GB2069524A - Pyrolysis of coal - Google Patents

Abstract

High yields of tar can be obtained from bituminous coals by pyrolysing the coal in a fluidised bed at 527-627 DEG C and 0.1 to 5 atmospheres with a gas residence time of 1 to 20 seconds and a coal/char residence time of 1 to 30 minutes. The bed is fluidised with a gas containing at least 80% hydrogen or an equivalent gas such as steam. The tar evolved can be substantially completely hydrogenated and lacks intractable polymeric or oligomeric material.

Description

SPECIFICATION Pyrolysis process This invention concerns a process for the pyrolysis of coal, especially in fluidised beds.

It is well-known that tars are produced by pyrolising coals. Old techniques which were primarily intented to yield coke or coal gas, gave low yields, about 5%, of a heavy intractable tar.

More recently, there has been considerable published research work on so-called "flash pyrolysis" involving very rapid heating of coal to maximise the tar yields, For example, American research workers have obtained up to 15% benzene in laboratory apparatus and 7.5% in pilot plants by hydropyrolysis of low-rank coals in hydrogen under pressure. The process calls for injection of coal into hydrogen at 100 bar and very rapid heating and cooling to give flash heating of the coal and decomposition of the tars to benzene.

Some very sophisticated apparatuses have been constructed in the search for increased liquid yields from flash pyrolysis but we believe that serious practical difficulties lie in the way of achieving yields in commercial size plant which are close to those achieved in the laboratory.

It is desirable to achieve high tar yields from coal using essentially available technology and avoiding high pressure apparatus into which solids are fed.

R. J. Tyler of CSIRO of Australia, published results of flash pyrolysis of coal in a fluidised bed reactor (Fuel, 1979, 58,680), in which a brown coal was fed into a sand bed fluidised with nitrogen, given a maximum tar yield of 23% w/w daf coal. The reaction conditions are specified in the paper and include pyrolysis temperatures of 606 and 6970C at atmospheric pressure and gas residence time of about 0.3 s. Hydrogen was also use as the fluidising medium, but little change in the quantites of tar or total volatile matter were found, although there was an increase in hydrocarbon gas yields.

The present invention provides a process for the production of high yields of tar from bituminous coal, comprising pyrolysing bituminous coal in a bed fluidised with at least 80% hydrogen or a gas equivalent thereto, at a temperature of from 800 to 900 K and at substantially atmospheric pressure, the gas residence time being in the range 1 to 20 seconds and the coal/char residence time being in the range 1 to 30 minutes.

This invention arose from studies of the process is carried out at "substantially atmospheric pressure with nitrogen and with hydrogen. The use of atmospheric pressure permits presently available technology to be used. Although the process is carried out at substantially atmospheric pressure", this is to be understood as permitting the use of pressures from 0.1 up to 5 atmospheres.

The bed used in the process may be a bed of inerts such as sand, ceramic spheres or refractory particles, or may be char particles from the process itself. Sand and the like have the advantage that the surface area is lower than that of char, and therefore cracking is reduced.

The fluidising gas is a gas containing at least 80% hydrogen, other possible components being one or more of CO, CO2, steam and methane.

Alternatively, the gas may be one equivalent to one containing at least 80% hydrogen; for example a gas containing or consisting of steam may be found to be equivalent to the hydrogen containing gas because of the production of mascent hydrogen by reaction with the char in the bed, and a stripping effect of tar from the surface of char particles.

Since the process is endothermic, heat must be supplied to the bed. This may be achieved by preheating the fluidising gas, and it is convenient to burn char from the process to provide this heat, or the bed may be heated by heat exchange from tubes or surfaces around or in the bed, or by inductive heating etc.

Bed temperatures are preferably in the range 820 to 870 K to achieve maximum tar yields, more preferably in the range 840 to 860 K.

We have found that tar yield in the process is inversely related to the vapour residence time at the pyrolysis temperature. On the other hand, extremely low residence times of less than one second, or very appreciably less than one second as has been suggested in the art, are not found to be necessary or desirable.

The bituminous coal is preferably a low rank high volatile coal having a volatile matter content in the range 30 to 45% daf coal. Suitable coals are those of 600 to 800 CRC (Coal Rank Code, as defined in "The Coal Classification System used by the National Coal Board [Revision of 1964]", published by National Coal Board, London.

Although the coal has to be fairly finely divided, that is, less than the maximum particle size for fluidisation, the effect of particle size on the process is complex. It is believed that larger particle sizes lead to increased vapour residence time within the particles and hence more extensive cracking of the tar. Conveniently, pulverised suitable particle fuel grade may be used, that is 85% or more is - 200 B.S. mesh.

The coal need not be dried, providing that it does not agglomerate during processing.

The product tars are carried out of the fluidised bed reactor as vapours by the fluidising gases, and may be condensed out in known manner. It may be desirable, especially with residence times towards the lower end of the range, to quench the vapours using a hydrogen donor. Hyrdogen donors are known to coal chemists and include hydroaromatics such as hydrogenated tar oils and tetralin. Preferably, the product tars are cooled, whether using a quench or not, to below 4000 C, more preferably to below 2000C.

