Abstract
In this paper, the experimental solubility of hydrogen sulfide in aqueous N- Methyldiethanolamine + Amine Functionalized UiO-66 (UiO-66-NH2) was studied. UiO-66-NH2 was produced using solvothermal process, and its physicochemical properties were investigated by different techniques including XRD, TGA, TEM, BET, and FTIR to realize its crystalline structure, morphology, thermal stability, and porous structure. The Zeta potential of the solution was turned out to be about 26.6 mV (millivolt), meaning that UiO-66-NH2 particles are moderately stable in aqueous 40 wt.% MDEA. The solubility of hydrogen sulfide has been carried out using the isochoric saturation / or static method in two concentration grades of 0.1 and 0.5 wt.% of UiO-66-NH2 in the aqueous solution of 40 wt.% MDEA known as nanofluid. Experimental measurements were carried out at temperatures of 303.15 through 333.15 K, and pressures up 1100 kPa. Results showed that the addition of UiO-66-NH2 nanoparticles to the MDEA solution altered the results less than 3% , while the mean value of uncertainty reported in this work is about 4% , meaning that the addition of nanoparticles do not have remarkable effect on H2S solubility. In contrast, it causes an increased capacity of CO2 absorption of that solution up to 10%.
Keywords
Introduction
Nanotechnology is a term linked to materials, devices, and systems functioning through the control of matter in the range of 1–100 nm [1–3]. Nanotechnology employs an ability to work at the atomic or molecular level whereby creation of large structures, design, assemblies, and applications can be emerged; in fact, it is an application of nanomaterials and the fundamental understanding of the relationships between physical properties and material dimensions [1, 4–6].
There are wide-ranging uses of nanotechnologies in the environment [7], treatment of polluted water [8, 9], medical industry and drug delivery [10, 11], gas sweetening [12], CO2 adsorptive absorption [13], CO2 adsorption [14, 15], food packaging, processing, production and sensors [16–19], and other applications [20–23].
A nanofluid is a diluted suspension of particles as dispersive nanoparticles [24] (with at least one dimension smaller than ∼100 nm), which may emerge some interesting properties in thermal conductivity, mass diffusion, and gas solubility, and Goharshadi et al. [25] have reported some special features of nanofluids.
Metal-organic frameworks (MOFs), due to their high specific surface area, large pore volume, molecular cell, homogeneous structure, and adjustable chemical functionality [26–30], have been known as potential materials for alternative adsorbents for the treatment of polluted water [31–33] and even separation of acid gases in sweetening processes [12, 34].
UiO-66-NH2 consists of a MOF containing highly Lewis-acidic ZrIV ions, in the form of ZrIV6O4(OH)412 + cluster (Fig. 1) [35]. The gas adsorption capacity of UiO-66 (Uio stands for the university of Oslo) is not high enough compared to other MOFs; however the advantage of UiO-66 is its structural stability and high thermal stability in water [36], which can make a nanofluid for absorption purposes. In literature, most of the studies have been carried out with the application of UiO-66 and NH2-UiO-66 for the adsorptive removal instead of the absorptive one. Among them, Chen and coworkers studied the adsorptive removal of various anionic and cationic dyes using UiO-66 and NH2-UiO-66 MOF [37] and in another work, Luu et al. [38] studied the application of NH2-UiO-66 for adsorption of CO2 and CH4.

