The Technology of Low Temperature Carbonization

Frank M. Gentry

[ Chapter 2: Low Temperature Coal Gas ]

Table of Contents

Preface & Table of Contents
Chapter I ~ Fundamentals
Chapter II ~ Low Temperature Coal Gas
Chapter III ~ Low Temperature Coal Tar
Chapter IV ~ Low Temperature Coke
Chapter V ~ Nitrogenous & Other By-Products
Chapter VI ~ Processes of Low Temperature Carbonization
Chapter VII ~ Operation, Design, & Materials of Construction
Chapter VIII ~ Economics & Conclusion
Name Index
Subject Index

[ Note: The quality of the scanned graphics (tables & figures) are "uneven" at best despite repeated efforts to scan and tweak the images. -- R.N. ]

Chapter II
Low Temperature Coal Gas

Evolution of Gas
American Coals
British Coals
Peat & Lignite
Time Effect
Vacuum Effect
Secondary Decomposition
Gaseous Atmospheres
Physico-Chemical Equilibrium

Evolution of Gas ~

It is worthwhile to examine the behavior of the lower types of carbonaceous fuels under destructive distillation as a preliminary to the discussion of the gaseous products evolved from coal under the influence of temperature. Bornstein (95) carefully examined the composition of gas evolved from distillation at various temperatures of a peat and a lignite, whose ash-free analyses approximated those shown in Table 2, and a wood whose ultimate analysis showed 4.8% carbon, 6.5% hydrogen, 40.8% oxygen, 0.63% nitrogen, 3.2% water, 0.10% ash, and 0.08% sulfur. These data have been compiled in Table 13. It should be observed that, in the case of wood, hydrogen is rapidly evolved at comparatively low temperatures and, as a matter of fact, it will be shown that the temperature of hydrogen evolution progresses with the geologic aging of the fuel. The percentage of carbon dioxide increases steadily, at any given temperature, with the position of the fuel in historical reckoning.

Lewes (17) carbonized bituminous coal at temperatures ranging from 400 C to 900 C and measured the volume of gas liberated at the respective temperatures. He found that the volume of gas liberated increased approximately 1,200 cubic feet per gross ton of coal for each 100 C rise in temperature throughout the range. A glance at Table 14 will show that, within the temperature interval 400 C to 900 C, the volume of gas liberated increases over 100% from the lower to the higher temperature. At the same time, the percentage composition of the gas undergoes a considerable change. The percentage of saturated and of unsaturated hydrocarbons each decreases about half. The amount of unsaturated hydrocarbons in the gas is small, compared to the other constituents.

American Coals ~

Studies of the gas obtained in the destructive distillation of coals found in the United States have been made by a number of experimenters. Particular mention should be made of the work of Porter and Ovitz (96), who used coals from Pennsylvania, Illinois, West Virginia, and Wyoming; Davis and Parry 97) who examined coal from Pittsburgh and Upper Kittanning beds of Pennsylvania; and Taylor and Porter (98), who investigated the coals of Pennsylvania, Illinois, West Virginia, and Wyoming. Representative analysis of the coals studied are given in Table 15 and Table 16. Comparison of these data with Table 1 and Table 2 seems to classify the West Virginia and Virginia coals as semi-bituminous, and the Illinois, Utah and Wyoming coals as sub-bituminous.

Tables 17, 18 and 19 give the quantity of gas evolved and its composition at three different temperatures of distillation for Virginia, West Virginia and Wyoming coals, respectively. In these three tables the Virginia coal is the oldest coal, the Wyoming coal the youngest, while the West Virginia coal occupies an intervening position historically. As a matter of fact, we shall see later that the geologic age of the coal has a direct bearing on the composition of the gas yielded at a given temperature.

Porter and Ovitz made a careful determination of the gas evolved at different temperatures with Pennsylvania and Illinois coals. The results of their determinations are shown in Figure 10 and Figure 11. It is plain from an examination of these diagrams that below 500 C there is practically no decomposition of the hydrocarbons, as evidenced by the absence of hydrogen, except for insignificant quantities. After a temperature of 600 C is reached, however, the decomposition of the hydrocarbons becomes exceedingly great.

