1175 
 y 1 
 
 SULPHITE WASTE LIQUOR 
 
 AND 
 
 ITS POSSIBLE UTILIZATION 
 
 DISSERTATION 
 
 Submitted in Partial Fulfillment of the Requirement 
 
 for the Degree of Doctor of Philosophy in 
 
 the Faculty of Pure Science, Columbia 
 
 University, in the City of 
 
 New York 
 
 By 
 
 GEORGE BARSKY, B. S., Ch. E., M. A. 
 
 New York City 
 
 1922 
 
SULPHITE WASTE LIQUOR 
 
 AND 
 ITS POSSIBLE UTILIZATION 
 
 DISSERTATION 
 
 Submitted in Partial Fulfillment of the Requirement 
 
 for the Degree of Doctor of Philosophy in the 
 
 Faculty of Pure Science, Columbia 
 
 University, in the City of 
 
 New York 
 
 By 
 George Barsky, B.S., Ch.E., M.A. 
 ^New York City 
 1922 
 
\\ 
 
 16 
 
 ACKNOWLEDGMENT 
 
 The author wishes to express to Professor Ralph H. McKee his 
 most sincere thanks for the direction and suggestion of this work and 
 for the many helpful suggestions made and encouragement offered 
 during its prosecution. 
 
 Chemical Engineering Department 
 Columbia University 
 
 Gift 
 University 
 
 m 131323 
 
 23-57 /.& 
 
5 
 
 Dedicated 
 to the late 
 
 Samuel Willard Bridgham, '67 Mines 
 
Sulphur in Sulphite Waste Liquor 
 
 Read at Annual Meeting of Technical Association of the Pulp and 
 
 Paper Association, April 15, 1922 
 Reprinted from the Paper Trade Journal, April 13, 1922, Vol. 74, No. 15, p. 315 
 
 The sulphite process for the production of pulp, involving as it 
 does the cooking of wood with calcium bisulphite solution con- 
 taining an excess of sulphur dioxide, may be regarded as a chemical 
 process having for its object the dissolving out of all of the con- 
 stituents of the wood except the cellulose. A part of the wood is 
 soluble in water 1 , *, without any noticeable chemical reaction. Ac- 
 cording to Klason," this amounts to about twelve per cent of the 
 weight of spruce wood. The reactions, whatever they are, result 
 in the addition of sulphur to the organic matter that goes into 
 solution. We have investigated the form of combination of the 
 sulphur in the liquor and from our results we have concluded that 
 there are at least three distinct forms in which the sulphur is 
 present in organic combination, in addition to that present as free 
 S0 2 , bisulphite, and normal sulphite. 
 
 It is commonly supposed that the waste liquor contains at least 
 two sulphonic acids of lignin. Klasoir, 3 has separated two "lignin- 
 sulphonic acid" salts from waste liquor, one by salting out with 
 calcium chloride and the other by treatment of the residual solution 
 from the salting out operation with naphthylamine hydrochloride. 
 He considers that the original lignin may be regarded as being 
 composed of 63 per cent of one and of 37 per cent of the other 4 , \ 
 It is possible that there may be several lignins, such as the "proto-, 
 hemi-, and ortholignins," suggested by Konig and Rump." There 
 may be several compounds iri the liquor all derived from the same 
 basic compound or compounds by some slight chemical change, 
 such as a mild oxidation, or the splitting out of sulphur dioxide 
 from the molecule, etc. 
 
 Indeed, despite the common statement that the principal con- 
 stituents of the liquor are ligninsulphonic acids or their salts, abso- 
 lute proof that some of the sulphur is present in the form of a 
 sulphonic acid grouping is lacking. Honig and Fuchs 7 found that 
 by alkali fusion of "barium ligninsulphonate" at 300° C. they could 
 split off 90 per cent of the sulphur, and could recover 15 to 19 per 
 cent of the organic matter as catechol and protocatechuic acid. In 
 the light of these results they maintain that the nucleus of the 
 lignin is similar to that of protocatechuic acid. They also con- 
 
 iShorger, A. W. J. Ind. Eng. Chem. 9, 560 (1917). 
 
 2 Klason, P. Beitrage zur Kenntniss der Chemischen Zusammensetzung der 
 Fichtenholzes. Berlin, 1911. 
 
 3 Klason, P. Arkiv. Kemi. Min. Geol., 6, No. 15, 1-21 (1917). C. A., 14, 
 2167. 
 
 4 Klason, P. Ber. 53B, 1862 (1920); C. A. 15, 859. 
 
 5 KIason, P. Ber. 53B, 1864 (1920); C. A. 15, 860. 
 
 "Konig, J. and Rump, E. Chemie und Struktur der Pflanzen-Zellmem- 
 bran. Berlin. 1914. 
 
 7 ir,nig, M. and Fuchs, W. Monatsh. 40, 341 (1919) 
 
 5 
 
sider that these results prove that the sulphur is combined in the 
 form of a sulphonic acid of an aromatic nucleus. However, this 
 is not conclusive, as they have no proof that the hydroxyl groups 
 are formed by alkali fusion in the same way that phenol is formed 
 from benzenesulphonic acid by alkali fusion. The catechol and 
 protocatechuic acid may be formed by a rearrangement during 
 fusion. Cellulose on alkali fusion will give these substances. For 
 example, Hoppe-Seyler 8 found that 50 grams of lignin-free filter 
 paper, 25 grams of KOH and 250 cc. of water, heated to 250° C, 
 gave .64 grams of almost pure protocatechuic acid and a trace of 
 catechol. 
 
 Practically all the evidence points to the view that lignocellulose, 
 or rather the lignocelluloses, since they vary somewhat among them- 
 selves, are chemical combinations of celluloses with other con- 
 stituents commonly known as lignins. There are, in all probability, 
 several lignins closely related in general structure. Concerning the 
 celluloses, there is a great deal of information available, yet the 
 subject is still one of much theorizing and experimentation." Con- 
 cerning the lignins, there is as yet very little, if any, really definite 
 information about their structures. 
 
 Several theories have been proposed, notably that lignin is a 
 product of the condensation of coniferyl and hydroxyconiferyl al- 
 cohols. This theory is generally attributed to Klason. 2 Another 
 theory is that of Cross and Bevan. 10 They are of the opinion that 
 the nucleus of the lignin molecule is a keto — R hexene group, 
 
 /CH=zCfK 
 OC<T ^CH 2 
 
 ^C— C/^ 
 (HO), (OH), 
 to which are attached other groups, such as methoxyl, acetyl, etc. 
 The form of combination between lignin and cellulose is not defi- 
 nitely known. Most authors 11 , 12 assume that it is either an ester or 
 an ether, since cellulose is known to contain alcoholic hydroxyl 
 groups, and lignin is assumed to be either an acid or an alcohol. 
 In addition to these two possible forms, ester and ether, we may 
 include another form, that of an acetal, since the ligneous material 
 present in sulphite waste liquor seems to have several carbonyl 
 groups in the molecule. These proposed forms of combination be- 
 tween cellulose and lignin are in accordance with the characteristic 
 property, 13 that of responding to hydrolytic treatment. 
 
 I. Determination of "Free S0 2 " 
 The sum of the sulphurous acid, the bisulphite, and the sulphite 
 present in sulphite waste liquor, all expressed as S0 2 , is what is 
 
 "Hoppe-Seyler. Zeit. Physiol. Chem. 13, 77-82 (1889). 
 
 8 Hibbert, H. J. Ind. Eng. Chem. 13, 256, 334 (1921). 
 "Cross and Bevan. Cellulose, 1918, p. 77. 
 ^Schwalbe, C. G. Chemie der Cellulose. Berlin, 1911. 
 12 Klein, A. Paper, 24, 351, 409 (1919). 
 "Cross and Bevan. Cellulose, 1918, p. 208. 
 
