GB1596809A - Process for treatment of exhaust gas - Google Patents

Description

(54) PROCESS FOR TREATMENT OF EXHAUST GAS (71) We, KUREHA KAGAKU KOGYO KABUSHIKI KAISHA, a company organised under the laws of Japan, of No. 8, Horidomecho, I-chome, Nihonbashi, Chuo-ku, Tokyo, Japan, do hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement:- This invention relates to a process for the treatment of an exhaust gas containing at least sulfur oxides (hereinafter referred to as "SOx") and oxygen.

Generally, combustion process exhaust gases such as those discharged from boilers and other combustion devices, contain SOx and oxygen.

Some, if not all, of these exhaust gases contain nitrogen oxides (hereinafter referred to as "NOx") in concentrations of the order of from 100 to a few hundred ppm in addition to SOx which are usually present in concentrations of the order of from 100 to 3,000 ppm.

It has been proposed to treat such exhaust gases e.g. by contacting it with the aqueous solution of an alkali metal sulfite or a slaked lime slurry thereby causing absorption of the SOx and thus freeing the exhaust gas from the SOx: such a process has been called a "wettype process". Where an exhaust gas is treated by a "wet-type process", 'however, it is not unusual that the residual SOx content amounts to e.g. 50 to 300 ppm. Moreover, the absorption of SOx gives rise to dithionic acid byproduct in the aqueous solution or the slurry.

Dithionic acid is formed when the SOx are absorbed in the form of sulfites and the sulfites are then oxidized into sulfates by the oxygen present in the exhaust gas. The dithionic acid thus formed accumulates in the form of a salt in the aqueous solution or the slurry and consequently brings about a reduction in the concentration of the SOx removing component.

The discharge, as waste, of an aqueous solution or slurry containing dithionic acid in unaltered form into natural water courses increased the COD (chemical oxygen demand) value thereof.

This is highly undesirable for environmental reasons.

The NOx which may be present in the exhaust gas is usually removed from the exhaust gas by contact with ammonia in the presence of a catalyst. This proposal however, has the drawback that the activity of the catalyst is impaired by residual SOx present in the exhaust gas. In order that the catalytic reduction of NOx by ammonia may be carried out over a long period at a stable level, therefore, the exhaust gas is desirably free from SOx prior to the treatment. It is known that, of the various SOx contained in the exhaust gas, SO3 can be removed to a fairly great extent by means of a wet type electrostatic precipitator.

This wet type electrostatic precipitator, however, is entirely ineffective with respect to SO2 .

This means that even a gas which has been treated by the electrostatic precipitator will poison the catalyst being used for the NOx reduction. Although the SO2 can be removed substantially completely from the exhaust gas when it is thoroughly contacted with the aqueous solution of an alkali metal salt, the gasliquid contact effected by previously proposed methods requires the use of a large volume of the solution and inevitably entails the occurrence of dithionic acid.

The present invention is predicated on our observation that if the aqueous alkali metal salt solution in the wet type treatment of an exhaust gas contains dissolved heavy metal ions such as iron, cobalt, nickel and vanadium these ions function catalytically in the by-production of dithionic acid; among these heavy metal ions, we observed that the catalytic activity of iron ion is more than ten times that of any other tested metal ion at equal concentration.

We further found that in cases where a heavy metal such as iron is left for a long period in an aqueous alkaline solution of an alkali metal salt, such as the solutions used for absorption of sulphur oxides, the metal ion is substantially undetectable in the solution. However, in cases where such a heavy metal is immersed in a neutral or acidic aqueous solution of the same alkali metal salt the heavy metal is detected in the solution and the lower the pH of the solution, the larger the amount of the ion in the solution, probably because of corrosion of the metal, for instance iron, by the neutral or acidic aqueous solutions. Consequently, the amount of by-produced dithionic acid increases proportionately as the pH of the aqueous solution shifts from neutral to acid values. We have additionally leamt that the incidence of this dithionic acid can be reduced if the oxidation of the sulfites to sulfates is caused to proceed rapidly.

Broadly speaking in our present process exhaust gas is brought into contact with aqueous alkali solution of pH at least 8 and substantially free from heavy metal ions.

According to the invention we provide a process for removing sulphur oxides from a gaseous exhaust mixture containing at least sulphur oxides and gaseous oxygen, wherein said gaseous exhaust mixture is contacted, in a spray column, with droplets of an aqueous solution of an alkali metal salt and/or an alkali metal hydroxide to absorb the sulphur oxides and to oxidize absorbed sulphur dioxide with the oxygen of the said gaseous exhaust mixture, the surfaces of the spray column contacted by said aqueous solution being of non-metallic material, the droplets having a particle size of 1-109CL diameter, the solution having a pH of at least 8, and the solution being introduced into the spray column at a rate of 0.5 to 5 litres per 100 Nmofthe gaseous exhaust mixture; and the thus treated exhaust mixture being introduced into an electrostatic precipitator in which finely divided droplets and dust are removed from the gaseous exhaust mixture.

