Checked by Frank C. Whitmore and Thos E. Hollingshead.
1. Procedure
A
5-l. flask, mounted on a
steam bath, is fitted with a wide (Note
1) inlet tube reaching to the bottom of the flask, a
reflux condenser, and a
delivery tube running from the top of the reflux condenser and reaching to the bottom of a
heavy glass cylinder containing a column of mercury about 10 cm. high. All stoppers and joints of the apparatus must fit tightly and be wired, as the reaction is carried out under a slight pressure. In the flask are placed
3180 g. (3.7 l., 30 moles) of xylene (Note
2) and
1 kg. of anhydrous aluminum chloride (Note
2). The inlet tube of the flask is then connected to a
tank of methyl chloride or to a methyl chloride generator (Note
3), the steam is turned on in the bath, and a fairly rapid stream of
methyl chloride is passed into the mixture (Note
4). At first there will be a rapid absorption, and the stream of gas must be regulated so that the
mercury in the pressure tube does not suck back. The
hydrogen chloride formed may be conveniently disposed of by a trap. As the reaction slows down, the pressure increases until both
hydrogen chloride and
methyl chloride begin to escape through the
mercury. At this point the current of incoming gas should be cut down so that undue loss of
methyl chloride is avoided. About one hundred hours is required for the completion of the reaction (Note
5).
When the reaction is completed, the steam is turned off and the flask is disconnected and loosely stoppered. After standing overnight the reaction mixture is decomposed by pouring it slowly onto 5 kg. of chopped ice. The greenish oil which separates from the aqueous layer is removed, dried over
calcium chloride, filtered and fractionated twice, using a
round-bottomed flask fitted with an efficient column and an air condenser (Note
6).
The fractions are cut as follows:
|
Fraction |
Temp. Range |
Distillate |
Yield, g. |
|
I and II |
Up to 150° |
Benzene, xylene |
Little |
III |
150 to 180 |
Trimethyl benzenes |
570 |
IV |
180 to 205 |
Tetramethyl benzenes |
2075 |
V |
Above 205 |
Mostly pentamethylbenzene |
815 |
|
Typical yields of the various fractions are given in the table; they may vary 10 to 20 per cent from these figures, owing to slight differences in procedure and materials, and especially to differences in the quality of the
aluminum chloride employed. The more efficient the
aluminum chloride, the greater is the percentage of the higher-boiling fractions.
The tetramethylbenzene fraction (IV) is rich in
durene, which may be frozen out and filtered because of its relatively high melting point (
80°). To isolate the
durene, fraction IV is thoroughly chilled in an
ice-calcium chloride pack, and filtered through a cold filter (Note
7), using good suction and pressing down the solid compactly. When no more liquid drips through the cold filter, the filtration apparatus is allowed to come gradually to room temperature (Note
7) and the suction is continued as long as any liquid drips through, after which the solid is removed and bottled. The yield is
540 to 610 g.
Fraction III may be methylated to the
tetramethylbenzene stage by heating on the steam bath with
100 g. of anhydrous aluminum chloride and passing in
225 g. of methyl chloride. The filtrates from the
durene likewise yield more
durene when heated on the steam bath with
50 g. of aluminum chloride. The products are worked up in the usual way; that is, they are decomposed by pouring onto twice their weight of chopped ice, separated from the aqueous layer, fractionated twice, and the
durene frozen out as before. By conducting one methylation of the
trimethylbenzene fraction, and one treatment of the
durene filtrates, the combined yield of crude
durene will average
1 to 1.4 kg. (
25–35 per cent of the theoretical amount based on the original
30 moles of xylene).
To purify the
durene, 200 g. of the crude product is placed in a
1-l. round-bottomed flask fitted with a reflux condenser, and melted in a
water bath at 95°;
200 cc. of warm (50°) 95 per cent ethyl alcohol is then added through the top of the condenser, and the mixture is carefully heated until homogeneous. The solution is filtered on a
hot-water funnel, allowed to stand tightly covered (Note
8) in a fairly warm place (35°) overnight, cooled to about 0°, and filtered on a
suction filter. The product (about
169 g.) melts at
74–78°. A second recrystallization yields
149 g. melting at
77–79°. A third recrystallization yields
140 g. having a melting point of
79–80°. The alcoholic filtrates are fractionated, and the crude
durene obtained is worked over with the isomers of
durene.
The
pentamethylbenzene obtained in this way is nearly pure, and one recrystallization from
95 per cent alcohol or from a mixture of equal volumes of alcohol and
benzene gives a snow-white product, but the product generally melts over too wide a range for practical purposes. However, if fraction VII is refractionated under diminished pressure and the fraction boiling at
123–133°/22 mm. (practically all at
127–129°) is collected and recrystallized as in the following paragraph, a product melting quite sharply at
52° (true m.p.
53°) is obtained.
Six hundred grams of crude pentamethylbenzene is heated to 100° and poured slowly with stirring into
1 l. of 95 per cent ethyl alcohol heated to 70° in a
2-l. beaker, and allowed to stand overnight at a temperature of approximately 30°. The crystals formed are collected on a suction filter and dried at room temperature overnight on a
porous plate. The yield is about
250 g. (Note
9).
