Ordered nanoporous polymer-carbon composites

Ordered nanoporous polymer-carbon composites

(Parte 1 de 2)

nature materials| VOL 2| JULY 2003| w.nature.com/naturematerials 473

Nanostructured organic materials,particularly those constructed with uniform nanopores,have been sought for a long time in materials science1–7.There have been many successful reports on the synthesis ofnanostructured organic materials using the so-called,‘supramolecular liquid crystal templating’route8–13.Ordered nanoporous polymeric materials can also be synthesizedthrough a polymerization route using colloidal14,15or mesoporous silica16,17templates.The organic pore structures constructed by these approaches,however,are lower in mechanical strength and resistance to chemical treatments than nanoporous inorganic,silica and carbon materials.Moreover,the synthesis ofthe organic materials is yet oflimited success in the variation ofpore sizes and structures,whereas a rich variety of hexagonal and cubic structures is available with tunable pore diameters in the case ofthe inorganic materials18–20.Here we describe a synthesis strategy towards ordered nanoporous organic polymers, using mesoporous carbon as the retaining framework. The polymer–carbon composite nanoporous materials exhibit the same chemical properties ofthe organic polymers,whereas the stability ofthe pores against mechanical compression,thermal and chemical treatments is greatlyenhanced.The synthesis strategy can be extended to various compositions ofhydrophilic and hydrophobic organic polymers,with various pore diameters, connectivity and shapes.The resultant materials exhibiting surface properties ofthe polymers,as well as the electric conductivity ofthe carbon framework,could provide new possibilities for advanced applications.Furthermore,the synthesis strategy can be extended to other inorganic supports such as mesoporous silicas.

Ordered nanoporous polymer–carbon composites

MINKEE CHOI AND RYONG RYOO* National Creative Research Initiative Center for Functional Nanomaterials,and Department of Chemistry (School of Molecular Science-BK21), Korea Advanced Institute of Science and Technology,Daejeon,305-701,Korea *e-mail:rryoo@kaist.ac.kr

Published online:2 June 2003; doi:10.1038/nmat923

Figure1The synthesis route to nanoporous polymer–carbon composite materials.a,Transmission electron microscope image of the CMK-3 mesoporous carbon,showing the hexagonal array of carbon nanorods.These rods are rigidly interconnected with uniform spacing using carbon spacers,which are random in thickness and location along the direction of the nanorods.b,Structural model of the carbon nanorods with microporosity (carbon spacers are omitted).c,The impregnation of organic monomers into CMK-3 carbon.The monomer loading is controlled so that the monomers should fill micropores and form a thin film on the nanorod surface,without causing capillary condensation in the mesopores between nanorods. d,Polymerization of monomers into crosslinked polymers.The resultant polymers form interpenetrating,inseparable composite frameworks with carbon.

50 nm dcba

© 2003 Nature Publishing Group

474 nature materials| VOL 2| JULY 2003| w.nature.com/naturematerials

The present synthesis strategy is to impregnate ordered mesoporous (or nanoporous) carbon materials21–25that are highly microporous in the mesopore walls (or frameworks) with organic monomers.The impregnation is controlled so that the micropores are completely filled with organic monomers,and also that the mesopore walls are coated with a thin layer ofmonomers.The monomers are subsequently converted to crosslinked polymers.The organic nanopores thus constructed exhibit highly enhanced stability,due to the formation ofinterpenetrating composite frameworks between the crosslinked polymers and carbon (see Fig.1).

The ordered mesoporous carbons suitable for this purpose are available with various pore arrangements (hexagonal and cubic structures) and diameters (typically,ranging from 2 to 10nm), through the synthesis route using mesoporous silica templates21–25. The structure ofthe mesoporous carbons is characterized by the three-dimensional interconnection ofnanorods or nanobeads, which seem to be uniform in diameter,typically in the range of several nanometres.The regular nanoscale spacings between the nanoframeworks impart these materials,referred to as CMK,with uniform mesoporosity.The carbon frameworks can be fabricated with high microporosity as illustrated in Fig.1.These micropores are the preferred adsorption sites for mesopores for many organic adsorbates.Organic adsorbates start filling the mesopores by capillary condensation after the micropores in the mesopore walls have been completely filled,and the wall surface is coated with a thin film ofadsorbates.Based on these adsorption phenomena,we have designed the synthesis procedure for a crosslinked polystyrene(PS)-CMK-3 composite (see Methods).We have confirmed that this procedure can be generalized for various hydrophilic and hydrophobic organic polymers,such as crosslinked poly(methyl methacrylic acid) and poly(2- hydroxyethyl methylmethacrylate),whereas the CMK carbons have high affinity with a wide variety ofmonomers ofdifferent hydrophobic nature.The polymer content can be increased to as much as 45 wt% ofcarbon without causing mesopore blockage. Our synthesis results have been confirmed by gravimetric analysis ofthe sample weight,pore-size distribution measurement,and most evidently,X-ray powder diffraction (XRD) intensity.

