MOF–graphite oxide nanocomposites:surface characterization and evaluation as adsorbents of ammonia
firmamentCamille Petit and Teresa J.Bandosz *
Received 5th May 2009,Accepted 13th June 2009
First published as an Advance Article on the web 15th July 2009DOI:10.1039/b908862h
Metal-organic framework (MOF-5)–graphite oxide (GO)composite was synthesized using
a solvothermal synthesis route.The parent materials (MOF-5and GO)and the nanocomposite were characterized using X-ray diffraction,SEM,TEM,FTIR and adsorption of nitrogen.They were also tested as adsorbents of ammonia in dynamic conditions.The composite material obtained had a unique layered texture with a prerved structure of MOF-5and GO.When tested as ammonia adsorbent,the composite showed some synergy enhancing the adsorption capacity in comparison with the
hypothetical physical mixture of the components.Although the removal capacity was high in the prence of moisture,water had a detrimental effect on the chemistry of materials and destroyed their porous framework.This caud ammonia retained on the surface to be progressively desorbed from th
e materials when the samples were purged with air.
Introduction
Graphite oxide (GO),formed by treating graphite with very strong oxidizing agents such as KClO 3/HNO 3,1has a layered structure and a non-stochiometric chemical composition,which depends on the level of oxidation.Recent advances in the synthesis and experimental characterization 2–5of GO provide the possibility of controlling the structure and surface chemistry of the materials,leading to renewed interest in them as adsor-bents,and particularly as reactive adsorbents.6,7The graphene layers of GO are arranged in an organized way with the inter-layer distance varying from 6to 12A
˚.The main reason for the variations in interlayer distance is the level of hydration.2Since epoxy and hydroxyl groups exist within the interlayer space,the water molecules are attracted there via hydrogen bonding.2Moreover,both functional groups and water molecules have different types of motion depending on the location of water,which can be either embedded or distributed in interlayer voids.The layered character of graphite oxide opened a new route for the synthesis of composite materials.8–19This path is possible owing to the hydrophilic character of GO,its easy dispersion in water and delamination in alkaline media or alcohols.20,21Moreover,graphite layers can be easily re
stacked and their degree of orientation depends on the method of drying.Our recent studies demonstrate that graphite oxide can be an efficient adsorbent of ammonia.6,7,22NH 3reacts with the acidic groups located on the edges of graphene layers and is also intercalated between the graphene sheets where hydrogen bonding with epoxy groups is the predominant adsorption mechanism.The amount of ammonia adsorbed on GO is very high in comparison with activated carbons.6,7,22,23
Another group of interesting materials are metal–organic frameworks (MOFs).24–34Their synthesis and properties were described in detail by Yaghi and co-workers.24–27They showed that intermolecular interactions and metal–ligand coordinations may be ud to design a wide variety of 2-D and 3-D metal–organic networks with high porosity,unusual ion exchange and adsorptive properties.35,36Examples are MOF-n materials which are built from the extended analogues of molecular metal carboxylate clusters.26They are stable at rather high tempera-ture and their porosity reaches 60%.35The preparation of MOFs involves reaction between solutions of metal species and organic linkers.The structure obtained is the result of maximum degrees of freedom of both components,spherical shape of metal ions and well-defined points of contacts for the organic linker.A detailed summary of the factors affecting the reticular chemistry of tho materials is prent
ed in ref.29.Besides a broad range of transition metals,which are ud in the synthesis,also noble metals can expand the catalytic features of the materials.30
Well-defined framework and high porosity with a broad range of pore sizes make MOFs potentially good adsorbents for air purification and paration.36–41So far adsorption of species such as hydrogen,37–39carbon dioxide 40or methane 41has been studied.Recently,Yaghi and co-workers studied the adsorption of various harmful gas using veral types of MOF-n materials.36Improvement compared to a common activated carbon (BPL carbon)was obrved,especially in the retention of ammonia.36An interesting summary of industrial applications of MOFs is prented in ref.38with examples of the superiority of this kind of materials toward adsorption of tetrahydrothiophene,amines,ammonia and alcohols.The materials also show a good performance in gas storage under high pressure.
Taking into account the above,a major goal of this rearch is to prent the synthesis of MOF–GO nanocomposite materials,which combine the favorable attributes of carbonaceous surfaces and MOFs.In particular,GO should potentially lead to an enhancement in nonspecific adsorption owing to the prence of
Department of Chemistry,The City College and the Graduate School of the City University of New York,160Convent Avenue,New York,NY,10031,USA.E-mail:tbandosz@ccny.cuny.edu;Fax:+12126506107;Tel:+12126506017
PAPER www.rsc/materials |Journal of Materials Chemistry
P u b l i s h e d o n 15 J u l y 2009. D o w n l o a d e d b y J i l i n U n i v e r s i t y o n 29/11/2013 00:03:07.
