WO1999041557A1

WO1999041557A1 - Air prepurification for a large scale cryogenic air separation plant

Info

Publication number: WO1999041557A1
Application number: PCT/EP1999/000962
Authority: WO
Inventors: Michiel Jan Groeneveld; Franciscus Gerardus Maria Niele
Original assignee: Shell Internationale Research Maatschappij B.V.
Priority date: 1998-02-12
Filing date: 1999-02-11
Publication date: 1999-08-19
Also published as: AU3327699A

Abstract

The invention relates to a process for the production of oxygen and/or nitrogen in a cryogenic air separation unit. It especially relates to a process which avoids as much as possible any build up in the process of unsaturated hydrocarbons or of any compounds in which unsaturated compounds have been incorporated. The process comprises the removal of at least unsaturated hydrocarbons from the feed air and separation of oxygen and nitrogen by cryogenic distillation, while at least producing 200 ton oxygen/day.

Description

- 1 -

AIR PREPURIFICATION FOR A LARGE SCALE CRYOGENIC AIR SEPARATION PLANT

The present invention relates to a process for the production of oxygen and/or nitrogen in a cryogenic air separation unit. It especially relates to a process which avoids as much as possible any build up in the process of unsaturated hydrocarbons or of any compounds in which unsaturated compounds have been incorporated.

Several cryogenic concepts have been developed over the years to liquefy and separate air into its main constituents nitrogen, oxygen and rare gases. Refrigeration for cryogenic applications is produced by absorbing or extracting heat at low temperature and rejecting it to the atmosphere at higher temperatures. Three general methods for producing cryogenic refrigeration in large-scale commercial application are the liquid vaporisation cycle, the Joule-Thomson expansion cycle and the engine expansion cycle. The first two are similar in that they both utilise irreversible isenthalpic expansion of a fluid, usually through a valve. Expansion in an engine approaches reversible isenthalpic expansion with the performance of work. For more detailed discussion reference is made to Perry' s Chemical Engineers Handbook, Sixth Edition, 12-49 ff. (McGraw-Hill, New York, 1984), Kirk-Othmer, Encyclopedia of Chemical Technology, Fourth Edition, Volume 7, p. 662 ff . (John Wiley and Sons, New York, 1993) and Ullmann' s Encyclopedia of Industrial Chemistry, Fifth Edition, Volume A 18, p. 332 ff . (VCH, Weinheim, 1991) .

Most commercial air separation plants are based on Linde's double distillation column process. This process is clearly described in the above references. In a typical example, feed air is filtered and compressed to a - 2 - pressure usually between 5 and 10 bara. The compressed air is cooled and any condensed water is removed in a separator. To avoid freezing of water and carbon dioxide in the cryogenic part of the plant, the feed air is further passed through an adsorbent bed, usually activated alumina and/or molecular sieves, to remove the last traces of water and carbon dioxide. The purified air is than cooled down further, and fed to a first cryogenic distillation unit, usually at an intermediate stage. Crude liquid material from the bottom section of the first distillation unit, usually comprising between 40 and 50 mol percent oxygen, is fed to the second distillation unit (which second unit is usually on the top of the first distillation unit, the condenser of the first column usually acting as the reboiler for the second unit), usually also at an intermediate stage. The second distillation unit is operated at relatively low pressure (usually 1 to 2 bara) . At the top of the first distillation unit almost pure liquid nitrogen is ob-tained which is typically fed to the second column at the top. Pure liquid oxygen is obtained at the bottom of the second distillation unit, while pure gaseous nitrogen is obtained from the top of the second column.

Many variations on the above concept are known. These include separation of air into gaseous products, liquid products and all kind of combinations thereof. Also the production of partly enriched oxygen and/or nitrogen streams together with almost pure oxygen and/or nitrogen streams, either in liquid or gaseous phase is well known. In addition there may be additional distillation units to separate any of the rare gases present in the feed air. Further, the methods for creating the low temperatures may vary in many ways. In this respect reference is made to the above cited literature references, and further to EP 798524, JP 08094245, EP 593703, EP 562893, US 5237822, - 3 -

JP 02052980, EP 211957, EP 102190, SU 947595 JP 71020126 and JP 71020125.