Molecular structure of UiO-66-NH2 [6].
The mixture of methanol and zeolitic imidazolate framework (ZIF-8) was proposed for CO2 capture at low temperatures [13]. Zirconium oxide (ZrO2) has stable properties, and the nature of zirconium atom connection with oxygen makes it attractive in the synthesis of MOFs. Adding functional groups is one of the methods to increase the adsorption capacity of MOFs. Vahidi et al. [12] recently have shown that adding UiO-66-NH2 nanoparticles to the MDEA solution, increases CO2 absorption of the solution up to about 10%.
In the present work, owing to this fact that H2S is another important impurity of sour gas in process of natural gas treatment, in continuing of what Vahidi et al. [12] have done, UiO-66-NH2 as a nanoparticle was synthesized and its effect on the solubility of hydrogen sulfide in MDEA solutions was studied. Experimental trials have been carried out using the isochoric saturation / or static method in two concentration grades of 0.1 and 0.5 wt.% of UiO-66-NH2 in the aqueous solution of 40 wt.% MDEA. Solubility measurements were done at temperatures of 303.15 through 333.15 K, and pressures up 1100 kPa.
Materials
The specifications and sources of the chemicals used in this work are listed in Table 1. All chemicals used in this work were reagent grade and used without further purification. Deionized water was obtained via ion exchange resin in RIPI, and its total dissolved substance (TDS) was measured by conductivity method (Mettler Toledo) (TDS < 10 ppm). Deionized water was degassed in an ultrasonic bath (FUNGILAB, model UA10MFD) at a temperature of 353.15 K and wave frequency of 40 kHz about half an hour before use.
Specifications and sources of chemicals used in this work
Specifications and sources of chemicals used in this work
1The total dissolved solid (TDS) was measured by conductivity method (Mettler Toledo).
The solvothermal process described by Katz et al. [39] was employed to prepare UiO-66-NH2 nano-particles. Here a short brief explanation of preparation is provided.
1.25 g (5.4 mmol) ZrCl4, 50 ml of the DMF, and 10 ml concentrated HCl were blended in a 500 cm3 reactor and sonicated to fully dissolved and homogenized. The amount of 1.34 g (7.5 mmol) 2-aminoterephthalic acid and 100 ml of the DMF were added and let the mixture be sonicated and heated at 120°C for about 24 hours. The deposit was filtered and washed with DMF (2×30 mL) and EtOH (2×30 mL) and again was filtered to remove all remaining solvent impurities. The yellow deposit was dried at a temperature of 120°C and low pressure.
Similar to previous work [12], a solution of aqueous 40 wt.% MDEA was considered to study the influence of UiO-66-NH2 addition on H2S absorption capacity. Analytical balance (Mettler model PE 200 with the accuracy of ±0.1 mg) was used to weigh a known amount of UiO-66-NH2 to prepare nanofluid with interest compositions (0.1 and 0.5 wt%). Ultrasonic bath was used for degassing of solutions. The behavior of dispersion systems (particles in liquid) is particularly sensitive to the ionic, and electrical structure of the particle interface. Zeta potential is a parameter that measures the electrochemical equilibrium at the particle-liquid interface. It measures the magnitude of electrostatic repulsion/attraction between particles and thus, it is a key factor in the stability of colloidal particles. Colloids with high zeta potential values (negative or positive) are electrically stable, but colloids with low zeta potentials values tend to coagulate or flocculate. The measured zeta potential for the specific solution of 0.1 wt% UiO-66-NH2 in aqueous 40 wt% MDEA was shown in Fig. 2. Zeta potential distribution is derived by dynamic light scattering that is usually presented with the level of intensity of signal strength which is the scattering intensity, and is measured in photons per second, or most often in kilo counts per second (kcps). As can be seen in Fig. 2, zeta potential distribution shows a maximum intensity value of about 26.6 mV (millivolt), meaning the solution has moderate stability of colloidal dispersions.

Zeta potential diagram of aqueous 40 wt% MDEA + 0.1 wt% UiO-66-NH2 solution.
The isochoric saturation method was employed to measure the gas solubility in the present work experimentally. The main feature of this method is that no liquid phase analysis is required, and only with the variation of gas pressure may one be able to obtain gas absorption. The procedure of the experimental trial is the same as that given in our previous papers [40–45], Fig. 3, but a short review is presented here.

