Available online at www.sciencedirect.com

SciVerse ScienceDirect

Energy Procedía 23 (2012) 23 - 32

Trondheim CCS Conference-6

Packing characterization: Mass Transfer Properties

Chao Wanga,b, Micah Perry0, Gary T.Rochellea, A. Frank Seibert0

aDepartment of Chemical Engineering, University of Texas at Austin, ¡University Station C0400, Austin, TX 78712, USA bProcess Science and Technology Center, Pickle Research Center, 10100 Burnet Road, Austin, TX 78758, USA

Abstract

Packing is widely used in post-combustion CO2 capture. This paper is focused on the measurement of three fundamental packing characteristics: effective gas-liquid contact area (ae), gas phase and liquid phase film mass transfer coefficient (kG) and (kL). Results show that contact area initially increases with liquid flow rate before it asymptotes to a value nearly equivalent to the nominal surface area. Initial attempts at constructing new mechanistic models have shown that existing kG and kL models can be improved. Both gas and liquid phase film mass transfer coefficients can be described as a power function of superficial gas or liquid velocity.

© 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of SINTEF Energi AS

Keywords: Post-combustion capture, structured packing, random packing, packing effective area, gas phase film mass transfer coefficient, liquid phase film mass transfer coefficient

1. Introduction

CO2-capture and storage (CCS) is a common term for the capture, transport and safe storage of carbon dioxide. Post-combustion technology is widely used because it can be added directly to existing power plants, minimizing capital costs. The majority of post-combustion technologies currently utilize a simple absorber/stripper configuration. Random and structured packings will be loaded into the absorbers, and they are the key to optimizing mass transfer efficiency of the post-combustion gas treating processes.

Packing is widely used in CO2 capture because of its relatively low pressure drop, good mass transfer efficiency, and ease of installation. The design of packed absorbers for carbon dioxide capture will require the reliable measurement and accurate prediction of the effective area, ae, as well as the gas and liquid phase film mass transfer coefficients, kG and kL. This paper is focused on the measurement of these

1876-6102 © 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of SINTEF Energi AS doi:10.1016/j.egypro.2012.06.037

important fundamental parameters and the construction of mechanistic design models capable of predicting these parameters for a wide variety of packings.

Nomenclature

ae effective area of packing, m2/m3

af fractional area of packing, ae/aP

ap specific area of packing, m2/m3

D diffusivity, m2/s

Gz Graetz number, ReLScL(5/H)

H Henry's constant, m3*bar/kmol

KG overall gas-side mass transfer coefficient, kmol/(m2*Pa*s)

kg' liquid-side mass transfer coefficient expressed in gas units, kmol/(m2*Pa*s)

kL physical liquid phase film mass transfer coefficient, kmol/(m2*Pa*s)

k2 2nd order reaction rate constant, kmol/(m2*Pa*s)

R ideal gas constant, (m3*Pa)/(kmol*K)

Re Reynolds number, upd/u

Sc Schmidt number, n/pD

Sh Sherwood number, kd/D

T absolute temperature, K

UG gas velocity, m/s

Ul liquid velocity, m/s

Yin/out mole fraction of solute at inlet/outlet

Z packing height, m

2. Experimental procedures

All experiments characterizing the mass transfer properties of the packings were conducted in the same column. The PVC column had an inner diameter of 0.42 m and a 3.05-m packed height. Operation was counter-current: ambient air entered below the packed bed and flowed upward through the bed of packing. The liquid was pumped in a closed loop and was distributed at the top of the column using a pressurized fractal distributor with 108 drip points per square meter. The experimental configuration is shown in Figure 1:

Distributor

Air Outlet

Packed Height-3m

Storage Tank

Fig. 1. Process flow diagram for the 420 mm ID packed column 2.1. Effective gas-liquid contact area measurement

The effective gas-liquid contact area of the packing is measured by the absorption of atmospheric carbon dioxide with a 0.1 gmol/L sodium hydroxide solution. According to the two film theory [1, 2], the overall mass transfer resistance is equivalent to the sum of the gas phase and liquid phase mass transfer resistance as defined below:

-= — + — (1)