View Article Online / Journal Homepage / Table of Contents for this issue
extended graphene type layers and also to an enhancement in strong specific adsorption owing to its acidic character.On the other hand,MOF can improve the kinetics of adsorption due to the structure of its framework.Moreover,the specific interactions and reactivity can also be enhanced owing to MOF’s chemical composition.MOF-5was chon in this study as the MOF-n compound.In this material,[Zn 4O]6+tetrahedra form the corners of the resulting primitive cubic structure,while benzene carboxylates (1,4-benzenedicarboxylate,BDC)allow the junction between the zinc oxide clusters.42The formula of the resulting material is Zn 4O(H-BDC)3.35Synthesis of the parent materials and the nanocomposite is fol-lowed by their surface characterization using a range of expe
ri-mental methods.Then,their performance in adsorption of ammonia is evaluated.To the best of our knowledge,we are the first group reporting the synthesis and characterization of such a material.
Experimental
Synthesis of materials
Graphite oxide was synthesized by oxidation of graphite (Sigma-Aldrich)using the Hummers method.43Briefly,graphite powder (10g)was stirred with cold concentrated sulfuric acid (230mL at 0 C).Then,potassium permanganate (30g)was added to the suspension slowly to prevent a rapid ri in temperature (less than 20 C).The reaction mixture was then cooled to 2 C.After removal of the ice-bath,the mixture was stirred at room temperature for 30min.Distilled water (230mL)was slowly added to the reaction vesl to keep the temperature under 98 C.The diluted suspension was stirred for an additional 15min and further diluted with distilled water (1.4L),before adding hydrogen peroxide (100mL).The mixture was left overnight.GO particles,ttled at the bottom,were parated from the excess liquid by decantation followed by centrifugation.The remaining suspension was transferred to dialysis tubes (MW cutoff 6000–9000).Dialysis was carried out until no precipitate of BaSO 4was detected by addition of BaCl 2.Then,the wet form of graphite oxide was centrifuged and freeze-dried.
A fine brown powder of the initial graphite oxide was obtained.The resulting material is referred to as GO.
MOF-5was prepared by mixing zinc nitrate hexahydrate (10.4g)and 1,4-benzenedicarboxylate (2g)in N ,N -dime-thylformamide (DMF,140mL)until complete dissolution of the solids.Then,the mixture was transferred into a round-bottom flask connected to a condenr and heated at 115–120 C for 24h.After cooling,the supernatant was removed and the crystals deposited on the bottom of the flask were collected,washed with DMF,and immerd in fresh chloroform overnight.Chloroform was changed twice over the cour of two days.Finally,crystals were collected,placed inside a clod filtering flask connected to an aspirator ud to create vacuum inside the flask,and heated at 130–135 C for 6h.The resulting crystals were then kept in a dessicator.
The composite material was prepared by dispersing GO powder in the well-dissolved zinc nitrate–BDC mixture.The resulting suspension was subquently stirred and subjected to the same synthesis procedure as for MOF-5.The added GO
consists of 5wt%of final material.The synthesized composite is referred to as MOF-5–GO.
泸江日语Methods
Ammonia adsorption
Adsorption capacity for the removal of ammonia was assd by carrying out dynamic tests at room temperature.In this process,a flow of ammonia diluted in air went through a fixed bed of an adsorbent sample.22,23The total flow rate of the inlet gas was 450mL min À1with an ammonia concentration of 1000ppm.The adsorbent bed contained about 1.5cc of the adsorbent powder (GO,MOF-5or MOF-5–GO)packed into a glass column.The size of the bed was 20mm (height)Â10mm (diameter).The conditions were chon to accelerate the time of the test and limit the exposure of the nsor,the lifetime of which is relatively short.The ammonia concentration in the outlet gas was measured using a Multi-Gas Monitor ITX system.The adsorp-tion capacity of each sample was then calculated in milligrams per gram of sorbent,as the difference between the inlet and outlet concentrations multiplied by the inlet flow rate,the break-through time and the ammonia molar mass in the experimental conditions.To evaluate the influence of water,the experiments for all carbon samples were performed with a flow of ammonia gas diluted either in dry air (ED)or in moist air (70%humidity)(EM).On all samples,the desorption of ammonia was evaluated when expod to 360mL min À1of the carrier air.Textural characterization
Textural characterization was carried out by measuring the N 2adsorption isotherms at À196 C.Befor
e the experiments,the samples were outgasd under vacuum at 120 C.The isotherms were ud to calculate the specific surface area,S BET ,total pore volume,V t ,volume of micropores,V mic ,volume of mesopores,V mes ,and pore size distributions.The latter was calculated using the density functional theory (DFT).44Surface pH
The pH of a sample suspension provides information about the acidity and basicity of the surface.About 0.15g of the initial and exhausted MOF-5,GO or MOF-5–GO powder was added to 7.5mL of distilled water.The suspension was stirred overnight to reach equilibrium before recording the pH.SEM
Scanning electron microscopy was performed on a Zeiss Supra 55instrument.The instrument has a resolution of 5nm at 30kV.Scanning was performed on a sample powder previously dried and sputter coated with a thin layer of carbon to avoid charging.TEM
Transmission electron microscopy (TEM)was performed on a Zeiss EM 902instrument.The microscope has a line resolution of 0.34nm and a point resolution of 0.5nm and operates in normal diffraction and low do modes at 50or 80kV.Analys were performed after the samples were resuspended in ethanol.
