Influence of Ammonia Concentration on Solvay Soda Process Parameters and Associated Environmental and Energetic Effects

Abstract

Modifying the absorption process in soda production by the Solvay method requires performing many calculations and determining a new equilibrium process. An increase in ammonia concentration in the reaction solution causes kinetic changes in equilibrium. Changes to the Solvay soda production technology were determined using chemical and instrumental analysis methods. A modification of the process in the form of SAB was introduced. Information allowing the design of an additional absorber and its location in the network of technological devices was presented in the form of parameters using typical chemical engineering assumptions. Spectroscopic and electrochemical techniques were used for this purpose. The increase in total alkalinity due to the addition of ammonia to 135 mmol·20 cm⁻³ resulted in cooling savings of about 152.4 MJ·Mg⁻¹ of soda. The ammonia desorption rate and process energy parameters were determined for the new system. The temperature requirements for the carbonation column were defined, and in particular, a technique was developed to minimize the cooling of the lower part of the reactor, which reduces the consumption of process energy. Emissions of CO₂ were reduced from 11.70 to 7.85% and NH₃ from 5.52 to 4.89% in exhaust gases from the carbonation column.

1. Introduction

The long-term development of the soda industry worldwide has led to a number of modifications to the Solvay process. The main one is the dual process, which allows manufacturing units to produce ammonium chloride (which can be used as a fertilizer) in almost equal amounts as the product. Several plants in the world operate in this mode; most of them are located in China. The Akzo dry lime process, which uses lime instead of lime milk to recover ammonia, is currently considered the most advanced technology. There are technologies in which soda is produced as a by-product [1]. Energy, water consumption, and environmental pollution are a difficult problem in the world. Due to the rapid expansion of technology, the demand for water and energy is increasing, which also enhances the dependence on the availability of these sources [2,3]. The Solvay soda production process is very energy-intensive [4] and is characterized by the large production of solid, liquid, and gaseous waste [5,6,7]. In many countries worldwide, especially in developing countries, severe water shortages have been caused by population growth, industrialization, global warming, and climate change. Water is the most important resource on Earth, and is essential to human life. It is estimated that approximately 1.8 billion people will be without access to clean drinking water by 2025. The proposed solution in the Solvay soda industry allows for large savings in the consumption of cooling water and energy [8]. There are several energy-consuming processes in the Solvay soda technology: distillation, carbonation (cooling), calcination, and calcination of limestone.

The main issue is the efficiency and energy of the carbonation process of ammoniated brine, especially based on seasonality. In summer, cooling the column is a big challenge. Considering the increasingly higher temperatures and availability of cooling water, this is one of the main economic factors in the operation of a carbonation column. To increase the efficiency of the process when there is less cooling of the column, studies on increasing the concentration of ammonia in the carbonation process were carried out [9].

One of the diffusion operations in the soda production process using the Solvay technique is the absorption of ammonia. Permeation, or mass exchange, is the basis of many unit operations, including distillation, rectification, drying, hydration, absorption, adsorption, crystallization, dissolution, etc. [10]. There is a gap in the research in terms of determining the influence of the ammonia concentration on the efficiency of the soda process under the real conditions of the Solvay process.

Absorption is the uptake of a component or components of a gaseous mixture by a liquid, which is the absorbent. The gas, due to diffusion, penetrates into the liquid through the interfacial surface to form a solution. A gaseous mixture can contain components that are soluble and components that are practically insoluble in the liquid. The former are called active components, the latter neutral or inert [11,12]. This is one of the main processes of Solvay soda technology.

The statics of the absorption process, i.e., the equilibrium between the liquid and the gas phase, depend on the composition of the phases, temperature and pressure, since the two-phase system under consideration has three degrees of freedom according to Gibbs’s phase rule [10]. Therefore, at a given temperature and pressure, a fixed composition of one phase corresponds to a well-defined composition of the other phase. Absorption will occur when the content of the component in the gas phase is greater than the corresponding equilibrium state.

Empirical formulas are known in the literature that allow for the approximate determination of the selected parameters of the Solvay process, but their technological usefulness has not been confirmed. Previous research has shown that ensuring the maximum value of the degree of transformation of the sodium ion W_Na requires the particularly careful operation of crude brine purification, ammonization, and carbonation.

