Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 30, 1897
Previous Article | Table of Contents | Next Article
This text is also available in PDF
(468 KB) Opens in new window

Art. LVIII.—On the Distillation Products of the Blackball Coal.

By Dr.W. P. Evans.

[Read before the Philosophical Institute of Canterbury, 3rd November, 1897.]

Part I.

The Blackball coal possesses physical characteristics agreeing in many respects with those of a cannel, and its relationship to the cannel family is also shown by the nature and amount of its distillation products. In some points, however, its behaviour is, as far as I am aware, unique.

The present paper deals with—(1) The distillation of the coal at a low temperature; (2) the subsequent fractionating of the crude tar so obtained; and (3) the distribution of sulphur between the main products.

(1.) The Distillation of the Coal.

As some quantity of tar was needed for the fractional distillation, a horizontal iron-tube retort (internal measurement, 40 cm. by 5 cm.) was used instead of the normal hard-glass retort (Schweelretorte), and a comparatively large charge of 200 grammes used at each operation. For this reason the results obtained represent far more closely those to be expected on the manufacturing scale. The head of the retort carried a diminisher and a T-piece, both of iron. One branch of the T was connected with a metallic condenser, with water-jacket, while the other allowed a thermometer to be fixed with its bulb directly in the path taken by the gases leaving the retort. The mouth of the retort was closed by an iron cap and stirrup-screw, and was rendered gastight by means of thin asbestos-paper washers.

The following table gives the results of the first eight experiments:—

– 488 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

No. of Experiment. Coal, in Grammes. Tar and Water, in Grammes. Coke, in Grammes. Gas and Loss (as Difference).
1 200 59 117 24
2 200 55·4* 117 27·6
3 200 57 116 27
4 200 59 113 28
5 200 56·5* 113·5 30
6 200 60 115 25
7 200 58 114 28
8 200 60·4 114 25·6
Totals 1,600 465·3 Consisting of 919·5 215·2
Tar. 314 Water. 151·3
Average percentage 100 19·62 9·45 57·46 13·45

It will be seen from above table that the yield of tar is remarkably high. Over twenty similar distillations have since been carried out, the maximum yield being 60·5 and the minimum 58 gr. of tar from 200 gr. of coal. The tar is almost entirely specifically lighter than water, but separates a smaller portion which is slightly heavier than water.

(2.) Fractional Distillation of Tar.

As the tar was nearly all lighter than water it was fractionated as a paraffin tar. The results show that its constitution is somewhat peculiar:—

Fraction 1 (below 250° C.). Fraction 2 (250°–300° C.). Fraction 3 (over 300° C.). Coke. Gas and Loss.
= Per Cent. of Tar. Specific Gravity at 20° C. Phenoles, &c., extracted by NaOH. =Per Cent. of Tar. Specific Gravity at 16·8° C. Freezing-point. = Per Cent. of Tar. Specific gravity at 48° C. Freezing-point. = Per Cent. of Tar. = Per Cent. of Tar.
24·7 0·881 19 per cent. vol. 15·2 0·967 Below 0° C. 54·9 0·991 23° C. 2·9 2·3

Specific gravity of tar at 25° C., 0·977, Correction for small differences, = ± 0·001 per 1° C.

[Footnote] * In these experiments (Nos. 2 and 5) leakage occurred at the mouth of the retort.

– 489 –

A second smaller quantity worked up to pitch at 400° C., when useful paraffin oils were still coming off, gave—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Total distillate = 81·86 per cent.
Pitch = 14·92 "
Gas and loss = 3·22* "

The tar is evidently an extremely useful one, giving clean distillates and an upper fraction which appears to be extraordinarily rich in valuable paraffin. The middle and upper fractions should furnish an excellent gas-oil.

(3.) Distribution of Sulphur.

This point seemed of especial interest, owing to the high percentage of sulphur contained in the coal. The results obtained are very striking:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Percentage of sulphur in coal (Eschka method) = 4·63
Percentage of sulphur in crude tar (Carius method) = 2·48
Percentage of sulphur in coke (of coal) = 3·14
∴ Percentage of sulphur originally in coal remaining in the tar = 10·5 (10·49)
Percentage of sulphur originally in coal remaining in the coke = 39 (38·96)
Percentage of sulphur originally in coal escaping during distillation = 50·5!

These figures point undoubtedly to the fact that the coal is far better suited for distillation at low temperature as here described (preparation of solvent oils, gas-oils, paraffins, &c.) than for use as a gas-coal at high temperature. At high temperatures the greater part of the sulphur is given off as carbon-bisulphide and escapes more or less the action of the purifiers, unless these are very carefully worked, while at a low temperature the sulphur, as direct tests showed, comes off almost entirely as sulphuretted hydrogen, and may be easily dealt with.

