Kinetics of Reactive Absorption of Carbon Dioxide with Solutions of

File failed to load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... Department of Chemical Engineering, Indian Institute of T...
0 downloads 0 Views 149KB Size
6632

Ind. Eng. Chem. Res. 2006, 45, 6632-6639

Kinetics of Reactive Absorption of Carbon Dioxide with Solutions of Aniline in Nonaqueous Aprotic Solvents Srikanta Dinda, Anand V. Patwardhan,* and Narayan C. Pradhan Department of Chemical Engineering, Indian Institute of Technology, Kharagpur 721302, India

Kinetics of reactive absorption of carbon dioxide (CO2) with solutions of aniline in commercial nonaqueous aprotic solvents, namely, acetonitrile, methyl ethyl ketone (MEK), toluene, and m-xylene was studied in stirred cell and model stirred contactor at 303 ( 1 K and atmospheric pressure. CO2 partial pressure was varied from 0.57 to 1 atm, and the concentration of liquid-phase reactant aniline was varied from 0.75 to 5 kmol/m3. In the case of all four solvents used, the reaction appeared to be first order with respect to CO2. The reaction order with respect to aniline was found to be -0.5, 1, 2, and 2.5 in the case of acetonitrile, MEK, toluene, and m-xylene, respectively. The intrinsic kinetic rate constants at 303 ( 1 K were found to be 1.40 × 102 (kmol/m3)0.5/s, 4.90 × 10-1 1/((kmol/m3) s), 2.02 × 10-1 1/((kmol/m3)2 s), and 2.45 × 10-2 1/((kmol/ m3)2.5 s), in the case of acetonitrile, MEK, toluene, and m-xylene, respectively, as solvents. Introduction The reactive absorption of CO2 with aniline dissolved in nonaqueous aprotic solvents, namely, acetonitrile, methyl ethyl ketone (MEK), toluene, and m-xylene was studied. Traditionally, the reactive absorption of acid gases with aqueous amines has been very important in gas purification processes.1-3 The most common absorbents used for reactive absorption of CO2 are aqueous solutions of amines and amine blends. Aqueous solutions of alkanolamines have been widely used in gas purification processes for the removal of acid gases such as CO2, hydrogen sulfide, carbonyl sulfide, etc.2 In recent years, CO2 is attracting the serious attention of research workers in synthetic chemistry and catalysis, for its realizable potential as a phosgene replacement in the synthesis of intermediates of high commercial value, such as carbamates, isocyanates, and urethanes, that are found in wide applications in areas such as polyurethanes, decorative paints, auto refinishes, wood adhesives, pesticides, rubber, and printing and printed packaging. Traditionally, organic isocyanates are manufactured by reaction of an amine with phosgene (COCl2), and the intermediate of this reaction is converted to the corresponding isocyanate in the presence of a nonaqueous aprotic solvent.4-14 This is reflected in a sizable number of recent research publications in the area of synthetic and mechanistic organic chemistry15-35 as well as recent patents.36-53 These recently patented “greener” methods do not require highly toxic COCl2. Out of all these methods, the most benign appears to be the one in which COCl2 has been replaced by CO2. Use of CO2 in place of COCl2 has the potential for cheaper, greener, inherently safer, and possibly more selective routes to produce industrially useful carbamates, isocyanates, and polyurethanes. For example, aniline is raw material for production of phenyl isocyanate. Hence, there is a tremendous potential for utilization of CO2 in the commercial production of phenyl isocyanate. However, the commercial utilization of this process makes the systematic kinetic study necessary, which has not yet been reported. In addition, there is scanty information in the literature on the kinetics of reactive absorption of CO2 with solutions of amines in nonaqueous solvents. It is, however, apparent that these reactions are fast and mass transfer may be * To whom correspondence should be addressed. Tel.: +91-3222283950. Fax: +91-3222-255303. E-mail: [email protected].

accompanied by chemical reaction in the diffusion film. It was, therefore, thought desirable to study the kinetics of reactive absorption of CO2 with solutions of aniline in commercially available nonaqueous aprotic solvents such as acetonitrile, methyl ethyl ketone (MEK), toluene, and m-xylene. Previous Studies Recently, Aresta et al.6 have reported the kinetics and mechanism of the homogeneous reaction of some organic amines with dimethyl carbonate (DMC) in the presence of dissolved CO2. Second-order kinetics was used in the analysis. Caplow54 has studied the homogeneous uncatalyzed and hydroxide-catalyzed reactions of some secondary amines with CO2. Second-order kinetics was observed. Crooks and Donnellan55,56 have reported the formation of N,N-dialkylcarbamate from diethanolamine and CO2 in anhydrous ethanol. The rate of formation of N,N-dialkylcarbamate from diethanolamine and CO2 in anhydrous ethanol was found to depend on the square of the amine concentration. The rate and activation data are consistent with the Danckwerts’ mechanism,54,57,58 in which a zwitterion intermediate is present at very low concentration and reacts with a second molecule of amine at the diffusioncontrolled limit to give a final salt in the rate-determining step. Ali et al.59 have reported the homogeneous kinetics of reaction of some primary amines with CO2 in ethanol solution, using the stopped flow technique. The highest conversion to carbamate ion was detected with hexamine. The results favor the zwitterion intermediate mechanism proposed by Caplow54 and Danckwerts.57,58 The reaction order was found to increase (ranging from 1 to 2) with the basicity of the amine. Very recently, Masuda et al.60 have studied the solvent dependence of the carbamic acid formation from ω-(1-naphthyl) alkylamines and dissolved CO2, in a variety of solvents such as dimethyl sulfoxide (DMSO), dimethyl formamide (DMF), pyridine, dioxane, acetonitrile, benzene, chloroform, 2-propanol, methanol, and water. Bubbling of CO2 through solutions of naphthyl alkylamines in DMSO, DMF, or pyridine (protophilic, highly dipolar, aprotic solvents) resulted in complete conversion of the amines to the corresponding carbamic acids. In dioxane (protophilic, dipolar, aprotic solvent), the carbamic acid and a small amount of the ammonium carbamates were formed. By contrast, in acetonitrile (protophobic, dipolar, aprotic solvent), in benzene

