Solubility of H2S in Aqueous Diisopropanolamine + Piperazine

Article pubs.acs.org/jced

Solubility of H2S in Aqueous Diisopropanolamine + Piperazine Solutions: New Experimental Data and Modeling with the Electrolyte Cubic Square-Well Equation of State S. H. Mazloumi,† A. Haghtalab,*,† A. H. Jalili,‡ and M. Shokouhi‡ †

Department of Chemical Engineering, Tarbiat Modares University, P.O. Box 14115-143, Tehran, Iran Gas Research Division, Research Institute of Petroleum Industry, Tehran, Iran

‡

ABSTRACT: The solubilities of H2S in aqueous solutions of diisopropanolamine (DIPA) and in aqueous mixtures of DIPA and piperazine (Pz) have been measured. The molality of DIPA was fixed at 2.96 m, and the concentration of Pz was (1.20 or 1.80) m. Experiments were carried out at (40, 60, and 80) °C over the pressure range (19 to 1554) kPa. For modeling of the DIPA + Pz + H2S + H2O system, the electrolyte cubic square well (eCSW) equation of state (EOS) (Haghtalab, A.; Mazloumi, S. H. Fluid Phase Equilib. 2009, 285, 96−104) was applied to predict the total and partial pressures of H2S over aqueous solutions of DIPA, Pz, and DIPA + Pz using only the interaction parameters, kij, of the H2S + H2O system. The reasonable agreement of the eCSW EOS and experimental results demonstrates the good accuracy of the eCSW EOS for thermodynamic modeling of the solubilities of acid gases in aqueous alkanolamine solutions.

■

first and main aim of this study was to obtain experimental data on the solubility of H2S in aqueous solutions of DIPA and activated DIPA (i.e., DIPA + Pz). The other objective of the present work was to model the experimental data using the equation of state (EOS) approach, which is a rather new application of EOSs for such systems.4−9 Thus, the electrolyte cubic square well (eCSW) EOS10,11 was applied to correlate and predict the solubilities of H2S in aqueous solutions of DIPA, Pz, and DIPA + Pz.

INTRODUCTION The experimental determination and modeling of the solubilities of acid gases in aqueous alkanolamine solutions, such as aqueous solutions of monoethanolamine (MEA), methyldiethanolamine (MDEA), diethanolamine (DEA), diisopropanolamine (DIPA), and diglycolamine (DGA), is one of the major fields in chemical engineering thermodynamics, because of their applications in the design of various chemical plants, such as natural gas, refinery, and tail gas treatments. DIPA is one of the major H2S-selective alkanolamines, and it is low-corrosive and has high potential for removal of the other sulfur compounds, such as COS and CS2. This amine is utilized in the sulfinol process, which is usually used for those streams that have a H2S/CO2 ratio greater than 1, where it is not necessary to remove CO2 at the same levels as required for H2S removal. Also, it is used in the Shell Claus off-gas treating (SCOT) process when selective elimination of H2S over CO2 is needed.1 One of the reasons for the applicability of this alkanolamine is that its reactivity with H2S is fast, as for the other amines, so the protonation reaction can progress instantaneously without kinetic control. On the other hand, the reaction of CO2 with DIPA, and alkanolamines generally, is kinetically controlled and slow. To improve the potential of DIPA for better removal of CO2, an alkanolamine with high reactivity with CO2 should be added to the solvent as an activator. One widely used activator that is already known to be a very good additive to accomplish this purpose is piperazine (Pz).2 In spite of the wide application of DIPA in acid gas removal units, experimental data on the solubilities of acid gases in this alkanolamine by itself or in its blends are rare,3 and thus, the © 2012 American Chemical Society

