WO2002040138A1

WO2002040138A1 - Process and plant for removing nitrogen oxides from flue gases

Info

Publication number: WO2002040138A1
Application number: PCT/NO2001/000457
Authority: WO
Inventors: Olav Martin Kvalheim; Edvard Sundby
Original assignee: Grønn Kraft As
Priority date: 2000-11-17
Filing date: 2001-11-16
Publication date: 2002-05-23
Also published as: AU2002218571A1; NO20005846D0; NO20005846L

Abstract

A method and an installation are described for combustion and/or thermal degradation (such as cracking) of organic/biological material, at temperatures in which waste gases containing nitrogen oxides (NOx) are formed, where the waste gases that are formed during the combustion of the thermal degradation are led in a closed circuit, in which peroxide is added to oxidise the nitrogen oxides which are formed.that are formed during the

Description

Process and plant for removing nitrogen oxides from flue gases.

The present invention relates to a method and an installation for combustion and/or thermal degradation (such as cracking) of organic material, at temperatures at which waste gases containing nitrogen oxides (N0_X) and/or sulphur oxides (SO_x) are formed.

The present invention may be used in different types of installations in which organic material is combusted or degraded, such as in ship and car engines, but, in the first instance, it is intended to be applied in thermal power plants/gas-driven power plants and installations for thermal degradation of organic or biological material to render these more discharge friendly and more economically viable. The following description, and the experiments that have been carried out to document the invention is therefore in particular based on gas-driven power plants.

Gas-driven power plants operated in a conventional way are not favourable because of the large discharges of the climatic gas C0₂, which makes it difficult to get political and public acceptance for gas-driven power plants. This is the case internationally, and in particular in Norway because of, for example, obligations associated with the Kyoto treaty. Operation of conventional gas-driven power plants generates three main components:

I Electrical energy

II Waste heat: a) Cooling water b) Air

III Harmful flue gases and climatic gases: a) C0₂ b) CO c) HC (non-combusted hydrocarbons) d) NOχ

Globally, for the last 20 years, there has been a steadily greater emphasis put on the environmental damage which is caused by combustion and other thermal degradation of fossil fuels. In spite of new technology, such as catalysts, flue cell technology, electrical engines, soot filters and so on, being put to use in vehicles and combustion plants with the resulting reduction in the discharges per unit, the discharges of polluting substances are still increasing in the global context because of increased industrialisation.

For a long time energy has been generated in coal-driven power plants. This type of power plant is a great source of pollution with considerable discharges of soot and carcinogenic hydrocarbons. Therefore, during the last years, focus has been directed at the generation of energy from gas-driven power plants which are almost pollution- free with regard to soot and hydrocarbons, but with the technology available today, still have considerable discharges of C0₂ and NO_x. In connection with the construction of gas-driven power plants, the main focus thus far has been on the large discharges of carbon dioxide. This is because this gas is assumed to be the main contributor to global warming. Furthermore, a great focus is put on discharges of NOχ.

According to the Gothenburg protocol, the signatories shall reduce their discharges of NO_x considerably within the year 2009. The background to the international agreement in this area is partly due to the fact that NO_x can form harmful ozone near the ground, and that the soil and water becoming acidified.

Cleaning technology based on catalysts can reduce the discharges of NO_x with up to 80-90%, but this is a cost- demanding technology. Therefore, there is a need for cheaper technology, which can quickly be developed and applied.

NO 19955026 describes a method to remove carbon dioxide and nitrogen oxides from combustion gases in that the combustion gas is first cooled to 50 - 100 °C, and that ozone is added to oxidise components containing nitrogen and carbon to the gases N0₂ and C0₂.

In contrast to this, the present invention relates to a conversion/oxidation of the mentioned carbon and nitrogen containing compounds to N0₂ and C0₂ by adding hydrogen peroxide. A considerable advantage one achieves by using hydrogen peroxide is that one does not need to cool the flue gases. The addition of hydrogen peroxide can take place at the temperature at which the flue gas is led out of the combustion chamber, normally at about 500 - 700 °C. Furthermore, ozone is very poisonous on inhalation, thermally unstable (explosive) and is also expensive to produce. Hydrogen peroxide can be produced cheaply from hydrogen gas, which in turn can be produced from natural gas in connection to the power plant. Furthermore, hydrogen peroxide is stably solubelized in water at concentrations which are considered for use in the present invention. A further advantage is that peroxide can be used in all steps of the process.

NO 305.154 describes a method to reduce the content of nitrogen oxides in polluted air by injecting ozone and ammonia. US 5,492,676 describes removal of nitrogen compounds from exhaust-containing tunnel air by adding inter alia ozone.

With the present invention a new method to remove, or reduce discharges of N0_X by combustion/conversion of organic material, is provided. Firstly, the use of peroxides, preferably hydrogen peroxide, as described above, has many advantages related to the known use of ozone. By addition of hydrogen peroxide to the flue gas, or waste gas, one will thus ensure an improved removal of NO . In addition, by adding hydrogen peroxide to different steps in the process, one achieves a complete installation, which is superior with regard to efficiency, small discharges, and economy compared to the conventional installations of today. In that NO_x is converted to nitric acid and any SO_x is converted to sulphuric acid, a commercially valuable product arises at the same time as harmful discharges are eliminated. Therefore, the purpose of the invention is to provide a method which makes it possible, in particular, to provide an improved combustion or general thermal conversion of organic material in which discharges of nitrogen oxides and any sulphur oxides are practically eliminated and discharge of carbon oxides to the air is considerably reduced through conversion to commercially valuable products, especially in bio-production for manufacture of feed products and the like. Furthermore, it is an aim of the invention that the chemical reactants, which are formed, can be used in further processing, and that the waste heat, which is generated, can be used for other related applications. With the improvements, which arise from the present invention, one has also found that the efficiency of the combustion can be increased. This is a result of the further refining of NOχ and SO_x to commercially valuable products. Thus, one can increase the efficiency of the turbines and take out more energy without it leading to increased discharges but, on the contrary, is balanced by increased nitrate and sulphate production. Thus, one achieves a combustion/thermal degradation in which the discharges of undesirable gases are eliminated (N0_X and SO_x) , or at least reduced considerably, at the same time as one ensures a favourable conversion of these undesirable gases to useful compounds.

