GB2052463A - Chlorination of Aluminous Material at Superatmospheric Pressure - Google Patents

Abstract

Aluminous material is chlorinated in the presence of a reducing agent at pressures greater than three atmospheres. A greater than linear increase in reaction rate obtained at superatmospheric pressures is a feature of the chlorination reaction which is employed to increase aluminum chloride production rate. Choice of pressure used is governed more by mechanical limitations than by an upper limit inherent to the chemical reaction. Alpha-alumina is preferably chlorinated to form satisfactory yields of aluminum chloride when pressures in excess of 10 atmospheres are used.

Description

SPECIFICATION Chlorination of Aluminous Material at Superatmospheric Pressure This invention relates to an improved method for chlorinating aluminous oxygen-containing material in the presence of a reducing agent.

Chlorination of aluminous, oxygen-containing materials in the presence of a reducing agent, such as carbon, to produce aluminum chloride is well known. Various methods have been devised to improve the extent or quality of chlorination, but possible use of superatmospheric pressures is one approach that has not been explored to any great extent until the present. Pressure has been used to condense aluminum chloride vapors to promote melting of aluminum chloride (French Patent 334,132), to force chlorine gas through molten aluminum (Brooks U.S. Patent 1,165,065) and to form briquettes from coking coal, and alumina or bauxite (McAfee U.S. 1,217,471).It was not recognized, however, until the present invention, that chlorinating an aluminous material in a reactor maintained at pressures greater than three atmospheres would have the advantages of both increasing the mass capacity of equipment by the factor F, reactor pressure [psig]+14.7 [psia] (where F= 14.7 [psia] and increasing the degree of approach to complete gas conversion at constant gas-solids contact time.

One explanation for this failure to seriously consider use of pressures greater than three atmospheres is that thermodynamic studies conducted at atmospheric pressure indicated that the reaction rate dependence on chlorine pressure was less than first order. Since a reaction which is less than first order with respect to chlorine pressure would be expected to yield a product at a rate-to-pressure ratio which diminishes with increasing pressure, those skilled in the art did not consider use of superatmospheric pressures as a worthwhile means for improving aluminum chloride production.

Russell et al U.S. Patent 3,842,1 63 is one example which illustrates that some in the art who had considered use of superatmospheric pressures did not believe that increased reactor pressure would improve chlorination efficiency. While stating that pressures between 0.1 and 10 atmospheres may be used, the Russell et al Patent indicates that chlorination is generally carried out at about one atmosphere, with pressures of 1-3 atmospheres described as preferred. While Russell et al stated that high pressures permit a greater throughput which normally offsets any decrease in efficiency, operation at pressures higher than three atmospheres was not considered a sufficiently attractive procedure such that we have been able to find reports of actual trials in the literature.

Atherholt (U.S. Patent 2,048,987) used pressures of two to three atmospheres to promote the reaction's approach to equilibrium and thereby cause a greater amount of aluminum chloride per pound of chlorine to be formed. Atherholt failed to realize, however, that by increasing mass flow through a reactor maintained at pressures greater than three atmospheres, the rate-to-pressure ratio would increase.

An embodiment of the invention relates to the production of aluminum chloride by the chlorination of alpha-alumina under high pressure conditions.

Alumina, or aluminum oxide, exists in various forms or phases. For example, the product of the Bayer process is Al(OH)3, commonly known as alumina trihydrate. To avoid the formation of undesirable amounts of HCI during chlorination, such hydrated forms of alumina are normally dehydrated or calcined at elevated temperatures to remove a substantial portion of the water therefrom. However, in the subsequent atmospheric chlorination of such dehydrated or calcined aluminas, it has been discovered that the most stable alpha-alumina phase has very little reactivity in the chlorination reaction. Thus, British Patent 668,620 cautions against using calcination temperatures above 10000 C. to avoid or inhibit production of alpha-aluminum oxide.The patentees state that the employment of aluminas produced below 100000. " ... . renders it possible to obtain considerably higher yields in respect of chlorine and carbon monoxide than is possible with alpha-aluminum oxide." The above mentioned Russell et al U.S. Patent 3,842,1 63 describes and claims a process for the production of aluminum chloride which involves the chlorination in a fluidized bed of an alumina having a carbonaceous reducing agent intimately associated therewith by the prior cracking and coking of a liquid hydrocarbon in the presence of alumina particles. Throughout the patent, however, the patentees stress the use of alumina having an alpha-alumina content of less than 5% by weight.The lack of reactivity on the part of alpha-alumina becomes of particular importance when a continuous process rather than a batch process is used because of the gradual buildup of unreactive alpha-alumina in the reactor which eventually necessitates shutdown of the reactor and removal of the nonreactive alphaalumina therefrom.