■

EXPERIMENTAL SECTION Materials. H2S (mass fraction purity > 0.999) was purchased from Linda Company, DIPA (mass fraction purity 0.98) from Merck, and Pz (mass fraction purity 0.99) from Acros, and they were used without further purification. Distillated, deionized, and degassed water were employed to prepare aqueous alkanolamine solutions. Apparatus. In this work, the solubility measurements were carried out using a synthetic−static apparatus in which a predefined amount of the solution is fed into a cell, the equilibrium temperature and pressure are monitored, and then the amount of the dissolved gas is calculated.5 Figure 1 presents the apparatus used in this study. The main part of the system is the equilibrium cell, which is made of 316 stainless steel, has one side glass window, has a volume is 117 ± 0.7 cm3, and is equipped with a heat jacket. Agitation of the liquid phase is Received: August 13, 2011 Accepted: August 1, 2012 Published: September 24, 2012 2625

dx.doi.org/10.1021/je2005243 | J. Chem. Eng. Data 2012, 57, 2625−2631

Journal of Chemical & Engineering Data

Article

Figure 1. Apparatus used in this work for measurement of H2S solubility in aqueous alkanolamine solutions.

carried out by a magnetic stirrer. The pressure of the equilibrium cell is measured using a pressure transducer with a range of (0 to 25) bar and an accuracy of 0.1 % of full scale (BD|SENSORS GmbH, Thierstein, Germany). The temperature of the cell is controlled and monitored using a thermostatted water bath (Thomson 2000, accuracy 0.1 K). A gas sample with a volume of 50 cm3 that is connected to gas storage tank with volume of 70 ± 0.6 cm3 is used to inject the H2S into the equilibrium cell. The pressure of the gas sample is measured using a pressure transducer (Druck PTX 1400) with a range of (0 to 40) bar and an accuracy of 0.1 % of full scale. The solution is prepared with a weight balance with an accuracy of 0.001 g. Procedure. First the equilibrium cell was evacuated using a vacuum pump for several minutes, so the alkanolamine solution (approximately 20 cm3) was injected into the cell under vacuum. The equilibrium cell was brought to the desired temperature by water circulation through the heat jacket. H2S gas was fed into the cell from the gas sample, and its amount was calculated using the known values of the pressure in the gas tank before and after the gas injection, the ambient temperature, the volume of the gas sample, and an appropriate equation of state. The precision of the ambient temperature was ± 0.2 °C. The mechanical stirrer was turned on, allowing the liquid and vapor phases to be in contact for about 2 h to ensure that the vapor−liquid equilibrium was achieved. The equilibrium pressure and temperature were then recorded. The acid gas (H2S) loading was calculated as follows: n Hl 2S = n H2S − n Hg 2S

n Hg 2S =

a H 2S =

namine

(3)

nlH2S

where is the number of moles of gas dissolved in the liquid phase, nH2S is the number of moles of gas injected from the gas tank, nHg 2S is the number of moles of H2S in the equilibrium vapor phase in the cell, PH2S is the equilibrium partial pressure of H2S (assumed to be the total equilibrium pressure minus the vapor pressure of the fresh solution), Vg is the volume of the gas phase in the cell (i.e., the total volume of the cell minus the volume of the liquid phase), ZH2S is the compressibility factor (calculated using the Peng−Robinson EOS), R is the gas constant, T is the absolute temperature, aH2S is the acid gas loading, and namine is the number of moles of the injected alkanolamines (DIPA and Pz). It should be noted that to determine PH2S, we did not use the apparent mole fraction of the solvent because the mole fraction of water in solution (xwater > 0.92) was much higher than that of the amine and the vapor pressures of the amines are much lower than that of water, so it was reasonable to consider the vapor pressure of the fresh solution to be that of water only. Thermodynamic Modeling. In this work, thermodynamic modeling of the solubility of H2S in the aqueous solution of DIPA and Pz was performed using the eCSW EOS.6 On the basis of the eCSW EOS, the residual molar Helmholtz free energy is expressed as a res(T , v , x) = a − a ideal = aCSW + aMSA + aBorn

(1)

(4)

where the subscripts “CSW” and “MSA” stand for the cubic square well EOS and the mean spherical approximation theory, respectively. The molar Helmholtz energy for the square well potential is given by11

PH2SVg Z H2SRT

n Hl 2S

(2) 2626



Article

Table 1. Coefficients for the Mole-Fraction-Based Chemical Equilibrium Constants of the Reactions A0 ln Kx,H2S

