water solution in a rotating packed bed

water solution in a rotating packed bed

Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 418–423 Contents lists available at ScienceDirect Journal of the Taiwan Institute of...

354KB Sizes 0 Downloads 26 Views

Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 418–423

Contents lists available at ScienceDirect

Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice

Absorption of ethanol into water and glycerol/water solution in a rotating packed bed Chia-Ying Chiang, Yu-Shao Chen, Mao-Shih Liang, Fang-Yi Lin, Clifford Yi-Der Tai, Hwai-Shen Liu * Department of Chemical Engineering, National Taiwan University, Taipei 106, Taiwan

A R T I C L E I N F O

A B S T R A C T

Article history: Received 20 July 2008 Received in revised form 27 November 2008 Accepted 28 November 2008

Rotating packed bed (RPB) in which the centrifugal force is employed plays an important role in the field of process intensification. With the help of centrifugal force, the wider operating range and the better mass transfer efficiency could be expected. Experimental results showed that the gas-phase mass transfer coefficient (KGa) increased with increasing rotational speeds, gas flow rates and liquid flow rates, but decreased with increasing the liquid viscosity. However, with the comparison of KGa in an RPB and the traditional packed columns, the enhancement of mass transfer coefficient was remarkable, up to 193-fold mass transfer efficiency in the viscous media. ß 2008 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Keywords: Rotating packed bed Absorption Mass transfer Ethanol Glycerol

1. Introduction In the past, water was used as the absorbent in most of the absorption researches by the so-called rotating packed bed (RPB) systems (Chen and Liu, 2002; Guo et al., 1997; Jassim et al., 2007; Lin et al., 2004a,b, 2006; Munjal et al., 1989; Tan and Chen, 2006) which were developed based on the concept of ‘‘process intensification’’ (Ramshaw and Mallinson, 1981). However, not all of the pollutants or materials can be trapped into water-phase effectively. An example is the absorption of some hydrophobic volatile organic compounds that need to be treated by alternative absorbents such as vegetable oil, mineral oil, diesel oil and silica oil (Heymes et al., 2006; Pieriucci et al., 2005; Poddar and Sirkar, 1996; Xia et al., 1999). Take those absorbents into consideration, it could be noted that they are viscous fluids, some of their viscosities even a hundred times of water. As a result, those viscous fluids might not be welcome to be introduced into the conventional packed column due to the hydrodynamic and mass transfer problems. In this point of view, rotating packed bed could be an excellent alternative with high mass transfer efficiency to handle those high viscosity fluids.

* Corresponding author. Tel.: +886 2 3366 3050; fax: +886 2 2362 3040. E-mail address: [email protected] (H.-S. Liu).

In a rotating packed bed, the gravity force is replaced with a centrifugal force up to several hundred folds of g to enhance mass transfer efficiency. Generally, an RPB consists a rotating doughnut-shaped packing element driven by a motor in a static housing. The liquid flowing through the packing elements radically is subjected to a high acceleration, resulting in thin liquid films, tiny liquid droplets and chaotic flow pattern. As a result, a dramatic increase in gas–liquid interfacial area and mixing efficiency can be achieved in an RPB. Moreover, the system can be operated at higher gas–liquid ratios because of the decreasing tendency of flooding. Therefore, 1–2 order of magnitudes enhancement in mass transfer could be observed frequently in an RPB, and the size of the equipments would be greatly reduced as compared with a conventional packed column. This would lead to a significant reduction in capital and operating costs. The RPB has also been proved to improve mass transfer in many processes other than absorption, such as stripping (Liu et al., 1996; Singh et al., 1992), distillation (Kelleher and Fair, 1996; Lin et al., 2002), adsorption (Lin and Liu, 2000; Lin et al., 2004a,b), and reactive precipitation (Chen et al., 2000, 2006; Tai et al., 2006). In 2005, Chen et al. studied the influence of liquid viscosity on the mass transfer rate for deoxygenation of glycerol solution (Newtonian fluid) and CMC solution (non-Newtonian fluid). They found that the centrifugal force was effective in enhancing the liquid-phase controlled mass transfer coefficient (KLa) in viscous