Giles and Vilbrandt (99) made a study of the gas distilled from Farmville, NC, coal at somewhat lower temperatures, the results of which are shown in Figure 12. This particular coal has the reputation of producing excessive smoke and of disintegrating when exposed to the atmosphere. These characteristics give disfavor to this fuel for domestic and industrial purposes. Reference to the illustration shows the large quantities of oxygen and nitrogen driven off in the initial stages of heating which, together with the small quantity of hydrocarbons evolved, demonstrates the reason for the ease with which which this coal disintegrates. From 200 C to 650 C, the proportion of ethane formed from this coal remains about constant. The formation of heavy hydrocarbons reaches a maximum at the very low temperature of 300 C, after which their formation gradually decreases to a second critical point just below 600 C, thereafter only a fraction of a percent being produced. Ti is an interesting fact that no hydrogen was formed up to 600 C, which shows that the volatile products were educed in their primary condition. It is to be expected, however, from the drop in the hydrocarbon curve at 600 C, that from that temperature onward hydrogen will be produced in quantities. Any decomposition of the heavy hydrocarbons below 660 C in this case evidently goes to form methane, for the percentage of this constituent of the gas has a striking inverse relationship to that of the heavy hydrocarbons. Although not shown in the illustration, carbon monoxide to the extent of 0.40% appeared at 420 C and remained constant throughout, while carbon dioxide to the extent of 0.65% appeared at 300 C and gradually increased in amount to 1.55% at 660 C. Ethylene and other unsaturated hydrocarbons, amounting to 0.74%, were produced at 200 C, but gradually decreased to 0.12% at 660 C. At 360 C the specific gravity of the total gas referred to air was 1.167, as compared with 0.952 specific gravity at 600 C. The laboratory gas yields of Porter and Ovitz and of Giles and Vilbrandt, as a function of the carbonization temperature, have been computed to a gross tonnage basis in Table 20.

Monett (100) examined 17 Utah coals under maximum carbonization temperatures of 1000 F, 12 of which were deemed worthy of further study. The average composition of the gas from these coals was: 6.7% carbon dioxide, 7.6% carbon monoxide, 28.8% hydrogen, 2.7 nitrogen, 39.6 methane, 7.1% ethane, and 6.7 illuminants. This agrees with the results of Parr and Layng (101), who also used Utah coals at maximum distillation temperatures of 1450 F and obtained a gas composition of 8.3% carbon dioxide, 8.9% carbon monoxide, 37.6% hydrogen, 3.9% nitrogen, 30.7% methane, 5.1% ethane, and 4.8% illuminants.

These data can be correlated to show the variation of the composition of low temperature gas with the age of the coal. While no definite facts are available, it has been pointed out that probably Virginia coal is the oldest and Wyoming the youngest of the 5 coals examined. Table 21 gives the comparative analyses of the gas obtained from the retort at 700 C. It is plainly evident that there is little change in the volume of gas evolved, that the percentage of the oxides of carbon contained in the gas greatly increases in the younger coals and that there is a corresponding decrease in the proportion of hydrocarbons and hydrogen liberated from the less consolidated fuels. The illuminants remain about constants. They consist of heavy hydrocarbons, which deposit soot at the temperature of the flame, thus providing minute incandescent particles which give illuminating properties to the flame.

British Coals ~

Burgess and Wheeler (57) carried on an extensive investigation of Welsh coals to determine their behavior throughout the temperature range from 450 C to 100 C. The average ultimate analysis of the coals used in their experiments was as follows: 89.9% carbon, 4.7% hydrogen, 5.1% oxygen, 1.6% nitrogen, 1.4% sulfur, and 5.14% ash. Comparison with Table 2 shows that these coals have an ultimate analysis almost identical with that of the semi-bituminous coals of the United States.

In 2-hour distillation experiments, they observed that the percentage of total gas evolved during the first 10 minutes increased rapidly with the temperature. Thus, at 600 C, 59.7% of the total gas produced in 2 hours was educed in 10 minutes; 64.6% at 700 C; 75.1% at 800 C; 88.9% at 900 C; 94.4% at 1000 C; and 94.3% at 1100 C. the commercial application of low temperature carbonization is confronted with an obstacle of first importance in the shape of retorting period, due to the low thermal conductivity of coal. The difference between the maximum and the minimum yields of various coals, examined by Burgess and Wheeler (57) is as much as 300% up to 700 C, after which the two values approach equality until at 1100 C the difference is les than 20%. Table 22 shows the range of yields obtained at various temperatures in the course of their experiments.

