Gas–liquid flow mass transfer in a T-shape microreactor stimulated with 1.7 MHz ultrasound waves

Mona Akbari,Masoud Rahimi*,Mahboubeh Faryadi

CFD Research Center,Chemical Engineering Department,Razi University,Kermanshah,Iran

1.Introduction

One of the most important operations in many industrial processes is gas–liquid contacting processes such as gas absorption,which is especially used for environmental protection fields[1–9].There are several conventional gas–liquid contactors used for these processes,including spray column,bubble column and packed tower.The main problems in these operations are limitations on mass transfer,low efficiency and high operational costs.Therefore, finding a convenient method to enhance mass transfer and solve these problems is quite necessary[10–14].In recent years,employing microreactors,as effective gas–liquid contactors,is the subject of many researches[1,15–18].Microreactors with micrometer dimensions and ultra-high surface to volume ratio can increase heat and mass transfer significantly compared with conventional contactors and reactors[1,15,19–21].In addition,gas–liquid phases mass transfer rates are greatly increased due to very large gas–liquid interfacial areas up to 20000 m2·m?3[22–24].On the other hand,the miniaturization of chemical reactors has many other bene fits including minimal environmental hazards and increase safety due to the smaller volume besides reduction of process costs[25,26].

Some researchers focused on employing the microtechnology with greater potential performance in many chemical processes such as gas–liquid reactions,liquid–liquid reactions and gas–liquid–solid reactions[27–36].However,for suitable designs of such microreactors,the key issues in them related to the characteristics of hydrodynamic and reaction process.Up to now,several researches have been conducted for a better understanding of hydrodynamics, flow pattern regime,heat and mass transfer rate for gas–liquid and liquid–liquid two-phase flows in microreactors.An experimental investigation on characterization of gas–liquid–liquid in T-shape microreactors was undertaken by Rajeshet al.[37].In particular,they studied the effects of distributors and addition of surfactant to understand controlled generation of various flow regimes in microchannels.In series of studies done by Yueet al.[38–40],gas–liquid flow and mass transfer characteristics in a rectangular microchannel for CO2absorption were investigated.Hesselet al.[41]studied the CO2absorption into NaOH solution in microreactors and compared their efficiency with a packed column.Their results showed that the micro-devices give a yield with a larger order of magnitude than the conventional packed beds.

On the other hand,the application of ultrasound wave irradiation technique is a well-known approach for improving a wide variety of processes.Ultrasonic wave propagation through liquid causes cavitation bubbles,which produce shock wave and micro-jetduring their collapse.Moreover,ultrasound energy dissipation in the liquid leads to the formation of a rather intense macroscopic liquid flow called the acoustic streaming.Ultrasonic waves are divided into two groups,high frequency and low frequency ultrasound.Waves with a frequency of 20 to 100 kHz are de fined as low frequency ultrasound.These waves need high power to transmit through the environment.At low frequency ultrasound,strong acoustic energy can be generated to induce cavitation in liquids.This has considerable mechanical and chemical effects on various chemical processes.Waves with frequencies in the range of 100 kHz–1 MHz called high frequency ultrasound.High frequency waves are used commonly in the medical industry and chemical analysis.High frequency ultrasound has weaker mechanical and chemical effects compared with low frequency one and consequently uses lower energy.

High frequency ultrasound,in the range of MHz,induces convective flows and micro-streams(continuous flow ofgenerated bubble induced by ultrasounic wave),simultaneously.Waves in this range are capable to produce cavitation,which can improve micromixing.The implosive collapse of microbubbles,result in a variety of mechanical effects such as micro-streaming and micro-jet that have ability to increase the mass transfer rate and causing turbulence-jet in a fluid.Therefore,it is possible to reach a more efficient macro and micromixing by using ultrasound waves[42,43].

An interesting study was carried out by Chenet al.[44]on the effect of low frequency ultrasound(20 kHz)on the mass transfer and fluid flow hydrodynamic.They found that the overall mass-transfer coefficient was improved by 3–20 times under ultrasonication.Moreover,they demonstrated that ultrasound oscillation disturbs the bubble formation process and changes the initialbubble length and pressure drop.

Despite of interesting studies on low frequency ultrasound in the microchannel,high frequency ultrasound do not investigate properly[44–47].Both low frequency and high frequency ultrasonic make bubbles in sonicated systems.Low frequency waves cause more strong bubble implosions and thereby increase dominant physical effects and mixing performance.However,higher frequency waves create more smaller bubbles which lead to the generation of acoustic streams in the bulk of liquid with mixing effect[48].In addition,these waves consume lower power compared with low frequency ones.

In our previous work[49],micromixing efficiency of homogenous liquid flowswere investigated in differenttypes ofmicrochannels irradiated with high frequency ultrasound.The results revealed that,the T-type microreactor in the presence of ultrasound wave irradiation could establish higher micromixing efficiency in comparison with plain microreactor and segregation index,improved up to 18%–36%.Based on this previous research,it can be said that development of the combined microreactor with high frequency ultrasound waves of 1.7 MHz can be effective in increasing micromixing and mass transfer rate between the liquid–liquid and gas–liquid interface.As already mentioned,many researches on gas–liquid phases in plain microreactors have been done.The effectof flow pattern and various geometries of microchannel on the mass transfer rate of gas–liquid two phase flows were investigated.However,there are limited studies on application of ultrasound waves for improvement of mass transfer efficiency and its effect on created flow patterns in microreactors.Therefore,in the present work,it has been tried to investigate the effect of high frequency ultrasound wave irradiation in the range of MHz on a micro-scale gas–liquid contactor.This work introduces a new technique for enhancing the mass transfer rate between gas–liquid phases in microreactors by employing high frequency ultrasound.Hence,the microreactor was put in an ultrasonic container equipped with high frequency(1.7 MHz)piezoelectric transducer.Mass transfer efficiency and mass transfer coefficient(kLa)of presenting microreactor in the presence and absence of ultrasound wave irradiation were evaluated using both chemicaland physicalCO2absorption.In these experimentalconditions,the in fluences of different parameters on two-phase flow pattern,pressure drop,mass transfer coefficient and interfacial area were investigated.In addition,a comparison study was carried out between the obtained results of both modes(sonicated and plain microreactor).

