PAPERmaking! Vol7 Nr2 2021

Polymers 2021 , 13 , 709

9of 13

Table4. ANOVA table for Col_4 Time-to-ignition by Col_1 heat flux.

Source

Sum of Squares

Df

Mean Square

F-Ratio

P-Value

Between Groups Within Groups

96,009.1 87,551.5 183,561

7

13,715.6

17.55

0.000

112 119

781.71

Total (Corr.)

Weight loss caused by thermal degradation of the board surface, which was exposed to radiant heat (Table 2), maintained the same course. Trend lines for the individual board thicknesses have a parallel direction, so it can be assumed that the weight loss is uniform per each unit of thickness (Figure 7). Difference can be seen in the samples with a thickness of 12mm.

Figure7. Graphical dependence of the weight loss Δ M and sample thickness H ratio on the heat flux. Legend: black point, 12 mm thickness; blue point, 15 mm thickness; red point, 18 mm thickness. The results clearly show the following dependence: with increasing thickness, the weight loss decreases, but the ratio of weight loss and sample thickness Δ M/H is rather similar in all heat fluxes (Figure 7). A similar scenario is observed at a thickness of 15 mm. The highest weight loss is observed at a thickness of 12 mm, where an area of light weight of the material burns down and it is possible to observe a slight increase in weight loss with increasing heat flux. There are several studies that confirm the reduction of the weight loss and prolongation of time-to-ignition by appropriate fire-retardant treatment of the boards [3,33,34]. Statistical analysis confirmed this fact. The heat flux (Table 5) and material thickness (Table 6) have no statistically significant effect on the change in sample weight.

Table5. ANOVA table for Col_3 Weight loss by Col_1 heat flux.

Source

Sum of Squares

Df

Mean Square

F-Ratio

P-Value

Between Groups Within Groups

5.11703E7 4.00277E8 4.51448E8

7

7.31005E6 3.5739E6

2.05

0.0554

112 119

Total (Corr.)

Made with FlippingBook Online newsletter maker