KG kg kg

When CO2 partial pressures are low and hydroxide ions are present in relative excess, the reaction can be treated as pseudo-first-order. According to wetted wall column experiments and experimental ko measurements conducted at UT-Austin, gas phase mass transfer resistance accounts for less than 10% of overall mass transfer resistance. Thus, the gas phase mass transfer resistance is negligible, and the overall mass transfer coefficient, KG, is equivalent to the liquid phase mass transfer resistance. According to Tsai's research [3, 4], kg' can be expressed by the following equation:

Ah - [OH ]D

The diffusivity of CO2, DCO2,L, and Henry's constant, HCO2, can be calculated using physical property data collected by several instruments during the experiment. The effective area, ae, can be calculated:

uG 1nC^02^ )

_yC02out

ZKQGRT

uG lnC^02^ )

_yC02out

Zk ' RT

In a typical experiment, 0.76 m of 0.1 gmol/L NaOH solution is prepared in the storage tank. A Horiba infrared carbon dioxide analyzer is used to evaluate inlet and outlet CO2 concentrations. The initial NaOH concentration is measured by manual titration. Three gas flow rate (0.6, 1, 1.48 m/s) and seven superficial liquid velocities (0.17, 0.33, 0.67, 1, 1.3, 1.6 and 2 cm/s) are used for a typical gas-liquid contact area study. The inlet and outlet CO2 concentrations are measured sequentially after the system has reach steady state. With the inlet and outlet CO2 concentration known, the effective area can be calculated by equation (3). Effective area results are shown in Section 3.

2.2. Gas phase film mass transfer coefficient (kG)measurement

The gas phase film mass transfer coefficient is measured by the absorption of sulphur dioxide with a 0.1 gmol/L sodium hydroxide solution [5]. The reaction between SO2 and NaOH is an instantaneous reaction, making the liquid phase resistance negligible [6]. When the liquid phase resistance term is removed from equation (1), it is apparent that the overall resistance, KG, is equivalent to the gas phase resistance. The gas phase film mass transfer coefficient can be calculated by the following equation:

k =-ySo2out (4)

G ZRTa„

Tandems of independent Thermo Scientific Model 43i SO2 analyzers are used to measure the inlet and outlet SO2 concentrations. The inlet analyzer is calibrated from 0-100 ppm while the outlet analyzer is calibrated from 0-100 ppb. A wider range of gas flow rates are usually studied (1.96, 3.25, 4.87, 6.50, 8.12 m/s) as ko is primarily a function of gas flow rate rather than liquid flow rate. For each condition, the steady state inlet and outlet SO2 concentration are measured simultaneously. With the inlet and outlet concentrations known, ko can be calculated by equation (4). Effective gas phase film mass transfer coefficient results are shown in the section 3.

2.3. Liquid phase film mass transfer coefficient (ki)measurement

The liquid phase film mass transfer coefficient is measured via air-stripping of toluene from water [7, 8]. As this is a non-reactive system, the two film theory can be simplified to the following equation:

-= — +--(5)

KL kL HtolkG

The liquid phase mass transfer resistance is significantly larger than the gas phase resistance, making kG negligible. The overall mass transfer resistance is then a sole function of the liquid phase resistance, kL, and it can be calculated by equation (6):

Toluene-in

\ Toluene-out J

Two liters of toluene are initially added to 760 liters of water in the storage tank. Toluene is then metered continuously into the sump tank to maintain a constant toluene concentration in the aqueous feed. Run conditions are the same as those used in the effective area experiment. For each point, the gas rate is fixed and the liquid rate is increased. An inlet water sample is taken from the pump discharge, and an outlet water sample is taken from a bayonet sampler located immediately beneath the packed bed after steady state has been reached. A Hewlett Packed 5890A FID gas chromatograph is used for the analysis. An analytical method has been devised ensuring accurate quantitative results.

3. Results and discussion

3.1. Effective gas-liquid contact area results

Measurements have been completed on five packings. The packings are abbreviated in the following manner: Mellapak, MP; Flexipac, FP; Raschig SuperPak, RSP; Raschig SuperRing, RSR; and Pall Ring, PR. Characteristic data for each of the packings is listed in Table 1.

