US1670053A - Dehydrating- alcohol - Google Patents

Description

May 15, 1928.

J. A. STEFFENS DEHYDRATING ALCOHOL Filed Aug. 16, 1923 5? Mcm? Patented. ay 15,. ieee.

anta

J'OHN A. STEFFENS, 0F BROOKLYN, NEW YORK.

nnHYnnATING ALCOHOL.

Application led August 16, 1923. Serial No. 657,814.

,This invention relates todehydrating alcohol; and it comprises u method of dehydrating alcohol (and like liquids) with the aid of water-immiscible volatile liquids,

such as benzol, wherein the alcohol to be.

dried is delivered to a column still containing benzol and dry alcohol is drawn off at the base of such still, and all vapors arising from the distilling operation are condensed to form a 2-layer condensate at the temperature of the head ofthe column, one such layer 'containing a greater amount of water than the other, and there is returned to the still head all ofthe layer relatively poor in water and about two-thirds of the layer rich in water, return being while at such temperature, the remaining 'third of the latter layer being withdrawn for appropriate treatment. elsewhere, conditions being so adjusted that the amount of water in the withdrawn liquid balances that entering 4the systcm with the alcohol to be dried; and it also comprises an assemblage of apparatus elements adapted for, use in the performance 4of the described method; such assemblage comprising acolu in still with appropriate heating means, means for introducing alcohol as vapor or liquid at a midpoint, means for withdrawing dry alcohol at the base, means for totally condensing vapors leaving the head of the still, means for gravit-ally separating the condensate into two liquid layers at the temperature ofthe still head, and means for withdrawing a portion of the heavier liquid while refluxing the residue of said liquid and all of the lighter liquid into said still head; all as more fully hereinafter set forth andas claimed.

As is well known, ethyl alcohol (grain alcohol) containing water cannot be brought to a strength beyond about -95 per cent by simple distillation in ordinary stills what.- ever the efficiency of the distilling means. It is easily possible to produce alcohol of about 95 per cent by many types of column still: but beyond this, the operator cannot go by operation. An alcohol of about 95 pery cent strengtlrhas a lower boiling point than either stronger or weaker alcohol; and spirit of this strength therefore goes to the head of the still and thence to the condenser.

There'is however a considerable commercial demand for absolute alcohol; or alcohol stronger than 95 per cent. In particusimple distillation in commercial lar it is desired for making composite motor I fuels; blended fuels of which alcohol is one component. and kerosene or some other petroleum oil is another'. Apart from the undcsirability of water in the fuel, 95 per cent `alcohol does not blend well with most petroleum oils and in general where it is used, it is necessary to have a third body present as a blending agent, benzol, ether, etc., in greater or less quantity. Where the alcohol is dry, this blending agent may be dispensed with, thereby cheapening and simplifying the manufacture of the fuel. With an alcohol which is more or less moist, the drier the alcohol, the less blending agent need'be used.

rlfhe commercial demand for absolute alcohol is in part met by modifications of the ordinary laboratory methods in which' alcohol is distilled to 95 per cent and then dried by hygroscopic chemicals; lime, calcium chlorid, potassium carbonate, sodium, calcium carbid, etc.' In part, it is met by modiications of the so-called Youngs method, wherein water is distilled out of strong alcohol 'by thev addition of a thi-rd volatile body, generally benzol. With a mixture containingalcohol, water and benzol, on distillation a mixture o'f vapors poor in alcohol is formed and goes to the head ofthe column, while anhydrous alcohol can De drawn off at the base. It is to this latter type vof process that the presentl invention pertains.

