Sodium silicate modification can improve the overall performance of wood. The modification process has a great influence on the properties of modified wood. In this study, a new method was introduced to analyze the wood modification process, and the properties of modified wood were studied. Poplar wood was modified with sodium silicate by vacuum-pressure impregnation. After screening using single-factor experiments, an orthogonal experiment was carried out with solution concentration, impregnation time, impregnation pressure, and the cycle times as experimental factors. The modified poplar with the best properties was selected by fuzzy mathematics and characterized by SEM, FT-IR, XRD and TG. The results showed that some lignin and hemicellulose were removed from the wood due to the alkaline action of sodium silicate, and the orderly crystal area of poplar became disorderly, resulting in the reduction of crystallinity of the modified poplar wood. FT-IR analysis showed that sodium silicate was hydrolyzed to form polysilicic acid in wood, and structural analysis revealed the formation of Si-O-Si and Si-O-C, indicating that sodium silicate reacted with fibers on the wood cell wall. TG-DTG curves showed that the final residual mass of modified poplar wood increased from 25% to 67%, and the temperature of the maximum loss rate decreased from 343°C to 276°C. The heat release and smoke release of modified poplar wood decreased obviously. This kind of material with high strength and fire resistance can be used in the outdoor building and indoor furniture.
The afforestation area in China has exceeded 6.7 × 106 hm2, the largest in the world [
Wood modification can change the structure or composition of wood by physical, chemical or biological methods to enhance the overall properties of fast-growing wood for more efficient utilization. Silicate has a wide range of sources and is low-cost, non-toxic and harmless. Compared with other inorganic modifiers such as montmorillonite, water-soluble silicate solution is easily immersed into wood [
Different impregnation methods can alter the impregnation effect of sodium silicate [
Fast-growing poplar (
For the preparation of impregnated wood, 10 pieces of each experimental material were selected, put into the impregnation tank (impregnation tank: 50 L, Changsha Juchuang Technology Co., Ltd., Changsha, China), and the vacuum was adjusted to −0.1 MPa and held for 30 min. The pressure in the impregnation tank was used to suck in the sodium silicate solution until the wood was completely immersed. After 5 min, the pressure relief valve was opened to slowly restore the pressure to normal pressure. The sodium silicate solution in the tank was discharged, and the experiment material was removed. The surface of the wood samples was cleaned with ultrapure water. After air-drying the surface of the samples, the samples were then aged for 24 h, put into an oven for gradient drying, and then put into a drying dish to cool to room temperature. The process flow chart is shown in
The effects of sodium silicate modulus, sodium silicate concentration, impregnation pressure, impregnation time, and cycle times on the impregnation of poplar wood were investigated by single factor experiment. The factors and levels of the single-factor experiment were designed as follows: the sodium silicate modulus values were 1.9, 2.5 and 3.4; the sodium silicate concentrations were 10%, 20%, 30%, and 40%; the impregnation pressure values were 0.1, 0.3, 0.5, and 0.7 MPa; the impregnation times were 1, 2, 3, and 4 h; and the number of cycles was 2, 4, 6, and 8. On the basis of single-factor experiments, sodium silicate concentration (A), impregnation time (B), impregnation pressure (C), and cycle times (D) were screened for the orthogonal experiment. An experimental table of four factors and four levels of L16(45) was used to optimize the process of sodium silicate modification of poplar wood, as shown in
Level | A | B | C | D | E |
---|---|---|---|---|---|
Sodium silicate concentration/% | Impregnation time/h | Impregnation pressure/MPa | Cycle times/time | Empty column | |
1 | 10 | 1 | 0.1 | 2 | — |
2 | 20 | 2 | 0.3 | 3 | — |
3 | 30 | 3 | 0.5 | 4 | — |
4 | 40 | 4 | 0.7 | 5 | — |
The experiment materials before and after modification were put into a drying oven at 103°C ± 2°C, dried to absolute drying, and then the absolute dry mass of untreated poplar wood Wc and modified poplar wood WT were obtained. The calculation using
Wood hardness was measured according to national standard GB/T 1941-2009, “Wood hardness experimental method.” The pressing speed was 3~5 mm/min uniform speed, the pressing depth was 5.64 mm, and the load reading was accurate to 10 N.