10.1021/ie060147k CCC: $33.50 © 2006 American Chemical Society Published on Web 08/30/2006

Ind. Eng. Chem. Res., Vol. 45, No. 20, 2006 6633

or chloroform (apolar, aprotic solvents), or in 2-propanol or methanol (dipolar, amphiprotic solvent), ammonium carbamates were formed, although the ammonium bicarbonates/carbonates were competitively formed in methanol. Most of the reactive CO2 absorption studies have been conducted in aqueous media except in recent research work by Alvarez-Fuster et al.61 and Sada et al.62,63 Alvarez-Fuster et al.61 found the reaction order with respect to cyclohexylamine in ethanediol to be 1. Sada et al.62 studied the heterogeneous kinetics of reactive absorption of CO2 with mono- and diethanolamine in solvents such as methanol, ethanol, 2-propanol, and water, using a stirred tank absorber with a plane gas-liquid interface at 303 K. The reaction was found to be first order with respect to CO2 for every solvent. The order of reaction with respect to ethanolamine was found to be 1 only for aqueous solutions of monoethanolamine; for the other solutions, the order ranged from 1.4 to 2, depending on the solvent species. Sada et al.63 studied the heterogeneous kinetics of reactive absorption of CO2 with solutions of monoisopropanolamine and cyclohexylamine in solvents such as methanol, ethanol, 2-propanol, toluene, and water, using a semibatch stirred tank with a plane gas-liquid interface at 303 K. The reaction was found to be first order with respect to CO2 for every solvent. The order with respect to monoisopropanolamine was 1 for aqueous solution and had a value between 1.46 and 1.93 for the other alcoholic solution systems. The order with respect to cyclohexylamine was almost 1 for aqueous and alcoholic solutions and 2 for a toluene solution. From the foregoing discussion, it is clear that very limited data are available on the kinetics of reactive absorption of CO2 with solutions of amines in nonaqueous solvents. These data will be essential for the possible utilization of CO2 in the commercial production of many commercially important carbamates/isocyantes. This work was, therefore, undertaken to study the kinetics of reactive absorption of CO2 with solutions of aniline (as a model amine) in commercially available nonaqueous aprotic solvents such as acetonitrile, methyl ethyl ketone (MEK), toluene, and m-xylene. It is expected that the data obtained for aniline will be useful for many more commercially important aromatic and aliphatic mono- and diamines, which are used in the industrial manufacture of carbamates and isocyanates. Experimental Apparatus and Procedure Stirred Cell. Pure CO2 from a cylinder was used. CO2 was stored in a balloon at atmospheric pressure. All the experiments were carried out in a 64 mm i.d. glass stirred cell. The design of the stirred cell was similar to that used by Patwardhan and Sharma.64,65 A glass stirrer with four blades, which just dipped into the liquid, was used. The gas phase in the stirred cell was also agitated by using a cruciform stirrer. The stirrer speed was varied from 45 to 95 rev/min. The effective interfacial area was 31.4 × 10-4 m2. A known amount of a solution was taken, and the volumetric rate of absorption of pure CO2, stored in a balloon at essentially atmospheric pressure, was noted, using a soapfilm meter of appropriate size. The concentration of aniline was varied from 0.75 to 5 kmol/m3. Prior to the measurement of the volumetric rate of absorption, the stirred cell was purged with pure CO2 for a sufficient time. The purging was stopped, and the unit was connected to a balloon containing pure CO2 at essentially atmospheric pressure. After about 100 s, the volumetric rate of uptake of CO2 was noted. Stirred Contactor with a Flat Gas-Liquid Interface. This type of contactor, which has independent stirrers for gas and

Figure 1. Variation in predicted solubility of CO2 in aniline solutions in different solvents at 303 K. Table 1. Predicted Solubility of CO2 in Aniline Solutions in Different Solvents at 303 ( 1 K predicted solubility of CO2, cAi × 102, kmol/(m3 atm)

[aniline], cB0, kmol/m3

acetonitrile

MEK

toluene

m-xylene

0.75 1.40 1.92 2.95 5.00

1.24 1.20 1.17 1.13 1.11

6.89 6.34 5.92 5.11 3.72

8.23 7.57 7.05 6.07 4.28

7.70 7.14 6.71 5.83 4.21

liquid phases and where the interface is kept flat and no gas dispersion is allowed, was employed to study the effect of CO2 partial pressure on the specific rate of CO2 absorption with aniline solutions in different solvents. The design features of this contactor were akin to those employed by Patwardhan and Sharma,64,65 and Yadav and Sharma.66 This model stirred contactor was operated at a gas-side stirrer speed of 2880 rev/ min and liquid-side stirrer speed in the range of 45 to 95 rev/ min. The mode of operation of contactor was semicontinuous. To assess the effect of gas-side resistance, the gas-side stirrer speed was varied from 720 to 2880 rev/min. All the experiments were conducted at 303 ( 1 K and essentially at atmospheric pressure by employing mixtures of CO2 and nitrogen. The specific rates of absorption were calculated on the basis of chemical analysis of carbamate formed in the liquid phase. Physicochemical Data Solubility of CO2 in Solutions of Aniline in Different Solvents. The equilibrium mole fraction (x2) of CO2 in the mixtures of aniline in acetonitrile, MEK, toluene, and m-xylene was predicted using regular solution theory,67 using the equation,

{

}

V2(δ1 - δ2)2φ12 1 fhyp,liq2 ) exp x2 fgas2 RT

(1)

The predicted values of the equilibrium mole fraction of CO2 in pure acetonitrile, MEK, toluene, and m-xylene were cross checked experimentally by measuring the dissolved CO2 concentration in CO2-saturated solutions of acetonitrile, MEK, toluene, and m-xylene at 303 ( 1 K. The predicted as well as experimentally measured values for toluene compared well with those reported in the literature.15 The corresponding values for the other solvents, namely, acetonitrile, MEK, and m-xylene were not available in the literature. Hence, the CO2 solubility (cAi) values for the mixtures of aniline in acetonitrile, MEK, toluene, and m-xylene were calculated from predicted mole fraction solubility

6634

Ind. Eng. Chem. Res., Vol. 45, No. 20, 2006

Figure 2. Variation in predicted diffusivity of CO2 in aniline solutions in different solvents at 303 K.