The accompanying drawing is a graph showing the relationship between the pH value of the aqueous solution of sodium sulfite containing dissolved iron (namely, iron ions) in various concentrations and the amount of sodium dithionate produced in the aqueous solution, when determined by blowing air into the aqueous solution..

A spray column in preferred embodiments of the invention has an inner wall coated with a synthetic resin rubber, or is of fibre reinforced plastics material ("FRP"). If a spray column includes surfaces which are directly contacted with the solution and which are not of a nonmetallic material, e.g. a spray column which is wholly made of iron, the iron is gradually dissolved out, thus providing ions in the aqueous alkali salt solution; the iron ions inevitably accelerate the formation of dithionic acid in the solution. For use in the practice of the present invention, the aqueous treatment solution of an alkali metal salt and/or alkali metal hydroxide may be any one of those which have conventionally been used in wet type processes for the treatment of exhaust gases; preferably sodium hydroxide or sodium carbonate solutions are used. In order that the amount of heavy metals dissolved in this treatment solution may be prevented from increasing, the treatment solution is required to have a pH maintained at a value of at least 8. Further, to ensure acceleration of the oxidation of the absorbed sulfites into sulfates the aqueous solution must be in the form of finely divided droplets. Heavy metals which may be found to be present in the treatment solution possibly originate by entrainment in the exhaust gas from the combustion process and as impurities in the alkali metal and/or alkali metal hydroxide used to prepare the treatment solution. The treatment solution may be carried out by any conventional procedure.

The finely divided droplets of the treatment solution can be produced by forcing the solution through an ultrasonic nozzle or two-fluid nozzle: this procedure may be carried out by forcing a compressed gas (e.g. an oxygencontaining gas or steam) through the nozzle in conjunction with the solution. After leaving the nozzle, the solution, in the form of finely divided droplets, is thoroughly mixed with the exhaust gas within the spray column, and suspended in the exhaust gas thereby to absorb the SO2 efficiently which is thence oxidised; this occurs, with rapidity due to the large surface area of contact, by reaction with the oxygen present in the exhaust gas. In order to increase the rate of the absorption of SO2 and the oxidation of the absorbed SO2, and to improve the operational efficiency of the spray column, the droplets produced by spraying the aqueous solution through the nozzle preferably have as small a particle diameter as possible. The droplets have a particle diameter of from lu to 100it, preferably in the range of from 1 to 30it, more preferably from 20 to 30cor. Bearing in mind the requirement that these droplets are to be collected by an electrostatic precipitator in the subsequent step of process, the particle diameter of the droplets is greater than lu.

Even after the aqueous solution has abasorbed the SO2 from the exhaust gas, it is still preferred that the solution has a pH value above 8. Any drop of the pH value below the level of 7.7, after absorption, has been found to be undesirable for the purpose of thorough removal of SO2, because at the reduced pH value, bisulfite is formed in the aqueous solution which creates a partial pressure of SO2.

Particularly, even if the droplets have a high pH value at the time of their charge through the spray nozzle, the pH value of the droplets nevertheless is liable to drop sharply owing to the absorption of SO2 and the subsequent formation of S03 solution. The aqueous solution of an alkali metal salt and/or alkali metal hydroxide to be used for spraying in the column, therefore, is preferred to have a sufficiently high concentration to ensure that the pH of the solution does not drop below the lower limit mentioned above. Where raw exhaust gas contains SOx in a concentration of 100 to 200 ppm, for example, the aqueous solution of an alkali metal salt and/or alkali metal hydroxide is preferred to have a concentration of from 2 to 10%, specifically about 5%.

In this case, the amount of the aqueous solution of alkali metal salt and/or alkali metal hydroxide to be used in preferably from 1 to 3 /hour, per 100 Nm3 /hour of the exhaust gas under treatment. This amount is only several percent, or even less, of the amount required by a conventional absorption method.

Where such a small amount of aqueous solution is sprayed in the form of finely divided droplets, the pressure loss within the spray column is so small as to make it possible to increase the flow rate of the gas to the order of from 5 to 20 m/sec, preferably from 8 to 15 m/ sec, a range which is large as compared with the range of from 2 to 3 m/sec which is usual within conventional practice. Better still, in this increased range, any slight variation in the flow rate has no appreciable effect on the overall operation. At such a gas flow rate, a residence time of about 0.1 second may suffice for desirable absorption of SO2.

The part of the spray column which does not come into direct contact with those droplets which have already been contacted with SO2 may be made of iron containing material.

Practically it is, therefore, permissible for the spray nozzle of the column to be made of stainless steel, for example.