According to the invention, this is achieved in that the nitrogen oxides and/or the sulphur oxides, which are formed during the combustion/thermal degradation, are led further and captured by conversion to nitrite and/or sulphate or the corresponding acids, and that the carbon oxide (s) is captured at least partially by use in the production of biomass and/or conversion to carbonate. With the harmful gases (NO_x and any SO_x) and climatic gas (C0₂) being captured, there is also a possibility of increasing the combustion temperature considerably, which gives increased efficiency for the power plant, which thereby achieves better profitability and reduced effluent discharge per produced unit of energy.

The present invention is thus characterised in that the waste gases, containing nitrogen oxides (NO_x) and/or sulphur oxides (S0_X) , which are formed during the combustion or the thermal degradation are led into a closed cycle in which peroxide is added to oxidise the harmful gases which are formed. It is especially preferred to use hydrogen peroxide.

A further preferred embodiment of the method according to the invention is characterised in that the waste gas is further bubbled through a solution of peroxide dissolved in water, wherein nitrogen oxides are oxidised to nitric acid, and any sulphur dioxide is oxidised to sulphuric acid. In this "bubble chamber", it can be relevant to pack glass balls, which increase the contact surface and the residence time for the waste gas.

A further preferred embodiment of the method according to the invention is characterised in that also during the combustion or the thermal conversion, which takes place in the combustion/conversion chamber itself, peroxide is added to increase the efficiency of the process.

The present invention further comprises an installation .for combustion and/or thermal conversion of organic material, characterised in that the installation comprises; a) chamber for combustion and/or thermal conversion of the organic material, and b) waste gas channel through which the waste gases which are produced in the chamber are led, and where this pipe comprises means, such as nozzles, for the addition of peroxide.

In a preferred embodiment, the installation comprises a bubble tank through which the waste gas is bubbled for further oxidation of the N0_X and any SO_x. In the installation, nozzles can also be placed in the combustion chamber itself, such that the peroxide can be feed into the flame area.

Thus, a preferred embodiment of the invention comprises oxidation in three different steps. An oxidation agent, preferably hydrogen peroxide, is added in all the steps.

The three oxidation steps in the process are as follows:

Step a - oxidation during the combustion itself

Hydrogen peroxide is added directly to the combustion chamber during the combustion process. The hydrogen peroxide is quickly converted to reactive hydroxyl radicals at the temperature in the combustion chamber. Further, the supply of air can be reduced such that the amount of waste gas is decreased and the biological material burns at a higher efficiency. This is not an obligatory step, but only an alternative embodiment. It must be pointed out that step b below can be used as the only step in the process. Step b - oxidation of components in the flue gas

Flue gases develop during the combustion process. The material is burning in the presence of air, and nitrogen oxides, among others, are developed as a consequence of the nitrogen in the air. Normally these flue gases are let out through the pipe and, in this way, represent a considerable pollution problem. These discharges are the reason to the political resistance to gas-driven power plants, and to reduce these discharges considerably has been the starting point and the motivation to provide a new, improved technical solution.

As mentioned above, adding ozone to the flue gas to promote an oxidation of NO to N0₂ and CO to C0₂, is known. To use ozone is not particularly favourable, and with the present invention we have shown that hydrogen peroxide can be used as an oxidation agent, and considerable advantages are achieved, as mentioned above.

By addition of hydrogen peroxide to the flue gases NO will be converted to N0₂. It must be pointed out that this addition of hydrogen peroxide can be carried out at the temperature which the flue gases may have, i.e. without a precooling of the flue gases.

Thus, this step represents a considerable improvement in relation to the use of ozone, where the oxidation agent is added at a temperature at about 100 to 150 degrees Celsius. Step c - oxidation of water-soluble components as the waste gas is led through a water reservoir

In this step, the waste gas is led, preferably after cooling by way of a heat exchanger, through a reservoir with hydrogen peroxide dissolved in water where N0₂ is converted to nitric acid, and any S0₂ to sulphuric acid. Nitric acid and sulphuric acid can thereafter be converted to nitrate and sulphate. Parts of C0₂ are possibly converted to carbonate in a separate step after this, in which a basic solution of hydrogen peroxide is used to oxidise parts of C0₂ in the waste gas to carbonate. These components are used as valuable components in the conversion to biological material, and thus the cycle is closed.

Detailed description of the invention

According to the invention, NO_x is captured by these components in the waste gas being oxidised to acid, in particular nitric acid, which can be converted further with a base to give a nitrate. The most relevant nitrates are potassium nitrate and ammonium nitrate.

The carbon dioxide which is captured is used either in biomass production, as it together with waste heat in the form of heated water from the gas driven power plant and nitrate is used in a land based site for production of micro-algae which are used as raw materials for feedstuffs or health food products or medicines, and/or it is led through a pipe installation in a enclosed ocean system or a part of a fjord, to algae and sea tare production which can yield increased production of zooplankton and wild fish. The waste heat can also be used in breeding installations. Thus, one achieves a closed system for the degradation and the re-building of organic material where a large part of the carbon is bio-fixed into useful products. Some of the carbon dioxide is used for the manufacture of carbonate in a known way.

A person skilled in the art will be able to decide, from analysis of demand for the various final products, the prices of the products and the reactants applied, etc. how one chooses to treat the waste gas. Also included in such a decision is whether one wishes to purchase reactants, or manufacture them in situ. It will be most relevant to manufacture H₂0₂, for example, on site, due to easy access to hydrogen gas .

The method according to the present invention makes it possible to operate nearly NO_x free (and possibly S0_X free) power plants with increased efficiency and reduced C0₂ discharges in that the otherwise undesired nitrogen oxides (and any sulphur oxides) which are formed, are made use of by supplying hydrogen peroxide containing liquid/gas to:

1) the flame area where the hydrogen peroxide gives improved combustion

2) the waste gas before heat exchange, and

3) the liquid which the waste gas is finally led through.

It is preferred that hydrogen peroxide is produced in connection with the installation, i.e. in situ, but this is not a requirement. If the hydrogen peroxide is manufactured in situ, this can best be carried out from the hydrogen gas which is either formed by reforming natural gas or by way of biological processes, for example from hydrogen producing micro-algae. As the temperature of combustion in the gas turbine increases a somewhat higher efficiency and increased production of NO_x are achieved.

The nitrogen oxides are converted to nitrate (and any SO_x to sulphate) and the salts formed are sold on or applied together with C0₂ and waste heat to biomass production by way of root fertilisation or a sprinkling or irrigation installation.