 6 
 
known as the "free S0 2 ." It is customarily determined" by titra- 
 tion with standard iodine solution, with the ordinary starch solution 
 as indicator. As the end point is a vanishing one, owing to the 
 progressive formation of more S0 2 by the decomposition of the 
 "loosely combined SO," (see below), practice is required before 
 results can be duplicated. 
 
 It was found that the primary variable factors that must be con- 
 sidered in studying this titration are the dilution and the tempera- 
 ture at which the titration is performed. 
 
 A. The Effect of Dilution 
 
 To determine the effect of dilution, 10 cc. portions were taken of 
 liquor A. 1G To each a measured amount of distilled water was 
 added. 2 cc. of starch solution (1 :200) were then added and the 
 titration performed in the usual manner. The effect of dilution 
 on the end point varies directly as the dilution, as it represents the 
 amount of standard solution required for the titration of amounts 
 of pure water. The corrections for this effect were determined by 
 running blanks with 10 cc. and 1U0 cc. of distilled water and draw- 
 ing a line, give the correction for any dilution. In this way the 
 following data were obtained and the curve plotted. (Table I, 
 Figure I.) 
 
 TABLE I. 
 
 cc. of 
 
 
 
 
 
 H 2 
 
 Cc. I 2 
 
 
 
 mg. S0 2 
 
 added 
 
 sol. 
 
 Cor. 
 
 cc. net 
 
 per cc. liq 
 
 
 
 0.6 
 
 0.0 
 
 0.6 
 
 0.07 
 
 50 
 
 0.7 
 
 0.1 
 
 0.6 
 
 O.07 
 
 1O0 
 
 0.9 
 
 O.l 
 
 0.8 
 
 0.09 
 
 150 
 
 1.0 
 
 0.1 
 
 0.9 
 
 0.1O 
 
 200 
 
 1.0 
 
 0.2 
 
 0.8 
 
 0.09 
 
 300 
 
 1.3 
 
 0.3 
 
 1.0 
 
 0.11 
 
 50O 
 
 1.6 
 
 0.5 
 
 1.1 
 
 0.12 
 
 1000 
 
 2.1 
 
 0.9 
 
 1.2 
 
 0.13 
 
 Blank 
 
 
 
 . . 
 
 
 10 
 
 0.0 
 
 
 
 
 1000 
 
 0.9 
 
 
 
 
 Titrations performed at room temperature, 20° C. 
 Iodine solution standardized against sodium arsenite. 
 1 cc. of I 2 solution equivalent to 1.09 mg. of S0 2 . 
 
 0.20 
 
 FIG. I 
 
 ZOO 400 600 800 
 
 cc. of Water Used 
 
 1000 
 
 "Sutermeister, E. Chemistry of Pulp and Paper Making, 1920, p. 191. 
 w By courtesy of Abitibi Power and Paper Company. 
 
These experiments show that the amount of free S0 2 increases 
 with the dilution at a decreasing rate. This would seem to indicate 
 that there is present some very unstable form on combination of 
 S0 2 . However,' the srhall increase in free S0 2 means that it is 
 either present in small amount or that an equilibrium exists be- 
 tween the free and combined S0 2 and that the equilibrium condi- 
 tions are satisfied by small concentrations of S0 2 . When we alter 
 the concentration of the combined S0 2 by diluting, we also alter 
 the concentration of free S0 2 required for the equilibrium. If such 
 an equilibrium exists, the concentrations of S0 2 required are very 
 small. 
 
 B. The Effect of Temperature 
 
 This effect was determined by taking 10 cc. portions, adding 50 
 cc. of distilled water and 2 cc. of starch solution, bringing to the 
 required temperature as registered by a thermometer immersed in 
 the liquid by externally heating or cooling the flask and titrating 
 immediately. Temperature has a slight effect upon the end point 
 when distilled water is titrated with iodine solution. To eliminate 
 this effect upon our data, a titration was performed with 50 cc. 
 of distilled water at the highest temperature involved. 0.7 cc. were 
 required. Thus this correction is slight and no appreciable error 
 is introduced if we consider that the correction for the other tem- 
 peratures is proportional to the temperature. From the data, Table 
 II, a curve, Figure IT, was plotted. 
 
 
 
 TABLE II. 
 
 
 
 
 Cc. l a 
 
 
 Net cc. 
 
 mg S0 2 
 
 Temp. 
 
 sol. 
 
 Cor. 
 
 I 2 sol. 
 
 per cc. 
 
 1°C 
 
 0.7 
 
 0.0 
 
 0.7 
 
 0.08 
 
 11°C 
 
 0.7 
 
 0.1 
 
 0.6 
 
 0.07 
 
 21 
 
 0.8 
 
 0.2 
 
 0.6 
 
 0.07 
 
 30 
 
 0.9 
 
 0.3 
 
 0.6 
 
 O.07 
 
 41 
 
 1.1 
 
 0.4 
 
 0.7 
 
 0.08 
 
 50 
 
 1.5 
 
 0.5 
 
 1.0 
 
 0.11 
 
 70 
 
 3.9 
 
 0.7 
 
 3.2 
 
 0.35 
 
 70 
 
 0.7 
 
 
 
 
 1 cc. Io sol. '= 1.09 mg. S0 2 . 
 
 These results indicate that profound decomposition of S0 2 bear- 
 ing substance occurs when sulphite waste liquor is heated to even 
 a moderate temperature. Therefore, for strictly concordant results 
 the free S0 2 must be titrated at a low temperature, and preferably 
 at the same temperature every time. This is most easily done by 
 adding ice to the solution. The curve, Figure 1, showing the effect 
 of dilution makes it clear that the error of dilution due to the addi- 
 tion of a small amount of ice is negligible. 
 
 II. Determination of "Loosely Combined SO2" 
 The sulphur dioxide mentioned above as easily split out of the 
 organic molecule is known as the "loosely combined." It is known 
 by this name because it is readily removable from the organic 
 molecule by treating with alkali solution. 
 
 8 
 
A. Reaction of Alkali at the Temperature Attained by Use of a 
 Water Bath 
 
 One hundred cc. of sulphite waste liquor and one hundred cc. of 
 sodium hydroxide solution (100 g. of NaOH per liter) were mixed 
 
 0.40 
 
 .0.30 
 
 8 
 
 ^020 
 
 FIG. IT 
 
 
 ' o o o —"^ 
 
 0.10 
 
 20 40 60 80 
 
 Temperature of Titration, °C. 
 
 and the flask containing the mixture placed upon a water bath. A 
 thermometer was immersed in the mixture. At intervals twenty- 
 five cc. portions were pipetted out and ice added in sufficient quan- 
 tity to bring the solution to zero degree centigrade. Sulphuric acid 
 solution (made by diluting acid of specific gravity 1.84 with an 
 equal volume of water) was added, with vigorous agitation until 
 the alkali in the liquor was just neutralized, the amount necessary 
 to neutralize 12.5 cc. of alakli solution having been previously de- 
 termined. After neutralization, the titration was proceeded with 
 immediately in the usual manner. 
 
 In this way, the data given below, Table III, were obtained, and 
 the curve representing them plotted, Figure III. Examination of the 
 curve shows that the mixture of alakli and waste liquor must be 
 allowed to remain on the water hath for at least ten minutes to 
 obtain concordant results, since the amount of SO. split out does 
 not become constant before that time has elapsed. Moreover, ii 
 is best to leave the mixture on the water bath for the same time 
 interval for each determination. 
 