Hexamethylbenzene.—Fraction VIII is fractionated in 250-cc. batches in a
Claisen flask at 20 mm. pressure, the following fractions being collected:
This fractionation may be carried out in an
ordinary Claisen flask, but there is some difficulty in maintaining the desired pressure owing to the solvent action of the hydrocarbons on the
rubber stopper. This difficulty may be avoided by the use of a
Claisen flask with very long necks and a wide side tube. The material should be distilled fast enough to prevent it from solidifying in the column and side tube. Prolonged heating of
hexamethylbenzene also causes a considerable amount of decomposition to tars.
The method for the production of large amounts of
hexamethylbenzene is the rapid methylation of
pentamethylbenzene or the
durene filtrates. A mixture of
378 g. of pentamethylbenzene and
200 g. of anhydrous aluminum chloride is heated on an
oil bath at 190–200° and a rapid stream of dry
methyl chloride is bubbled through for three to four hours, using the same apparatus as for the preparation of
durene. The mixture is allowed to stand overnight at room temperature.
One liter of hot xylene is added to dissolve the solidified material, and the reaction mixture is decomposed by pouring it onto 3 kg. of chopped ice. The resulting oil is separated, and the
xylene and other low-boiling material are removed by distillation under reduced pressure (Note
10). The fractions are divided as above. Two refractionations and two recrystallizations (Note
11) give
98–121 g. of white crystals, melting at
157–161° (Note
12).
2. Notes
1. The inner tube of a condenser makes a good inlet tube. It should be placed so that the wide end is inside the large flask.
2. The
xylene should be a good, colorless laboratory grade, b.p.
135–140°. Any moisture present may be removed by distilling and discarding the first 10 per cent of the distillate. The best
aluminum chloride available should be used, for the methylation is very unsatisfactory if the catalyst is of an inferior grade. It should be in small pieces but need not be powdered.
3. The methyl chloride generator consists of a 5-l. flask resting on a sand bath and fitted with a reflux condenser, with a delivery tube running from the top of the condenser to a train of wash bottles, two containing water and two containing concentrated
sulfuric acid, with three safety bottles, one at each end of the train, and one between the water and sulfuric acid bottles. To charge for about
45 moles (theoretical) of methyl chloride: 200 g. of water and
2.2 kg. (1.2 l.) of concentrated sulfuric acid are placed in the flask, and
1.4 kg. (1760 cc.) of methyl alcohol is added, with cooling, at such a rate that the temperature does not rise above 70°. Then
2.4 kg. of sodium chloride is added, the apparatus is tightly connected, and the flask is heated on the sand bath so that the gas is evolved at a fairly rapid rate. It has been found in practice that, using materials of the commercial grade, the yield of
methyl chloride is about
55 to 65 per cent of the theoretical amount, so that about double the calculated quantities must be used. This means that the generator has to be charged three times in order to convert
30 moles of xylene to
tetramethylbenzene. If a
tank of methyl chloride is available,
65–70 moles of methyl chloride should be used for this same amount of
xylene. The tank should be weighed before starting and the reaction stopped when the tank has lost the proper amount in weight.
4. Experiments have shown that the rapid current of
methyl chloride furnishes sufficient stirring.
5. The normal time of one hundred hours can be shortened by increasing the amount of
aluminum chloride. The product is then very viscous, however, and rather difficult to handle in large amounts.
6. Two systematic fractionations (not redistillations) with a good column are absolutely necessary in order to obtain good separations. The more efficient the column the better.
7. The material should be filtered through a
large Büchner funnel, which is immersed in a freezing mixture as long as any liquid drips through. It is stated in the literature that the first filtrate obtained in this way is mostly
isodurene (1,2,3,5-), whereas the second filtrate, obtained as the material warms slowly to room temperature, is
pseudodurene or prehnitene (1,2,3,4-), m.p.
−4°.
8.
Durene, being quite volatile, should not be allowed to remain exposed to the air any longer than necessary. It is also quite volatile with alcohol, and the mother liquors resulting from the recrystallizations should be distilled: the alcoholic distillate is used for further recrystallizations, and the residues may be worked up for
durene by heating with
aluminum chloride.
9. The melting point of
pentamethylbenzene is only slightly affected by recrystallization, because most of the impurity is
hexamethylbenzene, which can be removed only by fractionation.
10.
Hexamethylbenzene decomposes when heated very strongly for any length of time. Therefore better results are obtained if the distillations are carried out under reduced pressure.
11. Small amounts of impurities greatly influence the melting point of
hexamethylbenzene, and several recrystallizations of a fraction of close boiling range are necessary in order to prepare a sharply melting product.
12. Small amounts
(25 g. or less) of hexamethylbenzene which is nearly pure are best recrystallized from
ethyl alcohol. It requires about
600 cc. of boiling alcohol to dissolve 25 g., but on cooling
20 g. of pure product will result.
Ether and
benzene dissolve the substance much more readily, and larger amounts of materials are best recrystallized from either of these solvents, or from a mixture of one of them with alcohol.
One hundred and twenty-five grams of the hexamethylbenzene distillate which has been refractionated is melted and poured slowly with stirring into
1.5 l. of 95 per cent alcohol. A small amount which remains undissolved may be brought into solution by adding about
300 cc. of hot benzene, the beaker being heated on a steam bath and the mixture stirred constantly until all is dissolved. The solution is allowed to stand overnight at approximately 25°. The crystals are filtered by suction and washed with enough
95 per cent alcohol to moisten thoroughly (about 25 cc.). After drying, the crystals weigh approximately
112 g. and melt at
155–159°.
3. Discussion
Copyright © 1921-2002, Organic Syntheses, Inc. All Rights Reserved