Burgess and Wheeler (57, 102, 103) also made a careful study of the rate of gaseous evolution and of gaseous compositions as a function of the temperature of carbonization, with Altofts Silkstone coal, the results of which are shown in Tables 23 and 24. These are very interesting data. It should be noted that there is an initial large evolution of carbon dioxide, but that this constituent decreases as the temperature rises. At the same time, the percentage of carbon monoxide increases. It will be pointed out later that this is probably caused by the reduction of the dioxide to the monoxide at high temperatures in a reducing atmosphere. This same coal was distilled under reduced pressure and the gas collected in vacuo with the results shown in Table 24.

Carbonization in vacuo completely alters the character of the gas. The volume of gas evolved increases enormously above 300 C, at which temperature both the oxides of carbon reach their maximum proportions. In contrast to normal pressure distillation, small quantities of oxygen, probably occluded, are liberated. The gas obtained in vacuo has about the same hydrogen content at 400 C as normal pressure low temperature gas has at 700 C. Burgess and Wheeler (104) studied the decomposition of Lancashire coal under heat and vacuum. Table 25 shows that the results of their investigation on this fuel agree with the findings of Table 24.

The succession of events, as the temperature of the coal is advances during exhaustion, can be grouped into 5 stages. First, occluded gases, which cannot be removed under atmospheric conditions, are driven off up to about 150 C. These gases are largely paraffin hydrocarbons, principally belonging to the higher series. Second, a copious evolution of water appears at about 200 C and continues up to 450 C. this is evidently water of constitution and its evolution is accompanied with the formation of a large percentage of the oxides of carbon. Third, between 200 C and 300 C, organic sulfur compounds begin to decompose, as is manifest by the appearance of hydrogen sulfide. The hydrogen sulfide begins at about 270 C and practically ceases at 300 C. Within this stage, higher olefins begin to appear in the volatile products, but their evolution does not fall off until 350 C has been reached. Fourth, at 310 C, a reddish-brown oil is driven off and thereafter liquids other than water are produced. As this evolution of oil is not accompanied by a marked evolution, apparently oil is not necessarily to be considered a decomposition of the coal conglomerate. Fifth, at about 350 C, there is a rapid evolution of gas, accompanied by the formation of viscid oil, indicating a major decomposition stage, after which chemical transformations increase rapidly with the temperature rise.

It has been mentioned that the composition of the gas issuing from an internally heated retort is vastly different from that obtained in closed retort operation. Carbonization is effected in the Maclaurin retort by the sensible heat of producer gas. The gas obtained from a Maclaurin retort at 700 C by Burgess and Wheeler (105) and in an externally heated horizontal retort of the Fuel Research Board (114) at 600 C are compared in Table 26. The samples were taken from large-scale apparatus. The vast difference in the volume and calorific value of the gas obtained in the two processes is seen at once. The nitrogen and carbon monoxide are enormously increased by partial gasification, the former coming from the atmosphere and the latter from combustion of carbon in the manufacture of the producer gas.

The gas from low temperature coking can be passed through scrubbers and washed with certain solvents to extract the light oils, which may then be combined with the low-boiling cut from the tar fractionation to increase the yield of motor spirit. Scubbing with creosote oil removes certain of the hydrocarbons so that after washing, a low temperature coal gas contains about 4% carbon dioxide, 8% carbon monoxide, 35% nitrogen, 46% methane, and 4% other hydrocarbons. Unfortunately, no data are available on the composition of this particular gas before scrubbing, so that the conclusions drawn must be limited. By comparison, however, with the 700 C in Table 23, it will be seen that the amount of saturated and unsaturated higher hydrocarbons has been greatly reduced by the treatment.

Peat & Lignite ~

The Fuel Research Board (106) conducted a series of experiments on the winning of peat by carbonization in standard vertical retorts with the introduction of about 6% steam. This work was carried on at moderate temperatures of about 900 C in the heating flues, which for these retorts approximated low temperature conditions within the charge. The sample of peat tested had an ultimate analysis approximating that shown in table 2, but it had been air-dried to a moisture content of about 20%. The volatile matter present amounted to 49.5%, with 26.8% fixed carbon and 3.4% ash. As charged, the peat had a heating value of about 7,675 BTU/pound. The yield of products shown in Table 27 was obtained. The peat gas was very dense, owing to its high carbon dioxide content, but despite this, it burned with a satisfactory flame of slight luminosity. It was difficult to obtain scrub light oil from this gas because of the large amount of carbon dioxide present.