It is known that hydrocarbons in combination with pure oxygen might create safety problems. Therefore, when high amounts of hydrocarbons are present in the feed air, these hydrocarbons may be removed. Also, when extremely high purity oxygen and/or nitrogen has to be produced, e.g. for use in the production of micro-electronics components, hydrocarbons usually are removed from the feed air. However, in the (very) large scale production of oxygen of industrial quality, e.g. for use in steel production, for the conversion of coal and/or hydrocarbons into synthesis gas and for the production of ethylene oxide, it is extremely difficult to remove all hydrocarbons from the feed air. Usually, the higher hydrocarbons, especially C5+, are removed at the same time when water and/or carbon dioxide are removed. These latter two compounds have to be removed to avoid plugging of the cold process-lines. This is often done by refrigeration purification and/or (low temperature) solid adsorbent purification. This will also remove higher hydrocarbons. Further, often also acetylene is removed using adsorbent beds. Provided that the amounts of C1--.4 hydrocarbons in the feed air especially unsaturated C --. hydrocarbons, more especially ethane, are relatively low, e.g. less than 100 ppm for each hydrocarbon, often less than 40 ppm, it is generally accepted that these hydrocarbons do not create any safety problems, not in any gaseous oxygen streams produced, nor in any of the liquid oxygen streams or reservoirs. In this respect it is observed that these lower hydrocarbons are soluble in liquid oxygen, and thus are expected not to accumulate in the system. Even when higher concentrations of hydrocarbons would occur somewhere in the system, - 4 -

especially the distillation section, it would not be expected that this would result in any safety problem, especially in view of the low temperature. Further, a simple drain from a specific part in the distillation unit would be sufficient to reduce any increasing concentrations. Thus, it is general practice not to apply any special processes to remove any lower hydrocarbons from cryogenic air separation units producing large amounts of especially oxygen, more particularly oxygen of industrial quality. This is only done for small scale units when there is a clear need to do so.

We have now discovered that under specific conditions, which specific conditions are not clearly understood yet, there might be an accumulation of especially unsaturated lower hydrocarbons in the cryogenic section of an air separation unit. Without being bound by any theory, it is presumed that insoluble compounds may be formed, especially on the inside walls of the cryogenic distillation units. These insoluble compounds are expected to contain at least carbon and hydrogen, but very well may contain oxygen too (chemically bonded to the carbon) . These compounds may have the character of a polyperoxide of yet unknown structure, possibly alternating oxygen and carbon fragments. Such compounds may be formed via radical initiated polymerisation, perhaps initiated by a methylperoxyl radical or by a radical contained in a submicron aerosol particle. The insoluble compounds as mentioned above form a potentially very big hazard, although they will not necessarily always result in a disaster. It will be understood that for any explosion an ignition is necessary. A possible ignition source might be an internal friction, or the falling down of material, e.g. blocks of ice or other solid material, or the (spontaneous) falling down of grown insoluble compound at - 5 -

a higher stage in the reactor. Also chemical excitation might be possible, for instance initiated by a higher olefin molecule or by a transition metal. The build up of the insoluble material is presumably slow, and will depend on the presence of unsaturated hydrocarbons, especially ethene, and radicals in the feed air. In the case that there is no build up, or the distillation unit is derimed (which is usually done once every two or three years) before any ignition has occurred, there are no real problems. However, sometimes an ignition might occur, resulting in (small) damage to the internals, sometimes in large explosions. To overcome the formation and build up of these insoluble compounds, it is now proposed to remove at least a large part of small unsaturated hydrocarbons from the feed air for large scale cryogenic air separation units. Thus, the presently accepted amounts of ethylene present in the feed air in the ppm range, e.g. up 10 or even 25 or 40 ppm, should be diminished to amounts in the ppb range, e.g. less than 25 ppb, suitably less than 10 ppb or even less.

Preferably a corresponding reduction is also obtained for any other unsaturated hydrocarbons.

The present invention therefore relates to a process for the production of oxygen and/or nitrogen in a cryogenic air separation unit, comprising removal of at least unsaturated hydrocarbons from the feed air and separation of oxygen and nitrogen by cryogenic distillation, while at least producing 200 ton oxygen/day, especially at least 400 ton/day. This corresponds with an intake of at least about 858.4 ton/day air, especially at least about 1716.8 ton/day air.