At 190° C., by thermometer in neck of retort, the issuing gas was so highly charged with SH2 as to be of direct use in the laboratory!

Further experiments are now being carried out to determine the amount of volatile matter distilling at certain maximum temperatures; the nature of the gases given off; the composition of the various tar-fractions; and the effect of injecting superheated steam into the retort during the process of distillation.

[Footnote] * Caused, in part, by flask cracking at 370° C.

– 490 –

Part II.

The following table gives the results of the next ten distillations similar to those numbered 1–8 in the previous table. It will be seen that, though the yield of tar and water together is almost the same, a better method of separation raised the tar alone from 19·62 to 21·00 per cent., the water at same time sinking from 9·45 to 8·25 per cent.:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Distillation of Coal at Low Temperatures.—Second Series.
No. of Experiment. Charge of Coal, in Grammes. Tar and Water, in Grammes. Coke, in Grammes. Gas and Loss (as Difference).
9 200 57 115 28
10* 200 59 115 26
11 200 60·2 114 25·8
12† 200 58 114 28
13‡ 200 60·5 113 26·5
14 200 58 115 27
15§ 200 58·3 114·5 27·2
16 200 58 114·5 27·5
17∥ 200 57 115·5 27·5
18 200 59 114·5 26·5
Totals 2,000 585 Consisting of 1145 270
Tar. 420 Water. 165
Average percentage 100 21·00 8·25 57·25 13·50

The tar, when fractionated (light and heavy portions mixed in natural ratio), gave—

1. Fraction under 250° C = 23·20 per cent.
2. " 250° C. to 300° C. = 20·23 "
3. " over 300° C. = 48·36 "
Pitch (beginning to coke) = 7·06 "
Gas and loss = 1·14 "

That is, the addition of the heavy tar increased the upper fractions, as was to be expected.

[Footnote] * At 180° C., by thermometer in neck, 46 gr. had collected; at 220° C., 53 gr.; and at 250° C., 59 gr. No further appreciable yield of tar on heating to redness for some time.

[Footnote] † At 220° C. 53 gr. had collected.

[Footnote] ‡ Between 190° C. and 210° C. the gas given off consisted mainly of SH2.

[Footnote] § Slight leak at neck owing to stoppage in condenser. After 57 gr. had been obtained, retort kept at full heat (moderate red) for two hours. Strong evolution of ammonia, but only 1·3 gr. more tar.

[Footnote] ∥ Duration of distillation, 10·¼ hours. First nine hours between 166° C. and 245° C.; then raised to 330° C.

[Footnote] ¶ Duration of distillation, five hours. First four hours at 190°–220° C; then to 350° C.

– 491 –

Composition of Tar-fractions.

Fraction 1; Below 250° C.

Fraction 1 from the first series (light tar only) was washed, as usual, with caustic soda, and fractionated after neutralisation, while fraction 1 from the second series (light and heavy tars together) was fractionated without any previous treatment.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Fractionation of Crude-tar Fraction 1 (washed, and containing no Heavy Tar).
Number of Fraction. Temperature in Deg. C., Minimum to Maximum. Percentage of Whole Fraction. Colour.
1 Below 120 5·35 Light-yellow.
2 120–160 31·16 "
3 160–200 31.27 "
4 200–250 23·53 Brownish-yellow.
5 Over 250 8·29 Dark-brown.
6 Loss during distillation 0·40

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Fractionation of Crude-tar Fraction 1 (unwashed, and containing Heavy Tar).
Number of Fraction. Temperature in Deg. C., Minimum to Maximum. Percentage of Whole Fraction. Colour, Odour, &c.
1 80 3·79 White, somewhat milky; odour normal. Amount under 80°, very small. (So-called 90 per cent. benzole.)
2 80–100
3 100–120 1·21 White; odour normal. (Socalled toluole.)
4 120–160 14·65 White; odour normal. (Socalled solvent-naphtha.)
5 160–200 29·98 Light straw-yellow; odour scarcely perceptible. (Lighting oil.)
6 200–250 36·30 Brownish-yellow; odour decided. (Lighting oil.)
7 Residue 13·84 Dark-brown; odour strong. (To be carried on to fraction 2.)
8 Loss during distillation 0·23 Fluidity perfect in all cases.

The first fractions are small, the total quantity under 120° C. being only 5 per cent, of the main fraction 1, or 1·16 per cent, of the crude tar. The whole distillate from the tar could be used as a gas-oil without previous fractionation, if such a course were desirable. This, however, would probably be a wasteful method of treating the upper fractions.