The condensed tars are obtained in yields, under preferred conditions, of generally in excess of 20% by weight and generally greater than the Gray-King tar yields by 50-60%. Contrary to the disclosure of the Australian work in the Fuel paper, the use of hydrogen for fluidisation increases yields by about 4 percentage points over that when using nitrogen. The product tars are highly amenable to hydrogenation in manner known per se, for example by hydrotreatment using a NiCo or MoCo catalyst, yielding liquid fuels and chemicai feed stocks.Certain prior proposed processes in which coal is pyrolysed in a bed fluidised by inert gas or by mixtures of gases containing a relatively small proportion of hydrogen, yield tars of which a significant amount, for example in excess of 25%, is very difficult to reduce with catalysts and hydrogen under pressure.It is beiieved that the advantages of the present invention in high yield and useful product arise from the presence of hydrogen which "protects" the initial tar components which are in highly reactive free radIcal form. in the absence of hydrogen, it is believed that the tar components rapidly polymerise to give intractable material, yet the hydrogen consumption is ow indication that there is a stebilisation-type hydrogenation rather than any substantial hydrogenation significantly affecting the hydrogen content of the tar. Studies indicate that the mai" difference is in the sidechains, reducing the tendency to form free radicals.

The char remaining after pyrolysis, which is removed from the bed in known manner, e.g. over a -;jeir or pipe, is a good feedstock for gasification or for fluidised bed combustion . A proportion of the char will be consumed in practice to provide process heat and hydrogen, and the gases produced. largely methane and ethane, can also be consumed on site.

Ths invention will now be described by way of e; - ml ,ie.

EXAMPLE Coal from Linby mins a low rank high volatile bituminuous coat, CRC 802) was ground and sieved into two fractions + 180-355 m and + 355 700 m The aground coal was charged into a hopper attched to a 6 in ( 5.24 cm) internal diameter fluidised Led which was heated externally an an electric es-stance heater. A bed of sand on ceramic spheres was fluidised by gas fed from the bottom of the reactor and coal was fed into the bed at a rate of approximately 1.5kg/hr, after the reactor had reached thermal equi!ibrium at the desired reaction temperature.

Liquid products and water were collected in a series of three traps immersed in an ice-water bath. The gases emerging from the traps were passed through an electrostatic precipitator before being passed to a gas sampling section and a gas meter as samples were taken 5 minutes from start, 10 minutes from start and at 10 minutes intervals thereafter for analysis by gas chromatography.

The gas used for fluidisation was pure hydrogen or nitrogen at approximately atmospheric pressure. The maximum yield of tar in a hydrogen fluidised bed is approximately 24%, 5.5% higher than in nitrogen, at approximately 640 K.

However, it isbelieved that tar production is influenced by competing reactions, evolution of tar and cracking of tar, both increasing with increasing temperature. The hydrogen is though to exert a protective effect onthe tar yields, reducing the cracking (and opolimerisation) of tar and slightly increasing the temperature at which maximum yield is obtained, compared to nitrogen.

The Gray-King assay yield of tar with heating at 5 K min 1 to 873 K of a static bed of the Linby coal was only 13.3% @@@@ daf coal; the process of the invention using hydrogen achieves an 80% improvement while yielding a high quality tar suitable for further processing into synthetic fuels and chemical feddstockes.

Gas yields were found to increase generally with temperature du@to increased devo@illisation of coal and cracking of tar.

Variation in gas residinos time at the reaction temperature were studied and it was found that the extent of cracking and @erei@rs of yields is affected by residence time was found to favour tar at the exponse of gas in both hydrogen and nitrogen. Extremely low residence times as used by many researcheers were not studied, however, as it was considered that ths difficulties of attaining such low residence time in any practicable equipment outweighed the possible increases in tar yield.

Although only two particle sizes for the feed coal were used, and although the effect of particle size is complex, in general it was found that the smaller particle size gave increased yield of tar, by up to 2% in hydrogen.

Claims (8)

1. A process for the production of high yields of tar from bituminous coal, comprising pyrolysing bituminous coal in a bed fluidsed with at least 80% hydrogen or a gas equivalent thereto, at a temperature of from 890 to 890.. and at substantially @@m@epine@ie @@@@@@@ the gas residence time being in the range 1 to 20 seconds and the coal/char residence time being in the range 1 to 30 minutes.

2. A process according to claim 1, in which the bed comprises sand, ceramic spheres or refractory particles.

3. A process according to claim 1 or 2, in which the bed temperature is in the range 820 to 870 K.

4. A process according to claim 3, in which the bed temperature is in the range 840 to 830 @

5. A process according to any @@@ of the preceding claims, in which ths @@ minous @@@ @@ a low rank coal having a volatils matter content in the range 30 to 45% daf coal.

6. A process according to any one of the preceding claims, in which the vapours from the bed are quenched using a hydrogen donor.

7. A process according to claim 1, substantially as hereinbefore described.

8. A process according to any one of the preceding claims, comprising also the step of hydrogenation of the tar evolved.