Determination of Unsaturation in Organic Compounds - Analytical

---

INDUSTRIAL AKD ENGINEERING CHEMISTRY

140

VOL. 10, NO. 3

Oxidation of Hydrogen and Methane

ane. The recommended procedure is to absorb all possible components of the gas mixture, including carbon monoxide, and determine hydrogen and hydrocarbons over the catalyst a t 610" C. As carbon monoxide is oxidized a t 300" C., carbon monoxide, hydrogen, and methane may be

From the volume decrease and carbon dioxide the hydrogen

Conclusions

complete oxidation is obtained. However, larger percentages of excess oxygen permit faster passage of the gas through the catalyst tube.

Discussion

Literature Cited

Methane and ethanearecompletelyoxidizedovercommercial pl.ttinized silica gel a t 610" C. It may be assumed that higher hydrocarbons are also completely oxidized, as previous work ( 3 ) shows that they are more readily oxidized than meth-

(1)

Kobe, K. .4.,ISD. EKG.CREM., Anal. Ed., 3, 262-4

(1931)

(2) Kobe, K. A., and h v e s o n , E. J., I b i d . , 5, 110-12 (1933). (3) Kobe, K. A.2 andBrOokbank9 E. E., Ibid., 6, 35-7 (1934).

RECEIYED December IO, 1937.

Determination of Unsaturation in Organic Compounds By Means of the Mercury-Catalyzed Reaction with Standard Bromate-Bromide HOWARD J. LUCAS

AND

DAVID PRESSIIAN, California Institute of Technology, Pasadena, Calif.

I

T HAS been known for a long time that the reaction of bromine with compounds having a triple bond does not go to completion rapidly. I n the case of acetylene, Davis, Crandall, and Higbie ( 2 ) showed that the slowness and incompleteness of the reaction is due t o t'he presence of oxygen, and that the reaction is aided by aluminum, mercury, and nickel salts. h'lulliken and Wakeman (8) found that in general liquid alkynes and alkadienes do not react quantitatively with standard bromate-bromide solution. Recently (4) the analysis of acetylene in aqueous solution has been carried out quantitatively in the presence of mercuric sulfate, and without taking precautions against oxygen. I n the attempt to develop a method for determining unsaturation, a study has been made of the behavior of bromate-bromide solution, in the presence and in the absence of mercuric sulfate, F i t h a number of unsaturated compounds, some having a triple and some a double bond.

Materials c. P. glacial acetic acid was used in addition to the reagents previously described (4). Carbon tetrachloride was purified by saturating with chlorine, exposing t o direct sunlight for 10 to 12 hours, washing with aqueous sodium hydroxide, and drying with calcium chloride. I t distilled at 75.6" C. uncorrected. Several unsaturated ccmpounds were prepared for this work. The following were used after purification: dichloroethylene (Eastman's), distilled once; maleic acid (Pfanstiehl); fumaric acid (Eastman's) ; and cinnamic acid, crystallized twice.

had about two milliequivalents of unsaturation. Hydrocarbons and water-insoluble compounds were dissolved in carbon tetrachloride as follows: A sealed ampoule of the substance, having approximately 60 milliequivalents of unsaturation, is placed in a weighed, glassstoppered bottle of 150-ml. capacity and rather wide mouth. Some carbon tetrachloride is added, the ampoule is broken, and the bottle is filled with the solvent so that there is only a small air space, tightly stoppered, weighed, and shaken. From the weights of the two liquids, the density of the carbon tetrachloride at the temperature of pipetting and the approximate density of the solute, a volume of the solution, and likewise a concentration of the solute can be calculated, assuming additivity of volumes (perfect solution). By fitting the bottle with a two-hole rubber stopper carrying a separatory funnel (stem reaching to the bottom) and a tube through which a pipet can be inserted, a sample can be forced into the pipet by allowing mercury to flow from the separatory funnel. There should be a sniall air space between the liquid and the stcpper to prevent direct contact, but it should be small, to minimize errors due to dderence in volatility of solvent and solute. The pipet, with a three-way stopcock, is calibrated to contain a definite volume ( 3 ) . The stopcock is lubricated with a water-soluble grease. With a pipet holding 5.6 ml., and with the solution about 0.42 N in unsaturation, the sample delivered has about two milliequivalents of unsaturation. This procedure permits the removal of successive portions of the solution without any change in concentration.