Figure 4. Variation of specific absorption rate of CO2 with stirring speed at different aniline concentrations in acetonitrile, at 303 ( 1 K (stirred cell).

Physical Mass Transfer Coefficient. Values of the physical mass-transfer coefficient, kL, are needed to check the conditions for different controlling regimes.69 The kL values at different speeds of stirring in the stirred cell were determined by absorbing pure CO2 in deionized water. The kL values at different speeds of stirring in the model stirred contactor were obtained from earlier work by Yadav66 conducted on a model stirred contactor with identical configuration. The values of diffusivity of CO2 in aniline solutions were estimated from

DA µ ) constant T Materials Figure 3. Estimated variation of cAiDA1/2/(cAiDA1/2)cB0)0 with aniline concentration in different solvents at 303 K. Table 2. Predicted Diffusivity of CO2 in Aniline Solutions in Different Solvents at 303 ( 1 K predicted diffusivity of CO2, DA × 109, m2/s

[aniline], cB0, kmol/m3

acetonitrile

MEK

toluene

m-xylene

0.75 1.40 1.92 2.95 5.00

4.14 4.02 3.90 3.63 2.90

5.54 5.01 4.60 3.88 2.68

4.42 3.98 3.66 3.08 2.16

4.08 3.64 3.34 2.79 1.96

value (Table 1 and Figure 1). This variation of the solubility of CO2 can be explained on the basis of the polarity of aniline and solvent molecules. The order of polarity is acetonitrile > MEK > aniline > toluene > m-xylene. As the concentration of aniline increases, the solubility of CO2 decreases in all the solvents but the extent of decrease is different in different solvents. For acetonitrile, the extent of decrease is very low. In the case of MEK, toluene, and m-xylene, the extent of decrease is high. Diffusivity of CO2 in Solutions of Aniline in Different Solvents. The values of diffusivity of CO2 in acetonitrile, MEK, toluene, m-xylene and in the mixtures of aniline in acetonitrile, MEK, toluene, and m-xylene were predicted using the WilkeChang formula.68 Please refer to Table 2 and Figure 2. The diffusivity of CO2 decreases with aniline concentration due to an increase of viscosity and average molecular weight of the aniline solution. The variation of estimated combined physicochemical property (cAiDA1/2) of CO2 in the liquid medium with different aniline concentrations is shown in Figure 3.

Aniline of synthesis grade, and acetonitrile, toluene, m-xylene, and methyl ethyl ketone (MEK) of Laboratory Reagent grade were obtained from Merck Limited. A CO2 cylinder containing Analytical Reagent grade CO2 was obtained from BOC (Special Gases) Limited, Mumbai. A nitrogen gas cylinder was obtained from a reputed firm. Chemical Reaction The following chemical reaction was considered for the analysis of experimental results: km,n

A(dissolved gas) + zB(liquid-phase reactant) 98 product

(2)

where A ) CO2, B ) aniline, product ) carbamate, z ) stoichiometric factor ) 2, and km,n ) (m + n)th-order kinetic rate constant for the reaction between A and B. Results and Discussion Most of the experiments were conducted in the stirred cell. However, the solubility of CO2 in solutions of aniline in all the solvents used is not very small (≈1.24 × 10-2 kmol/(m3 atm) in acetonitrile, ≈6.89 × 10-2 kmol/(m3 atm) in MEK, ≈8.23 × 10-2 kmol/(m3 atm) in toluene, and ≈7.70 × 10-2 kmol/(m3 atm) in m-xylene); hence, it is expected that gas-side resistance may be present when CO2 is absorbed from mixtures of CO2 and nitrogen. Hence, to find out the order of reaction with respect to CO2, some experiments were conducted in the model stirred contactor as well, wherein gas-side resistance was

Ind. Eng. Chem. Res., Vol. 45, No. 20, 2006 6635

Figure 5. Variation of specific absorption rate of CO2 with stirring speed at different aniline concentrations in MEK, at 303 ( 1 K (stirred cell).

Figure 7. Variation of specific absorption rate of CO2 with stirring speed at different aniline concentrations in xylene, at 303 ( 1 K (stirred cell).

Figure 8. Variation of (RA/cAi)2/DA with aniline concentration, at 303 ( 1 K, for different solvents (stirring speed ) 95 rev/min, stirred cell).

Figure 6. Variation of specific absorption rate of CO2 with stirring speed at different aniline concentrations in toluene, at 303 ( 1 K (stirred cell).

eliminated by stirring the gas phase independently at a very high stirring speed of about 2880 rev/min. Absorption of CO2 with a Solution of Aniline in Acetonitrile. Figure 4 shows the variation of the specific absorption rate of pure CO2 with stirring speed in the stirred cell, at different aniline concentrations. The specific rate appeared to be dependent on the hydrodynamic factors in the range of stirring speed from 45 to 95 rev/min, for higher aniline concentrations. However, the specific rate was found to be almost independent of the hydrodynamic factors, for lower aniline concentrations. The specific rate of absorption of pure CO2 was found to be proportional to (cB0)-1/2, at a fixed stirring speed of 95 rev/ min. Here, instead of plotting ln(RA) versus ln(cB0), we have plotted ln[RA2/(cAi2DA)] versus ln(cB0)] in Figure 8, owing to our observation that both cAi and liquid viscosity (and hence DA) change with a change in cB0. It was found from this graph that n, namely, the order of reaction with respect to liquid-phase reactant B (aniline), is -0.53, which can be rounded off to -0.5. The negative order explained the observation that the specific rate appeared to be dependent on the hydrodynamic factors for higher aniline concentrations. The specific rate of absorption in the model stirred contactor was found to be directly proportional to the partial pressure of

Figure 9. Variation of specific absorption rate of CO2 with partial pressure of CO2 at 1.92 kmol/m3 aniline concentration, in different solvents, at 303 ( 1 K (stirring speed ) 95 rev/min, stirred contactor).

