Nov., 19511 HODGSON AND GLOVER 636 The Determination of Alcohol, Ether and Water in Ternary Mixtures BY H. W. HODGSON AND J. H. GLOVER A method is described for the analysis of all mixtures of alcohol, ether and water with an accuracy of 0.2 per cent. The composition is obtained from a calibration graph constructed from data obtained by boiling-point determina- tions and “cloud-point” titrations. A modification to the 3-tube for the boiling-point determination has effected an improvement in technique. The . method is simple and rapid in operation, and would be applicable to other ternary mixtures containing water. IN connection with the evaluation of the performance of rectification columns a rapid method was required that would be applicable to mixtures of all compositions of alcohol, ether and water.Existing methods1s2~3~4s6se~7 depend on separation or chemical means of estimation and consequently are time consuming and lack the flexibility necessary to cover all proportions of the constituents. Among the physical methods, Andress8 determined the specific gravity and surface tension and then evaluated the composition from a calibration diagram, and Scotta used refractive-index and specific-gravity measurements. These methods were found unsuitable owing to the volatility of mixtures that contain a high proportion of ether and to the fact that it is difficult to define the exact point of intersection of the lines on the graph owing to the small angle between them. The Andress method has the further disadvantage that several mixtures have the same specific gravity and surface tension.The advantage of indirect methods for the analysis of ternary mixtures lies in the accuracy and speed with which certain physical properties can be determined. The choice of properties for measurement is governed by the calibration diagram and so, to avoid difficulties such as are encountered in the Andress and Scott methods, two constants were selected, of which each is dependent on a different component of the mixture. Hence, in the calibration diagram, the two sets of lines cross almost at right angles, so permitting accurate location of any mixture. The properties determined are (i) a titration value that depends on the solubility of dibutyl phthalate in the sample and (ii) the boiling-point of the sample.The titration is carried out by adding a solution of dibutyl phthalate to the sample and then titrating, at constant temperature, with water until precipitation of the phthalate begins. This is a simple and reproducible procedure and the titration is influenced mainly by the water content of the mixture. The boiling-point of the sample is determined by measuring the temperature at which the vapour pressure is 760 mm of mercury. The apparatus is a modified &tube designed for easy filling. The accuracy of the proposed method depends on the scale and accuracy of the calibration diagram, and is 0-2 per cent. in the method described. However, for mixtures in which one component predominates a variation of the dibutyl phthalate reagent would provide greater accuracy over a narrower range.The proposed method could be applied to other ternary mixtures of which water is a component, and the titration technique alone can be employed to analyse binary mixtures containing water. METHOD APPARATUS- bore