Trenkler (107) has reported some tests by Muller on the coking of German brown coals at or below 500 C, which are reproduced in Table 28 to give some idea of the results to be expected from treatment of this class of fuel. A further consideration of brown coals will be given under the name of certain processes particularly designed for this purpose.

Benson and Canfield (108) investigated the possibility of winning Newcastle lignite, from the state of Washington, by low temperature distillation. This material, although of sufficient size for fuel, is generally considered too dirty to be worked. It contains about 12% moisture, 37% volatile matter, 41% fixed carbon, 10% ash, 1.4% nitrogen, and 0.34% sulfur. It has a calorific value of about 10,400 BTU/pound. The yield and composition of gas obtained from this lignite when distilled at various temperatures is shown in Table 29. The decomposition of the lignite became marked between 300 C and 400 C, as indicated by a large increase in the production of gas as well as by an increase in the proportion of hydrogen present within this temperature range.

The American lignites differ greatly from German brown coals in that it is impossible to form a stable and satisfactory briquette from them by the simple procedure of drying and pressing into shape under heat and pressure. The reason for this is that the former contains insufficient binder to consolidate and waterproof the mass. On the other hand, lignite char, when briquetted, forms quite a satisfactory fuel and, if by-products can be recovered from its carbonization, the cost of its production may be materially lessened. Further results of experience in the carbonization of lignite and other low-grade fuels will be given in connection with the discussion of the various processes with which experiments have been conducted and reported.

Shale ~

The distillation of oil shale has been carried on for many years in Scotland and in France, but the two industries are of independent origin. James Young, according to Redwood (109), while engaged in oil refining seems to have conceived the idea that oil originated from coal by distillation, due to subterranean heat. Young (110) tested a number of low-grade fuels, which resulted in the patent on his first oil shale retort. Since that time, the industry has become well established in Scotland, although the financial status of the many companies engaged in this enterprise did not generally become satisfactory until 1900. In 1919, however, changing economic conditions practically caused a shutdown of the Scottish shale oil plants, but some recovery has since taken place.

Naturally, the yield and composition of shale gas depends upon the composition of the shale and the conditions of retorting. Gas is evolved long before the first indications of oil appear and it continues for a long time after the eduction of oil ceases. However, the evolution of gas drops off rapidly after oil is no longer produced. According to Mills (111), Scottish shales have an average ultimate analysis of 25.3% carbon, 3.7% hydrogen, 5.7% oxygen, 1.1% nitrogen, 0.5% sulfur, and 63.8% ash. Most American oil shales contain from 1.5% to 6% hydrogen sulfide, the greater part of which is evolved in the early stages of distillation. Garvin (112) has reported the composition of shale gas from Scottish and American practice. The gases in Table 30 from the Colorado and Utah shales were obtained under conditions of dry distillation, whereas that from the Scottish shale was obtained from steaming the retorts. It will be observed later that the admission of steam during carbonization introduces several desirable complications into the process, among which is a partial formation of water gas, thereby accounting for the production of a large quantity of low thermal value gas.

Finley and Bauer (113), in distilling some American oil shales at 1000 F, found considerable effect in the analysis of the gas caused by oxidation of the shale before carbonization. It will be seen in table 31 that pre-oxidation reduced the oil output 55%, increased the amount of carbon dioxide in the gas 120%, and decreased the amount of hydrogen 37%, as well a causing an increase in the proportion of methane and a decrease in the proportion of ethane and illuminants.

Time Effect ~

In any industrial process, the time required for the necessary operations is of primary importance. The overhead charges demand the maximum output per unit of investment and this can be attained only through reduction of the time interval between raw material and finished product. In their work on the volatile products of coal carbonization, Taylor and Porter (98) made some important observations on the yield and composition of low temperature gas, as a function of the time of distillation. Tables 32 and 33 and Figures 13 and 14 have been compiled and plotted from their data, converted for convenience to units per gross ton of coal.

