A Comparative Study of Interfacial Area Obtained by Physical and Chemical Methods in a Bubble Column

Determination of interfacial areas in absorption processes have been experimentally investigated in a bubble column using a physical method (photographic method) and a chemical method (Danckwerts ́ method). Experiments were conducted in a vertical column 11.3 cm in diameter and 108.6 cm in height. The range of variables used is common for both methods. All experiments were carried out in batch regime. All values of area were correlated with variables grouped in dimensionless modules that reproduce experimental values with deviations below 9%, and these values obtained by the photographic method are of the same magnitude as those obtained by the chemical method, maximum deviation of these values being 10 %. INTRODUCTION For the experimental determination of the mass transfer coefficient (kL) it is inevitable to first know the gas-liquid interfacial area of the contact system. One could think that knowing the value of the product kLa would be sufficient for the calculation and design of gas-liquid absorbers and reactors. However, the knowledge of the area can be so important as the knowledge of the mass transfer coefficient to analyse thoroughly the mass transfer process and the influence of the different variables. Interfacial area can be determined of different ways depending on the gas-liquid contact system employed. In many cases it is possible to determine it by geometric considerations, while in other cases, it is necessary to determining it experimentally. The determination of interfacial areas is possible by one of the existing methods [15]. If the absorption process is accompanied by a fast chemical reaction, the interfacial area becomes the principal design criteria. The area can vary in a wide range, depending on the type of contact equipment, physical and chemical properties of the gas-liquid system, and on the operation conditions. Thus, specific techniques are necessary for its determination, numerous methods being proposed, each one with its advantages and disadvantages. The different methods used in the determination of interfacial areas can be classified as physical [6-11] or chemical methods [12-14]. The first methods are based on modifying some physical property, while the second methods are based on carrying out a chemical reaction of known kinetics, in such a way that the absorption rates be proportional to the area. The results obtained using different methods, physical and chemical, have sometimes given contradictory results [12, 15]. In the present study, the interfacial area is determined by two different methods, the Danckwerts ́ method and the photographic method, analysing the influence of the different variables on the area and obtaining the corresponding correlations. MATERIAL AND METHODS Mass transfer measurements The absorption experiments were carried out in the set-up that, with the exception of the contact device, has been described in detail in previous studies [16]. The bubble column, (Fig. 1), made of methacrylate, 108.6 cm high and 4 mm thick. The internal and external diameters of the column are 11.3 cm and 14.8 cm, respectively. The gas distributor was a porous plate, 4 cm in diameter. The size of the bubbles is modified by using porous plates of different equivalent pore diameter, allowing an important variation in the interfacial area. Table 1 shows the relation between porosity and the equivalent pore diameter for each of the porous plates employed. The liquid phase enters the column through the top via a vertical glass tube that is slightly bent at the lower end to avoid gas leakage. Table 1. Relation between porosity and the equivalent pore diameter for each of the porous plates employed Plate Equivalent pore diameter · 106 (m) 0 150-200 1 90-150 2 40-90 The gas, pure CO2, is led through a humidifier, where, by bubbling, is saturated water vapor at 25 oC. Once saturated, it passes through a bubble flow-meter, where the flow into the column is measured. The gas not absorbed, after making contact with the liquid phase, exits the column, passes to a second bubble flow-meter determining the gas flow at the exit. The amount of gas absorbed is determined by the difference of flow entering and exiting the column. In our experiments, we used inflow between 3 and 8.5 10 mol/s. The liquid, previously thermostated to 25 oC, is introduced into the column in loadings of 10.3 l. Photographic method This method consists in photographing different zones of the column during the absorption process and determining the diameter and number of bubbles in this zone. Once this is known, we can calculate the total number of bubbles and area occupied by these that would correspond to the gas-liquid contact area. This method was employed by several authors to determine interfacial areas in a column of spheres and cylinders [17], or to study the reaction of carbonic gas with carbonatebicarbonate solution in the presence of hypochlorite as catalyst [18]. Figure 1. Bubble column scheme: 1. Bubble column; 2. Thermometer; 3. Inflow of liquid ; 4. Outflow of gas; 5. Porous plate; 6 and 7. Outflow of liquid Danckwerts ́ method The method proposed by Danckwerts requires that the gas absorbed undergo a moderately fast pseudo-first order reaction with some of the solutes of the liquid phase. The value of interfacial area (A) is obtained from the slope of the straight line that results from the graphic representation of absorption flux per volume (N) versus pseudo-first order kinetics constant (k1), while the value of kL is calculated from the ordinates at the origin. This method has the advantage that appreciable differences of k1 are obtained using very small catalyst concentrations thereby not affecting the physical properties of the absorbent liquid [19]. RESULTS AND DISCUSSION Photographic method Upon application of the photographic method experiments were designed in advance to analyse the influence of the different variables on contact area and to determine the minimum number of experiments needed. In the last few years, several investigators have used factorial designs to cover problems of this type in other research fields [20, 21]. We have employed a two-level factorial design, 2, for the determination of area by the photographic method. We have focused the study on three independent variables of greater influence over interfacial area, superficial gas velocity (X1), pore size (X2) and surfactant concentration (X3), whose maximum and minimum values are: Superficial gas velocity: 0.8·10 1.8·10 m/s; Porous plate: 0 2; Concentration of sodium lauryl sulphate (SLS): 5·10 – 5·10 % mass. For mathematical studies it is necessary to use normalised independent variables. The correlation of the variables will be given, depending on the mathematical model, in the following form: 3 2 23 3 1 13 2 1 12 3 3 2 2 1 1 o X X b X X b X X b X b X b X b b y + + + + + + = (1) where: y is the dependent variable; Xi are the normalised independent variables; bi and bii are the model coefficients. Photographs were taken of the column during the absorption process, for different operating conditions, gas flow, liquid phase, and pore size. Several photographs were taken for each system and in different zones of the column. For each photograph, a count of the bubbles was taken in a prefixed interval, and their diameters were determined (see fig 2). Contact area was calculated with these data. Each experiment was performed five times. Figure 2. Photograph of bubbles of porous plate 2 and 0. The data obtained from the photographs is collected, along with those of calculation of the gas-liquid interfacial area. The proposed equation is the following: 3 2 3 1 2 1 3 2 1 X X 100 . 0 X X 310 . 0 X X 513 . 0 X 470 . 0 X 095 . 1 X 690 . 4 520 . 11 y + − + − + + = (2) Pareto Chart is shown in figure 3, it is a histogram showing the influence of each independent variable over the dependent. The principal effects are presented standardised (effects divided by standard error). All those effects that possess a value superior to level 0.05 (value indicated on the graph) are significant at that level.