An increase in the percent by weight chemical usually was accompanied by a decrease in 

 the peak rate of weight loss. However, cellulose treated with 0.050 percent (Nilt,) 2HP0tt 

 caused an increase over untreated in the maximum peak rate of weight loss during 

 pyrolysis (page 18). The only other peak rates which exceeded untreated occurred at the 

 low temperature peak (250° C.) with higher levels of (NHL,)2S0Lt treated cellulose during 

 pyrolysis. The maximum rate of weight loss for the low temperature peak increased with 

 percent by weight chemical until a 10-percent treatment was reached. The rate decreased 

 at treatment levels above 10 percent (page 19). The DTG curves for each particular 

 chemical are quite similar for nitrogen and air except that rates of weight loss are 

 somewhat higher in air. 



Tables 4 through 7 also give data taken from TGA curves showing the effect of 

 (NHi+)2HP0i+ and (NH(^)2S0t+ on residue at 450° C. in both nitrogen and air. Cellulose 

 residue can be estimated if the amount of inorganic chemical remaining at 450° C. is 

 known. The TGA curves, shown on page 11, for the two chemicals indicate a 100-pcrcent 

 weight loss for (NHi+)2S0t+ and 37-percent weight loss for (NH^)2HP04. Assuming the 

 inorganic chemicals decompose the same whether cellulose is present or not, the normal- 

 ized cellulose residue can be determined: 



Normalized cellulose residue at 450° C. (percent) = 



Percent residue - (percent chemical residue x 



fraction chemical treatment) 



Fraction cellulose in sample 



The effect of (NHt4)2HP04 and (NHlj)2S0|4 on cellulose residue after pyrolysis and 

 combustion is shown in figures 2 and 3. As the chemical is increased (percent by 

 weight), both chemicals increase residue at 450° C. In air and at the lower concentra- 

 tions in nitrogen, (NH4)2HP04 causes a greater increase in residue than (NH4)2S0^. 

 The rate of volatilization of residue at 450° C. is less for (NHL,)2HP0t^ treatments and 

 the temperature required for its complete volatilization is higher (compare pages 16 

 and 17) . 



COMPARISON OF METHODS AND RESULTS 



Comparison of DTA, TGA, and DTG curves for cellulose pyrolysis and combustion and 

 cellulose treated with various concentrations of (NH4)2HP0i, and (NH4)2S04 shows reason- 

 able agreement for each method. Small endothermic and exothermic reactions made 

 determination of maximum rate temperatures impossible in TGA. The use of DTG provided 

 information not otherwise obtainable and its sensitivity greatly added to ease of 

 interpretation. 



When cellulose is treated with (NH4)2HP04 and this retardant is increased between 

 and 25 percent, the following temperatures are lowered: the threshold temperatures for 

 pyrolysis and combustion (table 3) ; the temperature of maximum rate of weight loss 

 (tables 4 and 6) ; the temperature of the major pyrolysis endotherm (table 1) ; the tem- 

 perature of the combustion exotherm (table 1) . This increase in retardant exponen- 

 tially increases the amount of cellulose residue (tables 4 and 6) . Similar trends are 

 observable in (NHi+)2S0i+ treatments; furthermore, the effects of this retardant are 

 usually more pronounced for lower treatment levels than for (NHi^)2HP0i4 treatments. For 

 example, a treatment of 0.0500 percent (NHi4)2S0t4 lowers the threshold temperature for 

 pyrolysis 33° C. while the same treatment of (NH4)2HP0i+ lowers it only 4° C. TJiough 

 similar comparisons can be made for other treatments of less than 1.00 percent, treat- 

 ments from 1.50 to 25 percent result in nearly the same threshold temperature for both 

 (NHi,)2HP0i, and (NH4)2S0t, (table 3). 



The differences in chemicals and similarities in analysis methods are graphically 

 depicted in figure 4. 



7 



