Contents lists available at ScienceDirect

Microporous and Mesoporous Materials

journal homepage: www.elsevier.com/locate/micromeso

Tuning 1-hexene/n-hexane adsorption on MOF-74 via constructing Co-Mgbimetallic frameworks

Hui Suna,b,∗, Danni Rena, Ruiqi Konga, Dan Wanga, Hao Jianga, Jialun Tana, Di Wuc,d,e,Shengwei Chena, Benxian Shena

a Petroleum Processing Research Center, East China University of Science and Technology, Shanghai, 200237, Chinab International Joint Research Center of Green Energy Chemical Engineering, East China University of Science and Technology, Shanghai 200237, Chinac The Gene and Linda Voiland School of Chemical Engineering and Bioengineering, Washington State University, Pullman, WA, 99163, United StatesdDepartment of Chemistry, Washington State University, Pullman, WA, 99163, United StateseMaterials Science and Engineering, Washington State University, Pullman, WA, 99163, United States

A R T I C L E I N F O

Keywords:1-Hexene/n-hexane separationAdsorptionSelectivityMOF-74Bimetallic

A B S T R A C T

A series of bimetallic CoxMg1-x-MOF-74 (x=0, 0.12, 0.30, 0.57, 0.78, 1) structures were synthesized by usingfacile solvothermal systems having different binary metal compositions. All MOF samples were characterizedcarefully. Both adsorption affinity to n-hexane and 1-hexene and adsorption selectivity of 1-hexene over n-hexane were evaluated by static adsorption test coupled with computer simulation. It is indicated that theCoeMg bimetallic MOF-74 frameworks can be successfully constructed via using well-designed reactant com-positions. All bimetallic MOF-74 retain almost the same crystalline structure but higher porosity as compared totheir monometallic samples. In addition, both of the adsorption capacity and selectivity of CoeMg bimetallicMOF-74 samples are found strongly dependent on their metal compositions. Specifically, Co0.30Mg0.70-MOF-74exhibits the largest 1-hexene adsorption capacity of 152.7mg/g and the highest 1-hexene/n-hexane selectivity of9.74, which are 2.1–2.3 and 4.1–8.9 times higher than those of monometallic Coe or Mg-MOF-74 samples. Suchimprovement on adsorption ability and selectivity for olefin molecules can be attributed to the incorporation ofmore stable coordinatively unsaturated sites (CUS) within Mg-MOF-74 framework, leading to the synergeticeffect of pore structure evolution as well as higher density of CUS metal sites in bimetallic frameworks. Thetunable adsorption affinity to olefin/paraffin on MOF-74 by means of metal modification provides an approachto the efficient separation of olefins and paraffins from liquid hydrocarbon mixtures.

1. Introduction

Paraffins and olefins are commonly coexisting products from anumber of industrial processes such as steam cracking of hydrocarbons[1–4] and Fischer-Tropsch synthesis [5,6]. Economic and effective se-paration of paraffin/olefin containing streams is, therefore, realizedvery important to accomplish the industrial processes. Cryogenic dis-tillation is a traditional method for the separation of paraffin/olefinfrom light hydrocarbons. However, it is very energy-intensive and,therefore, a variety of energy consumption-reduced separation tech-nologies including adsorption separation [7,8], membrane separation[9–11] and liquid-liquid extraction [12,13] receive increasing interests.Among these methods, adsorption separation using high-performanceadsorbent is considered to be an effective method. Different adsorbentsinvolving molecular sieves [14,15], activated carbon [16–18], metal-

organic frameworks (MOFs) [19–21] have already been extensivelystudied and used for the separation of paraffin/olefin mixtures.

MOFs are a type of porous crystalline materials having periodicframework structures [22], which are composed of metal or metal oxideclusters and organic ligands. Owing to their orderly internal micro-channels, extremely high surface areas and highly adjustable pore di-mensions [22], MOFs have been realized to be promising functionalporous materials and are receiving increasing research and industrialinterests in various fields involving gas storage [23,24], separation[25,26], and catalysis [27,28] etc.

A variety of metal-organic framework structures [29–32] are con-sidered as adsorbents in the separation of paraffin/olefin mixtures.Specifically, a group of isostructural materials known as M-MOF-74 (Mrepresents metal), or M(dhtp)2 are characterized to be of remarkabletextural properties and accessible unsaturated metal sites for hosting

https://doi.org/10.1016/j.micromeso.2019.04.031Received 19 January 2019; Received in revised form 4 April 2019; Accepted 16 April 2019

∗ Corresponding author. Petroleum Processing Research Center, East China University of Science and Technology, Shanghai, 200237, China.E-mail address: sunhui@ecust.edu.cn (H. Sun).

Microporous and Mesoporous Materials 284 (2019) 151–160

Available online 19 April 20191387-1811/ © 2019 Elsevier Inc. All rights reserved.

T

hydrocarbon molecules. Their versatility can be achieved via introdu-cing different metal ions or mixture of them [33,34] at any extension.Furthermore, their stability under ambient conditions and in presenceof water can also be improved by using mixed metal species in the sameframeworks. As a result, these materials with open metal sites showdistinguished performance in adsorption and separation of gases havinghigh environmental and energy importance.

Botas et al. [35] synthesized ZneCo mixed MOF-74 via one-potsynthesis while Kahr et al. [36] used a postsynthetic incorporationmethod to form MgeNi bimetallic MOF-74 structure. It is reported thatWang et al. [37] conducted successful synthesis of MOF-74 containingas many as 10 different divalent metals (i.e. Mg, Ca, Sr, Ba, Mn, Fe, Co,Ni, Zn, and Cd) in a single framework structure. Previous studies in-dicate that MOF-74 with the expected topology can be synthesized viaincorporating different metals. The coexistence of multiple metals cannot only increase the complexity of the MOF structures, but also extendthe unique properties and applications brought about by the multi-metal ions in corresponding frameworks. The underlying effect of bi- ormulti-metal coexistence [22,38] on the structure-activity relationship ofmaterials is very important to their applications in diverse fields. On theother hand, Fischer-Tropsch synthetic liquid products containing par-affins and olefins with carbon number of 5–25 are valuable resources,whereas they are used as low-value fuels at present. A successful se-paration of paraffins and olefins from their liquid mixtures can be be-lieved to increase the utilization value of the liquid products of Fischer-Tropsch process largely.

