The present invention relates to a novel method for the re ^¬ moval of oxygen from an industrial gas through selective catalytic oxidation via reactant injection.

More specifically, the invention concerns an alternative route to reduce the oxygen content in industrial gases, where the reduction of the content of oxygen is crucial for the valorization of the gas. The method of the invention is focusing on landfill gas, digester gas and industrial CO ₂ off-gas. Today, oxygen removal is accomplished through PSA (pressure swing adsorption) , membrane or scrubber technologies with very high capital expenditure (CAPEX) and also a substantial loss of valuable components, such as methane in the main gas to the oxygen-containing off-gas. The present invention comprises addition of components, such as ¾, CO, methanol, ammonia or ethanol, to the main gas stream and leading the resultant gas stream to at least one catalytic reactor. In said reactor (s), the oxygen is converted selec ^¬ tively to CO ₂ and water across the catalyst.

Removal of oxygen from fuel gas streams is often a require ^¬ ment for distribution of the gas in the natural gas grid, and it is also a requirement when utilizing the gas as a vehicle transportation fuel. In addition, removal of oxygen is also critical for the utilization of other industrial gas streams, such as in producing merchant or industrial grade CO ₂ from oxygen-containing off-gases.

US 3.361.531 describes the removal of oxygen from oxygen- containing environments and gas mixtures by absorption in a solid material contact mass. More specifically, a compound selected from copper carbonate, manganese carbonate and iron carbonate is contacted with a hydrogen-containing gas at an elevated temperature below about 500 °C, thereby re- ducing the carbonate to the corresponding oxide compound.

This oxide compound is brought into contact with said oxy ^¬ gen-containing environment at around ambient temperature, thereby absorbing the oxygen and oxidizing the oxide compound .

The technologies dominating the industry today are PSA and membrane based technologies in small and medium sized pro ^¬ jects (typically up to 10,000 Nm ³ /h gas), whereas distilla ^¬ tion and cryogenic separation are the dominating technolo- gies in larger scale applications.

For applications in the digester gas and landfill gas puri ^¬ fication industry the gas flows are in the range of 500 to 10,000 Nm ³ /h, and technologies based on PSA and membranes are dominating. Apart from an often prohibitive CAPEX, PSA and membrane technologies have a high operation cost be ^¬ cause of their complexity and gas compression as well as a substantial loss of valuable hydrocarbons, such as methane, from the feed gas stream to the oxygen containing waste gas stream.

In the method according to the present invention, one or more components suitable for catalytic oxidation are in ^¬ jected into the oxygen-containing main gas stream after re- moval of sulfur-containing compounds and siloxanes from the gas. The components and the catalyst are chosen so that the catalyst oxidizes the injected components using the oxygen in the stream without substantially oxidizing the valuable components, such as methane, in the gas stream.

The components to be injected may comprise one or more of i.a. ¾, CO, ammonia, urea, ethanol and dimethylether

(DME) .

The active catalyst may comprise a metal selected among va ^¬ nadium, tungsten, chromium, copper, manganese, molybdenum, platinum, palladium, rhodium and ruthenium in metallic or metal oxide form supported on a carrier selected from alu ^¬ mina, titania, silica and ceria and combinations thereof.

Sulfur impurities in an industrial gas can create a corro- sive environment inside power generating equipment or even poison catalysts that may be present. In addition, hydrogen sulfide present in the feed gas to gas engines will cause degradation of the lubricating oil and lead to a need of frequent maintenance. Furthermore, ¾S needs to be removed if the gas is to be sent to gas pipelines or used as fuel in vehicles.

Another reason to clean the gas is that other impurities, such as siloxanes, can be deposited within heat and power generation equipment and cause significant damage to the internal components.

Siloxanes are organosilicon compounds comprising silicon, carbon, hydrogen and oxygen which have Si-O-Si bonds. Si- loxanes can be linear as well as cyclic. They may be pre ^¬ sent in biogas because they are used in various beauty products, such as e.g. cosmetics and shampoos that are washed down drains or otherwise disposed of, so that they end up in municipal wastewater and landfills. Siloxanes are not broken down during anaerobic digestion, and as a result, waste gas captured from treatment plants and land- fills is often heavily contaminated with these compounds.