except for the portion from E to H which is 1-mm capillary tubing. fitting stopcock. tions directly opposite each other. The boiling-point is affected most by the ether content of the sample. The &tube is constructed as shown in Fig. 1 of thin-walled Pyrex tubing of 4 to 5 mm Tap C is a well- A millimetre scale is etched on the U portion of the tube with the gradua-636 HODCSON AND GLOVER: THE DETERMINATIOK OF ALCOHOL, [I-ol. 76 REAGENT- Dibutyl PhthaZate solution-A 1 per cent. w/v solution in ethanol previously adjusted with water to sp.gr. (15.6" C/155" C) 0.8070 PROCEDURE- Determination of the titration value-Weigh 10 g * 0.1 g of the sample into a dry 150-ml conical flask modified by the addition of a vertical side arm to take a thermometer.Add by pipette 10 ml of dibutyl phthalate reagent, the temperature having been previously adjusted to 20" & 1" C. Insert a thermometer, adiust the temperature to 19" C, and then 0.0002. "t F- Imm I.D. 7mm O.D. \ A 5mm I.D. 7mm O.D. -G Scale graduated in mlllimetres D Fig. 1. Modified d-tube for boiling-point determination (half-scale) titrate wiih distilled water, maintaining the tem- perature at about 19" C, until one drop of water produces a permanent opalescence in the solution. Adjust the temperature to 20" & 0.1" C, when the solution should become clear, and then continue the titration dropwise until a permanent opalescence is again produced.This point is taken as the end- point. Duplicates should be reproducible within 0.1 ml. Determiizat ion of the boiling-Po i d . --Introduce pure, dry, mercury into the limb A of the J-tube, attach B to a vacuum line and carefully open tap C so that mercury is drawn up to the tap without being allowed to enter the tap itself. The amount of mercury used should be such that the tube is now filled from C to a point just round the bend at D. Close tap C, disconnect B from the vacuum line, arid apply the vacuum at A to bring the mercury to approximately the same level in both limbs. Gently waJm the U portion of the tube to remove occluded air, and then carefully break the vacuum at A so that the mercury returns slowly to its original position.All the air is now trapped below the stopcock C. Introduce two drops of the sample on to the mercury surface at D in limb A by means of a capillary pipette. Carefully tilt the tube to transfer the sample into limb B so that it is trapped at E. A capillary thread of mercury separates the sample from the trapped air at C. Immerse the filled tube in a bath of water to the level FG. Slowlv raise the temDerature of the water, stirring it continuously until the mercury levels in (he tube begin to'alter. At this point reduce the rate of heating to about 1°C per minute. Note the temperature at which the vapour in limb B is at 760 mm pressure, this point being determined from the difference in height of the mercury levels in relation to the atmospheric pressure.Duplicates should be reproducible within 0.1" C. CALIBRATION GRAPH- The calibration graph (Figs. 2 and 3) was prepared from the results described in the experimental section below. The ordinate of the graph is the percentage of water and the abscissa the percentage of ether in the mixture, the percentage of alcohol being obtained by difference. A series of titration lines connect points representing mixtures