In this study, a series of CoeMg bimetallic MOF-74 were preparedvia one-step solvothermal synthesis. Metal occupation and frameworkstructures of MOF-74 were finely tuned through using well-designedreactant compositions in order to achieve the effective separation ofolefin/paraffin from liquid mixtures. All synthesized samples werecarefully characterized by applying powder X-ray diffraction (XRD), N2

adsorption, thermogravimetric analysis and differential scanning ca-lorimetry (TG-DSC), scanning electron microscopy (SEM), energy-dis-persive spectrometer (EDS), Fourier transform infrared spectroscopy(FTIR) and X-ray photoelectron spectroscopy (XPS). In addition, theiradsorption performance was evaluated by using static adsorption testcoupled with computation simulation using 1-hexene/n-hexane modelcompounds. The influence of CoeMg composition in MOF-74 frame-works on 1-hexene/n-hexane adsorption on resulting adsorbents washighlighted.

2. Materials and methods

2.1. Chemicals

Co(NO3)2∙6H2O (with>99.0% purity), Ni(NO3)2∙6H2O (with>99.0% purity), Mg(NO3)2∙6H2O (with>99.0% purity), Zn(NO3)2∙6H2O(with> 99.0% purity), 2,5-dihydroxyterphthalic acid (DHTP) (with>98.0% purity), N, N-dimethylformamide (DMF) (with>99.5% purity),ethanol (with>99.5% purity), n-hexane (with> 99.0% purity) andcyclohexane (with>99.0% purity) were provided by Shanghai TitanScientific Co., Ltd (Shanghai, China). 1-Hexene (with> 99.0% purity)was provided by Beijing Huawei Ruike Chemical Industry Co., Ltd(Beijing, China). Deionized water was used in all cases. All the chemi-cals in this study were used as received.

2.2. Preparation of samples

2.2.1. Co-MOF-74 synthesisCo-MOF-74 samples were synthesized following a previously re-

ported [39–41], but slightly modified solvothermal procedure. 5.24 gCo(NO3)2∙6H2O (18mmol) and 1.19 g 2,5-dihydroxyterphthalic acid(DHTP)(6 mmol) were dissolved in a mixed solvent of DMF, ethanol andH2O (having total volume of 515.7 mL and volume ratioVDMF:Vethanol:VH2O of 15:1:1). The mixture was transferred into a teflon-

lined stainless steel autoclave and placed at 100 °C in an oven for 24 h.Then the autoclave was cooled down and the solid was separated bycentrifugation. After that, the resulting sample was kept in methanol for24 h. Finally, the product was dried at 150 °C under vacuum for 2 h,until the sample indicating a dark red color.

2.2.2. Mg-MOF-74 synthesis4.62 gMg(NO3)2∙6H2O (18mmol) and 1.19 g DHTP (6mmol) were

dissolved in the same mixed solvent to Co-MOF-74 synthesis. The nextsteps are the same to that of Co-MOF-74 synthesis. As the procedure iscompleted, a brown sample can be obtained.

2.2.3. CoxMg1-x-MOF-74 synthesisDifferent amounts of Mg(NO3)2∙6H2O and Co(NO3)2∙6H2O dissolved

together with 1.19 g DHTP (6mmol) in the same mixed solvent of DMF,ethanol and H2O. The next steps are also the same. Bimetallic MOF-74samples with Co and Mg were synthesized and labeled as CoxMg1-x-MOF-74, in which x represents the observed metal proportion in eachsample.

2.3. Computational simulation

All simulations were run using an Accelrys Material StudioSimulation Package. The starting unit cells of Co-MOF-74 and Mg-MOF-74 were taken from the Cambridge Structural Database (CSD). Based ona Co-MOF-74 lattice framework, we randomly replaced Co species withMg to produce a series of CoMg-MOF-74 frameworks having the Co toMg molar ratios consistent with experimentally determined values. Thefractional atomic positions were fixed during the simulation to restrictthe adsorption to a specific region. DFT calculations were employed tooptimize the structure of CoMg-MOF-74 using the DMol3 module in theMaterials Studio package. DNP basis set was adopted and GGA-PBE-Dexchange-correlation function with correction using Grimme methodwas chosen [42]. Then Grand Canonical Monte Carlo (GCMC) compu-tation was conducted to predict the adsorption capacities. In everyGCMC simulation, the temperature (T), volume (V) and chemical po-tential (μ) of the system were kept constant, while the total number ofmolecules (N) fluctuated. For each calculation point, the system wasequilibrated during 5,000,000 steps, followed by 5,000,000 samplingsteps for data collection. In order to compare the simulated results withexperimental isotherm data, the former would be converted from ab-solute values to excess adsorption amounts.

2.4. Characterization

Powder X-ray diffraction (XRD) patterns were collected at roomtemperature on a Rigaku D/max 2550 diffractometer (Rigaku IndustrialCorporation, Japan) using Cu Kα radiation operated at 18 kW and450mA. The 2θ scanning angle range was 3–80° with step sizes of 0.02and 5 deg·min−1.

The morphologies of samples were scanned by using a NOVA NanoSEM450 scanning electron microscope (SEM; FEI, USA) with a beamcurrent of 10 nA and an accelerating voltage of 15 kV. Element analyseswere carried out on a Falcon energy-dispersive spectrometer (EDS;EDAX Inc., USA). A sputter coating with a thin layer of platinum wasperformed to avoid charging. Homogeneity was checked using back-scattered electron (BSE) imaging.

The pore structures of all samples were determined by N2 adsorp-tion at 77 K using a 3H-2000PM2 automatic physisorption analyzer(BeiShiDe Instrument Co., Ltd., China). The specific surface area wascalculated using the Brunauer-Emmett-Teller (BET) method. Both themicropore volume and average micropore diameter were calculatedusing the Horvath-Kawazoe (HeK) method, respectively.

Fourier transform infrared spectroscopic (FT-IR) analyses wereperformed on a Nicolet 6700 FT-IR spectrometer (Thermo FisherScientific, USA) using a KBr tablet. Each spectrum was obtained from

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

152

the acquisition of 105 scans from 4000 to 400 cm−1 with a resolution of2 cm−1.

X-ray photoelectron spectroscopy (XPS) measurements were con-ducted with a Thermo Scientific ESCALAB 250Xi spectrometer (ThermoFisher Scientific, USA) equipped with Al Kα radiation as the X-raysource (14 kV, 250W). XPS spectra of the samples were corrected onthe basis of the reference of carbon (C 1s, 284.6 eV).