CO2, which was varied from 0.57 to 0.9 atm (Figure 9). It was found from this graph that m, namely, the order of reaction with respect to solute gas A (CO2), is 1. The dissolved gas undergoes a pseudo-first-order reaction when the concentration of the reactant in the neighborhood of the gas-liquid interface is not very different from that in the bulk and the concentration of the dissolved gas is very small

6636

Ind. Eng. Chem. Res., Vol. 45, No. 20, 2006

Table 3. Absorption of CO2 with Aniline Dissolved in Acetonitrile at 303 ( 1 K with a Stirring Speed ) 95 rev/min

a

contactor

cB0 (kmol/m3)

pA a (atm)

rate eq

k1,-0.5 × 10-2 (kmol/m3)0.5/s

remarks

stirred cell stirred cell stirred cell stirred cell stirred cell stirred contactor stirred contactor stirred contactor

0.75 1.40 1.92 2.95 5.00 1.92 1.92 1.92

0.86 0.86 0.87 0.87 0.90 0.57 0.67 0.77

5 5 5 5 5 5 5 5

1.39 1.46 1.41 1.46 1.38 1.37 1.35 1.39

fast reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime

Values are corrected for the vapor pressure of acetonitrile.

Table 4. Absorption of CO2 with Aniline Dissolved in MEK at 303 ( 1 K with a Stirring Speed ) 95 rev/min

a

contactor

cB0 (kmol/m3)

pA a (atm)

rate eq

k1,1 × 101 1/((kmol/m3) s)

remarks

stirred cell stirred cell stirred cell stirred cell stirred cell stirred contactor stirred contactor stirred contactor

0.75 1.40 1.92 2.95 5.00 1.92 1.92 1.92

0.86 0.86 0.87 0.87 0.90 0.57 0.67 0.77

6 6 6 5 5 5 5 5

5.30 4.00 4.34 5.32 5.54 4.60 5.00 5.10

intermediate reaction regime intermediate reaction regime intermediate reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime

Values are corrected for the vapor pressure of MEK.

Table 5. Absorption of CO2 with Aniline Dissolved in Toluene at 303 ( 1 K with a Stirring Speed ) 95 rev/min contactor

pA a rate cB0 k1,2 × 101 (kmol/m3) (atm) eq 1/((kmol/m3)2 s)

stirred cell

0.75

0.96

6

2.03

stirred cell

1.40

0.96

6

1.90

stirred cell

1.92

0.96

5

1.95

stirred cell

2.95

0.97

5

1.98

stirred cell

5.00

0.97

5

2.06

stirred contactor

1.92

0.66

5

2.02

stirred contactor

1.92

0.76

5

2.11

stirred contactor

1.92

0.86

5

2.11

a

Table 6. Absorption of CO2 with Aniline Dissolved in m-Xylene at 303 ( 1 K with a Stirring Speed ) 95 rev/min contactor

remarks intermediate reaction regime intermediate reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime fast reaction regime

compared to the reactant concentration. The conditions to be satisfied are69

(DAk1)1/2 .1 kL

(3)

and

( )

cB0 DB (DAk1)1/2 ,1+ kL zcAi DA

1/2

(4)

The specific rate of absorption, RA (kmol/(m2 s)), is then given as

RA ) cAi(DAk1)1/2

stirred cell

0.75

0.99

6

2.59

stirred cell

1.40

0.99

6

2.43

stirred cell

1.92

0.99

6

2.60

stirred cell

2.95

0.99

6

2.30

stirred cell

5.00

0.99

5

2.48

stirred contactor stirred contactor stirred contactor

1.92

0.69

6

2.45

1.92

0.79

6

2.40

1.92

0.89

6

2.35

a

Values are corrected for the vapor pressure of toluene.

(5)

If a part of the reaction occurs in the film and the rest occurs in the bulk, where the concentration of A is zero, then the

pA a rate cB0 k1,2.5 × 102 (kmol/m3) (atm) eq 1/((kmol/m3)2.5 s)

remarks intermediate reaction regime intermediate reaction regime intermediate reaction regime intermediate reaction regime fast reaction regime intermediate reaction regime intermediate reaction regime intermediate reaction regime

Values are corrected for the vapor pressure of m-xylene.

condition given by expression 3 is not satisfied but that given by expression 4 is satisfied. The specific rate of mass transfer is then given by

RA )

cAi(DAk1)1/2

[

]

(DAk1)1/2 tanh kL

(6)

The absorption of CO2 with aniline dissolved in acetonitrile, therefore, appears to conform to the fast pseudo-first-order reaction regime or to the intermediate regime where part of the CO2 reacts in the diffusion film and the rest reacts in the bulk liquid. Table 3 gives the pertinent details. The average value of the intrinsic kinetic rate constant (k1,-1/2), at 303 ( 1 K, was estimated at 1.40 × 102 (kmol/m3)1/2/s. Absorption of CO2 with a Solution of Aniline in MEK. Figure 5 shows the variation of the specific absorption rate of pure CO2 with stirring speed in the stirred cell, at different

