The Effect of Structure Modification of Sodium Compounds on the SO 2 and HCl Removal Efficiency from Fumes in the Conditions of Circulating Fluidised Bed

A. Pajdak,a,* B. Walawska,b and A. Szymanekc aThe Strata Mechanics Research Institute of the Polish Academy of Sciences, Reymonta 27, 30-059 Krakow, Poland bInorganic Chemistry Division “IChN” in Gliwice, New Chemical Syntheses Institute, Sowińskiego 11, 44-101 Gliwice, Poland cInstitute of Thermal Machinery, Faculty of Mechanical Engineering and Computer Science, Czestochowa University of Technology, Dąbrowskiego 69, 42-200 Czestochowa, Poland


Introduction
Sodium compounds including sodium bicarbonate (NaHCO 3 ) are widely used in the purification of gases created by incineration of municipal and medical waste, biomass as well as alternative fuels in the cement industry.They are more often used in the energy industry and in small solid fuel combustion units including circulating fluidised bed boilers.
A number of factors determines the efficiency of using sodium compounds to purify exhaust fumes.Apart from reactivity, the size and the porosity of grains are crucial.To reduce their size, they are mechanically modified by grinding.Smaller particles, thanks to their increased active surface and a better contact with the pollutant, react faster and more efficiently than larger particles 1 .The increase in porosity is achieved through thermal mod-ification.At high temperatures, NaHCO 3 decomposes into Na 2 CO 3.Under industrial conditions, such decomposition takes place after placing a dry sorbent in a stream of hot exhaust gases.Sodium bicarbonate is a sorbent, which is especially suitable to be used in the dry method of flue gas purification in small power plants [2][3][4][5][6] .Removing SO 2 from gases using sodium bicarbonate is possible in the range of temperatures between 413 K and 573 K, and goes in two ways depending on the temperature [7][8][9][10] .In the temperature range 343-393 K it goes according to the direct model.The non-porous sodium bicarbonate of a low specific surface area reacts with the sulphur compounds only on the surface according to the reaction: At temperatures above 393 K, the indirect model occurs and the direct model takes place only to a small extent.In the indirect model, the porous sodium carbonate with a developed active surface (2) is created as a result of the decomposition of sodium bicarbonate.In contact with sulphur dioxide, sodium(IV) sulpfate is created according to the reaction (3) and in contact with oxygen, sodium(VI) sulphate is created (4).2NaHCO 3 → Na 2 CO 3 + CO 2 (g) + H 2 O(g) (2)   Na 2 CO 3 + SO 2 → Na 2 SO 3 + CO 2 (3) Sodium bicarbonate also shows high reactivity against hydrogen chloride 4,5,[11][12][13] .Depending on the temperature, gas hydrogen chloride reacts with sodium bicarbonate in a direct (5) or indirect way after decomposition into sodium carbonate (2) according to the reaction (6): NaHCO 3 +HCl → NaCl + H 2 O + CO 2 (5)   2HCl + Na 2 CO 3 → 2NaCl + CO 2 + H 2 O (6) In the energy industry, sodium compounds are used to remove acid gas pollutants, such as sulphur dioxide (SO 2 ) and hydrogen chloride (HCl) from the flue gases 14 .Along with carbon dioxide, sulphur dioxide is one of the most dangerous pollutants emitted into the atmosphere.The problem of CO 2 pollution and its removal is an extensively described issue [15][16][17] .
Over 50 % of waste SO 2 comes from the power industry.The current concentration limits of SO 2 are regulated by EU directives including LCP directive 18 and IED directive 19 , which allow the emission of SO 2 from the combustion of solids, such as coal, of up to 250 mg m -3 n in plants of the nominal power of 100-300 MW and up to 400 mg m -3 n in plants of 50-100 MW.The latest MCP directive 20 establishes the requirements concerning SO 2 emission for medium size combustion plants with the nominal heat output of up to 50 MW.In plants of 5-50 MW, the allowed concentration is up to 400 mg m -3 n and in plants of 1-5 MW it is 1100 mg m -3 n .According to the draft directive, plants of the capacity above 5 MW will be obliged to adjust their emission levels by January 1, 2025, and plants of up to 5 MW by January 1, 2030.
Removing hydrogen chloride from gases takes place mainly in installations of thermal conversion of solid waste where concentrations above 1000 mgHCl m -3 are noted.However, the permitted emission to the air considering average daily values is 10 mgHCl m -3 21 .In installations of fuel combustion, purification of waste gases from HCl is not applied.It is connected with its low concentration in the fuels as well as the lack of proper legal acts regulating emission levels in those units.In its final preparation stage is Draft 1 LCP BREF concerning, among other, the issue of the levels of HCl and HF emitted during combustion.Its conclusions are to be published in 2016.It specifies the accepted hydrogen chloride emission for plants of the power heat of 100 MW th at the level of 1-6 mg m -3 n for new facilities and 2-10 mg m -3 n for the existing plants, and for the plants of over 100 MW th 1-3 mg m -3 n and 1-5 mg m -3 n respectively 22 .Taking into account the integrated protection of air and constant changes of EU acts regulating permissible emission levels, it should also be considered in such installations.
Currently in Europe, the largest application of sodium sorbents to remove SO 2 and HCl is connected with the process developed by SolvAir Solutions 23 .It involves reaction of ground bicarbonate particles with gas pollutants in a duct at temperatures around 413 K on a bag filter in which dust and reaction products are separated from the purified gas.This process provides high efficiency of exhaust purification with a low usage of the reagent as well as a low investment cost 24 .