It is known that siloxanes can be removed using non-regen ^¬ erative packed bed adsorption with activated carbon or po ^¬ rous silica as sorbent. Regenerative sorbents can also be used as well as units based on gas cooling to very low tem- peratures to precipitate the siloxanes out from the gas.

Further, liquid extraction technologies are used. In addi ^¬ tion, these technologies can be used in combination.

So a major issue in the utilization of raw gas from land- fills and anaerobic digesters is to provide a gas stream with a low sulfur content, i.e. less than a few hundred ppm, and with a very low content of siloxanes, typically linear or cyclic dimethyl Si-O-Si compounds. The pipeline specifications for natural gas are even stricter. In this case, ¾S must be removed to a residual concentration below 5 ppm, and CO ₂ and ₂ need to be removed as well. Combus ^¬ tion of sulfur containing compounds leads to formation of sulfur trioxide which will react with moisture in the gas to form sulfuric acid, which can condense in cold spots and lead to corrosion. However, particularly siloxanes give rise to problems because they are converted to S1O ₂ during combustion, leading to build-up of abrasive solid deposits inside the engine and causing damage, reduced service time and increased maintenance requirements for many components such as compressors, fans, blowers, burner nozzles, heat recovery surfaces in boilers and for gas engine components such as spark plugs, valves, pistons etc. In addition to causing damage and reduced service time to the engine, also any catalysts installed to control exhaust gas emissions are sensitive to S1O ₂ entrained in the gas stream, in fact even more so than the engine itself. For an SCR (selective catalytic reduction) catalyst, for example, the S1O ₂ toler ^¬ ance can be as low as 250 ppb .

For the reasons outlined above it is very desirable to re ^¬ move siloxanes and sulfur-containing compounds from gas streams.

Thus, the present invention relates to a method for the re ^¬ moval of oxygen from an industrial gas feed, said process comprising the steps of:

(a) removing sulfur-containing compounds and siloxanes from the feed gas,

(b) injecting one or more reactants for oxygen conversion into the heated feed gas,

(c) carrying out a selective catalytic oxygen conversion in at least one suitable reactor, and (d) cleaning the resulting oxygen-depleted gas, wherein the feed gas is heated either prior to or following step (a) . Conventionally the siloxanes and sulfur-containing com ^¬ pounds have been removed from gas streams at low tempera ^¬ tures, i.e. before heating the gas stream. However, it is also possible to fit the oxygen conversion more closely to a design in which the removal of the sulfur-containing compounds and siloxanes from the feed gas is part of the hot loop, i.e. with heating of the gas stream prior to removing the siloxanes and sulfur-containing compounds from it.

Preferably the gas feed, from which oxygen is to be re ^¬ moved, is a landfill gas, a digester gas or an industrial CO2 off-gas.

In a preferred embodiment of the method of the invention, a gas stream, such as a landfill gas containing ¾S and organic sulfur along with siloxanes, CO ₂ , ¾0, methane and various VOC (volatile organic carbon) compounds is treated.

The components to be injected in step (c) comprise one or more of ¾, CO, ammonia, urea, methanol, ethanol and di- methylether (DME) . Landfill gas of low quality, i.e. having a high content of nitrogen and oxygen, is more difficult and expensive to up ^¬ grade to pipeline quality than gases with a lower content of nitrogen and oxygen. Using the reactant injection to remove the oxygen from low quality landfill gases will lead to a high temperature increase in the reactor, which in turn will damage the catalyst. If, however, the reactant is dosed at two different points instead of one point, it is possible to use two reactors in series with cooling and re ^¬ actant injection in between. This approach has the added benefit that the energy recovered after each reactor can be used in a reboiler in the CO ₂ separation unit (amine wash) to regenerate the amine, and it can also be used as a feed preheater. The energy for the reboiler and for preheating of the feed would otherwise have to come from electricity or from combustion of landfill gas or natural gas. The heat coming from the oxidation can be transferred to an oil circuit which is used both to run a reboiler in the amine wash in the subsequent CO ₂ removal and to preheat the feed . The invention is illustrated further with reference to the Figures, where Fig. 1 shows the general outline of the method, while Fig. 2 illustrates a more attractive configu ^¬ ration . In the general case shown in Fig. 1, the process described above is applied after sulfur and siloxane removal using regenerative or non-regenerative adsorption technology along with gas cooling and refrigeration. After sulfur and siloxane removal, the gas is heated to between 150 and 450°C, and then ¾, CO, ammonia, urea, methanol, ethanol, DME or any combination thereof is injected into the main stream. Then the stream is led to the selective catalytic oxygen removal reactor, in which oxygen reacts with the injected component (s) to form CO ₂ and water. The hot reactor exit gas can be utilized to heat the reactor inlet gas by using a feed-effluent heat exchanger.

