Wei Wang ,Zhi Yun ,Zhigang Tang ,Xia Gui

1 College of Chemistry and Chemical Engineering,Nanjing University of Technology,Nanjing 210009,China

2 State Key Laboratory of Chemical Engineering,Department of Chemical Engineering,Tsinghua University,Beijing 100084,China

1.Introduction

As the rapid development of modern industry,a large number of the use of fossil fuels lead to increased CO2emissions in recent years,and the increasing atmospheric CO2level has been one of the main contributors to the greenhouse effect[1,2].This has caused a high level of concern and effort devoted to mitigating the emission of this gas to the atmosphere.At present,several different technologies including postcombustion capture,pre-combustion capture,and oxy-fuel combustion have been developed and designed to reduce CO2emissions associated with the production of electricity from coal- fired power stations.All these technologies can be achieved by several ways of carbon dioxide removal processes such as solid adsorption,solvent absorption,membranes or other biological separation methods[3].Among these techniques,carbon dioxide captured by solvent absorption is one of the most common and commercially mature technologies today and,in many cases,has been considered to be the most effective solution.The most typical and economical solvent-based processes in industry are still based on physical solvents,such as methanol[4-6],propylene carbonate[7,8],andN-methyl-pyrrolidone[9,10].However,the performance of these solvents is not satisfied,and the costs of existing CO2capture technologies are still too high.So,persistent efforts of predicting and searching for new gas absorbents are always being underway[11-13].Recently,some studies have shown by contrast that a macromolecule solvent with ether groups in molecules has a greater solubility and higher selectivity[13,14].

As ourcontinuous work[14-17],this study mainly focuses on the research of physical solvents with ether groups for CO2capture.In this work,isothermal VLE data for the binary systems consisting of carbon dioxide with diethylene glycol(monomethyl,monoethyl,monobutyl,dimethyl,diethyl,dibutyl)ether at high pressure were measured.The experiments were carried out at temperatures ranging from 288.15 to 318.15 K and pressures ranging up to 6 MPa.The experimental data were also correlated with a modified Patel-Teja equation of state(PT EOS)combined with the traditional van der Waals one- fluid mixing rules.

2.Experimental

2.1.Materials and apparatus

2.1.1.Materials

CO2with a volume fraction of0.9999 was supplied by BeiWen Gas in Beijing.diethylene glycol monomethyl ether(C5H12O3,with a mass fraction of 0.999,made in China),diethylene glycol monoethyl ether(C6H14O3,with a mass fraction of 0.999,made in China),diethylene glycol monobutyl ether(C8H18O3,with a mass fraction of 0.999,made in China),diethylene glycol dimethyl ether(C6H14O3,with a mass fraction of 0.999,made in China),diethylene glycol diethyl ether(C8H18O3,with a mass fraction of 0.999,made in China),and diethylene glycol dibutyl ether(C12H26O3,with a mass fraction of 0.999,made in China)were all obtained from Aladdin-Reagent Company in Shanghai.All components were used without further purification.

2.1.2.Apparatus and experimental procedure

The apparatus used in this work was based on the constant-volume method and often described in our previous work[14-17].Details of the experimental apparatus are given in Fig.1.

Fig.1.Schematic diagram of the apparatus.1—CO2 gas cylinder;2—buffer tank;3—water bath;4—equilibrium cell;5—liquid injector;6—waste container;7—vacuum connector;8—digital pressure transducer;9—digital temperature transducer;10—vent valve;11—magnetic stirrer;12—computer.

Compared with other measure techniques,the constant-volume method can be used to measure the VLE data simply and avoid some accumulated errors caused by the decompression stage in the direct measurement.Experiments were performed for different solvents at different temperature in the high-pressure equilibrium cell.After the pressure and temperature reading in the cell being invariable for a long time,the equilibrium can be considered to be reached.The pressure variation was recorded in the computerby the pressure transducer,and the equilibrium pressure could be read from it.The data calculating process is described in detail in the literature[14,16].

2.2.Modeling

The experimental data were correlated with a modified Patel-Teja equation of state(PT EOS)proposed by Yunet al.[18],which is established by a combination of the repulsive term of cubic chain-of rotators equation of state(CCOR EOS)[19]and the attractive term of PT EOS[20].The modified PT EOS has the following term:

wherezis the compressibility factor,vis the molar volume.a,b,andcare the parameters.The parameters are determined as follows:

The λ(Tr)in Eq.(8)is a function of reduced temperature.At critical point,λ(Tr)=1,ζ=Zc.