B. Reaction at Room Temperature 
 
 It was next decided to investigate the reaction at room tempera- 
 ture. Equal volumes of waste liquor and NaOH solution were 
 mixed as before and 25 cc. portions removed and titrated in the 
 
 2£l 
 
 ^/.74 
 5A 
 
 ^ 
 
 
 / 
 
 HG.M 
 
 
 
 
 
 _4r: 
 
 
 
 
 
 
 
 
 
 
 
 
 ol- 
 
 io 20 30 
 
 Time in Minutes 
 
 40 
 
 50 
 
 manner described above. Thus we obtain the data of Table IV, 
 shown graphically in Figure IV. Similarly, with liquor B, 18 data for 
 the same kind of a curve were obtained, Table V, Figure 5. 
 
 1.50 
 
 \l.00 
 ^6 .50 
 
 FIG. JV 
 
 r~~~* 
 
 10 20 30 40 
 
 Time in Minutes 
 
 50 
 
 60 
 
 Examination of these curves shows that the reaction is complete 
 or rather comes to final equilibrium in about ten minutes. 
 
 If the temperature is raised above room temperature, say, to the 
 temperature attainable by means of a water bath, an additional 
 amount of S0 2 will be split off. If the mixture is cooled to room 
 
 16 Bv courtesy of Price Bros. & Co. 
 
 10 
 
temperature again, we find that the amount of S0 2 split out is 
 still the same as that split out at the higher temperature. This 
 proves that the increase in amount is not due to a displacement 
 
 
 
 
 TABLE 
 
 Ill 
 
 
 Time 
 
 
 
 
 
 mg. SO. 
 
 elapsed, min 
 
 Temp. 
 
 °C 
 
 I 2 sol. cc. 
 
 per cc. liquor 
 
 
 
 
 44 
 
 
 12.3 
 
 1.07 
 
 11 
 
 
 88 
 
 
 21.9 
 
 1.90 
 
 21 
 
 
 89 
 
 
 22.2 
 
 1.93 
 
 31 
 
 
 89 
 
 
 23.7 
 
 2.06 
 
 45 
 
 
 88 
 
 
 25.2 
 
 2.19 
 
 Original 1 
 
 quor, 
 
 25 
 
 cc. 
 
 1.2 
 
 0.05 
 
 25 cc. of sodium hydroxide solution required 6.5 cc. of H 2 S0 4 solution for 
 neutralization with phenolphthalein as indicator. Therefore 3.2 v.c. were 
 added each time to neutralize the NaOII before titrating. 
 
 1 cc. of standard I» solution equivalent to 1.09 mg. of S0 2 . 
 
 
 TABLE 
 
 IV. 
 
 (Liquor 
 
 A) 
 
 
 Time 
 
 
 
 
 mg. 
 
 of S0 2 per 
 
 elapsed, min. 
 
 
 cc. I 2 
 
 sol. 
 
 cc 
 
 . of liquor 
 
 
 
 
 
 
 
 0.05 
 
 1 
 
 
 12.7 
 
 
 
 1.00 
 
 3 
 
 
 15.7 
 
 
 
 1.24 
 
 5 
 
 
 16.1 
 
 
 
 1.27 
 
 10 
 
 
 16.6 
 
 
 
 1.31 
 
 15 
 
 
 16.5 
 
 
 
 1.30 
 
 25 
 
 
 17.3 
 
 
 
 1.37 
 
 40 
 
 
 17.5 
 
 
 
 1.38 
 
 60 
 
 
 17.3 
 
 
 
 1.37 
 
 Room temperature 23° C. 
 
 1 cc. of standard I 2 solution equivalent to 0.987 mg. of S0 2 . 
 
 TABLE V. (Liquor B) 
 Time 
 elapsed, min. I 2 sol. cc. 
 
 ... 
 
 1 15.6 
 3 16.3 
 5 16.5 
 
 10 16.1 
 
 15 17.1 
 
 25 17.3 
 
 40 16.9 
 
 60 17.8 
 
 100 18.2 
 
 130 18.3 
 
 Ro< m temperature 23° C. 
 
 1 cc. of standard solution of I 2 equivalent to 0.835 mg SO 
 
 FIG. Y 
 
 mg. of S0 2 
 per cc. liquor 
 0.02 
 1.04 
 1.09 
 1.10 
 1.07 
 1.14 
 1.15 
 1.13 
 1.19 
 1.21 
 1.22 
 
 10 
 
 20 30 40 
 
 T/me in Minutes 
 n 
 
 50 
 
 60 
 
of the equilibrium with the change of temperature. Were the in- 
 crease in amount of S0 2 split out due to such a temperature dis- 
 placement of the equilibrium, then reversion to the original tem- 
 perature should result in a decrease to the amount split out at the 
 original temperature. Therefore, the reaction at the higher tem- 
 perature must include a different one from that at the lower tem- 
 perature. Table VI and Figure 6 show the splitting out of addi- 
 tional S0 2 when the temperature is raised. 
 
 
 
 TABLE VI 
 
 
 Time 
 
 
 
 mg. S0 2 
 
 elapsed, min. 
 
 Temp., 
 
 C C I2 sol. cc. 
 
 per cc. liquor 
 
 
 
 23 
 
 
 0.05 
 
 1 
 
 23 
 
 12.7 
 
 1.00 
 
 3 
 
 23 
 
 15.7 
 
 1.24 
 
 5 
 
 23 
 
 16.1 
 
 1.27 
 
 10 
 
 23 
 
 16.6 
 
 1.31 
 
 IS 
 
 23 
 
 16.5 
 
 1.30 
 
 25 
 
 23 
 
 17.3 
 
 1.37 
 
 40 
 
 23 
 
 17.5 
 
 1.38 
 
 60 
 
 23 
 
 17.3 
 
 1.37 
 
 90 
 
 23 
 
 17.9 
 
 1.41 
 
 91 
 
 26 
 
 17.9 
 
 1.41 
 
 93 
 
 32 
 
 17.9 
 
 1.41 
 
 95 
 
 38 
 
 17.7 
 
 1.40 
 
 102.5 
 
 53 
 
 18.9 
 
 1.49 
 
 105 
 
 58 
 
 19.1 
 
 1.51 
 
 115 
 
 74 
 
 20.8 
 
 1.65 
 
 130 
 
 85 
 
 22.1 
 
 1.75 
 
 140 
 
 87 
 
 23.0 
 
 1.82 
 
 150 
 
 89 
 
 23.2 
 
 1.83 
 
 160 
 
 89 
 
 23.4 
 
 1.85 
 
 170 
 
 89 
 
 24.4 
 
 1.93 
 
 1 cc. of Io solution equivalent to 0.986 mg of S0 2 . 
 
 The examination of the velocities of reaction also shows a dif- 
 ference of reaction under certain conditions. If the same reactions 
 proceed at a higher temperature, then the velocity of reaction is 
 greater at the higher temperature. Assume, for example, that we 
 
 FIG. VI 
 
 J. •— — • • • 
 
 ..v* 
 
 ioo no izo 130 
 Time in Minutes 
 
 have two reactions taking place at different temperatures. If the 
 velocity of the reaction at the higher is smaller than the velocity 
 of the reaction taking place at the lower temperature, then the re- 
 actions are not identical, since for a given reaction an elevation of 
 temperature results in an increase in velocity of reaction. 
 
 The ideal way of performing an experiment to determine the 
 velocity of the reaction between sulphite waste liquor and alkali 
 at a temperature higher than room temperature, would be to heat 
 
 12 ' 
 
up the alkali solution and the waste liquor separately to the de- 
 sired temperature and then to mix the two. Samples for analysis 
 could be withdrawn from time to time. However, this procedure 
 could not be followed because when waste liquor is heated up 
 separately, a loss of S0 2 occurs. 
 