2.Apparatus and Experimental Procedures

Two T-shaped glass microreactors were used in the experiments.The externaland internaldiameters ofchannels were 1 mm,and 800μm,respectively.The main channel lengths were considered 60 and 120 mm.Fig.1 shows a schematic view of the experimental setup and the snapshot of the container equipped with ultrasonic transducer.As shown in this figure,pure CO2from a gas cylinder was driven into A channelviaa pressure regulator and liquid solution was introduced to B channel from a liquid tank by a peristaltic pump(Qis?DSP100).A gas flow controllerwas used to ensure a constantgas flow rate into the microreactor.The pressure drop was measured using three pressure transducers(BD sensor,DMP 343,Germany)installed at the inlets and the end of the microreactor.The microreactor was immersed in a container,equipped with an ultrasound wave transducer,to evaluate the effectofultrasound waves on the mixing and mass transfer rate performance inside the microreactor.A high frequency piezoelectric transducer(1.7 MHz,model ANN-2517GRL,Annon Piezo Technology Co.Ltd.,China)with a diameter of 1.5 cm was installed at the bottom of the container.In order to use the maximum effect of ultrasound wave,the piezoelecteric transducer was placed under the contact point and mixing channel.In order to analyze gas–liquid flow pattern in the main channel and to evaluate the effect of ultrasonic wave on it,gas–liquid flow regimes were recorded with the aid of an electronic microscope(640× 480 pixels,magnification 10×–300×).The images were received by computer and used for later analysis.After fluid contacting in the main channel,two-phase mixture flowed downward into a phase separator and unabsorbed CO2was vented into the air.

Water and NaOH solution was selected as the liquid solution to investigate the physical and chemical mass transfer coefficient.Warder Method,the standard technique of titration liquid samples with HCl solution,was used to determine the amount of physically and chemically absorbed CO2.Phenolphthalein and methyl orange were chosen as the indicators for the firstand second endpoints,respectively.Atthe first titration point,all the hydroxide and part of the carbonate reacted:

Based on the two-titrant volumes,the carbonate concentration in the liquid sample collected in a separator was calculated.All experiments were conducted at(25 ±2)°C under atmospheric pressure.

3.Results and Discussion

Experiments were carried out in two approaches,with and without activation of the piezoelectric transducer.The in fluence of both modes on the various operating conditions was studied.

3.1.Two-phase flow patterns

In order to investigate two-phase flow pattern and the effect of ultrasound waves on it,water and CO2gas were selected.Moreover,in order to show clearer images of the flow pattern the water was colorized with pink color.A very little amount of pink ink was added to water to be sure that its effect on the water properties is negligible.By examining differentgas and liquid flow rates,typical flow patterns,generally reported in the existing literature,such as bubbly flow,slug flow,slug–annular flow,annular flow and churn flow were identified.Fig.2 shows some representative images at five types of flow pattern captured at the middle section of the microreactor for both modes.

Fig.1.Schematic diagram of the experimental setup and the snapshot of the container equipped with ultrasonic transducer:(1)CO2 cylinder,(2)pressure sensor,(3)microreactor,(4)piezoelectric transducer,(5)gas flow meter,(6)peristaltic pump,(7)phase separator,(8)electronic microscope,and(9)PC computer.

Fig.2.Representative photographs of CO2–water flow pattern in the microchannel.

Turning to Fig.2,distinct and small bubbles with diameters smaller than the channel size that dispersed in the continuous liquid phase characterize the bubbly flow(Fig.2(a)).In this flow pattern,due to surface tension force,the bubble almost seems spherical.By increasing the bubble diameter,the effectof surface tension decreases and the bubbles lose their spherical shapes.Slug flow is determined by elongated Taylor bubble with axial larger than the channel diameter(Fig.2(b)).As illustrated in this figure,the gasslugs were separated by the liquid segments in axial direction.With increasingUGand decreasingUL,gas slugs became longerwhile shorterliquid slugswasestablished.With the further increasing of gas flow,Taylor elongated bubbles eventually join and develop ring flow or slug-annular flow(Fig.2(c)).This flow pattern is formed by long gas segments that are distinguished by small liquid waves that are made due to the interfacial instability waves arising from the viscous shear stress.In the higher liquid flow rate,churn flow occurs,in which the bubbles are unstable at their trailing ends.Finally,in the higherUG,gas stream continuously flows in the center of channel and liquid flows as a thin layer on the wall(Fig.2(d)),named annular flow pattern.On the other hand,the ultrasonic wave can cause changes in the two-phase flow pattern that are also depicted in Fig.2.Based on this figure,it can be concluded that there is a regular stream in the plain microreactor while the stream behavior is more irregular in the sonicated layout.The presence of ultrasonic wave in the contact point makes irregular contact of two phases and changes the shape,the bubble size and the length of liquid and gas slugs.This can be explained by the fact that the performance ofsonicated microreactor on flow pattern can be due to stronger induced micro-streams and micro-jets by the collapse of microbubbles,generated by ultrasound wave propagation[50–52].Thereby,sonication increases the local turbulences inside the fluid.According to this figure,the effect of ultrasound wave is more significantatlower super ficialliquid and gas velocity such as bubbly and slug flow patterns.This might indicate that at a lower overall velocity,the ultrasound wave has a more significanteffect on micromixing with increasing localturbulence intensity.Howeverthe effect of ultrasound irradiation in the other flow patterns is lower than bubbly and slug flow.However,it still causes irregularities in the flow patterns.

