WILLIAM A. HORNE Ė "CATALYST MANUFACTURE FOR FISCHER-TROPSCH AND OTHER PROCESSES";

 

W. A. Horne: I will preface this discussion with the statement that all material presented has been abstracted from reports written in England by Technical Oil Mission members. However, some of the last written reports have apparently not been distributed in this country.

In discussing the Fischer-Tropsch catalysts, I will review their preparation in the order of their use in the process.

A. Preparation of Synthesis Gases

Synthesis gas is prepared from water gas or from coke oven gas.

1. Brown Oxide Catalysts

Water gas and steam are passed over the so-called "brown oxide" catalyst for CO conversion to CO2, the reaction being as follows:

CO + H20_______CO2 + H2

The resulting gas after CO2 removal is then blended back with water gas to obtain the desired ratio of carbon monoxide to hydrogen. This catalyst has the following composition:

Fe203  86.87%
Cr203 7-7.5%
Sodium and Iron Sulfate Balance
(approximately 1% of SO4)
Specific gravity about 1.17

The preparation of this catalyst is divided into three operations by I.G.:

(a) Preparation of the iron oxide
(b) Preparation of the chromium nitrate solution
(c) Mixing of a and b to yield the catalyst

For the iron oxide preparation, iron sulfate (FeSO4.7 H20) is dissolved in water to give a 20% solution and the solution is pumped into a storage vessel. Sodium carbonate (10-12% solution) is put into the precipitation vessel, and the iron sulfate is added with stirring at about 30-35oC. Quantities are adjusted to give a slight excess alkalinity at the end of the preparation. The iron carbonate precipitate is allowed to settle and is then well washed on a rotary filter with warm water (60-70oC.) It is then dried in a rotary kiln and decomposed in an annealing furnace. The iron oxide formed (about 95% Fe203) should have a weak red glow on leaving the furnace. It is then cooled in a screw-propeller cooler and stored in bunkers.

The chromium nitrate solution is prepared by dissolving a 34-40% wet slime of chromium oxide in nitric acid at about 70-90oC., with stirring. The nitrate solution should contain a minimum of 15% Cr203.

The further working up of the catalyst consists in filling a kneading machine with the previously prepared iron oxide and adding, with kneading, enough chromium nitrate solution to give the desired composition of the finished catalyst (86-87% Fe203, 7-7.5% Cr203). The product from the kneader is pressed into cakes (10-12 mm. Thick) and is roasted on plates in a tunnel kiln at 500-550oC. The dried cakes are broken in a cylindrical crusher. The fines (under 5mm.) are screened out and the material above 5 mm. Size is again calcined in a furnace at 550oC. It is then screened into small-kernal size (5-10 mm.) and large kernel-size (over 10 mm.) and packed in barrels.

The fines mentioned above (under 5 mm.) after a fine-grinding can be used when kneading up a new batch.

Water-gas shift catalyst obtained from Rheinpreussen, which was supposed to be fresh converter catalyst, was analyzed by the Fuel Research Station as follows: "This sample consisted of grayish-brown irregular-shaped pieces, approximate size range, Ĺ to ĺ", of great hardness and mechanical strength bulk density: 1,300 g/liter.

Analysis: %
Moisture (by distillation with xylene) 5.0
Additional moisture given off at 800oC 13.0
SiO2 5.1
Fe2O3 38.5
Cr2O3 5.4
A12O3 2.5
CaO 18.2
MgO 5.2
-CO2 5.0
-SO3 1.9
Difference (alkalies, etc.) 0.2
100.0

This analysis represents no unusual features. Chromium oxide is a normal promoter for the shift catalyst, and lime is a recommended addition, burnt dolomite being of ten used as the basis of the catalyst."

From seized documents it appears that the converter catalyst contained the following ingredients although its entire composition was not given:

Cr 3.7%
M1 0.1%
P 0. 1%
As Trace

2. Methane-Splitting Catalysts (I.G.)

This catalyst is used to convert methane or coke oven gas in the presence of steam at about 700-750o. to CO, CO2, and H2. Its composition is approximately:

SiO2 20-21%
A1203 17-18%
Fe203 5.0-5.5%
NiO 19-20%
CaO 10-11%
MgO 8.5-10%
The specific gravity is about 1.0

For the preparation of the catalyst, 184kg. of nickel powder as a 14.5% nickel nitrate solution is diluted with 1,000 liters of water, and is precipitated at 65-70oC. with 340 kg. of sodium carbonate (as a 10.5% solution). A light excess of sodium carbonate should be present. The total content of the precipitation vessel is separated in a filter press and the case washed free from carbonate and nitrate at about 40oC. The filter cake is blown dry with compressed air for about 15 minutes.

To about one-third of this filter cake (about 360 kg.), in a kneading machine, is added 178 kg. of Kaolin 76 kg. of magnesium oxide, and 36 kg. of nickel powder (as about a 15% solution). After kneading, the mass is spread about 20 mm. Thick on plates and ignited for about 7 hours at 500oC. The ignited material is ground to powder. Fifty kg. of this powder are mixed with 23 kg of alumina cement in a mixer, with the addition of 12 liters of water, screened through a 2 mm mesh sieve, and pressed in a tabletting machine to Raschig-ring shapes (16-18 mm. Diameter and 12-14 mm. Long). These rings are piled up and sprayed with water twice a day for three days to set the cement. They are then packed.

The methanization catalyst made by Ruhrchemie is somewhat similar to the above I. G. preparation. This catalyst was used to convert residual gas, CO, CO2, and H2 into methane. Its composition is as follows:

Nickel 60.5 wt. %
Magnesium oxide  9.0 wt. %
Kieselguhr 30.5 wt. %

For the precipitation of 1 kg. of nickel, 24 liters of a solution of nickel nitrate containing 42.0 g./1 of Ni and magnesium nitrate containing 6.2 g./1. of MgO is heated to boiling. Forty liters of a boiling sodium carbonate solution (80.o g./1.) is added slowly to the nitrate solution with efficient stirring (centrifugal stirrer). When precipitation is complete, 0.5 kg of kieselguhr is added and mixed well.

The hot mixture is then filtered rapidly in a filter press and washed with 120 1. of hot water.

The moist filter cake is shaped in an extruder and dried on a belt heated by an open flame. The catalyst is next brought to proper size (3-5 mm) in a special mill (KornmŁle) and screened.

The granular mass (called grŁnkorn) is reduced for one hour at 350oC. with a mixture of hydrogen and nitrogen (3:1) at a space velocity of 6000-8000 v/v/hr. (calculated at operating conditions). The reduced catalyst is swept by the stream of hydrogen and nitrogen until the temperature has fallen below 100oC, then it is purged with nitrogen and saturated with carbon dioxide. The catalyst now contain 60% of metallic nickel (based on total catalyst).

3. Synthesis Gas Purification

a) Coarse Purification

The converted water gas and/or cracked coked oven gas is treated over the "Grobreinigung" catalyst for removal of H2S. This catalyst was stated to be iron oxide (Luxmasse) and to have a life of about 12 weeks after which it was discarded. The gas velocity through the catalyst was stated to be 1 meter/second and air in the amount of 1.4% is added previous to the treatment. This air functions as an oxidizing agent in the subsequent fine purification step. A sample of the catalyst obtained at Rheinpreussen was analyzed by the Fuel Research Station with results as follows:

Moisture 49.5
Loss on ignition at 475o  5.4%

The ignited and air-dried catalyst showed the following analysis which is typical of Luxmasse:

Moisture 

3.4%

SiO2  

0.8

Fe203  

56.0

A1203  

27.4

CaO 

6.2

SO3  

1.9

CO2  

2.2

Difference 

2.1

100.0%

b) Fine Sulfur Purification (Feinreinigung)

The fine purification for the removal of organic sulfur takes place at 180oC, for fresh catalysts to 280oC for exhausted catalyst. A sample of this catalyst obtained at Rheinpreussen was analyzed by the Fuel Research Station with results as follows:

Moisture 

8.0

Fe203  

34.4

Na2CO3  

23.8

SO3  

3.5

Unidentified 

30.3

Documents indicate that the fine purification catalyst contains about 29% of Na2CO3. Apparently experiments on the regeneration of the fine purification catalyst with oxygen were conducted in 1941 from the records of which the following information was derived:

The fully spend mass showed as a typical analysis 33% Na2SO4, 0.3% Na2SO3, and 4% Na2CO3. This indicates that the reaction in Feinreinigung is essentially a catalytic oxidation and that 02 in the synthesis gas is essential, as other documents and data have indicated. Spent mass that had not been regenerated showed 16-30% Na2SO4, 0.4% Na2SO3, and 7-15% NasCO3. The above analyses indicate that the original mass contained about 30% Na2CO3, the remainder presumably being Fe203.