Wood bending strength was measured according to national standard GB/T1936-2009, “Wood flexural strength experimental method.” Briefly, the specimen size was 300 × 20 × 20 mm3, the indenter and bearing radius of curvature were 30 mm, and the distance between supports was 240 mm.
Wood along grain compressive strength was measured according to national standard GB/T1935-2009, “Wood along grain compressive strength experiment method.” Briefly, the specimen size was 30 × 20 × 20 mm3, with the length along the grain direction, uniform speed loading was used, and sample failure time was 1.5~2.0 min.
Samples of poplar wood were put into a wood mill and pulverized, and particles less than 0.074 mm in diameter were collected. The wood particles were evaluated using a Bruker Vertex 70 FT-IR spectrophotometer (FT-IR, Bruker Corp., Billerica, MA, USA) using an infrared spectrum that ranged from 400 to 4000 cm−1.
The crystallinity index of the samples was measured by X-ray diffractometer at a voltage of 36 kV and a current of 20 mA. The angle 2
where
After gold spraying treatment, standard samples of poplar were fixed on the platform with conductive adhesive. The end face and longitudinal face of poplar were scanned and observed using a scanning electron microscope (FEI Quanta 200) at a voltage of 20 kV.
An appropriate amount of wood powder was put into a micro crucible and tested with a 209 F3 TGA (Netzsch Instruments North America, Burlington, MA, USA). The experimental conditions were air atmosphere and heating from room temperature to 750°C at a heating rate of 10 °C/min.
Alkaline substances in sodium silicate solution will decompose poplar cellulose, hemicellulose and other components, which will affect the bending strength of modified poplar wood [
Next, the effects of different cycle times were tested. Poplar wood was modified by the vacuum-pressure impregnation method. The WPG, bending strength, compressive strength, and hardness of the modified poplar wood was tested after different cycle times and the results are shown in
The results of the single-factor experiments identified the optimal parameters for best modification effects as follows: impregnation time of 3 h, impregnation pressure of 0.5 MPa, modulus of sodium silicate of 3.4, 40% concentration of sodium silicate, and four cycles.
The results of the single factor test show that when the modulus of sodium silicate was 3.4, the effect was the best. Since the modulus values of sodium silicate range between 1.5 and 3.5, the modulus of sodium silicate was not varied in the orthogonal experiment. The orthogonal experiment was carried out with factors of sodium silicate concentration (A), impregnation time (B), impregnation pressure (C), and the cycle times (D). An orthogonal experiment table with four factors and four levels L16(45) was constructed and used to optimize the sodium silicate modification of poplar with WPG, bending strength, compressive strength and hardness as evaluation indexes.