Ind. Eng. Chem. Res., Vol. 45, No. 20, 2006 6637 Table 7. Intrinsic Kinetic Rate Constants, Solubility Parameters, and Relative Dielectric Constants for Aniline and the Different Solvents Used values properties

aniline

acetonitrile

MEK

toluene

m-xylene

δ1, at 298 K, MPa1/2 , at 298 K k1,n, at 303 ( 1 K, 1/((kmol/m3)n s)

23.00 7.30

24.60 36.60 1.40 × 102

19.05 18.50 4.90 × 10-1

18.30 2.40 2.02 × 10-1

18.06 2.30 2.45 × 10-2

aniline concentrations. The specific rate appeared to be dependent on the hydrodynamic factors in the range of stirring speeds from 45 to 95 rev/min, for lower aniline concentrations. However, the specific rate was found to be almost independent of the hydrodynamic factors, for higher aniline concentrations. The specific rate of absorption of pure CO2 was found to be directly proportional to (cB0)0.95, at a fixed stirring speed of 95 rev/min. Here, instead of plotting ln(RA) versus ln(cB0), we have plotted ln[RA2/(cAi2DA)] versus ln(cB0) in Figure 8, owing to our observation that both cAi and liquid viscosity (and hence DA) change with a change in cB0. It was found from this graph that n, namely, the order of reaction with respect to liquid-phase reactant B (aniline), is 0.95, which can be rounded off to 1. The specific rate of absorption in the model stirred contactor was found to be directly proportional to the partial pressure of CO2, which was varied from 0.57 to 0.9 atm (Figure 9). It was found from this graph that m, namely, the order of reaction with respect to solute gas A (CO2), is 1. The absorption of CO2 with aniline dissolved in MEK, therefore, appears to conform to the fast pseudo-first-order reaction regime or to the intermediate regime where part of the CO2 reacts in the diffusion film and the rest reacts in the bulk liquid. Table 4 gives the pertinent details. The average value of the intrinsic kinetic rate constant (k1,1), at 303 ( 1 K, was estimated at 4.90 × 10-1 1/((kmol/m3) s). Absorption of CO2 with a Solution of Aniline in Toluene. Figure 6 shows the variation of the specific absorption rate of pure CO2 with stirring speed in the stirred cell, at different aniline concentrations. The specific rate appeared to be dependent on the hydrodynamic factors in the range of stirring speeds from 45 to 95 rev/min, for lower aniline concentrations. However, the specific rate was found to be almost independent of the hydrodynamic factors, for higher aniline concentrations. The specific rate of absorption of pure CO2 was found to be directly proportional to (cB0)1.95, at a fixed stirring speed of 95 rev/min. Here, instead of plotting ln(RA) versus ln(cB0), we have

plotted ln[RA2/(cAi2DA)] versus ln(cB0) in Figure 8, owing to our observation that both cAi and liquid viscosity (and hence DA) change with a change in cB0. It was found from this graph that n, namely, the order of reaction with respect to liquid-phase reactant B (aniline), is 1.95, which can be rounded off to 2. The specific rate of absorption in the model stirred contactor was found to be directly proportional to the partial pressure of CO2, which was varied from 0.66 to 0.97 atm (Figure 9). It was found from this graph that m, namely, the order of reaction with respect to solute gas A (CO2), is 1. The absorption of CO2 with aniline dissolved in toluene, therefore, appears to conform to the fast pseudo-first-order reaction regime or to the intermediate regime where part of the CO2 reacts in the diffusion film and the rest reacts in the bulk liquid. Table 5 gives the pertinent details. The average value of the intrinsic kinetic rate constant (k1,2.5), at 303 ( 1 K, was estimated at 2.02 × 10-1 1/((kmol/m3)2.5 s). Absorption of CO2 with a Solution of Aniline in m-Xylene. Figure 7 shows the variation of the specific absorption rate of pure CO2 with stirring speed in the stirred cell, at different aniline concentrations. The specific rate appeared to be dependent on the hydrodynamic factors in the range of stirring speeds from 45 to 95 rev/min, for all aniline concentrations, but this dependence decreased with increasing aniline concentration. The specific rate of absorption in the model stirred contactor was found to be directly proportional to the partial pressure of CO2, which was varied from 0.69 to 0.99 atm (Figure 9). It was found from this graph that m, namely, the order of reaction with respect to solute gas A (CO2), is 1. Since absorption of CO2 depended on the hydrodynamic factors for the range of CO2 partial pressures and aniline concentrations studied, eq 6 was fitted numerically to the experimental data. It was found that n, namely, the order of reaction with respect to liquid-phase reactant B (aniline), is 2.45, which can rounded off to 2.5. The absorption of CO2 with aniline dissolved in m-xylene, therefore, appears to conform to the intermediate regime where

Figure 10. Variation of the intrinsic kinetic rate constant with the Hildebrand solubility parameter of the solvent.

Figure 11. Variation of the intrinsic kinetic rate constants with the relative dielectric constant of the solvent.