From the analysis of dependencies, we can formulate:

$$\left[N{H}_{3, bound}\right]=68.0+\left(c-95.0\right)+0.56\left(\left(o+e+a+m\right)-92.0\right)+0.57\left(R-190.0\right)-0.14\left(T-25.0\right)$$

and

$$W_{Na}={\mathrm{a}}_1·R+{\mathrm{a}}_2·\left[N{H}_{3, total}\right]+{\mathrm{a}}_3·{\left[C{l}^{-}\right]}_{total}-{\mathrm{a}}_4·T$$

It is apparent from the analysis that the figure in question is directly proportional to the value of the degree of carbonation (R) and the concentration of ammonia $\left(o+e+a+m\right)$ and total sodium chloride (c) in solution (a₁–a₄, numerical coeff.). An increase in the temperature of ammoniated brine at the exit from the carbonation column reduces the efficiency of the carbonation process [9].

Previous industrial experience indicates that maintaining the optimal level of process efficiency primarily depends on the magnitude of ammonia desorption in the carbonation apparatus, which is associated with temperature–hydrodynamic conditions and the distribution of the partial pressure of NH₃ over the carbonated ammonia brine solution.

When considering the equation determining the efficiency of the soda process, it should be noted that only one quantity can be changed, taking into account the other parameters. The concentration of chloride ions is the maximum concentration that occurs in saturated brine. The degree of carbonation depends on the amount of CO₂ introduced into the solution and is limited by the amount of CaCO₃ decomposition. A change in the concentration of total ammonia remains [13].

The kinetics of absorption, that is, the rate of mass exchange, are determined by how far the system is from equilibrium. Of course, the statics, as well as the kinetics, further depend on the properties of the absorbing liquid, the absorbed component, and the inert gas. The kinetics of the absorption process are also affected by the phase contact conditions, which are related to the design of the absorption device [14].

Absorption is accompanied by a thermal effect caused by a change in state of aggregation [15]. Isothermal absorption is possible when the amount of liquid supplied is large compared to the amount of gas (low solubility of the gas) or when the heat of absorption is properly dissipated from the absorption apparatus; otherwise, the absorption will proceed with a change in the temperature of the system. In many cases, it is difficult to delineate a strict boundary between physical absorption and absorption associated with a chemical reaction [16]. As part of ongoing research at the Inowrocław production plant, it was decided to introduce a new “small” absorber (SAB) to increase the concentration of ammonia in the ammonia brine, pre-carbonated. The optimal operation of the additional absorption process translates into higher utilization of CO₂ and the increased efficiency of the soda process.

First, the input and output parameters of the media for SAB were determined. Second, the number of mass transfer units was calculated, with one mass transfer unit corresponding to a section of the apparatus length over which the change in working concentrations of absorbate equals the average driving force over that section. The absorption process is an exothermic process [14,15].

The height of the mass transfer unit is calculated by the following formula, based on [10]:

$$h_y=\frac{V}{K_v·F_0}$$

(1)

The goal of the work carried out was to obtain a stable flux of ammoniated brine, characterized by a direct titer of about 130 mmol∙20 cm⁻³ and determine the chloride ion concentration and CO₂ before and after the absorption process.

2. SAB Calculation and Parameters in Modified Process

The main change in the process is the addition of ammonia absorber. This is an essential part of the modification to achieve the stated objectives of this paper. The input parameters of the ammonia brine were as follows: the absorber load was assumed at a level corresponding to 20–30% of the current load of the carbonation column, Vmax = 15–20 m³∙h⁻¹, ${\mathrm{C}}_{{\mathrm{NH}}_3}$ ≈ 100 mmol∙20 cm⁻³ ≈ 5.00 mol·dm⁻³, C_NaCl ≈ 88 mmol∙20 cm⁻³ ≈ 4.40 mol·dm⁻³, T ≈ 313 K. The output parameters of ammonia brine with higher ammonia concentration were as follows: V_max = 15–20 m³∙h⁻¹, ${\mathrm{C}}_{{\mathrm{NH}}_3}$ ≥ 130 mmol∙20 cm⁻³ ≈ 6.50 mol·dm⁻³, C_NaCl ≈ 83 mmol∙20 cm⁻³ ≈ 4.15 mol·dm⁻³, T ≈ 333K.

A general diagram of the proposed absorber to increase the concentration of ammonia in ammonia brine, pre-carbonated, is shown in Figure 1.