The α(Tr)is given to be Eq.(13),which is available in the region ofTr＜ 1.5,and whenTr＞ 1.5,α(Tr)can be defined as:

At critical points:

For non-polar substances,the generalized correlations between the parametersS1-S4and acentric factors ω are described as the following:

The critical properties(Tc,Pc,Zc)and acentric factor(ω)of the pure component used in this study are summarized in Table 1.

For calculations of the binary mixtures,the mixture parametersam,bm,andcmcan be obtained by using the traditional van der Waals one- fluid mixing rules as follows:

Table 1Physical properties of CO2 and diethylene glycol(monomethyl,monoethyl,monobutyl,dimethyl,diethyl,dibutyl)ether

wherexiandxjare molar fractions of componentsiandj,respectively.Andkijis the binary interaction parameter.The fitting was determined by minimizing the following objective function:

whereNprepresents the number of experimental data,andandrepresent the fugacity of gas-phase and liquid-phase,respectively.

3.Results and Discussion

3.1.Data processing

When the system attained equilibrium state,the absorbed amount of gasngis calculated from:

wheren1is the amount of gas in buffer tank before the gas charging step andn2is the amount of gas in buffer tank after the gas-charging step.nEis the amount of CO2in the gas phase at the equilibrium state,and the amount of solventnladded to equilibrium cell is calculated from:

where ΔVis the volume of the solvent,ρ is the density of the solvent,andMis the mean molecular weight of the solvent.

The solubility of CO2in the solvent expressed in mole fraction is calculated from:

At last,assume that the solvent follows Raoult's law,and then the equilibrium CO2partial pressurePcan be calculated from:

wherePis the pressure inside the equilibrium cell while phase equilibrium is being established,PSis the saturated vapor pressure of the solvent at the equilibrium temperature,andPiis the inert gas pressure.

The uncertainties of the measurement comprise system errors for pressure,temperature,and volume.The measurement errors of temperature,pressure,and volume areu(T)=0.1 K,u(P1)=u(P2)=0.01 MPa,u(PE)=0.001 MPa,andu(V)=0.05 ml.According to the method of the estimation of uncertainties,the overall uncertainty for the measured solubility of CO2can be calculated from:

in whichu(n1),u(n2),andu(ng)can be estimated by:

3.2.Results and discussion

The operating temperatureT,the total pressurePat the equilibrium state,CO2mole fractionx1in liquid-phase and the estimated uncertainties δ1for the binary systems CO2+diethylene glycol(monomethyl,monoethyl,monobutyl,dimethyl,diethyl,dibutyl)ether are presented in Tables 2 to 7 and shown in Figs.2 to 7.It can be seen from Tables 2 to 7 that the solubilities of CO2in glycol ethers decrease with increasing temperature and decreasing pressure,resulting in an obvious physical absorption process.

Figs.2 to 7 show that mole composition of gas increases linearly with the pressure increase,which conforms to Henry's law.Hence,Henry'slaw constant can be obtained by calculating the linearslope of the linePtox.As a result,Henry's law constants ofCO2in glycolethers at different temperature are presented in Table 8.The results illustrate that the Henry's law constants increase gradually with increasing experimentaltemperatures.As Table 8 shows,diethylene glycol dibutyl ether shows the best performance of CO2capacity under the same conditions,while the diethylene glycol monomethyl ether is the worst among the studied solutions.As previously discussed[13,14],it was known that the CO2-phobic hydroxyl group can hinder the absorption process.Compared with ‘monoether’compound,the hydroxyl group wasreplaced by the ether group in the ‘diether’compound,which can contribute to the high absorption capacity of absorbents.Furthermore,the experimental data also indicate that the solubilities of CO2show an upward trend with the increasing molecular lengths of solvents.

