity, plausibly caused by the decomposition of residual char. Simultaneously, there is the detection of a strong peak near 668 cm-1, which is due to C-H absorption of aromatic compounds generated in lignin degradation.
The release of carbon monoxide could be detrimental. A comparison between Fig. 6a and Fig. 6b reveals that there is lesser release of CO in the latter throughout the pyrolysis of FR-treated wood powder. At the same time, water, CO2, and carbonyl compounds were detected at relatively low temperatures. It is possible that the early decomposition of the cysteine-based FR produces acid that promotes the dehydration and decarboxylation of wood powder. Note that there is significant diminution of aliphatic C-H absorption near 2941 cm-1 and C-O-C absorption at ca. 1106 cm-1(Liu et al. 2018b). Also, there is a considerable decrease of aromatic C-H absorption at 668 cm-1, suggesting that the presence of FR prevents lignin from breaking down. In addition to the absorption peaks of NH3 (at 966 cm-1 and 930 cm-1), there are new absorption peaks ascribable to the stretching vibration of O=S=O group (at 1278 cm-1 and 1380 cm-1)(Li et al. 2020). The findings show that the generation of combustible vapors during the combustion of FR-treated wood powder was minimized.
3.6. Py-GC-MS analysis
To better understand the thermal degradation mechanism, we used the Py-GC-MS technique to identify the volatiles throughout the thermal decomposition of poplar wood powder with or without FR treatment.
The composition of wood powder is rather complex. In pyrolysis beside CO2 formation, there is the generation of others such as aliphatic compounds of simple structures, heterocyclic compounds, and aromatic compounds containing methoxy, alkyl, and hydroxyl groups. Furthermore, there are aliphatic oxides (including non-cyclic aliphatic acids, esters, ethers, ketones, and aldehydes), as well as oxygenated heterocyclic compounds (e.g., furan-like cyclopentanone and pyrone), most of which are flammable.
Shown in Fig. 7 is the total ion flow of cleavage products generated in the burning of pristine and FR-treated wood powder, where the peak intensity of the latter is significantly weaker than that of the former. The cleavage products of the two specimens are grouped into eight categories, and the results of product distribution are shown in Fig. 8. The main products from pristine wood powder are mainly those originated from cellulose, hemicellulose, and lignin, such as L-glucose generated in the depolymerization of cellulose and hemicellulose, and phenols from lignin. In contrast, there are more nitrogenous compounds and cycloalkanes in the case of FR-treated wood powder. A comparison between the Py-GC-MS data of the two specimens, the total nitrogen-containing compounds (27.41%) of FR-treated wood powder is much higher than that (10.63%) of pristine wood powder. Moreover, the FR-treated wood powder has cycloalkanes of 30.12%, which is 8.44 times higher than that of pristine wood powder. Note that the phenolic, aromatic, and fatty oxide contents in the FR-treated case are lower than that in the non-treated case. Overall, the results indicate that the addition of cysteine-based FR has caused a change in the decomposition pattern of wood powder, and the degradation of cellulose and hemicellulose to oxygenated heterocyclic compounds and sugar compounds followed by further degradation to aliphatic oxides was reduced. In the meantime, there is restriction of lignin degradation, and the char formation pattern suggests the formation of char with aromatic structure. The increase of nitrogenous compounds indicates that the FR degradation products promote the production of nitrogenous compounds from the cracking products of wood powder. It can be seen from the above that FR has a task to play in the gas as well as solid phase throughout the combustion of wood powder.
3.7. CONE analysis
Cone calorimetry (CONE) analysis was carried out on the wood powder density board and FR-treated wood powder density board to assess the heat release rate (HRR), total heat release (THR), and peak heat release rate (PHRR) as shown in Fig. 9 and Table 2. The tests were aimed to elucidate the combustion of wood powder density board treated with cysteine-based FR. The information in Table 2 reveals that the addition of cysteine-based FR reduced the highest peak heat release rate (PHRR) of FR-treated wood powder density board to 40.6 kW/m2, which is only ca. 14.6% of that of wood powder density board. Meanwhile, the fire growth index (FGI), which measures the fire spread ability (i.e., the heat released represented by the ratio of PHRR to the peak time of occurrence (time-to-PHRR)) is also considered. The FGI of FR-treated sample is 0.14 kW/s*m2, which is much lower than that of pristine wood powder density board (1.99 kW/s*m2), indicating that under the same thermal environment, the former has a better ability to resist the spread of fire. A similar deduction can be made based on the data of total heat release (THR) and mean effective heat of combustion (MEHC); that of FR-treated wood powder density board is only about one-third that of wood powder density board. However, there was a substantial increase in total smoke production (TSP) in the case of FR utilization, plausibly due to the change of pyrolysis pathway brought about by FR modification.
At high temperatures, FR releases phosphite and/or orthophosphoric acid, which promotes the charring of wood fibers, effectively reducing the total calorific value of combustion. With the minimization of peak heat release rate, there is effective inhabitation of flame spread. Referring to the heat release rate (HRR) curves (Fig. 9a), that of wood powder density board is bimodal with sharp heat release peaks, whereas that of FR-treated wood powder density board is small and smooth. The significant reduction of the latter suggests shortening of combustion time. Fig. 9b shows the THR curves, and that of wood powder density board represents a continuous and rapid exothermic process, and the THR is much larger than that of FR-treated wood powder density board. After ignition, the THR curve of FR-treated wood powder density board rose slowly, and then noticeably after 250 s, indicating that the cysteine-based FR can effectively retard the burning of wood density board. Overall, the HRR and THR results demonstrate that the FR could significantly affect the burning behavior of the material.
Table 2
Combustion parameters obtained from the CONE calorimeter of samples.