Number | WPG/% | Bending strength/MPa | Compressive strength/MPa | Hardness/kN | ||
---|---|---|---|---|---|---|
End face | Radial face | Bastard face | ||||
A1B1C1D1 | 10.10 | 61.90 | 58.80 | 4.28 | 2.44 | 2.54 |
A1B2C2D2 | 11.50 | 60.50 | 67.50 | 4.67 | 2.68 | 2.77 |
A1B3C3D3 | 12.90 | 63.00 | 73.20 | 4.73 | 2.72 | 2.84 |
A1B4C4D4 | 14.00 | 60.30 | 75.60 | 4.80 | 2.91 | 3.15 |
A2B1C2D3 | 24.20 | 71.30 | 79.00 | 5.05 | 2.97 | 3.05 |
A2B2C1D4 | 23.00 | 73.20 | 71.10 | 4.98 | 2.83 | 2.80 |
A2B3C4D1 | 30.00 | 63.50 | 83.50 | 5.62 | 2.89 | 3.28 |
A2B4C3D2 | 30.50 | 69.70 | 84.30 | 6.01 | 3.25 | 3.31 |
A3B1C3D4 | 43.00 | 65.90 | 80.40 | 7.30 | 3.20 | 3.63 |
A3B2C4D3 | 46.40 | 63.10 | 85.70 | 7.34 | 3.47 | 3.97 |
A3B3C1D2 | 39.40 | 62.70 | 79.60 | 6.86 | 3.12 | 3.57 |
A3B4C2D1 | 45.00 | 71.10 | 80.30 | 7.67 | 3.05 | 3.86 |
A4B1C4D2 | 59.20 | 79.20 | 87.00 | 8.93 | 3.39 | 3.77 |
A4B2C3D1 | 64.20 | 77.20 | 90.80 | 9.05 | 3.15 | 3.73 |
A4B3C2D4 | 62.30 | 67.70 | 78.90 | 8.93 | 3.35 | 3.84 |
A4B4C1D3 | 59.00 | 50.50 | 81.40 | 8.64 | 2.99 | 3.35 |
Unmodified poplar | — | 81.20 | 49.70 | 3.51 | 2.17 | 2.51 |
Average value | WPG/% | Bending strength/MPa | ||||||
---|---|---|---|---|---|---|---|---|
A | B | C | D | A | B | C | D | |
k1 | 12.125 | 34.125 | 32.875 | 37.325 | 61.425 | 69.575 | 62.075 | 68.425 |
k2 | 26.925 | 36.275 | 35.750 | 35.150 | 69.425 | 68.500 | 67.650 | 68.025 |
k3 | 43.450 | 36.150 | 37.650 | 35.625 | 65.700 | 64.225 | 68.950 | 61.975 |
k4 | 61.175 | 37.125 | 37.400 | 35.575 | 68.650 | 62.900 | 66.525 | 66.775 |
Excellent level | 4 | 4 | 3 | 1 | 2 | 1 | 3 | 1 |
Range | 49.050 | 3.000 | 4.775 | 2.175 | 8.000 | 6.675 | 6.875 | 6.450 |
Order | ACDB | ACBD | ||||||
Compressive strength/MPa | End face hardness/kN | |||||||
A | B | C | D | A | B | C | D | |
k1 | 68.775 | 76.300 | 72.725 | 78.350 | 4.621 | 6.389 | 6.191 | 6.654 |
k2 | 79.475 | 78.775 | 76.425 | 79.600 | 5.412 | 6.508 | 6.577 | 6.614 |
k3 | 81.500 | 78.800 | 82.175 | 79.825 | 7.290 | 6.534 | 6.772 | 6.439 |
k4 | 84.525 | 80.400 | 82.950 | 76.500 | 8.887 | 6.779 | 6.671 | 6.503 |
Excellent level | 4 | 4 | 4 | 3 | 4 | 4 | 3 | 1 |
Range | 15.750 | 4.100 | 10.225 | 3.325 | 4.266 | 0.390 | 0.581 | 0.215 |
Order | ACBD | ACBD | ||||||
Radial face hardness/kN | Bastard face hardness/kN | |||||||
A | B | C | D | A | B | C | D | |
k1 | 2.688 | 3.000 | 2.845 | 2.882 | 2.824 | 3.247 | 3.066 | 3.353 |
k2 | 2.985 | 3.033 | 3.012 | 3.110 | 3.111 | 3.319 | 3.379 | 3.354 |
k3 | 3.210 | 3.020 | 3.080 | 3.038 | 3.757 | 3.383 | 3.377 | 3.301 |
k4 | 3.220 | 3.050 | 3.165 | 3.073 | 3.672 | 3.416 | 3.542 | 3.355 |