Thermogravimetry and differential thermal scanning calorimetry(TG-DSC) analyses were carried out on a STA 409 PC/PG analyzer(NETZSCH, Germany). Synthesized sample was placed in a platinumcrucible and heated from room temperature to 800 °C at 10 °C·min−1

under the air flow (20mLmin−1).ICP-AES analysis was performed at a Varian 710-ES full-spectrum

direct reading plasma emission spectrometer (Varian, Inc., USA) withan analysis wavelength of 177–785 nm and an optical resolution of0.009 nm. Prior to measurement, MOF-74 sample was dissolved thor-oughly into a concentrated nitric acid solution under heating at 150 °C.

2.5. Adsorption experiment

The binary solutions of 1-hexene/cyclohexane and n-hexane/cy-clohexane with 1-hexene and n-hexane mass fraction of both about 5%were prepared to evaluate the adsorption performance of synthesizedMOF-74 samples.

All adsorption measurements were performed at 25 °C, a preciselyweighed 0.5 g sample was added into a 25mL beakerflask containing7.5 g previously prepared binary solution and then the beakerflask wassealed well. Hereafter, the beakerflask was placed in a constant tem-perature shaker. The concentrations of 1-hexene or n-hexane in thesolution were determined at intervals. The adsorption capacity Q(mg·g−1) can be calculated according to Q=m1(c0 - c)/m2× 103 untilthe adsorption amount reached a constant value, where m1 and m2 arethe masses of the solution and MOF-74 sample (g), respectively; c0 and care the initial and equilibrated mass fractions of 1-hexene or n-hexanein the solution (%). Compositions of binary solutions were examined byusing a GC-920 gas chromatography (Shanghai Haixin ChromatographyInstrument Co., Ltd., China) equipped with a flame ionization detectorand a high elasticity quartz capillary column (having a fixed phase ofOV-101, length of 50m and inner diameter of 0.2mm). In chromato-graphy analysis, we used the external standard compounds of 1-hexeneand n-hexane in order to determine the concentrations of 1-hexene andn-hexane in binary solutions.

3. Results and discussion

3.1. Sample characterization

3.1.1. XRDPowder XRD analyses were carried out on synthetic MOF-74 sam-

ples in order to confirm their crystalline phases. XRD patterns of Mg-MOF-74, Co-MOF-74 and a series of bimetallic CoMg-MOF-74 samplesare illustrated in Fig. 1. Based on their chemical compositions, the si-mulated XRD patterns are also given (see gray curves in Fig. 1). Bycomparing the observed patterns with the established crystal structuredata [20,43,44], all the synthesized samples are of single phase, theexpected crystalline structure of MOF-74.

3.1.2. XPSWe firstly performed XPS analyses on all synthesized MOF-74

samples to figure out their metal compositions (see Fig. 2 (a) for Mg andFig. 2 (b) for Co species). By integrating the XPS peaks of Mg and Cospecies, metal compositions for all samples were determined and listedin Table 1. Three MOF-74 samples having relatively high Mg contents(i.e. Co0.12Mg0.88-MOF-74, Co0.30Mg0.70-MOF-74 and Co0.57Mg0.43-MOF-74) display the same binding energy at 1304.6 eV, correspondingto Mg 1s. As for the Co0.78Mg0.22-MOF-74 sample, the binding energy

for Mg 1s shifts to a slightly higher value of 1304.9 eV. Meanwhile,these samples also show slightly shifting orbital peaks of Co 2p1/2 at797.5–797.9 eV and Co 2p3/2 at 781.7–782.0 eV with Co/Mg molarratio increasing. The peaks located at 802.5–802.9 eV and785.8–786.3 eV are the satellite peaks related to Co 2p1/2 and Co 2p3/2, respectively [45,46]. In order to confirm the analysis results of XPSand figure out the average metal composition of each bimetallicstructure, we performed the ICP-AES analyses on bimetallic CoMg-MOF-74 samples. All analysis results are listed in Table 1. The ICP-AESanalyses agree well with XPS results, confirming the reliable metalcomposition for each sample. As shown in Table 1, the CoMg-MOF-74samples show consistent changes on the Co to Mg molar ratios withtheir initial ratios used for solvothermal synthesis.

3.1.3. SEMThe morphologies of the synthesized MOF-74 samples were ob-

served from SEM images (see Fig. 3). It seems that the synthesizedCoeMg coordinated MOF-74 samples show significantly distinct crystalmorphologies. Specifically, it is indicated that the Mg-MOF-74 crystalexhibits petal-like morphology (see Fig. 3 (a)), which was demonstratedin previous publication [36] and the crystal size is found to be 5–10 μm.Meanwhile, the Co-MOF-74 crystal has a spherical structure formed ofrodlike crystal branches (see Fig. 3 (f)) [40] and the average size ofaround 8 μm. In cases of the CoxMg1-x-MOF-74 samples, their trans-formed morphologies evolves from petal-like structure (see Fig. 3 (b)for Co0.12Mg0.88-MOF-74) to hexagonal prism-like structure (see Fig. 3(c) and 3 (d) for Co0.30Mg0.70-MOF-74 and Co0.57Mg0.43-MOF-74) andthen a spherical structure (see Fig. 3 (e)) for Co0.78Mg0.22-MOF-74).Along with the increase of Co species in MOF-74 frameworks, the petal-like structures tend to disassemble into a number of hexagonal prism-shaped crystals and then evolve to spherical structures. In addition,these samples show increasing crystal dimension with Co/Mg molar

Fig. 1. Powder X-ray diffraction patterns of synthesized MOF-74 samples. (Thecurves in gray are the corresponding simulation results.)

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

153

ratio. As for the sample of Co0.78Mg0.22-MOF-74 with relatively highCo/Mg molar ratio (nCo/nMg= 3.5), the average particle size is up toalmost 13 μm.

3.1.4. EDSIn order to confirm the metal composition in each MOF-74 struc-

ture, we also conducted EDS analyses on all bimetallic CoMg-MOF-74samples (see Fig. 4). The observed Co/Mg molar ratios of CoMg-MOF-74 samples (e.g. 0.165 for Co0.12Mg0.88-MOF-74) agree with afore-mentioned XPS results (0.136 for Co0.12Mg0.88-MOF-74). Additionally,according to the EDS results, every sample is found to possess uniformcomposition of Mg and Co species, suggesting that the resulting bime-tallic MOF-74 samples are not mechanically composed of monometalliccrystals of Mg-MOF-74 and Co-MOF-74. Current results indicate thatMg and Co coexist in CoMg-MOF-74 samples. The Mg to Co ratio inresulting MOF structures can be expected to increase with increasingMg to Co ratio in synthesis system.