The process for a gas-driven power plant can be illustrated schematically by the following equations:

Gas + air + oxidant -» energy + C0₂ + CO + NO + N0₂ (combustion phase (step a) )

NO + oxidant —» N0₂ (combustion phase (step a) + flue gas phase (step b) + liquid phase (step c) )

N0₂ + oxidant -» HN0₃ (water phase (step c) )

and possibly with SO_x present

S0₂ + oxidant -> H₂S0 (water phase (step c) )

It must be pointed out that the preferred oxidant is hydrogen peroxide, but other peroxides may also be used. Step a

As mentioned above, this is an optional step. A liquid or gas containing hydrogen peroxide is supplied to the combustion area of the gas-driven power plant, which that results in an improved combustion process.

This method is used in engines and is known from NO 174301 C. Furthermore, J.A.J. Karlsson shows in his PhD thesis that part of the HOOH molecules which are fed early in the combustion process survive this first phase, and can thereby participate in reactions in the post-combustion phase/flue gas phase. This is a main element in Norwegian Patent 175 015. In this patent it is explained how a part of the hydrogen peroxide which is supplied to the suction side of a motor passes through the combustion chamber/combustion area out in the waste gas system where it reacts further with waste gas components in the exhaust and with a contaminating coating on metal surfaces and catalysts.

An important element in the process is the conversion of NO to N0₂. Addition of hydrogen peroxide to the flame zone has shown to increase the fraction of N0₂ significantly and gives a corresponding decrease in NO in the combustion of fossil fuel in car and ship engines. This process is a result of the general combustion methodology that is referred to in the above mentioned NO 174 301.

If the peroxide can be decomposed catalytically to hydroxy- radicals at the point of addition to the gas, the efficiency of this process can be increased. Thus, it is relevant to add means that can catalyse conversion of the peroxide, and this is described in more detail below.

As a second step in the integrated cycle, hydrogen peroxide is supplied to the flue gas to oxidise NO to N0₂.

In the flue gas NO is oxidised with hydrogen peroxide according to the following formula:

NO + H₂0₂ - N0₂ + H₂0

As for step (a) , the reaction can be made more efficient with the use of a catalysing agent, or with the use of UV light, plasma discharges, or other electro-magnetic effects. OH oxidises NO according to the following formula.

NO + OH → N0₂ + H

Vanderschuren and co-workers have reported on laboratory studies of oxidation of NO and N0₂ using peroxide in connection with combustion of carbon. As a third step in the integrated cycle the flue gas is bubbled from step 2 through a reservoir with peroxide, preferably after it is cooled in a heat exchanger. Small amounts of sulphuric acid can be added to increase the effectiveness of the conversion of N0₂ to nitric acid. After this process, the waste gas is led though a new reservoir with peroxide to convert parts of the carbon dioxide to carbonate.

To give a valuable final product, ammonia or an alkaline metal compound, preferably potassium hydroxide (KOH) , can be added. As the flue gas contains a lot of C0, the pH value of the liquid is set at around 5 because the carbon dioxide dissolves in the form of a carbonate.

Through effective oxidation and washing processes, NO_x is converted to water-soluble nitrates. Outgoing liquid can be used as nutrient salt/fertiliser in a bio-production process, or refined further for extraction of saleable fertiliser products.

The heated and nutrient rich liquid is thus led, after a possible use is made of parts of the heat energy for central heating purposes and the like, to a bio-production - installation which constitutes a new step in the integrated cycle.

In situ production of hydrogen peroxide

As mentioned above, it will be favourable to manufacture hydrogen peroxide in situ and a preferred embodiment of an installation according to the invention comprises therefore such a unit for in situ production of hydrogen peroxide. Furthermore, to realise this, it will be favourable to be able to synthesise an organic reductant and/or corresponding oxidant from natural gas in a smaller catalytic cracker installation, which is associated to the process. If, in such a cracker installation, one produces for example, isopropanol and acetone, then these can participate further in the process for the production of hydrogen peroxide, where hydrogen gas also possibly participates. Alternatively, these components can be bought . In connection with commercial large production of hydrogen peroxide, anthraquinone is normally used as a redox agent. In the method according to the invention it is considered to use the isopropanol method. This method consists of isopropanol being oxidised by air to acetone and hydrogen peroxide. Thereafter isopropanol can be regenerated in that acetone is reduced with the aid of hydrogen. Alternatively, acetone can either be burnt in the flame from the power plant where it, according to preliminary measurements can, together with hydrogen peroxide, support the conversion of NO to N0₂.

The main reaction can be described by the following redox pair, in which isopropanol is first oxidised by air to acetone and hydrogen peroxide, whereupon isopropanol is regenerated with the aid of hydrogen gas:

CH₃-CHOH-CH₂ + 0₂ → CH₃-CO-CH₂ + H₂0₂ oxidation

CH₃-CO-CH₃ + H₂ - CH₃-CH0H-CH₃ reduction

The advantage with this method is that the hydrogen peroxide produced does not need to go through wide reaching purification or stabilisation procedures as acetone and isopropanol have good combustion characteristics and can, as mentioned, act together with hydrogen peroxide in the process of the conversion of NO to N0₂ as described. Acetone is less polar than water and is therefore more strongly associated with NO in the gas phase, at the same time as it is a better solvent for NO in the liquid phase. If one burns or takes care of the acetone, which is formed, one can also run the process completely without hydrogen gas, but then at the price of the isopropanol production having to be increased considerably.

Anthraquinone is, however, a much heavier molecule with poorer combustion characteristics and can be expected to lead to tar or soot in the installation. This process will, therefore, demand greater extent of control and cleanliness, without giving the positive additional effects, which are achieved with the isopropanol method. The anthraquinone method can still come into consideration if it proves to be technically and economically favourable.

In situ production of hydrogen gas

In connection with production of hydrogen peroxide, hydrogen gas is normally used as a reduction agent. In the cycle, according to the invention, this gas can be produced by pyrolysis or incomplete combustion of methane with or without the presence of a catalyst.

In pyrolysis - methane is heated with no supply of air - graphite (soot) is formed as a by-product. If one supplies some air, one gets an incomplete combustion where besides formation of hydrogen, CO is also formed which is recycled to the gas flame for final combustion.

