The Domestic Tung-Oil Industry

R. S. McKinney, R. L. Holmes.

Tung oil is produced from the seed of the tung tree (Aleurites fordii and A. montana), a member of the Euphorbia family. The tung fruit is about 2 inches in diameter and normally has 4 or 5 seeds. The entire fruit is covered by a hull about a quarter of an inch thick. Each seed is coated by a hard shell; in it is the kernel, which constitutes about a third of the weight of the fruit and contains about 65 percent oil.

The Chinese have produced and used tung oil for centuries, but Europe and America did not know until the beginning of the twentieth century how to make use of its unique properties to produce water- and chemical-resistant coatings.

You would think that our young domestic tung-oil industry could find a solution to its problems in China, but explorers of the Department of Agriculture reported that the primitive methods used to produce much of the tung oil of China have changed but little since the sixteenth century, when Portuguese traders exchanged European goods for China wood oil (tung oil) at Canton. In China, the tung fruit, picked by hand, is piled into heaps to ferment. The seeds, husked by hand, after being roasted, are ground to a meal in stone mortars. The ground, steamed meal is shaped into cakes, which are wrapped in straw and placed in a hollowed-out log; the oil is pressed out by driving wedges between an end of the log and a wooden block placed against the cakes. Because of the crude methods, the oil has a dark color and a considerable amount of free acids, which may affect its use for some purposes. Obviously, such a process would not be economical in the United States.

The development of the domestic tung-oil industry resulted from the efforts of scientists in the Department of Agriculture and the agricultural experiment stations and individuals having commercial interests in tung production. Early studies showed that tung trees could be successfully grown within about 100 miles of the Gulf of Mexico. Needed are high rainfall, a mild climate, and soil that is a fertile friable loam and has good water and air drainage.

The first commercial planting of tung trees was made in 1925. The domestic commercial production of tung oil started in 1932, when two tank cars (120,000 pounds) of oil were shipped from the first tung mill near Gainesville, Fla. An estimated 192,000 acres of tung trees grow in the Coastal Plains of Florida, Louisiana, and Mississippi; Alabama, Georgia, and Texas have smaller acreages. About 25 million pounds of oil was produced in 1949 at 12 mills in several States.

So many plantings of tung trees had been made in the United States by 1938 and so much interest was shown in this new development that the Congress appropriated funds for investigations on the crop. Field laboratories were set up to carry out the investigations.

TUNG FRUIT is ordinarily processed in the United States by passing the partly dried fruit through decorticators and separators to remove the hulls and a portion of the shells from the seed and kernels. The oil is expressed from the separated, ground, and preheated kernels and seeds. That is done in a continuous press, which has an interrupted screw that moves the material forward within a cylinder of steel bars set close together against a restricted opening. The residue is discharged through the constricted opening at the far end of the cylinder in the form of a cake. The oil flows out between the bars of the cylinder. It is filtered and pumped into storage tanks.

DRYING TUNG FRUIT and seeds so that they can be milled or stored has been a major problem of the industry. The tung fruit that falls from the trees contains about 65 percent moisture and cannot be stored safely until it has dried to about 25 percent. The earliest practice in the domestic industry was to allow the fruit to lie on the ground until it had dried enough to be stored, without danger of heating or sprouting, in especially constructed barns. Under favorable weather conditions, several weeks are required for the newly fallen fruit in an orchard to dry to the proper moisture level; in wet seasons the tung fruit may not dry sufficiently on the ground to permit storage. For efficient oil extraction, the seed should contain 6 to 9 percent moisture. Fruit stored at 25 percent moisture requires several weeks or more to dry enough to permit pressing. A method of drying was needed so that the mills could start operating earlier and could accept wet fruit if they had to.

Tests were made on the artificial drying of the whole fruit, but that method does not seem economical in comparison with natural drying because of the excessive heat and time required. Because the hulls contain more than half of the moisture and have no value as a source of oil, their removal before drying or storage is desirable, so as to lower the heat needed in drying and to reduce the space required for storage.

The disc huller used in tung mills removes the hulls and removes or breaks a large proportion of the shells. It also damages some of the kernels. Experience with other oilseeds indicated that an increase in the free fatty acids through breakdown of the oil might be expected unless the broken seeds were expressed immediately. This was found to be the case. The moist broken seed developed free fatty acids rapidly and heated spontaneously unless forced ventilation was used for drying. However, tests in a pilot-plant tray drier at 165 F. showed that the hulled seed after drying to 10 percent moisture could be stored for several months with little or no increase in the content of free fatty acids of the oil and that the hulled dried seed could be processed efficiently soon after drying or after storage for several months.