Analytical Procedure

Preparation of Solutions for Analysis

The analytical procedure is based upon that described by Frieman, Kennedy, and Lucas (4).

Aqueous solutions of water-soluble compounds were made hy weighing sufficient material directly into a volumetric flask so that, when made to volume, the liquid was approximately 0.08 N in unsaturation and the sample taken, 25 ml.,

A calculated excess (10 to 15 per cent) of 0.1 N bromatebromide solution (about 25 ml.) is run from a buret into a 300ml. conical flask having a ground-glass stopper bearing a sealed-in stopcock. (The first analysis of the solution is approximate only, and should be carried out with a larger excess of bromate-bro-

MARCH 15, lY38

ANALYTICAL EDITION

141

mide solution. From this preliminary result the desired excess

can be calculated.) Folloxying the evacuation of the flafk by means of a n-ater aspirator, 5 ml. of 6 S sulfuric acid are added and the flask is allon-ed t o stand 2 to 3 minutes n-hile bromine is being liberated. Xext, 10 to 20 ml. of 0.2 iY mercuric sulfate and the solution to be analyzed, which should have about two milliequivalents of unsaturation, are run in. The volume of the carbon tetrachloride viash liquid should be about 15 ml. Follon.ing this, 20 nil. of glacial acetic acid are added. I n the case of n-ater-soluble substance.?, the acetic acid is omitted. After the flask, wrapped in a black cloth, has been shaken for about 7 minutes, 15 nil. of 2 Ai sodium chloride and 15 ml. of 20 per cent potasiuni iodide are added in succession, and the shaking is continud for 0.5 minute.

The vacuum is broken and the titration made with 0.05 .Y sodium thiosulfate, using starch. A blank, n-ithout the sample and n-it11 one-third of the amount of bromate-bromide solution used in the analysis, is run at the same time, and under the same conditions. The excess of bromate-bromide should not exceed 10 to 15 per cent; otherwise errors due to substitution may become appreciable. If the excess is less, the rate of addition may become so slow towards the last that the reaction is not quantitative in the time specified. The molar ratio of mercuric ion to final bromide ion should be greater than unity, otherwise the mercuric salt has insufficient catalytic effect. The presence of acetic acid greatly aids the reaction when carbon tetrachloride is present, no doubt because of the increased solubility of the unsaturated compound in the aqueous phase. The results are more reproducible in its presence, since the decrease in the time of bromination cuts down the amount of substitution. Sodium chloride is necessary, in order to liberate free bromine from its complex with mercuric sulfate. The iodine-starch end point, rather than the iodine-carbon tetrachloride end point, is used, because, in the presence of acetic acid, the solubility in the aqueous phase is so enhanced t h a t the iodine imparts but a faint pink color to the carbon tetrachloride while the aqueous phase is a deep yellow. Actually, this solubility relationship is an aid in the titration. The blank is run with only one-third, instead of the entire amount of bromate-bromide solution, in order better to approximate an average of the bromine concentration during a n analysis. The correction is usually about 0.5 per cent. CHAXGES OBSERVED DURING ASALYSES. The deep orange bromine color, which develops when sulfuric acid is added to the bromate-bromide solution, bleaches slightly as the mercuric sulfate is run in. Sometimes a precipitate, presumably mercuric bromide, forms a t the same time. dddition of the sample causes a furt'her bleaching, so that apparently little or no free bromine is present. The color deepens upon the addition of the sodium chloride. If one changes the order of addition in the case of the water-soluble compounds, propiolic, maleic, fumaric, and cinnamic acids, in the absence of both carbon tetrachloride and acetic acid, a sudden and complete bleaching occurs as soon as the molar ratio of mercuric ion to bromide ion reaches unity. This effect is explainable on the assumption that excess mercury is needed before the reaction is catalyzed appreciably. Under certain conditions, bromination of purely aliphatic alkynes is aided by the formation of a thick, viscous emulsion. The conditions are absence of acetic acid, a mercury-bromide ratio close to unity, and a bromate excess of about 5 per cent. VARIATIONSIX PROCEDURE. The importance of mercuric sulfate in the bromination is s h o w by comparing KO.3 with S o . 1 and No. 2 ,Table I, and also by comparing the results in Table 11, with and n-ithout mercury. From No. 1, Table I, it is evident that a large excess of mercury has no advantage over a small excess. I n general, the bromination of alkynes and alkenes proceeds quantitatively when the mercury-bromide ratio is greater than unity, and that of the alkynes is very unsatkfactorp in the absence of mercury. I n addition to the alkynes, the bromination of some other unsaturated compounds. which is unsatisfactory in the absence of mercury, is