Apparently this catalyst was prepared by kneading together precipitated ferric oxide with a concentrated solution of sodium carbonate in the proportion to give a final catalyst containing approximately 30% sodium carbonate. The product is dried, broken and used as lumps.

B. Commercial Fischer-Tropsch Catalyst for Synthesis reaction

1. Cobalt Catalyst

The commercial Fischer-Tropsch cobalt catalyst is apparently prepared in the same manner as it was prior to the war. The composition of the commercial catalyst prepared at Sterkrade was 100L5L8L180-200 cobalt:thoria:magnesia:kieselfuhr. The catalyst prepared at Lutzkendorf was 100:5:10:170-230cobalt:thoria:magenesia:kieselguhr. The variation in density, kieselguhr proportion is due to variation in density, kieselguhr being added in such amount as to maintain a constant cobalt content of 80 grams/liter of unreduced catalyst. The difference in the catalyst composition between these two plants is not believed to be significant.

The magnesia produces a harder catalyst particle which is less susceptible to powdering. The substitution of magnesia for thoria apparently decreased paraffin production but otherwise did not influence the course of the reaction; although, apparently the induction period was shortened, methane and carbon formation tendencies of the catalyst were decreased, and the life increased to 8months at normal pressure.

The iron content of the kieselguhr must be low (less than 1%) or methane production is excessive during the synthesis. If alumina is in excess of 0.4% in the raw uncalcined state, the catalyst has a tendency to gell which causes manufacturing difficulties. Calcium content should also be as low as possible. Acid washing of kieselguhr was avoided by selecting deposits which met the above conditions and had a desirable physical structure. The kieselguhr was then calcined at 600-700oC to reduce the volatile matter to less than 1%.

The procedure for the cobalt catalyst preparation at Sterkrade was as follows:

750 1. of a solution of the nitrates in the desired proportions, 100 Co:5 ThO2:10 Mg0 (2 parts Mg0 are left unprecipitated) containing 40-41 gm. Of cobalt/1. are heated to 100oC in an overhead, stainless-steel tank. The contents are run into a precipitating tank, fitted with a direct-drive, twin-screw stirrer, and containing 750 1 of a solution containing 104 gm. Na2CO3/1 also maintained at 100oC. Apparently the same results may be obtained by adding the sodium carbonate solution to the nitrate solution, the principle requirement being that the precipitation be carried out as rapidly as possible. The mixture for Ĺ minute and the dry kieselguhr added through a hopper, and stirring continued for 1 minute. The slurry is pumped to a standard-type filter press, and the cake washed with distilled water until the wash-water is neutral as determined by the addition of nitrophenol to 100 ml, wash-water plus 5 ml. 1/10 NH2SO4. Catalyst equivalent to 64 kg of cobalt requires about 10 m3 wash-water and the washing time is about 14-15 minutes.

The washed cake is then dropped into a "masher" situated below the press and is mixed with dust from the screening plant. (65 kg. total Co gives 45 kg Co as dust). The resulting paste is then pumped to a rotating vacuum filter. The thin cake scraped off the filter drum, containing approximately 70% water, falls into the extruder, where rotating arms force the past through 3 mm, diameter holes whence it falls into the drying chamber. The dryer is a cylindrical vessel 7 m in diameter and comprises 20 superimposed stages 20 cm apart. The catalyst is swept round each stage by rotating arms and falls down from stage to stage during a period of 1-1/2 to 2 hours. Drying is effected by steam heat and an air blast.

From the final stage of the dryer the rough granules containing 10% moisture are carried by a conveyor to the vibrating screens, and separated into Ė over 3 mm, 1-3 mm (desired size), fines, and dust which is sucked away in an air current. The air stream is filtered through a cloth filter and then scrubbed with water and steam. The fine catalyst recovered from the scrubber is not returned to the mashed for inclusion in the final catalyst because oxidation tends to convert carbonate to oxide which renders reduction more difficult and because of dirt and extraneous dust are apt to become concentrated in this fraction. This material, which only represents 0.1 to 0.2% of the whole, is therefore sent to the catalyst regeneration plant and treated as spent catalyst. The fines from the screens and cloth filters are returned to the masher. The particles above 3 mm pass to a further set of screens, where rotating arms force the granules through the 3 mm screen. The dust and fines from this process are returned to the masher and the 1-3 mm grade is blended with the stream from the first screens, and is bagged for transport to the adjacent reduction plant via a telfer conveyor. The bulk density of the granules is 320-350 gm/1. The maximum daily out-put of the plant is equivalent to 4 metric tons of cobalt.

A sample of cobalt catalyst, which was stated to be fresh, was removed from a kubel at Rheinpreussen, was analyzed by the Fuel Research Station, and tested for activity. The sample was taken under a CO2 atmosphere but even with this precaution the sample was observed to be at a dull red temperature. The results of analysis and test were as follows:

%

 %

Moisture  1.0 CaO 

 0.9

Loss on ignition 4.0 MgO 

 0.9(0.9)

SiO2   47.9(47.5) Ni 

 0.1

Fe2O3   6.5  -S02

 nil

ThO2   1.8(1.8)  -C02  

0.2

A1203  1.9 unaccounted for 

 5.5

Co203  29.2(28.7)

 100.00

The material "unaccounted for" may be undetermined elements or may be due to false assumptions as to the state in which the known elements are present in the sample.

The amount of material insoluble in acid, 52.7% gives a minimum figure for the kieselguhr content, and 100-(C0203aTh02+Mg0) gives a maximum figure.

The composition of the catalyst expressed in the usual manner would therefore appear to lie between the limits shown:

Co  Th02   Mg0  Kieselguhr
100  8.8 4.4 260 to 330

The result is in complete disagreement with information obtained from Sterkrade, which was that the catalyst used in all Western German Fischer-Tropsch plants had the composition:

100 Co.  5 Th02, 8 Mg0, 180-220 Kieselguhr

The above analytical results should therefore be regarded as tentative.

The sample as obtained was completely inactive for synthesis at 185oC. and 195oC. After reducing with hydrogen at 400oC in the usual manner, the catalyst showed gas volume contractions of 10% and 20% respectively when evaluated at 185oC and 195oC.

The catalyst manufactured at Lutzkendorf was prepared as follows:

The spent catalyst, from which the wax had been extracted, was ground with wash water from the succeeding stage. This fresh water contained about 5 gm. Per liter of Co. It was then extracted with boiling nitric acid (50%) in vessels of about 30 M3 capacity, filtered on a filter press, and washed. The first wash water was mixed with the filtrate. The second wash water was used for grinding the incoming spent catalyst, as mentioned above. The third wash water was used for the first wash of the next batch. It was claimed that the loss of Co in the filter-pressing operation was less than 0.1%. Joswig claimed that they had achieved an overall loss on the factory of only 0.3% compared with 2% at Ruhland and 4-5% at Sterkrade.

The Co solution was then treated with soda at 60oC to a pH of 4-5 to precipitate Fe, A1, and Th. The sludge obtained by filtration (Thoriumshclamm) was sent to Ruhland, Sodium fluoride solution was then added at 20oC. to precipitate calcium fluoride. To ensure complete removal of calcium, an excess of sodium fluoride was added which precipitated part of the magnesium. The liquor was then filtered using kieselguhr as filter aid. The filter cake was washed twice, the first washing being used for the first wash of the next batch. The Co concentration was now 4045 gm per liter. This was sometimes strengthened with fresh Co solution added as makeup. When preparing fresh Co solution, the Co was dissolved in nitric acid and treated with a small amount of soda and filtered before mixing with the recovered Co solution.

The Mg and Th as nitrates were then added to the Co solution to obtain a ratio of 100 Co:10 MgO:5 ThO2. The nitrate solution was heated to 105oC and mixed in a 5-minute interval with an equal volume of sodium carbonate solution (100 gm./a.) at 106oC. After precipitation, kieselguhr was added during a 5-minute interval. The filtrate from the precipitated catalyst contained about 30 gm. Per liter of sodium nitrate. The filtrate was treated with soda and settled to recover any Co which had passed through the filters. After settling, the solution was evaporated in a triple effect evaporator to recover sodium nitrate and produce distilled water for the catalyst manufacture. In the second evaporation stage, some caustic soda was added to ensure complete precipitation of any residual Co which was filtered out before the third stage. Joswig stated than an appreciable amount of Co may remain in solution as bicarbonate.

