Subscriber access provided by SOUTH CHINA UNIV OF TECH
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Rearch Article
Argon Adsorption on Cu(Benzene-1,3,5-tricarboxylate)(H O)
Metal−Organic Framework
V. Krungleviciute, K. Lask, L. Heroux, A. D. Migone, J.-Y. Lee, J. Li, and A. Skoulidas
Langmuir,
2007, 23 (6), 3106-3109 • DOI: 10.1021/la061871a
Downloaded from pubs.acs on November 26, 2008
More About This Article
geroAdditional resources and features associated with this article are available within the HTML version:•Supporting Information
•Links to the 11 articles that cite this article, as of the time of this article download
•Access to high resolution figures
•Links to articles and content related to this article
•Copyright permission to reproduce figures and/or text from this article
Argon Adsorption on Cu3(Benzene-1,3,5-tricarboxylate)2(H2O)3
letterMetal-Organic Framework
V.Krungleviciute,†K.Lask,†L.Heroux,†A.D.Migone,*,†J.-Y.Lee,‡J.Li,‡and
A.Skoulidas§
Department of Physics,Southern Illinois Uni V ersity,Neckers483A,Carbondale,Illinois62901, Department of Chemistry and Chemical Biology,Rutgers Uni V ersity,610Taylor Road,Piscataway,New Jery08854,and ExxonMobil Rearch and Engineering,3225Gallows Road,Fairfax,Virginia22037
Recei V ed June28,2006.In Final Form:October17,2006
Using volumetric adsorption techniques,we have measured the adsorption of argon on Cu3(BTC)2(H2O)3,(BTC )benzene-1,3,5-tricarboxylate),a microporous metal-organic framework structure,at temperatures between66and 143K.In addition to the experiments,we have ud Grand Canonical Monte Carlo simulations to calculate the adsorption isotherm of argon at87K.Our experimental and theoretical results are compared to tho of previous studies.The experiments were performed using a high density of points,allowing us to obtain,in detail,the isosteric heat’s coverage dependence.Our values from the simulations are in reasonable agreement with tho obtained in the experiments.
Introduction
Metal-organic frameworks(MOFs)consist of metal centers and/or metal clusters connected by organi
c linkers,forming3-D porous structures with1-D,2-D,or3-D channel systems.1-8 Through a careful lection of linkers and metallic clusters,the MOFs can be designed so that they have a porous structure of predetermined porosity,with very well-determined and uniform pore diameters(both of which are controlled by the structure of the material).1-20MOFs have attracted a great deal of interest from rearchers working in theory,experiment,simulation,and synthesis1-31becau they are among the most promising candidates currently being explored for their potential as storage materials for alternative ,hydrogen and methane).2,5-7,10,12-14,17,18,23,26,28Becau of the enormous societal and economic impact of a successful development in this area, it is important to thoroughly and rigorously test the adsorptive properties of the materials.
Cu3(BTC)2(H2O)3(benzene-1,3,5-tricarboxylate),e Figure 1,was one of the earliest studied metal-organic frameworks. Chui et al.9first reported information on this MOF in1999.In the framework of this material,two octahedrally coordinated Cu atoms are connected to eight oxygen atoms of tetra-carboxylate units to form a dimeric Cu paddle wheel.Each BTC ligand holds three dimeric Cu paddle wheels to form a microporous open framework with face-centered cubic symmetry.
Cu-BTC has a3-D channel structure connecting a system of tetrahedral-shaped cages accessible through small windows(∼3.5Åin diameter).The large cavities are connected through square-shaped
windows with a diameter of∼9Å.At moderate temperatures and pressures,the adsorption of molecules in the central cavity of the unit cell in Cu-BTC is localized clo to the internal surface.Most of the central cavity is filled only at low temperatures or high pressures.21
*Corresponding author.Fax:(618)453-1056.Phone:(618)453-1053. E-mail:aldo@physics.siu.edu.
†Southern Illinois University.
‡Rutgers University.
西班牙语学习§ExxonMobil Rearch and Engineering.