In preferred embodiments the absorption operation not only permits the absorption of SO2 and the oxidation of the absorbed SO2 to proceed with high efficiency but also offers economic advantage in other respects when compared with similar operations carried out using a conventional packed column, plate column, wetted-wall column or Venturi scrub ber. For example, in the practice of the invention a spray column may be used which involves a pressure loss therethrough of the order of from 10 to 20 mm. H2 0, whereas a similar operation using a Venturi scrubber involves a pressure loss of about 10 times as great. Compared with operations using other conventional absorption devices, the amount of absorption per unit internal volume of an absorption column in preferred operations can be quite large even though the amount of absorbate to be consumed is very small (of the order 10). Consequently, preferred operations in accordance with the present invention enjoy the advantages that the spray column is small in size and the effluent is small in volume in comparison with conventional procedures.

Further, preferred operations involve highly efficient gas-liquid contact and high rate of absorption; very short residence time (of the order of 0.1 second); and a small pressure loss which permits a high rate of flow of exhaust gas therethrough. The column height, need not therefore be considerable in order for the column to function effectively. The spray column may, therefore, readily be improvised out of a duct or some other similar article available to hand and there is no need to use in the installation any of the conventional absorption columns. A vertical spray colum is preferred.

However, we have found that even if the column is arranged horizontally, satisfactory absorption of SO2 may be achieved if the droplets of the aqueous solution are so fine that most of them are entrained in the current of the exhaust gas through the column.

Subsequently, in the practice of the present invention, the exhaust gas, whose SO2 content has, for instance, been lowered below 1 ppm, preferably below 0.1 ppm, is introduced into an electrostatic precipitator, wherein the exhaust gas is freed from fine particles (such as, for example, SO3 mist and the droplets of aqueous solution entrained by the exhaust gas) contained therein. Generally, after entry of the exhaust gas into the electrostatic precipitator, absorption of SO2 and the oxidation of the absorbed SO2 continue to proceed. In the gas passing from the electrostatic precipitator, the SOx (SO2 and SO3 combined) content is below 1 ppm. Preferably, a mist arrester is interposed between the spray column and the electrostatic precipitator and operated so as to remove from the current of the exhaust gas those floating droplets capable of being removed by a mechanical means, namely, those droplets having diameters exceeding 30 ,u. Consequently, the frequency with which short circuits may occur between the opposite poles of the electrostatic precipitator is lowered; this alleviates the load imposed upon the precipitator.

Through the treatments described above, the SOx content of the exhaust gas can be lowered to below the level of 0.1 ppm. At the time of the absorption of SOx, there is a possibility that part of the absorbed SOx will be converted into dithionic acid. As already described, dithionic acid is a substance which adds to the COD value of effluent and is fairly resistant to degradation; it is, therefore an important factor in the consideration of effluent disposal: this has been discussed above. For these reasons formation of dithionic acid must be avoided if at all possible.

The two essential conditions in the absorption stage of our process, namely pH and nonmetallic surfaces, assist in depressing the formation of dithionic acid. If SO2 absorption and/ or oxidation of the absorbed SO2 in the droplets is not completed within the spray column there is a possibility that dithionic acid will be formed when the exhaust gas enters the electrostatic precipitator. It is, therefore, preferred that the surfaces of the electrostatic precipitator to be directly contacted with the aqueous solution are formed of a non-metallic material. It is likewise preferred that the part of the electrostatic precipitator required to have high electroconductivity is made of a material chosen from mixtures of carbonaceous substances and any of synthetic resins, lead, titanium and tantalum. A composite material made up of a carbonaceous substances and a synthetic resin seems to be the most economically advantageous.

As mentioned above, if the removal of SO2 from the exhaust gas within the spray column is insufficient, it is possible that an increased amount of dithionic acid will be formed within the electrostatic precipitator. When SO2 was injected in an amount giving a concentration of 100 ppm, by way of a controlled experiment, into the line interconnecting the spray column and the electrostatic precipitator formation of dithionic acid was found definitely to occur within the electrostatic precipitator even when the part of the electrostatic precipitator directly contacted with the aqueous solution was made of a non-metallic material. The production of dithionic acid is ascribable to the heavy metal components such as Fe, V and Ni which collect within the electrostatic precipitator.