The catalyst sludge from the presses was mixed with water and with the dust screened out from the dried catalyst. The mixture was then filtered on a rotary vacuum filter. The filter cake was put through an extruder with 6 mm holes and then passed to a Buttner turbo-drier. The catalyst was dried at 120oC. (5-15% water) and screened to give a product of 1-3 mm. The dust was mixed with the new precipitate, as mentioned above. The over-size particles were further screened ad gradually broken down to the required size. The capacity of the plant was rated at 100- catalyst chamber fillings/month or approximately 300 tons.

The catalyst prepared at Lutzkendorf was reduced in a manner similar to that carried out by Ruhrchemie:

The reduction vessel comprises a central compartment of square cross-section containing the catalyst in a bed 30-35 cm deep and 2.1 m2 in area, with top and bottom fittings in the form of truncated pyramids. A sheet-iron grill is placed on top of the catalyst bed and sinks into it to a depth of about 10 cm. This device serves to distribute the gas stream entering the top of the reduction vessel. One charge of catalyst weighs 200-250 kg and occupies 800 liters.

The reduction is effected by passing downwards through the catalyst bed a rapid stream of ammonia synthesis gas (3 H2 + 1 N2_ which is preheated to 460oC in tubular heater fired with coke-over gas. The effluent gas at 250oC is reheated to 300oC and the CO2 present (ca. 2 gm,/m3) is converted to methane by passage through a bed of synthesis catalyst in another reduction vessel. The gas is then cooled, dried by ammonia refrigeration, passed through silica gel to reduce the water content to less than 0.1 gm./m3 and recycled to the preheater for the reduction where it is mixed with fresh make-up gas.

The recycle and fresh gas enter the reducer at about 7,000 m3/hr (S.V. 8,800). The period of reduction varies from 40 to 60 minutes, depending on the exact gas velocity which varies according to the number of reduction vessels in use. The temperature is varied according to the time and gas velocity as shown below:

Time  Temperature at Inlet To Reducer Gas Velocity
40 min. 435oC. 8,000 m3/hr.
60 min. 428oC.  6,000 m3/hr.

The temperature is controlled to within +/- 2oC. After reduction is complete, the catalyst is cooled to room temperature in a stream of nitrogen, the nitrogen then displaced by CO2 and the contents of the reduction vessel discharged into a kubel by removing the top cover plate and inverting the vessel. The vessels were so balanced that this inversion could be accomplished by a hand-operated mechanism. A total of 16 reduction charges are required to fill one kubel. In the reduction process, 50-60% of the cobalt is reduced to metal. If over 60% is reduced, a less suitable catalyst results. The extent of the reduction is determined by measuring the volume of hydrogen evolved when the reduced catalyst is treated with acid. The exact reduction conditions were said to depend on the nature of the kieselguhr used in the catalyst preparation. Dense catalyst caused difficulties in the reduction.

In partial disagreement with the above, Pichler stated that the reduction of the catalyst is carried out at 365oC at as high a gas rate as possible to keep the water vapor above the catalyst to a minimum. Nickel will be reduced almost completely but with cobalt the reduction is not complete at 365oC. A temperature of 400oC is too high for a catalyst having a Co:kieselguhr ratio of 1:1, while the higher kieselguhr contents a temperature of 400oC and even higher can be employed. A reduction time of 4-5 hours is suitable for a hydrogen rate of 2 liters/g. of Co in the catalyst. When longer reduction times are employed, the formation of methane during the synthesis is excessive.

According to Alberts, the optimum conditions for cobalt catalyst reduction are very difficult to establish. Generally speaking, the lower the reduction temperature the better, but lower temperatures required

Lower temperature required longer times. A difference of 10oC in the temperature of reduction made an important difference in the activity and life of the catalyst in the over a difference observable during the first three days of operation.

The normal catalyst life varies from 4 to 11 months with that used in medium pressure synthesis being longer than for atmospheric pressure synthesis. The life depends on degrees of sulfur purification, smoothness of operation, and operating history. During this life the catalyst may be intermittently partly activated by dewaxing for 12-15 hours at 160oC with a 140-180oC boiling range naphtha. This solvent treatment was preferred to hydrogen treatment for it recovered more of the valuable waxes as well as saving hydrogen. The hydrogenation treatment was performed at 200-225oC for 8 hours at 1,000 to 2,000 cu meters of hydrogen/hr/oven. A recent development never attempted on full scale was a high temperature re-reducing treatment of the catalyst. This consisted of treating for three hours with 2,000 cu meters/hr/oven with pure hydrogen at 400oC. The object of this treatment was to remove resistant carbonaceous deposits.

2. Iron Catalyst

No iron catalysts have been used commercially; however, a catalyst plant was to have been constituted at Sterkrade to produce a precipitated iron catalyst for use in a project planned for Italy. The catalyst was to be the same or similar to the Ruhrchemie iron catalyst described below under experimental catalysts.

C. Experimental Fischer-Tropsch Catalysts

1. Cobalt Catalyst

So far as it is known, no new cobalt catalysts were developed although investigations were still continuing on the effect of magnesium oxide. In general, the opinion was that a ratio of 1 thoria to 1.6 magnesia was satisfactory and produced about as good a catalyst as those containing a higher proportion of thoria. Further increases in magnesium oxide content were being investigated but no definite conclusion had been reached.

2. Iron Catalysts

No iron catalysts were actually in commercial use for the synthesis reaction although two types were under development. The first of these types was being developed to replace cobalt in conventional commercial equipment in order to relieve the shortage of cobalt. This group of catalysts must, therefore, operate at a maximum temperature in the range of 215-225oC. The second type of iron catalysts are those which were operated in experimental processes under development and generally at higher temperatures.

(a) Iron catalysts tested at Schwarzheide, Ruhland.

The object of the series of experiments made at Braunkohlen-Benzin A.G. (Schwarzheide, Ruhland) in 1942-1943 was to enable a group of companies to arrive at the best possible catalyst for iron medium pressure synthesis. Six firms submitted their preferred iron catalysts which had been prepared by different treatment prior to the actual synthesis. The conditions set for these tests were as follows:

Highest permissible temperature of synthesis-225oC.
Pressure of Synthesis Ė 10 atm.
Catalyst volume of converter (all converters of the same type) Ė 4.8 liters.
Period of operation (uninterrupted) Ė 3 months.
All of these conditions were observed by the participants.

(1) Kaiser Wilhelm Institute Catalyst

The catalyst submitted by the Kaiser Wilhelm Institute was prepared by precipitation from a boiling dilute ferric nitrate solution (1 kg. of iron/30 liters of solution) with boiling soda solution (1 kg. of soda/8-10 liters of water) at 100oC. One percent of copper (based on the iron) was added to the iron solution before precipitation. The precipitate was filtered, washed free of alkali with hot distilled water, the precipitate re-pulped with distilled water, and a solution of potassium carbonate (containing 1% K2CO3 based on iron) mixed with the paste. The paste was thickened on a water bath and drying completed at 110oC. The dried pieces were broken into two to four mm granules. Ferrous iron gave a less desirable catalyst, s well as ammonia precipitation or kieselguhr additions. The catalyst was pretreated with a gas rich in hydrogen (CO:H2 = 1:2) for 24 hours at a pressure of 0.1 atm. and temperature of 325oC (8 liters of gas/hour/10 grams of iron). The strongly pyrophoric ferric catalyst was then soaked in paraffin to protect it from oxidation. It was charged to the converter with as little contact with air as possible.

(2) Ruhrchemie Iron Catalyst

The dense sintered iron catalyst used by other was objectionable to Ruhrchemie for two reasons:

The temperature of favorable action was too high to be satisfactory in the standard converters; because of the high steam pressure necessary for the control of temperature; and the catalyst charge to the converter was too heavy for the converters in use. It was desirable that no change of converter design be included at the time, so research was started to obtain an iron catalyst free from these two objectives. It was not expected that a catalyst superior to the cobalt catalyst would be discovered, but the shortage of cobalt made it necessary to attempt to equal the regular catalyst by taking advantage of the proved satisfactory action of iron.

The iron catalyst contains also copper, calcium oxide, and potassium as active components and kieselguhr as a support. The solution employed contains iron, copper, and calcium oxide in the proportions 100:5:10. Since the precipitation of the calcium is not complete, the finished catalyst has the following proportions: Iron:copper:calcium oxide:kieselguhr:100:5:8:30.

The raw materials for the preparation are as follows:

Iron turnings free from such metals as chromium, molybdenum, nickel, vanadium, etc., and clean of oil and dirt. Scrap copper metal in the form of clippings of sheets, wire, etc., is used. Copper oxide of corresponding purity can be used. Calcium carbonate, quicklime or hydrated lime are equally good. A light voluminous kieselguhr is best Ė calcined at 700oC. Potassium hydroxide and nitric acid are technical grade.