1876-1070/$ – see front matter ß 2008 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jtice.2008.11.006

C.-Y. Chiang et al. / Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 418–423

Nomenclature a ac at A CG CG,i CG,o CL CL,i CG dp DG DL g G H kG kGa kL kLa KG a L QG QL r ri ro zb

effective gas–liquid interfacial area per unit volume (m2/m3) centrifugal acceleration (m/s2) total particle surface area per unit volume of the packed bed (m2/m3) absorption factor defined as Eq. (4) concentration of solute in the gas stream (mol/m3) concentration of solute in the inlet gas stream (mol/m3) concentration of solute in the outlet gas stream (mol/m3) concentration of solute in the liquid stream (mol/m3) concentration of solute in the inlet liquid stream (mol/m3) equilibrium concentration associate with the liquid concentration (mol/m3) Spherical equivalent diameter of the packing = 6(1  e)/atc (m) diffusion coefficient in gas (m2/s) diffusion coefficient in liquid (m2/s) gravitational force (m/s2) gas mass flux (kg/m2 s) Henry’s constant [(mol/m3)/(mol/m3)] gas-phase mass transfer coefficient (m/s) volumetric gas-phase mass transfer coefficient (s1) liquid-phase mass transfer coefficient (m/s) volumetric liquid-phase mass transfer coefficient (s1) overall volumetric gas-phase mass transfer coefficient (s1) liquid mass flux (kg/m2 s) gas flow rate (m3/s) liquid flow rate (m3/s) radius of the packed bed (m) inner radius of the packed bed (m) outer radius of the packed bed (m) axial height of the packing (m)

media. Besides, they also noted that the dependence of mass transfer coefficients on liquid viscosity was less in an RPB than in a conventional packed column. However, the effect of liquid viscosity on absorption processes which is predominately controlled by gas-phase resistance has not been evaluated yet. In this study, the influence of liquid viscosity on KGa was first examined with an ethanol/glycerol–water solution absorption process and then the gas-phase mass transfer coefficient (KGa) comparison between RPB and the conventional packed bed was also provided. 2. Experiments The main structure of the RPB is demonstrated in Fig. 1. The liquid is pumped into the system and flows out from the liquid distributor to the rotating packing. With the help of centrifugal force, liquid can move fast within the packing in the forms of thin films and tiny droplets which would increase the interfacial area of mass transfer. On the other hand, gas is introduced into the RPB by a compressor and flows from the stationary housing toward the central pipe of gas outlet radically. While gas and liquid contact each other counter-currently in the packing, a dramatic disturbance is expected, and thus resulting in good mass transfer performance. The packing used in this study was 0.22-mm diameter stainless steel wire meshes, whose interfacial surface area and porosity were 1024 m2/m3 and 0.944, respectively. The bed can be operated from 600 to 1800 rpm. The rotating packed bed with the inner and outer radii and the axial height were 2, 4 and 2 cm, respectively. In this experiment, two RPBs have been used, RPB-I and RPB-II, which represent two RPBs with 6 cm and 9 cm in radii of the stationary housing, respectively. Fig. 2 shows the diagram of the experimental setup in this study for ethanol absorption. The absorbents, glycerol/water solutions, at a temperature of 30 8C were pumped into the rotating packed bed. An air stream was introduced to a bubbler containing aqueous

Greek symbols viscosity of the gas (kg/m3) viscosity of the liquid (kg/m3) density of the gas (kg/m3) density of the liquid (kg/m3) liquid surface tension (N/m) critical surface tension of packing (kg/s2) sphericity of packing

mG mL rG rL s sc c

Dimensionless groups 2 Froude number of liquid, Fr L ¼ rL 2aat FrL L c ReG

Reynolds number of gas, ReG ¼ at Gm G

ReL

Reynolds number of liquid, ReL ¼ at Lm L

ScG

Schmidt number of gas, ScG ¼ r DG G G

ScL

Schmidt number of liquid, ScL ¼ r

WeL

Webber number of liquid, WeL ¼ r

m

mL

L DL L2

L at s L

419

Fig. 1. Main structure of an RPB.