Through the following formulae it is apparent how the reaction is influenced when the lambda value is increased from 0 to 1 where λ=l represents complete combustion. CH₄ → 2 H₂ + C λ=0 pyrolysis

CH₄ + H 0₂ — CO + 2 H₂ λ= incomplete combustion

CH₄ + 0₂ -» C0₂ + H₂0 + H₂ λ=*≤ incomplete combustion

CH + 2 0₂ -» C0₂ + 2 H₂0 λ=l complete combustion

It can be seen that to achieve favourable hydrogen production, λ should be in the area from 0 to H.

Alternatively, hydrogen is produced with the aid of micro- algae or bacteria. Ghirardi and co-workers have shown that strains of the genus Chlamodomonas can be used in this regard. These can be grown in connection with the installation and can also contribute to production of biomass. An advantage is that sunlight is not a limiting factor, something that can be the case for photosynthetic bio-cultivation.

In a conventional gas-driven power plant, the gas is burnt at a relatively low temperature to keep the formation of NO_x down.

Efficiency of combustion of natural gas

With the present invention, one solves the problem of NO_x formation as these gases are converted further to useful products, and one can therefore increase the combustion temperature. Presently, there are also new types of materials, which tolerate considerably more heat, so that increased combustion temperature does not represent a problem. Increased combustion temperature results in increased efficiency, and thus better economy for the installation.

Efficiency is defined as (T-T₀)/T. A typical combustion temperature in a thermal power plant is assumed to be about 1450 °C, i.e. about 1750 K. Theoretical efficiency: (1750- 300)/1750=82.8%.

In combustion, 270 kg air is used per second compared to 6- 6.5 kg per second natural gas (calorific value 40 MJ/Nm³) . The calculations are exemplified by using methane gas as an example of organic material

CH + 2 0₂ → C0₂ + 2 H₂0

Theoretically therefore, the following amount of oxygen is needed:

6x64/16 kg/s = 24 kg/s

The addition of oxygen is 270x0.21 kg/s * 55 kg/s, i.e. λ=2-2.5, i.e. a large excess of air which can be reduced to give a higher combustion temperature, especially if peroxide is added. A reduced total amount of air will also give less amount of fuel gas with higher temperature, which again gives a smaller volume of the waste gas. Heat of evaporation for water is 40 kJ/mol.

The fuel gas has a temperature of 610 °C from the gas turbine.

From the steam turbine the temperature is 70 °C. ϋn-co busted HC accounts for typically 3 ppm. In combustion of gas, the efficiency is increased by increasing the combustion temperature. However, this leads to increased discharges of NO_x with present technology.

One assumes that 270 kg air and 6 kg methane are used every second in the gas power process. For each mole methane (16 g) , 36 g water is formed, i.e. 36/16 x 6 kg = 14 kg water formed every second. This shows that water supplied to the fuel gas from the combustion is 14/270 = 0.05.

Thus the water content in the fuel gas increases with 5%. The total annual supply of water from the combustion area to the fuel gas phase is, therefore:

10^~3 x 270 x 0.05 x 3600 x 24 x 365 tonnes * 0.5 Mtonnes.

If the flue gases after removal of NO_x are led into a bio- production environment at a temperature of around 20 °C, one will be able to make use of the surplus heat from steam and air, and be able to make use of the C0₂ gas. The same amount of energy can also be liberated/made use of by passing the flue gas through a scrubbing liquid at a temperature of 20 °C.

The heat of evaporation for water is 2260 kJ/kg. At condensation of 0.5 Mtonnes steam, the following amount of energy is released:

2260 x 10⁹ kJ = 1130 TJ = 2260/3600 TWh = 0.3 TWh.

The annual amount of flue gas from the gas-driven power plant is about 10 Mtonnes. Heat capacity of air is: c_p = 1 kJ/(kgx°K)

If this gas is cooled from 200 to 20 °C, the following amount of heat will be liberated:

1 x (200-20) x 10 x 10⁹ kJ = 1800 TJ = 0.5 TWh

If the flue gas which is released is not at 200 °C, but 70 °C, an additional energy of about 0.15 TWh can be collected.

By being in a position to lower the surplus air, the amount of flue gas, which removes surplus heat, will also decrease.

According to SSB (Statistics Norway) 110 000 tonnes nitrogen is applied annually as nitrogen containing fertilisers by the Norwegian agricultural industry. At the same time, feed plants are being produced to an extent which corresponds to a feed value of 3.5 million tonnes barley - the nutritive value of one kg barley is defined as one feed unit.

Calculations of costs and income based on existing technology vis-a-vis the method according to the invention:

1. Consumption of hydrogen peroxide.

Industrikraft Midt-Norge presupposes an increase from 662 to 1100 tonnes NO_x per year at an increase of efficiency of 1.5%. If the efficiency increases by another 1%, one can assume that NO_x increases to about 1500 tonnes per year. If it is further assumed that the ratio between NO and N0₂ is 80/20, NO will account for 1200 tonnes. To convert from NO to N0₂ one hydrogen peroxide molecule is needed per nitrogen monoxide molecule:

NO + H₂0₂ -» N0₂ + H₂0

The molecular weight of NO is 30, while for hydrogen peroxide the molecular weight is 34. Consumption of peroxide in the first step is therefore:

1200x34/30 tonnes * 1350 tonnes peroxide

Assuming 100% conversion of NO to N0₂, there is formed

1200x46/30 tonnes * 1850 tonnes nitrogen dioxide

Together with the original 300 tonnes in the flue gas, this gives a total of 2150 tonnes N0₂. For further oxidation to HN0₃, a half molecule of peroxide is consumed:

N0₂ + H₂0₂ -» HN0₃

Consumption of peroxide in this step will be:

2150x17/46 tonnes = 800 tonnes peroxide

Thus, both steps together consume 2200 tonnes hydrogen peroxide.

The above calculations build on stoichiometric conditions where it is not taken into consideration that some of the peroxide is used in reacting with other smoke gas components such as CO and hydrocarbons (HC) , or in decomposing in the gas or on walls. In the calculations any additions of peroxide to the combustion process are not taken into account either. One must therefore calculate that an excess of about 50% hydrogen peroxide must be added to the process. In total, one can therefore calculate a consumption of 3300 tonnes hydrogen peroxide in a gas- driven power plant of this size.

2. Conversion of HNQ₃ to KN0₃.

In conversion of nitric acid to KN0₃, KCl or KOH, for example, can be used.

HN0₃ + KOH → KN0₃ + H₂0

From 2150 tonnes N0₂ one gets 2150x102/46 tonnes * 4700 tonnes potassium nitrate.