Table 32 admirably illustrates what a large part of the gas is composed of saturated hydrocarbons, a compared with the other constituents. The yields of the oxides of carbon, of hydrogen, and of hydrogen sulfide are not greatly increased beyond 25 hours of carbonization at 350 C with Pennsylvania coal. The unsaturated hydrocarbons increase but little beyond the 50 hour period, but the saturated compounds continue to be evolved in large quantities, although at a constantly decreasing rate, even to beyond 10 days.

The results with Wyoming coal are shown in Figure 13. Contrasted with the harder Pennsylvania coal, we find that carbon dioxide has moved from one of the minor to the chief constituent of the gas, and that carbon monoxide too has become a major constituent. With Wyoming coal, heated at 350 C over a long period, the amount of ammonia, hydrogen sulfide, and unsaturated hydrocarbons evolved increases only slightly beyond 25 hours of retorting. Hydrogen has a steady increase, but even after 10 days remains the smallest proportion of the gas.

Even after prolonged heating of 240 hours at 350 C, the entire volatile matter of coal is not removed. Subsequent treatment of the residuum at higher temperatures will yield further and even more voluminous products. The residuum from Pennsylvania coal, discussed in Table 32, was submitted to such an experiment and the composition of the gas is indicated in Table 33. By comparison of the two tables, it is seen that the yield of hydrogen becomes second in importance when the coke is subsequently heated at 450 C, after preliminary carbonization at 350 C. the saturated hydrocarbons are still the major components of the gas, while the other constituents remain of minor importance as before.

Figure 14 shows the total gas evolved from Pennsylvania and Wyoming coals and from the residuum of Pennsylvania coal, previously heated for 240 hours at 350 C. It will be seen that almost 500% more gas is evolved from the coke residuum than from the original coal, when the temperature is advanced but 100 C.

During the first few hours after charging in coke ovens operating at 1000 C, the gas evolved approximates the composition of low temperature gas. The layers of coal a short distance from the retort wall do not reach the higher temperatures until the lapse of considerable time. The illustration in Figure 15 shows the change in composition of the gas from an Otto coke oven operating at 1000 C as a function of the time after charging (89).

Examination of Figure 15 makes it evident that, during the first 2 or 3 hours after charging, the gas has all the characteristics of the low temperature product, that is, high calorific value, exceeding 800 BTU/cu ft; high percentage of saturated hydrocarbons, represented in this case by methane; relatively low hydrogen content; and minor percentages of carbon monoxide and unsaturated hydrocarbons. As the period of carbonization is extended, the interior of the mass of coal reaches the higher temperature and the composition of the gas alters accordingly. Finally, after 20 hours of retorting, the thermal value of the gas falls to about 400 BTU/cu ft and hydrogen becomes the main constituent. As the time proceeds and the average temperature rises, the percentage of carbon monoxide remains about constant, while the proportions of both the saturated and unsaturated hydrocarbons constantly decrease. The unsaturated compounds are represented mainly by ethylene. It is interesting to note, in the illustration, the parallelism existing between the percentage of methane present and the thermal value of the gas. This is easily understood when it is recalled that the calorific value of methane at constant pressure is 1072 BTU/cu ft, while that of hydrogen is but 347 BTU/cu ft, so that the resultant heat value of the has is largely dependent on the percentage of saturated hydrocarbons present.

No data are available on the initial gas evolution and the temperature adjustments when raw coal is introduced to a heated retort, but Taylor and Porter (98) have made experiments on small quantities of Pennsylvania coal plunged into a tube heated to 1000 C. The curves of Figure 16 show how the coal absorbs heat from the walls of the retort with a corresponding reduction in their temperature. As the charge absorbs heat, the wall temperature falls until at the end of about 30 seconds the layer of coal next to the wall has reached the equilibrium temperature, then the whole mass gradually absorbs heat and rises in temperature. Over 30% of the volatile matter was evolved during the first 30 seconds of heating and thereafter the increase in gaseous evolution was comparatively slow.

The Fuel Research Board (114) has investigated the effect of the time element in the low temperature carbonization of 5 British coals in horizontal retorts. These experiments were conducted on a large scale at 600 C. The average results are represented graphically in Figure 17. The initial evolution of gas during the first 40 minutes, as shown in the illustration, consisted mostly of carbon dioxide and some air expansion. After reaching a minimum, the rate of gas evolution attained a maximum after about 2 hours of retorting. At this point, the charge yielded gas at the rate of over 1.1 cu ft/minute/100 lb of coal. Thereafter, the rate of evolution gradually fell to less than 0.20 cu ft/minute/100 lb after 5 hours of carbonization. The calorific value of the gas had a rapid rise to a maximum of 1200 BTU/cu ft slightly before the rate of evolution reached its highest point. It then fell rapidly and flattened out after 3 hours at a thermal value above 600 BTU/cu ft.