Samples
|
PHRR
(kW/m2)
|
HRR
(kW/m2)
|
THR
(MJ/m2)
|
MEHC
(MJ/kg)
|
TSP
(m2/m2)
|
FGI
(kW/s*m2)
|
Wood powder
|
279
|
110
|
33.7
|
14.1
|
0.92
|
1.99
|
20% treated Wood powder
|
40.6
|
24.6
|
11.7
|
5.14
|
1.79
|
0.14
|
3.8. Limiting oxygen index (LOI) test
The limiting oxygen index (LOI) is a valid measure for the assessment of flame retardancy, and specimens of poplar powder density boards with or without FR treatment were tested (Table 3). The LOI of untreated poplar powder density board was 26.5%, and with the rise of FR concentration, there is increase of LOI. When the concentration of cysteine-based FR was 20%, the LOI value was 60.9%. Note that all the density boards treated with FR have a LOI value higher than that of minimum requirement (i.e., 30%) for flame resistance. The results demonstrate the high fire resistance quality of the FR-treated density boards. This is explained by the fact that the cysteine-based FR contains S in addition to P and N elements, and there is a synergistic flame retardant action among the P, N, and S components. It can be concluded that the cysteine-based FR is an effective material for fire resistance.
Table 3
Limiting oxygen test data of density boards treated with different concentrations of cysteine-based FR.
Serial number
|
Concentration of flame retardant (%)
|
LOI (%)
|
1
|
0
|
26.5
|
2
|
5
|
32.2
|
3
|
10
|
43.6
|
4
|
20
|
60.9
|
3.9. Surface morphology
Scanning electron microscopy was conducted on poplar wood powder, FR-treated wood powder, burnt wood powder, and FR-treated wood powder after combustion. The effect of FR on their structures was analyzed with a magnification of 2000 times (Fig. 10). The surface of the untreated wood powder was relatively smooth, and the wood powder particles were dispersed. As for the FR-treated wood powder, the surface was relatively rough, and the fiber surface was wrapped by a layer of FR. The wood powder particles were adhered to one another by the bridging effect of FR, indicating that there is the attachment of FR molecules on the wood powder. The wood powder without FR treatment burned rapidly and becomes fluffy ashes upon combustion (Fig. 10c). Depicted in Fig. 10d is the SEM image of the FR-treated wood powder after combustion. One can see skeleton structure of charcoal, indicating that the cysteine-based FR can effectively promote char formation on the wood powder. In addition, "flower pieces" with pore structures were formed on the surface of charcoal. This may be because during the burning, the FR releases CO2, NH3, and other gases, leading to pore generation on the surface of charred residue.
3.10. Raman spectrometer analysis
Raman spectroscopy was used to assess the residual char of FR treated wood powder after combustion to assess the influence of cysteine-based FR on char production on wood powder upon combustion (Fig. 11). The Raman spectrum exhibits two overlapping peaks, the D band at 1356 cm-1 and the G band at 1581 cm-1, which correspond to amorphous char and graphitic carbon, respectively(Chen and Wang 2009). Amorphous carbon is thermally less stable than graphitic carbon, and graphitic carbon is produced more favorably to suppress thermal diffusion during combustion. The intensity ratio of G-band to D-band (IG/ID), where IG and ID are the integrated intensities of the G and D bands, respectively, can be used to measure the degree of graphitization of charcoal. A higher IG/ID value is associated with higher degree of graphitization in the char residue(Su et al. 2021). The IG/ID of FR-treated wood powder is 0.33. There is almost no residual char left after the burning of pristine wood powder. It is envisaged that during burning, the cysteine-based FR promotes the dehydration of wood powder for the formation of charcoal, and the charcoal prevents the exchange of heat and generation of flammable gases, leading to the expected flame retardancy.
3.11. Determination of formaldehyde concentration
Figure 12 depicts the formaldehyde standard curve that was produced in accordance with GB/T 17657-1999 standard. The absorbance of the free formaldehyde absorber of the density board treated with 20% cysteine based FR was measured by spectrophotometer was 0.062 nm, and the free formaldehyde concentration of the treated density board was calculated to be 1.44 mg/L (< 1.5 mg/L) according to the standard curve. As a consequence, the density board treated with 20% cysteine based FR passed the free formaldehyde test and satisfied the standards for E1 grade products. Although formaldehyde is a component of the raw material, it is converted to the -CH2- group during the synthesis process(Chen et al. 2021c), so the treatment of the density board will not release formaldehyde.
3.12. Flame Retardant mechanism
Fig. 13 illustrates the flame-retardant mechanism. In the burning of the wood powder, there is the cleaving of lignin, cellulose, and hemicellulose molecular chains, and the formation of oligomers (e.g., cellobiose, L-glucose, and propyl coumarin). The oligomers are broken down into small molecular hydrocarbons and ultimately burnt with the release of H2O, CO, and CO2 gases. In the case of a wood powder treated with a cysteine-based FR, the combustion follows the condensed-phase and gas-phase flame retardant mechanisms. During combustion, inert gases such as H2O, CO2 and SO2 are produced, and the oxygen concentration is diluted(Schartel et al. 2017). Additionally, non-flammable gases containing N are formed because of NH4+ release, consequently lowering the concentration of combustible gases. Upon burning, the phosphate structure generates a variety of P-containing reactive radicals (e.g, PO· and HPO·), which can quench the H· and O· radicals(Qu et al. 2020). In the same time, the generated metaphosphate/polyphosphate and sulfur-phosphorus compounds interact with cellulose, lignin, and hemicellulose to promote the formation of carbon residues, which can function as a physical barrier on the surface of wood to significantly lessen the exchange of heat and oxygen during combustion. The cysteine-based FR is therefore highly efficient for fire resistance.