with the same titration value, and a similar series of boiling-point lines are drawn connecting mixtures of the same boiling-point. I t has been found more convenient to adopt the use of rectilinear co-ordinates rather than the more usual triangular form. Interpretation of results is simplified and the angle of intersection between the two sets of lines is larger.The scale adopted in the construction of the graph is made consistent with the accuracy of the method, 1 mm on the scale corresponding to 0.2 per cent. except in the region of high water content (50 to 100 per cent.), where the ether scale was doubled, 1 mm corresponding to 0.1 per cent., toALCOHOL - ETHER - WATER CALIBRATIQN DIAGRAM WATER 50 TO 100 PER CENT Titre, ETHER Fig. 2Tit re, Qc w I- s ETHER Fig. 3Nov., 19511 ETHER AND WATER IN TERNARY MIXTURES 637 facilitate the reading of the boiling-point lines. The boiling-point lines have been labelled at 5" C intervals and the titration lines at intervals of 1 ml, but intermediate values can easily be interpolated.EXPERIMENTAL DETERMINA4TION OF THE TITRATION VALUE- A large number of mixtures of alcohol, ether and water were prepared and titrated as Examination of the results indicated that the titre depended on two described above. factors- (a) the water content of the sample, and (b) the E value of the sample. The E value is defined by the following expression- 100 A E = A = the weight of ether in l o g of sample, and B = the weight of water in l o g of sample. (10 - B) + 7.6 where- The E value defines the slope of the line relating titration to water content. The constant 7.6 in the equation is the weight of alcohol added in 10ml of dibutyl phthalate reagent. The E value of any known mixture can thus be calculated. For mixtures containing up to 80 per cent.of water, the relation between the titre and water content was found to be linear for a fixed E value. For mixtures containing more than 80 per cent. of water, the relation between the titre and the water content is almost independent of the E value since, owing to phase restrictions, this cannot exceed 5; and the change in slope of the line relating titration to water content over a range of E from 0 to 5 is negligible for the small titrations obtained for mixtures containing more than 80 per cent. of water. Results illustrating these points are shown in Table I. The effect of a change in E value is illustrated by mixtures 22 to 27. For mixtures having E values of 0 to 15 the increase in titration is linear with increase in E value for a fixed water content, a titration maximum is reached at E = 18 and then the titration decreases in a smooth curve as E increases further.Mixtures of E value greater than 34 cannot be titrated, as phase separation occurs before the end-point is reached. Such mixtures may be analysed after addition of an equal weight of alcohol (of known water content) to a portion of the sample. This modified material is analysed by boiling-point determination and titration and its composition read from the calibration graph. The composition of the original sample can then be calculated by allowing for the alcohol added. The occurrence of phase separation is easily distinguished from the true titration end-point. Mixtures 28 to 37 illustrate the independence of titration and E value if the