The preparation of the solution of mixed nitrates is carried out in an acid-resistant vessel provided with heating and cooling coils and with an effective stirrer. The components are dissolved in the nitric acid. The copper is dissolved first, then the iron, and lastly the lime. The calculation of the quantity of nitric acid to be added is made on the basis of:

2 Fe requires 8 HNO3
3 Cu requires 8 HNO3
CaO requires 2 HNO3

The dissolving of the copper is begun in the cold. As the metal dissolves with strong evolution of oxides of nitrogen, the temperature rises. By heating, the temperature is brought toward the end to 60 to 70oC. Next, the iron is dissolved by adding the turnings gradually. The rate of addition is determined by the rate of evolution of oxides of nitrogen. The temperature of the solution rises to about 80oC., and as the reaction slackens towards the end, is held at this level by heating. The calculated amount of calcium oxide in the form of calcium carbonate, quicklime or hydrated lime, is then added. This component should in all cases be finely divided. It must be added slowly in order to avoid local over-neutralization. The solution is next heated to boiling and held at this temperature for several hours before cooling to room temperature. Prepared in this way, the solution contains little free nitric acid. The solution is stable and no deposit forms even if it is boiled for a long time. The content of iron, copper, and calcium oxide is within the following ranges:

Iron  115.0-125 g. per liter
Copper  5.0-7.0 g. per liter
Calcium oxide  11.0-13.Og per liter

Total nitric acid is between 410-450 g. If the iron and lime were free of insoluble material, the solution can be used directly. Otherwise, the solution must be filtered, a difficult operation because of the slimy character of the usual solid residue. Filters, if used, must be of material not attacked by the solution. Before precipitation, the concentration of the iron is brought to 50 to 55 g. per liter.

Soda ash is dissolved to a concentration of 90 to 100 g. per liter. The solution is not filtered.

Precipitation is effected by introducing, as quickly as possible, the nitrate solution heated to 98oC., into the boiling soda ash solution while stirring intensively. The quantity of soda ash used is such that, at the end of the precipitation, the pH is 6.8 (determined with indicator strips). If necessary, either nitrate solution or soda ash solution is added to produce the desired pH. The contents of the vessel are then stirred for half a minute; evolution of carbon dioxide being over in this time. The calculated amount of kieselguhr is then stirred into the liquid. Total time of precipitation must not exceed five minutes.

For the purpose of separating rapidly the mother liquor from the catalyst, the suspension is filtered by using pumps of large capacity. Washing with hot condensate (70-80oC) is continued until the cake will yield a finished catalyst containing 0.4 to 0.6% of sodium nitrate, based on the iron content. More complete washing is unnecessary and in fact lowers the desired content of calcium oxide. In general, 200-220 cu. m. of wash water/ton of iron in the cake is sufficient to give the desired result.

Impregnation follows the washing. Potassium hydroxide is used to give the desired content of alkali. Repeated experiments have proven that impregnation in the filter by pumping through the cake a caustic solution of the proper concentration does not give uniform product on account of channeling of the solution through the filer cake. To obtain a satisfactory impregnation, it is therefore necessary to paste the case and to add the caustic to this mass. The quantity of potassium hydroxide is such that the filter cake (moist) contains 3.0-3.5% of potassium hydroxide, based on the iron. This result is in general reached if the mother liquor of the suspension has a concentration of 6.0 g. per liter. After impregnation, the catalyst is again filtered in the press (iron). The cake is dried at 110oC. and subsequently shaped. The catalyst is hard and resistant to abrasion.

Reduction with a mixture of hydrogen and nitrogen follows at a temperature of 300oC. (maximum) Higher temperatures cause over-reduction and an inactive catalyst. Time for reduction is usually 30 minutes after the temperature is reached. This statement applies when H2 and N2 are used in the proportion of 3:1. Times are much shorter when hydrogen only is used. Reduction is most satisfactory when the mass is disposed in layers of 25 cm. Depth and when the gas flow is high (about 2,000 cu. m. per hour per sq. m. of cross section). A content of more than 50% of iron (metallic) in the finished reduced catalyst must be avoided. Higher contents of metallic iron produce a less desirable catalyst. Further, the iron soluble in a 2% solution of acetic acid shall be 60-70% of the total iron. This determination is made by boiling the reduced catalyst for about 2 hours with reflux, using a protective inert gas.

After reduction, the catalyst is treated with nitrogen (cold) and then saturated with carbon dioxide. Contrary to the behavior of cobalt catalysts, much heat is liberated when the iron catalyst is saturated with carbon dioxide. The saturation must, therefore, be conducted slowly.

(3) Lurgi Iron Catalyst

The best iron catalyst developed by Lurgi has the following composition:

100 Fe, 25 Cu, 9 A1203, 2K20, 30 SiO2

The catalyst is prepared by dissolving the copper and aluminum nitrates in a 10% solution of ferric nitrate in such quantities as will give the specified ratio of metals. The solution is heated to boiling and a 10% solution of sodium carbonate is added rapidly at about 70oC. in such quantity as required to precipitate the metals as hydroxides. The kieselguhr is then added rapidly, stirred for about 1 minute, and the mixture washed to a pH of 8.0 after which it is washed with a potassium carbonate solution to incorporate the specified quantity of K2O. The product is dried in a centrifuge sufficiently to permit its extrusion and is further dried on a conveyor belt by a blast of hot air to facilitate cutting into desired lengths. Final drying is done at 100oC. Synthesis is started at about 180oC.. and the temperature is raised to 220oC in two days. Over an operating period of three monthhs, the temperature is raised from 220oC to 230oC. Such a catalyst has not been run to exhaustion by Lurgi, but it is believed that its life would be about one year.

The washing in the filter press with potassium carbonate added to incorporate the specification quantity of K2O was considered inadvisable by Ruhrchemie personnel because the filter cake cracks and allows channeling of the carbonate solution and results in an uneven deposit of the K2O. However, Lurgi did absorb 1% carbonate and proved that the catalyst was active.

(4) Rheinpreussen Iron Catalyst

No information is available at the present time on the actual method of preparation of this catalyst. However, its composition was supposedly as follows:

Component  Weight Percentages
Fe  38.00
Cu  2.40
CaO  19.25
(CaCO3) (34.40)
Carbon  12.00

(5) Brabag Iron Catalyst

No information is available on the preparation of this catalyst. However, its composition was supposedly as follows:

Component  Weight Percent
Fe  47.75
Cu  6.00
Zn  5.04
Paraffin  present

(6) I. G. Ludwigshafen Iron Catalyst

I am uncertain as to the extent I.G. catalyst submitted for testing, but I believe it was an iron copper-potassium precipitated catalyst or one of the sintered catalysts. No detail are known concerning the preparation of the precipitated catalyst containing copper, however, a catalyst prepared by precipitation followed by sintering had been developed by I.G. for use in tubular reactors.

b) Miscellaneous Iron Catalysts

 (1) Catalyst for Tubular Reactors (Michael)

 Synthesis iron catalysts produced by precipitation in the same manner as the cobalt catalyst, are unsatisfactory because a "run-way" reaction develops: a sudden temperature rise occurs in which the formation of methane and carbon predominates. The catalyst preparation was changed, therefore, in order to obtain a solid catalyst with good heat conductivity and high activity. In this method, the sintering was carried out before the reduction and the reduction itself performed at as low a temperature as possible while maintaining a high hydrogen velocity.

In order to attain high activity, 5-25% of alkali-earth metal in the form of oxide or carbonate, (MgO or McCO3 preferably) is incorporated into the catalyst. The iron oxide or hydroxide is mixed with the alkali earth and 1-2% of K2CO3 or K2B407, pressed into pellets, heated to 850oC. in a non-reducing atmosphere, and then reduced with hydrogen after cooling to 350-450oC. The catalyst has a very high activity at 250oC. and lower temperatures, and excellent thermal stability in the tube reactor.

This catalyst may have been submitted to Brabag, however, it was stated to operated at 20 atm. and 230-250oC, which is somewhat more drastic than the limiting conditions set by the Reich Ministry.