Fig.3 reveals the developed flow pattern map for CO2–water flow.The flow transition lines predicted by Triplettet al.[53]for a 1.097 mm diameter circular microchannel are used for comparison with the obtained experimental data in this work.As shown in this figure,the production lines are generally in good agreement with the experimental observation's overall flow transitions.In addition,the results indicate that depending on the amount ofULandUG,various flow patterns are generated.At relatively lowUL,the slug flow,which includes the most data in the map,dominates at the low-to-moderateUG.With increasingUG,slug bubbles are leading to be unstable slug and in the higherUGtransition occurs from slug to slug–annular.In the higherULbubble and churn flow is formed at lowUGand moderate to highUL,respectively.Finally, flow pattern forms annular flow at higherUG.Moreover,by comparison of two modes,with and without activation piezoelectric transducer in microreactor,one can find that the ultrasound wave does not have a significant effect on changing of the flow pattern and transition lines.As farashigh frequency ultrasound has low power,it needs enough residence time to in fluence on flow regime.At higher flow rates residence time decreased.On the other hand,ultrasonic waves make local turbulence and irregularities in the microchannel.Athigher flow rate of liquid and gas flows,turbulence increased in the microchannel and the effect of ultrasound decreased compared with the effect of flow rate.This method is more effective to increase the local turbulence and internal circulation,and to change the size and shape of gas bubble and liquid slug.

On the other hand,with a comparison between high frequency waves in this paper and low frequency ultrasound waves achieved by Chenet al.[44],it can be concluded that the effect of low frequency ultrasonic is obvious by bubble oscillations.High frequency ultrasonic makes very small bubbles that make oscillation in the gas–liquid interface,so it cannot be obvious in Fig.2.On the other hand,piezoelectric transducer was placed below the inlet junction of microreactor,therefore the effect of high frequency ultrasonic was determined by the size of slugs,bubbles and the distance between them.

3.2.Pressure drop

Pressure drop,due to the direct relationship with energy consumption,is known as one of the most important parameters in the study of the microreactors.In order to ensure pressure results,pressure drop was measured three times foreachUGandUL.Itwas shown thatthe uncertainty of pressure is more in higher flow rates.The maximum uncertainty of pressure drop yields a measurement error of 5%.

Fig.4 illustrates the measured two-phase pressure drop results as a function ofULandUGfor CO2–water flow in the microreactor in the presence and without ultrasound wave irradiation for 12 cm microreactor length.Turning to this figure,the maximum difference between experimental data was selected for error bars.Therefore,the difference equal to 5%is a maximum possible error and experimental data are reliable.However,a few data points of pressure drop in Fig.4 differ significantly beyond the experimental error bars,also beyond the 5%level of measurement error.It seems that some unknown factor causes the abnormal measurements.From this figure,one can be clearly seen that pressure drop increases linearly with the increase inUGat the sameUL.The slopes of these lines obviously increased with the increase ofULvarying from 0.07 to 1 m·s?1.Furthermore,change inULhas a greater impact on pressure drop thanUG.The experimentalresults indicate that sonication has caused a very low difference in pressure drop across the microreactor.

Fig.3.Comparison between the observed flow pattern data for CO2–water flow in the microchannel with the transitional lines of Triplett et al.[29].

Fig.4.Measured two-phase pressure drop data as a function of U L and U G for CO2–water lf ow in 12 cm microreactor length for both modes.

In order to de fine how much ultrasonic power was dissipated in the microreactor,the calorimetric study was performed.Actual energy dissipated(Pdiss.)was calculated as follows[54,55]:where,mandCpdenote mass and specific heat capacity of water,respectively.tdenotes time and dT/dttime is the rate of water temperature increase.The transducer consumes the electrical power around 9.5 W.Based on our previous study[49],the amount of actual energy dissipated(Pdiss.)in the bulk of liquid(in the container)and in the microreactor were obtained 8.526 W and 4.45 × 10–4W,respectively.Therefore,ultrasound waves seem to be less effective on pressure drop in a broad domain of gas/liquid flow rates and flow pattern.However,ultrasound waves have a significant effect on the mass transfer efficiency that its further explanation will be presented in the consequent sections.

3.3.Mass transfer characteristics

3.3.1.Physical absorption

Experiments forphysicalabsorption were carried outto examine the mass transfer of the pure CO2into deionized water in the studied microreactor.Since the main part of mass transfer resistance is related to the liquid phase,the mass transfer resistance in the gas phase was considered negligible.Therefore,liquid side volumetric mass transfer coefficientkLawas obtained based on the following mass balance:

whereCinandCoutare molarconcentrations ofdissolved CO2in the inlet and outlet of the microreactor,respectively.Moreover,Vis the volume of the main channel andQLis the liquid flow rate.In some studies,the output was connected to a separator by a tube.So,another microchannel without the length of out channel was used and results were referred to second microchannel.In this study,the second microchannel was not needed and output directly entered to the separator.As the rate of CO2absorption is quite high,the equilibrium at the interface can be described by Henry's law[56]:

whereHandPCO2are the equilibrium solubility of CO2in the liquid phase and the partial pressure of CO2in the gas phase,respectively.The gas solubility into an electrolyte solution is in fluenced by various parameters as follows:

whereCiis ion concentration andhiis a parameter related to each ion in the solution.Moreover,hGdenotes the absorbed gas in the liquid phase.The values ofhiandhGwere presented by Schumpe[57].

Fig.5.Effect of super ficial gas and liquid velocities on liquid side volumetric mass transfer coefficient in the microreactor.