 For this reason the experiment was carried out by heating up 
 the alkali solution and then adding the waste liquor to it. This 
 method had the disadvantage that the temperature is not uniform 
 during the course of the experiment, but rises from the minimum 
 
 Z50 
 
 FIG. W 
 
 10 
 
 20 30 40 50 
 
 Time in Minutes 
 
 60 
 
 and finally reaches a maximum. However, since at all times the 
 temperature is higher than room temperature, the same data, though 
 not exactly quantitative, would indicate a difference of reaction. 
 Accordingly, the experiment was carried out in this fashion. 
 
 The apparatus used consisted of a 500 cc. short-necked round- 
 bottom flask, equipped with a stopper, through which passed a 
 thermometer, reaching almost to the bottom, and a short piece of 
 tubing serving as an air condenser to prevent excessive evaporation. 
 The flask was immersed to about three-quarters of its volume in a 
 boiling water bath. The samples for analysis were withdrawn by 
 temporarily removing the stopper and using a pipette. 
 
 The 100 cc. of alkali solution were first placed in the flask and 
 the temperature allowed to reach the desired point. The 100 cc. 
 of sulphite waste liquor were then added and the contents of the 
 flask thoroughly mixed. The temperature was noted then and at 
 the time of the removal of each sample. The initial time was taken 
 at the instant when the addition of the waste liquor was com- 
 pleted. 
 
 13 
 
In these experiments the molecular ratio of NaOH to S0 2 split 
 out is about 80 to one. In other words, the NaOH is present in 
 so large an excess that its concentration does not alter appreciably 
 during the course of the reaction and therefore we may neglect the 
 changes in concentration of NaOH in our discussion. 
 
 2.00 
 
 1.50 -, 
 
 F/G.VHL 
 
 WO 
 
 8 
 
 u O «l 
 
 7*" - 
 
 if 
 
 100 
 
 50 
 
 <o 
 
 I 
 
 10 20 30 40 
 
 Time in Minutes 
 
 50 
 
 60 
 
 In the manner described above, data were obtained for the re- 
 actions at the higher temperature of the two liquors previously 
 investigated. These data are shown in Tables VII and VIII, and 
 graphically in the corresponding curves. 
 
 Time 
 elapsed 
 min. 
 
 1 
 3 
 5 
 
 10 
 IS 
 25 
 40 
 65 
 90 
 
 Table VII (liquor A) 
 
 Temp. 
 ° C 
 
 65 
 74 
 87 
 94 
 97 
 97 
 97 
 97 
 97 
 97 
 
 Allowed to cool to 23 
 
 I 2 sol. 
 cc. 
 
 19.7 
 21.6 
 21.7 
 23.5 
 24.1 
 25.2 
 25.9 
 ■25.8 
 26.3 
 25.8 
 
 mg. S0 2 
 per cc. 
 
 liquor 
 
 0.05 
 1.65 
 1.81 
 1.82 
 1.97 
 2.02 
 2.11 
 2.17 
 2.16 
 2.20 
 2.16 
 
 1 cc. of standard I 2 solution equivalent to 1.047 mg. of S0 2 . 
 
 Time 
 
 
 elapsed 
 
 Temp. 
 
 min. 
 
 ° C 
 
 
 
 69 
 
 1 
 
 73 
 
 3 
 
 88 
 
 6 
 
 94 
 
 10 
 
 97 
 
 15 
 
 97 
 
 25 
 
 99 
 
 40 
 
 98 
 
 60 
 
 98 
 
 90 
 
 98 
 
 
 Cooled to 23 
 
 Table VIII (liquor B) 
 
 I 2 sol. 
 cc. 
 
 .21 A 
 22.7 
 23.4 
 24.3 
 24.6 
 25.7 
 26.5 
 26.6 
 27.4 
 27.0 
 
 mg. S0 2 
 per cc. 
 liquor 
 0.02 
 1.43 
 1.51 
 1.56 
 1.62 
 1.64 
 1.71 
 1.76 
 1.77 
 1.82 
 1.82 
 
 1 pp. of standard I 2 solution equivalent to 0.835 mg. of S0 2 . 
 
 14 
 
Let us examine these curves. At the low temperature, the re- 
 action is completed in approximately ten minutes. At the high tem- 
 perature, the reaction cannot be considered complete until about 
 forty minutes have elapsed. In other words, the net reaction at the 
 higher temperature is slower than that at the low. This proves 
 that there must be another reaction taking place at the higher tem- 
 perature. This reaction proceeds so slowly that the net velocity, 
 involving the low temperature reaction speeded up by the elevation 
 of the temperature, is less than the velocity of the low temperature 
 reaction. 
 
 In addition to the discussion of the data on the velocities of the 
 reactions at the different temperatures, we have available other 
 data supporting our conclusion, i. e., the data on the actual amounts 
 of S0 2 split out. For example, with liquor A the amount of SO, 
 split out at room temperature was 1.38 mg. per cc. The amount 
 split out at the higher temperature was 2.16. One and one-half 
 times 1.38 is 2.07. In other words, it looks as if for every two 
 moles of SO, split out at the low temperature, there is an addi- 
 tional mole which can be removed at the higher temperature. The 
 same ratio is true for liquor B. The figures for this liquor are 1.22 
 at the low temperature and 1.80 at the high. One and one-half 
 times 1.22 is 1.83. Tabulating these figures for convenience in 
 examining them we have : 
 
 -Split out at- 
 
 Low temp. High temp. Ratio 
 
 Liquor L H L:H 
 
 A 1.38 2.16 2:3.13 
 
 B 1.22 1.80 2:2.95 
 
 Considering the error in titrating and the probable presence of 
 many compounds in small amounts, the agreement of the figures is 
 quite good. 
 
 We have additional evidence of the presence of these two kinds 
 of S0 2 in combination. When sulphite waste liquor is boiled, there 
 occurs an escape of SO, from the liquor. For example, in a modi- 
 fication of one of the industrial processes for the production of 
 alcohol by the fermentation of sulphite liquor, one step consists in 
 boiling the liquor to lower its SO, content. This is assisted by the 
 passage of air through the liquor to carry off the SO,. Since the 
 waste liquor is acid, this may be considered to be the effect of acid 
 
 boiling commenced 
 
 
 TABLE IX 
 
 
 Time 
 
 
 mg. S0 2 
 
 elapsed 
 
 I, sol. 
 
 per cc. 
 
 hrs. :min. 
 
 cc. 
 
 liquor 
 
 0:00 
 
 17.7 
 
 2.42 
 
 1:25 
 
 16.9 
 
 2.32 b 
 
 2:00 
 
 13.8 
 
 1.89 
 
 2:36 
 
 12.9 
 
 1.77 
 
 3:14 
 
 12.2 
 
 1.67 
 
 4:10 
 
 11.8 
 
 1.62 
 
 5:16 
 
 11.9 
 
 1.63 
 
 6:10 
 
 11.5 
 
 1.58 
 
 6:38 
 
 11.8 
 
 1.62 
 
 7-15 
 
 11.5 
 
 1.58 
 
 1 cc. of standard T 2 solution equivalent to 1.37 mg. of SO, 
 
 IS 
 
treatment. After several hours of boiling, the S0 2 content reaches 
 a nearly constant figure. 
 
 We have performed the experiment in the laboratory and have 
 examined the SO, content during the course of boiling. Some of 
 the data are here reproduced, together with their graph, Table IX, 
 Figure 9. 
 
 In the analyses, the S0 2 was split out by alkali at a high tem- 
 perature, that attainable by a water bath. These data lead to the 
 same conclusion as the data on the velocity of reaction, namely, 
 that there are two kinds of SO, and that one kind is present in an 
 
 F/G.JX 
 
 o 
 
 3 4 5 6 
 
 Time in Hours 
 
 amount twice that of the other. For example, one liquor showed 
 2.42 mg. of SO., per cc. before heating was commenced, and after 
 about four hours, when the S0 2 content had reached a minimum, 
 or rather a nearly constant figure, 1.62 mg. per cc. The amount 
 removed by boiling was 0.80 mg. per cc, while the amount un- 
 affected was 1.62. One is approximately twice the other. 
 