Methods and apparatus
The chemical composition of the materials used in the tests was determined by classical methods as well as LECO analysers of two types.To specify the content of carbon and hydrogen, the method of selective absorption in infrared was applied using TruSpec CHN (LECO, Poland).The concentration of nitrogen was determined using a thermal conductivity detector after reducing nitric oxides and mixing with helium.The concentration of sulphur was determined by the combustion method with IR detection using analyser SC-144 (LECO, Poland).The concentration of oxygen was determined from the difference.
Sodium bicarbonate was mechanically modified by grinding in an impact mill 160 UPZ (Hosokawa-Alpine, Germany) with the rotor speed of 11400 rpm and the efficiency of 60 kg h -1 .Before and after grinding, the material was subjected to a grain size analysis using a laser diffraction particle size analyser LS 13320 (Malwern, United Kingdom).
The grain structure was analysed using a scanning electron microscope XL 30 Esem (Philips, The Netherlands) with analyser EDS (EDAX, Japan), which provides real space imaging of the sample topography.The samples were dusted with gold, and using a BSE detector, images at magnification of 1000x to 20000x were recorded.The tests were carried out in the high vacuum mode at the voltage of 15 kV.Structural parameters were measured applying porosimetric methods.The surface area and the total pore volume were defined.The measurements of microstructure of sodium bicarbonate were carried out using a Gemini VII Series Analyzer (Micromeritics Instrument Corporation, USA).The low-pressure nitrogen adsorption process (LPNA) was carried out at the temperature of 77 K.The multilayer adsorption theory Brunauer-Emmett-Teller (BET) 25 was used in the relative pressure range p/p 0 0.05-0.30.The analyses of macrostructure of NaHCO 3 used mercury porosimetry (MIP) 26 .The analysis were made with AutoPore IV 9500 (Micromeritics Instrument Corporation, USA).They involved the intrusion of mercury into the porous structures under the influence of strictly controlled pressure in the pressure range from 10 -5 MPa to 414 MPa.
Thermal decomposition of NaHCO 3 was tested by thermogravimetric methods (TG) and differential scanning calorimetry (DSC) with a thermogravimetric analyser TGA/SDTA 851e (Mettler Toledo, Poland) and a differential scanning calorimeter DSC 822e (Mettler Toledo, Poland).The analysis was carried out in a jet of air of 50 cm 3 min -1 , and the heating rate of 10 K min -1 .
To determine the correlation of the characteristics of specific surface area and the activation temperature, correlation calculations were made.Pearson correlation coefficient was used for the analyses (7): x y cov x y r x y = ⋅ s s where: where: r(x,y) -Pearson correlation coefficient between variables x and y, cov(x,y) -covariance between variables x and y, σ -standard deviation, E -expected value.
The correlation coefficient takes values between -1 to 1 and is determined according to the formula ( 9) and (10).The result close to 0 indicates a lack of correlation of the tested characteristics, and the result close to boundary values (1, -1) indicates their strong correlation.
where: r xy -Pearson correlation coefficient, x i -the results of particular measurements of the surface area, x -arithmetic mean from the measurements of the surface area, y i -the results of particular measurements of the activation temperature, y -arithmetic mean of the activation temperatures.