214.582

ln Kx,DIPAH+ ln Kx,PzH+

−13.2964 −18.135

A1

A2

−12995.4 −4214.0761 −3814.4

0 0

⎛ v ⎞ zRT ⎡ mv + v0w ⎤ ⎟ − ln⎢ aCSW = RT ln⎜ ⎥ ⎝ v − 4τv ⎠ 2 ⎣ mv + v0(l − m) ⎦

Γ=

(6)

∑ ∑ xixjwij

w=

i

j

⎡ ⎛ εij ⎞ ⎤ = ∑ ∑ xixj⎢exp⎜ ⎟ − mij ⎥ ⎢⎣ ⎝ kBT ⎠ ⎥⎦ i j

∑ xiv0i

v0 =

i

D= (7)

∑ xizi i

(8)

(9)

v0i = zi =

(10)

NAσi 3 2

(11)

4 2π 3 λii − 1 3

εiεj (1 − kij)

2Γ 3RTv ⎛⎜ 3 ⎞ 1 + σ Γ⎟ 3πNA ⎝ 2 ⎠

xiZi 2 NAe 2 ⎛ 1 ⎞⎟ ⎜1 − ∑ 4πε0 ⎝ D ⎠ ions σi

(19)

(20)

K x,DIPAH+

H 2O + DIPAH+ XooooooooY H3O+ + DIPA

(21)

K x,PzH+

(13)

H 2O + PzH+ XooooooY H3O+ + Pz

(22)

For the above reactions, the mole-fraction-based chemical equilibrium constants Kx can be expressed as (14)

Kx = e A 0 + A1 / T + A 2 ln T + A3T

and εi is the square well potential depth for species i, σi is its diameter, λiσi is its potential range, kij = kji is the binary interaction parameter, and NA is Avogadro’s number. It should be pointed out that in the above equations, the summations are over all ionic and nonionic species. For the long-range contributions of ions in the eCSW EOS, the explicit version of the MSA12 is used as follows: aMSA = −

(17)

(18)

(12)

σλ i i + σλ j j σi + σj

∑s xs

K x,H2S

in which λij =

(16)

∑s xsDs

H 2S + H 2O XoooooY H3O+ + HS−

and εij =

16, 17 15

It should be noted that the Born term has no influence on the pressure equation of the eCSW EOS; however, it contributes to the chemical potential equation. Expressions eCSW EOSs for the pressure and the chemical potential can be found in our previous work.6 For the solubility of H2S in aqueous solutions of DIPA and Pz, the following chemical reactions take place:

4 2 πλij 3 − 3 4 2 π (λij 3 − 1)

273 to 323 273 to 323

1 [ 1 + 2σκ − 1] 2σ

aBorn = −

where xi is the mole fraction of species i, kB is Boltzmann’s constant, and mij =

0.0099612 0.015096

where Ds is the dielectric constant of pure solvent s. The dielectric constants for the pure solvents were taken from the literature.5,6,13 The Born14 contribution to the molar residual Helmholtz energy is written as

and

z=

6

in which e is the unit of elementary charge, ε0 is the permittivity of the free space, Zi is the charge number of ion i, σ is the average ion diameter (calculated using the linear mixing rule σ = ∑ixiσi/∑ixi, where the summations are over ions), and D is the dielectric constant of the alkanolamine solution, given by

∑ ∑ xixjmij j

ref

273 to 498

⎛ e 2N 2 ⎞ 2 A ⎟ κ 2 = ⎜⎜ ⎟ ∑ xiZi ⎝ Dε0RTv ⎠ ions

where v is the molar volume, m is an orientational parameter, v0 is the close-packed volume, z is the maximum attainable coordination number, w is a related energy parameter, and τ = √2π/6. For a mixture, the following mixing rules are used: i

T/K range

0

and κ is the Debye screening length, given by

(5)

m=

A3

−33.5471

= ∏(xiγi)νi

(23)

i

where xi, γi, and νi are the mole fraction, activity coefficient, and stoichiometric coefficient, respectively, of species i. The values of Ai were taken from the literature6,15−17 and are given in Table 1. In this work, the symmetrical activity coefficient for water was used:

