Mass transfer performance of structured packings in a CO2 absorption tower☆

Wei Yang ,Xiaodan Yu ,Jianguo Mi,Wanfu Wang ,Jian Chen *

1 State Key laboratory of Chemical Engineering,Tsinghua University,Beijing 100084,China

2 State Key Laboratory of Organic–Inorganic Composites,Beijing University of Chemical Technology,Beijing 100029,China

3 CNPC Research Institute of Safety&Environment Technology,Beijing 102206,China

Keywords:Separation Absorption Mass transfer Packed bed Structured packing Carbon dioxide

ABSTRACT This paper studies the mass transfer performance of structured packings in the absorption of CO2 from air with aqueous NaOH solution.The Eight structured packings tested are sheet metal ones with corrugations of different geometry parameters.Effective mass transfer area and overall gas phase mass transfer coefficient have been measured in an absorption column of 200 mm diameter under the conditions of gas F-factor in 0.38-1.52 Pa0.5 and aqueous NaOH solution concentration of 0.10-0.15 kmol·m?3.The effects of gas/liquid phase flow rates and packing geometry parameters are also investigated.The results show that the effective mass transfer area changes not only with packing geometry parameters and liquid load,but also with gas F-factor.A new effective mass transfer area correlation on the gas F-factor and the liquid load was proposed,which is found to fit experiment data very well.

1.Introduction

At present,global extreme weather occurs frequently.Various natural disasters threat gravely the human survival and development.All these are related to the climate change which is caused by greenhouse gas emission,especially CO2emission[1].Reducing CO2and fighting against climate change have already become global consensus.

There are three main ways to reduce CO2.The first is the control from the source.It can reduce CO2and even achieve zero emission when developing clean energy and renewable energy[2].The second is the control during the process.Improving the energy efficiency can reduce energy consumption and thus CO2[3],which has great potential in practical engineering applications.The third is the control at the end.Currently CO2capture and storage(CCS)is the most directive and most effective way at the end of process[4-6].

CCS is of wide concern,but high cost and high energy consumption are the main bottlenecks of CCS[7,8].Researchers have done much work,and most work is focused on solvents and process simulation[9-14].Some research works about packings used for CO2absorption has been done.Aroonwilas et al.[15]researched about different commercial packings.Alix et al.[16]proposed a new structured packing which generated a low pressure drop and a very high interfacial area.

Mass transfer performance of packings was often evaluated by using the systems including air-H2O-CO2[17],air-H2O-O2[18]and air-H2ONH3[19],where only mass transfer process proceeds and no reaction occurs.In 2000,the model for predicting structured packing mass transfer performance in CO2chemical absorption was proposed[14].Later,the effects of operating parameters on CO2absorption in columns with structured packings were obtained by using CO2-MEA system[15],and the effective mass transfer area of random packings(Raschig super-rings)was studied with CO2-NaOH system[20].

In comparison with random packings,structured packings are made of thin corrugated metal sheets arranged side by side with an opposing channel orientation[14],which offer excellent mass transfer performance with a lower pressure drop[21].Their specific area is up to 250-750 m2·m-3with a void fraction of 90%.There are many kinds of commercial structured packings such as Gempak,Mellapak and Optifow.

The main objective of this study is to shed light on selection or development of a new structured packing used for CO2chemical absorption with a high efficiency.In this study,the effective mass transfer area and overall gas phase mass transfer coefficient of sheet metal structured packings with corrugations of different packing geometry parameters were measured.All of the data were analyzed with respect to designing high efficient packing suitable for CO2chemical absorption.

2.Mathematical Model

According to two- film theory,the interface mass transfer rate is

The material conservation equation for the absorbed amount can be written as

Because of low CO2concentration in the air,and OH?in the liquid phase in large excess,there is no free CO2in liquid,i.e.,

Substituting Eq.(3)into Eq.(2)followed by integration results in

Though some authors(for example[22])calculated KGAeby

For a tower with height of 1.77 m which was much longer than that of this study,it is believed that Eq.(4)(an integral formula)is suitable for calculating KGAein this study.

The mass transfer rate from the gas film equals to thatinto the liquid film and it can be expressed as

with

The enhancement factor(E)for a fast pseudo first-order reaction is given by the surface renewable theory[23]:

Because the reaction between CO2and NaOH is very fast,?1.Thus,the approximate equality holds:

The reaction rate constant(k)was calculated from the relations[24]:

Because the concentration of NaOH is small(0.10-0.15 kmol·m?3)and the viscosity of the solution is close to that of water,the effect of solution viscosity on the diffusion coefficient for CO2in the solution(DA)can be ignored.So DAwas calculated from the relation[23]:

Henry's constant(HA)for CO2in aqueous electrolyte solution could be determined by[23]:

The value of h+,h?and hgaswere cited from Danckwerts[23].