The sales price per tonne is about NOK 2-3000, which can result in a revenue of about 20 MNOK.

Potassium nitrate has the advantage that it can be used as a water based fertiliser, so-called root fertilising by way of hosepipes under ground. KOH or KCl must be bought and the cost for this must be deducted.

3. Electrolytic manufacture of peroxide.

Energy consideration

Hydrogen is normally produced by hydrolysis of water, a very energy demanding process: I ) 2 H₂0 ( 1 ) → 2 H₂ ( g) + 0₂ ( g) ΔH = 571 . 6 kJ

Thereafter, peroxide is produced by the net reaction:

II) 2 H₂(g) + 2 0₂(g) → 2 H₂0₂(1) ΔH = -375.6 kJ

This gives an energy consumption of (571.6-375.2) /2 kJ = 98 kJ per mole peroxide (34g) produced.

For 2200 tonnes peroxide, annual consumption is 2200xl0⁶x98/ (34x2200) kWh = 1.8 GWh per year.

For the use of 3300 tonnes, the energy consumption is 2.7 GWh per year. Total electricity production for the planned gas-driven power plant at Kollsnes is 3.1 TWh per year. Thus, the production of peroxide for cleaning of the flue gas consumes about 0.1% of the electrical energy. By increasing the efficiency by 1%, more peroxide can be added to the combustion chamber without the overall energy balance becoming negative, with the simultaneous achievement of an increased efficiency and low discharges of N0_X.

Thus, according to the invention, an effective combustion with increased efficiency is achieved in that the combustion temperature can be increased as a consequence of capturing the nitrogen containing gases, which are formed.

The various process steps, which are part of the method according to the present invention, are now explained, and we will describe below the units which are part of an installation according to the invention. Reference will be made to fig. 1 which shows a schematic arrangement of the units, which can be part of such an installation.

In a gas-driven power plant in Norway, gas from a gas field in the ocean (in the North sea) will be burnt. This gas is a mixture of hydrocarbons, but for simplicity it will be described as combustion of methane.

From a gas source, for example from a gas field in the ocean, methane is led to a gas-driven power plant. Methane is combusted in the flame zone of the turbines of the power plant and its after-flame zone, whereby mechanical energy, thermal energy, carbon dioxide and nitrogen oxides, predominately NO, are formed. In conventional gas-driven power plants, the waste gases, which are formed, are sent to the surrounding air by way of a chimney. According to the present invention, this chimney is removed and the waste gases from the combustion step (a) are led further into a closed pipe system, in which in one area is arranged means for supply of peroxide to the waste gas. An overview of the installation is outlined in fig. 11.

By the addition of oxidant, preferably H0, in the mentioned zones (step a, above) and furthermore in the waste gas zone of the gas works (step b, above) NO is oxidised to N0₂. NO is water soluble, something which is utilised in step (c) .

The waste gases, which are now oxidised, have now a temperature in the range 500-600 °C, and preferably are led through a heat exchanger. The thermal energy in the waste gas is transferred to the water in the heat exchanger and this is converted to steam which thereafter is led through the turbine for conversion to energy.

Thereafter the waste gases are led through a reservoir with peroxide.

By this oxidation, N0₂ is oxidised further to HN0₃. The nitric acid formed is converted after this water reservoir to nitrate by addition of a base, ammonia for making ammonium nitrate or potassium hydroxide for making potassium nitrate. The nitrate containing liquid can be used, possibly after cleaning, in biomass production. Three of the main components which are formed by a combustion and further conversion according to the invention, i.e. nitrate, C0₂ and waste heat are all central "components" in a subsequent conversion to biomass.

Experimental part

Example 1

Conversion of NO to N0₂, and CO to C0₂

The aim of the experiment was to check the degree of conversion of NO to N0₂ and CO to C0₂ under different conditions in the flue gas from the gas boiler. This is step 1 in our process, NO + H₂0₂ = N0₂ + H₂0.

To reach the aim of 80-90% total reduction in NO_x discharge into air (from 25 ppm to max 5 ppm) , we must therefore establish the conditions for a 90-95% conversion of NO to N0₂ in this step. The programme was run over a 1-2 week period and was carried out at Chalmers University in Gothenburg, Sweden.

Three different levels of N0_X were used, i.e. 50, 100, and 150 ppm. A gas-driven power station with the best economically defensible commercial technology produces about 25 ppm NO_x. The Norwegian Pollution Control Agency (SFT) has set 5 ppm as a limit, which implies a purchasing quota with the best available technology. The gas pilot cannot achieve less than about 50 ppm for NO_x. To check if the concentration of NO_x is of any significance, the different levels of 50, 100 and 150 ppm were examined.

For carbon monoxide, CO, the levels of 30, 60 and 90 ppm were examined. A typical level after combustion in a turbine will be around 50 ppm.

The excess air was set at 60, 90 and 120%. In one turbine, the excess air will be above 200%, but if the air excess can be reduced the efficiency can be increased in that the temperature of the flue gas will be higher and more energy can be taken out.

Three levels of the ratio between H₂0₂ and NO_x was examined, stoichiometrically the same, number of moles H₂0₂ injected equal to moles NO and 100% and 200% excess H₂0₂ in relation to NO, respectively.

The temperature of the injection point was kept at three different values, 550, 600 and 650 degrees Celsius, respectively. In flue gas in a gas turbine, the flue gas temperature from the gas turbine will typically be 600-650 degrees Celsius. The residence time is estimated, as this cannot be varied systematically in this test run.

The order of the experiments is randomised to remove the effects of systematic time variation in the gas pilot. All the experiments with C are so-called central runs or midpoint runs. They will expose time variation in the pilot and non-linearity in the test area. The results are given in table 1. The numbers are rounded off in the table, but the exact numbers are used in the data analysis.