(1)Yaghi,O.M.;O’Keeffe,M.;Ockwig,N.W.;Chae,H.K.;Eddaoudi,M.; Kim,J.Nature2003,423,705-714.
(2)Eddaoudi,M.;Kim,J.;Rosi,N.;Vodak,D.;Wachter,J.;O’Keeffe,M.; Yaghi,O.M.Science2002,295,469-472.
(3)Fletcher,A.J.;Thomas,K.M.;Rosinsky,M.J.J.Solid State Chem. 2005,178,2491-2510.
(4)Pan,L.;Liu,H.;Lei,X.;Huang,X.;Olson,D.H.;Turro,N.J.;Li,J. Angew.Chem.,Int.Ed.2003,42,542-542.
(5)Pan,L.;Sander,M.B.;Huang,X.;Li,J.;Smith,M.;Bittner,E.;Bockrath,
B.;Johnson,J.K.J.Am.Chem.Soc.2004,126,1308-1309.
(6)Lee,J.Y.;Li,J.;Jagiello,J.J.Solid State Chem.2005,178,2527-2532.
(7)Du¨ren,T.;Sarkisov,L.;Yaghi,O.M.;Snurr,R.Q.Langmuir2004,20, 2683-2689.
(8)Eddaoudi,M.;Li,H.L.;Yaghi,O.M.J.Am.Chem.Soc.2000,122,
1391-1397.
(9)Chui,S.S.-Y;Lo,S.M.-F.;Charmant,J.P.H.;Orpen,A.G.;Williams,
I.D.Science1999,283,1148-1150.
(10)Rowll,J.L.C.;Yaghi,O.M.Angew.Chem.,Int.Ed.2005,44,4670-4679.
(11)Li,H.;Eddaoudi,M.;O’Keeffe,M.;Yaghi,O.M.Nature1999,402, 276-279.
(12)Mueller,U.;Schubert,M.;Teich,F.;Puetter,H.;Schierle-Arndt,K.; Pastre´,J.J.Mater.Chem.2006,16,626-636.
(13)Yang,Q.;Zhong,C.J.Phys.Chem.B2005,109,11862-11864.
(14)Chen,B.;Ockwig,N.W.;Millward,A.R.;Contreras,D.S.;Yaghi,O. M.Angew.Chem.,Int.Ed.2005,44,4745-4749.
(15)Chen,B.;Eddaouadi,M.;Hyde,S.T.;O’Keeffe,M.;Yaghi,O.M.Science 2001,291,1021-1023.
(16)Chae,H.K.;Siberio-Pe´rez,D.Y.;Kim,J.;Go,Y.B.;Eddaoudi,M.; Matzger,A.J.;O’Keeffe,M.;Yaghi,O.M.Nature2004,427,523-527.
(17)Rosi,N.L.;Eckert,J.;Eddaoudi,M.;Vodak,D.T.;Kim,J.;O’Keeffe, M.;Yaghi,O.M.Science2003,300,1127-1129.
(18)Rowll,J.L.C.;Eckert,J.;Yaghi,O.M.J.Am.Chem.Soc.2005,127, 14904-14910.
成都理工学院分数线
(19)Vishnyakov,A.;Ravikovitch,P.I.;Neimark,A.V.;Bulow,M.;Wang, Q.M.Nano Lett.2003,3,713-718.
(20)Eddaoudi,M.;Moler,D.B.;Li,H.L.;Chen,B.;Reineke,T.M.;O’Keeffe, M.;Yaghi,O.M.Acc.Chem.Res.2001,34,319-330.
(21)Skoulidas,A.I.J.Am.Chem.Soc.2004,126,1356-1357.
(22)Pan,L.;Liu,H.;Kelly,S.P.;Huang,X.;Olson,D.H.;Li,J.Chem. Commun.2003,7,854-855.
(23)Lee,J.-Y.;Pan,L.;Kelly,S.P.;Jagiello,J.;Emge,T.J.;Li,J.Ad V.Mater. 2005,17,2703-2706.