-f fi

,

8 0 10 11

12 13 14 15

16 17 18 19 a

Phen>-lacetrleiie Phen~-Iacetglene l-Iie\->-ne I-HeYgne 1-Pentsne I-Pentyne I-Prntyne

.icetic .\?id 1 1 2; 1 1 ll 1 1 23 1 1 0 Excess Bromate 1 0 0

10 10 10

1

'0

J

23 0 12 1.7 0 a Time of Bromination I-Heptvne 1 5 1.5 1.5 I-Heptyne 1.5 15 15 I-Heptvne 1.5 1.5 15 Phenylacetglene 1.1 20 10 Plien~lacetylene 1.1 20 in Condition of Sample: Sunlight 5-Hepiyne 1.2 2n 15 2-Heptyne 1.2 2n 15 Condition of Sample: .iging 1-Heptpne 1.2 13 25 I-Heptyne 1,s 15 15 1.;

3

20 20 20 FO

30 15 3?

7 15 15

CC1: solution was exposed t o direct sunlight for 1 hour,

b Before exposure of same snluiion t o sunlight.

Sample of hydrocarhon was 3 ueeka old. d Hydrocarbon redistilled. C

made quantitative in its p r e s e n c e v i a . , dichloroethylene, maleic acid, and fumaric acid. On the other hand, cinnamic acid, which reacts satisfactorily in the absence of mercury, undergoes substitution when mercury is present. Dimethylbutadiene gives low results and propiolic acid gives very low results in the absence of mercury, while in its presence both give high results, the latter much the worse. I n more concentrated solutions, however, it has been found in this Iaboratory by Saul Winstein that dimethylbutadiene in carbon tetrachloride solution and in the absence of mercury gives fairly good results, but still low by 2 to 3 per cent. The effect of acetic acid is shown in Table I by comparing No. 4 with Xo. 5, and KO, 6 with No. 7. I n general, the reaction proceeds more rapidly and with greater reproducibility of results \Then acetic acid is present. An interesting case is that of dichloroethylene, which reacts rapidly in the presence of mercury when acetic acid is present and slowly when carbon tetrachloride is present, but scarcely reacts a t all in the absence of mercury and carbon tetrachloride. The effect of excess bromate is shown by Kos. 8, 9, and 10, Table I. These were among the first experiments. Since an excess was shown to be undesirable, subsequeut runs were carried out with only small excess. The effect of a large excess in the presence of acetic acid was not tested. Variation in the time, with the exception of dimethylbutadiene and propiolic and cinnamic acids, has little effect upon the results, provided the addition reaction is completed. I n general, 7 minutes are sufficient, when the general procedure is followed. However, the reaction with maleic and fumaric acids in water and in presence of mercury is so slow that a half hour should be allowed. Cinnamic acid dibromide is appreciably substituted in the presence of mercury. Freshly precipitated cinnamic acid reacts quantitatively with 1 mole of bromine in 3 minutes (Table 11), but with no more, even after 35 minutes. B u t when mercuric sulfate is added at the end of 4 minutes, a n additional 1/3 mole reacts within 1 minute. Increased substitution, with time, in the presence of mercury, takes place also Tyith dimethylbutadiene and with propiolic acid. I n the latter case the limit is one additional mole of bromine.