The amount of dithionic acid thus produced will decrease as the amount of SO2 entering the electrostatic precipitator decreases. It has been observed that the ratio of the amount of dithionic acid formed to the amount of SO2 injected as aforesaid decreases sharply as the SO2 content decreases to a certain level and that the amount of dithionic acid produced greatly decreases especially when the SO2 content of the exhaust gas at the outlet of the electrostatic precipitator is lower than 1 ppm. For the purpose of depressing possible formation of dithionic acid due to the heavy metals originating in the exhaust gas, therefore, it is preferred that the absorption of SO2 and/or the oxidation of the absorbed SO2 is thoroughly completed before the exhaust gas entraining the droplets enters the electrostatic precipitator. In trials we have carried out, when this condition is fulfilled, formation of dithionic acid can no longer be observed within the electrostatic precipitator. Since there is a possibility that the exhaust gas will be led into the electrostatic precipitator before the absorption of SO2 and/or the oxidation of the absorbed SO2 has been completed, the amount of the aqueous alkali metal salt and/or alkali metal hydroxide solution which is entrained in the current of the exhaust gas from the spray column into the electrostatic precipitator is preferably sufficient to provide thorough neutralization of S03 and at least avoid shifting the overall pH level toward the acidic side. It is preferred that this condition is fulfilled within the spray column. If this condition is satisfied within the electrostatic precipitator, it means that the condition is fully met also within the spray column.

The absorption of SO2 and oxidation of sulfite ions can be completed within the spray column, e.g. when the residence time within the spray column is greater than 0.1 second, preferably 0.2 second. In such a case, therefore, the surfaces of the electrostatic precipitator which come into direct contact with the aqueous solution need not be formed of a nonmetallic material.

The solution which collects in the bottom of the spray column may be made up with an alkali metal salt and/or an alkali metal hydroxide and recycled while at least a part thereof is transferred to the liquid collection unit of the electrostatic precipitator and neutralized.

Generally, the exhaust gas from a conventional boiler may contain SO3 approximately in an amount of from 20 to 50 ppm where the exhaust gas happens to be rich in SOx. If the greater part of the alkali metal salt and/or alkali metal hydroxide is used up in absorbing SO2 in the spray column the solution collecting in the bottom of the electrostatic precipitator may become acidic. Thus, the solution from this electrostatic precipitator is conveniently combined with the solution collecting in the bottom of the spray column, so that the resultant mixture will have a neutral to alkaline pH value when discharged.

Although the risk of the oxidation of sulfites inside the electrostatic precipitator is far less serious than in the absorption column, the droplets of the aqueous solution within the electrostatic precipitator tend to assume an acidic pH value. It is, therefore, advantageous that the process conditions are selected so that the droplets are entrained by the current of the exhaust gas from the spray column into the electrostatic precipitator there to be used for neutralizing collected SO3 mist. It is preferable that for the sake of precautions, the absorbate collecting in the bottom of the spray column should be injected into the bottom of the elec trostatic precipitator. It is important that the solution discharged from the bottom of the electrostatic precipitator is not allowed to be acidic.

In the case of an exhaust gas which contains NOx in addition to SOx and oxygen, the exhaust gas may be first treated as described above to be freed substantially from SOx and then passed, in conjunction with ammonia, through a catalyst bed. By this operation, the reduction of NOx can be contained for a long time without entailing any degradation in the activity of the catalyst. The catalyst reduction of NOx, can be carried out using any of the various known catalysts for NOx-reduction.

The catalytic reduction of NOx can be carried out at a temperature over the range of from 100 to 5000 C. For thermal economy, however, the catalytic reduction is preferred to be carried out at the lowest possible temperature.

For this reason, it is preferred that the reaction is carried out in the presence of a catalyst which is active at a relatively low temperature, for example, a catalyst made up of manganese oxides alone or in conjunction with Fe, Ni, Co, V. W. Cr, Sn, Ti and Zn which is active at temperatures in the range of from 100 to 220 C, among the various manganese oxides available, manganese carbonate and/or calcined rhodochrosite appear to give the best results.

The known catalysts such as those including noble metals like Pt, Ru and Rh, those using Co as a basic component, those made predominantly of activated carbon and those consisting of activated carbon coated with various metal oxides and with ammonium salts can be adapted for the reduction of NOx on condition that the exhaust gas has been treated so as to decrease its SOx content to an extremely low level.

The present invention can be practiced even on an exhaust gas which contains SOx in an amount exceeding 1000 ppm. Best results seem to be obtained in the treatment of exhaust gases whose SOx content is not more than 500 ppm, preferably not more than 100 ppm. For the removal of NOx from the exhaust gas this is advantageously accomplished by first subjecting the exhaust gas to SOx-removal so as to have its SOx content lowered below about 100 ppm, in accordance with the invention.

The invention will now be further described with reference to the following examples.

Example 6 is included to demonstrate the catalytic activity of various metals in the formation of dithionic acid and Example 8 is included to demonstrate the activity of various catalysts in removing NOx from the treated exhaust gas.

Example 1: The exhaust gas from a boiler using as its fuel a high-sulfur heavy oil containing 2.8% by weight of sulfur was subjected to a wet type treatment for SOx-removal using sodium sulfite, to afford an exhaust gas containing 44 ppm of SO2, 28 ppm of S03, 175 ppm of NOx, 4.5% of 02, 66 mg/Nm3 of soot and about 15% of water and having a temperature of 58 C.