Table 2Mole fraction(x1),partial pressure(P)of CO2 at equilibrium,and uncertainties(δ1)of CO2 in diethylene glycol monomethyl ether from 288.15 K to 318.15 K

Table 3Mole fraction(x1),partial pressure(P)of CO2 at equilibrium,and uncertainties(δ1)of CO2 in diethylene glycol monoethyl ether from 288.15 K to 318.15 K

Table 4Mole fraction(x1),partial pressure(P)of CO2 at equilibrium,and uncertainties(δ1)of CO2 in diethylene glycol monobutyl ether from 288.15 K to 318.15 K

Table 5Mole fraction(x1),partial pressure(P)of CO2 at equilibrium,and uncertainties(δ1)of CO2 in diethylene glycol dimethyl ether from 288.15 K to 318.15 K

Table 6Mole fraction(x1),partial pressure(P)of CO2 at equilibrium,and uncertainties(δ1)of CO2 in diethylene glycol diethyl ether from 288.15 K to 318.15 K

Table 7Mole fraction(x1),partial pressure(P)of CO2 at equilibrium,and uncertainties(δ1)of CO2 in diethylene glycol dibutyl ether from 288.15 K to 318.15 K

Fig.2.p-x diagrams of the binary mixture CO2(1)+diethylene glycol monomethyl ether(2).● 288.15 K;▲ 308.15 K;■ 318.15 K;— results obtained from the equation of state.

Fig.3.p-x diagrams of the binary mixture CO2(1)+diethylene glycol monoethyl ether(2).● 288.15 K;▲ 308.15 K;■ 318.15 K;— results obtained from the equation of state.

Fig.4.p-x diagrams of the binary mixture CO2(1)+diethylene glycol monobutylether(2).● 288.15 K;▲ 308.15 K;■ 318.15 K;— results obtained from the equation of state.

Fig.5.p-x diagrams of the binary mixture CO2(1)+diethylene glycol dimethyl ether(2).● 288.15 K;▲ 308.15 K;■ 318.15 K;— results obtained from the equation of state.

Fig.6.p-x diagrams of the binary mixture CO2(1)+diethylene glycoldiethyl ether(2).●288.15 K;▲308.15 K;■318.15 K;—results obtained from the equation of state.

Fig.7.p-x diagrams of the binary mixture CO2(1)+diethylene glycoldibutylether(2).●288.15 K;▲308.15 K;■318.15 K;—results obtained from the equation of state.

Table 8Henry's law constant of CO2 in diethylene glycol(monomethyl,monoethyl,monobutyl,dimethyl,diethyl,dibutyl)ether from 288.15 K to 318.15 K

Then,the experimental data were correlated by a modified PT EOS combined with the traditional van der Waals one- fluid mixing rules.Figs.2 to 7 show the comparison of the calculated and experimental data for the binary systems and it was found that there is a good conformity between experiment and model calculation.The fitted binary interaction parameterkijassociated to the EOS is shown in Table 9 with the average absolute deviation of pressure(dP).From Table 9,it was shown that the interaction parameters are not dependent on the temperature within the experimental temperature ranges.The largest relative error in pressure was 4.96%,which was in the allowable range.The results indicated that the parameters obtained were reasonable and the correlation methods were suitable to these two binary systems.

Table 9Fitted results for the binary systems CO2+diethylene glycol(monomethyl,monoethyl,monobutyl,dimethyl,diethyl,dibutyl)ether

4.Conclusions

The binary vapor-liquid equilibrium data of CO2in diethylene glycol(monomethyl,monoethyl,monobutyl,dimethyl,diethyl,dibutyl)ether were measured from(288.15 to 318.15)Katpressure up to 6 MPa based on the constant-volume method.The solubility of CO2in all solvents decreased with increasing temperature and decreasing pressure.The results also illustrate that the ether group has a better CO2absorption performance than the hydroxyl group.The experimental VLE data were correlated with a modified PT EOS with the traditional van der Waals one- fluid mixing rules.And there is a good conformity between experimentand model calculation.It was found that the largest relative error in pressure was 4.96%.With the relatively high partial pressures and low temperature,glycol ethers as a kind of physical solvent will be used for the pre-combustion capture techniques and may be applied on an integrated gasification combined cycle(IGCC)power plant.

[1]T.Kuramochi,A.Ramírez,W.Turkenburg,A.Faaij,Comparative assessment of CO2capture technologies for carbon-intensive industrial processes,Prog.Energy Combust.Sci.38(1)(2012)87-112.

[2]S.J.Jiang,W.Zhong,R.Peng,Y.Jiang,J.Zhang,Experimental studies on the influence of HCO3-on absorption and desorption of CO2from ammonia solution,Adv.Mater.Phys.Chem.2(4B)(2012)240-243.