大学英语学习
(24)Wang,Q.M.;Shen,D.;Bu¨low,M.;Lau,M.L.;Deng,S.;Fitch,F.R.; Lemcoff,N.O.;Semanscin,J.Microporous Mesoporous Mater.2002,55,217-230.
(25)Rowll,J.L.C.;Yaghi,O.M.Microporous Mesoporous Mater.2004, 73,3-14.
(26)Garberoglio,G.;Skoulidas,A.I.;Johnson,J.K.J.Phys.Chem.B2005, 109,13094-13103.
(27)Skoulidas,A.I.;Sholl,D.S.J.Phys.Chem.B2005,109,15760-15768.
(28)Sagara,T.;Klasn,J.;Ganz,E.J.Chem.Phys.2004,121,12543-12547.
(29)Sarkisov,L.;Du¨ren,T.;Snurr,R.Q.Mol.Phys.2004,102,211-221.
(30)Skoulidas,A.I.;Sholl,D.S.J.Phys.Chem.A2003,107,10132-10141.
(31)Centrone,A.;Siberio-Pe´rez,D.Y.;Millward,A.R.;Yaghi,O.M.;Matzger,
A.J.;Zerbi,G.Chem.Phys.Lett.2005,411,516-519.
3106Langmuir2007,23,3106-3109
10.1021/la061871a CCC:$37.00©2007American Chemical Society
Published on Web02/06/2007
In the prent work,we describe the results of experiments and computer simulations of Ar adsorption on Cu -BTC.We have analyzed the detailed features of the isotherms as a function of the amount of argon adsorbed and the temperature.We have also obtained experimental values for the isosteric heat of adsorption.The experimental data are compared with computer simulations.
Synthesis
The crystal structure of Cu 3(BTC)2(H 2O)3was first reported by Chui et al.9However,their reactions at 453K under hydrothermal conditions also generated a Cu 2O impurity pha as a byproduct.In our synthesis,the conditions were modified to produce a pure pha of Cu 3(BTC)2(H 2O)3.First,Cu(NO 3)2‚3H 2O (0.435g or 1.8mmol)was dissolved in 6mL of deionized water,and BTC (0.110g or 0.5mmol)was dissolved in 6mL of ethanol.The two solutions were then mixed in a 100mL beaker and stirred for 30min.The solution mixture was transferred into a Teflon-lined autoclave and heated in an
oven at 403K for 3h.After cooling the autoclave down to room temperature naturally,the solution was filtered,and the product was washed with ethanol (10mL ×3times).Shiny blue crystals of Cu -BTC were collected.
Adsorption Isotherm Measurements.We performed isotherm measurements on an in-hou designed and built volumetric adsorption apparatus.Pressures were measured using three different room-temperature pressure gauges (of maximum ranges 1,100,and 1000Torr,respectively).Gas was dod into the dosing volume in small steps through computer-controlled valves until a predetermined dosing amount was reached.The valve to the cell was then opened to allow the gas into the experimental chamber that contained the Cu -BTC.The temperature of the experimental cell was controlled in two stages.The temperature stability of the sample cell was (0.010K.The verification that equilibrium conditions were reached,and the recording of data was done through the u of a data acquisition program that we developed using LabView.
Prior to the start of the measurements,the Cu -BTC sample was placed into the stainless steel cell in which the measurements were to be conducted,and it was placed in an oven and outgasd under vacuum at 100°C for 6h.The cell was then transported,without breaching vacuum integrity,to the adsorption tup,attached to the clod-cycle refrigerator,and connected to the gas handling system.
The initial sample weight was 102.1mg.Our TGA measurements indicated that after 4h of heating at 100°C,the sample lost about 8.9%of its weight due to the removal of water molecules.The weight of the sample after outgassing was estimated to be 93mg bad on this consideration.