water content is above 80 per cent.Hence the results of the titration of known mixtures of alcohol, ether and water can be summarised as follows- (1) A plot of titration against water content for mixtures containing less than 80 per cent. of water gives a series of straight lines whose slopes depend on the E value of the mixtures, i.e., all mixtures having the same E value lie on the same line. (2) Mixtures containing more than 80 per cent. of water give titrations very nearly independent of their E value. (3) Mixtures with an E value greater than 34 cannot be titrated without prior dilution with alcohol. DETERMINATION OF THE BOILING-POINT- It was originally intended to determine the vapour pressure of the sample at a fixed temperature, but attempts to carry out such a determination in a simple manner led to results lacking in reproducibility.It was therefore decided that a logical alternative would be the determination of the temperature at which the vapour pressure of the mixture was a fixed value, and this was chosen as 760mm of mercury for convenience.638 HODGSON AND GLOVER: THE DETERMINATION OF ALCOHOL, [Vol. 76 The &tube apparatus was first investigated, but was found to be very difficult to fill. It was almost impossible to confine the sample above the mercury in the sealed limb without the introduction of air bubbles, and the initial charging of the tube with mercury was subject to the same difficulties. TABLE I DEPENDENCE OF TITRATION ON WAfER CONTENT AND E VALUE Mixture No. 1 2 3 4 6 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 26 26 27 28 29 30 31 32 33 34 35 36 37 Alcohol, 100 80 60 40 20 10 0 % Ether, Yo 0 0 0 0 0 0 0 Water, % 0 20 40 60 80 90 100 34 6 60 53 7 40 72 8 20 91 9 0 45.95 14.06 40 63.95 16.05 20 81-96 18-06 0 30.4 19.6 50 38.9 21.1 40 55.9 24.1 20 72.9 27.1 0 31.9 47.9 63.9 28.1 32.1 36.1 90 10 80 20 70 30 60 40 50 50 46 54 8.0 9.0 10.0 5.0 6.0 6.5 7.0 4-0 4.6 5.0 2.0 1.0 0 2.0 1.0 0.5 0 1.0 0.5 0 40 20 0 90 90 90 93 93 93 93 95 96 95 E value 0 0 0 0 0 0 0 Titration a t 20° c, ml 30.9 24.56 18.15 12-05 6-86 2.90 0.45 5.2 12.1 5.2 18.6 5.2 2646 5-2 31.6 10.4 10-4 10.4 19.0 26.6 32-3 15.6 15.9 15-6 18-95 15.6 26-1 15.6 32.8 20.8 20.8 20.8 19.25 26.1 32.85 5.7 31.7 11.35 32-75 17.05 32.85 22.7 32.6 28.4 31.15 30.7 30- 15 1.22 1.1 1 0 2.3 1-15 0.57 0 1.18 0.59 0 3.1 3.1 2.9 2.35 2-35 2.35 8.3 1.76 1.75 1.75 Finally, a modified &tube was evolved, the design of which allowed easy charging with mercury and removal of occluded air; this has been described on p.635. By the method described on p. 636, the boiling-points of a large number of ternary mixtures were determined. Mixtures were prepared by dilution with water of a series of solutions each containing a fixed ratio of alcohol to ether. As the work proceeded it became apparent that a number of unusual results were being obtained. For example, in the series containing alcohol and ether in the ratio of 9 to 1, maxima were obtained with mixtures containing 2, 54 arid 97 per cent. of water. With mixtures in the ratio of 8 to 2, maxima were obtained at water contents of 2, 4, 57 and 97 per cent.; similar maxima were obtained with the 7 to 3 and 6 to 4 ratios, except that the maximum at about 50 per cent.disappeared. The effect of ether on the boiling-point is quite considerable, but an increase in