Taylor and Porter (98) found that the rapidity with which the coal is heated has little effect upon the composition of the gaseous products, when the gas evolved is immediately removed from the carbonization chamber. When Pennsylvania coal was carbonized at a maximum temperature of 1050 C over a period of 270 minutes and compared with another sample brought to the same temperature in 6.5 minutes, it was found that the only appreciable change in the composition of the gas was a slight increase in the percentage of hydrocarbons present and a corresponding decrease in the proportion of hydrogen. Thus, in the first case, the hydrocarbons made up 21.3% and the hydrogen 69.1% of the gas, while in the second case, they composed 26.7% and 63.1% respectively. The other components of the gas remained practically unchanged in their proportions.

Vacuum Effect ~

The destructive distillation of US coals in vacuo at temperatures ranging from 250 C to about 1000 C, has also been investigated by Taylor and Porter (98). Due to the different periods of carbonization which they selected, no comparative data can be drawn, except for two temperatures, but these wills serve to show the effect of vacuum distillation at both high and low temperatures. The duration of carbonization was 6 hours at 600 C and one hour at 1050 C. We have already seen in Tables 24 and 25 the effect of vacuum distillation on two British coals at somewhat lower temperatures.

Table 34 affords another excellent example of the effect of the consolidation of the coal on the quality of the gas yielded, as pointed out in Table 21. Like the Welsh coals, vacuum distillation of US coals increases the hydrogen content of the gas evolved and decreases the percentage of hydrocarbons.

A critical comparison of Tables 34 and 35 shows that increasing the temperature of distillation in vacuo of American coals from 600 C to 1050 C almost doubles the amount of hydrogen present in the gas and approximately decreases by half the proportion of saturated hydrocarbons present. Very little change in the amount of unsaturated hydrocarbons and oxides of carbon was observed.

Secondary Decomposition ~

We have seen that the fundamental purpose of low temperature carbonization is to obtain primary decomposition products from coal and its related solid fuels. Superimposed upon this task is the equally difficult problem of preventing secondary decomposition of the volatile products after they have been educed from the charge, thus entirely defeating the purpose for which this special branch of destructive distillation was designed. If, after liberation, the volatile matter is allowed to become superheated or to come in contact with incandescent surfaces, secondary reactions set in and the composition of the products is changed accordingly. In continuing their research, Taylor and Pryor (98) made an investigation of this phenomenon. Table 36 gives some quantitative results obtained by superheating low temperature gas secured from the carbonization of Illinois coal at 450 C.

Superheating tends to crack the heavier saturated hydrocarbons, with the formation of hydrogen and unsaturated compounds. Similar observations were made on Pennsylvania coal gas obtained at 475 C, except that the area of heated surface to which the gas was exposed was considerably altered. In the first case, the heated area was reduced in size and increased in temperature by passing the gas over a glowing wire and, in the second case, the area was increased by the presence of a column of broken firebrick. The catalytic effect of even small incandescent surfaces is really interesting. In Table 37 we see that the saturated hydrocarbons have been dehydrogenated, as evidenced by the decrease of the percentage of saturated compounds present and by an increase of almost 30% in the proportions of hydrogen and unsaturated hydrocarbons.

Table 38 shows that the temperature at which cracking manifests itself, in the case of Pennsylvania coal, is fully 150 C lower than in the case of Illinois coal, discussed in table 36. It is unlikely that this difference in cracking temperature arises from the nature of the coal carbonized, but it is probably the result of the increased catalytic surface presented by the brick column. This investigation was extended to the case of reduced pressure, both with and without the presence of broken brick. It was found that, with a pressure below 4 cm Hg, the cracking temperature rose from 650 C to about 800 C in both cases, the only effect of the brick column being to increase the total volume of gas and the percentage of hydrogen.

