Acid Water

Fig. 5.—Flow Sheet of Hardwood Distillation Process.

from the neutralized pyroligneous acid. This results in a considerable saving, since both the lime-lee and the alcohol still operations are replaced by a single distillation and a larger proportion of acetate liquor is left to be evaporated in the efficient multiple effect evaporators.

In most wood distillation plants the refining of the wood alcohol is carried no further than the preparation of this concentrated crude with no attempt to separate a commercial grade of wood alcohol. The crude is shipped to central refineries where with the advantage of larger scale operation the complicated refining process can be more readily carried out. Some of the larger plants or groups of plants, however, do their own refining. The apparatus and the methods used and the products obtained are so variable that no typical process can be described, but the general principle involved is the separation by fractional distillation of three fractions besides the water. These are in order of boiling points (i) the "methyl acetone" fraction with lowest boiling point, (2) the middle fraction of refined wood alcohol and (3) the highest boiling fraction containing the allyl alcohol. Small amounts of oils soluble in alcohol but not in water are present in the crude alcohol and frequently these are more readily removed from certain fractions by diluting greatly with water, separating the oils, and redistilling the water solution. The only chemical treatments used are simple washings with caustic soda and with sulphuric acid to remove certain impurities but not to help in the separation of the final products. The grade of the alcohol obtained will depend on the efficiency of the apparatus and on the amount and composition of the methyl acetone fraction. With the same conditions otherwise the more "methyl acetone" obtained the higher will be the grade of the alcohol, that is, the less acetone i't will contain. The production of a wood alcohol with as little as 0.1 per cent acetone is a difficult operation.

Yields of Products

The yields vary with the quality of the wood and with the care :aken in the process 50 that it is difficult to determine the average. There are also a few kiln plants still in operation, so- that the average Df all plants is.lower than for oven and retort plants only. Census Report for 1919 shows average yields of 142 pounds of acetate of ime, 7.7 gallons of crude wood alcohol and 39 bushels of charcoal per cord. It is believed, however, that there must be some mistake n these figures and that 180 pounds of acetate, 10 gallons of crude vood alcohol and 48 bushels of charcoal per cord of wood is a better iverage. The Michigan and Wisconsin plants will commonly get less icetate than this and the New York and Pennsylvania plants more iltliough the wood seems to be very much the same and the apparatus and methods identical. Some plants get over 200 pounds o[ acetate and 12 gallons of crude alcohol per cord.

New Processes

There are several new processes of wood distillation which have been given some publicity but which are still in the experimental or development stage1 so that they ^ can not be classed witli the commercial processes and details in regard to the process and yields can not be given. They are, however, promising enough from ;i commercial standpoint or interesting enough from a technical standpoint so that they should be mentioned and briefly discussed.

It happens that all these new processes are designed for the distillation of sawdust or small chips. The advantages of this sort of material are (1) that it may be obtained cheaper than cordwood, (2) that it can be more rapidly and efficiently dried, and (3) that it can be handled by a continuous process. There are, however, many disadvantages in using sawdust or chips. In the first place, wood itself is a good non-conductor of heat and sawdust is even better in this respect, since a mass of sawdust contains many closed airspaces which are still better non-conductors. A stationary mass of sawdust is, therefore, very difficult to distill, since it is almost impossible for the heat required to pass through to the center of the mass. It is necessary, therefore, that the sawdust be stirred either by a rotating retort or by some kind of agitation within the retort. This immediately introduces other complications. Either a stirrer or a rotating retort requires moving parts which must be joined gas-tight to the stationary parts of the retort, and at the temperatures of destructive distillation this is not a simple problem. The agitation of the mass also stirs up a great, deal of fine charcoal dust, which tends to be carried out of the retort with the vapors and to clog the condensers. If the process is to be continuous, it is also necessary to introduce the sawdust without introducing too much air and without allowing any of the products of the distillation to escape. The charcoal must also be removed through some kind of a gas-tight seal The cooling of large quantities of finely divided charcoal is also a difficult problem and this, together with the marketing of large quantities of such a product, again adds to the difficulties of sawdimt distillation.