420

C.-Y. Chiang et al. / Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 418–423

Fig. 2. Diagram of the experimental setup for VOC absorption.

Table 1 Viscosities and Henry’s constants of glycerol/water solutions. Glycerol/water concentration (wt%)

Liquid viscosity (mPa-s)

H [(mol/m3)/ (mol/m3)]

0 26 62 80 88

1.04 1.95 9.32 40.5 102.8

5.50  104 4.47  104 4.17  104 3.28  104 2.75  104

transfer coefficient. CG represents the equilibrium concentration associated with the liquid concentration. And the overall mass balance can be written as for the case of solute-free liquid input (CL,i = 0):    C (2) Q G ðC G  C G;o Þ ¼ Q L ðC L  C L;i Þ ¼ Q L G  0 H i.e., CG ¼

ethanol, diluted with another air stream to the desired concentration. Meanwhile, in order to maintain stable gas concentration, a buffer flask was added into the system. Then the liquid and the gas streams were contacted counter-currently in a rotating packed bed. The concentrations of the inlet and outlet gas streams from gas-collecting tubes were measured by a gas chromatography (PerkinElmer Autosystem) equipped with a FID and a fused-silica capillary column (Supelco 2-5349). Nitrogen was used as the carrier gas. The injector, column, and detector temperatures were set at 190, 200, and 250 8C, respectively. The physical properties (viscosity and Henry’s constant) of glycerol/water solutions are presented in Table 1. The viscosities were measured by a viscometer (Brookfield, model DV-II+). Henry’s constants were determined by the bubbling method (Heymes et al., 2006). The gas stream with a fixed concentration of ethanol was introduced into a bubbler containing the absorbent at 30 8C, and the concentration of the outlet gas stream from the bubbler was measured by a gas chromatography until reaching saturation, i.e. equal to the inlet gas composition. Then with the concentrations of the outlet gas stream and the absorbent, Henry’s constant could be determined. 3. Results and discussion The overall volumetric gas-phase mass transfer coefficient (KGa) of an RPB can be obtained with the concept of mass balance and transfer unit. First, consider a differential volume with crosssection area 2przb and thickness dr. The mass balance of solute in this differential volume for a dilute system is Q G dC G ¼ K G aðC G  CG Þ2przb dr

(1)

where QG and CG are the gas flow rate and the concentration of solute (ethanol) in gas-phase, and KGa stands for the overall mass

1 ðC G  C G;o Þ A

(3)

where CG,o is the outlet ethanol concentration in the gas stream and H is the Henry’s constant listed in Table 1 for various conditions. Besides, A is absorption factor defined as A¼

QL HQ G

(4)

Then the mass transfer coefficient can be obtained by substituting Eq. (3) into Eq. (2) and integrating the equation from r = ri to r = ro with the boundary conditions CG = CG,o and CG = CG,i, respectively.   ln ð1  1=AÞðC G;i =C G;o Þ þ 1=A QG (5) KGa ¼ 1  1=A pðro2  ri2 Þzb The effects of liquid flow rate, ranging from 0.30 to 0.78 L/min, on KGa for different rotational speeds are shown in Fig. 3. As shown in the figure, the KGa values increased with an increase of the liquid flow rate. This was probably because that increasing liquid flow rates would lead to more liquid films spreading on the packing and more liquid droplets flying in the voidage of the bed, thus providing larger gas–liquid interfacial area. However, a smaller KGa was found for the liquid of 0.3 L/min. It might be explained by the uneven distribution of liquid at a lower liquid flow rate. As reported in the literature (Burns and Ramshaw, 1996), liquid tends to flow together and form a ‘‘pore flow’’ for low liquid flow rates, resulting in a decrease in gas–liquid contact area and a lower transfer coefficient. In addition, the exponent (y) of K G a / QLy was correlated and the y values were obtained as 0.51, 0.44, 0.41 and 0.33 at the rotor speeds of 900, 1200, 1500 and 1800 rpm, respectively. The exponent decreased as the rotational speed increased. Similar values and trends of the exponents were also observed by Lin et al. (2004a,b). However, a smaller y value, 0.36, was acquired at the rotor speed