[3]A.Kohl,R.Nielsen,Gas Purification,5th edition Gulf Publishing Company,Houston,2005.

[4]S.N.Joung,H.Y.Shin,S.Y.Kim,K.P.Yoo,C.S.Lee,W.S.Huh,Measurements and correlation of high-pressure VLE of binary CO2-alcohol systems(methanol,ethanol,2-methoxyethanol and 2-ethoxyethanol),Fluid Phase Equilib.185(1-2)(2001)219-230.

[5]K.Bezanehtak,G.Combes,F.Dehghani,N.Foster,D.Tomasko,Vapor-liquid equilibrium for binary systems of carbon dioxide+methanol,hydrogen+methanol,and hydrogen+carbon dioxide at high pressures,J.Chem.Eng.Data47(2)(2002)161-168.

[6]G.Hochgesand,Rectisol and purisol,Ind.Eng.Chem.67(7)(1970)37-43.

[7]R.W.Bucklin,R.L.Schendel,Comparison of fluor solvent and Selexol processes,Energy Prog.4(3)(1984)137-142.

[8]P.D.Mantor,O.Ablib Jr.,K.Y.Song,R.Kobayashi,Solubility of carbon dioxide on propylene carbonate at elevated pressures and higher than ambient temperatures,J.Chem.Eng.Data27(3)(1982)243-245.

[9]M.R.Bohloul,A.Vatani,S.M.Reyghambarzdeh,Experimental and theoretical study of CO2solubility in N-methyl-2-pyrrolidone(NMP),Fluid Phase Equilib.365(2014)106-111.

[10]R.Rajasingam,L.Lioe,Q.T.Pham,F.P.Lucien,Solubility of carbon dioxide in dimethylsulfoxide and N-methyl-2-pyrrolidone at elevated pressure,J.Supercrit.Fluids31(3)(2004)227-234.

[11]M.B.Miller,D.L.Chen,H.B.Xie,D.R.Luebke,J.K.Johnson,R.M.Enick,Solubility of CO2in CO2-Philic oligomers;COSMOtherm predictions and experimental results,Fluid Phase Equilib.287(1)(2009)26-32.

[12]K.Huang,S.Xia,X.M.Zhang,Y.L.Chen,Y.T.Wu,X.B.Hu,Comparative study of the solubilities of SO2in five low volatile organic solvents(sulfolane,ethylene glycol,propylene carbonate,N-methylimidazole,and N-methylpyrrolidone),J.Chem.Eng.Data59(4)(2014)1202-1212.

[13]Y.Li,W.J.Huang,D.X.Zheng,Y.Mi,L.Dong,Solubilities of CO2capture absorbents 2-ethoxyethyl ether,2-butoxyethyl acetate and 2-(2-ethoxyethoxy)ethyl acetate,Fluid Phase Equilib.370(25)(2014)1-7.

[14]X.Gui,Zh.G.Tang,W.Y.Fei,CO2capture with physical solvent dimethyl carbonate(DMC)at high pressures,J.Chem.Eng.Data55(9)(2010)3736-3741.

[15]X.Gui,Zh.G.Tang,W.Y.Fei,Solubility of CO2in alcohols,glycols,ethers and ketones at high pressures from 288.15 K to 318.15,J.Chem.Eng.Data56(5)(2011)2420-2429.

[16]X.Gui,W.Wang,Ch.W.Wang,L.Zhang,Zh.Yun,Zh.G.Tang,Vapor-liquid phase equilibrium data of CO2in some physical solvents from 285.19 K to 313.26 K,J.Chem.Eng.Data59(3)(2014)844-849.

[17]N.Ai,J.Chen,W.Y.Fei,Solubility of carbon dioxide in four mixed solvents,J.Chem.Eng.Data50(2)(2005)492-496.

[18]Z.Yun,M.R.Shi,J.Shi,H.Lee,A new quartic equation of state,Fluid Phase Equilib.187-188(15)(2001)275-298.

[19]H.Kim,H.M.Lin,K.C.Chao,Cubic chain-of-rotators equation of state,Ind.Eng.Chem.Fundam.25(1)(1986)75-84.

[20]N.C.Patel,A.S.Teja,A new cubic equation of state for fluids and fluid mixtures,Chem.Eng.Sci.37(3)(1982)463-473.