It is also known to chlorinate an aluminous material in a molten bath using a source of chlorine with a liquid or solid reducing agent. Such a process is described and claimed in Haupin et al U.S.

Patent 4,039,648. In the aforesaid molten bath process various types of alumina including alphaalumina can be chlorinated.

However, it would be of value to have a fluidized bed (non-moll:en) process or a modification of an existing fluidized bed process whereby unreactive alpha-alumina could be chlorinated thus eliminating the need for shutdown of fluidized bed reactors, and removal and disposal of accumuiated amounts of unreactive alpha-alumina.

In accordance with the present invention there is provided a method for chlorinating aluminous material containing oxygen comprising chlorinating the aluminous material in the presence of a reducing agent at a pressure of greater than three atmospheres.

As used herein: Metal-grade alumina (MGA) is an alumina suited for the production of aluminum by the Hall Process, which has an alpha-alumina content greater than 10%.

Partially calcined alumina (PCA) is alumina having greater than 99% gamma or other transitional phase alumina and an alpha-alumina content of less than 1%.

B.E.T. refers to a method for measuring surface area which is described in Brunauer et al, 60 Journal of American Chemical Society, 309-19 (1938).

Kg.-mol. is a unit of measure which, when multiplied by the molecular weight of a compound, gives the weight of the compound in kilograms.

The present process includes the feature of maintaining a reactor pressure greater than three atmospheres. This superatmospheric pressure increases the aluminum chloride production rate coefficient to an unexpected degree. The aluminum chloride production rate coefficient is analogous to a mass transfer coefficient and differs from a mass transfer coefficient only in that the aluminum chloride production rate is measured in terms of volume rather than area.

In the accompanying drawings: Figure 1 is a comparative graph showing the effect of pressure on the aluminum chloride production rate coefficient for a coked alumina system, a PCA-coke mixture, and a MGA-coke mixture in a continuous chlorination process. The aluminum chloride production rate coefficient in kg.-mols of AlClhr. m3. atm is plotted against the pressure in atomspheres; Figure 2 is a plot which shows the effect of pressure on the average aluminum chloride production rate coefficient in a batch high pressure reactor; Figure 3 shows a reactor system for chlorinating aluminous material at superatmospheric pressures; Figure 4 is a flow sheet illustrating the process of the invention.

In accordance with the invention, an aluminous oxygen-containing material is chlorinated at a pressure greater than three atmospheres in the presence of a reducing agent with the preferred range being from 5 to 15 atmospheres, and the most preferred range being 7 to 1 5 atmospheres. The aluminous material may be a refined alumina, such as that from the Bayer process, a coked alumina, such as that described and claimed in Russell et al U.S. 3,842,1 63, or a raw material, such as bauxite or clay used in conjunction with a separated reducing agent. Appropriate reducing agents include sulfur, coke, finely divided carbon, carbon monoxide, COCI2 and carbon tetrachloride.

In accordance with said embodiment of the invention a source of aluminous material containing alpha-alumina is chlorinated at a pressure greater than 10 atmospheres in the presence of a carbonaceous reducing agent in a fluidized bed. The use of the term "fluidized bed" is not intended to be limited to any special degree of fluidization but merely to distinguish from molten bath reactions.

While the source of aluminous material may vary greatly, it is most likely that refined materials, i.e., alumina from the Bayer process, will be the starting material since such materials are usually subjected to dehydration by calcinating at an elevated temperature which can result in the production of alphaalumina if the temperature is high enough. Most likely, the source of aluminous material will be from a chlorination process such as described in the aforementioned Russell et al U.S. Patent 3,842,1 63. As described earlier, such low pressure chlorination reactions result, when operated in continuous fashion, in the accumulation of unreactive alpha-alumina which must be disposed of in some manner.In accordance with the embodiment of the present invention, this may be conveniently accomplished either by transferring the unreactive alpha-alumina to a pressure reactor, or in designing the reactor conventionally operated at atmospheric pressure so that it may, alternatively, be utilized at high pressures, i.e., pressures above 10 atmospheres.