Gaseous Atmospheres ~

The effect of various atmospheres on the modification of Illinois coal by low temperature distillation has been investigated by Parr and Francis (30). They carbonized coal in atmospheres of oxygen, nitrogen and steam, and studied the character of the coke and gas under these conditions. Table 39 shows the result when Illinois coal was coked at various temperatures in an atmosphere of nitrogen. This table, as well as Table 40, has been computed to a nitrogen and oxygen-free basis. As may be expected, less carbon dioxide and more hydrogen was present in this case, than when an atmosphere of oxygen was used, as indicated by Table 40. The greatest effect on the composition of the gas evolved under nitrogenous atmospheres seems to be in the illuminants. The proportion of these heavy hydrocarbons was greatly increased, as may be seen by comparing Table 39 with Figure 11.

Table 40 shows the effect of carbonizing Illinois coal for 4 hours in an atmosphere of oxygen. The most important point to be noted here is the high percentage of carbon dioxide present which, coupled with the fact that temperature variations occurred which were wholly independent of the external heating, is positive evidence that oxidation of the charge has taken place. This should be compared with Figure 11, where it is seen that, in ordinary conditions, the percentage of carbon dioxide in the gas normally decreases with the temperature rise. No hydrogen was evolved up to 380 C.

Steaming ~

It has been pointed out that superheating the primary gaseous products introduces secondary reactions and that large moderately heated surfaces, as well as small incandescent surfaces, catalyze the cracking of the saturated hydrocarbons. The primary object of steam distillation in the carbonization of coal is to remove quickly the volatile products, so as to prevent their coming in contact with the hot retort walls. By suitably regulating the flow of fresh steam, the gases and vapors can be swept clear of the retort and the cracking effect largely reduced. Of secondary importance, and sometime of undesirable consequence, is the reaction between the steam and coal which tends to increase the gas evolved at the expense of the coke yield. The physical chemistry, determining the extent to which this action occurs, will be discussed later.

In low temperature carbonization, only a small amount of decomposition takes place when steam is introduced into the retorts and, if any reaction with the charge does occur, the dioxide, rather than the monoxide of carbon, is more likely to be formed. This will cause a decrease, rather than an increase, in the total calorific value of the gas. Therefore, any value that steaming may be found to have in low temperature methods must be of a physical rather than of a chemical nature. Thus, the steam absorbs heat in the hot bed near the bottom of the retort and distributes it through the cooler portions of the fuel bed near the top, thereby assisting in the transfer of heat throughout the charge. The function has been adequately illustrated in Figure 9.

The Fuel Research Board (115) as a result of their experiments on the steaming of coal during coking, recommend that high temperature vertical retorts should never be used with less than 5% of steam. This was found to increase greatly the heating quality of the gas and the yield of ammonia. Steaming has not proved practical in horizontal retorts in experiments so far conducted. The effect of various percentages of steam on high temperature gas, as determined by the Fuel Research Board using Arley coal, is shown in Table 41. The carbonization was carried on at 1170 C.

As the percentage of steam increases, the volume of gas becomes greater and the thermal value decreases. It will be shown later that in this case the temperature is too great to favor the formation of carbon dioxide in the presence of steam, but that there should be an increase in the carbon monoxide present. That this is the case, may be seen by reference to the table. The percentage of methane slightly decreases, and doubtlessly this in part accounts for the lower heating value of the gas. Carbon monoxide has a calorific value of only 341 BTU/cu ft as compared with 1072 BTU/cu ft for methane, so that the increase in the percentage of monoxide cannot be expected to compensate, from a calorific standpoint, for the decrease in the proportion of methane. The other constituents of the gas are not greatly influenced by the presence of steam.

Investigations, on a laboratory scale, of the effect of steam in low temperature carbonization were made by Davis and Parry (97), who studied Pennsylvania coal. In large-scale experiments of the Fuel Research Board, shown in Table 43, only up to 20% steam was admitted, while in this case up to 88% steam was passed through the retort to sweep the chamber clean of gas 5 to 7 times a minute. While the greatest possible effect would be obtained by agitating the coke, stirring of the charge was avoided in order not to introduce another variable and nullify the comparative value of the data. There are so many variables in low temperature carbonization that it is best to reduce these to a minimum for simplicity of interpretation.