This exhaust gas was fed downwardly into a cylindrical duct of FRP about 80 cm in diameter at a flow rate of 10,000 Nm3 /hour (with the flow rate of the gas fixed at about 7 m/sec.).

Simultaneously and co-currently with the flow of the exhaust gas, an aqueous caustic soda solution of pH 11 was spurted into the duct interior through an ultrasonic nozzle at a rate of 150 /hour so as to produce finely divided droplets about 1 to 30 ,ll in particle diameter.

The exhaust gas leaving the duct was led into another duct disposed horizontally at a distance of about 4.5m from the nozzle, then passed through a mist arrester and led into an electrostatic precipitator. With the gas-liquid contact effected in a very short time (about 0.6 second), the absorption and oxidation of SO2 were substantially complete. In the gas emanating from the mist arrester, the SO2 content was found to have been lowered to the order of 0.01 to 0.02 ppm. While the greater part of the sprayed droplets was collected in the bottom of the FRP duct, about one third of the droplets (particularly those of relatively small particle diameters) were entrained by the current of exhaust gas into the electrostatic precipitator.

The electrostatic precipitator having an interpole space of 100 mm was operated at 40 KV and 32 mA to free the exhaust gas from SO3, soot and sprayed droplets. The matter thus removed was allowed to flow down the collector interior and was accumulated in the bottom of the collector.

After the treatment given in the electrostatic precipitator the exhaust gas was found to contain less than 0.01 ppm of SO2 and less than 0.02 ppm of SO3 and an amount of soot incapable of detection by the method specified by JIS (Japanese Industrial Standard) and had a temperature of 56 C.

The absorbate which had been collected in the bottom of the FRP duct was found to contain by-produced Glauber's salt in addition to caustic soda and was confirmed to contain no discernible amount of sodium bisulfite, sodium sulfite or dithionic acid. The pH value of the absorbate was about 8.8 and the Na2 52 06/ Na2SO4 ratio was less than 1/10000. The absorbate in the bottom of the spray duct was wholly transferred with the SO3 collected by the precipitator. The neutralized effluent was fed to the side of the main SOx-removing unit of the sodium sulfite process. The effluent was found to contain dithionic acid, a substance liable to increase the COD value of the final effluent, less than 1/10000 in terms of the Na2 5206/Na2 S04 molar ratio.

In this case, the spray duct and the mist arrester were wholly made of a synthetic resin.

In the electrostatic precipitator, the main body was made of FRP, the precipator unit and other parts required to have high electroconductivity were formed of an uncorrodable unsaturated polyester and carbonaceous fibers and other parts required to have particularly high electroconductivity were made of titanium material.

The exhaust gas which had been treated as described above was heated and then fed at a space velocity (SV) of 7,500 hour , in conjunction with NH3 in a concentration of 170 ppm, to a catalyst bed packed with cylinders of catalyst 2mm in diameter and 15 mm in length and prepared from what had been obtained by baking rhodochrosite at 4000 C, so as to induce a catalytic reaction at 1600 C. Consequently, the NOx content fell to 2 ppm, indicating that 98.8So of the NOx was removed.

The operation was continued for a period of 850 hours. During this operation, the extent of NOx-removal and the pressure loss in the catalyst bed were found to involve variations which invariably fell within the ranges of allowable errors.

The gas which had undergone the NOx- removing treatment and was released into the atmosphere was found to contain extremely small amounts of harmful substances, i.e. not more than 0.03 ppm of SOx, 2 ppm of NOx and less than 1 ma/Nm3 of soot. It has thus been ascertained that, despite the use of a fuel of very inferior quality, the extent of air pollution by the gas resulting from the treatment in accordance with this invention was extremely small.

Example 2: The same exhaust gas as used in Example 1 was fed at a flow rate of 300 Nm3 /hour into a spray column of transparent PVC (Polyvinyl chloride) pipe about 10 cm in diameter and 2.3 m in length so that it flowed at a velocity of about 11 Nm/sec inside the column and the absorbent was sprayed downwardly through the two ultrasonic nozzles disposed one at the gas inlet of the duct and the other at a poin 1 m below the gas inlet to free the gas from its impurities.

An aqueous sodium carbonate solution of about pH 12 was used as the absorbent. This absorbent was sprayed through the lower nozzle at a rate of 5 /hour into the raw exhaust gas containing 44 ppm of SO2 so that most of the droplets resulting through the upper nozzle at a rate of 5 /hour.

