COAL-TAR AND WATER-GAS TAR CREOSOTES. 17 
escape valve closed. Steam is then blown through the coke bed, 
water gas being produced according to the equation C + H 2 = CO + H 2 . 
As this gas enters the carburetor, gas oil obtained from petroleum is 
sprayed on the checker brick and cracked to permanent gases. This 
cracking and fixing of the gases is continued in the superheater. 
The gas, as it leaves the superheater, passes through condensers and 
scrubbers, where the tar is removed. The run with steam is con- 
tinued until the temperature drops below a fixed point. This opera- 
tion is then stopped, and the temperature is again raised by blowing. 
Table 3 shows a satisfactory set of operating conditions. 
Table 3. — Operating conditions in the water-gas set (4). 
Size of set 8 feet 6 inches. 
Area of grate 33.2 square feet. 
Fuel used Oven coke. 
Oil used Lima gas oil. 
Temperature at base of superheater ] ,461° F. 
Temperature at top of superheater 1,300° F. 
Length of blow 3 minutes. 
Length of gas make 4 minutes. 
No tar is produced in the water-gas reaction itself, but is all obtained 
as a result of the cracking of petroleum oil. The same discussion 
which applied to the reactions taking place in coal-tar production 
apply equally well in this connection if we consider that the primary 
reaction as given under coal tar has already been performed by 
nature in the production of petroleum oil. Here again, as in pro- 
ducing coal gas, a gas manufacturer would prefer to have no Tertiary 
reaction; but, in the attempt to obtain a secondary reaction as com- 
plete as possible, some Tertiary products are formed. The presence 
of paraffin hydrocarbons in water-gas tar shows that, from the gas 
maker's point of view, a waste has resulted from insufficient cracking 
and the presence of aromatic tars shows a decrease in illuminating 
value on account of a recombination of some of the illuminants. 
The gas maker desires to obtain as small an amount of highly aro- 
matic tar as possible with complete cracking of the oil. The presence 
of the large amount of hydrogen in water gas probably suppresses to 
)me extent the Tertiary reaction which would normally be obtained 
y cracking petroleum in the presence of its own decomposition 
ipors, as is done in the manufacture of Pintsch and Blau gas; but 
e effect of this suppression would be no greater than a similar effect 
lich must take place in coal gas, for the latter contains more 
_, drogen than does water gas. 
The yield of tar depends, therefore, upon the completeness of 
cracking of the oil, which, in turn, is dependent upon the temperature 
to which the oil vapors are brought and the character of the oil used. 
Table 4 gives a comparative yield of water-gas tar from different oils. 
75536°— 22 2 