It will be understood that the present invention can be applied in all cryogenic air separation units, independent from the specific design. As indicated above, - 6 -

most cryogenic air separation units are based on the Linde's double distillation column process, but over the years many variations have been developed. The invention might be used in all these variations. A number of possible variations have been described above.

The removal of the unsaturated hydrocarbons as ethene, propene and butene may be done by methods well known in the art. A suitable process is an adsorption process, especially a pressure swing adsorption process. Materials as silica gel, carbon and natural and synthetic zeolites (e.g. mol sieves) may be used. Most of the gels and carbons have pores of varying size, but the synthetic zeolites are manufactured with closely controlled pore-seize openings ranging from 0.4 to 1.5 nm. This makes them even more selective than other adsorbents since it permits separation of gases on the basis of molecule size. The design of an adsorber can be made on the basis of the equilibrium data between the solid and the gas and the rate of adsorption. The rate of adsorption is usually rapid, and adsorption is essentially complete in a relatively narrow zone of the adsorber. If the concentration of the adsorbed gas is considerably more than a trace, it might be useful to carry out the purification in two steps. In usual plant operation at least two adsorption purifiers are employed: one in service, while the other one is being desorbed of its impurities.

Another way to remove the unsaturated hydrocarbons is the use of an oxidation catalyst, e.g. noble metal catalysts, especially porous, supported catalysts, especially at elevated temperature. Further oxidising agents as ozone, metal perchlorates and metal permanganates may be used as well as metal peroxides. It is important that a complete conversion into carbon dioxide and water is obtained. - 7 -

Another way to remove the unsaturated hydrocarbons is the chemical conversion of the unsaturated hydrocarbons, especially in a polymerisation reaction. In such a reaction an acidic catalyst might be used, preferably a zeolite. Further, also resins comprising acidic groups may be used.

It will be appreciated that at least a substantial amount of the unsaturated hydrocarbons present in the feed air will be removed, and that thus the oxygen produced will contain a substantially decreased amount of unsaturated hydrocarbons. Preferably at least 80 mol percent unsaturated hydrocarbons are removed from the feed air, preferably 90 percent, more preferably 96 percent. The unsaturated hydrocarbons to be removed from the feed air are especially Cι__4 hydrocarbons. As explained above, the higher hydrocarbons usually are already removed from the feed air. Especially the C_]__3 hydrocarbons, more preferably C^_2 hydrocarbons are removed from the feed stream. If possible, the best solution is the removal of all unsaturated hydrocarbons. It is observed in this respect that usually ethene is present in air in amounts much higher than propene and/or butene. The total amount of unsaturated hydrocarbons in the (gaseous) oxygen produced is suitably less than 12 volume ppb, preferably less than 1.2 volume ppb, more preferably less than 0.12 volume ppb. Especially substantially pure oxygen is produced, especially between 90 and 99.9 volume percent, preferably between 95 and 99.8 volume percent, the remainder being nitrogen and optionally rare gases. It is observed that it is expected that saturated hydrocarbons (methane, ethane, propane etc.) form a smaller safety risk than unsaturated hydrocarbons (ethene, propene etc.) in view of their - 8 - higher chemical inertness, and thus the need to remove saturated hydrocarbons is less urgent. In addition to the above proposed measures to remove unsaturated hydrocarbons from the feed air, it is advantageous that the cold liquid product from the first distillation unit is freed from any entrained hydrocarbons, especially unsaturated hydrocarbons, before introduction into the second distillation column. This can be obtained by (swing) adsorption using a suitable adsorbent, especially silica gel. A further improvement may be obtained when any recycle streams in the second distillation unit are freed from any entrained hydrocarbons, especially unsaturated hydrocarbons, before reintroduction into the distillation column. This can also be obtained by (swing) adsorption using a suitable adsorbent, especially silica gel.

In the process according to the present invention suitably at least 800 ton oxygen/day is produced, preferably at least 1200 ton/day, more preferably 2000 ton/day, still more preferably 2500 ton/day. The maximum amount may be up to 4000 ton/day, or even 6000 ton/day, although even higher amounts, e.g. 10,000 ton/day, are also possible. Preferably gaseous oxygen and/or nitrogen is produced, but the process applies also for the production of liquid oxygen and/or nitrogen.