3.1.5. FT-IRTo further confirm the structural evolution of these CoeMg coex-

isting MOF-74 samples, FT-IR analyses were also been carried out (seeFig. 5). In FT-IR spectrum of Mg-MOF-74 sample, the band near3385 cm−1 is assigned to the OeH stretching vibration of hydroxylgroup, which can be attributed to bulk water absorbed in the porousmaterials. A sharp peak at above 3500 cm−1, assigning to the vibrationof OH band of terminal H2O on metal sites can be recognized in Mg-MOF-74 and bimetallic samples. And the absorption peak shifts from3744 cm−1 for Mg-MOF-74 sample to 3738 cm−1 for the bimetallicCoxMg1-x-MOF-74, indicating that H2O species are located at differentmetal sites of mono- and bimetallic MOF-74 structures [47]. The bandat 1580 cm−1 is ascribed to the C]O of benzene ring in the organicligand DHTP. The bands at 1460 and 1413 cm−1 are attributable to theC]C vibration of the benzene ring skeleton, also indicating the organicligand in the resulting structures. A weak absorption peak at 1236 cm−1

is attributed to the stretching vibration of CeO band of phenolategroup. The band at 1208 cm−1 corresponds to beta eCeH stretching

Fig. 2. XPS results of synthesized CoMg-MOF-74. (a) XPS for Mg, (b) XPS for Co.

Table 1Chemical compositions of MOF-74 samples.

Sample Experimental n(Mg)/n(Co)a XPS-determined n(Mg)/n(Co) ICP-AES-determined n(Mg)/n(Co) Chemical Compositionb Structure model

Mg-MOF-74 / / / Mg(dhtp)2

Co0.12Mg0.88-MOF-74 9 7.33 7.64 Co0.12Mg0.88(dhtp)2

Co0.30Mg0.70-MOF-74 2.33 2.33 1.98 Co0.30Mg0.70(dhtp)2

Co0.57Mg0.43-MOF-74 1 0.75 0.64 Co0.57Mg0.43(dhtp)2

Co0.78Mg0.22-MOF-74 0.43 0.28 0.28 Co0.78Mg0.22(dhtp)2

Co-MOF-74 / / / Co(dhtp)2

a Metal compositions based on molar fraction.b According to XPS analysis.

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

154

Fig. 3. SEM images of synthesized MOF-74 samples. (a) Mg-MOF-74, (b) Co0.12Mg0.88-MOF-74, (c) Co0.30Mg0.70-MOF-74, (d) Co0.57Mg0.43-MOF-74, (e) Co0.78Mg0.22-MOF-74, (f) Co-MOF-74.

Fig. 4. EDS results of Mg and Co species of synthesized CoMg -MOF-74 samples.

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

155

vibration and the absorption peaks at 889 and 819 cm−1 can be at-tributed to the CeH bending vibration of benzene ring. Additionally,two absorption events appearing at 586 and 480 cm−1 correspond toMgeO vibration. As for the CoMg-MOF-74 samples, the M−O(M=Co or Mg) vibration can also be recognized at 586 and 476 cm−1.A feature peak at 1629 cm−1 corresponding to v(C]O) [48] exists inthe spectra of CoMg-MOF-74 samples but disappears in the spectrum ofmonometallic Mg-MOF-74. In a recent research, the same v(C]O) peakwas also found in FT-IR spectrum of Co-MOF-74, whereas it was absentfor Mn-MOF-74 [39]. It seems that our results agree well with previousfindings. Once Co species account for more than 30% of total metalspecies, the peak at 1580 cm−1 in spectrum of Mg-MOF-74 shifts to1550 cm−1. It is attributed to the conjugated effect of bimetallic MOF-74. When the C]O group conjugates with the C]C bond, the deloca-lization of the π electrons occurs between the two unsaturated bands.Then the double bond characteristics of C]O will be reduced, resultingin the shift of absorption frequency towards lower wave number. As aresult, strongly-coordinated olefin molecules to the metal centers inMOF-74 frameworks can be expected.

3.1.6. TG-DSCThermogravimetry and differential scanning calorimetry (TG-DSC)

analyses were conducted in order to examine the thermal stability ofsynthetic samples and understand their structural evolution on heating.All TG curves show two main mass loss events happening in varyingtemperature ranges for different samples (see Fig. 6 (a)). The first massloss event is found to start once heating and finish until the secondevent happens at different temperatures. For Co-MOF-74 sample, thesecond mass loss event starts at around 230 °C and finishes at 340 °C. Asfor Mg-MOF-74 sample, the starting and finishing temperatures for thesecond mass loss are found to increase up to 430 and 490 °C, respec-tively. Meanwhile, all bimetallic CoMg-MOF-74 samples show the

second event temperatures between those of Co-MOF-74 and Mg-MOF-74. The first mass loss event can be attributed to the removal of bulkspecies such as adsorbed water, residual methanol or DMF solvent insynthetic products. Water and methanol can be removed at relativelylow temperature while DMF removal requires a relatively high tem-perature due to its higher boiling point. The mass loss for the first eventof all MOF-74 samples shows very consistent values with the replace-ment of Mg with Co (ranging from 38% for Mg-MOF-74 to 36% for Co-MOF-74). Additionally, the second mass loss corresponds to the de-composition of the framework, exactly oxidation of organic ligands inair atmosphere. With the increase of Co replacement, the metal organicframeworks collapse at decreased temperatures (see the exothermicpeaks at 461 °C for Mg-MOF-74 and at 260 °C for Co-MOF-74 in Fig. 6(b)), indicating the metal-derived thermal stability reduction. All bi-metallic CoMg-MOF-74 samples have the collapse temperatures be-tween those of two monometallic MOF-74 structures. The significantdifference in thermal stability among these CoMg-MOF-74 architecturesmay be ascribed to the distinct metal configurational effects. As a result,it can be concluded that replacing Co with Mg can benefit the thermalstability of MOF-74 and help maintain a more stable framework onheating. To avoid possible structural decomposition, thermal treat-ments during whole synthesis process were strictly performed at nohigher than 150 °C.

3.1.7. Pore structure analysisWe further explore the influence of the metal compositions on the

pore structures of synthesized materials. Nitrogen adsorption isotherms

Fig. 5. FT-IR spectra of synthesized MOF-74.