The source of aluminous material containing alpha-alumina may be a coked alumina such as described and claimed in the aforesaid Russell et al U.S. Patent. Alternatively, the source of aluminous material may be a refined alumina, or a raw material such as bauxite or clay used in conjunction with a separate reducing agent such as coke, finely divided carbon, or gaseous reducing agents such as CO, COCI2 and Cm14.

The amount of alpha-alumina in the source of aluminous material may vary considerably.

However, in a continuous process, even very small amounts of alpha-alumina in the aluminous material could eventually build up if atmospheric pressure was used, i.e., amounts as small as 0.1 to 0.5 wt.%.

Thus, in accordance with the invention, the amount of alpha-alumina in the feed material may vary from 0.1 to 100 wit%. Thus, when the chlorination reaction is carried out, in accordance with the invention, at a pressure greater than 10 atmospheres, the differences in the mass transfer coefficient or production rate for AIC13 as a function of calcination temperature of the source of aluminous material fed to the chlorination reactor is not sufficient to interfere with satisfactory production.

The chlorination may be carried out under batch conditions or in a continuous manner.

In accordance with said embodiment of the invention, a pressure vessel capable of withstanding pressures in excess of 10 atmospheres up to as high as 20 atmospheres must be used. Preferably, the reaction is carried out at about 1 5 atmospheres in a pressure vessel having an outer shell capable of withstanding such pressures and lined with materials capable of withstanding high temperature corrosion from the reactives.

The reaction pressure, in accordance with said embodiment of the invention should be greater than 10 atmospheres up to about 20 atmospheres. Preferably, the pressure range is about 12 to 20 atmospheres and most preferably about 1 5 atmospheres.

The reaction is carried out at a temperature of about 550 to 800cm. The temperature may be varied somewhat, inversely with the pressure so that lower temperatures may be employed as the pressure is raised. Preferably a temperature of about 700cm. is used at a pressure of 1 5 atmospheres.

The chlorination can be carried out under either batch conditions or preferably operated in a continuous manner. When the chlorination is carried out in a continuous manner, suitable equipment well known to those skilled in the art is used to feed the reactants in under sufficient pressure.

Thus, the chlorination reaction may be carried out using any suitable apparatus which is capable of withstanding the superatmospheric pressures of this invention. The reactor system used to develop the data shown in the examples below will be described to illustrate one possible system.

The fluidized bed reactor 6 shown in Figure 3 was in an electric furnace which contained a chlorine-resistant metal (e.g., the alloy nominally containing 80% Ni, 1 5% Cr and 5% Fe, and sold under the Trademark Inconel) shell lined with graphite. The fluidizing grid was a 0.64 cm. thick mullite disc with three 0.32 cm. diameter holes for distributing the chlorine gas, and a hole in the center for inserting a thermocouple into the bed. Lock chambers were pressurized with nitrogen to introduce solid feed into the reactor through line 10, to return dust through line 11, and to discharge solid residue from the bed to solid residue tank 7 through line 12. A small, continuous discharge of bed solids was overflowed to insure a constant bed depth.A quartz baffle was inserted in the graphite liner between the feed and discharge arms to prevent short circuiting of the solid feed out the discharge arm. A nitrogen purge was added at the liner bottom to prevent leaking of chlorine behind the graphite liner. This nitrogen purge represented a slight but negligible dilution of the chlorine feed gas. The solid feed rate was controlled by a rotary vane feeder 1.A programmable controller regulated the opening and closing of the lock chamber valves on a one-minute cycle.

The chlorine gas flow was controlled by plug valve 8. The chlorine mass flow was measured by a differential pressure transmitter. The mass flow signal was corrected for pressure and temperature deviations.

The reactor was contained in a five-zone electric furnace with the temperature in each zone being controlled separately. The bed temperature was measured by an internal thermocouple extending up through the center hole in the fluidizing grid to 1 5 cm. above the fluidizing grid. The furnace zone temperatures were adjusted manually to maintain the desired reaction conditions.

