EXPERIMENTAL PROCEDURE 



The two flame retardants selected for investigation in this study were ammonium 

 phosphate (NHi^) 2HP0i(. and ammonium sulfate (NH4)2S0i+. In determining the effect these 

 additives have on the pyrolysis and combustion of cellulose, differential thermal (DTA) , 

 thermogravimetric (TGA) , and derivative thermogravimetric (DTG) analyses were used. 



Samples 



The cellulose samples used in this experiment were prepared from high purity 

 Munktell's chromatographic cellulose powder (smaller than 170 mesh) containing less 

 than 0.02-percent ash. After the cellulose was dried in a' desiccator over silica gel, 

 the moisture content was determined using thermogravimetric analysis. The desired 

 percent by weight chemical was obtained by adding a stock solution of known concentra- 

 tion to a previously weighed sample of the cellulose. The weight of cellulose was 

 corrected for 2.6-percent moisture content and weight percents expressed on a dry 

 basis. Each sample was prepared by combination of the desired chemical and cellulose 

 (the total equaled 1 gram) and addition of 10 milliliters of distilled water. After 

 thorough mixing, the samples were dried to approximately 2-percent moisture content in 

 an oven at 32° C. Twenty samples between 0- and 25-percent treatment of both ammonium 

 phosphate and ammonium sulfate were prepared. Both chemicals were analytical reagent 

 grade. An untreated sample was prepared by adding 10 milliliters of distilled water to 

 1 gram of cellulose and drying as for the treated samples. 



Differential Thermal Analyses 



Differential thermal analysis (DTA) data were taken using a DuPont 900 Thermal 

 Analyzer.^ The standard temperature cell (ambient to 500° C.) was used for all treated 

 cellulose samples. The intermediate temperature cell (ambient to 850° C), similar to 

 the standard cell, was used only to obtain thermograms of the inorganic compounds. 



The sample of 10 (±1) milligrams was placed in macrocapillary tubes, 4 millimeters 

 in diameter. Glass beads were used as the inert reference material. Chromel -alumel 

 thermocouples were used to determine the differential temperature of the sample and 

 reference material. A heating rate of 25° C. per minute was used for all analyses. 



Pyrolysis of all samples was studied using nitrogen to provide the inert atmosphere. 

 After the cell had been purged by evacuation and addition of N2, a flow rate of 1 liter 

 N2 per minute was established. An atmosphere of moisture-free air was used for study- 

 ing combustion. In order to increase the contact between the sample and reactive gas, 

 a fluidizer requiring a lower flow rate (100 cubic centimeters per minute) was used for 

 tests in air. 



Thermogravimetric and Derivative Thermogravimetric Analyses 



Thermogravimetric analyses (TGA) were made using the DuPont 950 Thermogravimetric 

 Analyzer while recording weight and temperature signals on the DuPont 900. Ten-milli- 

 gram samples were placed in a tared aluminum pan with a chromel-alumel thermocouple 

 about 1 millimeter above the center of the sample to record sample temperature. The 

 sample weight was adjusted to the 100-percent chart line following initial water loss 

 near 150° C. so that all sample weights would be on a percent-weight-remaining and 

 moisture-free basis. 



■^The use of corporation names in this publication is for the information and 

 convenience of the reader. Such use does not constitute an official approval by the 

 U.S. Department of Agriculture of any product to the exclusion of others which may be 

 suitable. 



2 