(2) Hot Gas Recycle Catalyst (Michael)

The best gasoline is obtained with iron catalysts when the synthesis is conducted above 300oC. Therefore, it is necessary that the catalyst be insensitive to such temperatures and a good conductor of heat. This type of catalyst is obtained by high-temperature sintering of iron powder, pasted with about 1% of alkali, during the reduction with hydrogen. An example of the catalyst preparation is as follows: Fine iron powder prepared by thermal decomposition of iron carbonyl is pasted with a concentrated solution of borax (1 gram of borax per 100 grams of iron) and formed into approximately 1 cm. Cubes. In order to insure loose packing, the cubes are made with slightly irregular edge lengths. The cubes are sintered and reduced with hydrogen at 800-850oC for 4 hours. The catalyst is cooled in a hydrogen atmosphere, which is replaced by carbon dioxide before charging to the reactor. In place of iron powder, Fe203 may be used in the preparation, but it yields a catalyst which is very porous and of relatively low stability.

(3) Foam Process Catalyst (Michael)

In contrast to the sintered catalyst, the reduction of the catalyst for the "Foam" process is effected at lower temperature (350-450oC) with a very high hydrogen throughput. Iron oxide is obtained in the finely divided state either by precipitation or by the decomposition of iron carbonyl. The oxide is stirred into a paste with alkali solution (K2CO3 or potassium borate) and formed into small granules which are dried, reduced, and ground fine in a ball mill in admixture with gas oil (250-300 kg. of iron/cu.m. of oil). Iron carbonyl can also be decomposed in the suspension oil, in which case, a smaller amount of iron suffices.

(4) Oil Circulation Process Catalyst (Duftschmid)

The same catalyst was used in this process as in the Hot Gas Recycle process.

(5) Synol Process Catalyst (Wenzel)

The catalyst is prepared by placing 19 kg. of iron from a wood charcoal preparation in a shallow water-cooled iron pan 50 cm in diameter and 15 cm high. Pure oxygen, supplied by two nozzles, is directed toward the mass which melts in 10 minutes. Sixteen kg. of aluminum nitrate, 4 kg. of potassium nitrate, and 2 liters 66o Be nitric acid is added along with 2-4 liters of water. The mixture is boiled until solid, and melting is continued from there on for 30 minutes. The melt is poured out on an iron pan and allowed to cool slowly. It is then broken up into granules (1-2 mm. Size). The dust is remelted electrically.

The following is the final analysis of the catalyst:

A1203  

2.5%

K2

0.2-0.6%

0.16%

0.03%

Fe304  

97%

The apparent catalyst density is 2.0.

The catalyst is reduced batchwise with pure hydrogen (less than 0.02 mg. of sulfur per m3). The pressure is substantially atmospheric and reduction temperature usually about 450oC. The reduction temperature does not appear to be critical; a range 380-650oC was stated to be satisfactory. The essential features of the catalyst reduction were said to be (a) a maintenance of high hydrogen rate corresponding to a minimum space velocity of 2,000 1/hour per liter of catalyst, and (b) efficient drying of the recycle and make up hydrogen.

The recycle hydrogen leaving the reduction oven contains 1-2 gms. water/m3> It passes through a water cooler at 20oC., an ammonia cooler at 4oC, and after mixing with fresh hydrogen, goes through a silica gel dryer.

The time required for reduction of a batch of catalyst is about 50 hours. Absence of water in the exit gas is used as a criterion of completion of reduction.

After reduction and cooling to 50oC., the catalyst was blanketed with CO2 and transferred to smaller steel containers. From these the catalyst, in a CO2 atmosphere, was measured into glass tubes which were used to charge the reaction tubes individually, so that exactly the same amount of catalyst was present in each tube, especially since the tubes were often only partially filled. This was to ensure equal pressure drop through each tube and therefore provide equivalent gas flow in each.

The used catalyst could be regenerated only by re-melting and re-reducing, after which it operated at a higher reaction temperature (225-250oC) and the product contained a smaller percentage of alcohols.

Except for the reduction part of the catalyst preparation, the catalyst employed for the Synol process is exactly the same as that used for the ammonia synthesis at Leuna.

(6) Ruhrchemie Alcohol Synthesis

The catalyst is a cerium or vanadium promoted iron catalyst which is prepared in a similar manner to the Ruhrchemie precipitated iron catalyst. The composition of the cerium catalyst was stated to be:

100 parts by weight

Iron

5 "

Copper

10 "

Cerium

50 " 

Kieselguhr

(7) K. W. I. Isosynthesis Catalyst

Thoria was found to be a particularly good catalyst. It can be improved by the addition of alumina particularly if the production of isobutene is desired. If the formation of liquid hydrocarbons is preferred, at the expense of the C4 fraction, it is necessary to add small proportions of alkali to the catalyst. In place of thoria, other metal oxides were used, such as zinc oxide and ceria.

Thorium oxide can be replaced partially by mixed catalysts. A combination of zinc oxide and alumina are, however, smaller for these catalysts than for thoria-based catalysts.

Thoria catalysts were usually precipitated by soda from nitrate solutions, with the practice of pouring rapidly the boiling soda solution into the boiling nitrate solution. In this way, for example, by using concentrations of 240 g. of thorium nitrate in two liters of water and two liters of soda solution (containing a small excess of soda), a hard granular catalyst is obtained after washing the precipitate free from alkali and drying at 110oC. (apparent density 1.3). If concentrated solutions are used, catalysts of low apparent density are produced that are not physically strong. Catalysts that are very hard and give a vitreous fracture are made if the mixing of the solutions is made slowly (e.g. in one hour). By precipitating with caustic soda or ammonia, such catalysts are produced even by rapid mixing (apparent density 2.3). The apparent density can be increased in each case by sintering in an air stream. A normal catalyst was raised in apparent density from 1.3 to 2.0 by heating in air at 300oC. The catalyst remains dull in appearance and hard.

Since a catalyst dried at 110oC shrinks during the synthesis, it is advantageous to perform the sintering before charging the catalyst. Up to 300o, shrinkage results from loss of water and carbon dioxide. At higher temperatures, firmly bound water is driven off without further shrinkage of the catalyst or lowering of its activity. On the contrary, a thorium catalyst heated to 1,000oC in air, showed an especially good converting power for synthesis gas. Since such a catalyst caused a somewhat increased production of methane, calcining was limited to 300oC.

For the preparation of mixed thoria-alumina catalyst, it was found to be advantageous to precipitate the oxides separately: thoria from the nitrate solution by soda, and alumina from sodium from solutions of thorium nitrate, with evolution of carbon dioxide. By heating at 240oC., ThO.CO3:2H2O is changed to 3Th02:ThO::CO3:H2O. This compound is then changed to 4ThO2:H2O at 300oC and at higher temperatures (above 400oC to ThO2.

Aluminate by acid; e.g. sulfuric acid. After washing, the two oxides are mixed thoroughly and dried. The preparation of a 20% alumina Ė 80% thoria catalyst is prepared as follows: Two hundred and forty grams of thorium nitrate dissolved in two liters of water are heated to boiling and precipitated (as explained) by boiling solution of 95 gm. of sodium carbonate in two liters of water. The filtered precipitate is washed fifteen times with 400 cc. of boiling water. Separately, 169 gm of aluminum nitrate are dissolved in a liter of water and mixed while boiling with 77 gm of sodium hydroxide first precipitated dissolves, the solution being only slightly turbid). Alumina is precipitated by adding to the boiling solution 17.2 cc. of concentrated sulfuric acid in 350 cc. of distilled water, settling, and decanting. It is then separated on a suction filter, and washed three times with 400-cc lots of boiling distilled water.

The two precipitates are mixed well and suspended in hot water, evaporated down on a water bath with constant stirring, and then dried at 110oC and calcined in air at 300oC.

Catalysts used in isosynthesis were regenerated from time to time by passing air over them at the temperature of the synthesis. The time between regeneration varies, depending upon the rate of formation of carbon (usually after several weeks). Activity was not impaired. The catalyst is not sensitive to sulfur compounds.

W. C. Schroeder. Thank you very much Dr. Horne. We have time for perhaps one or two questions. Are there any questions?

J. G. Allen. I would like to inquire as to the status of the reports mentioned. Dr. Horne didnít know if they had all come into the country. I believe they have all been brought back and should be in the process of reproduction.

L. L. Newman. Every one of the reports that was brought back to us has been copied. We made four copies of the, and the appropriate number of copies went out to the Technical Advisory Committee for duplication and to the Military for declassification.

J. G. Allen. These would have been reports that Dr. Faragher brought back on his return?

W.F. Faragher. I brought everything that we had when I came back. A supplement to the Ruhrchemie report, a couple of supplements to the Ludwigshafen report and the Leuna report had not gotten out of the section.

W. C. Schroeder. Presumably then, they are on the way and going through the mill.

W.F. Faragher. I believe they are.

J. G. Allen. I havenít seen them myself, although they may have been distributed. This is for my own information. I havenít seen them.