Table 1

Exp Inj NO CO H₂0₂/NO_x Excess Res %NO %CO nr temp in in o₂ time unconv unconv

1 550 159 33 0.98 8.53 0. 96 44 86

2 550 50 33 2 .96 8.56 0. 96 9 86

3 550 44 33 0 .84 12.25 0 .86 54 100

5C 600 104 61 1 .75 10.51 0 .87 17 85

6 650 160 33 0 .90 12.12 0 66 51 71

7 650 145 89 2 .85 12.05 0 66 24 84

8 650 51 33 0 .96 8.53 0 78 55 78

9 650 47 89 2 .90 8.53 0 78 23 87

IOC 600 105 61 1 .80 10.57 0 85 11 85

11 550 160 89 0 93 12.21 0 98 47 92

12 550 165 33 2 73 12.04 0 99 14 86

13 550 32 89 3. 10 12.16 0 98 11 95

14 550 166 89 2. 76 8.50 0 86 11 89

15C 600 104 61 1. 93 10.45 0 88 8 77

16 650 151 94 0 .91 8.55 0 83 36 80

17 650 160 94 2. 80 8.49 0 82 14 80

18 650 45 89 0. 94 12.08 0 66 34 89

20C 600 103 89 1 84 10.53 0 88 7 89

19 650 42 33 2 60 12.20 0 61 40 86

21C 600 104 61 1 83 10.56 0 88 7 92

4 550 53 89 0 93 8.61 0 92 48 94

From the table we see directly that % unconverted NO is unacceptably high with stoichiometric addition of H₂O₂ while 100% excess H₂0₂ gives 83-93% conversion. A further excess of H₂O2 does not appear to improve the degree of conversion. Multivariate regression analysis of the test runs.

The injection temperature was transformed to its inverse, but we get approximately the same results if we keep the temperature as in the table. The numbers were centralised and all interactions were included between the six input variables in the above table. The other data which were noted at Chalmers were also included in the first analysis, but it turned out that non-combusted HC was consistently lying at around 10-15 ppm below in all the runs, C0₂ was approximately constant and had no influence on the results. Conversion of CO to C0₂ varied from 0 to 31%.

The analysis showed that excess air has no significance for the conversion of N0_X and that it thus can be possible to reduce it and achieve a higher efficiency. The temperature of the injection point for H₂O₂ proved to be of little importance, but contributed indirectly by way of interaction with residence time.

When all the test runs in the programme were analysed together, the best model could explain about 75% of the conversion of NO to N0₂. The model which was obtained also showed systematic deviations in the central runs, by the model systematically giving too large values for these. The reason for this is non-linearities and two new models were therefore constructed.

In the first of these, all experiments with a stoichiometric ratio of 1:1 between NO_x and H₂0₂ were left out. This model gave a poor agreement between measured and predicted conversion of NO. Furthermore, it gave the result that the H₂0₂/N0_X ratio did not appear to have any significance for the degree of conversion of NO. From this it was concluded that little would be gained from increasing the excess H0₂ from 100 to 200%. From considerations of process economics this is a favourable result as we can reduce the use of the input factor H₂0₂.

Modelling of the runs with H₂0₂/NO_x ratio of 1:1 and 2:1 will be presented in more detail. The best model explained 85% of the variations in the conversion of NO. We can compare the values from the model with measured values and will then get the connection which is shown in figure 1.

The points up to the right in the plot have all a 1:1 ratio between H₂0₂ and NO, while the points down to the left are central runs with a 2:1 ratio between H0₂ and NO. These are all grouped together, apart from 5C which stands apart. We can see that all central points, except 5C, have given more than 90% conversion of NO to N0₂.

To see directly the relationship between predicted and measured values for the conversion in the gas pilot, we can plot measured and model values together, as shown in figure 2.

From figure 2 it can be seen that there is a close relation between measured and predicted values.

To decide which variables are included in the explanation model for the process, and how these contribute, two different types of diagrams are presented.

Fig. 3 shows in the form of a bar-diagram which variables are important for the conversion of NO. The grey areas show to what extent the variables are explained in the model. We can see that the ratio between H₂0₂ and NO_x is the most important as nearly the whole bar is grey. The fraction CO is much less important, while the cross-product of variables 1 and 6 is very important. Variable 1 is the inverse of the temperature, 1/T, while variable 6 is the residence time.

By plotting the regression coefficients in the model, we can decide which settings will be good for high conversion. This is shown in figure 4. The height of the bars indicates the importance of the variables. Therefore, H₂0₂/NO_x is the most important and CO the least important of the three variables .

We achieve good conversion with a 2:1 ration of H₂0₂ to NO_x, i.e. a high value. Similarly, we get good conversion when CO has a high value. This means that CO has a favourable effect on the conversion of NO to N0₂.

The cross-product of inverse temperature and residence time has an opposite effect, and this is shown in figure 5. When it has a high value, we get poorer conversion. However, it is difficult to provide a good explanation of this result as the temperature and residence time are correlated in the test programme:

Therefore, a test was run where this correlation was broken. In this test the ratio H₂0₂/NO_x is kept constant at 2:1, NO in is set at 100 ppm while CO in at 200 ppm, excess air is set at 90%, while the injection temperature is varied between 500 and 600 degrees Celsius and the residence time is varied as much as possible. All the responses and input variables, which are measured in the first series, are also measured in the new series. The results obtained are given in table 2.

Table 2

Exp Temp Residence %NO nr time unconverted

1 550 0.84 10

2 600 0.66 6

3 . 600 1.04 6

4-1 550 0.85 9

4-2 550 0.84 12

5 500 0.71 62

6 500 1.04 69

7 550 0.84 10

The results show that with the flue gas at 500 degrees Celsius, the degree of conversion of NO to N0₂ drops drastically. Therefore, it is obvious that we must lie close to 550 degrees to obtain a good conversion.

Modelling showed that the inverse of the temperature as input variable explains 79% of the degree of conversion of NO. A plot of predicted unconverted NO (grey) versus measured converted NO (black) is given in figure 6.

The model predicts somewhat low values for unconverted NO at 600 degrees and 500 degrees, while the central points have a better degree of conversion than the model predicts. This is because the degree of conversion of NO varies non- linearly with the inverse of the temperature. From this example 1, one can thus draw the following conclusions :

The test runs have shown that the degree of conversion drops drastically when the temperature falls below 550 degrees. The temperature in the flue gas from a gas turbine lies above 600 degrees so that this will not be a problem in real situations.

Furthermore, the tests have shown that the optimal ratio between H₂0₂ and NO for conversion to N0₂ lies at around 2:1. At this ratio, the degree of conversion is over 90% with a flue gas temperature of 550-600 degrees. With better mixing of peroxide, we can probably go down to 1-1.5 in peroxide in relation to NO.

The extent of excess of air has no significance for the conversion. The amount of CO is also without significance for the degree of conversion. Thus, it can be possible to increase the efficiency of the power plant by reducing the air excess.