 The total amount of sulphur is the liquor, expressed as milli- 
 grams of S0 2 per cc. is about 7 in the case of liquor A. Therefore, 
 in addition to the two kinds of SO, already discussed, there must 
 be at least one more kind, and there probably are several. These, 
 however, are not removed by alkali treatment nor by boiling. 
 
 16 
 
Summarizing: 
 
 1. The titrations for free S0 2 and loosely combined S0 2 have 
 been investigated, and the conditions laid down for the most ac- 
 curate titrations. 
 
 2. The splitting out of S0 2 by means of alkali has been in- 
 vestigated, and it has been found that there are two kinds of SO, 
 split out. One is split out at room temperature and the other only 
 when the temperature is elevated. 
 
 3. The removal of S0 2 from sulphite waste liquor by boiling 
 has been investigated and it has been found that after about four 
 hours the S0 2 content reaches a nearly constant figure. Of the 
 loosely combined, one part is removed by boiling while two parts 
 remain unaffected. 
 
 17 
 
Lignin Sulphonic Acids 
 
 Reprinted from Paper Trade Journal, May 18, 1922, Vol. 74, No. 20, p. 46. 
 
 The technology of the sulphite pulp process is quite involved. 
 The yield and the character of the pulp varies with the cooking 
 liquor and with the manner in which the cook is carried out. 
 Chemically, the process consists of the digestion of the wood in 
 calcium bisulphite solution containing an excess of sulphurous acid. 
 The digestion takes place under pressure and lasts about 10 hours.' 
 The net result of this operation is the conversion into soluble 
 materials of all the non-cellulose constituents of the wood. Some 
 of these constituents dissolve without reaction, others merely un- 
 dergo hydrolysis, while the balance react with the bisulphite and 
 sulphurous acid in the liquor. 
 
 In this investigation we have been concerned with the nature of 
 the substances which have reacted with bisulphite and sulphurous 
 acid. Our primary purpose was to ascertain whether there is but 
 one, or several such substances. Our results led to the conclusion 
 that there is a number of such substances. 
 
 A large number of woods are characterized by constituents of 
 an adventitious or transient nature, such as the tannins, gums, es- 
 sential oils, alkaloids, etc. In some cases these have been thorough- 
 ly investigated and a few are widely used in industry and in the 
 arts. But comparatively little is definitely known of the fundamental 
 tissues which we may speak of as the wood-substances. There 
 have been many attempts to resolve these wood-substances into 
 proximate constituents but so far these attempts have met with 
 but little success. The problem is fraught with many difficulties 
 and though it has been attacked by many investigators 2 the results 
 are as yet inconclusive. ' So far as they lend themselves to in- 
 terpretation they would seem to indicate that there is some degree 
 of uniformity in composition despite the structural disparity and 
 the widely varying character of the substances found in the woods 
 of different species. 
 
 The gradation from the cotton cellulose, through the different 
 kinds of cellulose, on through the so-called lignin or non-cellulose, 
 is really so gradual that it is not possible to make a sharp line of 
 demarcation. The celluloses obtained from woods, being products 
 of the resolution and decomposition of the wood, vary both in 
 character and proportion with the treatment by which they are 
 prepared. 
 
 1. A typical cook would be approximately for ten hours at 140°'C. with a 
 pressure of about 90 lbs. At the start of the cook the liquor would contain 
 about 4.4% total sulphur dioxide of which 1.3% would be combined as cal- 
 cium bisulphite. 
 
 2. The chemical literature in general and the literature on cellulose abounds 
 with references on this subject. 
 
 18 
 
Closely related to cotton cellulose, which is usually taken as 
 the standard type of cellulose, are the celluloses of flax, hemp, 
 china grass and others. These are obtained from the plants in 
 question by some purifying process. They present slight differ- 
 ences from cotton cellulose in external physical characteristics and 
 chemical properties. This indicates slight differences in chemical 
 constitution. 
 
 Next we have the celluloses which are characterized by a higher 
 percentage of oxygen, the presence of active carbonyl groups and 
 sometimes the presence of mcthoxyl groups. These are further 
 characterized by the splitting off of furfural on treatment with 
 hydrochloric acid. They may be termed natural oxycelluloses. 
 
 We have still a further group, that of the hemicelluloses, closely 
 resembling the true celluloses but easily hydrolyzed into simple 
 carbohydrates by the action of dilute acids or alkalies. Then we 
 have the carbohydrates known as the lignins which represent still 
 further variation of cellulose structure. 
 
 In brief, we have in wood tissues combinations of these sub- 
 stances and of the celluloses making up the fibers. These are 
 termed "lignocelluloses." 
 
 Most authors :i assume that the form of combination existing be- 
 tween cellulose and lignin (terms which we use in the generic 
 sense) is either (a) that of an ether, or (b) an ester, or a form 
 involving both. In support of (a) it may be said that the 
 ligneous material in sulphite waste liquor has alcoholic hydroxy! 
 groups. Cellulose is known to contain alcoholic hydroxyl groups. 
 It is the union of these groups that gives the ether linkage. For 
 (b) it is necessary to assume that lignin is an acid. Evidence of' 
 this acidic character is given by the solubility of lignin in alkalies 
 and by the acid character of groups, such as acetyl, that may be 
 split off from it. 4 In addition to these two linkages, we may add 
 the possibility of an acetal, since the waste liquor shows certain 
 aldehyde reactions, e. g., with phenylhydrazine. All these forms 
 of combination between lignin and cellulose are in accordance with 
 the characteristic property of the lignocelluloses, that of responding 
 to hydrolytic treatment. 4 
 
 We may regard the reactions in the digester as made up of a 
 hydrolysis, followed, or rather accompanied, by the interaction of 
 the non-cellulose products of the hydrolysis with the chemicals 
 of the cooking liquor. In this way the reverse reaction, that of 
 condensation, is prevented, the hydrolysis promoted, and the unde- 
 sirable substances made soluble. 
 
 In the sulphite process, consideration of the possible reactions 
 that may take place reveals that there may be any or several of 
 the following : s 
 
 3. Schwalbe, C. G Die Chemie der Cellulose. 1911. 
 
 4. Cross and Bevan Cellulose. 1918. 
 
 5. Schwalbe, C. G. Loc. cit. 
 
 19 
 
(1) The formation of an aldehyde or ketone addition product: 
 
 | OH 
 
 caHSO 3 + C = O -> C( 
 
 SO»ca* 
 
 (2) The saturation of a double bond : 
 
 caHS0 3 + C = C->-C— C— 
 
 | | H SO«ca 
 
 (3) E'sterification : 
 
 I I 
 
 caHS0 3 + — COH->— COSO,ca + H..0 
 
 , I, I 
 
 (4) Simultaneous oxidation and sulphonation : 
 
 I I 
 
 caHSO, + — CH -> CS0 3 ca + (H.,) 
 
 I I 
 
 In all cases where a separation is effected between the cellulose 
 and the non-cellulose, it is unsafe to assume that the bond between 
 the two classes of materials is merely broken. There probably 
 always occurs some alteration of one, and more likely both, of the 
 substances dissolved and the substances remaining insoluble in the 
 reagent. Of the two major constituent groups, the celluloses have 
 been much investigated because of their many uses. The lignin 
 compounds have had no uses and accordingly have been more or 
 less neglected. 
 