The gas phase mass transfer coefficient(kG)for structured packing could be calculated by the relation[25]:

According to Eqs.(4)and(6),the effective mass transfer area(Ae)can be calculated by

3.Experimental Facility and Methods

The experimental facility and process are shown in Fig.1.The column was about 1.2 m tall with an inside diameter of 200 mm,in which the height of packed section is 0.8 m.Sheet corrugated structured packing was packed in the column to promote contact between the gas and liquid phases.Air entered at the bottom of the column and flowed upward,while NaOH solution was pumped to the top of the column and flowed downward.After absorption,NaOH solution returned to the tank for cycling while the air exhausted from the top.

Fig.1.Flow diagram of experiment.

During steady operation,the concentrations of CO2in the gas entering and leaving the column were measured respectively using an infrared gas analyzer with the precision of±2 ml·m?3,while the concentration of NaOH solution was measured by double-tracer technique.From these measurements,the values of Aeand KGAecan be obtained.With the operation conditions changed,the mass transfer performance of different packings can be studied.Table 1 lists the conditions for the measurement.(See Table 2.)

Table 1 Parameters for the measurement

Table 2 Contributions of anion,cation and gas to Henry's constant for CO2

In order to study the effect of corrugated angle(α)on the mass transfer,packings with α of 30°,37.5°and 45°were used in this study at the condition of small liquid/gas flow ratios usually used in the process of CO2absorption.

4.Results and Discussion

4.1.Measurement of Ae

Under the condition of normal pressure and temperature,the gas flow rate was maintained and the liquid flow rate was altered.By measuring the CO2concentration in the gas entering and leaving the column and NaOH concentration of the solvent,the effective mass transfer area could be measured and calculated.Then,with the liquid flow rate maintained,the gas flow rate was changed.So the effective mass transfer area under different gas and liquid flow rates could be measured and calculated.

Fig.3 showed that Aewas affected positively by gas and liquid flow rates.Wilson[26]also discovered the same results.Alix et al.[16]discovered that Aeof a new structured packing 4D-50%also increased with the increase of gas and liquid flow rates,because an increase in wetted packing surface facilitated the mass transfer.It is worth noting that when the flow rate of gas and liquid increased to a certain value,Aewould exceed the specific area of packing material.This is consistent with the fact that Aewas larger than the specific area of packings at a sufficiently high liquid flow rate[20].At the high flow rate,the instabilities in the liquid flow(like ripples or waves,detachment and fragmentation of the film into copious liquid showers)led to this phenomenon.

The pattern of packing arrangement impacts the performance of mass transfer.At the condition of the same gas and liquid flow rate,Aewas obtained with two arrangement ways of successive packings as illustrated in Fig.2.From Fig.4 Aeof packings with a 90°rotation was over 30%bigger than that with a 0°rotation.The result was similar to that in[15].Aroonwilas et al.discovered that packings with a 90°rotation provided a great and comparable efficiency.It was possible that a 90°rotation arrangement allowed the liquid to spread not only from side to side but also from front to back in the column.So the wetted area increased.

4.2.Correlations for Ae

Correlations for Aeby Onda et al.[27]and Henriques de Brito et al.[25]listed in Table 3 were tested to calculate Aein this experiment.For comparison the parameters in the two correlations were regressed according to the results for the packing with α =45°,β =90°,h=9 mm of this study.The calculated value and experimental value under different gas F-factors were shown in Figs.5-7 respectively.It showed that although the trends were the same,results calculated by Onda et al.and Henriques de Brito et al.could not predict Aecorrectly with different gas flow rates.The average relative error of new correlated Onda et al.correlation(19)was 18.0%while for the new correlated Henriques de Brito et al.correlation(21)was 18.2%.

Onda et al.[27]considered that Aedid not change with gas flow rate.Henriques de Brito et al.[25]discovered that Aeremained unaffected by Fsfor packings with less specific area.So in their correlations,there was not a gas flow rate.However,it was found in this study that Aeimproved with the increase of gas flow rate.In this paper,the correlation by Henriques de Brito et al.[25]was modified so that the effect of gas flow rate could be considered.The new correlation was in the following form:

According to experimental results,the modification was conducted by multiple linear regressions to identify new coefficients x1,x2and x3.The correlations for the packings with different geometry parameters were shown in Table 4.The comparison of experimental Aeand calculated Aeby Eqs.(19),(21)and(22)were shown in Figs.5-7.