6638

Ind. Eng. Chem. Res., Vol. 45, No. 20, 2006

part of the CO2 reacts in the diffusion film and the rest reacts in the bulk liquid. Table 6 gives the pertinent details. The average value of intrinsic kinetic rate constant (k1,2.5), at 303 ( 1 K, was estimated at 2.45 × 10-2 1/((kmol/m3)2.5 s). Effect of Solvent on Intrinsic Kinetic Rate Constants. For the range of aniline concentrations studied in this work, the intrinsic kinetic rate constant k1,n varied from solvent to solvent (Table 7). The variation of k1,n with the Hildebrand solubility parameter is as shown in Figure 10. The Hildebrand solubility parameter is the total van der Waals force and is a numerical value that indicates the relative solvency behavior of a specific solvent. It is derived from the cohesive energy density of the solvent, which in turn is derived from the heat of vaporization. It can be seen that the value of k1,n increases with an increase in the value of the Hildebrand solubility parameter. The variation of k1,n with the relative dielectric constant is as shown in Figure 11. It can be seen that the value of k1,n increases with an increase in the value of the relative dielectric constant. Conclusions The absorption of CO2 with solutions of aniline in commercially available nonaqueous aprotic solvents such as acetonitrile, MEK, toluene, and m-xylene, under certain conditions, was found to conform to the fast pseudo-first-order reaction regime. The reaction was found to be first-order with respect to CO2. The reaction order with respect to aniline was found to be -0.5, 1, 2, and 2.5 in the cases of acetonitrile, MEK, toluene, and m-xylene, respectively, as solvents. Despite similar chemical structure of toluene and m-xylene, the intrinsic kinetic rate constants in these two cases were found to differ by 1 order of magnitude. These findings will have a direct bearing on the design and scaleup of the absorber in process development for carbamates/isocyanates. The negative order of reaction with respect to aniline in the case of the acetonitrile solvent needs to be investigated further. In addition, there appears to be a correlation between the intrinsic kinetic rate constant and the polarity of the solvent employed. Nomenclature A ) solute gas component (CO2) B ) liquid-phase reactant (aniline) cAi ) concentration of A at gas-liquid interface, kmol/m3 cB0 ) bulk concentration of B in liquid phase, kmol/m3 DA ) diffusivity of A in liquid phase, m2/s DB ) diffusivity of B in liquid phase, m2/s fhyp,liq2 ) fugacity of hypothetical liquid 2, for eq 1, kPa fgas2 ) fugacity of gas 2, for eq 1, kPa k1 ) pseudo-first-order rate constant, 1/s km,n ) intrinsic kinetic rate constants defined in eq 13, 1/((kmol/ m3)-1+(m+n) s) k1,n ) intrinsic kinetic rate constants defined in eq 14, 1/(kmol/ m3)n s) kL ) physical liquid side mass transfer coefficient, m/s m,n ) order with respect to solute gas A and liquid-phase reactant B, respectively pA ) partial pressure of A in gas phase, atm R ) universal gas constant RA ) specific absorption rate of solute gas, kmol/(m2 s) T ) temperature, K V2 ) molar liquid volume of the solute, m3/kmol x2 ) mole fraction of the solute gas in the liquid phase z ) stoichiometric factor for chemical reaction expressed by eq 2

Greek Symbols δ1, δ2 ) Hildebrand solubility parameter of solvent and solute gas respectively, MPa1/2  ) relative dielectric constant of solvent µ ) viscosity of the components, kg/(m s) φ1 ) volume fraction of the solvent Literature Cited (1) Bratzler, K.; Doerges, A. Amisol process purifies gases. Hydrocarbon Process. 1974, 53, 78-80. (2) Wall, J. Gas Processing Handbook. Hydrocarbon Process. 1975, 54, 79-138. (3) Sharma, M. M. Kinetics of reactions of carbonyl sulphide and carbon dioxide with amines and catalysis by Bro¨nsted bases of the hydrolysis of COS. Trans. Faraday Soc. 1965, 61, 681-688. (4) Aresta, M.; Quaranta, E. Role of the macrocyclic polyether in the synthesis of N-alkyl carbamate esters from primary amines, CO2 and alkyl halides in the presence of crown-ethers. Tetrahedron 1992, 48, 1515-1530. (5) Aresta, M.; Quaranta, E. Reaction of alkali-metal tetraphenylborates with amines in the presence of CO2: a new easy way to aliphatic and aromatic alkali-metal carbamates. J. Chem. Soc., Dalton Trans.: Inorg. Chem. 1995, 20, 3359-3363. (6) Aresta, M.; Dibenedetto, A.; Quaranta, E.; Boscolo, M.; Larsson, R. The kinetics and mechanism of the reaction between carbon dioxide and a series of amines: observation and interpretation of an isokinetic effect. J. Mol. Catal. A: Chem. 2001, 174, 7-13. (7) Belforte, A.; Calderazzo, F. Formation of alkylurethanes from carbon dioxide by regioselective O-alkylation of alkali-metal N,N-diethylcarbamates in the presence of complexing agents. J. Chem. Soc., Dalton Trans.: Inorg. Chem. 1989, 5, 1007-1009. (8) Bhanage, B. M.; Fujita, S.-i.; Ikushima, Y.; Masahiko, A. Synthesis of cyclic ureas and urethanes from alkylene diamines and amino alcohols with pressurized carbon dioxide in the absence of catalysts. Green Chem. 2003, 5, 340-342. (9) Jiyan, Z.; Xinqiang, Z.; Yanji, W.; Shufang, W.; Hongjian, Y. Synthesis of MDI from Dimethyl Carbonate over Solid Catalysts. Ind. Eng. Chem. Res. 2002, 41, 5139-5144. (10) Mahe´, R.; Sasaki, Y.; Bruneau, C.; Dixneuf, P. H. Catalytic synthesis of vinyl carbamates from carbon dioxide and alkynes with ruthenium complexes. J. Org. Chem. 1989, 54, 1518-1523. (11) McGhee, W. D.; Pan, Y.; Riley, D. P. Highly selective generation of urethanes from amines, carbon dioxide, and alkyl chlorides. J. Chem. Soc., Chem. Commun. 1994, 6, 699-700. (12) McGhee, W. D.; Riley, D. P.; Christ, K.; Pan, Y.; Parnas, B. Carbon dioxide as a phosgene replacement: synthesis and mechanistic studies of urethanes from amines, CO2, and alkyl chlorides. J. Org. Chem. 1995, 60, 2820-2830. (13) Mitsudo, T.; Hori, Y.; Yamakawa, Y.; Watanabe, Y. Ruthenium catalyzed selective synthesis of enol carbamates by fixation of carbon dioxide. Tetrahedron Lett. 1987, 28, 4417-4418. (14) Waldman, T. E.; McGhee, W. D. Isocyanates from primary amines and carbon dioxide: “dehydration” of carbamate anions. J. Chem. Soc., Chem. Commun. 1994, 8, 957-958. (15) Wilhelmi, E.; Battino, R. Thermodynamic functions of the solubilities of gases in liquids at 25 deg. Chem. ReV. 1973, 73, 1-9. (16) Chaturvedi, D.; Kumar, A.; Ray, S. A high yielding one-pot, novel synthesis of carbamate esters from alcohols using Mitsunobu’s reagent. Tetrahedron Lett. 2003, 44, 7637-7639. (17) Lu, X.-B.; Feng, X.-J.; He, R. Catalytic formation of ethylene carbonate from supercritical carbon dioxide/ethylene oxide mixture with tetradentate Schiff-base complexes as catalyst. Appl. Catal., A: Gen. 2002, 234, 25-33. (18) Srivastava, R.; Manju, M. D.; Srinivas, D.; Ratnasamy, P. PhosgeneFree Synthesis of Carbamates over Zeolite-Based Catalysts. Catal. Lett. 2004, 97, 41-47. (19) Casadei, M. A.; Inesi, A.; Moracci, F. M.; Rossi, L. Electrochemical activation of carbon dioxide: synthesis of carbamates. Chem. Commun. (Cambridge) 1996, 22, 2575-2576. (20) Yoshida, M.; Hara, N.; Okuyama, S. Catalytic production of urethanes from amines and alkyl halides in supercritical carbon dioxide. Chem. Commun. (Cambridge) 2000, 2, 151-152. (21) Abla, M.; Choi, J.-C.; Sakakura, T. Halogen-free process for the conversion of carbon dioxide to urethanes by homogeneous catalysis. Chem. Commun. (Cambridge) 2001, 21, 2238-2239.