C.-Y. Chiang et al. / Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 418–423

421

what the gas flow rate is, the mass transfer coefficient could be increased by raising the rotational speed because the centrifugal force was capable of reducing the mass transfer resistance for the ethanol absorption. Then, glycerol/water solutions were used as viscous absorbents for ethanol absorption. Fig. 5 showed the dependence of KGa on liquid viscosity varying from 1 to 102.8 mPa-s for different rotor speeds. The experimental results in Fig. 5(a)–(c) were performed with liquid flow rates of 0.1, 0.2 and 0.3 L/min, respectively. It is found in the figures that KGa would clearly decrease when the viscosity of the liquid increased. An increase in liquid viscosity would lead to thicker liquid films, resulting in less gas–liquid interfacial area. And the mass transfer efficiency was reduced with an increase of viscosity, consequently. To

Fig. 3. Dependence of KGa on liquid flow rate for different rotor speeds.

of 600 rpm. This is probably also because the liquid could not evenly spread in the bed at a low rotor speed, especially in the axial direction. Fig. 4 displayed the effects of gas flow rates, ranging from 10 to 20 L/min, on KGa for the rotational speeds, from 600 to 1800 rpm, providing roughly 12–109 of g force, based on the arithmetic mean radius. It is found in the figure that KGa would increase clearly with the increasing gas flow rate which implied that the gas-phase mass transfer resistance was reduced with the increasing gas flow rate. In addition, KGa was proportional to QG with the exponent z ranging from 0.65 to 0.87 for different rotor speeds. The exponent z would increase with an increase of the rotational speed. This characteristic is similar to the findings reported by Lin et al. (2004a,b). On the other hand, it is found that the KGa could be expressed as the form of vx. If the data of Fig. 4 are analyzed for different gas flow rates, it can be found that the exponent (x) increases from 0.55 to 0.69 while the gas flow rate increases from 10 to 20 L/min. This observation implied that the enhancement of mass transfer efficiency by the centrifugal force would be more pronounced at high gas flow rates. No matter

Fig. 4. Dependence of KGa on gas flow rate for different rotor speeds.

Fig. 5. Dependence of KGa on liquid viscosity for different rotor speeds; liquid flow rate are (a) 0.1 L/min, (b) 0.2 L/min and (c) 0.3 L/min.

C.-Y. Chiang et al. / Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 418–423

422

Table 2 The ratio of KGa’s in an RPB to those in a conventional packed column. Liquid viscosity (mPa-s)

Ratio of KGa’s in an RPB to those in a conventional packed column

1.04 1.95 9.32 40.5 102.8

3.25–8.96 2.40–10.6 10.1–25.8 42.9–138.5 81.4–192.6

understand about the influences of viscosity further, KGa was represented as mw where the exponent (w) ranging from 0.21 to 0.32. On the other hand, it is also noted in Fig. 5 that the mass transfer coefficient increased with increasing rotational speeds for all the viscosities. This implied that the centrifugal force could also reduce the gas-phase mass transfer resistance effectively in the viscous liquid media. Besides, though the effect of liquid viscosity on KGa has not been experimentally investigated in literatures, it is found that the dependence of KGa on the liquid viscosity is less than those proposed by Liu et al. (1996) (w = 0.631) and Lin et al. (2004a,b) (w = 0.507), but similar to the exponent (w = 0.328) proposed by Chen and Liu (2002). To compare the effects of the liquid viscosity on KGa in an RPB with that in a conventional packed column, the correlations, i.e. Eqs. (6)–(9), based on two-film theory which were provided by Onda et al. (1968) were used to calculate the corresponding KGa in a conventional packed column. K Ga ¼

1 1=kG a þ H=kL a

(6)

1=3

2

kG ¼ 2at DG Re0:7 G ScG ðat dp Þ

(7)

   s 0:75 c 0:05 a ¼ at 1  exp 1:45 Re0:1 We0:2 L FrL L

(8)

s

kL ¼ 0:0051



rL mL g

1=3 

L

mL a

2=3

0:4

ScL0:5 ðat dp Þ

(9)