water content has little effect; the boiling-point for mixtures containing up to 70 per cent. of water changes by only a few degrees.Nov., 19511 Alcohol, % 100 95.5 91 80 74 69 64 59 ETHER AND WATER I N TERNARY MIXTURES TABLE 11 RESULTS OF BOILING-POINT DETERMINATIONS Binavy mixture (no . h e y ) - Water, B.p. (760 mm), % " C 0 78.4 4.5 78.1 9 78.2 20 78.8 26 79.4 31 79.8 36 80.2 41 80-6 Mixtures of alcohol - ether ratio 9 to 1- Alcohol, Ether, 90 10 88.85 9.9 88-2 9.8 85.5 9.5 82.8 9.2 81.0 9.0 72.1 8.0 62.9 7.0 53.8 6-0 49.6 5-5 44.9 5.0 42-65 4.85 41.4 4.6 40.5 4.5 38.25 4-26 36.2 4-0 264 3.0 18.0 2.0 13.6 1.5 % % 9.15 1 -05 6.76 0.76 4.5 0.5 3.6 0.4 2-7 0.3 2.25 0.25 Mixtures of alcohol - ether ratio 8 to 2- Alcohol, Ether, 80 20 78.8 19.7 78.1 19.5 77.6 19.4 76.7 19-2 76-2 18.8 73.6 18.4 71.7 17.9 63.8 15.9 56.7 13.9 48.0 12-0 40.0 10.0 36.4 9.1 36.0 9.0 36-2 8-8 34.2 8.6 33.5 8.4 32.0 8.0 24.0 ' 6.0 16.0 4.0 8.0 2.0 4.0 1.0 76.8 19.0 % % Water, 0 1.25 2.0 5.0 8.0 10.0 19.9 30.1 40.2 44.9 60.1 52-6 54.0 55.0 57.5 59.8 70.1 80.0 86.0 89.8 92.6 96.0 96-0 97.0 97.5 % Water, 0 1.5 2.4 3.0 4.1 6.0 8.0 10.4 20.3 30.4 40.0 60.0 54.5 65.0 66.0 67.2 68.1 60.0 70.0 80.0 90.0 96.0 6-2 30.1 % 56-9 14.0 -~ ~ 634 16.0 20.1 Alcohol, Water, % 50 45 41 30 20 15 10 5 % Fi 0 55 59 70 80 85 90 95 639 B.p.(760 mm), "C 81-2 81.6 82-0 83.5 86.0 88.0 90-0 94.5 B.p. (7,gO mm), C 68.0 68-3 (maximum) 68-0 67.8 68.0 68.3 68.8 69.0 69.3 69.1 68.7 69.8 70.4 (maximum) 68.3 67.7 68-0 69.8 74.7 78.5 81.0 86.0 90.9 92.8 95.2 (maximum) 94-6 B.p. (760 mm), O C 59.2 60.2 (maximum) 59-9 58-9 60.8 (maximum) 60.4 60.4 60.3 60-3 59.5 57.6 57-1 67.3 57.6 68.2 59-6 (maximum) 67.6 66.9 58.8 63.8 70.4 86.2 60.6 69.5 60.3640 HODGSON AND GLOVER: THE DETERMINATION OF ALCOHOL, [Vol. 76 TABLE I I-continued Mixtures of alcohol -ether ratio 7 to 3- Alcohol, Ether, 70 30 68.7 29.5 67.9 29.1 66.9 28.7 66.2 28.3 644 27.6 62.7 26.9 69.2 25.3 56.9 24.1 35.1 15.1 33.3 14.3 31.7 13.6 29-4 12.6 28.2 12.1 20.8 8.9 14.2 6.1 7.3 3.2 10.5 4.6 % % Mixtures of alcohol - Alcohol, 60 59.7 59.3 68.8 58-5 57.9 67.4 66.6 66.9 86.2 64.0 52-6 50.4 45-6 40.8 39.6 38-4 37.2 34-8 32.4 30.0 27.6 3.6 3.0 1-8 1.6 % ether ratio 6 lo 4- Ether, % '40 39.8 39.6 39.2 39.0 38.6 38.2 37.8 37-3 36.8 36.0 35.0 33.6 30.4 27.2 26.4 26.6 24.8 23.2 21.6 20.0 18.4 2-4 2.0 1.2 1.0 Water, 0 1-8 3.0 4-4 5.6 8.0 10-4 15.5 19.0 49.8 52.4 54.7 58.0 59.7 70.3 79.7 89.6 85-0 % Water, 0 0.5 1.2 2.0 2-5 3.5 4.4 6.6 6.8 8.0 10.0 12.4 16.0 24.0 32.0 34.0 36.0 38.0 42.0 46.0 60.0 54.0 94.0 95.0 97.0 97.6 % B.p.(760 mm), OC 62.5 64.2 64.3 53.9 63-2 64-1 54.5 54.3 63.8 48.5 48.0 47.7 48.1 48.3 49.8 51.5 69.5 53.6 B.p. (7060 mm), C 48.2 48.4 48.6 48.8 48.9 49.4 49.4 49.1 49.1 49.5 48.9 49.2 48.6 47.1 46.2 46.1 46-0 45.5 45.0 43.6 42.6 41.6 76-6 80.0 90.0 92.5 Mixtures of alcohol - ether ratio of less than !3 to 1 show a decrease in boiling-point with increasing water content up to about 50 per cent.of water. Difficulty was found in repro- ducing the boiling-points of 8 to 2 ratio mixtures in the vicinity of the maximum at 57 per cent. of water; this maximum is presumably due to an unstable phase equilibrium at this composition. Results are shown in Table 11. All other points were found to be reproducible. CALCULATION OF TITRATION LINES- The fact that the titration bears a