An analysis of the adsorbed gas was performed using FT-IR spectroscopy (Vertex 70V FT-IR spectrometer, Bruker Optics, Ettlingen, Germany). The content of 1 Nm³ of the analyzed gas was 0.6482 Nm³ NH₃, 0.1804 Nm³ CO₂, and 0.1714 Nm³ H₂O. The ammonia content of the input gas (bottom of the absorber) was $Y_{down}=1.8425 \frac{{\mathrm{mol} \mathrm{NH}}_3}{\mathrm{inerts} \mathrm{mol}}$. It was assumed that 90% of ammonia would be absorbed; therefore, $Y_{up \mathit{90}}=0.18425 \frac{{\mathrm{mol} \mathrm{NH}}_3}{\mathrm{inerts} \mathrm{mol}}$. The flow rate of the carbonated brine was set at 15–20 m³·h⁻¹, that is, 16,935.0–22,580.0 kg·h⁻¹. The density of the ammonia brine (with values of 100.76 mmol∙20 cm⁻³ NH₃ and 89.93 mmol∙20 cm⁻³ Cl⁻) was 1.1290 kg·dm⁻¹.

Identifying the solution included determining direct alkalinity, total alkalinity (n), the amount of CO₂ (d), and the content of chloride ions (Cl⁻) (c) and ammonium chloride (NH₄Cl) (b).

The method involves distilling off the ammonia and absorbing it in a standard solution of sulfuric(VI) acid. Excess sulfuric(VI) acid is titrated with a standard solution of sodium hydroxide in the presence of methyl orange. Ammonia contained in liquids in the form of carbonate and bicarbonate salts passes into the gas phase as a result of thermal decomposition of these compounds; this is called free ammonia.

Bound ammonia in the form of NH₄Cl decomposes with a strong alkali such as sodium hydroxide.

2.1. Direct Alkalinity (n)

To determine alkalinity, the sample was titrated with 0.5 M sulfuric(VI) acid in the presence of methyl orange. The titration was carried out until the color of the solution changed from yellow to orange-yellow.

2.2. Amount of CO₂ (d)

The determination was performed on a Scheibler apparatus. The carbonate and bicarbonate ions contained in the sample were decomposed using hydrochloric acid, with the release of free CO₂, which displaces the liquid from the gas burette tube so that the amount of gas released can be determined.

2.3. Content of Cl⁻ (c)

The determination of chloride ions was performed using a Metrohm Ti-Touch 916 compact potentiometric titrator. To perform the measurement, the sample was neutralized, then diluted and acidified with sulfuric(VI) acid. The sample was then titrated with 0.03333 M silver nitrate solution in the presence of a silver electrode, using the appropriate instrument software.

2.4. Content of NH₄Cl (b)

The potentiometric titrator was used to determine the amount of ammonium chloride in the sample. To carry out the measurement, the sample was boiled in lye; this process involved adding the sample to a specific amount of sodium hydroxide and then leaving it to stand on a hot plate to boil off the ammonia. The remaining excess alkali was titrated with 1 M sulfuric(VI) acid in the presence of a glass electrode, using the appropriate program of the instrument.

2.5. Content of NH₂COONH₄ (e)

The presence of carbamate ions was confirmed by ¹³C NMR (Avance 300NMR spectrometer, Bruker, Germany), and the ion content was analyzed using IR spectrometry at a wavelength of 687 cm⁻¹.

All analyses were performed using the standards and methodologies presented in [17].

For the absorption process, based on Whitman–Lewis boundary layer theory and Fick’s first law [18], the mass exchange equation can be formulated as:

$$G=K·F·\Delta y$$

(2)

The driving force of the mass exchange process can be expressed in terms of difference:

$$\Delta y=\left(y-y^{\ast }\right)$$

(3)

As the absorption process proceeds, the driving force of the mass exchange process changes along the absorption apparatus (i.e., the distance of the system from equilibrium changes). For a sufficiently small mass exchange area dF, the mass exchange equation can be written as:

$$dG=K·dF·\left(y-y^{\ast }\right)$$

(4)

From the material balance, it is known that:

$$G=V·\left(y_1-y_2\right) \mathrm{or} dG=Vdy$$

(5)

From the comparison of the above quantities we obtain:

$$dG=K·dF·\left(y-y^{\ast }\right)=Vdy$$

(6)

Considering V from the material balance equation, we obtain:

$$K·dF·\left(y-y^{\ast }\right)=\frac{G}{y_1-y_2}dy$$

(7)

Transforming the equation and integrating the limits from 0 to F and from y₂ to y₁, we obtain:

$$\underset{0}{\overset{F}{\int }}dF=\frac{G}{y_1-y_2}\underset{y_2}{\overset{y_1}{\int }}\frac{dy}{K\left(y-y^{\ast }\right)}$$

(8)

Therefore, we can obtain:

$$G=F·\frac{y_1-y_2}{{\int }_{y_2}^{y_1}\frac{dy}{K·\left(y-y^{\ast }\right)}}=K·F·\Delta y_{sr}$$