For the packing with α =45°,β =90°,h=9 mm in Figs.5-7,the modified correlation Eq.(22)obtains the average relative error for Aeas 1.3%,compared with 18.0%of Eq.(19)and 18.2%of Eq.(21).So the modified correlations with the consideration of effects of gas flow rates showed better agreement.

4.3.The effect of geometry parameters on Ae

There are three main geometry parameters of sheet corrugated structured packings.They are corrugated angle(α),corrugated peak height(h)and addendum angle(β)respectively.This paper tried to change one parameter while fixing the other two parameters.Packings with different geometry parameters were tested so that the effect of α and β on Aecould be studied.

When β of 75°was fixed,an example for the results of α variation was shown in Fig.8.At the Fsof 0.38 Pa0.5when Lwwas below 41.4 m3·m-2·h packing with α of 45°exhibited the biggest Ae,while Lwwas above 41.4 m3·m-2·h packing with α of 30°performed best.At the Fsof 0.76 Pa0.5or 1.14 Pa0.5,when Lwwas below 15.9 m3·m-2·h packing with α of 37.5°exhibited the biggest Ae,while Lwwas above 15.9 m3·m-2·h packing with α of 30°performed best.

Two examples for the results ofβvariation were shown in Figs.9 and 10.At α of 45°,packing with β of 75°performed best except when Fs=0.76 Pa0.5and Lw=44.6 m3·m-2·h.However at α of 37.5°,when Fs=0.76 Pa0.5or 1.14 Pa0.5,packing with β of 90°performed best.In brief the effect of α and β was also affected by gas/liquid flow rates.Under most gas/liquid flow rates in this study,α =30°and β =75°were the best parameters.Aroonwilas et al.[15]also observed that corrugated angles had important impact on mass transfer performance.

Fig.2.Geometry and arrangement of sheet corrugated structured packings.(a)90°rotation;(b)0°rotation.

Fig.3.A e for packings with α =45°and β =90°vs.liquid load for different gas F-factors with a packing arrangement of 90°rotation.

Packings with different geometry parameters exhibited different effective mass transfer area.In this study,eight different packings were measured so that the packings with the biggest Aecould be verified.The results were shown in Figs.11 to 13.It was apparent that when the gas F-factor was small(Fs=0.38 Pa0.5),the packing with α =30°and β =75°had the biggest Ae.However,when the gas F-factor increased to 0.76 Pa0.5or 1.14 Pa0.5,the packing with α =37.5°and β =90°exhibited the biggest Ae.

Fig.4.A e of different packing arrangement patterns.

4.4.Discussions on KGAe

It is important to note that corrugated angle had an impact on the sensitivity of KGAeto liquid load variation.From Fig.14,it could be seen that the packings with corrugated angle of 30°was more sensitive to the change in liquid load than the packings with corrugated angle of 37.5°and 45°.The result was the same as in[15]that packings withsmaller corrugated angle were expected to a greater change in KGAewith increasing liquid load.The possible reason was that the lower corrugation angle had a tendency to allow liquid at a given flow rate to spread in a greater extent over the packing surface.

Table 3 Effective mass transfer area correlations

Fig.5.Comparison of different correlations to experimental results at the gas F-factor of 0.38 Pa0.5 for packings with α =45°and β =90°.

Fig.6.Comparison of different correlations to experimental results at the gas F-factor of 0.76 Pa0.5 for packings with α =45°and β =90°.

Fig.7.Comparison of different correlations to experimental results at the gas F-factor of 1.14 Pa0.5 for packings with α =45°and β =90°.

Table 4 Modified effective mass transfer area correlations for different packings with a packing arrangement of 90°rotation

Fig.8.Corrugated angle vs.A e at gas F-factor of 0.38 Pa0.5 and β =75°.

Fig.10.Addendum angle vs.A e at gas F-factor of 0.76 Pa0.5 and α =37.5°.

Fig.11.Comparison of A e for different packings at the gas F-factor of 0.38 Pa0.5.

With the gas flow rate fixed,and liquid flow rate was altered,the CO2concentration in the gas entering and leaving the column was obtained and substituted into Eq.(4)so that KGAecould be calculated.It was apparent that KGAeincreased with the increase of liquid and gas flow rates.As the liquid and gas flow increased,more CO2molecules were allowed to travel to the reaction zone,which would result in the higher mass transfer performance.Zeng et al.[28]observed that KGAeincreased with the increase of gas flow rate using CO2-NH3·H2O system.Aroonwilas and Tontiwachwuthikul[22]also used CO2-NaOH system and discovered that KGAeincreased with liquid flow rate increasing while remained unaffected by a gas flow rate.The liquid flow range and gas flow range of Aroonwilas and Tontiwachwuthikul[22]experiment were 4 to 14 m3·m?2·h?1and 0.31 to 0.66 Pa0.5,which were smaller than that of this study.This indicated that liquid film was controlling the resistance but when gas flow rate increased to a certain degree liquid film was struck into liquid drops so that Aeincreased thus improving KGAe.So in this study KGAeincreased upwards when Fsincreased from 0.38 to 1.14 Pa0.5.