Dust was filtered from the reactor product gas in two dust collectors 2 connected in parallel. The two collectors were both equipped with stone filters and hot nitrogen blowbacks for filter cleaning.

A sample stream 1 3 of the filtered process gas was bled off directly downstream of the dust collectors. The sample gas was passed through a quartz wool filter to remove final dust traces and then cooled to 50 C. in a water jacketed heat exchanger to condense and remove the metal chlorides before entering a gas chromatograph 9 for analysis of CO, CO2, Cl2, COCI2 and N2.

Electric heaters were installed on the offgas piping, dust collectors, gas sample line, and reactor sidearms to prevent condensation and plugging by metal chlorides. All offgas piping was Inconel metal pipe, including the gas sample line. All process valves were nickel ball valves with either metal or polytetrafluoroethylene seats depending upon the temperature service. Pressure taps were located throughout the reactor, offgas lines and dust collectors to measure pressure drops. All taps were purged with nitrogen to prevent plugging.

Reactor pressure was measured at the chlorine gas inlet 4 at the bottom of the reactor. The pressure control valve was positioned between the dust collectors and the de-sublimer.

A desublimer 3, such as that disclosed in U.S. Patent 3,930,800 of January 6, 1976, with a water-cooled heat exchanger and fluidized with air was used to condense and collect the aluminum chloride product. The desublimer offgas was sent to a caustic scrubber installation 5 for final treatment.

The runs described in Examples I through XVIII correspond to four to six hours of steady-state operation with continuous feeding and total dust return to the reactor. A constant bed depth of 70 cm.

was maintained, as mentioned above, by discharging a small amount of reactor bed solids (1-2 kg./hr.) out of the reactor side arm discharge. The grid and fluidized bed pressure drops were typically 6 and 60 cm. WC, respectively.

Examples 1-IX Coked alumina having a carbon content of 18 wt.%,0.27 wt.% hydrogen, 0.6 wt.% alpha alumina and surface area of 8 mg. (B.E.T.), was charged to the above-described reactor and chlorinated.

Sodium chloride was added to the reactor with the alumina feed to react with some of the aluminum chloride product and thereby insure a high level of NaAICI4 catalyst in the chlorination reactor. The process parameters and results which are seen in Table 1 are visually represented in Figure 1.

Table 1 Coked Alumina Examples Superficial Al203 A10I3 Production Temp. Pres. Velocity Rate Rate Coefficient Expt. No. (0C.) (atm.) (cm./sec.) (kg./hr.) (kg.-mol/hr. m3. atm.) 635 4.6 12.1 6.3 13.6 II 625 1.4 13.8 1.7 8.4 lil 625 1.6 16.1 2.1 8.3 IV 625 3.4 6.5 1.8 7.5 V 625 3.3 13.2 6.2 10.7 VI 700 3.4 6.1 2.6 12.6 VII 625 4.4 8.3 3.1 10.5 VIII 700 4.6 6.2 2.2 10.1 IX 700 4.7 11.6 5.6 13.2 Calculation of the aluminum chloride production rate coefficient shown in the last column of Table 1 will be demonstrated using the data of Example The pertinent experimental conditions and data of Example I are as follows: Reactor Pressure 4.6 atm.

Bed Depth 0.7 m.

Bed Diameter 0.0825 m.

Cl2 Feed Rats 9.8 kg./hr.

N2 Feed Rate 0.14 kg./hr.

Solids Feed Rate 6.3 kg./hr.

H2 Content of Solid Feed 0.27% Reactor Temperature 635 C.

Reactor Offgas Analysis CO2 11.2% CO 0.0% Cl2 1.19/0 COCI2 0.8% The AICI3 volume per cent concentration in the reactor product gas was calculated from the offgas analysis: AICI3=4/3 CO2+2/3 (CO+COC12) AICI3=1 5.5% The product gas concentrations (without the N2 content from pressure taps, filter blowbacks and purges) were normalized to 100%.

Actual Vol. % Normalized Vol. % CO2 11.2 39.2 CO 0.0 0.0 Cl2 1.1 3.8 LOCl2 0.8 2.8 AICI3 15.5 54.2 100.0% The mols of product gas were calculated from the above normalized percentages and a Cl2 mass balance.