P u b l i s h e d o n 15 J u l y 2009. D o w n l o a d e d b y J i l i n U n i v e r s i t y o n 29/11/2013 00:03:07.
XRD
X-Ray diffraction(XRD)measurements were conducted using standard powder diffraction procedures.Adsorbents were ground with DMF in a small agate mortar.The mixture was smear-mounted onto a glass slide and then analyzed by Cu K a radiation generated in a Philips X’Pert X-ray diffractometer.
A standard glass slide was run for the background correction. FTIR
Fourier transform infrared(FTIR)spectroscopy was carried out using a Nicolet Magna-IR830spectrometer using the attenuated total reflectance(ATR)method.The spectrum was generated and collected16times and corrected for the background noi. The experiments were done on the powdered samples,without KBr addition.
Results and discussion
Metal–organic frameworks or graphite oxides are expected to have a distinct structure,which provide
sfingerprints of their textural and chemical nature.To ensure that the MOF-5–GO composite has the elements of both components,the X-ray diffraction patterns should be analyzed.They are collected in Fig.1.For GO,the well-defined peak at2Q about9.29 repre-nts an interlayer distance of9.50A˚.For MOF-5,various sharp diffraction peaks are en which are characteristic of this mate-rial structure and are in agreement with the data published in the literature.45–47The XRD pattern of the nanocomposite does not differ significantly from that for MOF-5.Only the sharpness of the peak at2Q about9.7 is slightly reduced and the splitting of that peak is noticed.This split was obrved by Lillerud and co-workers on MOF-5and attributed to a distortion of the cubic symmetry.46Finding it for our composite suggests that the prence of GO in the sample increas the distortion in the MOF-5cubic arrangement,as one could expect owing to addi-tional constraints in the degrees of freedom during synthesis. Nevertheless,X-ray analys indicate that the major structural and chemical features of MOF-5are prerved in MOF-5–GO.The predominant features of MOF-5are expected since it consists of95wt%of the nanocomposite content.
The texture of the materials studied is en on SEM micro-graphs prented in Fig.2.For MOF-5,besides the well-defined cubic crystals,some remains of an amorphous pha can be distinguished.The particles of GO look very den with the layers stacked together as a result of disp
ersive forces and strong specific interactions between the surface groups on the graphene-like layers.On the other hand,MOF-5–GO exhibits totally different surface features.The layers of sandwich-like structures can be clearly en and the regular structure of layers suggests that they might be layers of MOF-5crystallites parated by the layers of GO.In spite of the fact that only5wt%GO were added during the MOF-5synthesis,its effect on the texture of materials is very pronounced.It is interesting that in the synthesis crys-tallites of similar sizes are formed and then restacked into larger particles with GO layers acting as dividers.The differences in the texture between MOF-5–GO composite,MOF-5and GO precursors are also visible in TEM micrographs(Fig.3).On MOF-5–GO,ordered layered units with random orientation
can Fig.1X-Ray diffraction patterns for the parent materials and the
nanocomposite.
Fig.2SEM micrographs for the parent materials and the nano-
composite:(a)MOF-5,(b)GO,(c)and(d)
MOF-5–GO.
Fig.3TEM micrographs for the parent materials and the nano-
composite:(a)MOF-5,(b)GO,(c)and(d)MOF-5–GO.
P
u
b
l
i
s
h
e
d
o
n
1
5
J
u
l
y
2pasio
9
.