Schematic diagram of absorption set-up system.
The double-wall equilibrium cell was connected to a water recirculation bath (model T 2500 PMT Tamson) with temperature stability within ±0.02 K. The temperature was measured using a model TM-917 Lutron digital thermometer with a 0.01 K resolution equipped with a Pt-100 sensor inserted into the cell. The pressure of equilibrium cell was measured using a model PA-33X KELLER pressure transmitter sensor in the range of (0 to 25) bar, which was accurate to within 0.01% of full scale, and that of the gas container was measured using a Baroli type BD SENSORS digital pressure gauge in the range of (0 to 25) bar, which was accurate to within 0.01% of full scale.
Via solution tank (Fig. 3), a known amount of degassed solvent was introduced into the evacuated equilibrium cell (Vcell ≈131.8 cm3). The water recirculation bath was set at the desired value, and then a known amount of the gaseous solute, H2S, through the gas container was injected into the equilibrium cell. In the equilibrium cell, stirring with magnet stirrer and speed of 300 rpm was continuously performed to increase the mass transfer rate and decrease the time required to reach the equilibrium state. It generally takes the system about 2 hours to reach the equilibration state. The difference between two pVT measurements was used to calculate the amount of gas introduced in the equilibrium cell [46, 47]:
where Vgc is the gas container volume, ρ
i
, ρ
f
are mole density of H2S (mole per volume) corresponding to the temperature of the gas container and initial pressure, pi and final pressure, pf in the gas container before and after transferring the gas. Accordingly, n
i
and n
f
are respectively initial and final mole of H2S in the gas container before and after transferring into the equilibrium cell. The total amount of gas, ntotal, introduced in the equilibrium cell can be added cumulatively. After gas injection into the equilibrium cell, the pressure drop of gas was begun until no pressure drop was observed, meaning that equilibrium state may be established. The amount of H2S in the liquid phase was then determined:
where Pt and PVP denote the total absolute pressure and vapor pressure of the solution. The amount of remaining H2S in the gas phase
where Vg is the gas-phase volume,
where Vunloaded solvent is the volume of the unloaded solvent estimated by the density of aqueous MDEA solution [49].
Finally, using equations 1 through 5, loading of gas in the liquid phase was measured:
in which,
where, wunloadedsolvent is mass of unloaded solvent in kg. The error propagation theory was employed to obtain the uncertainties of results [50]. According to this theory, the overall uncertainty for the measured solubility of H2S can be calculated:
in which n
i
and n
f
are defined in eq 1 and u (n
i
), u (n
f
) and
and u (nMDEA) was accounted by the purity of MDEA reported by the supplier.
The main part of the uncertainty of the solubility is attributed to errors in the pressure sensor for equilibrium cell and gas container (both are equal to u (P k )=±0.0025 MPa), temperature sensors (u (T k )=±0.10 K), and scale for solvent in equilibrium cell (±0.1 mg). The purity of MDEA as reported by the supplier in Table 1 is 99 wt%. Thus, the uncertainty of the mass of MDEA in preparation of solvent is ±0.01. wMDEA g. The volumes of the gas container (107.2±1.3 cm3) and equilibrium cell (131.8±1.5 cm3) were obtained by the pressure swing method [44].
UiO-66-NH2 characterization
The physico-chemical properties of UiO-66-NH2 were studied by different techniques including XRD, TGA, SEM, BET, and FTIR to understand its crystalline structure, morphology, thermal stability, particle size, and porous structure.
XRD analysis
The XRD pattern in Fig. 4A allowed us to confirm that UiO-66-NH2 was synthesized successfully with characteristic peaks of UiO-66 at 2θ values of 7.34° and 8.48°. Figure 4A displayed narrow lines and high crystallinity of the sample and showed good agreement with that reported in the literature [51–54]. In addition, the powder X-ray diffraction pattern of the sample after mixing with water for10 days was shown in Fig. 4(B). As this figure shows, UiO-66-NH2 showed no changes in its crystallin structure after treating with water during 10 days. The highly stable UiO-66 was reported in 2008 [55]. Even if water molecules can come sufficiently close to the metal center, the presence of steric factors can reduce the reaction kinetics by providing a significant activation energy barrier to overcome. Despite the relatively low pKa of carboxylate ligands, the high coordination number of the metal in the well-known Zr-based UiO-66-NH2 MOF helps make this structure highly stable even after adsorbing large amounts of water [56–59] In addition, UiO-66-NH2 due to thermodynamic and kinetic feature is stable in water [60]. Using the XRD pattern, the crystal size is determined as 85 nm.