Excellent level | 4 | 4 | 4 | 2 | 3 | 4 | 4 | 4 |
Range | 0.532 | 0.050 | 0.320 | 0.228 | 0.933 | 0.169 | 0.476 | 0.0540 |
Order | ACDB | ACBD |
It can be seen from
Orthogonal experiment can analyze the optimal process of multiple factors for one index, and fuzzy mathematics can be used to obtain the optimal process of multiple factors for multiple indexes. Firstly, each index value was fuzzified, and the precision of the index in the orthogonal experiment was mapped to a fuzzy subset and fuzzified. Then, the corresponding set of fuzzy membership degree was found by this transformation [
Let the set of evaluation indexes be U = {u1, u2,
Let the judgment set be V = {v1, v2,
where “rij” represents the membership degree of the ith index ui on the jth rating vi, i = 1, 2,
The results of the orthogonal experiment in
where t1, t2, t3, and t4 represent the average values of each level, and the evaluation membership degree ri can be calculated by the membership function uR:
All calculation results are listed in
Index | X1 | X2 | ||||||
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 1 | 2 | 3 | 4 | |
u1 | 0.084 | 0.187 | 0.302 | 0.426 | 0.238 | 0.252 | 0.252 | 0.258 |
u2 | 0.069 | 0.317 | 0.300 | 0.314 | 0.262 | 0.259 | 0.242 | 0.237 |
u3 | 0.069 | 0.301 | 0.309 | 0.320 | 0.243 | 0.251 | 0.251 | 0.256 |
u4 | 0.053 | 0.238 | 0.320 | 0.390 | 0.244 | 0.248 | 0.249 | 0.259 |
u5 | 0.222 | 0.247 | 0.265 | 0.266 | 0.248 | 0.251 | 0.250 | 0.252 |
u6 | 0.211 | 0.233 | 0.281 | 0.275 | 0.243 | 0.248 | 0.253 | 0.256 |
Index | X3 | X4 | ||||||
1 | 2 | 3 | 4 | 1 | 2 | 3 | 4 | |
u1 | 0.229 | 0.249 | 0.262 | 0.260 | 0.260 | 0.245 | 0.248 | 0.248 |
u2 | 0.234 | 0.255 | 0.260 | 0.251 | 0.258 | 0.257 | 0.234 | 0.252 |
u3 | 0.231 | 0.243 | 0.261 | 0.264 | 0.249 | 0.253 | 0.254 | 0.243 |
u4 | 0.236 | 0.251 | 0.258 | 0.255 | 0.254 | 0.252 | 0.246 | 0.248 |
u5 | 0.235 | 0.249 | 0.254 | 0.262 | 0.238 | 0.257 | 0.251 | 0.254 |
u6 | 0.229 | 0.253 | 0.253 | 0.265 | 0.251 | 0.251 | 0.247 | 0.251 |
Each evaluation object Xj in the table satisfies normalization for the evaluation subset of any evaluation index, that is:
Fuzzy comprehensive evaluation is carried out in the space composed of the evaluation object including the experimental factor set X, the evaluation index set U, and the evaluation matrix R, that is, the evaluation space S, S = (X, U, R). R is in essence, a fuzzy relation between set U and set V, so we can consider R as a fuzzy converter that can map the fuzzy subset A on U to B on V, where A and B are the preimage and image of the mapping, respectively.