Mols Cl2 in product gas=Mols Cm, fed (Cl2+COCI2+(1.5 AICI3)) (mols of gas)/1 00=(0l3 rate)/70.9 Mols of product gas=0.1 5725 kg.-mols/hr.

The N2 dilution and HCI generation were next added to the above to determine the total mols of reactor offgas.

kg N2 dilution=(0.14-)/28=O.005 kg.-m./hr.

hr To determine HCI generation, it is assumed that the entire hydrogen content of solid feed reacts to form HCI.

HCI generated=0.0027x6.3 =0.01701 kg.-m./hr.

Total mols reactor offgas=0.15725+0.005+0.01701 =0.17926 kg.-m./hr.

The product gas concentrations including the N2 and HCI values were renormalized to 100%.

Normalized Vol. % CO2 34.4 CO 0.0 Cl2 3.3 COCI2 2.5 AIC13 47.5 HCI 9.5 N2 2.8 100.0% The normalized product gas concentrations, reactor pressure and reactor temperature were input into a computer program which estimates equilibrium partial pressures of the various gases based on thermodynamic values.

Normalized Partial Pressures at Actual Vol. % Temperature & Pressure (atm.) CO2 34.4 1.75 CO 0.0 0.120 C12 3.3 0.292 COCI2 2.5 0.00512 AICI3 47.5 1.19 Al2CI6 - 0.615 0.615 HCI 9.5 0.483 N2 2.8 0.142 100.0% 4.6 atm.

The log mean pressure of the chlorinating gas was calculated.

P,-P, Pim= P, ln( P0 P, (N2 corrected)=(CI2 rate/70.9)/((Cl2 rate/70.9)+(N2 rate/28))Rx pressure P,=4.44 atm.

Po=Pcl2+pcocl2 =0.292+0.00512 =0.29712 atm.

4.44-0.29712 Pirn =1.532 atm.

4.44 0.29712 The AICI3 production rate was calculated from the Cl2 rate, and the Cl2 conversion was corrected for HCI generation.

Cl2 rate-(Solid Feed Rate) (%H/100) (35.45) AICI3 rate 70.9 (Cl2 conversion) (PAICI3+2PAI2CI6) 1 .5PAICI3+3PAI2CI6 AICI3 rate=0.079906 kg.-m./hr.

The bed volume for the pilot reactor was determined: n(0.082 5)2 Bed Volume= xO.7 4 Bed Volume=0.003742 m3 The aluminum chloride production rate coefficient (PRC) was then calculated: PRC=(AICI3 rate)/(Bed VolumexPIm) PRC=1 3.9 kg./mol. AlClhr.m3. atm.

The significance of the aluminum chloride production rate coefficient may best be appreciated in terms of chlorine pressure and a simple rate equation. Where the pressure effect of chlorine is less than first order, ti.e., rate=k[Pci2] < 1), as is the case with chlorination of alumina at atmospheric pressure, the aluminum chloride production rate coefficient will decrease with increasing pressure. If the reaction rate is first order with respect to chlorine pressure, (i.e., rate=k[Pct2]'), the aluminum chloride production rate coefficient will remain constant with increasing pressures.When reaction rate dependence is greater than first order, (i.e., rate=k[Pci,l > l), as is the case in chlorination of alumina at superatmospheric pressures, the aluminum chloride production rate coefficient will increase with increasing pressure.

The aluminum chloride production rate coefficient was affected by fluidizing velocity, i.e., higher production rate coefficients are realized at higher fluidizing velocities. Apparently more efficient fluidization was achieved in the experiments performed at velocities of 12.1-13.8 cm./sec. than those run at 6.1-8.3 cm/sec. The fact that the aluminum chloride production rate coefficient was still significantly higher (in all but one case) than that obtained at atmospheric pressure despite lower fluidizing velocities clearly indicates that increased pressures (i.e., pressures greater than three atmospheres) significantly affect the rate of aluminum chloride production as indicated by the aluminum chloride production rate coefficient.