Next, the coefficient of determination r 2 was calculated from the formula (11) and t value conclusive as for the significance of the correlation ( 12): where: r 2 -coefficient of determination, t -value conclusive as for the significance of the correlation, n -the number of measurements.t value was compared with critical value t kr calculated from Student's t-distribution of the degrees of freedom v = n -2 on the basis of which the significance level a = 0.05 was assumed.The significance of the characteristics was determined by the fulfilment of the condition (13): To determine the efficiency of sodium compounds to remove SO 2 and HCl simultaneously from flue gases, the tests were carried out on a 0.1 MW CFB test rig.This original test rig is placed in the Institute of Advanced Energy Technologies in Czestochowa University of Technology in Poland.The basic element of the rig was the combustion chamber and a cyclone where the grain separation of the fluidized layer material took place.The scheme of the test rig is shown in Fig. 1.
The air and fuel were fed continuously.The fuel was supplied at the rate of 4.5 kg h -1 .The material of the fluidized bed in the combustion chamber was quartz sand of the grain size from 2.5 to 6.3 µm and the average grain diameter of 4.0 µm.The sodium sorbents were fed continuously into the jet of fumes in a few molar ratios (2Na/S) at a temperature of around 573 K in the second draft at Sorbent Dosing Port.After each measurement, the samples of the fly ash and bottom ash were taken at FA and BA, respectively.
To measure the concentration of gases, a GAS-MET DX-4000 analyser was used, which was installed at Flue Gas Sampling Port.Thanks to the application of the 'hot' method (maintaining the measurement duct at 453 K), it selectively measured the level of pollutants i.e., SO 2 , HCl and O 2 , CO, CO 2 as well as the moisture.The collected gas sample spectrum was analysed in CALCMET software using the FT-IR (Fourier Transform Infrared) spectroscopy method.
Based on the decrease in the concentration of the pollutants in the flue gases, the average efficien-cy in removing SO 2 and HCl (14) as well as the maximum temporary efficiency in removing SO 2 and HCl (15) were determined according to the following formulas: where: ϕ(t) -average efficiency in purifying fumes, %, C 0 -the average concentration of the pollutants in the fumes before purification (background), mg m -3 n , C s -the average concentration of the pollutants during the process, mg m -3 n . (15) where: ϕs(t) -the maximum temporary efficiency in purifying fumes, %, C 0 -the average concentration of the pollutants in the fumes before purification (background), mg m -3 n , C min -the minimum concentration of the pollutants during the process, mg m -3 n .Equations ( 14) and (15) were formulated based on the general equation for the efficiency of the process and the Clapeyron's equation.To compare the concentration levels, some simplifications were made, taking into account the conditions in which the tests were carried out, i.e. continuous flow of fumes measured in standard conditions, constant chemical composition of gases (constant concentration of pollutants (C 0 ), constant pressure (atmospheric), constant temperature of the gases).Due to the low concentration of pollutants, it was also assumed that the content of the remaining ingredients of gas does not change in the purification process.The concentration of SO 2 and HCl was measured continuously as the volume concentration, which was converted to mass concentration (mg m -3 n ), and 6 % content of oxygen.