When the flow rate of the exhaust gas was 300 Nm3 /hour or over, part of the exhaust gas was separated and treated by the electrostatic precipitator in the same manner as above. In all the test runs, the exhaust gas at the outlet was found to contain less than 0.1 ppm of SO2, less than 0.1 ppm of SO3 and less than 1 mg/Nm3 of soot. The liquid finally disposed of was found to contain dithionic acid in an amount of less than 1/10000 in terms ofNa2 S2 6 /Na2 S04 molar ratio.

Example 4: With the same apparatus and under the same conditions as used in Example 3, excepting the SO2 concentration in the raw exhaust gas which was fixed at 150 ppm and the flow velocity of the gas inside the spray column which was fixed at 11 Nm/sec., the treatment of the exhaust gas with the absorbent was carried out with test pieces of various materials placed at prescribed positions in the reaction systems, to determine the effect of the presence of the materials in the system upon possible byproduction of dithionic acid.

Each test run was continued for the fixed period of five hours and the data of the test run were those obtained over the entire period of five hours. On completion of each test run, the interior of the entire system was cleaned and the absorbate which had been collected therein was thoroughly removed before the subsequent test run was started.

All the test pieces used in the test runs were prepared in the shape of a plate 3 cm in width and 30 cm in length. In the spray column, one test piece was hung longitudinally down from the upper portion along the wall. In the precipitator, one test piece was hung down along the wall extending from the gas inlet on the spray column side to the discharge zone.

In the test runs using test pieces made of carbon, titanium, tantalum and FRP incorporating carbonaceous fibers, no increase in the dithionic acid content was observed in any of the absorbates.

a) In the test run wherein test pieces of ordinary steel were placed in the spray column and the precipitator, the molar ratio of N2 S2 06 /Na2 SO4 was found to be 4/10000.

b) In the test run wherein a test piece of stainless steel (SUS-304) was placed only in the precipitator, the molar ratio was found to be 2/10000. c) The result was the same when only the discharge pole of the precipitator was made of the same stainless steel (SUS-304). In all these test runs, the conditions on the exhaust gas side were kept on similar levels to those of Example 2.

Example 5: Various hevy metals believed to function catalytically in the formation of dithionic acid within the spray column and the precipitator were dissolved in solutions of SO2 with pH values in the range of from pH 3 to pH 8 and subjected to a flask test, to determine the effect of the dissolved heavy metal upon the by-production of dithionic acid.

Each flask was charged with 400 m of an aqueous solution containing 10% by weight of sodium sulfite and a prescribed amount of a given heavy metal. The liquids in the flasks were heated to 580 C, adjusted in pH value by use of caustic soda and, while under continuous stirring, were aerated by air which was blow upwardly from the bottom at a rate of 15 Qlhour.

The contents of the flasks were thus allowed to undergo reaction for about one hour. At the end of the treatment, the contents were analyzed.

The accompanying graph represents the relationship between the amount of iron dissolved in the liquid andthe amount of dithionic acid produced in the liquid possibly as a function of the pH of the liquid. In the drawing, the vertical axis shows the amount of Na2 S2 06 produced (x10-3 mol/ h), and the horizontal axis pH of the aqueous solution. It can be seen from this graph that, at pH 5-5.5 or below the presence of iron accelerated the formation of dithionic acid. The amount of dissolved iron sharply decreased as the pH value of the liquid roas. At pH 6 or over, the solubiluty of iron decreased below 1 ppm and substantially no production of dithionic acid was observed.

Where the pH value of the liquid is low, particularly where it falls to the neighbourhood of 4, a treatment which is effective in rendering the structural material of the system incapable of releasing iron into the liquid becomes important. Where the pH value is 6 or over, the amount of dissolved iron is so small it may be thought that special attention may not always be required to be directed to the selection of structural material of the system. However, in view of the possibility that the nature of the liquid will be locally varied or temporarily affected by a variation in the operational status, it is important to pay due attention to the selection of the structural material of the system, especially of the precipitator. The test was also conducted with Mn, Cu, V, Ni, Ca and Mg besides iron, from which it was ascertained that the effects of these metals on the by-production of dithionic acid were less than 1/10 of that by iron.

Example 6: With the same apparatus as used in Example 3, the raw exhaust gas fed at a flow velocity of 11 Nm/sec was treated with the absorbent sprayed into droplets of various particle diameters. The treatment was performed under various conditions such that most of the sprayed droplets had particles falling (a) in the range of from 1 to 30 u similarly to Example 3, (b) in the range of from 50 to 100 p (by use of an ultrasonic nozzle) and (c) in the range of from 70 to 300 p (bu use of an ordinary nozzle). In the test run (b), the exhaust gas at the outlet was found to have a SO2 content of 4.8 ppm and a Na2 S2 O6 /Na2SO4 molar ratio of 9/1000. In the test run (c), the SO2 content was 11.2 ppm and the Na2 S2 06 /Na2 S04 molar ratio was 24/1000. The SOB and soot contents in the test runs (b) were the same as those in the test run (a). In the test run (c), however, the soot content was slightly greater (though not determinable quantitatively) and the SOs content was OA ppm. These results suggest that the finely divided droplets of the absorbent contribute to the aggregation of SO3 mist, etc.

to some extent. In the test run (c), the amount of sprayed droplets which were entrained by the current of exhaust gas and drifted in the form of fog into the precipitator was smaller than in the test runs (a) and (b).