– 492 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Fractionation of Crude-tar Fraction 2 (250° C. to 300° C.).
Number of Fraction. Temperature in Deg. C., Minimum to Maximum. Percentage of Whole Fraction. Colour, &c.
1 250 32·75 Clear brown-yellow.
2 250–275 34·75 "
3 275–300 20·33 Clear brown.
4 Residue 11·80 Solid at ordinary temperature; to go to next fraction.
5 Loss during distillation 0·37

Left in melting ice for several hours fractions 1 and 2 remained perfectly fluid and clear; fraction 3, however, became buttery, owing to the separation of crystalline paraffin.

Estimation of Paraffin in Tar-fraction 3 (over 300° C.).

N.B.—Owing to want of proper material the following results can only be looked upon as roughly approximate. The laboratory results are here also as a rule not so close to the results obtained on the manufacturing scale as in other cases.

For hard paraffin the fraction was filtered (under 0·7 atmosphere) at 10° C. The well-defined crystalline mass formed 23 per cent. of the total fraction.

For remaining soft paraffin the filtrate was treated by the ether-alcohol precipitation method, and gave another 14 per cent, of paraffin (vaselines), with only slight evidence of crystallization.

An attempt was made to filter at a low temperature, but even at 0° C. the whole fraction became too solid to filter except under high pressure.

Part III.

Distillation at Low Temperatures by Aid of Superheated Steam.

Apparatus.—The retort (f)* used in the first series of experiments was adapted to distillation in a current of steam by brazing a 5 mm. pipe through the cap (d). This pipe (e) extended right along the floor of the retort, and carried twenty steam-jets. Its outer end was connected by a brass unionjoint to a thermometer-box (c) and the lower end of the superheating coil (b). The upper end of the coil was in its turn connected directly with the glass flask (a) in which steam

[Footnote] * See accompanying diagram, which shows apparatus as set up for use.

– 494 –

was generated. The thermometer (g) and metal condenser (h) at the neck of the retort were again used, but the condenser was supplemented by a second longer one (k) of glass. Less than 1 gramme of distillate was got from this second condenser.

The retort was filled in an inverted position, the cap (with steam-pipe now, of course, at the top) screwed firmly home, and the retort then gently turned through 180° into the right position for work. The temperature of the steam could, by means of a Fletcher's Argand-Bunsen, be easily raised to over 400° C, and could be kept fairly uniform for a number of hours.

The charge of coal used was, as before, 200 grammes, the temperatures of the steam and the retort, as well as the time of distillation, being varied considerably.

The following table gives the results of six typical experiments.

Experiments 7 and 8 of the table are types of the check experiments carried out as a help towards estimating the quantities given in column V.

Column VI. gives the minimum yield on the supposition that no steam reacts with the carbon.

No 4 experiment probably represents the effective working conditions for the coal—viz.: (1) Fairly quick distillation; (2) temperature of gases not rising much above 250° C.; (3) injected water nearly equal to the weight of the coal, and so forming about 75 per cent, of the total distillate.

The coal is evidently eminently suitable for distillation purposes.

– 493 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Number of Experiment. I. II. III. IV. V. VI. VII. VIII. Conditions of Distillation.
Charge of Coal. in Grammes. Coke. in Grammes. Total Liquid Distillate. in Grammes. Water injected as Steam. in Grammes. Estimated Number Grammes Water in Combination. Minimum True Distillate = (III.–IV.). Estimated True Distillate = (III.–IV. + V.). Gas and Loss, estimated = (I.–II.–VII.). Temperature of Injected Steam in Deg. C. Temperature of Retort-neck in Deg. C. Duration. in Hours. of Distillation. Water injected per Hour as Percentage of Charge.
Initial. Final. Initial. Final.
1 200 111 312 255 7 57 64 25 140 220 180 320 5·¾ 22·15
2 200 123 271 214 5 57 62 15 170 190 190 230 5·¼ 20·3
3 200 133·5 242 187 5 55 60 6·5 110 160 200 220 3·¾ 24·9
4 200 114 272·5 209 4·5 63·5 68 18 130 320 195 320 3 34·7
5 200 119·5 349 290 6 59 65 15·5 100 260 200 260 3·½ 41·5
6 200 124·5 387 334 8·5 53 61·5 14 180 360* 200 280 2·⅓ 71·5
Actual Loss.
7 Sides of the retort lined with coke. 110 115 5 about 280°. Passed at average rate.
8 92·6 95 2·4 about 230°

[Footnote] * Note.–Tar actually separated from the total distillate of Nos. 1–6 weighed 282 grammes, giving an average yield of 23·5 per cent.

Previous Article | Table of Contents | Next Article