Since the two-phase mixture in the main channeldirectly entered to the separator,the end effect was considered negligible.Fig.5 describes the measuredkLain the microreactor as a function ofULandUGwith and without ultrasound wave irradiation.The observed flow patterns in the considered range are slug and slug–annular flow patterns.The presented results for two different lengths of microreactor in Fig.5(a)and(b)indicate thatkLavalue increases significantly with the increase inULandUG.However,both super ficial liquid and gas velocities increase the value ofkLa,but the effect of these parameters on the absorbed concentration of CO2is different.By increasing the value ofULat a fixedUG,the residence time and contact time between two phases decrease.Increase inkLais mainly attributed to the fact that the increase ofULcan increase the local turbulence and subsequently cause thinnerthickness ofthe liquid boundary layer,which results in reduction of mass transfer resistance between gas–liquid phases.This can cause an increase in the liquid side mass transfer coefficient.On the other hand,it can also see that with the increase inUGat a fixed value ofUL,kLawas increased.This can be due to the enhancement of gas holdup and turbulence of the microreactor,which increases contact between two phases and thereby,the rate of mass transfer.From the obtained results,it seems thatUGchanges have a significant effect onkLavalue compared with changes inUL.For example,with the increase in theUGfrom 0.17 to 0.33 m·s?1in the fixed value of 0.17 m·s?1forUL,kLaincreases about three times.On the other hand,with change inULfrom 0.17 to 0.33 m·s?1inUGof 0.17 m·s?1caused an increase of almost 1.5 times in liquid side mass transfer coefficient.

As depicted in Fig.5,with an increase in the length of microreactor from 6 to 12 cm at a fixedUGandUL,thekLadecreased.However,with an increase in microreactor length,contact time of two phase increases,while the value ofkLahas decreased.By taking more distance from the con fluence zone,the concentration of CO2solved in the liquid phase increases.So the decline ofdifference betweenCandC*decreases mass transfer rate andkLaand consequently averagekLa.

According to the results it can be concluded that ultrasound wave propagation in the microreactor has a significant effect on thekLavalues.The obtainedkLain the layout with this approach for microreactor withUL=0.17 m·s?1was increased up to 34%and 28%for length of 6 and 12 cm,respectively.The collapse of microbubbles,generated by the high frequency wave propagation,leads to induce strong micro streams and more local turbulence in the microreactor.In addition,ultrasound wave assists to better internal circulation within slugs[50].In addition,ultrasound wave may disrupt the interface and guide jets of one phase into the other.

These phenomena can effectively increase the solution mixing and more contact of between CO2gas and water.These can increase the mass transfercoefficientand improvementofabsorption ofCO2in water.

3.3.2.Chemical absorption

3.3.2.1.Mass transfer coefficient determination.In this paper,NaOH solution was chosen to study the mass transfer coefficient by chemical absorption.Therefore,in order to find thek*La,mass transfer parameters should be examined in the microreactors.The global reaction that happens in the microreactor is[58]:

Based on the achieved results by Robert and Danckwerts[59],the apparent first-order rate constantfor this reaction,k1,app,is 0.86 s?1.According to absorption values presented in Fig.5,it can be expected that the following inequality should be ful filled for CO2absorption into buffer solution and physicalabsorption underthe same ranges ofULandUG:

It indicates that the present microchannel reaction rate is much smaller than mass transfer rate of CO2into solution,so absorption process in buffer solution can be assumed as a physical one.Therefore,k*Lacan be derived askLa.

Fig.6 demonstrates the chemical volumetric mass transfer coefficient in liquid phase as a function ofUGandULat a constant initial concentration of NaOH solution.The results are presented for both modes,sonicated and plain microreactor,in different lengths of microreactor.Changes in chemicalk*Laat variousUL,UGand ultrasound wave irradiation are similar to physicalkLachanges that were mentioned above.In addition,the results for this method indicated that sonication through the microreactor has a significant impact on thek*Lain comparison with the plain microreactor atvarious examined operationalconditions.The collapse of cavitation microbubbles at the interface of two phases makes more irregularities as wellas more efficientmixing,which causes higher mass transfer rate.Therefore,ultrasound can promote chemical reaction by its mechanical effect.The measuredk*Lawith this approach forthe microreactorlength of6 and 12 cm inULvalue of0.17 m·s?1was increased up to 31%and 24%,respectively.However,with a comparison betweenkLaandk*Lavalues,shown in Figs.5 and 6,it can be obviously seen that the obtained mass transfer coefficient by chemical absorption has lower value than thatofphysicalone.This can be due to the factthat the diffusivity of CO2in water is more than electrolyte solutions,such as NaOH solution[60].

In Fig.7 the values of thek*Laat variousUGfor two initial NaOH concentrations are shown.The in fluence of NaOH concentration onk*Lais illustrated in this figure.As depicted in this figure,for both modes,an increase in the concentration ofNaOHcaused a decrease ink*Lavalue.This can be because of lower diffusion coefficient in electrolyte solution with higher concentration[60].The amount of absorbed CO2increased with the increase in solution concentration,but it decreased in comparison with the initial solution concentration.Therefore,according to Eq.(5),the ratio of concentration differences is reduced during the microreactor and leads to a reduction ink*La.

Fig.6.Effect of super ficial gas and liquid velocities on liquid side volumetric mass transfer(C NaOH=0.5 mol·L?1).

Moreover,an increase in super ficial gas velocity caused an increase in gas holdup and flow turbulence intensity.In addition,the gas–liquid mass-transfer resistance decreases with the increasing ofUGvalue,resulting in increase of CO2mass transfer and CO2concentration in the liquid phase.On the other hand,thek*Lavalue of sonicated microreactor is greater than that of the plain microreactor at various conditions.The enhancement of mixing and local turbulence by ultrasound waves increases the CO2solubility and decreases the gas–liquid mass transferresistance.Therefore,higherk*Lawas achieved in the sonicated microreactor.