The present invention further relates to a process for the production of oxygen and/or nitrogen in a cryogenic air separation unit, comprising removal of at least unsaturated hydrocarbons from the feed air and separation of oxygen and nitrogen by cryogenic distillation, in which process the concentration of ethene in the produced (gaseous) oxygen is less than 40 percent of the concentration of ethene in the feed air calculated with respect to the oxygen, preferably less - 9 -

than 15 percent of the feed air, more preferably less than 3 percent of the feed air. In this process the purity of the oxygen produced is suitably between 90 and 99.9 volume percent, preferably between 95 and 99.8 volume percent. Suitably, in the process the concentration of ethene in the feed air after the removal of unsaturated hydrocarbons is less than 100 volume ppb, preferably less than 1 volume ppb, more preferably less than 0.01 volume ppb. In the process suitably at least 200 ton oxygen/day is produced, preferably 400 ton/day, more preferably 800 ton/day, even more preferably at least 1200 ton/day, still more preferably 2000 ton/day, even still more preferably 2500 ton/day. Further details for this process correspond with the details as described in claims 2 to 12, and the further technical details as described in the description herein before.

All percentages and other fractions as ppm and ppb are based on volume unless indicated different, and are based on the total volume of all components.

Claims

- 10 -C L A I M S

1. Process for the production of oxygen and/or nitrogen in a cryogenic air separation unit, comprising removal of at least unsaturated hydrocarbons from the feed air and separation of oxygen and nitrogen by cryogenic distillation, while at least producing 200 ton oxygen/day, especially 400 ton oxygen/day.

2. Process according to claim 1, in which the unsaturated hydrocarbons are removed by adsorption.

3. Process according to claim 2, in which a pressure swing adsorption process is used, preferably using molecular sieves.

4. Process according to claim 1, in which the unsaturated hydrocarbons are removed in an oxidation process, preferably a process in which an oxidation catalyst or an oxidising agent is used, more preferably a noble metal catalyst.

5. Process according to claim 1, in which the unsaturated hydrocarbons are removed by in a polymerisation reaction, preferably a process in which an acidic catalyst is used, more preferably a zeolite.

6. Process according to any of claims 1 to 5, in which at least 80 mol percent unsaturated hydrocarbons are removed, preferably 90 percent, more preferably

96 percent.

7. Process according to any of claims 1 to 6, in which the unsaturated hydrocarbons to be removed are C^_4 hydrocarbons, preferably C]__3 hydrocarbons, more preferably C╬╣_2 hydrocarbons.

8. Process according to any of claims 1 to 7 , in which the cryogenic distillation process comprises a double - 11 -

distillation column process, preferably a process in which the cold liquid product from the first distillation unit is freed from entrained hydrocarbons, especially unsaturated hydrocarbons, before introduction into the second distillation column.

9. Process according to any of claims 1 to 8, in which any recycle in the second distillation unit is freed from entrained hydrocarbons, especially unsaturated hydrocarbons, before reintroduction into the distillation column.

10. Process according to any of claims 1 to 9, in which substantially pure oxygen is produced, especially between 90 and 99.9 volume percent, preferably between 95 and 99.8 volume percent.

11. Process according to any of the preceding claims, in which at least 800 ton oxygen/day is produced, preferably at least 1200 ton/day, more preferably 2000 ton/day, still more preferably 2500 ton/day.

12. Process for the production of oxygen and/or nitrogen in a cryogenic air separation unit, comprising removal of at least unsaturated hydrocarbons from the feed air and separation of oxygen and nitrogen by cryogenic distillation, in which process the concentration of ethene in the produced oxygen is less than 40 percent of the concentration of ethene in the feed air calculated with respect to the oxygen, preferably less than 15 percent of the feed air, more preferably less than 3 percent of the feed air.

13. Process according to claim 12, in which the purity of the oxygen produced is between 90 and 99.5 volume percent, preferably between 95 and 99 volume percent.

14. Process according to claim 12 or 13, in which process the concentration of ethene after the removal of unsaturated hydrocarbons is less than 100 volume ppb, - 12 -

preferably less than 1 volume ppb, more preferably less than 0.01 volume ppb.

15. Process according to any of claims 12 to 14, in which at least 200 ton oxygen/day is produced, preferably

400 ton/day, more preferably 800 ton/day, even more preferably at least 1200 ton/day, still more preferably

2000 ton/day, even still more preferably 2500 ton/day.