Fig. 6. TG-DSC results of synthesized MOF-74 under air atmosphere. (a) TG curves, (b) Heat flow curves.

Fig. 7. N2 adsorption isotherms on synthesized MOF-74 samples.

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

156

at 77 K for all evacuated MOF-74 samples are typical type I (see Fig. 7).The surface area of Co0.30Mg0.70-MOF-74 calculated according toLangmuir method is about 15.7% higher than that of Mg-MOF-74(1265m2·g−1 vs. 1093m2·g−1 in Table 2). Moreover, the Co0.30Mg0.70-MOF-74 sample also gives the largest micropore volume of0.45 cm3·g−1. The increase in specific surface area of CoeMg bimetallicMOF-74 can be attributed to the slight difference in cation radius andcoordination modes between Mg and Co cations. When constructinginfinite structural units, bimetallic structures contribute increasingporosity and surface area. Present results are consistent with thoseexpected, because heavier transition metal substitution leads to an in-crease in surface area [41]. And the values for parent MOF-74 are

consistent with previous findings [38]. All MOF-74 samples are con-firmed to have a microporous structure with comparable micropore sizeof around 0.65 nm and 1.30 nm (see Fig. 8 for the pore size distribu-tion).

3.2. Effect of metal composition on olefin/paraffin adsorption

To evaluate the effect of metal composition on adsorption perfor-mance, the static adsorption of 1-hexene and n-hexane on differentMOF-74 samples was tested at 298 K using 1-hexene/cyclohexane andn-hexane/cyclohexane binary solutions. The adsorption capacities for1-hexene and n-hexane and selectivity of 1-hexene over n-hexane are

Table 2Pore structure analyses of MOF-74 samples.

Sample BET specific surface area, m2·g−1 Langmuir specific surface area, m2·g−1 Micropore volume (HeK), cm3·g−1 Average pore size (HeK), nm

Mg-MOF-74 940 1093 0.37 0.66Co0.12Mg0.88-MOF-74 1094 1174 0.43 0.66Co0.30Mg0.70-MOF-74 1055 1265 0.45 0.67Co0.57Mg0.43-MOF-74 1135 1220 0.43 0.66Co0.78Mg0.22-MOF-74 1016 1214 0.42 0.63Co-MOF-74 1057 1209 0.42 0.66

Fig. 8. Pore size distributions of MOF-74 samples. (a)∼ 0.65 nm and (b)∼ 1.30 nm.

Fig. 9. (a) Adsorption capacities of 1-hexene and n-hexane on MOF-74 samples, (b) 1-hexene/n-hexane adsorption selectivity on MOF-74 samples.

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

157

presented in Fig. 9. All bimetallic MOF-74 samples show closer affinityto 1-hexene but lower adsorption ability for n-hexane than bothmonometallic samples. Specifically, the highest adsorption capacity of152.7 mg/g for 1-hexene (2.1–2.3 times higher that of Coe or Mg-MOF-74 samples) can be obtained at the Co0.30Mg0.70-MOF-74 sample (seeFig. 9 (a)). As a result, the Co0.30Mg0.70-MOF-74 sample provides thehighest 1-hexene/n-hexane selectivity of 9.74, which is 4.1–8.9 timehigher than that of Mg or Co occupied MOF-74 framework. The resultsare supported by previous N2 adsorption measurements. In comparison,Li et al. [49] reported the extractive separation of a 1-hexene/n-hexanemixture based various ionic liquids. The extraction selectivity of 1-hexene to n-hexane was in the range 1.5–3.0. Wentink et al. [50] at-tempted to separate 1-hexene from other C6 isomers by Ag-DBPA andobtained the highest 1-hexene/n-hexane selectivity of 1.42 with a sol-vent-to-feed ratio of 3. Nunez et al. [51] found that Au-PCM-10 couldexhibit a 1-hexene/n-hexane adsorption selectivity of 4.5 while theadsorption capacities for 1-hexene and n-hexane were 1.30mmol/g and0.28mmol/g, respectively. All CoMg-MOF-74 samples show the

expected strong affinity for 1-hexene molecules, illustrating that CoMg-MOF-74 samples possess a larger pore void as well as higher density ofcoordinatively unsaturated sites (CUS) after incorporating mixed metalsinto the MOF-74 structure to host more olefin molecules. Earlier re-searches have proven that the porosity and CO2 capacity of Mg-MOF-74can be enhanced through the incorporation of more stable CUS metalsites (Ni or Co) within the MOF-74 framework [38].

The adsorption isotherms of 1-hexene on various MOF-74 sampleswere also been measured using a vapor sorption analyzer in order tounderstand the adsorption thermodynamic characteristics of olefin

Fig. 10. 1-Hexene adsorption isotherms on MOF-74 samples.

Fig. 11. Simulation results of 1-hexene adsorption on MOF-74.

Fig. 12. The experimental and simulated adsorption capacities of 1-hexene onMOF-74 samples.

Fig. 13. Adsorption energy of 1-hexene on MOF-74 samples.

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

158

compounds on resulting MOF-74 structures. 1-Hexene shows type Iadsorption isotherms on all MOF-74 samples (see Fig. 10). As the in-crease of relative pressure, the equilibrium gas adsorption amount isalso increased. It can be seen that the adsorption capacity of bimetallicCoMg-MOF-74 for 1-hexene is larger than that of single metal MOF-74.In case of Co0.30Mg0.70-MOF-74 sample, the equilibrium adsorptionamount for 1-hexene under relative pressure P/P0 of 0.951 is de-termined to be 155.6 mg/g, which is the maximum and is agreeablewith the saturated adsorption capacity of 1-hexene from uptake resultsin binary solution. The corresponding equilibrium adsorption amountsof Co0.57Mg0.43-MOF-74, Co0.78Mg0.22-MOF-74 and Co0.12Mg0.88-MOF-74 are 143.1, 135.4 and 106.5 mg/g, respectively. By comparing theadsorption isotherms among MOF-74 samples with different metalcompositions, it can be found that the adsorption capacities of 1-hexeneamong six samples are in the same order to aforementioned results fromliquid phase adsorption tests (see Fig. 9 (a) for the adsorption capacityfor 1-hexene). The work of Geier et al. has shown that Co-MOF-74 hashigher adsorption ability for several light hydrocarbons on a basis ofper M atom compared to Mg-substituted MOF-74 structure [52]. Alladsorption measurements jointly indicate that the bimetallic CoMg-MOF-74 having well-controlled metal compositions show largely en-hanced 1-hexene adsorption selectivity over n-hexane, and can bepromising materials for effective selective adsorption separation ofolefins from paraffin/olefin liquid mixtures.

