PROCESS FOR THE PREPARATION OF CHLORINE DIOXIDE
The present invention relates to a method for continuous preparation of chlorine dioxide from alkali metal chlorate and sulphuric acid using sulphur dioxide as a re- ducing agent. The invention relates especially to enhanced efficiency of the method owing to formic acid.
Chlorine dioxide is used commercially in considerable amounts in the following fields, for instance: water purification, fat bleaching, waste treatment and especially cellulose bleaching. At cellulose mills, chlorine dioxide is the only possible chlorine compound in Elemental Chlorine-Free Bleaching (ECF), for instance. Thus it is desirable to provide optimal efficiency in chlorine dioxide preparation.
Due to the explosion risk, chlorine dioxide is usually prepared in the vicinity of the site of application. Industrial preparation of chlorine dioxide is based on the reduction of chlorate. It is known to use chlorates as the raw material, and chlorine diox- ide is formed in extremely acidic solutions. For economic reasons, either hydrochloric acid, methanol, hydrogen peroxide or sulphur dioxide is usually used as the reducing agent.
A known method for preparing chlorine dioxide is the Mathieson method, in which sulphur dioxide is used as the reducing agent. The method usually implements two serially disposed reactors, which operate under atmospheric pressure or near atmospheric pressure.
The total reaction equation is the following:
2 NaC103 + H2S04 + S02 > 2C102 + 2 NaHS04 (1)
However, this reaction equation does not describe intermediate phases and partial reactions occurring in chlorine dioxide preparation. It is known that a chloride ion acts as a transmitter agent in any chlorine dioxide process, even if it has not been added separately to the reaction system. The stoikiometry of the process has been described in the paper (Dennis Owen, Operation and Maintenance of Chlorine Dioxide Generators, Tappi 1989 Bleach Plant Operations, p. 157) by the following reaction equation:
2 HCIO3 + 2 HC1 > 2 C102 + Cl2 + 2H20 (2)
in which chloric acid HC103 has been produced from sodium chlorate under acid conditions. The chlorine produced reacts further with the sulphur dioxide reducing agent into hydrochloric acid:
Cl2 + S02 + 2H20 >2HC1 + H2S04 (3)
For this reason, it is important that the portion of the chlorine produced in reaction (2) that gets into a gaseous phase after it has left the reactor and is reacted into hydrogen chloride and then returned to the reactor, as well as the hydrochloric acid admixed with chlorine dioxide gas.
In an acid solution, chlorine forms hypochlorous acid HOCl as an intermediate phase.
If the amount of chloride is too low in the reaction solution, the reaction medium will not contain a sufficient amount of chloride ions for the reaction to be maintained, and then chloride ions have to be formed from chlorate by means of "over- reduction", resulting in reduced efficiency of chlorine dioxide formation. For this reason, sodium chloride is frequently added to the reactor at the start of the chlorine process. If, again, the amount of chloride feed accounts for more than 6% by weight of the supplied amount of chlorate, chlorine may remain in the final chlorine dioxide solution.
There will also be significant yield loss due to secondary reactions occurring in extremely acid solutions, which produce sodium chloride.
To resolve this problem, the Mathieson method maintains a high acid concentration, 450 to 500 g/1, but the yield is typically less than 87%. In the practice, chlorine . dioxide yield calculated per chlorate feed may be even less than 80%. Low efficiency entails high raw material cost and a large amount of waste acid per produced amount of chlorine dioxide.
Since the adoption of a vacuum process with higher efficiency would require significant investments, cellulose mills are still interested in increasing the output of two conventional serially disposed reactors operating under atmospheric pressure or near atmospheric pressure in a continuous process such as the Mathieson method, for instance, and in reducing the drawbacks of this, with subsequent higher effi-
ciency. In addition, higher production capacity at cellulose mills has become a problem by chlorine dioxide production forming a limiting factor.
The applicant's previous patent FI 108536 discloses a solution for increasing the output of the Mathieson process. Hydrogen peroxide was used as a reducing agent alongside sulphur dioxide in this solution.
It is also known to enhance efficiency by adding chloride ions such as e.g. sodium chloride to the process. However, this results in chlorine formation and increased amounts of waste acid, and hence it is not a very useful method.