(9)

where:

$$\Delta y_{sr}=\frac{y_1-y_2}{{\int }_{y_2}^{y_1}\frac{dy}{y-y^{\ast }}}$$

(10)

Most often, however, strict determination of the size of the surface area of the separation of phases is not possible; therefore, in the formulas used for practical calculations, this size is replaced with a readily available volume size. Assuming that the magnitude of the phase separation area is proportional to the volume, we can calculate:

$$F=\alpha ·F_0·H_w \mathrm{or} dF=\alpha ·F_0·d{H}_w$$

(11)

Eventually, combining the above equations with the basic differential equation, we obtain:

$$d{H}_w=\frac{Vdy}{K_v·F_0·\left(y-y^{\ast }\right)}$$

(12)

Integrating the above relationship from 0 to H_w and from y₂ to y₁, we obtain:

$$H_w=\frac{V}{K_v·F_0}{\int }_{y_2}^{y_1}\frac{dy}{y-y^{\ast }} \mathrm{or}, \mathrm{in} \mathrm{short}, H_w=h_y{m}_y$$

(13)

As an example, it is useful to present the process of calculating the absorber for water. The solubility of ammonia in water at 313 K at a pressure of 1 atm assuming an isothermal process is shown in

Table 1. Solubility of ammonia in water at 313 K at pressure of 1 atm.

$\mathbf{X}=\frac{{\mathbf{mol} \mathbf{NH}}_3}{{\mathbf{mol} \mathbf{H}}_2\mathbf{O}}$	$\mathbf{Y}=\frac{{\mathbf{mol} \mathbf{NH}}_3}{\mathbf{mol} \mathbf{of} \mathbf{inert}}$
0.2647	2.3628
0.2117	1.0822
0.1588	0.5605
0.1058	0.2816
0.0794	0.1865
0.0530	0.1112
0.0423	0.0869
0.0318	0.0629
0.0264	0.0520
0.0211	0.0410
0.0169	0.0327
0.0127	0.0246
0.0105	0.0207

.

We can graph $\frac{\left({\mathrm{Y}}_{\mathrm{down}}{-\mathrm{Y}}_{\mathrm{up}}\right)}{\left({\mathrm{X}}_{\mathrm{down}}^{\ast }{-\mathrm{X}}_{\mathrm{up}}\right)}$, where X_up $=\frac{1.8425-0.18425}{\left(6.5-5.0\right)}=\frac{1.65825}{1.5}=1.1055\times 1.2=1.3266$; hence, X_down = 0.2217. By drawing a graph of $\frac{{\mathrm{mol} \mathrm{NH}}_3}{\mathrm{mol} \mathrm{of} \mathrm{inert}}=f\left(\frac{{\mathrm{mol} \mathrm{NH}}_3}{{\mathrm{mol} \mathrm{H}}_2\mathrm{O}}\right)$, the number of mass transfer units can be determined graphically, as shown in Figure 2.

The column fill height is the product of the number and height of mass transfer units:

$$H_w=h_y{m}_y$$

(14)

The number of mass transfer units is 0.2217.

We assume that the height of mass penetration for the liquid phase is:

$$h_y=\frac{L}{k_{x}aS}$$

(15)

From here, the parameters of the absorber can be easily calculated.

3. Experimental Results

In order to effectively increase the concentration of ammonia in the pre-carbonated brine, it is necessary to determine the effect of the changes on the solubility of NaCl in the pre-carbonated brine so as not to cause NaCl precipitation.

In order to obtain information on the solubility of NaCl, equilibrium solubility tests were conducted on sodium chloride at 303 K in ammonia solution with equilibrium concentrations of 118 and 106.5 mmol∙20 cm⁻³ (initial concentrations 136.5 and 121.6 mmol∙20 cm⁻³, respectively), depending on the degree of carbonation of the solution (R). The results of the experiments are shown in Figure 3 and Figure 4.

The graphs in Figure 3 and Figure 4 show that the solubility of NaCl decreases with increasing NH₃ in solution, and simultaneously increases as the degree of carbonation of the solution rises. At the same time, the trend of increasing solubility with increasing R is more noticeable. That is, considering the analyzed system, with an increase in the degree of carbonation, the solubility of NaCl increases, and this increase is more characteristic when R is in the range of 42–46%.

The traditional Solvay process absorption system implemented at CIECH Soda Polska S.A.’s Inowrocław soda plants is shown in Figure 5.