Example 2 Conversion of NO to N0₂ and S0₂ to S0₃ in the flue gas

The aim of the experiment was to check the degree of conversion of NO to N0₂ and S0₂ to S0₃ under different conditions in the flue gas from the gas boiler.

This is an extension of the investigation in step 1 of our process for a gas-driven power plant, NO + H₂0₂ = N0₂ + H₂0 The presence of S0₂ can influence the process, and it can also lead to corrosion problems if a large part of S0₂ is converted to S0₃.

The tests were carried out at Chalmers University in Gothenburg, Sweden over a 1-2 week period. The following conditions were used:

NO at the three levels of 100, 200 and 300 ppm.

S0₂ at the three levels of 200, 300 and 400 ppm.

Carbon monoxide, CO, at the levels of 15, 30 and 45 ppm.

Three values of the ratio between H₂0₂ and NO were used, equal stoichiometric ratio, injected number of moles H₂0₂ equal number of moles NO+S0₂ and 50% and 100% excess H₂0₂ in relation to NO+S0₂, respectively.

Three temperatures of the injection point were also used, 550, 610 and 670 degrees Celsius respectively.

Excess air was set at 60%.

The residence time before measuring S0₂ and NO was set to about 0.8 seconds .

The order of test is randomised to eliminate systematic time variation effects in the gas pilot. All experiments with C are so-called central runs or centre- point runs. They will disclose time variation on the pilot and non- linearity in the test area. Experiment 15 failed, otherwise the experimental number gives the order in which the tests were carried out. Table 3 shows the results of the tests. The numbers are rounded off in the table, but exact numbers were used in the data analysis.

Table 3

Exp Inj NO CO H₂0₂/ so₂ %NO %S0₂ nr temp in in NO_x+S0₂ in unconv unconv

1 550 311 15 0.93 202 20 94

2 549 107 15 1.92 199 11 93

3 550 108 15 0.99 395 12 100

4 549 104 45 0.97 204 11 100

5C 612 217 30 1.47 290 13 99

6 671 305 15 1.02 389 58 99

7 670 328 45 1.94 408 50 98

8 670 105 15 1.06 205 58 101

9 670 98 45 2.15 208 51 97

10C 610 230 30 1.40 294 10 100

11 550 315 45 1.16 384 14 96

12 550 324 15 2.01 391 13 95

13 551 102 45 1.96 409 12 98

14 550 317 45 1.90 205 13 97

16 671 324 45 0.95 198 54 95

17 670 314 15 1.95 192 42 97

18 670 102 45 0.75 417 33 100

19 670 108 15 1.96 414 23 101

20C 610 216 30 1.47 293 14 96

From table 3 it appears directly that % unconverted NO is unacceptably high at max temp of 670 degrees Celsius whatever the stoichiometric addition of H₂0₂. This is probably because that peroxide is broken down into non- reactive components when the temperature is too high. At lower temperatures we have 85-90% conversion, except in the first experiment which only gives 80% conversion.

Furthermore, it can be seen that S0₂ is not very reactive in the gas phase reaction with peroxide. This is very favourable as we thereby obtain an approximately selective reaction for NO. Thereby, we can reduce consumption of chemicals and minimise corrosion problems.

The data are analysed with multivariate regression analysis, and a short summary of this analysis is given below, including the most important conclusions from this series .

The injection temperature was transformed to its inverse, but about the same results appear if we keep the temperature as in the table. The data were centralised and all interactions were noted between the five input variables in the above table 3.

The temperature of the injection point for H₂0₂ showed to be of little significance, but contributes indirectly by way of interaction with residence time (discussed below) . When all the test runs were analysed together, the best model explains about 81% of the conversion of NO to N0₂.

Figure 7 shows the deviation between measured and predicted unconverted NO. The model shows systematic deviations in the central runs, in that the model systematically gives too high values for these. The reason is non-linearities.

To decide which variables that are included in the explanation model, and how these contribute, two different diagrams were constructed.

A plot of regression coefficients is shown in figure 8. The height of the bars describes the importance of the variables. We can see that the inverse of temperature dominates completely. By going up to maximum temperature, unconverted NO increases drastically.

Fig. 9 shows a bar diagram that demonstrates which variables are important for the conversion of NO. The grey parts show to what extent the variables are explained by the model. We can see that temperature is clearly the most important as nearly the whole bar is grey.

To directly see the relationship between predicted values for the conversion and measured values in the gas pilot, we can plot measured and predicted values together, and this is shown in figure 10. We can see that there is a very good agreement between measured and predicted values.

From the tests carried out in this example 2, we can draw the conclusion that the degree of conversion drops drastically when the temperature rises up towards 670 degrees Celsius. From earlier experiments we know that a reduction in temperature to 500 degrees has the same effect. However, a very dilute solution of peroxide was used here (much water) in the pilot, which leads to a reduction in flue gas temperature of at least 50 degrees. The temperature is thereby much lower after mixing and when the peroxide is activated. We will not get the same cooling effect in a full-scale installation because of the large amounts of gas present.

Example 3

Conversion of N0₂ and S0 by leading the gases through a liquid

Table 4 shows the results from a set-up in which gas containing N0₂ is bubbled through a liquid with 1% H₂0₂ (in a flask) . The temperature in the liquid is varied, as is also the volume of liquid.

Table 4

From the results given in the above table it can be seen that the conversion to nitric acid can be increased by increasing the volume of the absorption liquid. Furthermore, the conversion is most effective at the lowest temperature, which was used in this test, and this indicates that the waste gases, which are led out from the combustion/conversion chamber, should preferably be cooled. Furthermore increased contact time will give increased conversion of N0₂ to HN0₃.

To test the conditions which can influence the contact time, and to test how effective the conversion of S0₂ to sulphuric acid is, a test was set up in which a column was used as a model for the bubble tank. The column is a glass cylinder with a diameter of about 5 cm. Glass balls with a diameter of 10 mm are placed in the glass cylinder to a height of 100 mm. The cylinder is filled with liquid to a height about 1 cm above the balls. Gases are led into the column through a feeding tube at the bottom of the column. The gases rises up through the column and leaves the column by way of a tube which is placed a distance above the surface of the liquid. The whole arrangement is immersed in a water bath to ensure constant temperature. The rate of flow of gas was 5.7 1pm.