For better expression of the effect of ultrasound wave on improving mass transfer coefficient in the microreactor,the relative mass transfer coefficient enhancement(E)is de fined as follows:

Fig.7.The values of the k*L a at various super ficial gas velocities for two initial NaOH concentrations.

where(kLa)USis the value of mass transfer coefficient in the sonicated microreactor and(kLa)plainis that of the plain one.This dimensionless parameter has been used in order to clarify the contribution of the ultrasound effect onkLaat various operating conditions.Chemical method was considered for measuring ofk*Lavalue.Fig.8 illustrates the values ofEfor variousULandUGat a fixed length of microreactor.The results indicate that with an increase inULin variousUG,the relative mass transfer coefficient enhancement decreases almost significantly.As shown in this figure,with reduction inUL,the effect of ultrasound irradiation on obtaining value ofEis increased and the ultrasound wave at lowerULhas a more significant effect on obtainingk*Laby chemical method.The reason can be explained by the fact that at higher velocity,the role of macromixing and convective flow circulation is more significant than extra mixing and local turbulence induced by ultrasound waves.In addition,the weakening effect of ultrasound under higher flow rates can also be attributed to the shorter irradiation time.

The effect of ultrasound wave on the value ofEin various lengths of microreactor andUGwas also investigated.The obtained results for differentUGand microreactor length in a constantULare listed in Table 1.The tabulated values show that with the increase in the microreactor length,the contacttime increases while the driving force(concentration difference)decreases along with the microreactor length.It is evident that with the increase in the microreactor length,the relative mass transfer coefficient enhancement decreased.This may be explained by a fact that with an increase inUGat a fixed microreactor length,the local turbulence and concentration of CO2in the two-phase interface enhanced,while the effect of ultrasound waves reduces.This leads to the reduction of theEvalue.In addition,the ultrasound wave effect is more significantatlower velocities.This mightindicate thatatlowervelocities the ultrasound waves,which caused extra local turbulence during the microreactor,have a more significant effect on the increasing of mass transfer coefficient.Consequently,theEvalue increases at lower velocity and shorter microreactor length.

3.3.2.2.Interfacialareadetermination.There are two methods to measure the interfacial area in a microreactor:

(1)Danckwerts method

(2)Calculation directly from the recorded image

Fig.8.Effectofsuper ficialgas and liquid velocities on the relative mass transfer coefficient enhancement(L=12 cm,C NaOH=0.5 mol·L?1).

In the sonicated microreactor,ultrasound wave makes fluctuations in the shape and the size of slugs and increases the internal circulation in the slugs.Hence,the interfacial area cannot be calculated precisely from the recorded image and Danckwerts method is more suitable.

The chemical method developed by Danckwerts[58]was used to measure the two-phase interfacial area in the microreactor.Thismethod is based on the absorption of CO2into a 1 mol·s?1NaOH solution.The reaction considered between CO2and NaOH solution can be expressed as follows:

Table 1Effect of super ficial gas velocities and length of main channel on E

The rate of reaction(13)is significantly higher than that of reaction(12)because of ionic transfer.Therefore,the reaction(12)that of second order is rate controlling.A general correlation of this reaction,kOH?,was given by Pohorecki and Moniuk[61]:

Fig.9.Effect of super ficial gas and liquid velocities on interfacial area in the microreactor for both modes(L=12 cm,C NaOH=1 mol·L?1).

Reaction(13)can be considered as a fastpseudo- first-orderreaction under the following conditions:

In these conditions,the gas side mass transfer resistance is negligible and the concentration of OH?is the same in the liquid side.

Sharma and Danckwerts[62]proposed a specific rate of absorption per unitarea for a second-order reaction satisfying the above conditions as:

Table 2Calculated inequalities(17)and(18)in different conditions

The diffusivities of CO2and of NaOH in the liquid phase are calculated using Eqs.(22)and(23)as follows[56]:

Inequalities(17)and(18)in various operational conditions are listed in Table 2.As shown in this table,both of inequalities are true for all conditions,so consideration of fast pseudo- first-order reaction for reaction(8)is correct.

Fig.9 reveals the obtained interfacial area in layouts with and without ultrasound wave irradiation at different value ofUG.Since literatures indicated that the value ofULhas notsignificanteffect[23,34 and 47],UL=0.07 m·s?1was selected for evaluation of interfacial area values.As illustrated in this figure,with increasingUGvalue,the interfacial area in both modes increases.In addition,the results demonstrate that the ultrasound wave increases the gas–liquid interfacial area about 30%at variousUG.Moreover,the effect of sonication is more significant at lowerUG,which can be due to irregularities made by ultrasonic on the flow pattern.

In order to ensure about calculated interfacial data,in Table 3 the data in this work are compared with two other studies.As shown in this table,interfacial area in three systems is in the same range and the difference between them is due to channel diameter,length of main channel,the shape of cross-section,super ficial gas/liquid velocity and other operational parameters.

To further investigate of interfacial area,the ratio of the measured interfacial area to the geometrical wall surface(RA)was calculated.Table 4 demonstrates this parameter atUL=0.07 m·s?1.As depicted in this table,RAwithUSbecomes more than 1 at the maximum velocity of the gas.However,the layout without ultrasound irradiation could not reach to this value in the whole range of examined gas velocities.

Table 3The comparison between the interfacial areas in different microreactors

Table 4Calculated RA at various gas velocities

4.Conclusions

In this study,a novelmethod presented based on application ofultrasound wave for improving the mass transfer rate in a glass microreactor with a diameter of 800 μm.For this purpose,absorption of CO2in water and NaOH solution(0.5 and 1 mol·L?1)was investigated.Experiments for both modes were carried out,without and with activation of 1.7 MHz ultrasound piezoelectric transducer.In the experiments,the super ficial velocities were chosen in the range of 0.17–15 m·s?1for gas flow and 0.07–1 m·s?1for liquid flow.Water and CO2were entered to the microreactor and various two-phase flow patterns were observed using an electronic microscope.Experimentaldata showed a good agreement with the transition lines suggested by Triplettet al.[42].By investigating the effect of sonication,it was found that ultrasound wave did not change the flow patterns,but it disturbs the bubble formation process,changes the liquid slug lengths and size of bubbles in bubble and slug flow regimes.However,it does not have any significant effect on other flow regimes.Pressure drops were measured in both cases,with and without ultrasound wave.Results revealed that the pressure drop increased linearly with an increase in the amount ofUGat the sameULand its changes are more dependent onULthanUG.In addition,it was found that sonication changes the pressure drop slightly,almost less than 10%,which can be neglected.In addition,the effect of ultrasound wave on gas–liquid flow and mass transfercharacteristics was evaluated.For investigation of the effect of ultrasound wave irradiation on mass transferrate,liquid side volumetric mass transfercoefficientsforphysical and chemical absorption process were calculated.The results revealed that with increasingUGandUL,kLaincreased significantly.