The process of making ammonia brine containing about 4.8 mol NH₃·dm⁻³ in a saturating apparatus can be carried out with an ammonia concentration of up to 6.5 mol·dm⁻³. In view of the above, ammonia gases from the distillation process should be directed as a gas phase to a small countercurrent absorber with a cooler (SAB). Pre-carbonated ammonia brine is the absorbing liquid. The media are contacted in countercurrent. The cooled liquid after absorption is dispensed into the selected area of the carbonation column. A diagram of how the task is carried out is shown in Figure 6.

The choice of dosing sites for enriched ammonia brine to Cl is a matter of research, which will be published in a subsequent paper. To properly characterize the procedure, the process of ammonia desorption from the system must also be analyzed. Ammonia desorption measurements were carried out to identify the ammonia loss in the sodium process for different ammonia concentrations in the starting solution, in an open system using our apparatus, shown in Figure 7.

The apparatus made it possible to perform the tests at a strictly assumed volume rate of inert q_in, flowing from a buffer tank through a thermostated reactor containing a well-defined volume of ammonia brine of known chemical composition. The pressure and gas flow control block allowed full control of hydrodynamic parameters during the testing of the ammonia desorption process under intensive barbotage conditions in the reactor. The scrubber system, with a standard solution of sulfuric acid(VI) and sodium lye, provided measurements of the amount of ammonia and carbon dioxide in the effluent gas.

In order to determine the effect of the concentration of ammonia in the solution and the degree of carbonation on the intensity of the process under consideration, six solutions of ammonia brine with various chemical compositions were prepared: ${\mathrm{C}}_{{\mathrm{NH}}_3}=4.404\text{–}6.258\mathrm{mol}·{\mathrm{dm}}^{-3}$, ${\mathrm{C}}_{\mathrm{NaCl}}=4.147\text{–}4.267\mathrm{mol}·{\mathrm{dm}}^{-3}$, and ${\mathrm{C}}_{{\mathrm{CO}}_2}=0.000\text{–}1.966\mathrm{mol}·{\mathrm{dm}}^{-3}$.

An investigation of the ammonia desorption process at 303, 313, and 323 K was performed, analyzing the change in concentrations of ammonia, carbon dioxide, and sodium chloride in the liquid phase at specific time intervals.

Table 2. Effect of temperature, gas flow rate, and chemical composition of ammonia brine on ammonia desorption rate (experimental data).

Brine Chemical Composition of coNH3 coNaCl coCO2	Ammonia Concentration in ctNH3 Solution	Temperature T (K)
		303		313		323
		Gas flow Rate (qin dm3 h−1)
		8.0	12.0	8.0	12.0	8.0	12.0
		Desorption Time
(mol·dm−3)/(mval/20 cm3)		t (s)
1	2	3	4	5	6	7	8
4.404/88.08 4.267/85.34 0.000/0.00	4.404/88.08 4.286/85.72 4.167/83.34 4.048/80.96 3.930/78.60	0 665 1210 1765 2320	0 370 740 1090 1490	0 380 765 1095 1460	0 190 385 605 800	0 175 350 515 685	0 100 200 305 415
4.394/87.88 4.247/84.94 0.953/38.12	4.394/87.88 4.276/85.52 4.157/83.14 4.038/80.76 3.920/78.40	0 1100 2190 3355 4565	0 680 1330 2035 2705	0 610 1165 1740 2355	0 525 1025 1640 2215	0 425 920 1330 1775	0 200 385 560 772
4.384/87.68 4.217/84.34 1.375/55.00	4.384/87.68 4.266/85.32 4.147/82.94 4.028/80.56 3.910/78.20	0 1270 2470 3640 4880	0 890 1770 2650 3675	0 940 1870 2790 3800	0 570 1125 1715 2270	0 502 989 1559 2101	0 360 658 1163 1541

and

Table 3. Effect of temperature, gas flow rate, and chemical composition of additionally ammoniated brine on ammonia desorption rate (experimental data).

Brine Chemical Composition of coNH3 coNaCl coCO2	Ammonia Concentration in ctNH3 Solution	Temperature T (K)
		303		313		323
		Gas Flow Rate (qin dm3 h−1)
		8.0	12.0	8.0	12.0	8.0	12.0
		Desorption Time
(mol·dm−3)/(mval/20 cm3)		t (s)
1	2	3	4	5	6	7	8
6.258/125.2 4.148/82.96 0.000/0.00	6.258/125.2 6.140/122.8 6.021/120.4 5.902/118.0 5.784/115.7	0 460 980 1350 1720	0 275 550 780 1080	0 222 455 690 910	0 140 275 415 550	0 120 250 370 500	0 85 170 250 340
6.175/123.5 4.147/82.94 1.265/50.6	6.175/123.5 6.057/121.1 5.938/118.8 5.819/116.4 5.701/114.0	0 520 1030 1565 2095	0 345 685 1030 1385	0 370 715 1090 1460	0 245 495 730 980	0 205 420 615 815	0 145 295 440 590
6.141/122.8 4.209/84.18 1.966/78.64	6.141/122.8 6.041/120.8 5.903/118.1 5.785/115.7 5.666/113.3	0 665 1335 1990 2665	0 475 940 1340 1795	0 500 990 1485 1995	0 370 755 1105 1470	0 330 670 995 1315	0 240 490 725 965