D
o
w
n
l
o
a
d
e
d
b
y
J
i
l
i
n
U
n
i
v
e
r
to be by your sides
i
t
y
o
n
2
9
/
1
1
/
2
1
3
:
3
:
7
.
be distinguished.We hypothesize,taking into account the chemistries of both components of nanocoalongside
mposites,that the building process of the ordered structure of MOF-5–GO is bad on the attachments of MOF-5‘‘blocks’’to a graphene layer by reactions with epoxy groups on GO.This follows the analysis of water interactions with MOF-5prented by Greathou and Allendorf48where a replacement of an oxygen atom in ZnO4 tetrahedron by an oxygen atom from water is a very important step leading to the decomposition of MOF-5by water.In our view oxygen atoms in epoxy groups on GO play the same role as oxygen atoms in water.49In the subquent steps of the nano-composite building process,an alternation between attachment of graphene layers and MOF-5‘‘blocks’’takes place.
The nitrogen adsorption isotherms measured on MOF-5and MOF-5–GO exhibit a typical Langmuir-type shape suggesting the predominant microporosity(Fig.4).The surface of GO is inaccessible for the nitrogen molecule and thus this material is considered as nonporous.22The parameters of the porous structure calculated from the isotherms and pore size distri-
butions(PSDs)are prented in Table1and Fig.5, respectively.Although the surface area of MOF-5is smaller than 3000–3500m2gÀ1,obrved for some MOF-5,26,45surfaces of similar magnitude to our material were reported by Huang and co-workers,and Panella and Hirscher.47,50This low surface area can be related to the method of outgassing during the prepara-tion and thus the completeness
of solvent removal.The solvent ud to prepare the materials as well as the temperature at which crystals were formed might also have an influence on their resulting porosity.38,45The materials obtained are predominantly microporous with pore size between5and10A˚(Fig.5),which is in agreement with the structural model of cavities in MOF-5in which the passage of spheres of diameter up to8.0A˚has been defined.42Pores with sizes between16.0and23.5A˚are also found for our materials and they might be due to a distortion in the structure of MOF-5.When the nanocomposite is formed,the structural parameters decrea by about10%,and the pore size distribution is prerved.Results for PSDs must be considered with caution since the pore model ud for DFT calculation reflects the slit-shaped pores of carbonaceous materials.44 Nevertheless,since the same model is ud for the ries of materials the trends in PSDs can be analyzed.
The ammonia adsorption capacities calculated from the breakthrough curves(Fig.6)in milligrams per gram of the materials and in milligrams per unit volume of the adsorbent bed and the surface pH values for the initial and exhausted samples
Table1Parameters of porous structure calculated from nitrogen adsorption isotherms
Sample S BET/
m2gÀ1
V t/
cm3gÀ1
V meso/
cm3gÀ1
V mic/
cm3gÀ1
V mic/
V t
MOF7930.4080.0230.3850.94 MOF–ED7390.3990.0100.3890.97 MOF–EM100.0570.0520.0050.09 MOF–GO7060.3650.0240.3410.93 MOF–GO–ED7100.3650.0240.3410.93 MOF–GO–EM80.0250.0210.004
0.16Fig.5Pore size distributions for the initial
samples.
Fig.4Nitrogen adsorption isotherms for the initial
samples.
Fig.6Ammonia breakthrough curves and desorption curves for MOF-
5,GO and MOF-5–GO.
P
u
b
l
i
s
h
e
d
o
n
1
5
J
u
l
y
2
9
.
D
o
w
n
l
o
a
d
e
d
b
y
J
i
l
i
n
U
n
i
v
e
r
s
i
t
y
o
n
2
9
/
1
1
/
2
1
人教版高一英语教材3
:
3
:
7
.
are summarized in Table 2.In this table,we also list the capacities calculated by considering the physi
cal mixture of the adsorbents (5wt%GO and 95wt%MOF-5).It has to be noted that the error in the NH 3breakthrough capacities is around 10%.Even though the capacity in moist conditions on the MOF-5–GO is smaller than that on GO,the value obtained is about 12%greater than that expected when the structural synergy between the compo-nents of the nanocomposite does not exist.High adsorption on GO was explained by interactions of ammonia with acidic groups and its intercalation between the distorted graphitic layers.6,22In dry air,the performance of MOF-5is very poor and the capacity measured is consistent with that reported by Yaghi et al.36That low capacity is a result of the lack of strong chemical interactions between NH 3and MOF compound 36and relatively large pores
compared to the size of the ammonia molecule (3A
˚51).The much higher capacity in wet conditions must be related to adsorption of large quantities of water on MOF-552and dissolution of ammonia in the water prent in the pore space.Another possible scenario is a change in the mechanism of adsorption caud by the formation of ammonium ions simultaneously with changes in the chemistry of materials caud by water 50as will be discusd later in this paper.The low capacity of the composite at dry conditions is governed by the behavior of the predominant pha of MOF-5.The prence of GO in the composite slightly incre
as the performance but the analysis is difficult owing to a very short breakthrough time.