By assuming above equations valid for viscous absorbent and substituting Eqs. (7)–(9) into Eq. (6), the mass transfer coefficient for a conventional packed column can be obtained. The ratio of KGa values in an RPB to those in a conventional packed column was evaluated as a function of liquid viscosity, shown in Table 2. For example, the mass transfer coefficient in an RPB was 3.25–8.96 times higher than in a conventional packed column when taking water as the absorbent. However, once the liquid viscosity was increased to 102.8 mPa-s, a much higher ratio (81.4–192.6) could be achieved. It could be found that as liquid viscosity increased, the ratio became larger which implied that operating under a higher liquid viscosity by an RPB could provide a more remarkable enhancement on mass transfer. Moreover, when considering the effects of rotor speed, it is found that as a higher rotor speed was applied, a higher (KGa)RPB/(KGa)c ratio could be obtained. This also reinforced the capability of centrifugal force to intensifying the mass transfer efficiency. 4. Conclusion It has been already proved experimentally that mass transfer could be enhanced by the so-called ‘‘Higee’’ system. Intuitively, with the unique trait of high mass transfer efficiency, a Higee

system may be applicable to handle viscous fluids. However, very little investigation on this respect has been reported, thus this work examined the absorption performance of ethanol in rotating packed beds using glycerol/water solutions as viscous absorbents whose viscosity ranging from 1 to 102.8 mPa-s. The overall mass transfer coefficient (KGa) was found as a function of rotor speed with the exponent of 0.55–0.69, and it was proportional to the gas flow rate and liquid flow rate with the exponent of 0.65–0.87 and 0.33–0.51, respectively. It was also found that KGa decreased with liquid viscosity by the exponent of 0.21–0.32. The higher gas flow rate could help the system with higher mass transfer efficiency by increasing the turbulence of the system. Meanwhile, increasing the rotor speed and liquid flow rate or decreasing liquid viscosity could force liquid to flow in evenly thin films or as extremely tiny drops flying in the voidage of the packing, so that mass transfer efficiency could be enhanced. Moreover, by comparing the experimental data of the RPB with the estimations based on the correlations of conventional packed columns, it was found that an RPB could provide remarkable enhancement on mass transfer in the viscous media, up to about 193 times. Acknowledgement The support from Ministry of Economic Affairs, Taiwan, Republic of China, is greatly appreciated. References Burns, J. R. and C. Ramshaw, ‘‘Process Intensification: Visual Study of Liquid Maldistribution in Rotating Packed Beds,’’ Chem. Eng. Sci., 51, 1347 (1996). Chen, J. F., Y. H. Wang, F. Guo, X. M. Wang, and C. Zheng, ‘‘Synthesis of Nanoparticles with Novel Technology: High-Gravity Reactive Precipitation,’’ Ind. Eng. Chem. Res., 39, 948 (2000). Chen, Y. S. and H. S. Liu, ‘‘Absorption of VOCs in a Rotating Packed Bed,’’ Ind. Eng. Chem. Res., 41, 1583 (2002). Chen, Y. S., C. C. Lin, and H. S. Liu, ‘‘Mass Transfer in a Rotating Packed Bed with Viscous Newtonian and Non-Newtonian Fluids,’’ Ind. Eng. Chem. Res., 44, 1043 (2005). Chen, Y. S., C. Y. Tai, M. H. Chang, and