Many oil-extraction tests carried out on both laboratory-scale and commercial screw presses indicated that the oil is expressed most efficiently when the ground meal entering the barrel of the press contains 3 to 5 percent moisture. The tempering bin attached to the press at the first tung-oil mill built in the United States was found to remove about 4 percent moisture from the meal, while the stack cookers placed over the press at later. mills could reduce the moisture in the meal as much as 8 percent. The meal cooker and drier attached to the newest type of press removed about 5 percent moisture from the meal. Because of the limited capacity of the meal driers, other methods of drying were sought.

Several tung-oil mills have built bin driers similar to those used for drying cottonseed, in which hot air is drawn through the seed for about 24 hours. The driers were found to be somewhat unsatisfactory, because drying is not uniform and the possibility exists of development of free fatty acids in the seeds and kernels during the prolonged drying period. Several producers have installed vertical seed driers, in which warm air is drawn through a layer of seed 12 or 24 inches deep. It has been found that better yields of oil are obtained by keeping the temperature of the drying air in the vertical driers below 160 F. and drying the seed to a content of about 9 percent of moisture. Very satisfactory results have been obtained with the vertical driers, and their use has increased the capacity of the mill and the efficiency of oil expression.

WHEN THE FIRST commercial planting of tung was made in the United States, tung oil had not been recovered in any country by modern equipment. Early tests of a number of mechanical devices for hulling and separating tung seeds showed that that could be accomplished by first passing the tung fruit through a disc huller to remove the hulls and part of the shell from the kernels and discharging the hulled fruit over shaker screens to separate the hulls and broken shells from kernels and seeds. Other experiments indicated that the continuous press was better than the hydraulic press for expressing the oil.

A study of the hulling of tung fruit showed high losses of oil from kernel fragments in the separated hulls when the disc huller was operated on very dry fruit, and low losses on fruit containing 15 to 20 percent moisture; fruit containing much more than 30 percent moisture could not be hulled. A portable drum-type huller developed in the Department of Agriculture permits the hulling of fruit that contains 30 to 45 percent moisture without breaking many shells.

If the intact seed is dried and stored, it is hardly feasible to remove more of the shell just before pressing, and it would be necessary to allow all the shell to remain in the meal. Most mills remove about half the shell, but opinions differ as to the amount of shell that should be left with the kernels to provide friction and drainage during pressing.

Tests were made on a laboratory press and a commercial screw press to determine the effect of varying the proportion of shell on the yield of oil. Meal containing all the shell ( about 40 percent of the weight of meal) processed as efficiently as that containing half the shell. Kernels completely freed of shells yielded a press cake very high in oil content. Apparently a certain amount of shell is necessary to develop sufficient friction for efficient operation.

It was also found to be impossible to obtain efficient recovery of oil with seeds carried over from one season to the next. The difficulty appeared to result principally from the fact that the meal from old kernels does not become plastic when subjected to heat and pressure. When the old seeds are mixed with new seeds, no difficulty is experienced in expressing the oil from the meal.

During the filtration of crude tung oil, filter cake is produced which may contain as much as 50 percent oil and may constitute 20 percent or more by weight of the expressed crude oil. To recover as much oil as possible, the material is either added to fresh tung meal entering the press or is allowed to accumulate and then mixed with press cake and hulls and pressed without addition of fresh meal. Neither process has been satisfactory, as the oil in the filter cake oxidizes readily at high temperatures and the cake ignites spontaneously if not handled carefully. Besides, this fine material tends to pass through the cylinder bars of the press and increases the amount of foots of the crude oil.

Tests made on a commercial press and a laboratory press showed that the oil content of the press cake increased when filter cake was added to the meal and re-pressed. It was also impossible to obtain a press cake of low oil content when the filter cake was mixed with press cake and hulls and processed separately. A better method of, handling filter cake is needed; solvent, extraction seems to be the answer.

During the Second World War, the supply of tung oil from China was cut off, and the stocks in the United States were allocated for military purposes. Imported oil stored for 4 years in-creased in acid value from 4.0 to 8.0 and its rate of polymerization increased as shown by a decrease in the time of the Worstall test from 7 to 5 1/2 minutes. In several instances, domestic oil stored in small tanks exposed to the weather formed a polymerized surface layer several inches thick. The layer was useless for the preparation of paints or varnishes. Such layers did not form in large tanks or in a tank under a roof. In a large tank, only a slight increase in the acid value of the oil occurred in about 2 years, the color of the stored domestic oil was good, and there was no evidence of gel formation. In tests at the Tung Oil Laboratory in Bogalusa, La., in 1946-49, domestic tung oil sealed in gallon cans was stored under different conditions and tested at intervals of 3 months. No measurable deterioration was detected, even after storage for 3 years, in cans painted black and standing in the sun.