Adsorption Isotherm Simulations.Cu -BTC was modeled as a rigid framework,with the atoms held fixed in their experimentally determined crystallographic positions.19,21Cu -BTC has a cubic lattice with a unit cell dimension of 26.343Å.The structure of Cu -BTC was taken from Chui et al.9The crystal structure of Chui
et al.includes axial oxygen atoms bonded to the Cu atoms,which correspond to water ligands.We simulated dry Cu -BTC with the oxygen atoms removed.19,21
To model Ar,we ud a spherical Lennard-Jones potential;the values of the parameters ud (which are the same as tho ud previously)30are listed in Table 1.Interactions between adsorbed Ar molecules and Cu -BTC were modeled using pairwi interactions between adsorbates and each atom in the metal -organic framework.Only Lennard-Jones interactions were considered for the interac-tions.The parameters were taken from the UFF potential.32Mixed-atom interactions were defined using the Lorenz -Berthelot mixing rules.In the simulations,a cutoff distance of 17Åwas ud
for the Lennard-Jones interactions.Long-range corrections were included in the adsorbate -MOF interactions by assuming that the MOF was isotropic at distances beyond the cutoff.This application of long-range corrections inside what is an intrinsically structured material is of cour an approximation,a point discusd carefully by Macedonia and Maginn.33
The interactions with Cu -BTC were pretabulated on a 0.2Ågrid.During the simulations,a 3-D cubic Hermite polynomial interpolation scheme was ud to calculate the potential at each point in space.30
berlinAdsorption isotherms of Ar were computed using GCMC simulations.34,35The chemical potential of the bulk gas was related to its pressure by the virial equation of state,which was fitted to experimental data by NIST.36A minimum 2×2×2unit cell simulation box was ud.The adequacy of the super cell size was tested,and a 2×2×2super cell was found to be sufficient for the whole pressure range considered in this study.At the lowest densities,the size of the simulation volume was incread so as to contain at least 50particles during the simulations.At least 2million equilibration and 5million production steps were ud for each loading.At the highest loadings,as many as 10million equilibration and 25million production steps were ud.
In addition to calculating the equilibrium pore loading at each t of bulk pha conditions we considered,we also computed isosteric heats of adsorption,Q st .Assuming ideal behavior in the bulk pha,the isosteric heat can be determined in GCMC simulations in terms of fluctuations in the internal energy and number of adsorbed molecules.37,38
Results and Discussion
In one of the earlier studies of Cu -BTC,19Ar isotherm experiments were conducted at 87K,and the GCMC technique was ud to simulate veral possible adsorption isotherms employing four different force fields to describe adsorbent and adsorbate interactions.Three different substeps were prent in the simulated data.The first,low coverage,substep was
(32)Rappe,A.K.;Cawit,C.J.;Colwell,K.S.;Goddard,W.A.,III;Skiff,W.M.J.Am.Chem.Soc.1992,114,10024-10035.
(33)Macedonia,M.D.;Maginn,E.J.Mol.Phys.1999,96,1375-1390.(34)Frenkel, D.;Smit, B.Understanding Molecular Simulation:From Algorithms to Applications ;Academic Press:London,1996.
(35)Heuchel,M.;Snurr,R.Q.;Buss,E.Langmuir 1997,13,6795-6804.(36)Lemmon,E.W.;McLinden,M.O.;Friend,D.G.Thermophysical Properties of Fluid Systems.In NIST Chemistry WebBook,NIST Standard Reference Databa Number 69;Linstrom,P.J.,Mallard,W.G.,Eds.;National Institute of Standards and Technology:Gaithersburg,MD,2005;p 20899(v).
(37)Nicholson,D.;Parsonage,G.Computer Simulation and the Statistical Mechanics of Adsorption ;Academic Press:New York,1982.
(38)Karavias,F.;Myers,A.L.Langmuir 1991,7,3118-
3126.
Figure 1.Cu 3(BTC)2(H 2O)3(BTC )benzene-1,3,5-tricarboxylate)metal -organic framework.In the figure,Cu is shown in green,O is shown in red,C is shown in gray,and H is omitted for clarity.