To further evaluate the effect of mixed metal sites on 1-hexeneadsorption, we carried out the computation simulation of 1-hexeneadsorption on six MOF-74 structures (see Fig. 11 for the simulationresults) having samples presented in Table 1. Considering the very lowadsorption capacity of n-hexane on bimetallic CoMg-MOF-74 series (seeFig. 9-a), the performance of synthetic samples for 1-hexene/n-hexaneseparation can be mainly determined by the adsorption affinity to 1-hexene. Therefore, the evaluation on adsorption performance wasmainly focused on the adsorption of 1-hexene on various samples. Ineach GCMC simulation, the metal species were randomly distributed.According to the analysis results (metal composition determined fromXPS and ICP-AES in Table 1), we defined Co to Mg molar ratios as 2:16,6:12, 10:8 and 14:4 for the synthetic bimetallic samples of Co0.12Mg0.88-MOF-74, Co0.30Mg0.70-MOF-74, Co0.57Mg0.43-MOF-74 and Co0.78Mg0.22-MOF-74, respectively.

The experimental and simulated adsorption capacities for 1-hexeneon CoMg-MOF-74 samples are collected and presented in Fig. 12. Thesimulation results show consistent fluctuation with experimental valueswhen changing metal compositions of MOF-74 samples. Furthermore,the energy changes for 1-hexene adsorption on different MOF-74structures were also obtained from simulation results (see Fig. 13). TheMg-MOF-74 sample releases the highest adsorption energy of 17.4 kcal/mol, followed by Co-MOF-74 showing adsorption energy of 16.5 kcal/mol. In contrast, the Co0.12Mg0.88-MOF-74 sample exhibits the smallestadsorption energy. In the cases of CoMg-MOF-74, the adsorption energyis lower than that of monometallic samples. It seems that incorporationof more stable CUS Co metal sites within the original Mg-MOF-74 fra-mework can lead to the synergetic effect of pore structure evolution aswell as higher density of CUS metal sites in bimetallic frameworks, andtherefore benefits 1-hexene adsorption into MOF-74 structures with lessexothermic effect. Such stepwise adsorption energy of 1-hexene on bi-metallic structures follows a similar trend to the adsorption of hydrogenon NieCo bimetallic MOF-74 [41]. In addition, our results clearly in-dicate that there are strong relationship between adsorption perfor-mances (i.e. the adsorption capacity and selectivity) and metal com-position of MOF-74. Tuning metal centers of MOF-74 is realized aneffective strategy in order to manipulate its adsorption affinity to par-affin and olefin compounds and, therefore, to construct adsorption se-lectivity-oriented porous architectures.

4. Conclusions

We synthesized a series of bimetallic CoxMg1-x-MOF-74 (x=0,0.12, 0.30, 0.57, 0.78, 1) structures via using well-controlled binarymetal compositions in solvothermal systems. Applying XRD, N2 ad-sorption, TG-DSC, SEM, EDS and FT-IR, we confirmed the successfulsynthesis of bimetallic MOF-74 structures and examined structural di-versity among these synthesized samples. As compared to the mono-metallic Mge or Co-MOF-74, bimetallic MOF-74 samples have uniformcrystalline structure but more pore void. Furthermore, the selectiveadsorption of 1-hexene on MOF-74 can be largely enhanced via co-ordinating CoeMg bimetal in the organic frameworks. Specifically,Co0.30Mg0.70-MOF-74 exhibits the largest 1-hexene adsorption capacityof 152.7mg/g and the highest 1-hexene/n-hexane selectivity of 9.74,which are 2.1–2.3 and 4.1–8.9 times higher than those of monometallicCoe or Mg-MOF-74 samples. Manipulation on metal composition ofmetal-organic frameworks is realized an effective strategy to optimizeits adsorption performance, therefore, to construct adsorption se-lectivity-oriented porous crystalline materials.

Notes

The authors declare no competing financial interest.

Acknowledgements

This work is financially supported by the National Natural ScienceFoundation of China (Grant 91634112 and 21878097), the NaturalScience Foundation of Shanghai (Grant 16ZR1408100) and the OpenProject of State Key Laboratory of Chemical Engineering (SKL-ChE-16C01). D.W. acknowledges the institutional funds from the Gene andLinda Voiland School of Chemical Engineering and Bioengineering atWashington State University.

References

[1] Z.B. Bao, J.W. Wang, Z.G. Zhang, H.B. Xing, Q.W. Yang, Y.W. Yang, H. Wu,R. Krishna, W. Zhou, B.L. Chen, Q.L. Ren, Molecular sieving of ethane from ethylenethrough the molecular cross-section size differentiation in gallate-based metal-or-ganic frameworks, Angew. Chem. Int. Ed. 57 (2018) 16020–16025.

[2] E.D. Bloch, W.L. Queen, R. Krishna, J.M. Zadrozny, C.M. Brown, J.R. Long,Hydrocarbon separations in a metal-organic framework with open iron (II) co-ordination sites, Science 335 (2012) 1606–1610.

[3] D.S. Sholl, R.P. Lively, Seven chemical separations to change the world, Nature 532(2016) 435–437.

[4] T. Ren, M. Patel, K. Blok, Olefins from conventional and heavy feedstocks: energyuse in steam cracking and alternative processes, Energy 31 (2006) 425–451.

[5] A. Khodakov, W. Chu, P. Fongarland, Advances in the development of novel cobaltFischer-Tropsch catalysts for synthesis of long-chain hydrocarbons and clean fuels,Chem. Rev. 107 (2007) 1692–1744.

[6] M.E. Dry, The fischer-tropsch process: 1950-2000, catal, Today Off. 71 (2002)227–241.

[7] K. Adil, Y. Belmabkhout, R.S. Pillai, A. Cadiau, P.M. Bhatt, A.H. Assen, G. Maurin,M. Eddaoudi, Gas/vapour separation using ultra-microporous metal-organic fra-meworks: insights into the structure/separation relationship, Chem. Soc. Rev. 46(2017) 3402–3430.

[8] Z.B. Bao, G.G. Chang, H.B. Xing, R. Krishna, Q.L. Ren, B.L. Chen, Potential of mi-croporous metal-organic frameworks for separation of hydrocarbon mixtures,Energy Environ. Sci. 9 (2016) 3612–3641.