Materials used for testing
Sodium bicarbonate (NaHCO 3 ) known as baking soda, high-purity product of standard class, comprising 99.8 % sodium bicarbonate and 0.2 % sodium carbonate (Na 2 CO 3 ) (calculated on a dry where: BA -bottom ash, CM -circulating material, FA -fly ash, HE -heat exchanger, MFC -mass flow controller, PAH -primary air heater, SAH -secondary air heater The air and fuel were fed continuously.The fuel was supplied at the rate of 4.5 kg h -1 .The material of the fluidized bed in the combustion chamber was quartz sand of the grain size from 2.5 to 6.3 µm and the average grain diameter of 4.0 µm.The sodium sorbents were fed continuously into the jet of fumes in a few molar ratios (2Na/S) at a temperature of around 573 K in the second draft at Sorbent Dosing Port.After each measurement, the samples of the fly ash and bottom ash were taken at FA and BA, respectively.
weight basis) was used as the sorbent.The material was obtained through saturation of the calcined soda produced in the Solvay process with gas CO 2 28  .
Coal from Janina coal mine in Libiąż was used as the fuel.Its characteristics are presented in Table 1.

Experimental
Sodium bicarbonate is a fine-grained material classified in accordance with the IUPAC guidelines as macroporous material 29 .It occurs in the both single grain form and as agglomerate since it has the tendency to create conglomerates.The measured mean diameter of the analysed NaHCO 3 was 45.9 µm (d 3,2 ).
As the grain size of sodium bicarbonate decreases, its reactivity increases, which is why it was ground in an impact mill.Before that it was mixed with magnesium stearate (magnesium salt of stearic acid C 36 H 70 MgO 4 ) -concentration 0.25 %.Adding the magnesium stearate prevented caking without affecting the physicochemical properties of NaHCO 3 .After grinding, the obtained grains were of average diameter over ten times smaller and a larger surface area.Table 2 shows structural parameters of NaHCO 3 before and after grinding.
The SEM images show the grain surface before grinding (Fig. 2) and after grinding (Fig. 3).The SO grains had smooth surface and compact, crystalline structure.The SOm grains were of various sizes and irregular shapes.Furrows and cracks can be seen on their surfaces.
Grinding sodium bicarbonate, apart from developing its surface area, decreased the temperature of its decomposition.The thermogravimetric analysis showed that the maximum decomposition of the sample SO took place at 431 K (peak), whereas SOm at the temperature 414 K. Fig. 4 shows the thermogravimetric analysis of the samples.properties of NaHCO 3 .After grinding, the obtained grains were of average diameter over ten times smaller and a larger surface area.Table 2 shows structural parameters of NaHCO 3 before and after grinding.The SEM images show the grain surface before grinding (Fig. 2) and after grinding (Fig. 3).The SO grains had smooth surface and compact, crystalline structure.The SOm grains were of various sizes and irregular shapes.Furrows and cracks can be seen on their surfaces.Before and after grinding (SO, SOm), the sodium bicarbonate was subjected to thermal decomposition by heating in the temperature range of 373-523 K for 0.5 h.After heating, LPNA and MIP methods were used to determine the following parameters: the surface area and the total pore volume.Their values changed together with the change in the activation temperature.Above the temperature of 293 K, the values of the parameters increased and in the temperature range of 423-523 K they were maximum.The surface area of the microstructure of SO sample was up to 3.2 m 2 g -1 (LPNA) at 473 K and SOm-6.0 m 2 g -1 (LPNA) at 523 K.
The maximum surface area of the sample microstructure determined by MIP method was 4.3 m 2 g -1 (473 K) and the maximum pore volume was 0.77 cm 3 g -1 (473 K).Above 523 K, the values of these parameters decreased, and at 673 K they reached the level similar to the level before the thermal modification.It was the effect of softening and closing of the pores developed in the temperature range of 293-523 K.The changes of the structural parameters are presented in Fig. 5.
The SEM analyses showed that heating the SO sample at 373 K led to the development of micrometric cracks parallel to the crystallographic directions (Fig. 6a).Small pores appeared on the surface, which resulted from partial decomposition of NaHCO 3 (Fig. 6b).During activation at 523 K, the SO sample decomposed totally and the pores developed more intensively than after activation at 373 K.The pores are of a more elongated shape and of a considerably bigger size, which is shown in Fig. 7. Grinding sodium bicarbonate, apart from developing its surface area, decreased the temperature of its decomposition.The thermogravimetric analysis showed that the maximum decomposition of the sample SO took place at 431 K (peak), whereas SOm at the temperature 414 K. Fig. 4 shows the thermogravimetric analysis of the samples.Grinding sodium bicarbonate, apart from developing its surface area, decreased the temperature of its decomposition.The thermogravimetric analysis showed that the maximum decomposition of the sample SO took place at 431 K (peak), whereas SOm at the temperature 414 K. Fig. 4 shows the thermogravimetric analysis of the samples.