linear relation t o the water content for mixtures containing less than 80 per cent. of water can be used to calculate the -titration lines up toNov., 19511 ETHER AND WATER IN TERNARY MIXTURES 641 this water content. The equation relating titration to water content over the range 0 to 80 per cent.of water can be written- where- T = T O - m W T is the titration value; W is the water content; To is the titration when water = 0; m is a constant depending on the E value of the mixture. where- T, is the titration for mixtures containing 80 per cent. of water. This is constant Values of m and To for various E values calculated in this manner are shown in Table 111. for all values of E. TABLE I11 CALCULATED VALUES OF To AND m FOR VARIOUS VALUES OF E E To m 0 30.9 0.3131 6 31.55 0.3212 10 32.2 0.3294 15 32.8 0.3369 18 32-9 0.3381 20 32.8 0.3369 22 32-06 0.3350 25 32.1 0.3281 28 31.26 0.3175 30 30.45 0,3075 32 29.3 0.2931 34 27-6 0.2719 TABLE IV CALCULATED VALUES FOR TITRATION LINES (VALUES OF W) Titra- tion 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 E=O - - 2.9 6.1 9.3 12.5 15-6 18.8 22.0 25.2 28-4 31.6 34.8 38.0 41.2 44.4 47.6 50.8 54.0 57.2 60.4 63.6 66-8 70.0 73.1 76-3 79.5 82.5 86.2 89.6 93.2 97.2 E=;S 1.7 4-8 7.9 11.1 14.2 17.3 20.4 23.5 26.6 29.7 32.8 35.9 39.1 42.2 46.3 48.4 51.5 54.6 57.8 60.9 64.0 67.1 70.2 73.4 76.4 79.6 82.5 86.2 89.6 93-2 97.2 - E=lO 0.62 3.6 '6.7 9-7 12.7 15-8 18.8 21.8 24.9 87.9 30-9 34.0 37-0 40.1 43.1 46.1 49.2 52-2 55.3 68.3 61.3 64.4 67-4 70.4 73.5 76.5 79.5 82.5 86-2 - - - E= 15 2.4 5.3 8.3 11.3 14.2 17.2 20.2 23.1 26.1 29.1 32.0 35.0 38-0 40.9 43.9 46.9 49.9 62.8 55.8 68.7 61.7 64-7 67.7 70.6 73.6 76.6 - - - - - E=18 2.7 5.6 8-6 11.6 14.5 17.5 20.5 23.4 26.4 29.3 32.3 3 5 3 38.1 41.1 44.1 47.0 50.0 63-0 55.9 58.9 61.8 64.8 67.7 70.7 73.7 76.6 - - - - - - E=20 2.4 5.3 8.3 11-3 14.2 17.2 20.2 23.1 26.1 29.1 32.0 36.0 38.0 40.9 43.9 46.9 49.9 52.8 55.8 58.7 61.7 64.7 67.7 70.6 73.6 76.6 - - - - - - E=22 1.9 4.9 7.9 10.9 13.9 16.9 19.9 22.8 25.8 28.8 31.8 34-8 37.8 40.7 43.7 46.7 49.7 52.7 55.7 58.7 61.6 64.6 67-6 70.6 73.5 73.5 - - - - - - E=26 0.3 3.3 6.4 9.4 12.6 15.5 18.6 21-6 24.7 27.7 30.8 33.8 36.8 39.9 42.9 46.0 - - - - - - - - - - - - - - - -642 HODGSON AND GLOVER: THE DETERMINATION OF ALCOHOL, [Vol.76 Titration lines are drawn from points calculated by substituting the above values for m and To in the equation TO - T w =-- m This gives W for fixed E values, and hence the percentage of ether can be calculated. For example, if the titration line for 20 ml is being constructed, points are obtained as (a) E = 0, To = 30.9, m = 0.3131.follows- 30.9 - Water = W = - = 34.8 per cent. As the E value = 0, the percentage of ether = 0 Therefore, by difference, the percentage of alcohol = 65.2 per cent. (b) E = 5, To = 31-66, m = 0.3212 Water = W = o.3212 = 35.95 per cent. Since the E value = 6, the percentage of ether = 7.0 per cent. Therefore, by Merence, the percentage of alcohol = 57-06 per cent. 31.55 - 20 A series of points can be obtained in this way and the titration line constructed. The calculation of the ether content of a mixture from the E value and water content can be done graphically. This is the most convenient way when a large number of points are required. Water, % 0 1 2 3 4 6 8 10 12 16 20 24 28 32 36 40 44 60 52 54 56 