The powder X-ray diffraction pattern of synthesized UiO-66-NH2, A: as synthesized (before mixing with water), B: after 10 days mixing with water.
The SEM morphology gives us pertinent information about the shape and particle size of the crystalline structure of substances. The picture of the sample of UiO-66-NH2 is shown in Fig. 5. As shown in Fig. 5, the shape of UiO-66-NH2 crystals is spherical, and its average particle size is approximately below 100 nm.

Scanning electron microscopy (SEM) morphology of UiO-66-NH2.
Thermo-gravimetric analysis (TGA) is another tool to receive information about the thermal stability of substances. In Fig. 6, the diagram of weight loss and derivative weight loss of UiO-66-NH2 with temperature was depicted. As shown in Fig. 6. There are three weight loss steps in the TGA curve. The first weight loss is between 303 K and 403 K that is indicative of the vaporization of water. The second weight loss step is at 403–673 K due to dehydroxylation of OH– at 523 K and was completed at 573 K [61]. The third step of weight loss is above 618 K because of the decomposition of the material. The result indicated that UiO-66-NH2 was stable up to 673 K.

TG analysis of UiO-66-NH2.
The IR spectrum of UiO-66-NH2 is shown in Fig. 7. Generally, it looked similar to the spectrum reported by Abid et al. [61] and Luu et al. [38]. The appearance of the absorption band at 1575.66 cm–1 indicated the existence of the reaction of –COOH with Zr4 +. The band of 1497.59 cm–1 is referred to as C = C from aromatic. The carboxyl groups from free aromatic carboxylic acid were observed at 1656.08 cm–1 [62]. The peak appearing at 3430.08 cm–1 is NH2 on the organic linker, referring to symmetric and asymmetric vibrations of NH2 groups. Low intensities of these peaks may be attributed to the strong bonding between the amino groups in the coordinated acid, with C = O groups of free NH2–BDC inside the pores and bridging OH groups in the metal center (bridging OH groups can interact with amino groups on the organic linker by hydrogen bonding).

FTIR spectrom of UiO-66-NH2.
Synthesized UiO-66-NH2 showed high surface area up to 842 m2 g–1, micropore volume of 0.340 cm3 g–1, pore with 19.0871 Å, the average particle size of 71.249 Å, and thermal stability up to 618 K. Figure 8 shows the pore size distribution of UiO-66-NH2. This figure represents that 62.5% of pore width is in the range of 0–5 nm.