list the measurement results of two series of tests of the kinetics of the ammonia desorption process with the extreme concentration of NH₃ in the brine.

In the following, we summarize the chemical composition of the ammonia brine entering the reactor, the change in concentration of total ammonia in solution, the temperature of the process, the value of the volumetric flow rate of the inert gas, and the time of ammonia desorption.

The effect of individual process parameters on the studied ammonia desorption was evaluated by function analysis:

$$K_{N{H}_3}^{des}=f\left(T, R,c_{N{H}_{3c}}^O,q_{in}\right)$$

(16)

Within the studied range of temperature changes, ammonia brine chemistry, and inert volume flow rate, the desorption intensity of NH₃ was approximated by the equation:

$$N_{N{H}_3}^{des}=K_{N{H}_3}^{des}·c_{N{H}_{3c}}$$

(17)

The experimental data of the desorption process confirm the assumptions of the kinetics of this process. The higher the temperature, the greater the desorption. Because the stream of desorbed ammonia is not linear, experimental studies of the desorption rate were conducted. The desorption tests confirmed that in order for additional ammonia to be bound, it must be dosed to the zone of the carbonation column, where the total CO₂ concentration is as high as possible. As a result, the desorption time will be the longest and the concentration of carbamate ions the highest. Research on this process was carried out at the Faculty of Chemistry of the Nicolaus Copernicus University in Toruń.

4. Solution Implementation and Energy Changes

In order to implement the ammonization process of ammonia brine and pre-carbonated ammonia brine using the SAB plant, a number of technical, technological, and analytical experiments were carried out. In the course of the work, a well-defined flux of ammoniated brine was obtained with direct alkalinity of 130–140 mmol∙20 cm⁻³ (i.e., 6.50–7.00 mol∙dm⁻³).

In the course of the work, favorable conditions were created to carry out the research on a higher concentration of ammonia in the brine, and thus a higher concentration of CO₂ and the co-crystallization of NH₄HCO₃ or NaCl. The absorption of ammonia in aqueous solution can be represented by the following equation:

NH_3(g) + H₂O_(c) ⇌ NH₃∙H₂O_(r) − 35.9 kJ

(18)

The absorption of carbon dioxide in an aqueous ammonia solution can be presented as follows:

2NH_(r) + CO_2(g) + H₂O_(c) ⇌ (NH₄)₂CO_3(r) − 91.7 kJ

(19)

Ammonia is a gas that is easily soluble in water. In this case, the absorption of the gas in water is affected only by the resistance of the gas boundary layer. The rate of this process can be expressed by the formula:

N = K_g S (P_g − P_r) (kmol·h⁻¹·m⁻²)

(20)

The expression (P_g − P_r) is referred to as the driving force of the absorption process, or absorption potential. This potential depends on the partial pressure difference of the absorbate (P_g) in the feed gas and its equilibrium pressure over the liquid (P_r). These values are expressed in mm Hg. In the above equation, K_g expresses the absorption constant in the gaseous boundary layer, the unit of which is kmol·m⁻²·h⁻¹·mmHg⁻¹. For the product of K_g S, an approximation is assumed, which leads to a constant value during the absorption process at each stage under certain conditions. Given this assumption, it can be concluded that the rate of absorption is affected only by the value of the potential. The matter gets a bit more complicated if we also take into account the influence of CO₂. As can be seen from the above reaction equations, the absorption of both components into the liquid causes an exoenergetic reaction, which results in the release of heat and ultimately an increase in the temperature of the solution. This adversely affects the equilibrium pressure of carbon dioxide, by which its absorption efficiency decreases. An increase in temperature significantly accelerates the rate of reaction and reduces the viscosity of the solution, causing favorable conditions for ammonia diffusion, thereby facilitating the absorption of CO₂ according to the chemical reaction. To avoid this, it was assumed that the absorption process would take place at an elevated temperature not exceeding 343 K.