In this manner of drying alcohol, not all the alcohol supplied to the still is drawn olf at the base; some inevitably distils over with the benzol and water. The vapors coming from the head of thestill, on condensation and standing, separate intotwo superimposed liquid phases or strata: an upper or= overlying stratumr of benzol containing a.

little water and a little alcohol, and an 11nderlying layer of water and alcohol containing some benzol kept in solution by the alcohol. The upper layer, is, or may be, returned to the still to serve anew. It is customary to dilute the lower aqueous layer to separate the dissolved benzol and then rerectify the diluted liquid to obtain 95 per cent alcohol which can re-enter the operation. Obviously, the greater the proportion which the dry alcohol obtained at the base of the still bears to the-alcohol which goes over with the vapors of Water"- and ben.-

zol and which must be reconcentrated, the greater the economy of the operation. By the most approved methods now in use, it is dilicult however in practical operation to obtain more than to 80 per cent of the alcohol directly as dry alcohol; 30 to 20 per cent going over with the benzol and water and requiring redistillation and reconcentration.;

The yields stated are below those theoretically possible and there is room for much improvement in the practical operation of the process. It is the object of the present invention to provide such an improvement. The diiliculty lies in therelatively complex nature of the operation and the nice regulation of conditions necessary to ensure that neither benzol (which has about the same boiling point as alcohol) nor water passes out with `the dried alcohol at the base of the still. The difliculties of operation are accentuated in methods of operation where the layer rich in benzol is fed directly back to the still as it separates from the'cold condensate. Under these. conditions, with any slight change in the composition or proportion of the alcohol feed, either benzol or water tends to creep downward in the column and reach the absolute-alcohol discharge.

In the stated methods of the prior art, the column still and its condenser are used in the ordinary way; the vapors reaching the top of the column or the still head being withdrawn through a suitable vapor outlet and sent to4 a condenser. This 'condenser is intended to, and usually does, deliver a relatively cold condensate; a condensate at a temperature. considerably less than that of the still head. This condensate on standing in suitable settlingtanks or devices, separates into two liquid layers or phases which are individuallyV treated. The upper, or lighter layer, which contains most ofthe benzol, together with some water'and some alcohol is often directly returned to the still.

In the present invention, I employ a column still, but I use itin a different way and in lieu of forming and stratifying a condensate in a cold state as two distinct layers, I

condense the vapors coming from the still head hy a device in the' nature of a reflux condenseigthereby producing a 2-layer condensate and, while at stillfhcad temperature., refiux directly back to Vtle still the portion ofthe condensate containing but little Water,

' together with a major fraction of the portion containing more water. While both layers contain all three components, Water, alcohol and benzol, for the sake-of convenience, the layer containing the major portion of water is hereafter called the aqueous layer. and the other the non-aqueous layer. In effect, however, I perform a gravital separation of the two immiscible liquids orl liquid phases at still head temperature.-

There are several important'and advanta- 'light liquid is richer in benzol and alcohol and, therefore, relatively lighter. Since the object of the process is the elimination of water, this production of a heavy layer richerin Water is in itself desirable while there is the further advantage of the stated increased difference in specific gravity. This gives a quicker and easier separat-ion; and I find that in practice by operating on a condensate at still head temperature, it is not necessary to Wait forethe production of two well stratified and distinct layers; that I canoperate on the condensate as fast as it is delivered.

In any distillation of this type, in normal operation there is a certain necessary lixed relationship between the three component bodies, alcohol, benzol and Water, in the composition of the vapors in the still head; water vapor being as necessary as the other vapors. The condensate yielded, as a whole, has of course'vthe same iixed composition. And I find that in practice, I can-best preserve the normality of' Vapor conditions in the still head by returning to it all the lighter portion. of the condensate given by the reflux and about two-thirds of the heavier. portion. The 1: 2 ratio given is of course approximate; in practice I -aim to withdraw such a proportion -of the heavier liquid as will contain water in amount corresponding to .that enterin the system with the alcohol to be dried. n so doing there is a minimum of interference yyithi normal vapor conditions in the still Because of the quick and clean separation y which is practicable at still head temperature I find that I can deliver the Whole body of condensate into and somewhat -bel'ow the surface of' a body of liquid in or near the still head and overflowing into it. In so doing the light liquid, rich in benzol, rises through the body and overfiows with it,

,while thel heavier liquid separates at the bottom. Such portion of this latter as is to be withdrawn can Abe tapped off cleanly at or near the bottom.