Table 1.Parameters of the Lennard-Jones Potential
(K)
σ(Å)Ar -C 70.27 3.44Ar -Cu 15.315 3.28Ar -H 45.493Ar -O 53.12 3.28Ar -Ar
119.8
3.4
Argon Adsorption on Cu 3(BTC)2(H 2O)3Langmuir,Vol.23,No.6,20073107
attributed to adsorption in the tetrahedral side pockets,the cond substep to adsorption in the main channels,and the third one to the solidification of argon inside the main pores.The experimental isotherm measured on Cu -BTC only displayed the first two substeps,not the third.
We measured Ar isotherms at eight different temperatures,from 66up to 143K.Unlike the previous experiments,we were able to obrve all three substeps in the experimental data (e Figure 2).
The first substep (A in Figure 2)occurs at around 24cm 3STP/g.This step is attributed to adsorption in the tetrahedral side pockets of the Cu -BTC metal -organic framework.The pockets have ∼3.5Ådiameter windows that are large enough to let the argon atoms through and be adsorbed inside.This feature was also obrved in previous theoretical and experimental work done on this system.19The cond substep (B in Figure 2)starts at around 55cm 3STP/g and ends at around 180cm 3STP/g.It is attributed to adsorption in the main channels.The third substep (C in Figure 2)occurs around 215cm 3STP/g.This feature was attributed in the previous computer simulation studies 19to the solidification of argon in the pores of the metal -organic framework.While a feature that corresponds to this behavior was obrved in our data (e C in Figure 2),we note that no such feature was obrved in the previous experimental measurements.19Our simulations show that between the cond and the third substeps,the majority of the empty space is filled with argon molecules.It is possible that an ordering of the absorbed molecules or a complete saturation of the center regions of the large cages lead to the third obrved step.However,the order parameter has not been calculated,and the origin of the third substep was not analyzed in terms of molecular configu-ration.bookstore
scheduled
We have computed the effective specific surface area (ESSA)for this sample from the isotherm measured at 78.5K.We obtained 886m 2/g for this quantity;this value is 59%of the 1500m 2/g measured by Vishnyakov et al.using argon 19but is in good agreement with the value of 917.6m 2/g determined by Chui et al.9using nitrogen.At least a portion of the difference in the ESSA values can be attributed to the different values ud for the cross-ctional area of argon.
We also obtained the total pore volume for this sample.We ud a density of 1378.5kg/m 3for Ar at 90.03K.36We obtained a total pore volume of 0.32cm 3/g.This is value is 86%of the 0.37cm 3/g reported by Vishnyakov et al.,1980%of the 0.40
cm 3/g reported by Lee et al.,6and in excellent agreement with the 0.333cm 3/g determined by Chui et al.9using nitrogen isotherms.We note that all of the previous quoted results,including our own,are significantly lower than the total pore volume found by Wang et al.,who reported a volume of 0.658cm 3/g for this material.24(They note that the level of microporosity depends on the activation conditions of the sample.)We carried out GCMC simulations for argon adsorbed on Cu -BTC as described previously.The results from the calcula-tions are shown as the red curve in Figure 2.There are three substeps prent in the isotherm simulated at 87K.The first one occurs at around 4.5×10-5atm,the cond at 8.1×10-4atm,and the third at around 5×10-3atm.All of the pressure values are
grace什么意思in excellent quantitative agreement with the experimental adsorption isotherms we measured at 90K.The amount adsorbed in the simulation data was scaled down by a factor of 1.6to match our experimental isotherm (no adjustment was applied to the pressure values).The origin of this difference is unclear at this time.We note that this factor is within the range of the ratio of experimentally reported values for the pore volume.6,9,19,24One of the important quantities that can be obtained from adsorption isotherm results is the isosteric heat of
angra mainyuadsorption
Figure 2.Experimental argon adsorption isotherms (black symbols)on the Cu -BTC sample and simulated adsorptin isotherm (red symbols)for the same sample at 87
K.
Figure 3.Logarithm of pressure vs inver of the isotherm temperature for different amounts of argon adsorbed on the substrate.The slope of the line is proportional to the isosteric heat of adsorption for each amount
adsorbed.