The Briquetting Process

The American Wood Reduction Co. avoided many of the problems in connection with sawdust distillation by making the sawdust into

Ihîs wïwrUt«."6 proce88M Imv8 hecn 1,1,1 în i" t-umuu-rdal ai™,I plant» «1,,™

a solid block of wood before distillation. In other words, the sawdust or very small-sized wood was briquetted and distilled in the form of briquets. When sufficient pressure is applied to wood a very solid briquet can be formed without the use of any binding material. These briquets, however, when made of hardwood, tend to fall to pieces during destructive distillation; consequently the benefits of briquetting would be largely lost. It was found, however, that, if these sawdust briquets could be subjected to slight mechanical pressure during the distillation, they remained in one piece and fairly firm, regularly shaped pieces of charcoal could be made by this method.

The process finally developed by this company included the conversion of the waste wood in the form of sawdust or small chips into 4-in. briquets, which were loaded from an automatic loader into retorts, consisting of a series of horizontal tubes 20 ft. long and with an inside diameter slightly greater than that of the briquets. Each of these tubes was provided with a piston, by means of which a small pressure of 8 or 10 lbs. per sq. in. could be exerted continuously against the column of briquets during the distillation. On account of the small diameter of the tubes, each of which really acted as a single retort, the complete distillation could be finished in about three hours, and the temperature control was good enough so that unusually high yields were obtained. The cost of preparing the wood for distillation and the rather complicated and small-capacity apparatus was, therefore, compensated by the high yields and short length of time in the retort. A plant to use this process was under construction during 1918, but was never put into operation, and the process never had a trial on a commercial scale.

The Sawtelle Process

The Sawtelle process also avoided some of the usual complications of sawdust distillation by a combination of the distillation of the wood and the manufacture of producer gas from the charcoal. This was accomplished by using an updraft gas producer with wood as fuel. The wood in the form of sawdust and fine chips forms a cone-shaped pile in the gas producer, the upper part of which is wood in the drying stage, the next layer wood in the stage of destructive distillation, and the lower layers charcoal in the process of manufacture into producer gas. The heat required for distilling and drying is furnished by the hot gas. Since charcoal is not a product of this process, the difficulties with charcoal dust and with cooling the finely divided charcoal are not encountered. This process has operated successfully from the mechanical standpoint in a small commercial unit producer, but accurate yield figures could not be obtained on account of the lack of a gas scrubber. This process produces probably $

to 10 times as much gas as the ordinary process and the losses in the unscrubbed gas are, therefore, very high, especially the loss of alcohol. With the use of gas scrubbers there seems to be no reason, however, why this method should not give the maximum yields, since the process is a continuous one and. the temperature can readily be controlled. In localities where there is a demand for the producer gas obtained, this process should be very promising, but under present conditions the presence of large quantities of hardwood waste in localities where there is also a demand for power is uncommon.

The Stafford Process

This process was developed with the idea of utilizing fully the exothermic reaction of wood distillation. The process and apparatus is of the simplest, consisting merely of a large, well-insulated chamber into which the dry, warm wood is introduced continuously. The heat of decomposition of the wood is sufficient to bring an equal quantity of dry, warm wood to the distillation point; and the process is, therefore, continuous without the addition of heat from the outside. A thick layer of charcoal is kept in the bottom of the retort and partly cooled charcoal can be withdrawn continuously or at intervals. This process requires no moving parts inside the retort since the heat interchange between incoming wood and the hot vapors and hot charcoal in the retort is rapid enough without any stirring. A large plant using this process is now in commercial operation but no details of the results are available. There seems to be no reason why this process should not obtain maximum yields.

The Seaman Process

The Seaman process was developed and demonstrated in a commercial-sized unit

TABLE i.

Hardwoods Consumed in Distillation, by States: 1919.

TABLE i.

Hardwoods Consumed in Distillation, by States: 1919.

State

Number of establishments

quantity

cost

Cords

Per cent distribution

Total

Average per cord

United States

Michigan

Pennsylvania

New York

Wisconsin

All other states

0 0

Post a comment