Table 1. Characteristic data for packings tested

Packing name

FP 1.6 Y HC

RSP 250

RSR#0.5

1" Plastic PR

Surface Area, m2/m3 Corrugation angle Channel length, m

Metal, Structured 205 60 0.02

Metal, Structured 295 45 0.017

Metal, Hybrid 250

N/A N/A

Metal random 250

N/A N/A

Plastic random 210

N/A N/A

Figure 2 shows the effective area results for MP2X. At low liquid rates, the effective area increases sharply with liquid flow rate and shows a mild dependence with liquid rate. The figure also shows that the effective area increases slightly with gas flow rate. The gas-liquid contact area results for other structured packings show similar trends. Random packings display a larger dependence on gas rate. All of the effective area results are presented in Figure 3.

Results are limited to one gas rate (uG=0.989m/s) in Figure 3. To eliminate the influence of the surface area of metal, normalized liquid flow rate (uL/aP) is plotted as the independent variable. Figure 3 shows that the effective area is a function of liquid flow rate. Effective area is not a function of gas flow rate for structured packings and is a weak function of gas flow rate for random packings. At the same run conditions, the magnitude of fractional area follows this trend: RSP 250>MP2X~FP1.6 Y HC>RSR#0.5>>1 inch Plastic Pall Ring. The plastic packing provides 20% less fractional area than the metal packings. This may be due to the increased hydrophobicity of polypropylene relative to stainless steel combined with the aqueous nature of the experiments. Figure 3 also compares the experimental data

with a widely used area model proposed by Tsai [9] for MP2X. The experimental results are within 20% deviation of the Tsai model for that specific packing.

TO TO 01

4-» (J TO

♦ uG=1.484m/s ■ uG=0.989m/s i uG=0.597m/s x uG=1.98m/s X uG=2.476m/s

Liquid Flow Rate/(m/s)

Fig. 2. Gas-Liquid contact area results for MP2X

5.0E-6 2.5E-5 4.5E-5 6.5E-5 8.5E-5 1.1E-4

Generalized Liquid Superficial Velocity uL/aP (m2/s)

Fig. 3. Gas-Liquid contact area results summary

Chao Wang et al. /Energy Procedía 23 (2012) 23 - 32 3.2. Gas phase film mass transfer coefficient (kG) results

♦ uL=0.0034m/s ■ uL=0.0067m/s A uL=0.0134m/s

♦ ■

uG/(m/s)

Fig. 4. Gas phase film mass transfer coefficient kG results for MP2X

Gas Superficial Velocity uG (m/s)

Fig. 5. Gas phase film mass transfer coefficient kG summary. Liquid Velocity = 0.67 cm3/cm2-s.

The effects of gas velocity on the gas phase film mass transfer coefficient (kG) are shown in Figure 4. The kG increases with the gas flow rate and is essentially independent of the liquid flow rate. The remainder of the packings display similar behavior as shown in Figure 5. The kG varied with gas velocity to the n-

power where n ranged from 0.7 to 1.0. This range is slightly greater than exponents reported in the literature [10, 11, 12, 13, 14].

3.3. Liquid phase film mass transfer coefficient (kL)results 8E-5

0.000 0.010 0.020

uL/(m/s)

Fig. 6. Liquid phase film mass transfer coefficient (kL) results for MP2X

Figure 6 shows the relationship between superficial liquid velocity and the liquid phase film mass transfer coefficient (kL) for MP2X. Contrary to kG, kL shows dependence on liquid velocity and essentially no dependence on gas velocity. The liquid phase film mass transfer coefficient (kL) data for the other packings followed a similar trend, and they are summarized in Figure 7.

It is interesting to note that the liquid phase film mass transfer coefficient for the 1"Plastic Pall Ring is larger than the metal packings at the same flow conditions. This may indicate that the rougher surface of the polypropylene packing is increasing the turbulence in the liquid film. The liquid phase film mass transfer coefficients measured in this work vary with the liquid velocity to the n power where n ranges from 0.6 to 0.8. This range is slightly larger than the range of 0.5 to 0.67 reported in the literature [10, 11, 12, 13, 14].