Examples X-XlV A mixture of PCA (79.4 wt.%), petroleum coke (19.9 wt%), and sodium chloride (0.7 wt.%) was chlorinated in the reactor system seen in Figure 3. The coke had been calcined at 8250C. for 30 minutes and sized to -210+105 microns (-65+150 mesh). The coke had a carbon content of 96.8 wit.%, 0.79 wt.% hydrogen, 0.52 wt.% nitrogen, 0.99 wt.% sulfur, 0.25 wt.% ash and surface area of 10 m2/g.The alumina had the following analysis: Moisture 1.30 wt.% Loss on Ignition (LOI) 1.29 Alpha-alumina 0.60 Surface area 98 m2/g. (B.E.T.) Screen Analysis +149 microns-2% +74 microns72% +44 microns96% Sodium chloride was added to the alumina to insure a high NaAICI4 catalyst level. The experimental conditions and aluminum chloride production rate coefficient are shown in Table 2.

Figure 1 , which includes the data presented in Table 2, graphically shows the effect of pressure on aluminum chloride production rate.

Temperature does influence the aluminum chloride production rate coefficient when PCA is chlorinated at these pressures, but this temperature effect is not significant enough to seriously interfere with the chlorination. The fact that the aluminum chloride production rate coefficient decreased as the temperature was increased from 615 to 720 C. should not be construed as indicating that at higher temperatures and pressures, the aluminum chloride production rate coefficient is less than that obtained at atmospheric pressure. In fact, the aluminum chloride production rate coefficient is significantly greater at superatmospheric pressures despite this temperature effect. One possible explanation for this temperature effect is that for a mixture of coke and alumina, one stage of the reaction could be adsorption of chlorine on carbon to form a chlorinated carbon radical which then reacts with alumina. If sorption is involved, the reaction rate would decrease with increasing temperature. Another possibility is that the reaction proceeds through an intermediate, such as carbon tetrachloride, which is less stable at the higher temperature.

Table 2 Coke+PartiallyCalcinedAlumina Examples Bed Depth=O. 70 m.

Aluminum Chloride Production Rate Temp. Pres. Velocity Cl2 Conversion Coefficient Ex. No. lOC.) (atm.) (cm./s.) (%) Ikg.-mollhr.m3. atm) X 615 3.5 6.4 89.2 6.5 Xl 643 3.5 11.3 62.3 4.9 XII 635 4.4 6.4 98.7 11.6 XIII 650 4.5 11.5 82.5 8.9 XIV 620 1.8 9.9 23.4 1.3 Examples XV-XXl A mixture of petroleum coke (19.9 wt.%), alumina (79.4 wt.%) and NaCI (0.7 wt.%) was chlorinated according to the procedure of Examples I through IX.The coke had the following analysis: Size fraction -210+105 microns Carbon 96.80% Hydrogen 0.79% Nitrogen 0.52% Sulfur 0.99% Ash 0.28% Surface area 10 m2/g. (B.E.T.) The alumina (MGA) had the following analysis: Moisture (250--3000C.) 1.96% Loss on Ignition (LOI) (1 1000C.) 1.28% Alpha-phase content 18.00% Surface area 54 m2/g. (B.E.T.) Screen Analysis + 149 microns-4% +74 microns56% +44 microns-76%.

The results shown below in Table 3 are visually represented in Figure 1.

An approximate threefold increase in the aluminum chloride production rate coefficient was realized by increasing reaction pressure from 1.9 to 3.4 atmospheres. The slower rate of increase of the production rate coefficient between 3.4 and 4.5 atmospheres seen in Figure 1 may have been caused by varying levels of reactive alumina and/or NaAICI4 in the reactor bed. Another possible explanation for this decreased improvement in production rate coefficient is the influence of fluidizing velocity upon the reaction. In both experiments at 3.5 atmospheres, the velocities exceeded 10 cm./sec. while the two experiments at 4.5 atmospheres with the lowest aluminum chloride production rate coefficients were performed with fluidizing velocities between 6 and 7 cm./sec.More efficient fluidization of the beds with high sodium concentrations at velocities exceeding 10 cm./sec. or the generation and return of larger quantities of reactive dust at the higher velocities could also explain this leveling.

