36 
EXPERIMENTAL WORK IN FUEL TESTING. 
Determinations of ash and sulphur in coal sample* variously treated — Continued. 
SPECIAL TEST ON. REFUSE FROM A WASHING. 
[Original refuse— ash, 47.50; sulphur, 1.71.] 
Portion. 
Ash. 
Sulphur. 
Compared with 
original sample. 
Ash. 
Sulphur. 
85.2 
15.0 
17.7 
51.4 
12. 23 
28.13 
64.88 
0.88 
.76 
1.79 
1.83 
4.98 
33.35 
0.13 
1.35 to 1.65 . 
.13 
.92 
14.5 
3.0 
11.6 
1.43 
3.83 
.05 
33. 27 
1.48 
.17 
45.42 
1.40 
« Estimated from results on original refuse. 
VOLATILE MATTER IN COALS AND LIGNITES. a 
The official method of determining volatile matter, recommended 
by the committee on coal analysis appointed by the American Chem- 
ical Society, is as follows: 
Place 1 gram of fresh, undried, powdered coal in a platinum crucible weighing 20 
or 30 grams and having a tightly fitting cover. Heat over the full flame of a Bunsen 
burner for seven minutes. The crucible should be supported on a platinum triangle, 
with the bottom 6 to 8 centimeters above the top of the burner. The flame should 
be fully 20 centimeters high when burning free, and the determination should be 
made in a place free from draughts. The upper surface of the cover should burn 
clear, but the under surface should remain covered with carbon. To find "volatile 
combustible matter, ' ' substract the per cent of moisture from the loss found here. 
This method has been used regularly in the volatile determinations 
made in the laborator} T , the only modification being that the flame is 
protected from air currents by inclosing the apparatus in a cylindrical 
asbestos shield 15 centimeters long and 7 centimeters in diameter, the 
platinum triangle being located 3 centimeters below the top of the 
shield. The use of the shield gives more uniformity in the heat 
treatment, with a corresponding greater uniformity of results. 
In most coals the routine results obtained in the laboratory have 
checked to within less than 0.3 or 0.4 per cent; occasionally a sample 
has given trouble, and the variation between duplicates, without any 
apparent reason, was as great as 1 per cent. On some lignites it has 
been found impossible to obtain close duplicates, and on a few samples 
the official method gives very inaccurate determinations — as may be 
shown by the following results obtained in the laboratoiy upon two 
different samples, Nos. 2734 and 2764, of Texas No. 3 lignite, which 
differed only in the amount of moisture remaining in the air-dried 
sample, and perhaps in the fineness of grinding: 
a By permission of Dr. J. A. Holmes, in charge, and the director of the laboratory, the results of 
this investigation of the volatile matter was published by Prof. E. E. Somermeier as a paperin the 
Journal of the American Chemical Society, August, 1906. 