In addition, the hot exit gas from the catalytic reactor can be used to heat the feed gas to the sulfur and siloxane removal steps, so that these steps may operate at an ele ^¬ vated temperature. Downstream from the catalytic reactor, CO2 is removed using amine-based CO2 removal technology, CO2 water scrubbing technology or solvent-based CO2 removal technology. Alter ^¬ natively, CO2 can be removed using PSA or membrane technol- ogy .

In an alternative configuration the nitrogen removal unit is positioned downstream from the water removal unit. Nitrogen can be removed through PSA or membrane-based technology, and water is removed through cooling and condensa ^¬ tion followed by a molecular sieve unit, alternatively op ^¬ erating in a TSA (thermal swing adsorption) configuration. In the more attractive configuration shown in Fig. 2, for siloxane and sulfur removal the present invention is com ^¬ bined with Applicant's GECCO™ technology for digester and landfill gas conditioning. The feed gas is heated to 200- 450°C and fed to a siloxane absorption bed comprising alu- mina, alumina with nickel, silica or combinations thereof.

After siloxane removal, the gas is fed to a catalytic reac ^¬ tor containing a catalyst selected from tungsten, vanadium, molybdenum, platinum and palladium in metallic or metal ox- ide form supported on a T 1 O2 carrier. In this catalytic re ^¬ actor, the catalyst converts the sulfur compounds to S O2 and the VOC compounds (not methane and light [i.e. C3 and lower] hydrocarbons) to CO2 and water and also hydrogen halides if some of the VOCs are halogenated.

One or more components suitable for catalytic oxidation, i.e. ¾, CO, ammonia, urea, methanol, ethanol, DME etc., is/are injected into the main gas stream containing oxygen, and the gas stream is fed to the catalytic reactor contain ^¬ ing a catalyst such as vanadium, tungsten, chromium, copper, manganese, molybdenum, platinum, palladium, rhodium or ruthenium in metallic or metal oxide form supported on a carrier selected from alumina, titania, silica and ceria or combinations thereof. In the reactor, the injected compo ^¬ nent (s) is/are selectively oxidized to ¾0 and CO2 , while the valuable hydrocarbons, such as methane and light [i.e. C3 and lower] hydrocarbons, are substantially not con ^¬ verted. It is preferred that the catalyst comprises tung ^¬ sten, vanadium, molybdenum, platinum or palladium in metallic or metal oxide form supported on a T 1 O2 carrier. The hot reactor exit gas can be utilized to heat the reac ^¬ tor inlet gas by using a feed-effluent heat exchanger.

The additional heat generated in the oxygen removal step will provide a higher temperature difference for the feed- effluent heat exchanger, which reduces the CAPEX.

Downstream from the heat exchanger, the S O2 is removed in a wet caustic or H2O2 scrubber or a dry scrubber using a caustic sorbent. After the S O2 removal, CO2 is removed by using amine-based technology, solvent-based CO2 removal technology, water-based CO2 removal technology or alterna ^¬ tively PSA and/or membrane technology.

Nitrogen removal can be accomplished using membrane or PSA based technology. Then water is removed by using cooling and condensation followed by a molecular sieve, alterna ^¬ tively in a TSA configuration. Alternatively, the nitrogen removal unit is positioned downstream from the water re ^¬ moval unit.

It is further preferred that the catalyst is monolithic.