L. L. Newman. The Leuna report has been mimeographed that is being held up pending repair of the stapling machine. The report has been stenciled, assembled, and, I think, by the end of this week it will be completed.

V. Haensel. I wanted to ask Dr. Horne as question about this high space velocity reduction with hydrogen. As far as I can see from what I have been told, they use it to avoid a high water vapor concentration in contact with the catalyst. Do you actually mean high space velocity or high linear velocity?

W. A. Horne. High linear velocity as well as high space velocity. One meter per second linear velocity.

V. Haensel. They were emphasizing this fact. I wonder if any quantitiative or comparable data is available on that point, which I imagine is important, because theyíve been using the techniques for quite some time and it does appear to be very essential.

W. A. Horne. I havenís checked into the microfilms but feel sure there should be something on the reason for this practice.

H. V. Atwell. On the question that Dr. Haensel raised, the high space velocity in reduction was started by Ruhrchemie during 1944 and when we have time to analyze that data, we can probably obtain an answer. I havenít seen any statement summarizing the work which would indicate whether it is high linear velocity or high space velocity that is necessary.

W. A. Horne. I would like to fine out if it is only the reduction of water concentration over the catalyst or if there is some other reason?

V. Haensel. So far as I could find out, that was the only reason for the high linear velocity. Where there was particular damage to the catalyst, apparently they went to extremes by using refrigeration and silica gel for drying the hydrogen recycle. Apparently water vapor in the hydrogen during reduction is one of the principal factors that effect the catalystís inactivity.

H. V. Atwell. Somewhere in the microfilm documents there is a statement to the effect that Rheinpreussen developed an iron catalyst that would operate at atmospheric pressure around 200oC. This catalyst produced water instead of carbon dioxide which was a very unusual characteristic, but I never found any description of that catalyst and I wondered if anybody else had run across it.

W. A. Horne. I was unable to get to Rheinpreussen myself, however, I did hear something about an iron catalyst they were working on but could obtain no details.

H. V. Atwell. Iím rather embarrassed I asked that question. It was probably under our feet there at shaft I.

W. C. Schroeder. I wouldnít be surprised.

W. C. Schroeder. Dr. Horne, in view of the fact that your next subject will deal with the Hydrogenation Catalysts, I think we might bring that in right now and finish Hydrogenation. Will you go ahead.

W. A. Horne. I will review the Hydrogenation Catalysts in the order of their use in the process. Many of these have already been mentioned by Dr. Hirst.

A. Liquid or "Sumpf" Phase Catalysts

1. Tin Catalyst

For bituminous coal hydrogenation at 300 atmospheres acid catalyst conditions are employed. Tin oxalate (0.06 weight % on a.m.f coal) is added to the feed paste and ammonium chloride (0.7-1.0 weight % on a.m.f. coal) is injected in the first converter. Sodium carbonate is injected at the last converter outlet to prevent corrosion in the hot separator.

2. Bayermasse ĖIron Catalyst

Bituminous coal hydrogenation at 700 atmospheres employed alkaline catalyst conditions. A solution of hydrated iron sulfate (.1.2-1.7) weight $ on a.m.f. coal) is added to the coal before drying and prior to pasting 1.5-2.5 weight % of a.m.f. coal of Bayermasse (an iron-rich residue from aluminum manufacture containing 40-50% Fe) and 0.3 weight % of a.m.f. coal of sodium sulfide is added.

3. Molybdenum Ė Grude Catalyst

This catalyst for hydrogenation of heavy oils and tar had been replace by the iron-Grude catalyst on account of the shortage of molybdenum. It was prepared as follows: Grude containing 40-50% ash (obtained by carbonization of brown coal in a Winkler generator) is treated with sulfuric acid to effect 90% neutralization of its alkalinity. The neutralization is made necessary by the fact that alkali poisons the molybdenum catalyst. The Grude is impregnated with ammonia molybdate solution to give 2% MoO3 in the finished catalyst. It is then dried at 140oC. and ground to a fine powder (60% through 10,000 mesh). The catalyst is used as a 40% paste in asphalt free oil.

4. Iron-Grude Catalyst (Leuna No. 10927)

This catalyst had replaced the molybdenum-Grude catalyst. It was prepared as follows: The Grude is not pre-neutralized but is impregnated with hot concentrated solution of F3SO4.7 H2O in a screw conveyor. An equivalent amount of sodium hydroxide solution is added followed by drying to 18% water content. The concentration and amount of FeSO4.7 H2O solution is adjusted to yield a dried catalyst containing 5 weight % iron.

B. Saturation or Prehydrogenation Catalysts (I. G.)

The purpose of the saturation is to hydrogenate nitrogen and oxygen compounds Which damage the splitting catalyst used subsequently.

1. Catalyst 5058

It was prepared as follows: In a supply tank of about 2,000 liters capacity 500 kg. of tungstic acid (WO3H2O), alkali and chlorine free, that contains 92-93 percent of WO3 is dissolved at 60-70oC. in 1,500 liters of mother liquor. The solution is effected by stirring for about 1-1/2 hours. The mother liquor which contains 9-10% NH3, 10-12% H2S, and 2-3% WO3 as (NH4)2WS4 in solution, is obtained from a previous lot of catalyst and is first adjusted to a concentration of 13% NH3 by passing in gaseous ammonia. The solution is allowed to stand without stirring for 1 hour at 150 mm. Hg and 60-70oC to give a solution containing 27% WO3 then is pumped into the saturating vessel through a cloth filter (three quarters hr. to one hr.) where 3-4% WO3 is removed. The solution is saturated with H2S at 200 mm. Hg and thereafter the solution, which has been cooled to about 55oC is heated in an atmosphere of hydrogen sulfide with stirring to about 70oC (3-5 hrs). This charge is slowly cooled with stirring to 50oC, and then more rapidly to about 20oC (time cooling 6-7 hours). The precipitate of yellow salt (NH4)2WS4) is fed to the suction filter while stirred. Nitrogen under pressure of 0.5 atm. is used in the filter, followed by a current of nitrogen at 60oC to partially dry the cake, then dried in a stirred drier by nitrogen at 100-120oC. The mother liquor containing 4 gm WO3/100 cc is collected in a stirred storage vessel and is used subsequently as mentioned above. The decomposition of the yellow salt is carried out in a screw-conveyor furnace in a 100% excess stream of hydrogen at 400-430oC (about 1 hour). The black powder (73-74% W, 25-26% S, 0.1% H2O, 0.1-0.2%H2SO4) is cooled at the end of the furnace by a stream of nitrogen. The capacity of the furnace is 1.2 to 1/5 tons per day. The black powder is then ground in a hammer-mill until 70-80 percent passes through a 100-mesh screen (important not to grind too fine). Ten-mm. Pellets are then made in a Kilian press flushed with nitrogen. The sharp corners of the pellets are removed in a rotating screen drum, and the finished catalyst is packed in barrels that are flushed with nitrogen, since air oxidation results in a loss of activity which cannot be restored.

The used catalyst is crushed to pieces of about 2-5mm and calcined at 6-800oC. in air in a revolving oven that is heated externally. The pro-duct is ground and dissolved in the supply vessel in mother liquid that is then discharged into a settling vessel. Further operations are the same as in the method of preparation from tungstic acid. The crushing strength of the pellets is 250-300 kg. per sq. cm. The life of this 100% WS2 catalyst is normally about two years.

2. Catalyst 7846

This catalyst was first substituted for 5058 due to a shortage of tungsten And was later superseded by 8376. It was prepared as follows: The carrier is prepared from technical alumina containing 60% A1203 and 40% water. This alumina is dissolved in NaOH, and reprecipitated with nitric acid, the acid additions being made as quickly as possible at pH of 5.5-6.5 at a temperature not exceeding 50oC. The precipitate is filtered through a filter press and washed with water for a short time. After re-slurrying with water, the alumina is again filtered and washed with water until alkali-free. The filter cake is dried at 200oC and then powdered. After peptizing with about 2% HNO3, the paste is spread on aluminum trays, partially dried, cut into cubes of 1-2 cm. Sides, slowly dried, and then calcined at 400-500oC. The cubes are washed with 5% ammonia solution and again ignited. They are then soaked with an ammoniacal molybdic acid solution, dried, soaked with a solution of nickel carbonate in acetic acid, and finally ignited at 500-600oC. If sulfuric acid is used in place of nitric acid, NiSO4 is converted to NiS in the presence of H2 at 400oC. The final composition is 87% A1203, 10% MoO3, and 3% Ni203.