Ind. Eng. Chem. Res., Vol. 45, No. 20, 2006 6639 (22) Dell’Amico, D. B.; Calderazzo, F.; Labella, L.; Marchetti, F.; Pampaloni, G. Converting carbon dioxide into carbamato derivatives. Chem. ReV. 2003, 103, 3857-3898. (23) Sasaki, Y.; Dixneuf, P. H. Ruthenium-catalyzed reaction of carbon dioxide, amine, and acetylenic alcohol. J. Org. Chem. 1987, 52, 43894391. (24) Sasaki, Y.; Dixneuf, P. H. Ruthenium-catalyzed synthesis of vinyl carbamates from carbon dioxide, acetylene, and secondary amines. J. Org. Chem. 1987, 52, 314-315. (25) Katritzky, A. R.; Black, M.; Fan, W. Q. Carbon dioxide: a reagent for the simultaneous protection of nucleophilic centers and the activation of alternative locations to electrophilic attack. 17. Substitution of N-methyl1- and N-methyl-2-naphthylamine and side-chain functionalization of o-toluidine. J. Org. Chem. 1991, 56, 5045-5048. (26) Nomura, R.; Hasegawa, Y.; Ishimoto, M.; Toyosaki, T.; Matsuda, H. Carbonylation of amines by carbon dioxide in the presence of an organoantimony catalyst. J. Org. Chem. 1992, 57, 7339-7342. (27) Casadei, M. A.; Moracci, F. M.; Zappia, G.; Inesi, A.; Rossi, L. Electrogenerated Superoxide-Activated Carbon Dioxide. A New Mild and Safe Approach to Organic Carbamates. J. Org. Chem. 1997, 62, 67546759. (28) Saylik, D.; Horvath, M. J.; Elmes, P. S.; Jackson, W. R.; Lovel, C. G.; Moody, K. Preparation of Isocyanates from Primary Amines and Carbon Dioxide Using Mitsunobu Chemistry. J. Org. Chem. 1999, 64, 3940-3946. (29) Salvatore, R. N.; Shin, S. I.; Nagle, A. S.; Jung, K. W. Efficient Carbamate Synthesis via a Three-Component Coupling of an Amine, CO2, and Alkyl Halides in the Presence of Cs2CO3 and Tetrabutylammonium Iodide. J. Org. Chem. 2001, 66, 1035-1037. (30) Bratt, M. O.; Taylor, P. C. Synthesis of Carbonates and Related Compounds from Carbon Dioxide via Methanesulfonyl Carbonates. J. Org. Chem. 2003, 68, 5439-5444. (31) Dinsmore, C. J.; Mercer, S., P. Carboxylation and Mitsunobu reaction of amines to give carbamates: Retention vs inversion of configuration is substituent-dependent. Org. Lett. 2004, 6, 2885-2888. (32) McGhee, W. D.; Riley, D. P. Palladium-mediated synthesis of urethanes from amines, carbon dioxide, and cyclic diolefins. Organometallics 1992, 11, 900-907. (33) McGhee, W. D.; Riley, D. P.; Christ, M. E.; Christ, K. M. Palladium-catalyzed generation of O-allylic urethanes and carbonates from amines/alcohols, carbon dioxide, and allylic chlorides. Organometallics 1993, 12, 1429-1433. (34) Dean, D. C.; Wallace, M. A.; Marks, T. M.; Melillo, D. G. Efficient utilization of [14C]carbon dioxide as a phosgene equivalent for labeled synthesis. Tetrahedron Lett. 1997, 38, 919-922. (35) Horvath, M. J.; Saylik, D.; Elmes, P. S.; Jackson, W. R.; Lovel, C. G.; Moody, K. A Mitsunobu-based procedure for the preparation of alkyl and hindered aryl isocyanates from primary amines and carbon dioxide under mild conditions. Tetrahedron Lett. 1999, 40, 363-366. (36) Choi, J.-C.; Mahamdi, A.; Sakakura, T. Preparation of carbamate esters from amines, alcohols, and carbon dioxide. Japanese Patent JP2004189728, 2004. (37) Sakakura, T.; Mahhumdy, A.; Tsui, C.-C. Preparation of carbamate esters as intermediates for polyurethanes from amines, alcohols, and carbon dioxide. Japanese Patent JP2002212159, 2002. (38) Lee, B. Phosgene-free process for preparing carbamates using dry ice under standard temperature and pressure. International Patent WO 2002034698, 2002. (39) Barth, H.; Betche, H.-J.; Steiner, K.; Schneider, S.; Bayer, U.; Westermayer, M.; Wolfsperger, U. Preparation of carbamates from indoles and other nitrogen heterocycles using carbon dioxide. German Patent DE19913483, 2000. (40) Jung, K. W. Efficient carbamate synthesis by reaction of amines, electrophiles and carbon dioxide in the presence of a cesium compound. International Patent WO2000050389, 2000. (41) Orth, J. H. Preparation of aminocarbamates from diamines and carbon dioxide. International Patent WO9729083, 1997. (42) McGhee, W. D.; Paster, M. D.; Riley, D. P.; Ruettimann, K. W.; Solodar, A. J.; Waldman, T. E. Process for preparing isocyanates. International Patent WO9518098, 1995. (43) Heuer, L. Preparation of carbamates. German Patent DE4318889, 1994.