F i g . 5 -Changes in the structural parameters of SO and SOm modified mechanically and thermally
The SEM analyses showed that heating the SO sample at 373 K led to the development of micrometric cracks parallel to the crystallographic directions (Fig. 6a).Small pores appeared on the surface, which resulted from partial decomposition of NaHCO 3 (Fig. 6b).During activation at 523 K, the SO sample decomposed totally and the pores developed more intensively than after activation at 373 K.The pores are of a more elongated shape and of a considerably bigger size, which is shown in Fig. 7.The SEM analyses showed that heating the SO sample at 373 K led to the development of micrometric cracks parallel to the crystallographic directions (Fig. 6a).Small pores appeared on the surface, which resulted from partial decomposition of NaHCO 3 (Fig. 6b).During activation at 523 K, the SO sample decomposed totally and the pores developed more intensively than after activation at 373 K.The pores are of a more elongated shape and of a considerably bigger size, which is shown in Fig. 7. Modification of ground NaHCO 3 (SOm) at 373 K resulted in developing numerous small pores which only partially covered the surface (Fig. 8).As the thermogravimetric analysis and the other tests showed 30 , the decomposition of the ground sodium bicarbonate was not total, hence, the porous structure did not develop on the whole surface of the grains.Also, numerous lamellar and columnar crystals appeared of the size of several micrometers.After activation at 523 K, the crystals disappeared and the pores became more prominent, creating a well-developed, thick, porous layer (Fig. 9).
To determine the correlations of the characteristics of the surface area and the activation temperature, correlation coefficients were calculated, presented in Table 3.The calculations included SO and SOm thermally activated in the temperature range of 373-673 K.In such a temperature range, NaHCO 3 decomposed into Na 2 CO 3 .In both cases, the significance of the correlation was met (r > 0), which showed a strong correlation (0.7 < r xy < 0.9).
The studies on circulating fluidized bed combustion rig verified the effect of grinding and heating of sodium reagents on their structural parame-124 Also, numerous lamellar and columnar crystals appeared of the size of several micrometers.
After activation at 523 K, the crystals disappeared and the pores became more prominent, creating a well-developed, thick, porous layer (Fig. 9).To determine the correlations of the characteristics of the surface area and the activation temperature, correlation coefficients were calculated, presented in Table 3.The calculations included SO and SOm thermally After activation at 523 K, the crystals disappeared and the pores became more prominent, creating a well-developed, thick, porous layer (Fig. 9).To determine the correlations of the characteristics of the surface area and the activation temperature, correlation coefficients were calculated, presented in Table 3.The calculations included SO and SOm thermally F i g .9 -Surface of the SOm grain after thermal activation at 523 K, magnification a) 5000x, b) 10000x Ta b l e 3 -Correlation coefficients of the specific surface area and the activation temperatures in the sodium compounds SO  During combustion of coal only, the average emission of SO 2 at the outlet of the combustion chamber was 2829 mg m -3 and 24 mg m -3 of hydrogen chloride.After the samples were placed in the installation, a sudden drop in the pollutants followed.The SO and SOm sorbents were fed into the installation in three molar concentrations: 2Na/S: 0.5; 1.0 i 2.1.The use of SO sorbent resulted in the drop in the mean SO 2 concentration to the level of 2690-1830 mg m -3 .
The use of SOm sorbent decreased the mean SO 2 concentration to the level of 1660-490 mg m -3 .The achieved reduction levels of SO 2 are presented in Figs. 10 and 11.After the application of SO sorbent, the concentration of hydrogen chloride de-creased from 24 mg m -3 to 6-7 mg m -3 and after the application of SOm sorbent it decreased to the level of 10-11 mg m -3 .Due to the significant excess of the sorbent in relation to the concentration of HCl in the gases, the change of the molar ration 2Na/S and not Na/Cl was taken into account.The effects of the application of SO and SOm in various molar ratios 2Na/S are presented in Figs. 12 and 13.
Each time after the introduction of sodium compounds into the flue duct, samples of fly ash (FA) were taken.In all the tested ash samples, chemical analyses confirmed presence of the products of the reaction in the form of sodium(IV, IV), sulphates, and sodium chlorides.The level of sodium(IV) sulphates were in the range of 0.03-0.2%.The level of sodium(VI) sulphates in FA rose along with the increase in the molar ratio of 2Na/S and the decrease in the grain size from 4.6 % (for SO 2Na/S = 0.5) to 21.3 % (for SOm 2Na/S = 2.1).In all the ash samples, presence of the product of the reaction of NaHCO 3 with hydrogen chloride in the form of sodium chloride was determined.
The following SEM images show the grains of sodium carbonate and embedded in them are smaller particles of sodium(VI) sulphate, calcium, crumbs of aluminosilicate enamel and efflorescence of sodium chloride forming several-micrometer-sized crystals.
Fig. 14a shows small particles of sulphate and sodium chloride of a size below a micrometer forming on a sodium carbonate grain.Some of the sodium sulphate grains combined (Fig. 14b) creating complex structures embedded in sodium carbonate grains (Fig. 14c).