Table 3 Aluminum Chloride Production Rate Temp. Pres. Velocity C4 Conversion Coefficient Expt. No. (0C.) (atm.) (cm./s.) fO/o) (kg.-mol/hr.m3. atm.) XV 700 1.9 9.9 28.0 1.4 XVI 605 1.9 9.5 21.1 4.0 XVII 620 3.4 11.4 62.3 5.3 XVIII 700 3.6 12.5 47.5 3.5 XIX 625 4.6 6.1 77.3 4.0 XX 700 4.5 6.6 77.1 4.1 XXI 630 4.4 10.7 67.0 5.4 Examples XXII through XXV were run in a bench scale quartz fluid bed reactor on a batch basis.

Example XXII Two experiments using coked alumina (17.9% carbon) at 0.3 m. initial bed height and 8 cm/sec.

chlorine superficial velocity were run, one at five atmospheres and the other at 1 5 atmospheres. The coked alumina had the following analysis: Alpha alumina 0.3% Na2O 0.23% LOI (0--11000C.) 19.5% Surface Area 8 m2/g. (B.E.T.) At five atmospheres and 7000C., the chlorine conversion was 97.2% yielding an aluminum chloride production rate coefficient of 29.1 kg.-mols aluminum chloride per meter3 hour atmosphere.

At 15 atmospheres and 71 50C., the chlorine conversion was 99.7% yielding an aluminum chloride production rate coefficient of 52.1 kg.-mols aluminum chloride per meter3 hour atmosphere.

Example XXIII Two experiments using partially calcined alumina (79.4 wt.%) and separate petroleum coke (19.9 wt.%) and sodium chloride (0.7 wt.%) were run at 0.27 meter initial bed height and 8 cm./sec. chlorine superficial velocity, one at five atmospheres and the other at 10 atmospheres. The alumina had the following analysis: Alpha alumina 0.4% Na2O 0.35% Moisture (O-3000C.) 1.11% LOl (300-12000C.) 1.02% The coke used had been calcined at 8250C. for 30 minutes and sized to 150 mesh.At five atmospheres and 6950C., the chlorine conversion was 59.6% yielding an aluminum chloride production rate coefficient of 7.69 kg.-mols aluminum chloride per m3 hour atmosphere. At 10 atmospheres and 6900 C., the chlorine conversion was 69.6% yielding an aluminum chloride production rate coefficient of 10.7 kg.-mols of aluminum chloride per m3 hour atmosphere.

Example XXIV Four experiments using alumina (MGA) (79.4 wt.%), sodium chloride (0.7 wt.%) and separate petroleum coke (19.9 wt.%) were run at 8 cm./sec. chlorine superficial velocity, two at five atmospheres and two at 10 0 atmospheres.The alumina had the following analysis: Surface area 59 m2/gm. (B.E.T.) Alpha alumina 17.0% Na2O 0.54% Moisture (O3000 C.) 1.68% LOI (300--12000C.) 0.88% The coke which had been calcined at 8250C. for 30 minutes had the following analysis: Size fraction -210+105 microns Surface area 10 m2/g. (B.E.T.) Carbon 96.8% Hydrogen 0.79% Nitrogen 0.52% Sulfur 0.99% Ash 0.28% One of the two samples chlorinated at five atmospheres was run at 0.53 m. initial bed height and 6500C. The chlorine conversion was 96% yielding an aluminum chloride production rate coefficient of 1 5.3 kg.-mols aluminum chloride per meter3 hour atmospheres. The other at five atmospheres was run at 0.56 m. initial bed height and 6500C.The chlorine conversion was 97.2% yielding an aluminum chloride production rate coefficient of 1 6.2 kg.-mols aluminum chloride per meter3 hour atmosphere. One of the two samples chlorinated at 10 atmospheres was run at 0.53 m. initial bed height and 6650C. The chlorine conversion was 99.6% yielding an aluminium chloride production rate coefficient of 26.7 kg.-mols aluminum chloride per meter3 hour atmosphere.

The other at 10 atmospheres was run at 0.56 m. initial bed height and 6550C. The chlorine conversion was 99.6% yielding an aluminum chloride production rate coefficient of 25.2 kg.-mols aluminum chloride per meter3 hour atmosphere.