(15)

γwater =

where Γ is the MSA screening parameter, given by 2627

φ

(T , P , x water)

water 0 φwater(T ,

P , x water → 1)

(24)



Article

the following: (1) At fixed partial pressure and constant concentration of DIPA or DIPA + Pz, with decreasing temperature the H2S loading is enhanced. (2) At high loadings at a fixed temperature, DIPA and DIPA + Pz are saturated, so the slope of the pressure-versus-loading curve is decreased. When the amine is saturated, the solubility decreases, and the slope of the curve increases. As shown in the figures, a decrease in the slope was observed because a logarithmic scale was used. (3) At fixed temperature, constant DIPA concentration, and different concentrations of Pz, the curves of partial pressure versus loading cross at the higher acid gas loadings. At the beginning of acid gas solubility in the alkanolamine solution, the molecules of H2S are hydrolyzed to give hydrogen sulfide ion and hydronium ion, and the latter is consumed by the amine molecules, so one molecule of H2S is needed to produce one protonated DIPA. (4) At acid gas loadings below the crossing point, the H2S solubility is enhanced with increasing Pz concentration, and above that point, the reverse phenomenon takes place. Thermodynamic Modeling. As mentioned above, the eCSW EOS10 was applied to model the solubility of H2S in aqueous solutions of DIPA, Pz, and DIPA + Pz. The parameters of the eCSW EOS for pure compounds are the size parameter, σ, the square well depth, ε, and the interaction range parameter, λ. The parameters for pure H2S and H2O were taken from our previous work.6 To obtain the parameters for pure DIPA and Pz, vapor pressure data for these compounds were correlated using the SW EOS.5,18 Table 4 presents the eCSW EOS parameters for the pure and ionic species and also the percent absolute average deviation (%AAD) of the SW EOS in correlating the experimental data for the pure compounds. The next step in the modeling was determining the interaction parameters of the binary subsystems. The values of kij for H2S + H2O systems, which were obtained in our previous work6 by fitting experimental bubble pressure data,19,20 are presented in Table 5. The other binary interaction parameters were assumed to be equal to zero. One should note that all of the binary mixtures were considered as nonideal, but the nondeal behavior of the Pz + H2O and DIPA + H2O systems over the range 0 < xamine < 0.1 was taken into account with the parameters of the pure components using the SW EOS. To demonstrate the predictive capability of the eCSW EOS, the bubble pressures of the ternary H2S + DIPA + H2O and H2S + Pz + H2O systems and the quaternary H2S + DIPA + Pz + H2O system were calculated using only the kij values for the H2S + H2O system. The experimental data for the ternary H2S + Pz + H2O system were those reported by Xia et al.15 and Speyer and Maurer.2 For modeling of the H2S + DIPA + H2O system, the experimental data of Issacs et al.1 together with the experimental data from this work were used. For the solubility of H2S in aqueous DIPA + Pz, the data shown in Table 3 of this work were used. The results predicted by the eCSW EOS are shown in Table 6 and presented in Figure 7 for Pz solutions, Figure 3 for DIPA solutions, and Figures 4 and 5 for DIPA + Pz solutions. As one can see, the eCSW EOS was able to predict the partial and total pressures of H2S over aqueous alkanolamine solutions well. As shown in Figure 7 and Table 6, for the ternary H2S + Pz + H2O system, the prediction results of the model are in very good agreement with the data of Xia et al.,15 but the agreement with the data of Speyer and Maurer2 is only fairly good. It should be noted that the experimental data of Xia et al.15 are total pressures, which indicates that the eCSW EOS generally has more capability to represent total pressures than

For the other species, including the alkanolamine, the reference state of infinite dilution in water (i.e., the unsymmetrical activity coefficient) was used: γi* =

φi(T , P , xi) ∞

φi (T , P , xi → 0)

(i ≠ water) (25)

It should be noted that the coefficients given in Table 1 are consistent with the above definitions of the activity coefficients of the species.