The basic factors that guided the design, construction, and operation of the SAB absorber were as follows:

(a)

Method of conducting the process, continuous or semi-continuous;

(b)

Mass exchange area;

(c)

Liquid retention time;

(d)

Determination of the phase that controls the speed of the process;

(e)

Exoenergetic process: heat removal, use of cooling;

(f)

Probability of corrosion of apparatus; and

(g)

Rate of gas and liquid flow entering the apparatus.

The implementation of the task was carried out according to the layout shown in Figure 6 and Figure 8.

The basic scheme of the apparatus was rearranged in the following manner. Gas after RG-CC (FIC 4080.4) was introduced into the countercurrent SAB absorber, with Pall fill and cellular fill at the top of the apparatus, in countercurrent to the brine solution introduced through two stubs: onto the cellular fill (FIC 4080.9) and onto the Pall fill (FIC 4080.13), with the flux directed onto the Pall fill connected to the return flux (FIC 4080.15). The waste gas was directed to the corresponding absorber.

The ammoniated brine was directed to the cooler (A1.4082.1), from which the stream was divided into two parts: the turnback (FIC 4080.15), which is part of the flux directed to the Pall fill, and the flux of ammoniated brine (FIC 5104.15), which is directed in the start-up and test stage to the absorber tank, and in the carbonation stage to the corresponding CL location. The parameters of the SAB were as follows: The column was installed at +9900, designed as an apparatus consisting of three segments and a gas discharge head. The diameter and height of the apparatus were D = 1000 mm and H = 8000 mm. In the lower segment, there was a brine tank with a volume of V = 0.78 m³, above which is a pre-absorption zone on a bed of Pall rings with a contact area of 240 m². The filling in the lower segment had the following characteristics: Pall rings, 50 × 50 × 2, 316 L, V = 1.80 m³. In the middle segment, flow was designed through two layers of cellular fill. Absorption of about 65% of the ammonia contained in the gas after passing through the Pall rings was assumed on the cellular fill. The characteristics of the fill in the middle segment were as follows: process cell fill of 2 × 3 layers, 316 L, φ 600. The apparatus was designed so that it would also be possible to remove the cell fill and replace it with another bed of Pall rings.

In the upper segment of the column was a designed demister, comprising a separation cell fill of three layers, 316 L, φ 600. A tray was placed under the column, from which all possible leaks from the stubs of the column could be drained by gravity.

The implementation of the SAB operation was carried out in such a way that the gas was started first and the brine flow was second. Once the flows were established, the appropriate amount of cooled brine per turn was determined and the process parameters were observed.

The research objective of the absorption process was to determine the motion parameters of the SAB absorber, in particular the relationship of total alkalinity = f (V brine V1, V brine V2) (Figure 9 and Figure 10), and the parameters under production conditions.

The analysis of the data, presented in Figure 9, allowed us to determine the optimal volume ratio for SAB liquids in order to obtain the highest possible alkalinity. The green–red area is optimal. For the selected area in Figure 9, the volume of the recycle stream was determined. From Figure 10, it can be concluded that in the range of 124–135 mmol∙20 cm⁻³, it increased linearly. The volume of the recycle stream (except from the recycle stream temperature) may be a parameter for controlling the direct alkalinity of the additional ammoniated brine.

The carbamate ion concentration is responsible for the increase in efficiency of the carbonation process. As seen in the results of the data presented in

Table 4. Results of analysis of and calculations for ammoniated brine.

Analyte (mval∙20 cm−3)	Brine after SAB					Brine before SAB
	1	2	3	4	5
a ≡ [NH3]	130.6	132.8	135.0	129.1	126.6	98.9
c ≡ [Cl−]	83.1	81.1	80.9	81.2	80.8	87.6
q ≡ [Na+]	83.1	81.1	80.9	81.2	80.8	87.6
m ≡ [NH3(r)]	54.4	55.3	81.9	103.4	76.9	79.7
n ≡ [alkalinity]	130.6	132.8	135.0	129.1	126.6	98.9
o ≡ [NH4+]	49.5	50.3	36.1	19.3	34.3	12.9
pH	9.61	9.62	9.93	10.30	9.92	9.40
d ≡ [CO2]	78.3	79.7	54.0	25.8	50.6	19.8
w ≡ [CO32−] + [HCO3−] + [CO2(r)]	25.1	25.2	20.1	13.4	19.7	7.3
e ≡ [NH2COO−]	53.2	54.5	33.9	12.4	30.9	12.5
f ≡ [CO32−]	20.7	20.8	18.3	12.9	17.9	6.0
g ≡ [HCO3−]	4.4	4.4	1.9	0.56	1.9	1.3
k ≡ [acidity]	78.33	79.7	54.0	25.8	50.6	19.8
I (ionic force)	197.2	198.4	168.9	129.5	165.9	95.8

, the concentration of carbamate ions increased significantly during the addition of ammonification, while the parameters of the brine were maintained. The higher alkalinity due to the higher concentration of ammonia caused a greater fixation of the CO₂ contained in the reaction solution. Increasing the concentration of NH₂COO⁻ ions caused an increase in supersaturation in the crystallization area of NaHCO₃. This effect is very favorable for the efficiency of the soda process.