The effect of steam in this case is plainly evident. Comparing the distillation with and without steam at corresponding temperatures, it will be observed that the percentage of carbon dioxide just about doubles in the presence of steam, while the carbon monoxide is only slightly increased at 650 C and remains practically unchanged at 550 C. This, together with the increase of hydrogen content, is indicative of the following reactions:

[12] H2O + C => CO + H2

[13] 2H20 + C => CO2 + 2H2

The data in Table 42 show that the first reaction, represented by the foregoing chemical equation, is hardly apparent at all at 550 C, but takes place to a minor extent at 650 C. On the other hand, the second reaction occurs to a considerable degree even below 500 C.

The experiments of the Fuel Research Board (92) on the effect of steam in low temperature carbonization were carried on in Glover-West vertical retorts, using a blend of 60% Mitchell Main, a coking coal, and 40% Ellistown Main, a non-coking coal. The temperature of distillation ranged from 850 C at the bottom of the retort to 700 C at the top. Examination of Table 43 discloses that the introduction of steam has little effect on the percentage of carbon monoxide, but that there is a slight increase in the proportions of carbon dioxide and hydrogen. It may be concluded, therefore, that there was a slight reaction between the steam and coal and that this took place in accordance with the principles deduced and in agreement with the data derived from the work of Davis and Parry.

It cannot be concluded that steam has any effect on the remaining components of the gas, but it will be observed later that, under certain conditions and in certain processes the introduction of steam during carbonization has a beneficial effect on the removal of sulfur from the coke.

Physico-Chemical Equilibrium ~

From the standpoint of physical chemistry, it is necessary to consider four reactions when steam is passed into a hot coal bed. Of these, two are primary reactions, represented by Equation (12) and Equation (13), previously mentioned, and two are secondary reactions, indicated as follows:

[14]  C + CO2 <=> 2CO

[15] H20 + CO => H2 + CO2

With reference to Equation (15), we have by the law of mass action,

[16] (H2O) x (CO) / (H2) x (CO2) = K

In other words, the product of the concentrations of the reagents divided by the product of the concentrations of the products is a constant at any given temperature of the products is a constant at any given temperature, regardless of the proportion of the components present at the beginning of the reaction. The constant K, however, is a function of temperature, as expressed by Van't Hoff's equation:

[17]  d logeK / d T = Q / RT2

Where T is the absolute temperature; Q, the heat involved in the reaction, and R, the universal gas constant (1.985 calories per degree). The integral,


cannot be solved directly because Q is a function of T and this relationship must be established by experiment. Such experiments have been performed and it has been determined that the constant has the following values, according to Landolt-Bornstein (116):

Temperature:    686 C      886 C      1005 C     1205 C
K                      0.534        1.197         1.620         2.600

Equation [15] is a first order reaction, and in such a case the rate of formation of the product is proportional to its own concentration and to a coefficient known as the specific reaction time, that is:

[19] d (CO) / dt = k1 (CO)

As the equilibrium is a dynamic one, the reverse reaction must be considered also:

[20] d (CO2) / dt = k2 (CO2)

This too is a first order reaction, because in the carbonization of coal, sufficient hydrogen is liberated to cause no appreciable decrease in its concentration. The resultant rate at which the monoxide is formed, therefore, is the difference between these two equations:

[21] [d (CO / dt] r = d (CO / dt - d (CO2) / dt = k1 (CO) - k2 (CO2)

Now the equilibrium constant K is equal to the ration of the specific reaction rates of the direct and reverse reaction, K = k1 / k2, from which it will be seen that, if the value of the equilibrium constant is greater than unity, the rate of formation of the monoxide, in accordance with Equation 15, is positive, and if K<1, then the rate of formation is negative. In other words, it becomes possible at any temperature to determine the extent to which the reaction occurs. In the simultaneous occurrence of reactions [12] to [15], therefore, it may be determined, with sufficient data available, which reaction dominates. At any given temperature, it depends merely upon the rate at which equilibrium is established.

At temperatures below 700 C, reactions [13] and [15] predominate, so that the steam reacts with the coal to produce the dioxide and any monoxide that is formed is oxidized to the higher compound. The reaction velocity of Equation [14] is slow in any case, so that its effect is negligible. Above 700 C reaction [15] is very rapid, while beyond 1000 C reaction [12] becomes of first importance. These theoretical deductions are substantiated by the data given in Tables 41 and 42.

Your Support Maintains this Service


The Civilization Kit

... It's Your Best Bet & Investment in Sustainable Humanity on Earth ...
Everything @ on a Data DVD