142

INDUSTRIAL AXD ENGINEERISG CHEMISTRY TABLE11. BROMISATIOX WITH

Suhstance

I-Pentyne 1-Hexyne 1-Heptyne I-Heptyne 2-Heptyne 2-Heptyne Phenyl acetylene Propiolic acid Cyclohexene 1-Hexene 2,3-Dimethylbutadiene Dichloroethylene

HpSOii Br -

Ratio H O h c CCIr

1.3 1.1 1.2

1.5 1.2 1.2 1.1 1.3

1.2 1.3 1.1 1.4 1.2

Time Taken Found X1. .Vi. .\fin. Milliequzcalents Mercuric Sulfate Present n 20 2 1.83 1 89 0 3 20 1 66.5 1 70 0 35 20 2.22 2 25 20 15 7 1.78 1 i6 0 20 2.05 1 9i 20 20 2.15 2 08 25 20 3.03 3 02 0 0 2 2.21f 3 26 15 20 3 2.275 2 285 15 3 20 2.95 2 9s

'0

;

15 0 20

20 20 0

5 70

1.60 1.73 1.97

Mixture of phenyl acetylene and 1.3 15 20 J cyclohexene 1.93 Maleic acid 1.3 0 0 25 l.i9 1.7 Fumaric acid 0 1.401 0 30 Cinnamic acid 1.2 0 1.35 0 a Mean of four results, average deviation 0.5. b Mean of two results, average deriation,'O.Z. 0 Alean of four results, arerage deviation, 1.3. d Mean of two results, average deviation, 0.5.

Error

% i2 . 2 0

+ 0.9:

-i-1 . 3

-

l.6d

- 4.0e - 3 5 - 0.3

i-56

++ 01 .4g0

1.97 l.i2 1.97

+20 - 0.5 0 0

1.96 1 i7 1,407 2 25

+ 1.6 ++-6 i01 .. 41

The bromine absorbed by a compound having a triple bond may run low if the substance has been exposed to air for Some time. Presumably, this is due to absorption of oxygen. The effect is intensified by exposure to sunlight (16 and 17, Table I). Distillation removes the material which is responsible for the low result (18 and 19, Table I). In Table 11are listed some typical results obtained by brominating in the presence and in the absence of mercuric sulfate. CONCLUSIONS. The procedure proposed constitutes a method of determining unsaturation in many unsaturated compounds, and is valuable for alkynes, which react but slowly in the absence of mercuric sulfate. It is important. that these alkynes be freshly distilled, and that their solutions in carbon tetrachloride be not exposed to direct sunlight. However, the fact that the mercuric salt catalyzes substitution in some cases, indicates that broad generalizations cannot be drawn. Instead it will be necessary to study each unsaturated compound separately.

Preparation of Materials Pentyne-1 was made according to the method of Bourguell (1) by refluxing 2,3-dibromopentane (from 2-pentene and bromine) with xylene and sodamide, the last prepared according to Vaughn, Vogt, and Nieuwland (18); boiling point 38-40' C., uncorrected. The method of Lebeau and Picon (6, IO) was followed in the preparation of 1-hexyne and 1-heptyne by reacting sodium acetylide with I-bromobutane and 1-bromopentane, respectively. Alkene present was largely removed along with the ammonia, by allowing the mixture to come to room temperature before decomposing the sodium alkynylide with ice. The temperature was kept low, in order to avoid the isomerizing effect of concentrated sodium hydroxide solution. Two fractionations through a 45-cm. (18-inch) column tf lass helices (IS)gave 1-hexyne of boiling point 69.7' to 70.1 uncorrected, yield 30 per cent. The 1-heptyne was fractionated twice as above, boiling point 98" to 100" C., uncorrected, yield 37 per cent, and later, in order to remove oxygen absorbed on standing, through a Vigreux column in an atmosphere of nitrogen; boiling point 97.5" to 97.9" C., uncorrected. For the preparation of 2-heptyne, sodium hexynylide was reacted with methyl sulfate following the general procedure of Meinert and Hurd (7). However, methyl sulfate is less satisfactory than the iodide, probably because it reacts more rapidly with ammonia. The inferiority of sulfates to bromides and iodides has been shown by Vaughn, Hennion, Vogt, and Nieuwland (11). After drying and fractionating twice through the column, the product, in 10 per cent yield, boiled at 109.3-109.9" C . Vaughn et al. (11) give 107-111" C. Phenylacetylene was synthesized by the method of Hessler (5). It was distilled through a Vigreux column in an atmosphere of nitrogen. The fraction distilling at 139.8-140.6° C. was used.