(44) McGhee, W. D. Process for preparing isocyanates. International Patent WO9417032, 1994. (45) McGhee, W. D.; Waldman, T. E. Phosphazene induced preparation of isocyanates. US Patent US5298651, 1994. (46) McGhee, W. D.; Talley, J. J. Preparation of urethanes and carbonates. US Patent US5302717, 1994. (47) McGhee, W. D.; Stern, M. K.; Waldman, T. E. Process for preparing isocyanate and carbamate ester products. US Patent US5233010, 1993. (48) McGhee, W. D.; Waldman, T. E. Process for preparing isocyanates from amines and carbon dioxide. US Patent US5189205, 1993. (49) McGhee, W. D.; Parnas, B. L.; Riley, D. P.; Talley, J. J. Phosgenefree process for preparing urethane and carbonate monomers and polymers. European Patent EP511948, 1992. (50) McGhee, W. D.; Riley, D. P. Preparation of urethane and carbonate products. European Patent EP477159, 1992. (51) Riley, D. P.; McGhee, W. D. Preparation of urethane products. US Patent US5055577, 1991. (52) Kawanami, H.; Ikushima, Y.; Hatakeda, K.; Torii, K. Preparation of cyclic urethanes from amino alcohols and carbon dioxide under mild and environmentally friendly conditions. Japanese Patent JP2003064062, 2003. (53) Kubota, Y.; Odaka, M.; Tomohiro, T.; Okuno, H. Manufacture of cyclic urethanes. Japanese Patent JP06092945, 1994. (54) Caplow, M. Kinetics of carbamate formation and breakdown. J. Am. Chem. Soc. 1968, 90, 6795-6803. (55) Crooks, J. E.; Donnellan, J. P. Kinetics of the reaction between carbon dioxide and tertiary amines. J. Org. Chem. 1990, 55, 1372-1374. (56) Crooks, J. E.; Donnellan, J. P. Kinetics of the formation of N,Ndialkylcarbamate from diethanolamine and carbon dioxide in anhydrous ethanol. J. Chem. Soc., Perkin Trans. 2: Phys. Org. Chem. 1988, 2, 191194. (57) Danckwerts, P. V. Absorption accompanied by chemical reaction. AIChE J. 1955, 12, 456-463. (58) Danckwerts, P. V. The reaction of carbon dioxide with ethanolamines. Chem. Eng. Sci. 1979, 34, 443-446. (59) Ali, S. H.; Merchant, S. Q.; Fahim, M. A. Kinetic study of reactive absorption of some primary amines with carbon dioxide in ethanol solution. Sep. Purif. Technol. 2000, 18, 163-175. (60) Masuda, K.; Ito, Y.; Horiguchi, M.; Fujita, H. Studies on the solvent dependence of the carbamic acid formation from ω-(1-naphthyl)alkylamines and carbon dioxide. Tetrahedron 2005, 61, 213-229. (61) Alvarez-Fuster, C.; Midoux, N.; Laurent, A.; Charpentier, J. C. Chemical kinetics of the reaction of CO2 with amines in pseudo m-nth order conditions in polar and viscous organic solutions. Chem. Eng. Sci. 1981, 36, 1513-1518. (62) Sada, E.; Kumazawa, H.; Osawa, Y.; Matsuura, M.; Han, Z. Q. Chemical kinetics of the reaction of carbon dioxide with ethanolamines in nonaqueous solvents. AIChE J. 1985, 31, 1297-1303. (63) Sada, E.; Kumazawa, H.; Osawa, Y.; Matsuura, M.; Han, Z. Q. Reaction kinetics of carbon dioxide with amines in nonaqueous solvents. Chem. Eng. J. 1985, 33, 87-95. (64) Patwardhan, A. V. Heterogeneous Reactions. Ph. D. (Technology) dissertation, Department of Chemical Technology, University of Mumbai, India, 1988. (65) Patwardhan, A. V.; Sharma, M. M. Kinetics of absorption of oxygen in aqueous alkaline solutions of polyhydroxybenzenes. Ind. Eng. Chem. Res. 1988, 27, 36-41. (66) Yadav, G. D.; Sharma, M. M. Effect of diffusivity on true gasside mass transfer coefficient in a model stirred contactor with a plane liquid interface. Chem. Eng. Sci. 1979, 34, 1423-1424. (67) Dack, M. R. J. Solution and solubilities (Part 1); John Wiley and Sons: New York, 1974. (68) Danckwerts, P. V. Gas-Liquid Reactions; McGraw-Hill: New York, 1970. (69) Doraiswamy, L. K.; Sharma, M. M. Heterogeneous Reactions: Analysis, Design, and Examples; John Wiley and Sons: New York, 1984; Vol. II.

ReceiVed for reView February 4, 2006 ReVised manuscript receiVed June 25, 2006 Accepted July 21, 2006 IE060147K