Besides,with the enhancement in the microreactor length,driving force of mass transfer decreased and therebykLadeclined.The mass transfer coefficient for microreactor with the length of 6 cm is more about 30%compared with other microreactor.On the other hand,the value ofkLain physical absorption was higher than that of chemical absorption.Based on these results,for microreactor with the length of 12 cm,mass transfer coefficients in physical absorption are higher about 10%and 100%compared with mass transfer coefficients in chemical ones in NaOH concentrations equal to 0.5 mol·L?1and 1 mol·L?1,respectively.This was explained by lower diffusion coefficient in the electrolyte solution compared with pure water.The results indicated that for both modes increase in concentration of NaOH cause a decrease ink*Lavalue.

In another part of this study,gas–liquid interfacial area was measured by the Danckwerts pseudo- first order reaction method for both layouts,the sonicated and plain microreactor.The results showed that the interfacial area in both modes increased with an increase in theUGvalue.In addition,the results revealed that the ultrasound wave could increase the interfacial area up to 30%.For better understanding of the ultrasound irradiation effect on mass transfer,a dimension-less parameter,relative mass transfer coefficient enhancement(E),was presented.Ithas been illustrated thatthe sonication enhances the mass transfer coefficient up to 34%.The results showed that the high frequency ultrasound can generate acoustic streams,which have the ability to increase mixing and local turbulence inside the microreactor.

Nomenclature

ainterfacial area,m2·m?3

Cconcentration,mol·m?3

C* physical solubility of CO2in the liquid,mol·m?3

Cinconcentrations of dissolved CO2in the inlet liquid phase,mol·m?3

Coutconcentrations of dissolved CO2in the outlet liquid phase,mol·m?3

Cpspecific heat capacity,J/(kg·°C)?1

DCO2diffusivity of CO2in the liquid,m2·s?1

Erelative mass transfer coefficient enhancement,dimensionless

Hequilibrium solubility of CO2in the liquid phase,

Pa·(mol·m?3)?1

HaHatta number,dimensionless

hGabsorbed gas in the liquid phase

hiparameter related to each ion in the solution

Iionic strength,mol·m?3

kLaliquid side volumetric mass transfer coefficient,s?1

k*Laliquid side volumetric mass transfer coefficient obtaining by chemical absorption,s?1

kOH– rate constantforreaction between carbon dioxide and hydroxyl ion,m3·(mol· s)?1

Lmain microchannel length,m

mmass,kg

PCO2partial pressure of CO2in the gas phase,Pa

Pdiss.energy dissipated,W

QLliquid phase volumetric flow rate,m3·s?1

Rreaction rate,mol·(m3·s)?1

Ttemperature,K

ttime

UGsuper ficial velocity of gas,m·s?1

ULsuper ficial velocity of liquid,m·s?1

Vvolume,m3

zvalence of ion I,dimensionless

ΔPtwo phase pressure drop,Pa

Subscripts

G gas

in microchannel inlet

L liquid

mc main channel

out microchannel outlet

[1]G.Chen,J.Yue,Q.Yuan,Gas–liquid microreaction technology:Recent developments and future challenges,Chin.J.Chem.Eng.16(5)(2008)663–669.

[2]C.W.Choi,D.I.Yu,M.H.Kim,Adiabatic two-phase flow in rectangular microchannels with different aspect ratios:Part I–Flow pattern,pressure drop and void fraction,Int.J.Heat Mass Transf.54(1–3)(2011)616–624.

[3]H.Ganapathy,E.Al-Hajri,M.Ohadi,Mass transfer characteristics of gas–liquid absorption during Taylor flow in mini/microchannel reactors,Chem.Eng.Sci.101(2013)69–80.

[4]K.Yamamoto,S.Ogata,Effects of T-junction size on bubble generation and flow instability for two-phase flows in circular microchannels,Int.J.Multiphase Flow49(2013)24–30.

[5]A.?zkan,E.Yegan Erdem,Numerical analysis of mixing performance in sinusoidal microchannels based on particle motion in droplets,Microfluid.Nanofluid.19(2015)1101–1108.

[6]Z.Zhang,Z.Qian,L.Xu,C.Wu,K.Guo,Deviation of carbon dioxide–water gas–liquid balance from thermodynamic equilibrium in turbulence I:Experiment and correlation,Chin.J.Chem.Eng.21(7)(2013)770–775.

[7]F.Yang,S.Zhou,X.An,Gas–liquid hydrodynamics in a vessel stirred by dual dislocated-blade Rushton impellers,Chin.J.Chem.Eng.23(11)(2015)1746–1754.

[8]W.Li,X.Geng,Y.Bao,Z.Gao,Micromixing characteristics in a gas–liquid–solid stirred tank with settling particles,Chin.J.Chem.Eng.23(3)(2015)461–470.

[9]Z.Jia,Q.Chang,J.Qin,A.Mamat,Preparation of calcium carbonate nanoparticles with a continuous gas-liquid membrane contactor:Particles morphology and membrane fouling,Chin.J.Chem.Eng.21(2)(2013)121–126.

[10]E.Santacesaria,M.Di Serio,P.Iengo,Mass transfer and kinetics in ethoxylation spray tower loop reactors,Chem.Eng.Sci.54(10)(1999)1499–1504.

[11]J.A.Delgado,M.A.Uguina,J.L.Sotelo,V.I.águeda,A.Sanz,Simulation of CO2absorption into aqueous DEA using a hollow fiber membrane contactor:Evaluation of contactor performance,Chem.Eng.J.152(2–3)(2009)396–405.

