Return to Table of Contents


F. Gas Recycle Operation

Both Lurgi (13) and Ruhrchemie did considerable experimental work on the recycle operation of the Fischer-Tropsch process using a conventional externally cooled middle pressure reactor and a relatively low recycle ratio.  This is to be distinguished from the I. G. process developed by Michael, which involves a very high recycle ratio, so that the heat of reaction is carried away by the circulating gas.  Dr. Herbert of Lurgi stated that this I. G. process has been operated only on a pilot plant scale and in his opinion the cost of such recycling on a commercial scale would be prohibitive.

The Lurgi low recycle operation  was developed in a pilot unit installed at Hoesch-Benzin at Dortmund in 1938.  This unit consisted of a single tube of the standard middle pressure reactor as built by Gutenhoff-nungshute.  The annular layer of catalyst had an inside diam. of  24 mm. and an outside diam. of with an overall length of 5 meters.  The tube was water-jacketed with a vapour chamber connected to the top by means of which the steam pressure could be controlled to give the desired synthesis temperature.  This unit was operated under varied conditions with a Ruhrchemie cobalt catalyst for about two years.  Single pass operation with this catalyst gave a yield of 155-162 grams of liquid plus "gasol" per cubic meter and the liquid product had the following composition:

Gasoline boiling below 200°C 45%
Oil boiling 200/320°C 25%
Wax boiling above 320°C 30%
(Wax boiling above 460°C 12%)

The octane number of the gasoline was 40 and the centane number of the fraction boiling from 120° to 280°C. was above 100.  With a recycle ratio of about 3:1 it was possible to increase the throughput by 30% without sacrifice of yield in grams per cubic meter or to realize a yield of 170 gm/m3 at the same throughput.  With recycle operation charging water gas instead of ideal gas the olefin content of the liquid was 50-60% in stead of 20% as normally.  Typical conditions and results of the Lurgi operation using recycle in only the first of three stages as follows:

Catalyst 100 Co: 5ThO2:8 MgO:200 Kieselguhr
Pressure 7-10 atm.
Temperature 190-225°C
Fresh gas charge 1200 m3/hr.
Recycle gas rate 3000 m3/hr.
Gas rate from first (recycle) 
stage to second stage
540 m3/hr.
End gas from last stage 33 m3/hr.
Overall product per 1000m3 charge gas:
"Gasol" 15 Kg.
Gasoline 73 kg. (50% olefins)
"Oil" 36 kg.
Wax 36 kg
  160 kg.


Charge to
 Stage 1
Stage 1
Effluent from
CO2 10.5 19.7 23.5 44.8
CO 31.8 28.3 26.9 9.5
H2 51.4` 35.4 28.9 6.4
CH4 0.4 5.2 7.0 16.3
N2 5.9 10.7 12.7 22.5
CnHm -- 0.7 1.0 0.5

The pilot plant work at Hoesch Benzin had demonstrated the merits of middle pressure recycle operation so that both Hoesch and Schaffgotsch had given orders for the conversion of their commercial plants to this type of operation.  However, these conversions had not been accomplished because of bombings of the plants in question and scarcity of necessary materials and equipment.

With regard to Ruhrchemie gas recycle operations with a cobalt catalyst, Martin (9) stated that experiments had been made on a laboratory scale only, but that these experiments had been promising enough to justify installation of the necessary plant equipment for commercial operation at Sterkrade.  This equipment was destroyed by bombing before it could be put into operation.

Alberts (2) stated that the Ruhrchemie-Sterkrade plant involved the use of all the medium-pressure ovens in one stage with water-gas as the feed material and a 3:1 recycle gas-fresh gas ratio, the unrecycled residual gas being passed to a normal 2-stage atmospheric-pressure section after adjustment of the H2:CO ratio to 2:1 in a CO-conversion unit.  The system is illustrated by the flow diagram of Figure 5.   