 The sulphite waste liquor remaining after the digestion of the 
 wood and the removal of the cellulose varies in color from a light 
 yellow to a dark brown, is slightly acid and smells slightly of SO;.. 
 It has a specific gravity of about 1.05 and contains about 10 per 
 cent of solids. It is strongly reducing to Fehling's solution and 
 reacts with phenylhydrazine to give a copious tarry mass. 
 
 Lindsey and Tollens 7 found that after removal of sulphuric 
 acid from the sulphite waste liquor by means of barium hydroxide, 
 they could obtain a heavy precipitate with lead acetate. It had a 
 composition, the organic portion of which was expressed as 
 C 2( ,H 3 o0 12 . They also obtained a precipitate with hydrochloric acid 
 and assigned the formula C 20 H 3 o SO 10 to it. A brominated derivative 
 was also prepared to which they assigned the formula C 2G H 2S Br 4 
 SO n . They concluded that the major part of the dissolved organic 
 substances behaves as a homogeneous complex as they were unable 
 to resolve it into its proximate constituents. 
 
 Seidel and Hanak, 8 after removal of sulphuric acid precipi- 
 tated certain material by adding alcohol. The percentage of CaO 
 was determined and a part of the precipitate then converted into 
 
 6. ca — yi C-i. 
 
 7. Lindsey and Tollens. .. Ann. 267, 341 (1892). J. Soc. Chei'n. Ind. 
 
 11, 835 (1892), 12 287 (1893) Z. angew. Chem. 
 5 154 (1892). 
 3. Seiden and Hanak J. Soc. Chem. Ind. 17 596, 863 (1898). 
 
 20 
 
the barium and a part into the zinc salt. The authors found that 
 the ratio of the metals combined in the salts was practically the 
 same as that of their atomic weights. From this they concluded 
 that the precipitate obtained was a salt of a definite organic acid 
 which is the principal constituent of the organic matter in the 
 waste liquor. 
 
 Krause 9 obtained a chlorine derivative by adding bleaching 
 powder to the liquor. This derivative, after purification with 
 alcohol and ether, showed on analysis a composition corresponding 
 to the formula C* H 2 o C1S0 3 ,. 
 
 Klason 10 separated "calcium lignosulphonate" by adding 
 crystallized calcium chloride as long as any went into solution. He 
 obtained a heavy precipitate which was filtered and washed with 
 alcohol. This salt was decomposed by adding sulphuric acid to its 
 solution in just sufficient quantity to react with all the lime present. 
 The calcium sulphate formed was filtered off and the solution con- 
 centrated. Alcohol was then added to complete the precipitation of 
 the calcium sulphate. The alcohol was evaporated from the filtrate, 
 the solution diluted and neutralized with barium hydroxide. Any bari- 
 um sulphate found was allowed to settle out. The addition of alco- 
 hol then precipitated the so-called barium lignosulphonate. Analysis 
 of this salt pointed to the formula d H M 0, 7 S_. Ba. Molecular 
 weight determinations gave results in the neighborhood of 6,000. 
 Accordingly Klason assigns the formula (Cm T J ,- O u ) a to the 
 lignin. To the C, H J2 0„ Klason assigns 3.7 methoxyl groups and 
 1.1 hydroxyl groups. Of this work it may be said that there is 
 nothing involved in the method that would allow us to consider 
 this material a compound. In fact the analytical data immediately 
 indicate that it must be a mixture. Hence it is a little far fetched 
 to make deductions concerning lignin with this as a basis. 
 
 Klason 11 later discovered that there was a "calcium lignosul- 
 phonate" which was not precipitated upon the addition of calcium 
 chloride but which remained in solution. It could be precipitated 
 by means of naphthylamine hydrochloride. This latter he terms 
 a |3 lignin derivative. That precipitable by calcium chloride he 
 called an a lignin derivative. 
 
 Honig and Spitzer 1 " attempted to separate the material ol 
 the liquor by fractional precipitation with alcohol but all their 
 fractions with one exception, calculated as salts of lignosulphonic 
 acid, showed sensibly the same composition. 13 Melander 14 found 
 that the product salted out of waste liquor with sodium chloride 
 
 9. Krause, H T. Sec. Chem. Ind. 25 493 (1906). 
 
 10. Klason, P Beitrage zur Kenntniss der Chcniisclicn Zusam- 
 
 mcnsetzmi" der Ficlitcnholzes. 1911. 
 
 11. Klason, P Chem. Zentr. 90 92 (1919). J. Soc. Chem. 
 
 Ind. 38 570A (1919). 
 
 12. Hiinig, M Monatsh. 39 871 (1918). 
 
 13. Klason, P Ber. 53?! 1864 (1920). 
 
 14. Melander, K. H. A Cellulosechem. 2 41, 69, (1921), Paper 28 
 
 No. 21 p. 19 (1921) T. Soc. Chem. Ind. 40. 
 620A (1921). Chem. Soc. Abs. 116Pt, (1919). 
 
 21 
 
differed from the product obtained by Klason with calcium chloride. 
 
 The precipitation schemes given above are in agreement with the 
 customary schemes for the precipitation of an emulsoid colloid by 
 the use of a strong electrolyte or by the addition of alcohol. Where 
 we have present a mixture of substances in the colloidal state, in 
 general such methods would result in the precipitation of mixtures. 
 
 By precipitating, dissolving and reprecipitating, Hofmeister™ 
 succeeded in obtaining pure albumen (i. e., crystallized) from 
 colloidal albumen. Von Weimarn, 16 adopting the same principle, 
 prepared crystalline gelatine and agar, typical colloids. We have 
 adopted a similar procedure to separate the material precipitated 
 with calcium chloride, with the idea of studying its purity. We 
 have fractionally precipitated and then refractionated after dis- 
 solving. These experiments were as follows : 
 
 Three liters of liquors 17 were evaporated to about 80Occ. and 
 the calcium sulphate filtered off. The resultant solution was placed 
 in a beaker and stirred mechanically. Calcium hydroxide suspen- 
 sion was added until the solution was neutral to litmus paper. 
 Crystallized calcium chloride was then added in 50 g portions until 
 a precipitate appeared, and the' solution heated on the water bath 
 for about two hours, i. e., until the precipitate coagulated. There- 
 upon it was filtered off with suction, sucked as dry as possible, and 
 weighed. In all cases the procedure was exactly the same, so that 
 the percentage of moisture in the precipitate was the same. An 
 additional SO g of calcium chloride were added to the filtrate and 
 the precipitate so obtained treated in the same manner as the 
 preceding one. The addition of calcium chloride was continued 
 until no further precipitation took place. In this way, a fractiona- 
 tion of the calcium chloride precipitate was effected and the data 
 given in Table I obtained and curve in the figure drawn. The 
 wide range during which precipitation takes place, 150 g to 400 g of 
 calcium chloride, would seem to indicate that there is a mixture 
 being precipitated. 
 
 Table I 
 
 'Total g 
 
 
 
 Grams 
 
 CaCI 2 
 
 
 
 total 
 
 added 
 
 Ppt. 
 
 Ppt. 
 
 Ppt. 
 
 SO 
 
 None 
 
 
 
 
 
 100 
 
 None 
 
 
 
 O 
 
 ISO 
 
 None 
 
 
 
 
 
 200' 
 
 A 
 
 95 
 
 95 
 
 250 
 
 B 
 
 85 
 
 180 
 
 300 
 
 C 
 
 80 
 
 260 
 
 350 
 
 D 
 
 45 
 
 305 
 
 400 
 
 E 
 
 15 
 
 320 
 
 450 
 
 
 
 
 320 
 
 The precipitates A, B, C, D, and E were dissolved in propor- 
 tionate amounts of water, 2 cc. per gram. The appearances of the 
 
 15. Hofmeister, F Z. physiol. Chero. 14, 165 (1889), 16—187 
 
 1892). 
 