[12]S.Krumdieck,J.Wallace,O.Curnow,Compact,low energy CO2management using amine solution in a packed bubble column,Chem.Eng.J.135(1–2)(2008)3–9.

[13]C.P.Stemmet,M.Meeuwse,J.van der Schaaf,B.F.M.Kuster,J.C.Schouten,Gas–liquid mass transfer and axial dispersion in solid foam packings,Chem.Eng.Sci.62(18–20)(2007)5444–5450.

[14]F.Yi,H.-K.Zou,G.-W.Chu,L.Shao,J.-F.Chen,Modeling and experimental studies on absorption of CO2by Ben field solution in rotating packed bed,Chem.Eng.J.145(3)(2009)377–384.

[15]H.Niu,L.Pan,H.Su,S.Wang,Flow pattern,pressure drop,and mass transfer in a gas–liquid concurrent two-phase flow microchannel reactor,Ind.Eng.Chem.Res.48(2009)1621–1628.

[16]T.Y.Chan,G.H.Priestman,J.M.MacInnes,R.W.K.Allen,Development of a microchannel contactor–separator for immiscible liquids,Chem.Eng.Res.Des.86(1)(2008)65–74.

[17]R.Luo,L.Wang,Liquid flow pattern around Taylor bubbles in an etched rectangular microchannel,Chem.Eng.Res.Des.90(8)(2012)998–1010.

[18]M.Heuberger,L.Gottardo,M.Dressler,R.Hufenus,Biphasic fluid oscillator with coaxial injection and upstream mass and momentum transfer,Microfluid.Nanofluid.19(3)(2015)653–663.

[19]W.Ehrfeld,V.Hessel,H.Lowe,Microreactors New Technology for Modern Chemistry,Wiley-VCH,Weinheim,2000.

[20]X.Wang,G.Liu,K.Wang,G.Luo,Measurement of internal flow field during droplet formation process accompanied with mass transfer,Microfluid.Nanofluid.19(3)(2015)757–766.

[21]M.Rahimi,M.Akbari,M.A.Parsamoghadam,A.A.Alsaira fi,CFD study on effect of channel con fluence angle on fluid flow pattern in asymmetrical shaped microchannels,Comput.Chem.Eng.73(2015)172–182.

[22]A.Schuster,K.Se fiane,J.Ponton,Multiphase mass transport in mini/micro-channels microreactor,Chem.Eng.Res.Des.86(5)(2008)527–534.

[23]B.Xu,W.Cai,X.Liu,X.Zhang,Mass transfer behavior of liquid–liquid slug flow in circular cross-section microchannel,Chem.Eng.Res.Des.91(7)(2013)1203–1211.

[24]Y.Zhao,G.Chen,Q.Yuan,Liquid–liquid two-phase flow patterns in a rectangular microchannel,AIChE J.52(12)(2006)4052–4060.

[25]S.Ferrouillat,P.Tochon,H.Peerhossaini,Micromixing enhancement by turbulence:Application to multifunctional heat exchangers,Chem.Eng.Process.45(8)(2006)633–640.

[26]M.Kashid,A.Renken,L.Kiwi-Minsker,Mixing efficiency and energy consumption for five generic microchannel designs,Chem.Eng.J.167(2–3)(2011)436–443.

[27]M.Darekar,K.K.Singh,S.Mukhopadhyay,K.T.Shenoy,S.K.Ghosh,Solvent extraction in microbore tubes with UNPS–TBP in dodecane system,Sep.Purif.Technol.128(2014)96–105.

[28]U.Novak,A.Pohar,I.Plazl,P.?nidar?i?-Plazl,Ionic liquid-based aqueous twophase extraction within a microchannel system,Sep.Purif.Technol.97(2012)172–178.

[29]H.Su,S.Wang,H.Niu,L.Pan,A.Wang,Y.Hu,Mass transfer characteristics of H2S absorption from gaseous mixture into methyldiethanolamine solution in a T-junction microchannel,Sep.Purif.Technol.72(3)(2010)326–334.

[30]C.Ye,G.Chen,Q.Yuan,Process characteristics of CO2absorption by aqueous Monoethanolamine in a microchannel reactor,Chin.J.Chem.Eng.20(1)(2012)111–119.

[31]B.Agostini,J.R.Thome,M.Fabbri,B.Michel,D.Calmi,U.Kloter,High heat flux flow boiling in silicon multi-microchannels–Part I:Heat transfer characteristics of refrigerant R236fa,Int.J.Heat Mass Transf.51(21–22)(2008)5400–5414.

[32]T.Harirchian,S.V.Garimella,A comprehensive flow regime map for microchannel flow boiling with quantitative transition criteria,Int.J.Heat Mass Transf.53(13–14)(2010)2694–2702.

[33]W.Li,Z.Wu,A general correlation for adiabatic two-phase pressure drop in micro/mini-channels,Int.J.Heat Mass Transf.53(13–14)(2010)2732–2739.

[34]S.-S.Hsieh,C.-Y.Lin,Correlation of critical heat flux and two-phase friction factor for subcooled convective boiling in structured surface microchannels,Int.J.Heat MassTransf.55(1–3)(2012)32–42.

[35]T.Harirchian,S.V.Garimella,Flow regime-based modeling of heat transfer and pressure drop in microchannel flow boiling,Int.J.Heat Mass Transf.55(4)(2012)1246–1260.

[36]E.Jafarifar,M.Hajialyani,M.Akbari,M.Rahimi,Y.Shokoohinia,A.Fattahi,Preparation of a reproducible long-acting formulation of risperidone-loaded PLGA microspheres using micro fluidic method,Pharm.Dev.Technol.(2016)1–8.

[37]V.M.Rajesh,V.V.Buwa,Experimental characterization of gas–liquid–liquid flows in T-junction microchannels,Chem.Eng.J.207-208(2012)832–844.