Process energy changes are best presented in the form of a Sankey diagram [20,21]. The energy effect of the carbonation process is presented in Figure 11.

In the experiment, after increasing the concentration of ammonia in the carbonation process, there was an increase in temperature at the exit from the column. The efficiency of the soda process increased slightly, due to the suction effect on NaHCO₃. This also has a beneficial effect on the process in summer conditions, when it is difficult to cool the carbonation column.

From the data presented in Table 4, it can be seen that the composition of the ammoniated brine strictly depends on the composition of the absorbed gas, and in particular on the amount of NH₃ and CO₂ in the gas fed into the SAB. The fact that the amount of CO₂ in the gas after RG-CC translates directly into the amount of carbamate ions in the solution after SAB should receive special attention.

For the carbonation process, ammoniated brine solutions containing the highest possible concentration of NH₂COO⁻ ions are preferred, which will cause a direct increase in supersaturation in CL.

5. Environmental Aspects

Minimized cooling water consumption resulting from the reduced cooling by 152.4 MJ∙Mg⁻¹ was the main effect of the change in process regime. An additional ecological effect is presented in Figure 12. The concentrations of CO₂ and NH₃ in the exhaust gas were measured by infrared absorption using an LDS-6 laser process gas analyzer (Siemens, Germany). The diagram shows that during dosing of additional ammoniated brine, the CO₂ and NH₃ concentrations in the exhaust gases decreased. Additional ammoniated brine dosing was alternated with periods of no dosing. The CO₂ concentration exceeded 25% v/v without additional ammoniated brine and was on average 3.85% lower with ammoniated brine. Likewise, the concentration of NH₃ was 0.63% lower on average. Lowering the concentration of CO₂ and NH₃ in the waste gas has a very large ecological benefit in terms of total NaHCO₃ production.

As a result of extensive analytical and industrial research at CIECH Soda Polska S.A.’s Inowrocław soda plants using ammoniated brine, the soda yield of the process was increased using the developed SAB device. Acknowledging that the soda industry is troublesome in terms of waste [1,16,18], this reduction in the amount of substrate at higher yields also minimizes the amount of process by-product. An additional environmental aspect, due to the thermodynamics of the process, is the increase in temperature at the outlet of the carbonation column, with unchanged crystal quality. This results in minimized cooling water consumption, which minimizes the cost of producing NaHCO₃.

6. Conclusions

The study shows that the rate of ammonia desorption increases exponentially with increasing temperature and inert volumetric flow rate, while it decreases with increasing carbonation of the system. The value of the kinetic constant of the ammonia desorption process at an inert volume flow rate of 8.0 dm³·h⁻¹ varied in the range 2–24 × 10³ mmol·dm⁻³·s⁻¹, while at a volumetric flow rate of 12.0 dm³·h⁻¹ the range was 4–27 × 10³ mmol·dm⁻³·s⁻¹. The above relationships provide the basis for determining the optimal range of process parameters for the initial carbonation of ammonia brine using Solvay technology. Considering the hydrodynamic data of the SAB absorber, after conducting a series of experiments, the results of which are presented above, it can be assumed that the model parameters of the system are as follows: Average flux of pre-carbonated brine directed to the Pall fill (V2) is about 8 m³∙h⁻¹; average flux of pre-carbonated brine directed to the cell fill (V1) is about 12 m³∙h⁻¹; average turning flux is about 10 m³∙h⁻¹; and increase in total alkalinity is about 36 mmol∙20 cm⁻³. The increase in overall CO₂ in ammoniated brine can be described by the relation: y = 0.86x − 29.2 where y is overall CO₂ and x is general alkalinity; the average gas flow for the above streams is about 2800 Nm³∙h⁻¹. The MAB operating temperature is 334–340 K, and the mass exchange flux of ammonia between gas phase and liquid is 61.2 kg NH₃∙h⁻¹∙m⁻³; thus, this is the value of the transfer coefficient of ammonia under fixed hydrodynamic conditions. The temperature of the bottom of the carbonation column may be 307 K without additional cooling. This saves energy due to the need for less cooling water.