In practical embodiments of my inven tion, I use a vcolumn still. delivering vapors to a reflux condenser having liquid, as well as ivapor, communication wlth the head of the still. In the liquid communication line, however, I interpose a liquid collecting device which may be simply the top plate in vapors leaving the head of the still and dclivers them to the collecting device. This condensate comprises bony-ol, alcohol and water and it tends tov separate into two immiscible portions of liquid; this separation however, being at still head temperature, say, about C. The proportions and composition of the two liquids at this temperature are, as stated, different from the proportions and compositions secured in settling at lower temperatures or at the ordinary temperature. The collector I lProvide with means for withdrawing a certain amount of the heavy liquid settling out of the mixtureof liquids flowing back from the reflux. The rest of the condensed liquids are allowed to overflow back into the still by appropriate connections. IVithdrawals are so controlled that about one-third of the4 more aqueous and heavier liquid coming from the reflux is withdrawn, while twothirds go back into the still system. In operatingin the way described, lthe withdrawal of about one-third of the total watery condensate will make the Withdrawal of water from the system balance the input of water with the alcohol to be dried. This input of wet alcohol may be as vapor or as liquid; the particular way of feeding the alcohol into the still forming no vpart of the present invention. Put in another way, for

every. three parts by volume of wateryv liquid condensing in the reflux in a time unit, about one part by volume is-withdrawn in a more or less continuous manner, and the residue of the reliuxed condensate is returned to qthe still. The withdrawal may be vcontrolled eitherin the manner stated: by balancing water output against water input, or, and more simply, by watching a thermometer somewhat below the collecting device stated. As long as the temperature in this thermometer stays constant, the amount of liquid withdrawn is correct. The particular temperature registered by this thermoineter willvary somewhat, according to its location in the column. Atv lower points the temperature will be greater than athigher points. Generally the temperature to be established in this. thermometer is somewhere between 72 and 74 C. Other indicating means, such as the usual continuous testers, may be employed.

Absolute alcohol is drawn olf from the base of the still in the usual manner.. It may be absolute, that is dry, or substantially so, or it may be of any controlled proportion of moisture between the 94.-95 per cent alclohol enter-ing the still and absolute alcoho The withdrawn third of the total aqueous liquid delivered by the reflux condenser to the collecting device is sent to -another column still. Here it is distilled in the usual manner; that is, water withdrawn at the base of the still and a concentrated alcoholic liquid is delivered by the usual type of condensers. All the water in the.` alcohol-supplied to the primary still passes 'off as such at the base of this .secondary still.

A condensate may be obtained containing all the alcohol and benzol; and if distillation be so conducted as to deliver a condensate-containing about 10 per cent benzol to 90 per cent of alcohol and water, the whole condensate may be then returned tothe primary still as it is, without disturbance in the operation of such primary still.

The withdrawn watery liquid, and particularly with some rather impure alcohols, generally contains some aldehyde; and it is a useful expedient in operating the second column to use a little permanganate in admixture with the li uid to be distilled, thereby oxidizing the al ehyde'.

It is found that in the practice of the present method some 90 per cent of the alcohol delivered to the primary or drying column is recovered at its base as dry alcohol, leaving only l0 per cent to be strengthened in the secondary column.