ExamDle 7: With the same apparatus as used in Example 3, the raw exhaust gas was fed at a flow velocity of 8 Nm/sec. and SO3 from fuming sulfuric acid was added to the exhaust gas so as to increase the SO3 concentration in the exhaust gas to 80 ppm. The exhaust gas at the outlet and the liquid finally disposed of were found to contain SOx and Na2 S2 06 at the same tolerable levels as those obtained in Example 3.

Example 8: At the entrance to the catalyst bed in the system used in Example 1, the exhaust gas (containing NH3 gas) was diverted and various catalysts indicated in Table 1 below were fed at a fixed feed rate of 300 /hour, one each into as many test catalytic reactors. The results were as shown in Table 1 below.

Table 1

Catalyst Reaction temperature SV (h-1) Percentage o eC) NOx- removal % a Obtained by adding 10% by weight of CuO to 140 7,000 96 what had been produced by baking rhodo chrosite at 400us and baking the resultant mixture at 4000C b Obtained by adding 10% by weight of ZnO to 140 7,000 96 what had been produced by baking rhodo chrosite at 400 C and baking the resultant mixture at 4000 C.

c Obtained by baking manganese carbonate at 130 5,000 97 350"C.

d Manganese dioxide obtained by electrolysis 150 5,000 92 technique.

e. Obtained by baking manganese dioxide ore 200 5,000 72 at 4000 C.

f Obtained by coating activated carbon with 110 5,000 93 platinum.

g Obtained by coating activated carbon with 120 2,500 84 ammonium bromide.

h Obtained by baking Co3 04 at 3000C. 180 3 000 88 All these test runs were continued over a period of 850 hours. It was only in the test runs of f and g that the percentage of NOx-removal showed a variation in excess of 2%. In the test runs of f and g, since the catalytic activity was impeded owing to decomposition of the formed ammonium nitrate, the temperature of the exhaust gas had to be temporarily raised to about 200 C for the treatment to restore the original catalytic activity. This mean that the catalysts in these test runs were not permanently poisoned.

WHAT WE CLAIM IS: 1. A process for removing sulphur oxides from a gaseous exhaust mixture containing at least sulphur oxides and gaseous oxygen, wherein said gaseous exhaust mixture is contacted, in a spray column, with droplets of an aqueous solution of an alkali metal salt and/or an alkali metal hydroxide to absorb the sulphur oxides and to oxidise absorbed sulphur diocide with the oxygen of the said gaseous exhaust mixture, the surfaces of the spray column contacted by said aqueous solution being of non-metallic material, the droplets having a particle size of 1-100 diameter, the solution having a pH of at least 8, and the solution being introduced into the spray column at a rate of 0.5 to 5 litres per 100 Nm of the gaseous exhaust mixture; and the thus treated exhaust mixture being introduced into an electrostatic precipitator in which finely divided droplets and dust are removed from the gaseous exhaust mixture.

2. A process according to Claim 1, wherein the flow velocity of the exhaust gas within the spray column is in the range of from 5 to 20 m/sec.

3. A process according to Claim 1 or 2, wherein the surface of the electrostatic precipitor to be contacted with the aqueous solution are formed of a substantially non-metallic material.

4. A process according to any one of Claims 1 to 3 wherein the non-metallic material of which the part of the spray column contacted by said aqueous solution are formed are a synthetic resin rubber or a fibre reinforced plastics material.

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (16)

1. **WARNING** start of CLMS field may overlap end of DESC **.

   the outlet was found to have a SO2 content of 4.8 ppm and a Na2 S2 O6 /Na2SO4 molar ratio of 9/1000. In the test run (c), the SO2 content was 11.2 ppm and the Na2 S2 06 /Na2 S04 molar ratio was 24/1000. The SOB and soot contents in the test runs (b) were the same as those in the test run (a). In the test run (c), however, the soot content was slightly greater (though not determinable quantitatively) and the SOs content was OA ppm. These results suggest that the finely divided droplets of the absorbent contribute to the aggregation of SO3 mist, etc.

   to some extent. In the test run (c), the amount of sprayed droplets which were entrained by the current of exhaust gas and drifted in the form of fog into the precipitator was smaller than in the test runs (a) and (b).