■

RESULTS AND DISCUSSION Experimental Results. To validate the experimental setup and the procedure used in this work, a comparison between the results of the present work and those obtained by Xia et al.15 and Speyer and Maurer2 for the solubility of H2S in about 2 m aqueous solutions of Pz at 313 K was made, as shown in Table 2 and Figure 2. As one can see, good agreement between the experimental results of this work and those of Xia et al.15 and Speyer and Maurer2 was obtained. Table 2. Solubility of H2S in Aqueous 2 m Pz Solution at (40 ± 0.2) °C (Present Work) loading

PH2S

(mol of H2S)·(mol of Pz)−1

bar

± ± ± ± ± ±

0.26 0.50 1.02 1.95 3.46 5.54

0.918 1.096 1.233 1.391 1.518 1.589

0.003 0.003 0.004 0.004 0.005 0.005

Figure 2. Comparison of experimental results from the present work and the literature for the solubility of H2S in aqueous 2 m Pz solution at 313 K: □, Xia et al.;15 ◇, Speyer and Maurer;2 ●, this work.

Table 3 shows the experimental results for the solubility of H2S at (40, 60, and 80) °C in aqueous 2.96 m DIPA solutions and in activated DIPA + Pz solutions in which the concentration of DIPA was 2.96 m and that of Pz was (1.2 or 1.8) m. The loadings of H2S and their uncertainties are given, and the apparent solubilities of H2S (expressed as the mole fraction in the liquid mixture) are also presented. Considering Table 3 and Figures 3 to 6, one can conclude 2628



Article

Table 3. H2S Solubility in Aqueous DIPA or DIPA + Pz Solutions at (40, 60, and 80) °C (40 ± 0.2) °C loading (mol of H2S)·(mol of DIPA+Pz)−1

(60 ± 0.2) °C

uncertainty in loading apparent xH2S

(mol of H2S)·(mol of DIPA+Pz)−1

uncertainty in loading

PH2S

loading

bar


0.619 0.769 0.885 0.972 1.054 1.145 1.260 1.370

0.0304 0.0375 0.0429 0.0469 0.0507 0.0548 0.0600 0.0649

± ± ± ± ± ± ± ±

0.001 0.001 0.002 0.002 0.002 0.002 0.003 0.003

0.26 0.50 1.02 1.95 3.46 5.54 8.68 11.74

0.604 0.740 0.847 0.932 1.019 1.104 1.219 1.323

0.631 0.731 0.859 0.990 1.078 1.162 1.234 1.296

0.0421 0.0485 0.0565 0.0646 0.0699 0.0749 0.0792 0.0829

± ± ± ± ± ± ± ±

0.003 0.003 0.004 0.004 0.005 0.005 0.005 0.006

0.21 0.37 0.76 2.25 4.40 6.76 9.00 11.26

0.621 0.717 0.832 0.957 1.046 1.125 1.189 1.249

0.661 0.747 0.840 0.935 0.998 1.067 1.129 1.205 1.300

0.0496 0.0557 0.0622 0.0688 0.0731 0.0777 0.0819 0.0869 0.0931

± ± ± ± ± ± ± ± ±

0.004 0.005 0.005 0.006 0.006 0.007 0.007 0.008 0.009

0.19 0.36 0.63 1.35 2.44 4.24 6.18 8.66 11.90

0.651 0.732 0.817 0.905 0.966 1.033 1.093 1.163 1.249

apparent xH2S


2.96 m DIPA 0.0297 ± 0.002 0.0361 ± 0.003 0.0411 ± 0.003 0.0451 ± 0.003 0.0491 ± 0.004 0.0529 ± 0.004 0.0581 ± 0.005 0.0628 ± 0.005 2.96 m DIPA + 1.20 m Pz 0.0415 ± 0.003 0.0476 ± 0.003 0.0548 ± 0.004 0.0626 ± 0.005 0.0680 ± 0.005 0.0727 ± 0.005 0.0766 ± 0.006 0.0801 ± 0.006 2.96 m DIPA + 1.80 m Pz 0.0489 ± 0.002 0.0547 ± 0.003 0.0606 ± 0.003 0.0667 ± 0.003 0.0709 ± 0.004 0.0754 ± 0.004 0.0795 ± 0.004 0.0841 ± 0.005 0.0898 ± 0.005

Figure 3. Experimental results from the present work (□, T = 353 K; △, T = 333 K; ○, T = 313 K) and the prediction of the eCSW EOS6 (solid line) for H2S solubility in aqueous 2.96 m DIPA solutions.

