16 BULLETIN 312, U. S. DEPARTMENT OF AGRICULTURE. 
silicate which remains in the furnace may be converted into a soluble 
silicate by the addition of an anhydrous salt of potash or soda. 
Levi states that it is preferable to have a furnace of such a type 
that the reacting mass is not in contact with the carbon electrodes, 
so that the phosphoric acid formed will not be reduced to phosphorus. 
Washburn, on the other hand, states that he has carried on experi- 
ments like the above on a large scale and claims that unless a reducing 
agent, such as carbon, be added to the mixture of phosphate rock and 
silica, the volatilization of phosphoric acid is very incomplete. He 
heats a mixture of phosphate rock, silica, and carbon until the mass is 
entirely fused and the phosphorus driven off as such, and also in the 
form of oxide. He then exposes the gases to an oxidizing atmosphere 
and converts any phosphorus present to phosphoric anhydride and 
subsequently to phosphoric acid. He claims that under the proper 
conditions 90 per cent of the phosphoric acid present in the rock is 
volatilized. 
In a more recent patent, Haff describes a process very similar to the 
above, but states that pieces of broken carbon should be placed on the 
bath to form a conducting path of increased current density between 
the electrodes and thus allow of a higher temperature in the furnace- 
Another process of Haff’s, as well as a patent taken out jointly by - 
Wilson and Haff,? might be included in the subsequent table dealing 
with processes for the production of two or more fertilizer elements, 
but since they depend on the volatilization of phosphoric acid they 
are placed under this head. In these processes a mixture of phos- 
phate rock and feldspar is heated in an electric furnace to a tempera- 
ture at which both the phosphoric acid and the potash are driven off. 
The products are then condensed and collected in some suitable 
manner. 
It is generally conceded that cheap hydroelectric power is essential 
for the commercial success of any of these processes. At present the 
demand for hydroelectric energy is so great that in any of the locali- 
ties where it has been developed so far it brings a higher price than 
can possibly be paid for power in the economic production of phos- 
phatic fertilizers- The enormous energy that can be developed at 
certain sites, as, for instance, on the St. Lawrence and the Saguenay 
Rivers in Canada and on the Columbia River in Oregon, energy 
which, owing to the distance of the power sites from large industrial 
centers, can not be absorbed through the usual channels, may in 
the future cause a revolution in the fertilizer industry by making it 
feasible to produce concentrated phosphates from relatively. low- 
grade materials at considerably less cost than is possible by methods 
now in general use. 
1 United States Patent No. 1018186 (1912). 2 United States Patent No. 1103910 (1914). 
— 