Figure 4.Experimental and theoretical isosteric heat data.Arrows that point at three peaks correspon
d to the steps A -C,also pointed to by the arrows,in Figure 2.
3108Langmuir,Vol.23,No.6,2007Krungle V iciute et al.
that,in terms of isotherm data,is given by eq1(ref39)
We have determined the isosteric heat of adsorption for argon adsorbed on Cu-BTC using this expression.In Figure3,we prent a plot of the logarithm of the pressure versus the inver of the temperature.Each line reprents data taken at the same amount adsorbed on the substrate,for the various temperatures measured.The slope of this line is proportional to the isosteric heat of adsorption corresponding to that value of the amount adsorbed(e eq1).
In Figure4,we show both computer simulated and experimental results for the isosteric heat’s dependence on the total amount of argon adsorbed(the simulated coverages have been scaled down,as they were in Figure2).The isosteric heat decreas significantly as the amount adsorbed increas.At the lower coverages,where the isotherms indicate that there is adsorption in the side pockets,the experimental q st value is around145 meV.At coverages corresponding to the cond substep,the experimental value of q st is around112meV.Finally,for coverages corresponding to the third substep,which in the theoretical calculations was attributed to the solidification of argon,we e
a sharp peak in the isosteric heat.This peak is consistent with capillary condensation occurring in the pores (our simulations do not allow us to definitively identify this feature with a solidification transition,as was done in previous simulations).19
From the GCMC simulations,we found Ar isosteric heat values of136and110meV for the tetrahedral side pockets and main pores,respectively.Using the GCMC technique,Vishnyakov et al.19obtained a value of150.8meV for the depth of the potential on the side pockets(corresponding to the first substep).The simulation values are very clo to the values of145and112 meV that we obtained from our experimental measurements.
To obtain an idea of the strength of the binding of argon on Cu-BTC,we calculated the binding energy of argon on the Cu-BTC sample using eq2(ref40)
We obtained a value of124meV for the binding energy.As a comparison,the value of for argon on the grooves of single walled carbon ,on the highest binding energy sites of clod-ended nanotubes)is163meV,41and the value of for Ar on planar graphite is99meV.42
Since the value of the binding energy of Ar on Cu-BTC is smaller than that of Ar on clo-ended nanotubes,and since both a large adsorptive capacity and a high binding energy sites are needed for
an effective gas storage material,it is unlikely that Cu-BTC will be able to fulfill the practical requirements that an effective storage material must meet.
Conclusion
We have studied the adsorption of argon on a Cu-BTC metal-organic framework using experimental and theoretical techniques. We found three substeps prent in the sorption isotherms.The first two correspond to the adsorption in the tetrahedron side pockets and main channels.The third substep prent in the experimental data(that was not found by Vishnyakov et al.)and also in GCMC simulations corresponds to capillary condensation. The values obtained in the simulations for the isosteric heat of adsorption and its dependence on adsorbent loading are in very good agreement with the experimental values we measure. Acknowledgment.J.Y.L.and J.L.acknowledge the donors of The Petroleum Rearch Fund administrated by the ACS for the partial support of this work.
LA061871A
(39)Dash,J.G.Films on Solid Surfaces;Academic Press:New York,1975.
(40)Wilson,T.;Tyburski,A.;DePies,M.R.;Vilches,O.E.;Becquet,D.; Bienfait,M.J.Low Temp.Phys.2002,126,403-408.
(41)Krungleviciute,V.;Heroux,L.;Migone,A.D.;Kingston,C.T.;Simard,
B.J.Phys.Chem.B2005,109,9317-9320.
(42)Vidali,G.;Ihm,G.;Kim,H.-Y.;Cole,M.W.Surf.Sci.Rep.1991,12, 133-181.
q st )-k
B(∂ln P∂1/T
)n(1)
)q
st
-2kT(2)
Argon Adsorption on Cu3(BTC)2(H2O)3Langmuir,Vol.23,No.6,20073109