[9] P. Bernardo, E. Drioli, G. Golemme, Membrane Gas separation: a review/state ofthe art, Ind. Eng. Chem. Res. 48 (2009) 4638–4663.

[10] M. Carta, R. Malpass-Evans, M. Croad, Y. Rogan, J.C. Jansen, P. Bernardo,F. Bazzarelli, N.B. McKeown, An efficient polymer molecular sieve for membranegas separations, Science 339 (2013) 303–307.

[11] M. Carta, M. Croad, R. Malpass-Evans, J.C. Jansen, P. Bernardo, G. Clarizia,K. Friess, M. Lanc, N.B. McKeown, Triptycene induced enhancement of membranegas selectivity for microporous troger's base polymers, Adv. Mater. 26 (2014)3526–3531.

[12] J.F. Huang, H.M. Luo, C.D. Liang, D.E. Jiang, S. Dai, Advanced liquid membranesbased on novel ionic liquids for selective separation of olefin/paraffin via olefin-facilitated transport, Ind. Eng. Chem. Res. 47 (2008) 881–888.

[13] R.L. Li, H.B. Xing, Q.W. Yang, X. Zhao, B.G. Su, Z.B. Bao, Y.W. Yang, Q.L. Ren,Selective extraction of 1-hexene against n-hexane in ionic liquids with or withoutsilver salt, Ind. Eng. Chem. Res. 51 (2012) 8588–8597.

[14] J. Padin, R.T. Yang, New sorbents for olefin/paraffin separations by adsorption via

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

159

pi-complexation: synthesis and effects of substrates, Chem. Eng. Sci. 55 (2000)2607–2616.

[15] M. Selvaraj, A. Pandurangan, K.S. Seshadri, P.K. Sinha, K.B. Lal, Synthesis, char-acterization and catalytic application of MCM-41 mesoporous molecular sievescontaining Zn and Al, Appl. Catal., A 242 (2003) 347–364.

[16] D.M. Zhao, J. Cheng, C.D. Vecitis, M.R. Hoffmann, Sorption of perfluorochemicalsto granular activated carbon in the presence of ultrasound, J. Phys. Chem. A 115(2011) 2250–2257.

[17] C. Pevida, M.G. Plaza, B. Arias, J. Fermoso, F. Rubiera, J.J. Pis, Surface modificationof activated carbons for CO(2) capture, Appl. Surf. Sci. 254 (2008) 7165–7172.

[18] D. Mohan, C.U. Pittman, Activated carbons and low cost adsorbents for remedia-tion of tri- and hexavalent chromium from water, J. Hazard Mater. 137 (2006)762–811.

[19] A. Cadiau, K. Adil, P.M. Bhatt, Y. Belmabkhout, M. Eddaoudi, A metal-organicframework-based splitter for separating propylene from propane, Science 353(2016) 137–140.

[20] Y.B. He, R. Krishna, B.L. Chen, Metal-organic frameworks with potential for energy-efficient adsorptive separation of light hydrocarbons, Energy Environ. Sci. 5 (2012)9107–9120.

[21] R.B. Lin, L.B. Li, H.L. Zhou, H. Wu, C.H. He, S. Li, R. Krishna, J.P. Li, W. Zhou,B.L. Chen, Molecular sieving of ethylene from ethane using a rigid metal-organicframework, Nat. Mater. 17 (2018) 1128–1133.

[22] H. Furukawa, K.E. Cordova, M. O'Keeffe, O.M. Yaghi, The chemistry and applica-tions of metal-organic frameworks, Science 341 (2013) 1230444.

[23] Pascal D.C. Dietzel, V. Besikiotis, R. Blom, Application of metal-organic frameworkswith coordinatively unsaturated metal sites in storage and separation of methaneand carbon dioxide, J. Mater. Chem. 19 (2009) 7362–7370.

[24] N.T.X. Huynh, V. Chihaia, D.N. Son, Hydrogen storage in MIL-88 series, J. Mater.Sci. 54 (2019) 3994–4010.

[25] L.Y. Li, L.D. Guo, S.Y. Pu, J.W. Wang, Q.W. Yang, Z.G. Zhang, Y.W. Yang, Q.L. Ren,S. Alnemrat, Z.B. Bao, A calcium-based microporous metal-organic framework forefficient adsorption separation of light hydrocarbons, Chem. Eng. J. 358 (2019)446–455.

[26] J.R. Li, J. Sculley, H.C. Zhou, Metal-Organic Frameworks for separations, Chem.Rev. 112 (2012) 869–932.

[27] J. Lee, O.K. Farha, J. Roberts, K.A. Scheidt, S.T. Nguyen, J.T. Hupp, Metal-or-ganic framework materials as catalysts, Chem. Soc. Rev. 38 (2009) 1450–1459.

[28] J.S. Seo, D. Whang, H. Lee, S.I. Jun, J. Oh, Y.J. Jeon, K. Kim, A homochiral metal-organic porous material for enantioselective separation and catalysis, Nature 404(2000) 982–986.

[29] A. Luna-Triguero, J.M. Vicent-Luna, P. Gomez-Alvarez, S. Calero, Olefin/paraffinseparation in open metal Site Cu-BTC metal-organic framework, J. Phys. Chem. C121 (2017) 3126–3132.

[30] L. Chen, S. Yuan, J.F. Qian, W. Fan, M.Y. He, Q. Chen, Z.H. Zhang,Effective Adsorption Separation of n-Hexane/2-Methylpentane in facilely syn-thesized zeolitic imidazolate frameworks ZIF-8 and ZIF-69, Ind. Eng. Chem. Res. 55(2016) 10751–10757.

[31] Y.W. Chen, Z.W. Qiao, D.F. Lv, H.X. Wu, R.F. Shi, Q.B. Xia, H.H. Wang, J. Zhou,Z. Li, Selective adsorption of light alkanes on a highly robust indium based metal-organic framework, Ind. Eng. Chem. Res. 56 (2017) 4488–4495.

[32] G.G. Chang, Z.B. Bao, Q.L. Ren, S.G. Deng, Z.G. Zhang, B.G. Su, H.B. Xing,Y.W. Yang, Fabrication of cuprous nanoparticles in MIL-101: an efficientadsorbent for the separation of olefin-paraffin mixtures, RSC Adv. 4 (2014)20230–20233.