58 60 64 70 75 80 86 90 92 94 96 97 98 48 46' C 50.4 - - 48.0 50.0 (46" C) 51.0 (46" C) 50.4 48.5 39.6 42-5 33-8 31.5 29.4 27.1 24.3 21.8 19.0 17.7 16.7 16.1 14.4 11.8 14.2 12.4 10.5 - - - - - - - - - - 50" C 35.6 36.9 36.5 34.6 37.0 35.0 35.8 34.2 33.2 3 1.4 29.4 26-2 24.3 22.1 20.1 18-4 17.0 15.1 14.0 13.2 12.0 12.4 11.8 10.9 10.2 8.8 7.5 6.3 - - - - - - - 55" c 25.9 28e.5 28.3 26.8 27.1 26.6 26.0 26.0 25-7 24.0 22.0 20.1 18.7 17.0 15.4 13.9 12.7 11.6 11.1 10.5 10.2 10.3 9.7 8.4 8.0 7.2 6-4 5.5 4.4 - - - - - - TABLE V BOILINGPOINT LINES Values for ether contents 60" C 19.0 20.2 20.1 18-0 19.6 19.4 19.4 18.4 18.0 17.3 16-3 15.1 14.1 12.8 11.6 10.6 9.6 8.8 8.6 8.3 8.2 5-8 7.3 6.4 6.0 5.6 5.2 4.6 4.1 - - - - - - 66" C 13.0 13.6 13.5 12.8 12.8 12-7 12.8 12.4 12.2 11.8 11.5 10.7 10.1 9.3 8.5 7-8 7.1 6.5 6.5 6.2 6.4 5.5 5 .2 4.8 4.4 4.0 3.8 3.8 3-6 - -- - - - - 70' C 7.8 7.9 7.9 7.5 7.3 7.0 7.0 7.0 7.0 7.0 7.0 6-8 6.4 6.2 3.8 5.5 5.1 4.5 4.5 4.3 4.8 3.6 3.4 3.3 3.0 2.7 2.8 2.8 5.7 - - - - - - 76" C 2-7 2.8 2.7 2-4 2.2 2.2 2.2 2.3 2.4 2.9 2.8 3-1 3.1 3.1 3.0 2.9 2.7 2.7 2.6 0.5 3.0 2.0 2.0 9.0 1.8 1.6 1-8 1.8 2.0 2-4 2.6 2.8 - - -. 80" C 85" CNov., 19511 ETHER AND WATER I N TERNARY MIXTURES 643 The titration lines of mixtures containing more than 80 per cent. of water are most easily obtained from a graph of experimental results. I t will be noted that the higher titrations will give negative values for W when the above equation is used at various E values. A negative result indicates that the titration line does not cover mixtures of the E value in question.Table IV shows points for titration lines calculated in this manner. CONSTRUCTION OF THE BOILING-POINT LINES- The boiling-points of known mixtures already listed in this paper are used in the con- struction of the boiling-point lines on the calibration graph. First the boiling-points of mixtures are plotted against their water content for fixed ratios of alcohol to ether. From the resulting curves, points may be obtained for the construction of a series of graphs relating the percentage of ether to the boiling-point for fixed water content mixtures. Finally, the boiling-point lines may be constructed from points obtained from the series of graphs. The composition of all mixtures boiling at any given temperature can readily be found by this method. Where maxima and minima are encountered in the first set of curves a sufficient number of points must be taken for the fixed water content curves to cover the effect of the minima and maxima on the boiling-point lines. Points obtained by this method are shown in Table V. The authors thank the Directors of the British Oxygen Company Limited for permission to publish. REFERENCES 1. Juzikhin, A. N., Zavod. Lab., 1937, 6, 1013. 2. “Allen’s Commercial Organic Analysis,” Fifth Edition, J. & A. Churchill Ltd., London, 3. Szeberenyi, P., 2. anal. Chem., 1916, 54, 409. 4. 5. 6. 7. 8. Andress, G. M., Chcm. Met. Eng., 1944, 114. 9. RESEARCH AND DEVELOPMENT DEPT. BRITISH OXYGEN COMPANY LTD. Volume I, p. 160. Johnson, J. J., Ind. Eng. Chem., Anal. Ed., 1932, 4, 20. Bonner, T. G., Analyst, 1947, 72, 47. Shaefer, W. E., Znd. Eng. Chem., Anal. Ed., 1944, 16, 432. - , A7tal. Chem., 1948, 20, 661. Scott, T. A., J. Phys. COIL Chent., 1946, 50, 406. ANALYTICAL LABORATORY LONDON, S.W.19 March, 1961