Example XXV Alumina (80 wt.%) and petroleum coke (20 wit%) (-65 to + 1 50 mesh) were reacted at 8 cm./sec. chlorine superficial velocity, 7000 C., 0.58 m. initial bed height at 5, 7, 9, 11 and 13 atmospheres. The alumina had the following analysis: Surface area 0.3 m2/g. (B.E.T.) Alpha alumina 85.0% Beta alumina 15.0% Na2O 0.61% Moisture (0--3000C.) 0.25% LOI (3000--12000C.) 0.06% The coke had the same analysis as that of Example XXI. Each run was terminated when 90% of theoretical chlorine had been delivered. The results are graphically illustrated in Figure 2 where the average aluminum chloride production rate coefficient is plotted against pressure in atmospheres.The average aluminum chloride production rate coefficient is an arithmetic average of all the instantaneous readings taken calculated by assuming a constant bed volume.

These comparisons show that the rate dependence on chlorine pressure is greater than one since both chlorine conversions and aluminum chloride production rate coefficients increased substantially as the pressure increased.

Example XXVI 640 Grams of Alcoa grade A3 alumina containing 70% by weight alpha-alumina was charged to a reactor with 1 60 grams of Great Lakes Carbon 8250C. petroleum coke having a particle size range of -65 to + 1 50 mesh. The reactor which had a bed 0.54 meters deep was maintained at a temperature of 7000 C. and a pressure of 1 5 atmospheres while chlorine was flowed through the bed at a rate of 3.02 kg. per hour. After 35 minutes, analysis of residue indicated that 90% of the original alumina charge had been chlorinated, thus indicating the chlorination of substantially all of the alpha-alumina, i.e., that the mass transfer coefficient was not seriously affected by the temperature at which the aluminous material was previously calcined.

It will be apparent to those skilled in the art that numerous variations of the illustrated details may be made without departing from this invention.

Claims (24)

Claims

1. A method for chlorinating aluminous material containing oxygen, comprising chlorinating the aluminous material in the presence of a reducing agent at a pressure of greater than three atmospheres.

2. A method according to claim 1, wherein the aluminous material is chlorinated at a temperature of from 550 to 8000C.

3. A method according to claim 1 or 2, wherein the reducing agent consists of coke, carbon, carbon monoxide, COCI2 or CCl4.

4. A method according to any one of claims 1 to 3, wherein the aluminous material is alumina.

5. A method according to any one of the preceding claims, wherein the aluminous, oxygencontaining material is calcined before being chlorinated.

6. A method according to any one of the preceding claims, wherein the pressure is at least five atmospheres.

7. A method according to claim 6, wherein the pressure is at least seven atmospheres.

8. A method according to claim 7, wherein the pressure is at least nine atmospheres.

9. A method according to claim 8, wherein the pressure is at least 10 atmospheres.

10. A method according to claim 9, wherein the pressure is at least 11 atmospheres.

11. A method according to claim 10, wherein the pressure is at least 13 atmospheres.

12. A method according to claim 11, wherein the pressure is at least 1 5 atmospheres.

13. A method according to any one of the preceding claims 1 to 5, wherein the pressure is at most 1 5 atmospheres.

14. A method according to claim 13, wherein the pressure is between 3.3 and 4.7 atmospheres.

1 5. A method according to claim 13, wherein the pressure is between 5 and 1 5 atmospheres.

1 6. A method according to claim 15, wherein the pressure is between 5 and 10 atmospheres.

1 7. A method according to claim 15, wherein the pressure is between 10 and 1 5 atmospheres.

1 8. A method according to any one of claims 1 to 5, in which the aluminous material contains alpha-alumina and is chlorinated at a pressure of greater than 10 atmospheres.

1 9. A method according to claim 18, wherein the aluminous material contains at least 0.1 wt.% alpha-alumina.

20. A method according to claim 18 or 19, wherein the pressure is from above 10 atmospheres up to 20 atmospheres.

21. A method according to claim 20, wherein the pressure is from 12 to 20 atmospheres.

22. A method according to claim 21, wherein the pressure is about 15 atmospheres.

23. A method according to any one of claims 1 8 to 22, wherein the mass transfer coefficient is not seriously affected by the temperature at which the aluminous material was previously calcined.

24. A method according to claim 1 , for chlorinating aluminous material containing oxygen, substantially as hereinbefore described with reference to the Examples.