(80 ± 0.2) °C uncertainty in loading

PH2S


bar

PH2S

loading

bar


apparent xH2S

0.47 0.90 1.58 2.58 4.13 6.42 9.78 13.15

0.578 0.700 0.798 0.882 0.969 1.064 1.286

0.0284 0.0342 0.0388 0.0427 0.0468 0.0511 0.0611

± ± ± ± ± ± ±

0.004 0.004 0.005 0.006 0.006 0.007 0.008

0.86 1.50 2.34 3.41 5.05 7.36 14.53

0.41 0.68 1.32 3.03 5.31 7.91 10.45 12.90

0.604 0.692 0.796 0.916 1.008 1.085 1.152 1.212

0.0404 0.0460 0.0526 0.0600 0.0657 0.0703 0.0743 0.0779

± ± ± ± ± ± ± ±

0.003 0.004 0.004 0.005 0.006 0.006 0.006 0.007

0.80 1.24 2.15 4.05 6.41 9.22 11.87 14.50

0.43 0.69 1.15 2.06 3.26 5.22 7.33 10.10 13.75

0.634 0.709 0.786 0.868 0.927 0.996 1.056 1.125 1.208

0.0477 0.0530 0.0585 0.0642 0.0682 0.0729 0.0770 0.0816 0.0871

± ± ± ± ± ± ± ± ±

0.003 0.003 0.003 0.004 0.004 0.004 0.005 0.005 0.005

0.84 1.26 1.92 3.02 4.33 6.36 8.59 11.56 15.54

Figure 4. Experimental results from the present work (□, T = 353 K; △, T = 333 K; ○, T = 313 K) and the prediction of the eCSW EOS6 (solid line) for H2S solubility in aqueous 2.96 m DIPA + 1.20 m Pz solutions.

partial pressures. In the case of the H2S + DIPA + H2O system, the overall %AAD in the prediction of partial pressures of H2S over aqueous DIPA solutions is about 20 %. Figure 3 compares the results predicted using eCSW EOS with the experimental data obtained in this work for the partial pressures of H2S over

aqueous 2.96 m DIPA solutions at (313, 333, and 353) K. It can be seen the predicted results are better at 313 K. The partial pressures of H2S over various aqueous solutions of DIPA + Pz were predicted with %AAD of 17 % by the eCSW EOS. Figures 2629



Article

Table 5. Fitted kij Values and %AAD for Correlation and Prediction of H2S + H2O Binary Mixtures kija

a

T/K range

Np

b0

b1

%AAD

refs

298 to 411

53

0.2607

−43.87

2.3

6, 19, 20

kij = b0 + b1/(T/K).

Table 6. %AADs of the Predictions for the Ternary and Quaternary Systems Using Only the kij Values Given in Table 5 loading range system Isaacs et al.1 this work

Figure 5. Experimental results from the present work (□, T = 353 K; △, T = 333 K; ○, T = 313 K) and the prediction of the eCSW EOS6 (solid line) for H2S solubility in aqueous 2.96 m DIPA + 1.80 m Pz solutions.

T/K range 313, 373 313 to 353

Xia et al.15 Speyer and Maurer2

313 to 393 313

this work

313 to 353

amine conc.

mol·mol−1

H2S + DIPA + H2O 2.5 M 0.098 to 1.414 2.96 m 0.58 to 1.37 H2S + Pz + H2O ca. 2, 4 m 0.597 to 2.429 ca. 2 m 0.144 to 1.037

H2S + DIPA + Pz + H2O 2.96 m DIPA 0.604 to 1.296 + 1.20 m Pz 313 to 353 2.96 m DIPA 0.634 to 1.300 + 1.80 m Pz

Np

% AAD

26

21.5

23

17.4

82 10

8.7 31.8

24

12.1

24

10.5

Figure 6. Comparison of the experimental results from the present work for H2S solubility in different aqueous solutions at 353 K: ◇, aqueous 2.96 m DIPA; ■, aqueous 2.96 m DIPA + 1.20 m Pz; △, aqueous 2.96 m DIPA + 1.80 m Pz.