 16. Von Weimarn, P. P. . .Grundzrige d. Dispersoid Chemie. 1911. 
 
 17. By courtesy of the Hammermill Paper Co. 
 
 22 
 
different solutions were quite distinctive: A, muddy; B, black; 
 C, dark wine color; D and E, a lighter brown. These solutions 
 were treated in a manner similar to the original evaporated liquor 
 in an attempt to accomplish a still further separation of the com- 
 pounds. The data are shown in the Tables II, III, IV, from 
 which precipitation curves similar to that in the figure can be 
 drawn. 
 
 
 Table II 
 
 
 
 Total g 
 
 
 
 Grams 
 
 CaCI 2 
 
 
 Grams 
 
 total 
 
 added 
 
 Ppt 
 
 Ppt. 
 
 Ppt. 
 
 
 
 
 
 
 
 
 25 
 
 a; 
 
 25 
 
 25 
 
 50 
 
 A 2 
 
 17 
 
 42 
 
 75 
 
 A 3 
 
 7 
 
 49 
 
 100 
 
 Table III 
 
 
 
 49 
 
 Total g 
 
 
 
 Grams 
 
 CaCl 2 
 
 
 Grams 
 
 total 
 
 added 
 
 Ppt. 
 
 Ppt. 
 
 Ppt. 
 
 
 
 
 
 
 
 
 10 
 
 
 
 
 
 
 20 
 
 
 
 
 
 
 30 
 
 
 
 
 
 
 40 
 
 B, 
 
 40 
 
 40 
 
 50 
 
 B 2 
 
 10 
 
 so 
 
 60 
 
 B 3 
 
 6 
 
 56 
 
 70 
 
 Table IV 
 
 
 
 56 
 
 Total g 
 
 
 
 Grams 
 
 CaCl 2 
 
 
 Grams 
 
 total 
 
 added 
 
 Ppt. 
 
 Ppt. 
 
 Ppt. 
 
 
 
 
 
 
 
 
 10 
 
 
 
 
 
 
 20 
 
 Ci 
 
 14 
 
 14 
 
 30 
 
 Co 
 
 10 
 
 24 
 
 40 
 
 c 3 
 
 4 
 
 28 
 
 50 
 
 
 
 
 28 
 
 A„ B 1( Ci, D and E (D and E were not fractionated because of 
 the small quantity) were then converted to the barium salt by the 
 method described by Klason. The method used for the several 
 fractions was the same. In every case the final precipitation was 
 accomplished by pouring the aqueous solution into twice its volume 
 of 95 per cent alcohol. The precipitates were sucked dry, washed 
 with 95 per cent alcohol, and dried for several days over con- 
 centrated sulphuric acid. 
 
 These barium salts were then analyzed by organic combustion^ 
 to determine the percentage of carbon and hydrogen. The 
 sample subjected to combustion was contained in a platinum boat. 
 To provide against any sulphur that might be burned off, a plug 
 of lead peroxide-minimum mixture was inserted in the combustion 
 tube in the manner usual for such compounds. The percentage of 
 ash was determined by weighing the residue from the combustion. 
 The analytical data are given in Table V. 
 
 18. Fisher, H. L Laboratory Manual of Organic Chemistry, 1920. 
 
 23 
 
Table V 
 
 Wt. of Wt. of H 2 CO. 
 
 sample ash found found % Ash % C % H 
 
 Ai .2116 .1136 .0547 .2132 53.69 27.48 2.89 
 
 B, .2035 .0398 .0848 .3559 19.56 47.69 4.66 
 .2181 .0431 .0936 .3858 19.75 48.24 4.80 
 
 Av. 19.66 
 
 C, .2368 .1016 .3975 
 
 .2121 .0453 .0849 .3531 21.36 
 
 Av. 21.36 
 
 I) .2257 .0950 .3358 
 
 .2220 .0592 ' .0915 .3278 26.66 
 
 Av. 26.66 
 
 K .2028 .0510 .0824 .3184 25.14 
 
 .2029 .0524 .0790 3189 25.82 
 
 Av; 25.48 42.85 
 
 Because of the likelihood of there being extraneous mineral 
 matter such as barium chloride, etc., contaminating the barium 
 salts, the analyses were calculated to the ash-free basis. The re- 
 sults are given in the following table. (Table Va). 
 
 47.97 
 
 4.73 
 
 45.67 
 
 4.80 
 
 45.40 
 
 4.48 
 
 45.54 
 
 4.64 
 
 40.55 
 
 4.41 
 
 40.26 
 
 4.61 
 
 40.41 
 
 4.51 
 
 42.82 
 
 4.5 4 
 
 42.88 
 
 4.36 
 
 
 
 Table Va 
 
 
 
 
 
 
 
 
 
 
 Ash-f 
 
 ree 
 
 Basis 
 
 
 
 
 % Combustible 
 
 % c 
 
 % II 
 
 
 A 
 
 
 Ratio C/ 
 
 
 % Ash 
 
 % C 
 
 
 \ 
 
 %H 
 
 11 
 
 Ai 53.69 
 
 46.31 
 
 27.48 
 
 2.89 
 
 59.4 
 
 
 6.2 
 
 9.7 
 
 
 tii 19.66 
 
 80.34 
 
 47.97 
 
 4.73 
 
 59.7 
 
 
 5.9 
 
 in.1 
 
 
 Ci 21.36 
 
 78.64 
 
 45.54 
 
 4.64 
 
 57.9 
 
 
 5.9 
 
 9.8 
 
 
 D 26.66 
 
 '73.34 
 
 40.41 
 
 4.66 
 
 55.1 
 
 
 6.4 
 
 8.6 
 
 
 E 25.48 
 
 74.52 
 
 42.85 
 
 4.45 
 
 57.6 
 
 
 6.0 
 
 9.6 
 
 
 Cellulose 
 
 
 
 
 44.2 
 
 
 6.3 
 
 7.0 
 
 
 We can now plot a curve showing the composition of the ma- 
 terial fractionated by the calcium chloride precipitation. For ex- 
 ample, Aj is the material which comes down when precipitation 
 starts, Bi the material precipitated when 200 grams of calcium 
 chloride have been added, and so on. This curve is plotted on the 
 same sheet as the one showing the course of the precipitation, with 
 the grams of calcium chloride as absiccas. 
 
 After sufficient calcium chloride has been added to start precipi- 
 tation, there occurs a steady precipitation of lignin mixtures for 
 the range that the precipitation curve is a straight line. When 
 the lignins have all been precipitated, or nearly so, the most diffi- 
 cultly precipitable material comes down. This stage is represented 
 by the section of the curve from D to E. During this interval the 
 curve of precipitation is not a straight line, but indicates from its 
 shape that the substances coming down are not as readily precipi- 
 tated as those preceding it, and consequently might be expected to 
 differ from them in composition. 
 
 24 
 
The elementary analyses show percentages of carbon varying 
 from 597 to 55.1, percentages of hydrogen varying from 6.4 to 5.9. 
 The ratio of percentage of carbon to that of hydrogen varies from 
 8.6 to 10.1. This ratio for cellulose is 7.0. In other words, the 
 ratio of carbon to hydrogen is higher in lignin than in cellulose. 
 