Discussion
When removing SO 2 from flue gases, it was clearly observed that the grain size and the amount of the sorbent (molar ratio 2Na/S) affected the sorp-tion effect.The small-grained SOm sorbent removed SO 2 better than SO sorbent.As the molar ratio increased, so did the efficiency of purification.The biggest drop in the SO 2 concentration was achieved at the molar ratio 2Na/S = 2.1 and the smallest at 2Na/S = 0.5.
When removing HCl, no clear influence of the grain size and the amount of the sorbent on the purification efficiency was observed.The application of SO sorbent with larger grains resulted in a greater drop in the HCl concentration than the application of SOm.Increasing the amount of the reagent only slightly improved the efficiency of removing HCl in SO sorbent, whereas in the case of SOm sorbent, it had no effect.Table 4 presents the achieved average and the minimum momentary concentrations of pollutants after placing the sorbents in the gas pipe.
The SO sorbent showed the smallest average SO 2 removal efficiency (equal to 17 %) at the smallest molar ratio (2Na/S = 0.5).At the 2Na/S = 2.1 ratio the efficiency was 35 %.The sorbent after forming on a sodium carbonate grain.Some of the sodium sulphate grains combined (Fig. 14b) creating complex structures embedded in sodium carbonate grains (Fig. 14c).grinding -SOm had the average sorption efficiency twice as great as that of SO at the molar ratio of 2Na/S = 2.1.
Increasing the amount of the reagent improved the purification efficiency up to as much as 89 % for 2Na/S = 2.1.The maximum momentary purification efficiency by SO sorbent was 25-42 %, and by SOm from 55 % to 97 %.The values of the average and the maximum SO 2 removal efficiency is presented in Fig. 15.
The average HCl removal efficiency was above 50 %.When using the SO reagents, the average efficiency was from 66 % to 78 %, and when using SOm it was in the range of 54-57 %.The maximum efficiency was at the level of 79-95 %.The values of the average and the maximum HCl removal efficiencies are presented in Fig. 16.