The gas mixtures which were used in the experiment are generated in a flue gas generator and consist of S0₂ and N0₂. The experiments were carried out with different concentrations of H₂0₂ and at different temperatures. Furthermore, different liquids were tested, namely brackish water, fresh water and seawater. The results are given in table 5. Table 5

Conversion of S0₂ to sulphuric acid works very well, with approximately all S0₂ being converted. Furthermore, the test shows that the contact time in the "bubble tank" is too short for an effective conversion of N0₂ to nitric acid, and we are presently carrying out further tests to establish which contact time is sufficient. Increased contact time can be achieved, for example, by increasing the height of the column.

Claims

1. Method for combustion and/or thermal degradation (such as cracking) of organic/biological material at temperatures where waste gases containing nitrogen oxides (N0χ) are formed, characterised in that the waste gases, containing nitrogen oxides (NO_x) and/or sulphur oxides (SO_x) , which are formed during the combustion or the thermal degradation are led into a closed circuit, in which peroxide is added to oxidise the nitrogen oxides that are formed.

2. Method in accordance with claim 1, characterised in that the peroxide is preferably hydrogen peroxide but can also be derived from H₂0₂ in which one of the hydrogen atoms is substituted, for example, with an alkyl group.

3. Method in accordance with claim 1, characterised in that the addition of peroxide to the waste gases occurs at the temperature the waste gases have after the combustion/conversion and before any cooling of the waste gases in a heat exchanger, by quenching or the like.

4. Method in accordance with one of the claims 1-3, characterised in that the peroxide is pre-heated to 150-700 °C before being added to the waste gases.

5. Method in accordance with one of the claims 1-4, characterised in that the waste gas, at the addition of hydrogen peroxide, has a temperature in the area 500-700 °C.

6. Method in accordance with claim 1, characterised in that the waste gas, after oxidation in the flue gas channel (12), is led by way of a heat exchanger (16), in which water is converted to steam which thereafter is led through a steam turbine (18) for recovery of energy.

7. Method in accordance with claim 1, characterised in that the waste gas is bubbled through a liquid of peroxide dissolved in water, in which nitrogen oxides are oxidised to nitric acid and any sulphur dioxide is oxidised to sulphuric acid.

8. Method in accordance with claim 7, characterised in that to the liquid is added an alkaline compound, preferably potassium hydroxide or ammonium hydroxide, to convert nitric acid to potassium nitrate and ammonium nitrate, respectively, and any sulphuric acid to potassium sulphate and ammonium sulphate.

9. Method in accordance with one of the claims 1-8, characterised in that peroxide is added during the combustion or the thermal conversion which occurs in the combustion/conversion chamber (12) to increase the efficiency of the process.

10. Method in accordance with one of the claims 1-9, characterised in that the access to oxygen is reduced by using a reduced air excess to increase the efficiency of the process.

11. Method in accordance with one of the claims 1-10, characterised in that peroxide is added to the process stepwise as follows:

a) directly to the combustion/degradation chamber (12) , b) to the waste gas which is led out of the chamber (12), and c) to the liquid in the bubble tank (20) through which the waste gas, preferably after cooling by way of a heat exchanger, is led.

12. Method in accordance with one of the claims 1-11, characterised in that a catalytic agent is also added in one or more of the steps for addition of peroxide, or that one physically ensures a catalytic conversion of peroxide to the corresponding radicals.

13. Method in accordance with claim 11, characterised in that the catalysis of the conversion of peroxide is brought about by a supply of UV light, plasma discharges or other electromagnetic effects.

14. Method in accordance with one of the claims 1-13, characterised in that parts of the C0₂ which are formed during the process are converted to carbonate or bicarbonate by the waste gas, after conversion to and removal of nitric acid and any sulphuric acid, being bubbled through an alkaline solution with peroxide.

15. Method in accordance with one of the claims 1-14, characterised in that addition of peroxide to the waste gas is regulated so that S0₂ is not converted to S0₃, but that conversion of S0₂ to S0₃ takes place as the waste gas is bubbled through the bubble tank (20) with an aqueous solution containing hydrogen peroxide, to thereby produce H₂S0₄ in this tank (20) .

16. Method in accordance with claim 15, characterised in that small amounts of sulphuric acid (H₂S0₄) are added to the bubble tank to catalyse the conversion of N0₂ to HN0₃.

17. Installation (10) for combustion and/or thermal degradation of organic material, characterised in that the installation (10) comprises the following units:

a) chamber (12) for combustion and/or thermal conversion of the organic material, b) pipe through which the waste gases which are produced in the chamber (12) are led, and in which this pipe comprises means, such as nozzles, for addition of peroxide.

18. Installation in accordance with claim 17, characterised in that the installation further comprises: c) a heat exchanger (16) through which the hot waste gases which are produced in the chamber (12) are led, for exchange of heat with a medium, preferably water in liquid form, which at the heat exchange is converted to steam.

19. Installation in accordance with claim 13, characterised in that the steam from the heat exchanger (16) is converted to energy in a steam turbine.

20. Installation in accordance with one of the claims 17- 19, characterised in that the waste gas, preferably after having been cooled by way of the heat exchanger (16), is bubbled through a bubble tank (20), and that means for addition of peroxide are arranged in this bubble tank (20) .

21. Installation in accordance with claims 17-20, characterised in that a supply tube is arranged in the chamber (12) for addition of peroxide to the flame zone in the chamber (12) .

22. Installation in accordance with one of the claims 17-

21, characterised in that the installation (10) is a gas- driven power plant.

23. Installation in accordance with one of the claims 17-

22, characterised in that the organic material, which is combusted/broken down, is natural gas or components of natural gas such as methane.

24. Installation in accordance with one of the claims 17-

23, characterised in that the installation comprises a unit for manufacture of hydrogen peroxide in situ.

25. Installation in accordance with one of the claims 17- 24, characterised in that the different components of the installation are connected together in a closed cycle.

26. Installation in accordance with one of the claims 17- 25, characterised in that in connection with the installation are arranged facilities which can convert further the chemical compounds which are generated and/or use the energy which is produced, and/or use the water masses that are generated at a temperature above ambient temperature .

27. Installation in accordance with claim 26, characterised in that the mentioned facilities comprise, for example, units for production of biomass such as algae.

28. Installation in accordance with claim 27, characterised in that the installation comprises facilities for utilisation of the produced algae as raw materials for feed production and/or health food products and/or medicinal products.

29. Installation in accordance with claim 26, characterised in that the mentioned facility is a fish farm.