 50 100 150 200 250 300 350 400 450 
 Grams of Calcium Chloride Added 
 
 The above all leads to the interesting conclusion that the barium 
 salt accepted by Klason and others as a more or less single definite 
 compound is not by any means a single substance, but on the 
 contrary, contains several substances of varying composition. This 
 mixture accounts for the strange values Klason obtained for the 
 hydroxyl number, etc., of his salts and for the discrepancies that 
 have been found by various workers. 19 
 
 19. Loc. cit. 7, 8, 9, 12, 14. 
 
 25 
 
Sulphite Waste Liquor as a Fuel 
 
 In the manufacture of alcohol 1 from sulphite waste liquor it -is of 
 advantage to concentrate the liquor to about 20% solids to reduce 
 the size of the installation required for fermentation and distilla- 
 tion. In conjunction with the alcohol process, the liquor might be 
 further evaporated and utilized as a liquid fuel. It could be burned 
 as a pitch containing 50 per cent solids by spraying through a 
 burner in a way similar to that in which the waste liquors of the 
 sulphate process are burned in one recovery process 2 . This would 
 be an addition to the power plant of the pulp mill. 
 
 When the sulphite liquor obtained after the removal of the al- 
 cohol by distillation is evaporated, there remains a residue showing, 
 on the dry basis for a typical liquor, a heat value of 7950 B. t. u. 
 per pound with an ash content of about 13.9 per cent. For each 
 ton of pulp there is obtained about 1,100 gallons of liquor after the 
 removal of the alcohol. This weighs 9,900 pounds and contains 14 
 per cent of solids, i. e., 1,386 pounds. To concentrate this to 50 
 per cent solids requires the removal of 7,128 pounds of water with 
 the formation of a mobile pitch containing 50 per cent solids and 
 weighing 2,772 pounds. 
 Heat value of the pitch containing one pound of solid : 
 
 B. t. u. developed by burning the solid 7,950 
 
 B. t. u. required 
 
 to heat the water from 120° to 212° F 92 
 
 to evaporate the water 970 
 
 to heat this vapor to flue temperature estimated 
 
 as 500° F. .47 (500-212) 135 
 
 B. t. u. unavailable because of presence of water 1,197 
 
 B. t. u. available 6,753 
 
 If this pitch is burned under a boiler to give steam at 150 pounds 
 pressure, there will be obtained an efficiency of say 60 per cent. 
 Then there will be obtained from feed water at 120° F : 
 6,753 X -60 
 
 = 3.66 pounds of steam at 150 pounds. 1,107 is 
 
 1,107 
 the number of B. t. u. required to convert 1 pound of water at 120° 
 F. to one pound of steam at 150 pounds gauge pressure. The steam 
 generated can be used for the development of power in non-con- 
 densing steam engines. The exhaust of such engines can be used 
 to avaporate the liquor in multiple effect evaporators. On the basis 
 of this system these calculations follow : 
 A simple non-condensing steam engine, using steam at 150 pounds 
 
 1. McKee, R. H. U. S. Pat. No. 1,273,392, July 23, 1918. 
 Paper, 24, 584 (1919). 
 Pulp Paper Mag., Canada 18, 715 (1920). 
 2. Moore, H. K. Trans. Amer. I. Chem. Eng., 10, 177 (1917). 
 Paper 25, 1157, 1197, 1241 (1920). 
 
 26 
 
pressure, has a steam consumption oi about 30 pounds per indi- 
 cated horsepower hour. Therefore from the steam there can be 
 obtained 
 3.66 
 
 = .122 I. H. P. per pound of residue. 
 
 30 
 The exhaust steam from this type of engine is at a pressure of 
 three to five pounds and is about 88 per cent dry. Assuming that 
 owing to pipe condensation and other heat losses it delivers but 75 
 per cent of the theory to the evaporators, there are delivered to the 
 evaporators 
 
 .88 X -75 X 3.66 = 2.42 pounds of low pressure steam per 
 pound of residue. 
 In a plant producing one hundred tons of pulp per day, there can 
 be obtained 
 
 .122 
 
 100 X 1,386 X = 703 I. II. P. per 24 hour day 
 
 24 
 100 X 1,386 X 2.42 -— 335,000 pounds of low pressure steam 
 To evaporate the 1,100 gallons to 50 per cent solids as calcul- 
 ated above requires the removal of 7,128 pounds of water. A triple 
 effect evaporator of the ordinary type requires about .372 pounds 
 of low pressure steam per pound of water evaporated and there- 
 fore can evaporate 7,128 pounds of water if furnished with 2,650 
 pounds of steam. For a hundred ton plant this amounts to 265,000 
 pounds. Therefore we have an excess of steam above the require- 
 ments for the evaporation of the liquor to the state of 50 per cent 
 solids. This amounts to 335,000 less 265,000 or 70,000 pounds of 
 low pressure steam per hundred tons of pulp. 
 
 In other words, for a plant producing one hundred tons of pulp 
 per day, the fuel available will give, in addition to the energy re- 
 quired for the evaporation of the discharge from the alcohol still 
 to a concentration adapted for burning, 703 I. H. P. for 24 hours 
 and 70,000 pounds of low pressure steam. 
 
 Although such an installation has the important advantage of dis- 
 posing of the waste liquor nuisance, it might be argued that it would 
 prove a financial burden to the pulp mill because of the operating 
 charges. The following approximate tabulation of costs and credits 
 shows that such an argument is groundless. 
 Triple-effect evaporator to evaporate 30,000 pounds of water 
 
 per hour $75,000 
 
 Boiler 1,000 H. P. installed 40,000 
 
 Engines, 700 H. P. installed 14,000 
 
 $129,000 
 
 20 per cent interest, depreciation, repairs, etc 25,800 
 
 Labor, 2 x 24 x 50 x 300 7,200 
 
 $33,000 
 27 
 
With the horsepuwei .year at $190 and coal at $6, the yearly- 
 credit value would be : 
 
 For 703 H. P 63,300 
 
 For 70,000 pounds of steam per day at 10 pounds per pound 
 
 of coal for 300 days 6,300 
 
 Total yearly credit value $69,600 
 
 Cost of operation 33,000 
 
 Net yearly credit value $36,600 
 
 28 
 
CONCLUSIONS 
 
 In this investigation the following were accomplished: 
 
 (1) It was discovered that the loosely combined sulphur dioxide 
 is present in sulphite waste liquor in two distinct forms of com- 
 bination. This is a contribution to the knowledge of the chemistry 
 of the sulphite process as well as to the knowledge of the com- 
 position of the waste liquor. 
 
 (2) A method has been devised for effecting at least a partial 
 separation of the lignin compounds. This is an important advance 
 in the study of their structure. 
 
 (3) A method has been proposed, which in conjunction with 
 the manufacture of alcohol, utilizes a part of the material, disposes 
 of the remainder, and eliminates the nuisance, at a profit. 
 
 29 
 
VITA 
 
 George Barsky was born August 26, 1896, in the City of New 
 York. He received his elementary education in the New York 
 City Public Schools from which he graduated in 1910. He gradu- 
 ated from Townsend Harris Hall in 1913 and entered Columbia 
 College, from which he received the B. S. degree in 1916. He en- 
 tered the Engineering School in 1916 and was awarded the degree 
 of Chemical Engineer in October, 1918. During the year 1917- 
 1918 he held the position of Research Assistant in the Department 
 of Chemistry, Columbia University. On October 1, 1918, he was 
 called into active service in the U. S. Army, of which he had been 
 a member in reserve. Upon his honorable discharge in January, 
 1919, he was employed by Professor Ralph H. McKee as a research 
 assistant. 
 
 From July, 1919, to July, 1921 he held the Samuel Willard Bridg- 
 ham Fellowship. Under the grant of this fellowship, the foregoing 
 investigation was pursued. In October, 1921, he received the degree 
 of M. A. from Columbia University. 
 
 He is the author of the following articles : 
 
 Fuel Value of Volatile Liquid Mixtures, J. Ind. Eng. Chem. 12, 
 77 (1920). 
 
 Fuel From Sulphite Waste Liquor (with Ralph H. McKee), 
 Paper 26, 368 (1920). 
 
 30 
 
018 370 8bl h