[33] J.W. Yoon, H. Chang, S.J. Lee, Y.K. Hwang, D.Y. Hong, S.K. Lee, J.S. Lee, S. Jang,T.U. Yoon, K. Kwac, Y. Jung, R.S. Pillai, F. Faucher, A. Vimont, M. Daturi, G. Ferey,C. Serre, G. Maurin, Y.S. Bae, J.S. Chang, Selective nitrogen capture by poroushybrid materials containing accessible transition metal ion sites, Nat. Mater. 16(2017) 526–531.

[34] F. Luo, C.S. Yan, L.L. Dang, R. Krishna, W. Zhou, H. Wu, X.L. Dong, Y. Han, T.L. Hu,M. O'Keeffe, L.L. Wang, M.B. Luo, R.B. Lin, B.L. Chen, UTSA-74: a MOF-74 isomer

with two accessible binding sites per metal center for highly selective gas separa-tion, J. Am. Chem. Soc. 138 (2016) 5678–5684.

[35] J.A. Botas, G. Calleja, M. Sanchez-Sanchez, M.G. Orcajo, Effect of Zn/Co ratio inMOF-74 type materials containing exposed metal sites on their hydrogen adsorptionbehaviour and on their band gap energy, Int. J. Hydrogen Energy 36 (2011)10834–10844.

[36] J. Kahr, R.E. Morris, P.A. Wright, Post-synthetic incorporation of nickel into CPO-27(Mg) to give materials with enhanced permanent porosity, CrystEngComm 15(2013) 9779–9786.

[37] L.J. Wang, H.X. Deng, H. Furukawa, F. Gandara, K.E. Cordova, D. Peri, O.M. Yaghi,Synthesis and characterization of metal-organic framework-74 containing 2, 4, 6, 8,and 10 different metals, Inorg. Chem. 53 (2014) 5881–5883.

[38] Y. Jiao, C.R. Morelock, N.C. Burtch, W.P. Mounfield, J. Hungerford, K.S. Walton,Tuning the kinetic water stability and adsorption interactions of Mg-MOF-74 bypartial substitution with Co or Ni, Ind. Eng. Chem. Res. 54 (2015) 12408–12414.

[39] H.X. Jiang, Q.Y. Wang, H.Q. Wang, Y.F. Chen, M.H. Zhang, MOF-74 as an efficientcatalyst for the low-temperature selective catalytic reduction of NOx with NH3, ACSAppl. Mater. Interfaces 8 (2016) 26817–26826.

[40] H.Y. Cho, D.A. Yang, J. Kim, S.Y. Jeong, W.S. Ahn, CO2 adsorption and catalyticapplication of Co-MOF-74 synthesized by microwave heating, Catal. Today 185(2012) 35–40.

[41] J.A. Villajos, G. Orcajo, C. Martos, J.A. Botas, J. Villacanas, G. Calleja, Co/Ni mixed-metal sited MOF-74 material as hydrogen adsorbent, Int. J. Hydrogen Energy 40(2015) 5346–5352.

[42] P. Valvekens, M. Vandichel, M. Waroquier, V.V. Speybroeck, D.D. Vos, Metal-di-oxidoterephthalate MOFs of the MOF-74 type: microporous basic catalysts withwell-defined active sites, J. Catal. 317 (2014) 1–10.

[43] H.F. Yao, Y. Yang, H. Liu, F.G. Xi, E.Q. Gao, CPO-27-M as heterogeneous catalystsfor aldehyde cyanosilylation and styrene oxidation, J. Mol. Catal. A Chem. 394(2014) 57–65.

[44] P.D.C. Dietzel, Y. Morita, R. Blom, H. Fjellvag, An in situ high-temperature single-crystal investigation of a dehydrated metal-organic framework compound and field-induced magnetization of one-dimensional metal-oxygen chains, Angew. Chem. Int.Ed. 44 (2005) 6354–6358.

[45] J.D. Howe, C.R. Morelock, Y. Jiao, K.W. Chapman, K.S. Walton, D.S. Sholl,Understanding structure, metal distribution, and water adsorption in mixed-metalMOF-74, J. Phys. Chem. C 121 (2017) 627–635.

[46] D.W. Wang, Q.H. Wang, T.M. Wang, Morphology-controllable synthesis of cobaltoxalates and their conversion to mesoporous Co3O4 nanostructures for applicationin supercapacitors, Inorg. Chem. 50 (2011) 6482–6492.

[47] K. Tan, S. Zuluaga, Q.H. Gong, P. Canepa, H. Wang, J. Li, Y.J. Chabal,T. Thonhauser, Water reaction mechanism in metal organic frameworks with co-ordinatively unsaturated metal ions: MOF-74, Catal. Commun. 80 (2016) 24–27.

[48] H.X. Jiang, Q.Y. Wang, H.Q. Wang, Y.F. Chen, M.H. Zhang, Temperature effect onthe morphology and catalytic performance of Co-MOF-74 in low-temperature NH3-SCR process, Catal. Commun. 80 (2016) 24–27.

[49] R.L. Li, H.B. Xing, Q.W. Yang, X. Zhao, B.G. Su, Z.B. Bao, Y.W. Yang, Q.L. Ren,Selective extraction of 1-hexene against n-hexane in ionic liquids with or withoutsilver salt, Ind. Eng. Chem. Res. 51 (2012) 8588–8597.

[50] A.E. Wentink, N.J.M. Kuipers, A. B. De Haan, J. Scholtz, H. Mulder, Effects of ligandstructure on reactive vapor-liquid distribution ratio and selectivity for C6-olefinisomers, Ind. Eng. Chem. Res. 44 (2005) 9221–9229.

[51] A.J. Nunez, L.N. Shear, N. Dahal, I.A. Ibarra, J. Yoon, Y.K. Hwang, J.S. Chang,S.M. Humphrey, A coordination polymer of (Ph3P)AuCl prepared by post-syntheticmodification and its application in 1-hexene/n-hexane separation, Chem. Commun.47 (2011) 11855–11857.

[52] S.J. Geier, J.A. Mason, E.D. Bloch, W.L. Queen, M.R. Hudson, C.M. Brown,J.R. Long, Selective adsorption of ethylene over ethane and propylene over propanein the metal-organic frameworks M2(dobdc) (M = Mg, Mn, Fe, Co, Ni, Zn), Chem.Sci. 4 (2013) 2054–2061.

H. Sun, et al. Microporous and Mesoporous Materials 284 (2019) 151–160

160