In WO 9005698, the Mathieson method recycles reaction solution from the reactor to the reactor through the gas scrubber. The gas scrubber provides a large area, so that unreacted sulphur dioxide may react with chlorine during the formation of sulphuric acid and hydrochloric acid. This equipment variant has provided chlorine dioxide that is substantially freer from chlorine, while achieving also higher yield and output capacity.
In FI patent specification 89474 chlorine dioxide has been prepared by a vacuum process of SVP type, using methanol as the reducing agent. The efficiency of the production process has been enhanced by decreasing the amount of methanol used as the reducing agent, so that the formic acid concentration in the reactor produced as a by-product in the process has been raised by adding a surplus of formic acid. Enhanced efficiency relates to the consumption of methanol used as the reducing agent by manipulation in the reaction solution of the amount of formic acid produced as a by-product in the reduction reaction.
The purpose of the present invention is to enhance the efficiency of conventional processes of the Mathieson type, in which operations are carried out under atmos- pheric pressure or near atmospheric pressure.
In the method of the invention, formic acid or a salt of formic acid was added to the reaction solution, and then an increase in the efficiency of the process was surprisingly observed.
In accordance with the invention, a method has thus been achieved for continuous preparation of chlorine dioxide by reacting alkali metal chlorate, sulphuric acid and sulphur dioxide acting as a reducing agent in a reactor under atmospheric pressure or near atmospheric pressure, during continuous feed of inert gas to the reactor, and at a temperature in the range from 30 to 100°C, the method being characterised in
the addition of formic acid or its salt to the reaction solution, resulting in a over 87% efficiency of chloride dioxide.
In the method of the invention, the reactor, which is maintained under atmospheric pressure or near atmospheric pressure and at a temperature less than 100°C, is con- tinuously supplied with the raw materials needed in the reaction. Advantageously, alkali metal chlorate and sulphuric acid are fed through the reactor top and sulphur dioxide diluted with inert gas is fed through the reactor bottom. At the same time, a formic acid solution or a solution of a salt of formic acid is introduced through the reactor top, the solution reacting with the sulphuric acid in the reaction solution, forming formic acid.
The alkali metal chlorate of the invention may be potassium or sodium chlorate, preferably sodium chlorate. Chlorate is preferably used in the reaction solution in a concentration from 15 to 25 g/1. Typical sodium chlorate consumption is in the range from 1.65 to 1.90 kg/kg of chlorine dioxide.
The process of the invention can also be carried out as a two-stage process, in which the reaction solution is conducted from the first stage as overflow to the second stage. This is usually done in a smaller reactor, in which chlorate that was unreacted at the first stage reacts completely as a mixture of inert gas and sulphur dioxide is fed through the reactor bottom. In this situation, formic acid can be supplied either to the first stage or the second stage or to both of these.
In a two-stage process, the chlorate concentration at the first stage may be from 10 to 50 g/1, preferably 15 to 25 g/1. At the second stage, chlorate can be almost totally consumed, so that its concentration is 0.5 to 5 g/1, preferably 1 to 2 g/1. The residence time at the first stage is typically 10 to 20 h and 1 to 3 h at the second stage.
To ensure optimal operation of the process, the acid concentration should be sufficiently high, typically from 450 to 500 g/1. In the method of the invention, the sulphuric acid concentration of the reaction mixture may be in the range from 100 to 650 g/1, preferably from 400 to 500 g/1.
Air is usually used as inert gas in the method of the invention. For safety reasons, nitrogen or a mixture of nitrogen and air is preferably used, especially at the start of the process. In addition, e.g. carbon dioxide or process exhaust gases can be used. Inert gas can be used for mixing the reaction medium and also for diluting the chlorine dioxide produced.
The proportion of sulphur dioxide acting as the reducing agent and fed to the reactor along with inert gas may account for 5 to 15 % by volume, preferably 8 to 12% by volume of the feed. With too small an amount of sulphur dioxide, the amount of chloride ions produced in the reaction medium will not be sufficient for the reaction to be completed, and chloride will have to be formed from chlorate by means of over-reduction, which deteriorates efficiency. Sulphur dioxide used in too large amounts may result in a situation, in which the tower scrubber contains a surplus of sulphur dioxide relative to chlorine to be reduced. At the subsequent absorption stage, this surplus of sulphur dioxide will react with chlorine dioxide, resulting in lower yield.