Fig.12.Comparison of A e for different packings at the gas F-factor of 0.76 Pa0.5.

Fig.13.Comparison of A e for different packings at the gas F-factor of 1.14 Pa0.5.

Fig.14.Liquid load vs.K G A e for structured packings with different corrugated angles.

Fig.15.Proportion which gas phase mass transfer resistance accounts for in total mass transfer resistance vs.gas F-factor.

The total mass transfer resistance and gas phase mass transfer resistance can be calculated by KGand kGrespectively so that the ratio(KG/kG)which gas phase mass transfer resistance accounts for total mass transfer resistance can be obtained in Fig.15.It was evident that KG/kGwas less than 50%and decreased with the increase of gas flow rate.So the process of NaOH solution absorbing CO2could be defined as liquid- film control.This is consistent with the literature conclusion that the major mass transfer resistance lay in the liquid film[29].The same result was also obtained by the Zeng et al.[28].When the gas flow rate was very high,the gas phase mass transfer resistance became very small.In this case,there were two methods to improve mass transfer rate.First increasing the concentration of NaOH solution could improve the enhancement factor of reaction which caused the decrease of liquid phase mass transfer resistance.Secondly,Aecould be improved by increasing the liquid load.

5.Conclusions

The performance of sheet metal corrugated structured packings with different geometric parameters was studied for CO2absorption by aqueous NaOH solution.The results show that effective mass transfer area of packings increases along with the increase in gas and liquid flow rates,and even surpass the specific area of packings.The process of NaOH solution absorbing CO2belongs to the process of liquid- film control,and as the gas flow rate increases,the fraction gas phase mass transfer resistance in the total resistance decreases.Effective mass transfer area of packings is also related to geometric parameters,and usually when corrugated angle α equals to 30°and addendum angle β equals to 75°,the performance of mass transfer is excellent.A modified correlation considering the effects ofboth liquid and gas flow rates have been proposed with a good agreement.

As for CO2scrubbing,amine aqueous solutions are frequently used as absorbent.In the future aqueous amine solutions(MEA,MDEA,etc.)will be used to evaluate the performance of mass transfer with different packings.

Nomenclature

Aeeffective mass transfer area,m2·m?3

Apspecific area of packings,m2·m?3

CAIconcentration of A at gas-liquid interface,kmol·m?3

CALconcentration of A in the liquid subject

DAdiffusion coefficient of A in the liquid,m2·s?1

dhhydraulic diameter of packing,m

E chemical enhancement factor

Fsgas F-factor,Pa0.5

G total mole gas flow rate,kmol·m?2·s?1

Gwgas mass flow rate,kg·m?2·s?1

HAHenry's law constant for CO2in aqueous electrolyte solution,kPa·m3·kmol?1

HaHatta number

HwHenry's law constant for CO2-water,kPa·m3·kmol?1

h corrugated peak height of sheet corrugated structured packings,mm

hgascontributions of gas to Henry's law constant

hicontributions of i to Henry's law constant

h+contributions of positive ion to Henry's law constant

h?contributions of negative ion to Henry's law constant

Icionic strength,kmol·m?3

KGoverall gas phase mass transfer coefficient,kmol·m?2·s?1·kPa?1

kGgas phase mass transfer coefficient,kmol·m?2·s?1·kPa?1

L liquid mass flow rate,kg·m?2·s?1

Lwspraying density,m3·m?2·h?1

NAinterface mass transfer rate of A,kmol·m?2·s?1

P total system pressure,kPa

S gas-liquid contact area,m2

T temperature,K

uLflow rate of liquid,m·s?1

yAmole ratio of CO2in the gas phase

yemole fraction of component in the gas phase in equilibrium with liquid subject concentration

yinmole fraction of component in the gas phase entering the column

youtmole fraction of component in the gas phase leaving the column

Z height of packing layer,m

α corrugated angle for sheet corrugated structured packings,(°)

β addendumangle for sheet corrugated structured packings,(°)

ε mean relative deviation for effective mass transfer area,%

ρLdensity of liquid,kg·m?3

μGviscosity of gas,Pa·s

μLviscosity of liquid,Pa·s

σCcritical surface tension for packing material,mN·m?1

σLsurface tension of the liquid,mN·m?1

Subscripts

A CO2

e vapor liquid equilibrium

G gas phase

i composition of the fluid

in gas phase entering the column

L liquid phase