The comprehensive evaluation of the multi-factor experiment was composed of the comprehensive evaluation of each single factor, and the fuzzy comprehensive evaluation B of each factor was a fuzzy subset of V, so the evaluation result needs to be obtained by the synthesis operation of fuzzy matrix as
The symbol “
When judging different factors, different converters R were used in the same judging space. Here, each element in comprehensive evaluation B was solved according to the fuzzy matrix multiplication rule, where bi was
Before the evaluation of impregnation process parameters, the weight vector A of index was established. The weight coefficient can be set according to different needs. The WPG directly reflects the filling effect. Modified poplar wood is usually used in furniture products, wooden handicrafts, and structural parts, so there may be higher requirements for load-bearing and elastic properties of the modified poplar wood. We determined the weight coefficients of the six indexes as a1 = 0.40, a2 = 0.26, a3 = 0.28, a4 = 0.20, a5 = 0.18, and a6 = 0.18. The weight vector A was used as the common criterion for each object, where A = (0.35, 0.29, 0.30, 0.20, 0.18, 0.18). The steps of comprehensive evaluation of each factor are shown using X1 (solution concentration) as an example:
The evaluation vectors of the four levels of solution concentration on the experiment indexes u1, u2, u3, u4, u5 and u6 were:
Thereby forming a fuzzy evaluation matrix R1 for X1:
According to fuzzy evaluation
There, b1 = (0.35 Λ 0.084) V (0.29 Λ 0.069) V (0.30 Λ 0.069) V (0.20 Λ 0.053) V (0.18 Λ 0.222) V (0.18 Λ 0.211)
= 0.084 V 0.069 V 0.069 V 0.053 V 0.18 V 0.18 = 0.18
The results of normalized processing gave B1 = (0.159, 0.265, 0.267, 0.309).
According to the above calculation method, B2 = (0.254, 0.251, 0.245, 0.250), B3 = (0.231, 0.251, 0.258, 0.260), B4 = (0.254, 0.251, 0.248, 0.247). Because b14 > b13 > b12 > b11, b21 > b22 > b24 > b23, b34 > b33 > b32 > b31, b41 > b42 > b43 > b44, the best process of impregnation modification was the combination of A4B1C4D1 under the condition of integrating the overall properties of poplar.
Wood is a combustible material, so there is a need to improve the fire resistance of modified poplar. The poplar modified by sodium silicate exhibited improved flame retardancy.
TG and DTG curves show that the thermal stability of the modified poplar was significantly improved. Compared with the unmodified poplar, the mass loss rate of the modified poplar was much slower. The final residual quality of the modified poplar wood increased from 25% for the unmodified poplar wood to 67%, an increase of 168%. The maximum loss rate temperature was also reduced from 343°C to 276°C. This shows that the immersion of sodium silicate destroyed the structure of wood, reduced the combustible matter, and shortened the time of wood burning. The sodium silicate treatment removed some hemicellulose and lignin and replaced them in the wood. The mass loss rate of modified poplar wood was reduced because the wood was fixed with non-combustible sodium silicate crystals. In addition, the chemical reaction between sodium silicate and wood changed the composition of wood, so the overall flammability was reduced. In
In this experiment, poplar wood was modified by vacuum-pressure impregnation with sodium silicate. After single factor screening, the optimum process parameters of a single index were determined by orthogonal experimental analysis. The optimum process parameters were determined under different factors of multiple indexes by the fuzzy mathematics method: 40% solution concentration, 1 h impregnation time, impregnation pressure of 0.7 MPa, and two cycles. Sodium silicate can evenly fill the vessels and cell lumen, remove part of hemicellulose and lignin, and make the orderly crystallization area became disorderly, resulting in the reduction of the crystallinity of modified poplar wood. Sodium silicate can hydrolyze in wood to form polysilicic acid, which combined with the fibers on the cell wall of poplar wood to form Si-O-Si and Si-O-C structures. The pyrolysis process before and after the modification of poplar wood was nearly the same, with faster pyrolysis of modified poplar wood. The final residual mass of the modified poplar wood was 168% higher than that of the unmodified poplar wood, the maximum loss rate was advanced from 343°C to 276°C, and the flame retardancy was significantly improved. This work optimized the process of poplar modified by sodium silicate, and the fuzzy mathematics method was used for the analysis of orthogonal results to more easily determine the optimal parameters, which provided a new idea for future research.
This work was financially supported by
The authors declare that they have no conflicts of interest to report regarding the present study.