Formic acid can be added to the reactor as 50 to 100% aqueous solution or by dissolving a salt of formic acid, preferably alkali metal salt, such as sodium or potassium formate, to a water concentration of 5 to 75% by weight, and by adding the solution thus obtained to the reactor. The feed rate of formic acid to the reactor is de- termined by the desired formic acid concentration in the reaction mixture. The greater the formic acid addition to the reaction mixture, the higher the chlorine dioxide efficiency. However, in the practice, the amount of formic acid used is appropriately optimised with respect to the efficiency goal and the reactant cost. In practical operation, this will restrain the desire to raise efficiency to nearly 100%, even though this would be technically feasible. The determined formic acid concentration in the reaction mixture is preferably over 1 g/1, and most advantageously 5 to 35 g/1. The efficiency ratios are then above 87% and 88 to 96%, respectively, when calculated on chlorine dioxide formed per reacted chlorate.
The mechanism that enhances the efficiency of formic acid is not exactly known. Formic acid may act as a reducing agent, and in that case, it is consumed in the reaction. In the method of the invention, about 30% of formic acid is decomposed, implying that it participates in a reaction.
The method of the invention can be implemented in any type of known continuous reaction devices, in which atmospheric or near atmospheric pressure is applied. The reactors used in the Mathieson method are particularly suitable. There may also be two or more reactors if increased production output is required.
If necessary, chloride ions, e.g. sodium chloride, can be added to the reaction solution, especially at the activation stage, for instance. However, the amount of chloride ions should be small enough for excess chlorine not to remain in the final chlo- rine dioxide solution.
The method of the invention is applicable also as a reaction step in other production methods. It may replace the first reduction step described in FI 108536, for instance, followed by a second reduction step using hydrogen peroxide as the reducing agent.
The method of the invention achieves a notable improvement of process efficiency, allowing higher production output and reduced production cost. At the same time, the proportion of secondary reactions decreases. The amount of waste acid decreases in the same ratio as efficiency increases. This is an advantage at cellulose mills, which have problems caused by the utilisation of waste acid.
In one embodiment of the invention, acid gases are advantageously returned to the reactor for further improvement of efficiency. Due to practical problems, this has not been done in the following examples. The examples exemplify advantages achieved by the invention, without being restricted to these.
Comparative example 1
10 continuous tests were conducted in a reactor operating under atmospheric pres- sure and having a diameter of 80 mm and a height of 4500 mm. The liquid surface height in the reactor was 3500 mm. The reactor was supplied with a 47 % by weight chlorate solution at a rate of 1.26 kg/h and with a 93% sulphuric acid solution at a rate of 0.4 kg/h. The reactor was also supplied at a rate of 243 g/h with a 9 % by volume of sulphur dioxide admixed in nitrogen. As the reactor had reached equilib- rium, its chlorate concentration was 30 to 40 g/1 and acid concentration 450 to 480 g/1. The reactor temperature was maintained at 45 °C. The chlorine dioxide amount formed was determined and, as an average of ten tests, 86% of chlorine dioxide was recovered from reacted chlorate, this percentage being the efficiency of the process.
Example 2
In the test of the invention, 85% by weight formic acid was added to the reaction solution of example 1 through the reactor bottom, so that the formic acid concentration of the reaction mixture was 12 g/1. The results of four different tests were 85.2%, 88.3%, 90.4% and 94.4%, respectively. The method of the invention shows relatively large deviation due to the great number of reaction steps typical for the Mathieson process, which allow for several secondary reactions. The mean efficiency obtained for the process was 89%.
Example 3
In the test of the invention, 85% by weight of formic acid was added to the reaction solution of example 1, so that the reaction mixture had a formic acid concentration of 36 g/1. The efficiency obtained for the process was 98%.
Comparative example 4
In order to test the impact of the carboxylic acid quality used to enhance efficiency, formic acid was replaced with acetic acid. For the sake of comparison, 85% by weight of acetic acid was added at a rate of 60 g/1 to the process of example 1. The efficiency obtained for the process was 87%. This test indicates that it is essential to add precisely formic acid as carboxylic acid, otherwise there will be no enhancement of the efficiency.
Example 5
Formic acid was added in accordance with the invention to a reactor in the same way as in example 2. The amount of formic acid addition varied in the range from 5 to 35 g/1, and the test series comprised several repetitions. The mean efficiency of the process was over 87% in all of the tests, and the dependence between efficiency and formic acid addition of figure 1 was determined on the basis of the tests.