[38]J.Yue,G.Chen,Q.Yuan,L.Luo,Y.Gonthier,Hydrodynamics and mass transfer characteristics in gas–liquid flow through a rectangular microchannel,Chem.Eng.Sci.62(7)(2007)2096–2108.

[39]J.Yue,L.Luo,Y.Gonthier,G.Chen,Q.Yuan,An experimental investigation of gas–liquid two-phase flow in single microchannel contactors,Chem.Eng.Sci.63(16)(2008)4189–4202.

[40]J.Yue,L.Luo,Y.Gonthier,G.Chen,Q.Yuan,An experimental study of air–water Taylor flow and mass transfer inside square microchannels,Chem.Eng.Sci.64(16)(2009)3697–3708.

[41]V.Hessel,W.Ehrfeld,T.Herweck,V.Haverkamp,H.Lowe,J.Schiewe,C.Wille,T.Kern,N.Lutz,Gas/liquid microreactors:Hydrodynamics and mass transfer,4th International Conference on Microreaction Technology,IMRET 4,Atlanta,USA 2000,pp.174–186.

[42]M.Rahimi,N.Azimi,F.Parvizian,Using microparticles to enhance micromixing in a high frequency continuous flow sonoreactor,Chem.Eng.Process.70(2013)250–258.

[43]Z.Yang,S.Matsumoto,H.Goto,M.Matsumoto,R.Maeda,Ultrasonic micromixer for micro fluidic systems,Sens.Actuators A9(2001)266–272.

[44]Z.Dong,C.Yao,Y.Zhang,G.Chen,Q.Yuan,J.Xu,Hydrodynamics and mass transfer of oscillating gas–liquid flow in ultrasonic microreactors,AIChE J.62(4)(2016)1294–1307.

[45]Z.Dong,S.Zhao,Y.Zhang,C.Yao,Q.Yuan,G.Chen,Mixing and residence time distribution in ultrasonic microreactors,AIChE J.63(2017)1404–1418.

[46]Z.Dong,C.Yao,X.Zhang,J.Xu,G.Chen,Y.Zhao,Q.Yuan,A high-power ultrasonic microreactor and its application in gas-liquid mass transfer intensification,Lab Chip15(4)(2015)1145–1152.

[47]M.Rahimi,B.Aghel,B.Hatamifar,M.Akbari,A.Alsaira fi,CFD modeling of mixing intensification assisted with ultrasound wave in a T-type microreactor,Chem.Eng.Process.86(2014)36–46.

[48]M.Rahimi,S.Safari,M.Faryadi,N.Moradi,Experimental investigation on proper use of dual high-low frequency ultrasound waves—Advantage and disadvantage,Chem.Eng.Process.78(2014)17–26.

[49]M.Faryadi,M.Rahimi,S.Safari,N.Moradi,Effect of high frequency ultrasound on micromixing efficiency in microchannels,Chem.Eng.Process.77(2014)13–21.

[50]S.Aljbour,T.Tagawa,H.Yamada,Ultrasound-assisted capillary microreactor for aqueous–organic multiphase reactions,J.Ind.Eng.Chem.15(6)(2009)829–834.

[51]A.Brotchie,F.Grieser,M.Ashokkumar,Effect of power and frequency on bubblesize distributions in acoustic cavitation,Phys.Rev.Lett.102(8)(2009)084302.

[52]M.D.Luque de Castro,F.Priego-Capote,Ultrasound assistance to liquid–liquid extraction:A debatable analytical tool,Anal.Chim.Acta583(1)(2007)2–9.

[53]K.A.Triplett,S.M.Ghiaasiaan,S.I.Abdel-Khalik,D.L.Sadowski,Gas–liquid two-phase flow in microchannels part I:Two-phase flow patterns,Int.J.Multiphase Flow25(1999)377–394.

[54]M.Rahimi,M.Dehbani,M.Abolhasani,Experimental study on the effects of acoustic streaming of high frequency ultrasonic waves on convective heat transfer:Effects of transducer position and wave interference,Int.Commun.Heat Mass Transfer39(5)(2012)720–725.

[55]P.R.Gogate,I.Z.Shirgaonkar,M.Sivakumar,P.Senthilkumar,N.P.Vichare,A.B.Pandit,Cavitation reactors:efficiency assessment using a model reaction,AIChE J.47(2001)2526–2538.

[56]M.Zan fir,A.Gavriilidis,C.Wille,V.Hessel,Carbon dioxide absorption in a falling film microstructured reactor:Experiments and modeling,Ind.Eng.Chem.Res.44(2005)1742–1751.

[57]A.Schumpe,The estimation of gas solubilities in salt solutions,Chem.Eng.Sci.48(1993)153–158.

[58]P.V.Danckwerts,Gas–Liquid Reactions,McGraw-Hill,New York,USA,1970.

[59]D.Roberts,P.V.Danckwerts,Kinetics of CO2absorption in alkaline solutions—I.Transient absorption rates and catalysis by arsenite,Chem.Eng.Sci.17(1962)961–969.

[60]H.Hikita,S.Asai,H.Ishikawa,M.Seko,H.Kitajima,Diffusivities of carbon dioxide in aqueous mixed electrolyte solutions,Chem.Eng.J.17(1979)77–80.

[61]R.Pohorecki,W.Moniuk,Kinetics of reaction between carbon dioxide and hydroxylions in aqueous electrolyte solutions,Chem.Eng.Sci.43(1988)I677–1684.

[62]M.M.Sharma,P.V.Danckwerts,Chemical methods of measuring interfacial areas and mass transfer coefficients in two fluid systems,Braz.J.Chem.Eng.15(1970)522–528.

[63]J.F.Chen,G.Z.Chen,J.X.Wang,L.Shao,P.F.Li,High-throughput microporous tubein-tube microreactor as novel gas-liquid contactor:Mass transfer study,AIChE J.57(1)(2011)239–249.