Pore size distribution of UiO-66-NH2.
The solubility of H2S in the aqueous solution of 40 wt.% MDEA, 40 wt% MDEA + 0.1 wt% UiO-66-NH2 and 40 wt% MDEA + 0.5 wt% UiO-66-NH2 have been measured and listed in Tables 2, 3 and 4 respectively. Experimental data graphically were shown and compared with each other for a typical temperature of T = 313.15 K in Fig. 9. As can be seen from Fig. 9, addition of UiO-66-NH2 in MDEA aqueous solution as a nano-fluid has no significant effect on H2S solubility (less than 3%); in fact, all depicted graphs are within their uncertainty limits. In Fig. 10, using data reported by Vahidi et al. [12] for the solubility of CO2 in 40 wt.% MDEA and 40 wt% MDEA + 0.1 wt% UiO-66-NH2, the solubility behavior of CO2 in two solvents at T = 313.15 K were depicted and compared. As can be seen, the positive effect of UiO-66-NH2 on CO2 absorption is significant. By more inspection of CO2 solubility in both solvents, the loading of CO2 in nano-fluid at three typical pressures of 0.735, 1.52 and 2.20 MPa are respectively equal 1.059, 1.166 and 1.212, as opposed to values of 0.978, 1.058 and 1.107 in 40 wt% MDEA showing an absorption rising of 8.5% , 10.2% , and 9.5% respectively.
Solubility data of H2S in 40 wt% MDEA solution
Solubility data of H2S in 40 wt% MDEA solution
T is temperature in K, PH2S is partial pressure of H2S in MPa, Pt is total pressure in MPa, m is molality of H2S in solvent in mole per kg., ±Δm is uncertainty of molality and αH2S is loading of H2S in solvent (mol H2S/mol MDEA).
Solubility data of H2S in 40 wt% MDEA + 0.1 wt% UiO-66-NH2 mixtures
T is temperature in K, PH2S is partial pressure of H2S in MPa, Pt is total pressure in MPa, m is molality of H2S in solvent in mole per kg., ±Δm is uncertainty of molality and αH2S is loading of H2S in solvent (mol H2S/mol MDEA).
Solubility data of H2S in 40 wt% MDEA + 0.5 wt% UiO-66-NH2 mixture
T is temperature in K, PH2S is partial pressure of H2S in MPa, Pt is total pressure in MPa, m is molality of H2S in solvent in mole per kg., ±Δm is uncertainty of molality and αH2S is loading of H2S in solvent (mol H2S/mol MDEA).

Comparison of solubility (loading equal mol H2S/mol MDEA) data of H2S in 40 wt% MDEA, 40% wt MDEA + 0.1% wt NP and 40% wt MDEA + 0.5% wt UiO-66-NH2 nano-fluid at a typical temperatures (313.15) K.

Comparison of solubility (loading equal mol CO2/mol MDEA) data of CO2 in 40 wt% MDEA and 40% wt MDEA + 0.1% wt UiO-66-NH2 nano-fluid at T = 313.15 K (data are from reference [6]).
In this work, UiO-66-NH2 was synthesized with the solvothermal method and analyzed with XRD, SEM, BET, TGA, and IR spectroscopy. Because of the presence of the NH2 group in UiO-66-NH2 and its interaction with water and MDEA, the UiO-66-NH2 nanoparticle is stable in the solution of MDEA according to the Zeta potential of the mixture. The effects of the presence of UiO-66-NH2 nanofluid on the solubility of H2S in aqueous 40 wt% MDEA solutions were studied. All experiments have been carried out in two concentration grades of 0.1 and 0.5 wt% of UiO-66-NH2 in the aqueous solution of 40 wt% MDEA and at temperatures of 303.15 through 333.15 K and pressures up 1100 kPa.
Results show that the addition of UiO-66-NH2 nanoparticles to the MDEA solution has a negligible effect on the solubility of H2S. A nearer prospect on the results reveals that difference between solubility of H2S in aqueous 40wt% MDEA and aqueous 40wt% MDEA + 0.1wt% nano-particle at the same temperature and H2S partial pressure is lower than the average uncertainty value reported in Tables 2–5 (∼ 4.0%), meaning that it does not have remarkable effect on H2S solubility, however as shown in the previous section and reported by Vahidi et al.[12], the addition of UiO-66-NH2 nanoparticles into the aqueous 40 wt% MDEA solutions increases the capacity of CO2 absorption up to about 10% , and owing to this fact; they reported their experimental solubility with an average uncertainty of 5.5% , the effect of nanoparticles on solubility would be meaningful.
Footnotes
Acknowledgment
We are thankful to the Research Council of the Research Institute of Petroleum Industry (RIPI) and the Research and Development of the National Iranian Oil Company (NIOC) for their support of this work.