The object of this method of operation was the production of olefines by maintaining a high concentration of CO in the gas mixture.  If this is attempted by, for example, using an undiluted synthesis gas of composition 2CO + H2 there is a tendency to get carbon deposition.  Alberts stated that the use of water gas with recycling was the best method of achieving the desirable effects of high CO concentration while avoiding carbon deposition.  The principle of the process was illustrated by Alberts as follows:

  Parts H2 Parts CO
Fresh water gas 1.25 1.00
Consumed in reaction 1.00 0.50
Residual gas 0.25 0.5
3 volumes recycle gas 0.75 1.5
1 volume fresh water gas 1.25 1.0
Total 2.0 2.5

i.e. total inlet gas has the inverse H2:CO ratio to that of water gas.

The reaction temperature for the process was higher than usual viz. 220-225°C.  The contraction in the process was 50% and the yield 100-110 gm. C3 and higher hydrocarbons /m3 ideal gas.  In the pilot plant experiments a catalyst life of 6-7 months had been obtained.  The products were as follows:

C3 + C4 =

8% (olefines 60-65%)

Gasoline C5-C10 =

30% (olefines 60-65%)

Middle Oil C10-C17/18 =

28% (olefines 40-65%

Wax C18 =

  34% (olefines small %)

 The gasol fraction formed an excellent raw material for polymer gasoline production.  the gasoline had an octane of 50-55 but this could be increased to 70 by an isomerisation process involving no gas formation or change in boiling range and olefine content.  It consisted of treatment at atmospheric pressure and 300°C. in vapour phase over Floridin activated by treatment with HCl.  The space velocity and clay life were the same as for normal clay refining.  Alternatively the gasoline could be polymerised with aluminium chloride to give a lubricating oil of viscosity pole height 1.7.

The unsaturated middle oil would have formed the main raw material for the OXO plant.  The paraffinic residue, after the latter process had removed the olefines, would be sold as diesel oil.

Martin (9) stated that the oxygen products from recycle operation with water gas were both carbon dioxide and water instead of substantially all water as from the hydrogen-richer normal synthesis gas.  The life of the catalyst was not changed by this recirculation system, being still five to six months.

Alberts considered that this recycle process was the best method of conducting the synthesis with a cobalt catalyst.  He stated that the recycle of total products had been tried on a commercial unit but was found to lead to lower conversions and an increase in saturated hydrocarbons.

In working along these Lurgi appears to have been ahead of Ruhrchemie or more aggressive (or both) and developed a patent position which led to a contract with Ruhrchemie giving Lurgi the exclusive right to build Fischer-Tropsch plants under Ruhrchemie patent licenses in all countries except the United States and Japan.  Lurgi's favorable position in the field of high pressure gasification was also a factor in reaching this agreement.  Up to the end of the war Lurgi had exercised the rights thus acquired only in the building of city gas conversion units under the Geilenberg plan.

G. Products and By-Products

    1. General

In general, C3 and C4 olefins from the "Gasol" product were converted to alcohols or other chemical derivatives and the residual paraffins were liquefied and distributed in cylinders as "Treibgas" for motor fuel.  The light gasoline was shipped to government depots for blending in motor gasoline or aviation gasoline, but was usually not better than 50 O. N., I. G. Research. The heavy gasoline and light Kogasin went into Diesel fuel, partly blended with aromatic wash oil to increase the quantity.  Heavier Kogasin was subjected to sulfochlorination by I. G. to make "Mersol" detergents, or was converted to synthetic lubricants by various processes.  Gatch (paraffin wax) was sent mostly to Witten for conversion to soaps.  The yield and quality of products from all of the commercial plants investigated did not differ materially from previously published information.

2. Olefins in Fischer-Tropsch

According Dr. Koch, (3) the Fischer-Tropsch synthesis gives alpha-olefins as the most likely primary products.  The beta-olefins are formed  from the alpha-olefins through an isomerization reaction, the degree of this isomerization being dependent upon the ratio of hydrogen to carbon monoxide, pressure and catalyst.  An investigation of the C4 cut for alpha and beta olefins showed that by increasing the ratio of hydrogen to carbon monoxide, a greater concentration of beta-olefins is obtained, presumably by means of atomic hydrogen.  A higher operating pressure gives more alpha-olefins.  The use of the iron catalyst produces a higher concentration of alpha-olefins.  This is shown in the following table:


Cobalt at normal pressure 0.25
Iron at 10 atms. 0.67-1

It is realized that the comparison is not strictly accurate because the two catalysts have been used at different pressures, and higher pressures do give a higher concentration of alpha-olefins.  No information on the ratio of butene-1 to butene-2 produced by the cobalt catalyst at medium pressure is available.  No information is available on the effect of temperature used in the synthesis.