Table 4. Parameters of the eCSW EOS for Pure and Ionic Species ε/kB

σ

species

K

λ

Å

H2O H2S DIPA Pz HS− DIPAH+ PzH+

772.657 524.895 1890.666 847.891 524.895 1890.666 847.891

1.4641 1.3330 1.2232 1.4341 1.3330 1.2232 1.4341

2.317 2.885 1.633 3.384 2.885 1.633 3.384

Figure 7. Predictions of the eCSW EOS6 (solid lines) for H2S solubility in aqueous 2 m Pz solutions at various temperatures. Experimental data:15 ◇, T = 313 K; □, T = 333 K; *, T = 353 K; ○, T = 373 K; △, T = 393 K.

%AADa Tr range 0.44 0.73 0.58 0.59

to to to to

0.94 0.96 0.95 0.97

P

ρ

5.8 2.6 0.8 1.4

3.5 1.9 − −

333, and 353) K. It can be observed that the accuracy of the eCSW EOS in representing the experimental partial pressures of H2S over 2.96 m DIPA + 1.20 m Pz and 2.96 m DIPA + 1.80 m Pz solutions is very good.

■

%AAD = (100 %)·(∑j|Xcalc,j − Xexptl,j|/Xexptl,j)/Np, where Np is the number of data points and X = P or ρ. Tr: reduced temperature; ρ: liquid density. a

CONCLUSION The solubility of H2S in 2.96 m aqueous solutions of diisopropanolamine (DIPA) and aqueous solutions containing 2.96 DIPA m + 1.20 m piperazine (Pz) or 2.96 DIPA m + 1.80 m Pz at various H2S loadings were measured using a static equilibrium cell over the partial pressure range from (19 to 1554) kPa at temperatures of (313, 333, and 353) K. The

4 and 5 present comparisons of the predictions of the present model and the experimental data for 2.96 m DIPA + 1.20 m Pz and 2.96 m DIPA + 1.80 m Pz solutions, respectively, at (313, 2630



Article

amine, Triethanolamine, and 2-Amino-2-methyl-1-propanol Solutions. J. Chem. Eng. Data 2007, 52, 619−623. (14) Born, M. Volumen und Hydrationswarme der Ionen. Z. Phys 1920, 1, 45. (15) Xia, J.; Pérez-Salado Kamps, Á .; Maurer, G. Solubility of H2S in (H2O + piperazine) and in (H2O + MDEA + piperazine). Fluid Phase Equilib. 2003, 207, 23−34. (16) Lee, L. B. A Vapor−liquid equilibrium model for natural gas sweetening process. Ph.D. Thesis, University of Oklahoma, Norman, OK, 1996. (17) Blauwhoff, P. M.; Bos, M. Dissociation Constants of Diethanolamine and Diisopropanolamine in an Aqueous 1.00 M Potassium Chloride Solution. J. Chem. Eng. Data 1981, 26, 7−8. (18) Hysys; Hyprotech Ltd.: Calgary, AB, 1995. (19) Chapoy, A.; Mohammadi, A. H.; Tohidi, B.; Valtz, A.; Richon, D. Experimental Measurement and Phase Behavior Modeling of Hydrogen Sulfide−Water Binary System. Ind. Eng. Chem. Res. 2005, 44, 7567−7574. (20) Selleck, F. T.; Carmichael, L. T.; Sage, B. H. Phase behavior in the hydrogen sulfide−water system. Ind. Eng. Chem. 1952, 44, 2219− 2226.

addition of Pz increases the chemical solubility of H2S in aqueous solutions of DIPA. The experimental data from the present work and those for H2S solubility in aqueous Pz have been used to validate the electrolyte Cubic Square Well (eCSW) equation of state. The interaction parameters in the eCSW EOS, kij, for the binary H2S + H2O system, which were determined previously using experimental bubble pressure data, were employed in the eCSW EOS to predict the total and partial pressures for the H2S + DIPA + H2O, H2S + Pz + H2O, and H2S + DIPA + Pz + H2O systems. In comparison with the experimental data, percent absolute average deviations of 20 %, 11 %, and 17 %, respectively, were obtained, demonstrating the good predictive ability of the eCSW EOS for the studied solutions.