In spite of the lower NH 3removal capacity in wet conditions on MOF-5than that on GO,the performance is still better than that on unmodified activated carbons,52–54comparable to tho on carbons modified with metal chlorides,23and slightly better than that measured on carbons modified with poly-oxometalates 55,56or metal oxides.57–59Since the capacity is high even though the pH is much higher than that for GO,other mechanisms than simple acid–ba interactions must be involved in the retention of ammonia on tho materials.It is interesting that on the exhausted samples (for MOF-5and MOF-5–GO),the pH increas by only one pH unit after exposure to ammonia in spite of the high capacity.For this,a neutralization reaction or removal of a significant amount of ammonia dissolved in water by purging with dry air after the adsorption process is the possible explanation.
Analysis of the shapes of the NH 3breakthrough curves and the desorption curves indicates the differences in the perfor-mance of materials.It has to be mentioned here that due to the nsor performance limitations,the desorption tests were per-formed only for 2h after the adsorption tests and that this period might not be sufficient to obrve the full desorption.The
the final destination
relatively small area under the desorption curve in the ca of GO run in moist and dry conditions indicates a significant quantity of strongly adsorbed ammonia.22For MOF-5,initially a sharp decrea in the concentration is noticed on the desorption curve and then ammonia concentration in the air stream stays at a more or less constant level during the duration of the desorp-tion experiment.A similar pattern is noticed for the nano-composite.This must be related to the removal of water with dissolved ammonia.That process is expected to last a long time when very small size pores are prent.60All the are signs of a weak retention of ammonia.Moreover,as the parameters of porous structure indicate,after ammonia adsorption in moist conditions,the materials (MOF-5and MOF-5–GO)become practically nonporous (e Table 1).This loss of porosity is due to a collapsing of the structure,as a result of water exposure,as already obrved by Long and co-workers.45As mentioned above,Greathou and Allendorf showed that water leads to the destruction of the MOF-5structure owing to its specific inter-actions with the zinc oxide clusters.48Oxygen atoms in water progressively replace oxygen atoms in ZnO 4tetrahedra.Hydrogen bonding between hydrogen atoms in water and oxygen atoms in ZnO 4tetrahedra reprents another way of interactions leading to the collap of the MOF-5structure.That destruction might affect the kinetics of water desorption (with dissolved ammonia).It is also possible that the destruction happens gradually when ammonia is prent in the system.Indeed,ammonia might compete with water for reaction sites owing to some si
milarities in the chemistries of the two species.In particular,hydrogen bonding between hydrogen atoms in NH 3and oxygen atoms in ZnO 4tetrahedron appears as a plau-sible scenario.Support for this might be that obrved increa in the ammonia concentration at the end of the desorption experi-ment.It is interesting that the structure does not collap when it is expod only to ammonia.
Another interesting feature on the breakthrough curves of the composite tested in moist conditions is a well-pronounced change in the shape of the curve with the progress of adsorption starting at a concentration of 50ppm.This behavior is real and is not en for GO or MOF-5.This suggests changes in the mech-anism of adsorption or appearance of additional adsorption centers with the duration of the experiment.The changes in the chemistry can be only caud by either water or ammonia,or ammonia and water.什么是宾语从句
How ammonia changes the chemistry of the composite is en on the FTIR spectra prented in Fig.7.The spectra for GO before and after adsorption of ammonia were analyzed in detail previously.58,61,62As expected bad on the composition of the
catch22Table 2Measured ammonia breakthrough capacity,calculated hypothetical capacity and the surface pH for the initial and exhausted samples
Sample NH 3breakthrough capacity
Calculated hypothetical capacity mg g À1of material pH mg g À1of material mg cm À3of material Initial Exhausted GO-ED 55.536.9— 2.47 6.24GO-EM 61.039.8— 2.47 6.66MOF–ED 5.9 2.9— 5.64 5.80MOF–EM 42.523.0— 5.64 6.63MOF–GO–ED 6.9 3.38.4 6.09 6.10MOF–GO–EM
53.3
28.8
43.4
6.09
7.06
P u b l i s h e d o n 15 J u l y 2009. D o w n l o a d e d b y J i l i n U n i v e r s i t y o n 29/11/2013 00:03:07.
男生英文名字大全
