In addition to the fundamental methods of distillation described in the preceding chapters, attention must be drawn to various methods not at present in general use, some of which possess features rendering them specially applicable in certain cases.

In the case of the Merrill process (Pat. U.S.A., 1918) oil circulating in pipes is employed as the heating agent. It is claimed by the inventor that the process is particularly valuable for replacing the usual steam heating. In the case of steam stills, for example, hot oil, previously heated in a well-designed tubular furnace, is circulated through the heating pipes of an ordinary steam still.

The advantages claimed are the following: (a) The heating medium is at ordinary atmospheric pressure, leakages thus being minimised.

(b) Ease of control.

(c) Ample "Thermic Head," i.e. the temperature of the heating medium is well above that of the oil to be heated.

(d) Low loss of heat, as the heating oil is returned to the furnace, whereas in the case of condensed steam much of the heat is necessarily lost even if the hot water is returned to the boiler plant.

The system is already largely used in the United States for melting asphalt, greases, waxes, etc., and in a few cases for distillation.

The method appears to offer many advantages over steam heating, and should doubtless find extensive application in the petroleum industry.

A difficulty which is frequently met with in the distillation of certain crude oils is that of the formation of coke on the still bottoms or on the internal walls of tubular retorts. This occurs particularly when the oil is heated to high temperatures, as is the case in cracking plants. Many devices have been proposed for overcoming this difficulty; for example, Wells and Wells (Am. Pat. 1296244 of 1919) patented a process in which the oil to be heated lies on a bath of molten metal, an alloy of suitable melting point being chosen. As no coke adheres to the surface of the molten metal, but merely floats on the top and can be removed, any burning of the still bottom is avoided.

A similar patent has been granted to Coast (U.S. Pat. 1345134 of 1920), who suggests covering the bottom of an ordinary still with a melted alloy. The still is provided with an arrangement by which the molten metal can be circulated through the oil to be heated.

A very obvious method of overcoming this difficulty is that of applying the well-known principle of heating by direct contact with gases, the transference of heat through still bottoms being thus avoided. This method is well known and largely used in other industries, e.g. for the concentration of sulphuric acid.

The idea of utilising gases as direct heating agents has indeed been adopted. Wells and Wells in the above-mentioned patent also claim the passing of a permanent gas into the molten metal bath, and thence into the oil to be distilled. Day and Day (Eng. Pat. 119440 of 1918) claim distillation by introducing into a body of oil gases composed of products of combustion mixed with cracked vapours, these vapours being introduced by a perforated pipe on the bottom of a still.

As far back as 1862, Trachsel and Clayton (Eng. Pat. 2966 of 1862) patented arrangements for distilling liquids by the passage through them of hot gases. In 1862 Broonan (Eng. Pat. 3037 of 1863) distilled liquid bituminous substances by direct application of heated gases, specifying products of combustion among others.

Wells (Fr. Pat. 379521 of 1907) adopted a similar idea in utilising the volatile products of petroleum in place of direct steam in the distillation of the heavier fractions. The same idea was utilised in the plant of Alexieef, which was operated for several years in Baku, without, however, any degree of success.

It seems strange that this method, which apparently offers certain decided advantages, has so far never been successfully developed in the petroleum industry.

Scores of patents have been granted for various types of distillation plant and fractionating apparatus. Only those methods which have so far found general application in the industry, or which seem to present prospects of future application, have been described. Intensive fractional distillation finds as yet little application in the petroleum industry, possibly a future demand for pure light products may some day necessitate it.