H. S. Liu, ‘‘Characteristics of Micromixing in a Rotating Packed Bed,’’ J. Chin. Inst. Chem. Engrs., 37, 63 (2006). Guo, F., C. Zheng, K. Guo, and N. C. Gardner, ‘‘Hydrodynamics and Mass Transfer in Crossflow Rotating Packed Bed,’’ Chem. Eng. Sci., 52, 3853 (1997). Heymes, F., P. M. Demoustier, F. Charbit, J. L. Fanlo, and P. A. Moulin, ‘‘New Efficient Absorption Liquid to Treat Exhaust Air Loaded with Toluene,’’ Chem. Eng. J., 115, 225 (2006). Jassim, M. S., G. Rochelle, D. Eimer, and C. Ramshaw, ‘‘Carbon Dioxide Absorption and Desorption in Aqueous Monoethanolamine Solutions in a Rotating Packed Bed,’’ Ind. Eng. Chem. Res., 46, 2823 (2007). Kelleher, T. and J. R. Fair, ‘‘Distillation Studies in a High-Gravity Contactor,’’ Ind. Eng. Chem. Res., 35, 4646 (1996). Lin, C. C. and H. S. Liu, ‘‘Adsorption in a Centrifugal Field: Basic Dye Adsorption by Activated Carbon,’’ Ind. Eng. Chem. Res., 39, 161 (2000). Lin, C. C., T. J. Ho, and W. T. Liu, ‘‘Distillation in a Rotating Packed Bed,’’ J. Chem. Eng. Jpn., 35, 1298 (2002). Lin, C. C., T. Y. Wei, W. T. Liu, and K. P. Shen, ‘‘Removal of VOCs from Gaseous Streams in a High-Voidage Rotating Packed Bed,’’ J. Chem. Eng. Jpn., 37, 1471 (2004a). Lin, C. C., Y. S. Chen, and H. S. Liu, ‘‘Adsorption of Dodecane from Water in a Rotating Packed Bed,’’ J. Chin. Inst. Chem. Engrs., 35, 531 (2004b). Lin, C. C., T. Y. Wei, S. K. Hsu, and W. T. Liu, ‘‘Performance of a Pilot-Scale Cross-Flow Rotating Packed Bed in Removing VOCs from Waste Gas Streams,’’ Sep. Purif. Technol., 52, 274 (2006). Liu, H. S., C. C. Lin, S. C. Wu, and H. W. Hsu, ‘‘Characteristics of a Rotating Packed Bed,’’ Ind. Eng. Chem. Res., 35, 3590 (1996). Munjal, S., M. P. Dudukovic, and P. Ramachandran, ‘‘Mass-Transfer in Rotating Packed Beds—II. Experimental Results and Comparison with Theory and Gravity Flow,’’ Chem. Eng. Sci., 44, 2257 (1989). Onda, K., H. Takeuchi, and Y. Okumoto, ‘‘Mass Transfer Coefficient between Gas and Liquid Phases in Packed Columns,’’ J. Chem. Eng. Jpn., 1, 56 (1968). Pieriucci, S., R. D. Rosso, D. Bombardi, A. Concu, and G. Lugli, ‘‘An Innovative Sustainable Process or VOCs Recovery from Spray Paint Booths,’’ Energy, 30, 1377 (2005). Poddar, T. K. and K. K. Sirkar, ‘‘Henry’s Law Constant for Selected Volatile Organic Compounds in High-Boiling Oils,’’ J. Chem. Eng. Data, 41, 1329 (1996). Ramshaw, C. and R. H. Mallinson, ‘‘Mass Transfer Process,’’ U.S. Patent, 4, 283, 255 (1981).

C.-Y. Chiang et al. / Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 418–423 Singh, S. P., J. H. Wilson, R. M. Counce, J. F. Villiersfisher, H. L. Jennings, A. J. Lucero, G. D. Reed, R. A. Ashworth, and M. G. Elliott, ‘‘Removal of Volatile Organic-Compounds from Groundwater Using a Rotary Air Stripper,’’ Ind. Eng. Chem. Res., 31, 574 (1992). Tai, C. Y., C. T. Tai, and H. S. Liu, ‘‘Synthesis of Submicron Barium Carbonate Using a High-Gravity Technique,’’ Chem. Eng. Sci., 61, 7479 (2006).

423

Tan, C. S. and J. E. Chen, ‘‘Absorption of Carbon Dioxide with Piperazine and Its Mixtures in a Rotating Packed Bed,’’ Sep. Purif. Technol., 49, 174 (2006). Xia, B., S. S. Majumdar, and K. K. Regenerative, ‘‘Regenerative Oil Scrubbing of Volatile Organic Compounds from a Gas Stream in Hollow Fiber Membrane Devices,’’ Ind. Eng. Chem. Res., 38, 3426 (1999).