3. Catalyst 8376 (7846W250)

This catalyst was prepared as follows: Commercial aluminum sulfate (A12(SO4)318 H20) that contains about 18% A1203 is dissolved in water at 50-70oC. to an almost saturated solution (about 10% alumina in the solution). This solution is allowed to flow simultaneously with a 20% ammonia solution into a steam-jacketed stirred vessel to precipitate aluminum hydroxide. Solutions are so added to the vessel that there is always a small excess of ammonia in the mixture, which is controlled by an antimony electrode (pH between 8 and 10). Twenty cc. of the filtrate is equivalent to about 10 cc. of N/10 H2SO4. The precipitate is pumped from a storage vessel into the filter press and washed sulfate-free with weak ammonia solution (0.1%). The filter cake (about 19% solids) is dried in a drying over or a rotating-tube dryer until the ignition loss of the product is 15-20%. The dried product is then ground with 1% graphite and made into 10-mm. Pellets in a Kilian press. The pellets are calcined in an electrically heated vertical furnace in the presence of air at 450oC.

The calcined pellet (800 liters) are impregnated in a vessel that can be flooded with solution and can subsequently (i.e. after discharging the excess of the solution and draining) serve as a dryer. An ammoniacal solution of WO3 (20%) and NiSO4 is used several times with discharges of solution and drying between each operation. The number of stages is determined by the desired percentage of tungsten and nickel. The dry pellets are then sulfurized in an electrically-heated vertical furnace through which a mixture of hydrogen and hydrogen sulfide is passed at 400-450oC. The sulfurized pellets are ground and reformed in the Kilian press, and again sulfurized in the furnace. The final composition is 70% A1203, 27% WS2 and 3% NiS.

Very little of this catalyst has been regenerated. For regeneration of catalyst that has had normal use, the pellets are carefully roasted at 600oC., impregnated with 1% W03, and the corresponding quantity of nickel sulfate, and sulfurized to a sulfur content of 9-10% in a stream of hydrogen plus hydrogen sulfide. Small-scale experiments showed that the regenerated material had good activity. None, however, has been used on a commercial scale.

For the recovery of tungstic acid from catalysts that cannot be regenerated, the roasted product is ground and dissolved in concentrated sulfuric acid. The solution is diluted and filtered from the solid residue. Tungstic acid is dissolved in ammonia or ammonium sulfide solution and is used either for the preparation of 5058 or 8376. The aluminum sulfate solution can be used for the preparation of activated alumina. This recovery of tungstic acid has not been made commercially.

C. Splitting or Hydrocracking Catalyst (I. G.)

1. Catalyst 6434

This was the only completely satisfactory splitting catalyst used by the I.G. and was occasionally used directly on petroleum fractions and brown coal tar middle oil. It was prepared as follows: 300 kg. of Terrana A extra (Deggendorf) of proven activity is etched with 344 kg. of 8% hydrofluoric acid in a stirred pan for about 15-20 minutes at ordinary temperature. 500 liters of a 10% solution of yellow salt (ammonium thiotungstate) is added slowly, and the pan is heated with steam. In 8-10 hours, the charge is dry. During a further two hours, it is cooled. The cooled product is broken up in a special hammer-mill fitted with a 5-mm. Screen (Schlagkreuzmuhle) and is then decomposed in a screw-conveyor furnace at 400-420oC.in the presence of hydrogen and hydrogen sulfide. The discharge end of the furnace is cooled with nitrogen. The capacity of the furnace is 1-1/2 tons/day. The cooled product is again ground in a hammer-mill, passing out through a 1-mm. Screen. The ground product is wetted in an Eirich mixer, each 20 kg of product receiving 6.2 to 7 liters of water. The mass is then pressed through a 3-mm. Screen. This product is fed to the Kilian press, where it is made into 10-mm. Pellets (capacity 700 kg./day). The pellets are allowed to stand in the air for several hours, are tumbled in a screening apparatus and then dried in a drying over or an electrically-heated vertical furnace (up to 200oC.). The final operation is calcining at 450oC in a treating furnace in the presence of hydrogen and hydrogen sulfide. The finished catalyst is cooled with nitrogen and packed in drums under nitrogen. The final composition is 90% activated clay and 10% WS2. The used catalyst has been regenerated only in a few instances by very careful roasting at 550-600oC, impregnating with a solution of yellow salt (1% WS2) and sulfurizing at 450oC. The greater part of this used catalyst (several hundred tons) was worked up electrothermally into ferro-tungsten at Bitterfield. In this operation, the catalyst is roasted and then reduced electrothermally, alone or after admixture with ore.

D. Welheim Catalyst

This catalyst operates at 700 atmospheres in one stage to produce aviation gasoline, whereas the I.G., process is three stages, i.e. Saturation, Splitting, and D.H.D. reforming. This catalyst was prepared as follows: About 90 kg. of AD 5 paste (exact composition unknown, average water content is 60%; therefore, 36 kg. of dry material) is mixed for half hour in a kneading machine with 4.1 kg. of hydrofluoric acid (70-72%). After adding 10 kg. of Terrana, the kneader is operated for a further 10 minutes. The zinc oxide (3.7 kg.), flowers of sulfur (3.0 kg.) and chromic acid anhydride (2.3 kg), with an equal weight of water, are added and the mixing continued for 10 minutes more. A further portion of Terrana (10 kg) is added, and the mixture is neutralized with 9 kg. of technical ammonium hydroxide (sp. Gr. = 0.91). Finally, the ammonium thiomolybdate (1.4 kg.) dissolved in 7 kg of the ammonia solution is added, together with the remaining Terrana, and mixing is continued for a further with the remaining Terrana, and mixing is continued for a further 20 minutes. The finished mass is then made into cylindrical pills of 10 mm. Diameter, and dried in an oven at 70-75oC for four days.

Activation with dry hydrogen is effected by reduction according to the following schedule:

Increasing temperature to 350-360oC 8 hours

Maintained at 250-360oC 4 hours

Cooling to atmospheric temperature 10 hours

The catalyst is then placed in barrels. Contact with air does not decrease the activity but moisture does.

The above preparation is for catalyst K-534, which has a final analysis of 0.6% Mo, 2% Cr, and 5% Zn.

The same method is used to prepare catalysts K-413 and K-536, except that the molybdenum contents are o.4% and 0.7%, respectively.

The exact nature of the ADS paste was not known by Dr. Frese, except that it was an active clay. How it differs from Terrana is not known.

E. D. H. D. Catalysts

These catalysts are quite similar to the Hydroforming catalyst, consisting of molybdic oxide on activated alumina.

1. Catalyst 7935 (I. G.)

The activated alumina is prepared as described for catalyst 8376. The calcined pellets are treated in the combination impregnating the drying apparatus in batches of 800 liters with an ammoniacal solution of MoO3 (about 5% of ammonia and 12-15% MoO3) until the finished catalyst contains 15% MoO3. After drying in the apparatus at 190o, the catalyst is calcined in an electrically-heated vertical furnace in a stream of air heated to 400oC.

2. Catalyst 7360 (I. G.)

Activated alumina is prepared in the form of cubes by the aluminate process and the cubes treated in the same way as for catalyst 7935, until the finished catalyst contains 10-12% Mo3.

To regenerate catalysts 7360 and 7935, the used catalyst is roasted and ground. The powder is then moistened and dissolved in concentrated sulfuric acid; the solution is diluted and saturated with hydrogen sulfide. The impure precipitate of molybdenum sulfide is filtered, roasted, and dissolved in ammonium hydroxide.

W. C. Schroeder. Does that complete the paper?

W. A. Horne. Well, there are quite a few additional catalysts but they are not primarily hydrogenated catalysts.

W. C. Schroeder. Well, thank you, Dr. Horne, perhaps we can call for questions.

W. A. Horne. This catalyst information is contained in the Ludwigshafen report, a supplement to the Ludwigshafen report, the Leuna report, and a supplement to the Welheim report.

L. L. Hirst. Mr. Newman gave me 15 to 20 pages of report by you on catalysts preparation to be duplicated and as soon as we get the Leuna report out of the, we can start on that.

W. A. Horne. I assume it will come out under the TAC distribution, submitted for distribution by PAW and the Bureau of Mines to TAC. Is it for TAC distribution?

W. C. Schroeder. I want to get some of this on the record. Discussion by Horne, Atwell, and Hirst. Do I understand now that there is a report being prepared specifically on catalysts that is going to TAC and will be distributed through the regular TAC channels. Is this right?

L. L. Newman. Thatís correct.

W. A. Horne. There is a report covering hydrogenation catalysts produced at Ludwigshafen Oppau as a single report showing all the equipment, the sizes, power, and so forth.