3. Diesel Oil:

Despite the excellent quality of the Diesel oil recovered from Fischer-Tropsch primary products, Martin (9) considered this a low-grade use of the product, emphasizing again his opinion that Fischer-Tropsch primary products should be converted into more valuable materials, such as special chemicals.  for example, he mentioned that I. G. had purchased the Diesel oil fraction (B. P. 230°-320° and about up to C18 or C19) for the purpose of manufacturing detergents.  the fact that this had been done in war time, when Diesel oil was in critical demand illustrated his point, Martin believed.  At this point Martin referred to an article by Dr. Asinger (Zeit. Angen. Chem. about Dec. 1944) as a source of information on the chlorination and nitration of these higher hydrocarbons.

Questioned as to the possible increase in yield of the Diesel oil fraction by some modification of the normal Fischer-Tropsch process, Martin said every effort to accomplish this had failed.  In general, the maximum yield of this fraction by any type of operation was about 35 per cent.

4. High-Melting-Point Wax:

Martin was asked about the possible commercial production of high-melting-point wax by a ruthenium catalyst as proposed by Pichler.  He said that he did not consider this commercially feasible because of the extreme scarcity of ruthenium,  Martin said that such waxes could be produced more economically by means of the usual cobalt catalyst, although, of  course, not in 100% yield as had been claimed by Pichler for the ruthenium catalyst.

Martin stated that an alternative method for producing hard wax by means of the usual cobalt catalyst, although, of course, not in 100% yield as had been claimed by Pichler for the ruthenium catalyst.

5. Synthetic Lubrication Oil:

Martin said that the quality of the lubricating oil made form Fischer-Tropsch primary products had been much improved since the beginning of the war.  The oxidation test had been improved by the addition of inhibitors, the best one being phenothiazine.

The best stock for lubricating oil was cracked material made from those fraction of the primary product boiling between 220° and 320°C. and wax with a melting point up to 30°C.  All material intended for this cracking step was filtered to remove the small amount of cobalt that had entered the oil from the catalyst, since it had been found that even the small amounts of cobalt caused undesirable side reactions during cracking   The cracking was carried out in a Dubbs unit in the presence of steam at a temperature under 500°C..

The production of synthetic lubrication oil at the Sterkrade-Holten plant of Ruhrchemie had averaged 1400 metric tons per month.  The yield of lubricating oil from the cracked product was about 55%, with about 25% going to gas.  Most of the lubricating oil made had a viscosity of 6 to 7 degrees Engler and a viscosity pole height 1.7, although attempts had been made to make higher viscosity oils by operating at lower temperature.

H. Miscellaneous Activities.

Martin reported that pilot plant research had been conducted on the reaction of water gas-acetylene mixtures over a cobalt catalyst.  Possibly it was expected that acrylic aldehyde or its homologs might result.  The total pressure was 20 atmospheres and , of this, acetylene accounted for about one atmosphere partial pressure.  Water gas was present in about stoichimetrical quantity, while the remainder was inert gas..  It was understood that this investigation was still in progress when the bombing stopped all research and that the results are far from complete.

Martin was asked why Ruhrchemie had not made further progress in the design of catalyst chambers in view of the fact that the present two standard designs (plate type and concentric-double-tube type) had been known before the war.  He said that the four construction companies making catalyst chambers (Gutehoffnungshütte, Mannesmann, Krupp, etc.) had decided to freeze the design in order to avoid changes in tools, patterns, etc.  He said that Ruhrchemie itself had considered other designs, such as cooling liquid flowing over the catalyst, catalyst suspended in liquids, but after consideration they always came back to present designs as preferable.

Return to Table of Contents