Summary
The laboratory studies and the studies on the circulating fluidized bed combustion rig CFB showed that properly prepared sodium sorbents work very well in the dry method of purification of fumes from the combustion of coal.The average HCl removal efficiency was above 50 %.When using the SO reagents, the average efficiency was from 66 % to 78 %, and when using SOm it was in the range of 54-57 %.The maximum efficiency was at the level of 79-95 %.The values of the average and the maximum HCl removal efficiencies are presented in Fig. 16.Fig. 15 Average and maximum sorption efficiency of HCl after the application of sodium compounds in various molar ratios

Summary
The laboratory studies and the studies on the circulating fluidized bed combustion rig CFB showed that properly prepared sodium sorbents work very well in the dry method of purification of fumes from the combustion of coal.
Grinding and heating resulted in the change of the surface topography of the grains of the sodium sorbents.The unground grains (SO) had smooth surface without pores and after grinding they changed into forms that had furrows and cracks.After heating in the temperature range of 373-673 K, the grains intensively developed their surface area.Their pore structure was thick and extensive, and it covered almost the whole surface of the grains.
After grinding, the microstructure (LPNA) of the surface area did not change significantlyfrom 0.09 m 2 g -1 do 0.73 The results of the analyses prove that properly prepared sodium compounds have the ability to almost completely simultaneously remove SO 2 and HCl from purified flue gases.They can therefore be successfully used in conventional power industry.

Fig. 4
Fig. 4 DSC analysis of sodium compounds SO and SOm

F i g . 6 -
Surface of the SO grain after thermal activation at 373 K, BSE, magnification: a) 5000x, b) 10000x

F i g . 8 -
Surface of the SOm grain after thermal activation at 373 K, BSE, magnification a) 5000x, b) 10000x 124

F i g . 1 0
-Boxplot chart for the distribution of SO 2 concentration after using SO in different molar ratios F i g .11 -Boxplot chart for the distribution of SO 2 concentration after using SOm in different molar ratios F i g . 1 2 -Boxplot chart for the distribution of HCl concentration after using SO in different molar ratios F i g . 1 3 -Boxplot chart for the distribution of HCl concentration after using SOm in different molar ratios

F i g . 1 4 -
Surface of the sodium compound, magnification: 5000x: a) Na 2 CO 3 grain with embedded small particles of Na 2 SO 4 and NaCl, b) sodium sulphate grain, c) expanded particles of Na 2 SO 4 and NaCl embedded in Na 2 CO 3

Fig. 14 3 DiscussionTa b l e 4 -
Fig. 14 Surface of the sodium compound, magnification: 5000x: a) Na 2 CO 3 grain with embedded small particles of Na 2 SO 4 and NaCl, b) sodium sulphate grain, c) expanded particles of Na 2 SO 4 and NaCl embedded in Na 2 CO 3 130

Fig. 15
Fig.15 Average and maximum sorption efficiency of SO 2 after the application of sodium compounds in various molar ratios

F
Fig. 16 Correlation of the average sorption efficiency of SO 2 and HCl after the introduction of sorbents Ta b l e 2 -Structural characteristics of sodium bicarbonate

Table 2
Structural characteristics of sodium bicarbonate Their reactivity was determined in relation to sulphur dioxide and hydrogen chloride contained in the flue gases from combustion.