8.,

VOL. 10, NO. 3

ASD R-ITHOCT hIERCCRIC SULFATE HgSOi/ Br Ratio HOAc CClr Time Taken Found MI. M1. M i n . Miltiequivalents Mercuric Sulfate Absent ~. I-Hexyne 0 20 30 1.685 0.99 I-Hexpne ... 20 20 10 1.26 2.26 I-Heptyne ... 0 20 2 .21 1.34 30 1-Heptyne ... 15 20 0.94 1.78 2-Heptyne ... 0 20 10 1.84 1.35 2-Heptyne 20 20 2.05 1.60 Phenyl acetylene 20 3.03 20 2.71 Propiolic acid 0 0 15 2.211 0.38 Cyclohexene 20 3 0 2.27 2.275 I-Hexene 0 20 5 2.95 2.95 1-Hexene 15 20 3 2.94 2 95 2,3-Dimethylbutadiene 0 20 5 1.60 1.13 2,a-Dirnethylbutadiene ... 15 20 5 1.60 1.04 Dichloroethylene ... 0 0 100 1.12 1.73 0 ... 20 20 1.97 0 05 LIaleic acid 0 20 ... 0 1.79 0 03 Fumaric acid 1.40 0 0 10 0,015 Cinnamic acidh 3 0 0 1.355 1.40 6 Mean of three results, average deviation, 1.4. Z Amount determined by acidimetry. 0 Mean of t w o results, average deviation, 0.1. h As the sodium salt. Substance

;

Error

% -41

-44 -39 -50 -26 -23 -11

-78 - 0 2 0 0 - 0 3 -29

-35 -30 -9s -98 -99 3 :i

+

Propiolic acid was prepared from succinic acid through dibrOmosuccinic acid (9). The pure compound was not obtained; the distillate containing somewithout water, coming at 40" to 83" C. under 130 mm., was used furtherover purification. The propiolic acid was determined by titration against standard alkali. It was assumed that no other organic acid was present. Cyclohexene, 1-hexene, and 2,3-dimethylbutadiene were kindly supplied by Saul Winstein, of this laboratory. Their boiling points ere, respectively, 82,6-82.7~, 63,243,70, and 68.3-68.4' C., corrected. The first t5-o t\'ere distilled from sodium before use.

Summary I n the presence of mercuric sulfate several alkynes and alkenes were found to react quantitatively with bromine, two moles and one mole, respectively. These are 1-pentyne, 1hexyne, 1-heptyne, 2-heptyne1 phenylacetylene, 1-hexene, cyclohexene, and dichloroethylene. A general analytical procedure which is applicable to such compounds is described. In some cases, however, the reaction, although quantitative, may be slow-for example, maleic and fumaric acids. I n other cases substitution as well as addition may take p l a c e for example, propiolic acid, 2, 3-dimethylbutadiene, and cinnamic acid. I n the case of alkynes, the analysis should be carried out on freshly made solutions of recently distilled products, in order to avoid the errors due to absorbed oxygen.

Literature Cited

(3) (4) (5) (6)

(7) (8) (9) (10) (11) (12) (13)

Bourguell, Ann. Chim. [lo], 3, 191, 325 (1925); Compt. rend., 179, 686 (1934). Davis, Crandail, and Higbie, IND.E m . CHEM., Anal. Ed., 3, 108 (1931). Eberz and Lucas, J. Am. Chem. Soc., 56, 1232 (1934). Frieman, Kennedy, and Lucas, Ibid., 59, 722 (1937). Hessler, "Organic Syntheses," Collective Vol. 1, p. 428, New York, John Wiley & Sons, 1932. Lebeau and Picon, Compt. rend., 156, 1077 (1913). Meinert and Hurd, J . Am. Chem. Soc., 52, 4540 (1930). Mulliken and Wakeman, IND. ENQ.CHEM.,Anal. Ed., 7, 59 (1935). Perkin and Simonsen, J. Chem. Soc., 91, 834 (1907). Picon, Compt. rend., 169, 32 (1919). Vaughn, Hennion, Vogt, and Nieuwland, J . Org. Chem., 2, 1 (1937). Vaughn, 'Vogt, and Nieuwland, J. Am. Chem. Soc., 56, 2120 (1934). Kilson, Parker, and Lauphlin, Ibid., 55, 2795 (1933).

RECEIVEDJuly 8, 1937. Contribution from the Gates a n d Crellin Laboretories of Chemistry, California Institute of Technology, No. 609.