W. C. Schroeder. I see.

W. A. Horne. The other catalysts, the ones investigated at Leuna which we didnít have time to make a thorough study of, were just given paragraph descriptions, extending to a page in some cases. These thirty to forty catalysts made at various places. I think the total numbers comes out somewhere about sixty or sixty-five. As I said previously, they are distributed in a number of different reports, and there has been no attempt to bring them all together.

W. C. Schroeder. Is there any more discussion on this hydrogenation process? I want to keep the discussion in order for the purpose of the record.

W.F. Faragher. One thing I want to point out that seem important to me. Dr. Hirst mentioned the increase in throughput of the Leuna plants which had to be operated at low pressure because of the physical conditions of the plants and heís made a careful study of levels and temperatures which have been very helpful, but the thing Dr. Herold emphasized most in their improvements was maintaining a falling temperature in the converters rather than the normal rise in temperature. In other words, they have the high temperature at the point where the asphalt destruction seemed to be desired. That was in his opinion a very important factor in bringing out about the increased throughput. Iím no worshipper of Germans and was quite happy to learn, when I went to Billingham, that the British had gotten increased throughput without a falling temperature gradient and had surpassed the Germans by some other means. I would wager that standards wouldnít suffer by comparison with the improvements in Germany. In other words, the Germans were still not supermen.

Don S. Fraser. While discussing catalysts, it might be appropriate at this time to point out to those that are not familiar with what weíre doing that we have arranged to send to the United States any collection of catalysts and oil samples. There seems to be a possibility that the analytical work on the catalysts might get sidetracked and when they do get back here will not be put to the use for which we collected them. Iíd like to know what is going to be done with those samples now that C.I.O.S. is almost disbanded. Who is going to take charge of them when they arrive here in Washington, or will be advised of their whereabouts if they come in at New York or some other port.

W. C. Schroeder. Some of those samples have already arrived and most of them are going to Pittsburgh, from where it will be distributed for examination and analysis. Most of those have been rather small size sample and one thing or another that came over. But about three weeks ago, a carbon sample arrived in Boston, and that is now on its way to Pittsburgh. Itís been on its way for about three weeks, and I suppose eventually it will arrive, but there hasnít been any notification of arrival yet. However, it will be handled and distributed by Pittsburgh. The only thing I can see at the moment is to have the Bureau of Mines undertake the responsibility of handling these samples and getting the analysis done. Now there has been a mechanism set up for distribution of samples for analysis and it will probably follow that original chart of idea that was laid out. I donít see any way to do the job except to handle it through Pittsburgh. Any suggestions that will free Dr. Storch of work will be very welcome.

E. B. Peck. On this analysis of catalysts, I would like to know if youíre going any further on this than just the chemical analysis, or will you use X-ray and other methods that have been recently developed for analyzing catalysts in this country.

W. C. Schroeder. Dr. Storch, will you reply to that? I think you had a hand in outlining that scheme for catalysts.

H. H. Storch. Well, I can tell you what is definitely known so far. We got a group of catalysts from the Navy, small samples about 100 cc. and those have been analyzed in the Bureau of Mines, chemical analysis, and their composition followed very closely their recorded compositions that were cited here today. There didnít seem to be any reason to do any more as it was such a small sample. Then we got a second batch of samples, most of which were equally small in size as the first batch, and a good many of them were duplicates of these Navy samples. We donít propose to do much with those except in cases where the samples were not duplicates of the Navy samples. Now in this second batch, we also got a keg about 1 to 200 pounds of DHD catalyst. That will be forwarded to the company that was agreed upon in the TAC meetings for proper evaluation and tests. There are also a few small samples of miscellaneous materials that will be sorted out and sent to the proper companies. Now we have one lot of material which contains something in the order of a ton of unreduced cobalt Fischer-Tropsch catalyst, that will be treated according to the TAC outline. These consist of several drums of these hydrogenated cabinets and there is where the question that Dr. Peck asked comes into the picture. Thereís enough catalyst there to run some service tests and sufficient for any extensive physical examination that may be desired. Those will be routed to the Standard Oil Development Company and other companies that are interested. The extent of the tests will depend very largely on the judgment of people like Standard Oil Development Company as to how far it will go.

W. A. Horne. I think for the clarification of everybody, that first samples that have been mentioned were principally samples picked up in the field by the early investigators who came in from London and sat around until finally we got them all shipped back to the United States, whereas, the last batch of samples, the large quantities (which I think is nearer three tons than one ton) of cobalt, plus 500 to 1,000 liters of several other hydrogenation catalysts, were collected along with all the oil samples we could get our hands on Ė a total of 30 tons of material from three major expeditions after samples. Those were handled through the Ordnance Department, who took care of this, going to several consignees. One portion went direct to the Bureau of Mines as originally scheduled; second group was addressed to T.I.I.C. in Washington, which I believe are also now being consigned to the Bureau of Mines for handling; third group went to Ordnance direct, and Ordnance representatives specifically requested and helped gather the samples; fourth group went into the British to be distributed to various organizations in Great Britain through the Fuel Research Station in Greenwich.

Now I think the presence of these samples should be advertised specifically so that any one who is interested in them will know they are at a certain place and can officer to do work on them. I donít know what the original plans were, I donít know what the original plans were, I may be crossing the trail now. You can plan to get samples and when you get there maybe you get what you planned, maybe you donít; maybe you find some others you never heard about that are still of interest. So my suggestion would be that as these samples come in that they be advertised either through TAC chancels of Bureau of Mines, or whatever the proper agency is. Let everybody know theyíre here, and then see who is interested in what sample. Make it entirely fair for all people concerned. Iíd like to have some comments on that.

W. C. Schroeder. I think we should follow the procedure which had been developed by Dr. Delbridge and Mr. Snodgrass and a group here in Washington while the team was in Europe. They selected certain companies for doing certain types of analysis. I think those samples should first be distributed to those companies and that analysis secured. The results should be disseminated through TAC or through any other appropriate agency. If there are any samples remaining, I think we will keep them on file and allow anybody to write in and request such samples as he may want. I feel we should go through this orderly procedure first to insure the fact that weíve got complete results before we start scattering the samples around and broadcast because there are a number of people who want quantities of those and I think weíll hold back on that until weíre sure weíve got our job done.

H. V. Atwell. Well, I agree with the plan of saving samples until the specified analyses is finished, I donít see any harm in publishing a list of whatís here already. It will give the men a chance to check on samples they picked up and may have disappeared.

W. C. Schroeder. Havenít you got a list through TAC already?

H. V. Atwell. I think itís very important that we all know whatís in the country, how much, what itís description is, and what the disposition is.

J. G. Allen. Well, Iím reasonably sure that no such list has been made up as yet because I talked to Dr. Delbridge at the A.P.I. last month and he didnít know that these last samples were coming in, and when they were coming in.

W. C. Schroeder. There is a list of samples that had been picked up. I donít know as it specifies the quantities, but apparently as Mr. Allen has it right here, and I think that list can be disseminated and weíll see that it is soon, and sent out to the group.

W. A. Horne. Mr. Newman told me in Pittsburgh that all the Welheim samples were being sent to Ordnance, and I as one of the individuals that picked up those samples, would hate like the devil to see them become lost in Ordnance with men retiring from the service and the responsibility of those samples shifting from man to man, and they might never get looked at.

W. C. Schroeder. Well, Ordnance hasnít received anything so far, for they think the Bureau of Mines has their samples. I donít know exactly what is happening to them, but weíll keep our eyes open and see if we can get them all into Pittsburgh.

W. A. Horne. These were marked plainly from Paris where they were going, and very specifically each package was marked and each bill of lading should be with it, and could be checked very easily.

W. C. Schroeder. I suspect what is happening is that the Navy seems to have taken over the shipment of these things, and I donít know exactly what happened but I rather imagine theyíve been put on a boat and brought over and thrown all into cars. Theyíre going to send the whole works to Pittsburgh as the easiest means of getting rid of them. From there on, if they are plainly labeled weíll see what we can do with them.

J. G. Allen. Well we divided them into four groups and bills of lading were all made out and the first boatload left there about the middle of September-left LeHavre. One set was obtained by the Navy.

W. C. Schroeder. Well, if the Navyís got a set, I think weíll be able to get it. Thereís a good deal of cooperation there.

*************

W. C. Schroeder. We really are getting seriously behind schedule. I want to put this thing back into some sort of order.

The next discussion is on "Lubricants, at Schkopau, Leuna, and Ruhrchemie," by Hans Schindler.

1-Alkali metal carbonates precipitate basic salts