   ExamDle 7: With the same apparatus as used in Example 3, the raw exhaust gas was fed at a flow velocity of 8 Nm/sec. and SO3 from fuming sulfuric acid was added to the exhaust gas so as to increase the SO3 concentration in the exhaust gas to 80 ppm. The exhaust gas at the outlet and the liquid finally disposed of were found to contain SOx and Na2 S2 06 at the same tolerable levels as those obtained in Example 3.

   Example 8: At the entrance to the catalyst bed in the system used in Example 1, the exhaust gas (containing NH3 gas) was diverted and various catalysts indicated in Table 1 below were fed at a fixed feed rate of 300 /hour, one each into as many test catalytic reactors. The results were as shown in Table 1 below.

   Table 1

   Catalyst Reaction temperature SV (h-1) Percentage o eC) NOx- removal % a Obtained by adding 10% by weight of CuO to 140 7,000 96 what had been produced by baking rhodo chrosite at 400us and baking the resultant mixture at 4000C b Obtained by adding 10% by weight of ZnO to 140 7,000 96 what had been produced by baking rhodo chrosite at 400 C and baking the resultant mixture at 4000 C.

   c Obtained by baking manganese carbonate at 130 5,000 97 350"C.

   d Manganese dioxide obtained by electrolysis 150 5,000 92 technique.

   e. Obtained by baking manganese dioxide ore 200 5,000 72 at 4000 C.

   f Obtained by coating activated carbon with 110 5,000 93 platinum.

   g Obtained by coating activated carbon with 120 2,500 84 ammonium bromide.

   h Obtained by baking Co3 04 at 3000C. 180 3 000 88 All these test runs were continued over a period of 850 hours. It was only in the test runs of f and g that the percentage of NOx-removal showed a variation in excess of 2%. In the test runs of f and g, since the catalytic activity was impeded owing to decomposition of the formed ammonium nitrate, the temperature of the exhaust gas had to be temporarily raised to about 200 C for the treatment to restore the original catalytic activity. This mean that the catalysts in these test runs were not permanently poisoned.

   WHAT WE CLAIM IS: 1. A process for removing sulphur oxides from a gaseous exhaust mixture containing at least sulphur oxides and gaseous oxygen, wherein said gaseous exhaust mixture is contacted, in a spray column, with droplets of an aqueous solution of an alkali metal salt and/or an alkali metal hydroxide to absorb the sulphur oxides and to oxidise absorbed sulphur diocide with the oxygen of the said gaseous exhaust mixture, the surfaces of the spray column contacted by said aqueous solution being of non-metallic material, the droplets having a particle size of 1-100 diameter, the solution having a pH of at least 8, and the solution being introduced into the spray column at a rate of 0.5 to 5 litres per 100 Nm of the gaseous exhaust mixture; and the thus treated exhaust mixture being introduced into an electrostatic precipitator in which finely divided droplets and dust are removed from the gaseous exhaust mixture.
2. 2. A process according to Claim 1, wherein the flow velocity of the exhaust gas within the spray column is in the range of from 5 to 20 m/sec.
3. 3. A process according to Claim 1 or 2, wherein the surface of the electrostatic precipitor to be contacted with the aqueous solution are formed of a substantially non-metallic material.
4. 4. A process according to any one of Claims 1 to 3 wherein the non-metallic material of which the part of the spray column contacted by said aqueous solution are formed are a synthetic resin rubber or a fibre reinforced plastics material.
5. 5. A process according to any preceding

   claim wherein said aqueous solution is of sodium hydroxide or sodium carbonate.
6. 6. A process according to any preceding claim wherein the concentration of sulphur oxides in the raw exhaust gas is not more than 500 ppm.
7. 7. A process according to Claim 6, wherein the concentration of sulphur oxides in the raw exhaust gas is in the range of from 100 to 200 ppm.
8. 8. A process according to Claim 5 wherein the concentration of sulphur oxide in the raw exhaust gas is less than 100 ppm.
9. 9. A process according to Claim 6 wherein the concentration of said aqueous solution is in the range of from 2% to 10% by weight.
10. 10. A process according to Claim 8, wherein the concentration of said aquous solution is 5% by weight.
11. 11. A process according to any preceding claim wherein said aqueous solution has no more than 1 ppm of dissolved iron.
12. 12. A process according to any preceding claim wherein the finely divided droplets of said aqueous solution have a diameter in the range of from 1 to 30.
13. 13. A process according to Claim 11 wherein the finely divided droplets of said aqueous solution have a diameter in the range of from 20to30.
14. 14. A process according to any preceding claim wherein after said aqueous solution has adsorbed the sulphur oxides it has a pH above 8.
15. 15. A process according to any preceding claim wherein a mixture arrester is interposed between the spray column and the electrostatic precipitator.
16. 16. A process according to Claim 1 for the removal of sulphur oxides from an exhaust gas mixture containing at least sulphur oxides and oxygen, substantially as described in any one of the examples 1 to 4, 6 and 7.