♦ uG=0.597m/s ■ uG=0.989m/s uG=1.484m/s A

1 ▲ ■ ▲

5.0E-6 1.0E-5 2.0E-5 4.0E-5 8.0E-5

Generalized Liquid Superficial Velocity uL/aP (m2/s)

Fig. 7. Liquid phase film mass transfer coefficient (kL) summary

4. Conclusions

Three fundamental mass transfer properties for packings have been measured and analyzed in this paper. They include effective gas-liquid contact area (a«), gas phase film mass transfer coefficient (kG), and liquid phase film mass transfer coefficient (kL). Five packings, including random and structured packing geometries, have been tested. Results show that the effective area (aj increases with liquid flow rate and is essentially independent of gas flow rate. This behaviour is observed with all packings. Experimental results for MP2X are within 20% deviation of the effective area predicted by the Tsai model.

Experiments to characterize the gas phase film mass transfer coefficient for each packing reveal that kG is a function of gas flow rate with no dependence on liquid rate. This dependence can be expressed as a power function, and the exponents regressed for the five packings studied range from 0.7 to 1. These exponents are slightly larger than values reported in the literature.

The liquid phase film mass transfer coefficient shows a large dependence on liquid flow rate and little dependence on gas flow rate. Like kG, the relationship between the liquid flow rate and the liquid phase mass transfer coefficient can be expressed as a power function. The exponents regressed for the five packings studied range from 0.6 to 0.8, which is slightly higher than the values reported in the literature.

Future work will include similar measurements of additional random and structured packings as well as the development of new mechanistic models for the prediction of gas and liquid phase film mass transfer coefficients.

Acknowledgements

Separations Research Program Process Science and Technology Center Luminant Carbon Management Program

References

[1] Danckwerts PV, Kennedy AM, Roberts D. Kinetics of CO2 absorption in alkaline solutions-II: Absorption in a packed column and tests of surface-renewal models. Chem EngSci 1963;18:63-72.

[2] Danckwerts PV. Gas-Liquid Reactions. New York: McGraw-Hill Book Company; 1970.

[3] Tsai R, Schultheiss P, et al. Influence of Surface Tension on Effective Packing Area. IndEng Che. Res 2008;47:1253-1260

[4] Tsai R. Mass Transfer Area of Structured Packing. The University of Texas at Austin. Ph.D Dissertation. 2010

[5] Sharma MM, Mehta VD. Effect of Diffusivity on gas-side mass transfer coefficient. Chemical Engineering Science 1966;66:361-365

[6] Yaici W, Laurent A. Determination of gas-side mass transfer coefficients in trickle-bed reactors in the presence of an aqueous or an organic liquid phase. International Chemical Engineering 1988;28(2):299-305

[7] Akita K, Yoshida F. Gas holdup and volumetric mass transfer coefficient in bubble columns. Effects of liquid properties. Industrial & Engineering Chemistry Process Design and Development 1973;12(1):76-80

[8] Linek V, Petericek P. Effective interfacial area and liquid side mass transfer coefficients in aborption columns packed with hydrophilised and untreated plastic packings. Chem Eng Res Des 1984;62:13-21

[9] Tsai R, Seibert F, et al. A dimensionless model for predicting the mass transfer area of structured packing. AIChE J 2011; 57(5): 1173-1184

[10] Bravo JL, Rocha JA, Fair JR. Mass transfer in Gauze Packings. Hydrocaborn Process 1985;64:91-95

[11] Rocha JA, Bravo JL, Fair JR. Distillation columns containing structured packings: a comprehensive model for their performance. 2. Mass-Transfer model. Ind. Eng. Chem. Res 1996;35:1660-1667

[12] Onda K Takeuchi H, Okumoto Y. Mass transfer coefficients between gas and liquid phases in packed columns. J Chem. Eng. Jpn 1968;1:56-62

[13] Bravo JL, Fair JR. Generalized correlation for mass transfer in packed distillation columns. Ind Eng Chem Process Des. Dev. 1982;21:162-170

[14] Billet R, Schultes M. Advantage in correlating packing column performance . Inst. Chem Eng Symp Ser. 1992;128:129-136