In the accompanying illustration, I have shown, more or less diagranimatically, cer-A tain assemblages of apparatus elements embodying my invention and capable of use in the performance of the described process. In this showing Figure l is a central vertical section, certain parts being shown in elevation, and

Figure 2 is a similar view of a modificatemperature of about 78.3 C. Materially above the base, but below the point of alcohol input, is the control thermometer 4 pre-v viously referred to. With the type of column shown, its readings are generally somewhere between -7 2 and 74 G. As stated, otherl indicating devices may be used but a simple thermometer suiiices.y As shown, alcohol to be dried is introduced into the still through inlet 5, shown as communicating' with the still at a midpoint. In Figure 1, above the top plate 6 of a regular conventionally shown liner system is collecting device formed by plate 7 Ithrough which pass vapor conduit 20 and overflow pipe 8. The latter extends above this plate a sufficient distance to i give a pool or body of liquid thereon. Valved outlet pipe 10 communicates withv the bottom of this body. Above this point, the still head is shown as provided with vapor outlet 11 leading to reflux condenser 12. As shown, this condenser has a Water jacket surrounding vertical tubes 13; the side and end walls being removed to show the tubes'.- Trapped liquid return pipe 14 delivers the entire condensate of the reiiux condenser into the stated collecting device somewhatbelow the level of the body of liquid therein. A vent pipe, cooled atl' is provided to take care of any permanent gases or uncondensable vapor.

Where the liquid withdrawn atf10 is sent to a second or recovery column (not shown) to remove the water and recover the alcohol and Ybenzol in concentrated form, the condensate there produced ma be returned to the column 1 either throug 4the alcohol inlet means 5 previously mentioned, or through a special inlet system 16.

The apparatus of Figure .2 is exactly the same as that of Figure 1, with the exception that the collecting device 9, in lieu of being located Within the head of the column, is4

located outside and is surrounded by heat insulating lagging 17. Overow communication with the head of the still is throughl trapped pipe 18.

The resent invention may be used indehydratlng other hygroscopic'liquids, such as the propyl and butyl alcohols,.the glycols, etc. Instead of. using benzol, other volatile liquids, insoluble, or substantiall insoluble,

in water, such as gasoline, petro eum ether, etc., may be-employed. v

What I claim is 1. In the dehydration of alcohol contain- `ing water in a column still with the aid of benzol, the process which comprises so operating'such column as to produce a' mixture of the vapors of alcohol, water, and benzol at its head having such aA composition that two liquid layers would be formed upon condensing said mixture, condensing substantiallv the entire body of vapors from the head to form a condensate at substantially still head temperature, separating and withdrawing from the condensate a minor fraction of `the layerof liquid occurring therein which contains a greater per cent of water than the other layer therein, and lreturning the rest of the 'condensate to the still.

tially still head temperature, said conden.

sate containing a layer rich in water but including a small amount of alcohol land benzol, and a second layer immiscible with 'the layer. rich in water, returning to the still -the said second layer and a major fraction of the layer rich in water, introducing fresh alcohol containing water to the still, and

'withdrawing from the system the remaining minor fraction of the layer rich in water, the amount of the minor fraction so withdrawn being that which wlll carry water in an amount equal lto the water entering the still with the alcohol containing water to be dehydrated.

3. In the dehydration of alcohol containing water in a column still Withthe aid of benzol, the process which comprises so operating such column still as to produce a mixture of the vapors of alcohol, water, and benzol at its head having such a composition that two liquid layers would be formed upon condensing said mixture, Icondensing substantially the entire body of vaporsfrom the head to form a condensate at substantially, still h ead temperature, separating and withdrawlng from the condensate about onethird of the layer rich in water, and returning to the still the other layer/and therepnaining two thirds of the layer rich in wa- 4. In the Vdehydration of alcohol containing water in a columnstill with the aid of benzol, the process which comprises so operating'such column still as to produce a mixture of the vapors of alcohol, water, and benzol at its head having such a composition that two liquid layers would be formed upon condensing said mixture, condensing substantially `the entire body of vapors from the head,vto form a condensate at substantially still head temperature, separating and withdrawing about one-third of the layer rich in water and returning to the still the other layer and the remaining *..vo thirds of Athe layer rich in water, redistilling the

Images