■

AUTHOR INFORMATION

Corresponding Author

*Department of Chemical Engineering, Tarbiat Modares University, P.O. Box 14115-143, Tehran, Iran. Tel.: (09821) 82883313. Fax: (09821)82883381. E-mail: haghtala@modares. ac.ir. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Isaacs, E. E.; Otto, F. D.; Mather, A. E. Solubility of Hydrogen Sulfide and Carbon Dioxide in an Aqueous Diisopropanolamine Solution. J. Chem. Eng. Data 1977, 22, 71−73. (2) Speyer, D.; Maurer, G. Solubility of Hydrogen Sulfide in Aqueous Solutions of Piperazine in the Low Gas-Loading Region. J. Chem. Eng. Data 2011, 56, 763−767. (3) Dell’Era, C.; Uusi-Kyyny, P.; Pokki, J.-P.; Pakkanen, M.; Alopaeus, V. Solubility of carbon dioxide in aqueous solutions of diisopropanolamine and methyldiethanolamine. Fluid Phase Equilib. 2010, 293, 101−109. (4) Chunxi, L.; Fürst, W. Representation of CO2 and H2S solubility in aqueous MDEA solutions using an electrolyte equation of state. Chem. Eng. Sci. 2000, 55, 2975−2988. (5) Derks, P. W. J.; Dijkstra, H. B. S.; Hogendoorn, J. A.; Versteeg, G. F. Solubility of Carbon Dioxide in Aqueous Piperazine Solutions. AIChE J. 2005, 51, 2311−2327. (6) Haghtalab, A.; Mazloumi, S. H. Electrolyte Cubic Square-Well Equation of State for Computation of the Solubility CO2 and H2S in Aqueous MDEA Solutions. Ind. Eng. Chem. Res. 2010, 49, 6221−6230. (7) Huttenhuis, P. J. G.; Agrawal, N. J.; Solbraa, E.; Versteeg, G. F. The Solubility of Carbon Dioxide in Aqueous N-Methyldiethanolamine Solutions. Fluid Phase Equilib. 2008, 264, 99. (8) Huttenhuis, P. J. G.; Agrawal, N. J.; Versteeg, G. F. Solubility of Carbon Dioxide and Hydrogen Sulfide in Aqueous N-Methyldiethanolamine Solutions. Ind. Eng. Chem. Res. 2009, 48, 4051. (9) Archane, A.; Gicquel, L.; Provost, E.; Fürst, W. Effect of methanol addition on water−CO2−diethanolamine system: Influence on CO2 solubility and liquid phase separation. Chem. Eng. Res. Des. 2008, 86, 592. (10) Haghtalab, A.; Mazloumi, S. H. A square-well equation of state for aqueous strong electrolyte solutions. Fluid Phase Equilib. 2009, 285, 96−104. (11) Haghtalab, A.; Mazloumi, S. H. A new coordination number model for development of a square-well equation of state. Fluid Phase Equilib. 2009, 280, 1−8. (12) Harvey, A.; Copeman, T.; Prausnitz, J. Explicit Approximation of the Mean Spherical Approximation for Electrolyte Systems with Unequal Ion Sizes. J. Phys. Chem. 1988, 92, 6432−6436. (13) Hsieh, C.-J.; Chen, J.-M.; Li, M.-H. Dielectric Constants of Aqueous Diisopropanolamine, Diethanolamine, N-Methyldiethanol2631


Solubility of H2S in Aqueous Diisopropanolamine + Piperazine

Recommend Documents