Generally,the XRD pattern for ordered mesoporous materials is composed ofdistinct diffraction lines appearing at low angles (typically, where 2θ< 5°),which are used for the determination ofthe structure. These XRD lines decrease in intensity as guest species are loaded in the mesopores,and such an intensity change is a useful means ofjudging the location ofthe guest species.However,contrary to most cases,our CMK carbons exhibit a significant increase in the diffraction intensity when organic polymers are synthesized within the structure following the procedure described in the Methods section.The XRD result for the PS in CMK-3 carbon is shown in Fig.2a.The XRD data indicate that the polymerization inside micropores results in an increase ofthe apparent density ofthe mesopore walls.This distinctive phenomenon is similar to the intensity increase in the neutron-diffraction when N2is preferentially adsorbed,at low pressure,inside micropores that are present in the mesopore walls ofsome mesoporous silicas26,27. The mesopore vacancy in the carbon–polymer composite has been confirmed by pore-size analysis using the Barrett–Joyner–Halenda (BJH) algorithm (Fig.2b).For example,the CMK-3 carbon indicates only a small decrease from 4.2 to 3.9 nm in the median mesopore diameter,even when the PS loading is increased up to 45 wt% ofcarbon. We have measured the mesopore volume corresponding to the capillary condensation around P/P0= 0.4 (see inset ofFig.2b).This experiment revealed that the polymer loading caused the mesopore volume corresponding to 2–10 nm pores to decrease by only 7% (based on the same carbon weight),despite the 45 wt% polymer-loading.

Despite the highly porous architecture,the CMK carbons exhibit high mechanical stability under compression.Interestingly,the mechanical stability increases further with the incorporation of polymers (see Supplementary Information),showing a synergistic enhancement between the carbon framework and the polymer.The PSCMK-3 composite material maintained the highly ordered nanostructure after being pressed for 10 min at 1,500MPa.In addition, the structure was maintained after heating at 150°C,or washed with chloroform,as shown by the same XRD pattern and N2adsorption isotherm as those for the pristine sample.The high stability is a remarkable advantage in that it permits safe chemical functionalization under harsh conditions,such as heating in strong acids and solvents. For example,the polymer in the PS-CMK-3 system could be sulphonated using concentrated sulphuric acid at 100°C (1.0 g ofthe

Intensity PS-CMK-3


Pore size distribution (cm

–1 nm

Volume (ml g


Figure2X-ray powder diffraction (XRD) patterns and pore-size distributions for CMK-3 mesoporous carbon and its composite material with crosslinked polystyrene (PS-CMK-3,45 wt% polymer-loading on carbon).a,The XRD intensity of the polymer–carbon composite is significantly increased from that of the original carbon, due to the polymer filling in micropores of the carbon frameworks.b,The pore-size distribution of CMK-3 and PS-CMK-3 determined by Barrett-Joyner-Halenda (BJH) analysis of the Nadsorption isotherm (inset).The mesopore diameters changed from 4.2 to 3.9 nm with the polymer loading.P= Npressure in contact with the sample; P= vapour pressure of liquid Nat 7 K.

© 2003 Nature Publishing Group

nature materials| VOL 2| JULY 2003| w.nature.com/naturematerialss 475 composite was stirred for 4 h at 100°C in 20 ml 98% sulphuric acid). The resulting product exhibited the specific BET surface area of 650m2g–1,and about the same pore-size distribution as the material before the sulphonation.Titration with NaOH revealed that the sulphonated PS-CMK-3 contained acid groups amounting to 3.5×10–3molg–1polymer (after subtraction of10.3% ofthe total value for the acid density ofCMK-3 treated under the same condition). This result is very remarkable for a PS containing as much as 20 mol% divinylbenzene (DVB),whereas the ion-exchange capacity ofordinary commercial crosslinked PS resins ranges from 1.8–4.8 ×10–3mol g–1,and decreases rapidly as the DVB content exceeds the commercial range of 2–8 mol%.

The PS-CMK-3 nanoporous material has quite a distinct surface nature,compared with the mesoporous carbon.As shown in Fig.3, a column packed with sulphonated PS-CMK-3 gave different eluate concentrations depending on the solvent composition.In the case of CMK-3 carbon,the dye was so strongly adsorbed that it was difficult to remove from the column at any solvent composition.The dye adsorption data indicate that the nanopore walls are effectively coated with a thin layer ofpolymer.Interestingly,such a thin layer ofpolymer does not cause a significant decrease in the electric conductivity.This has been confirmed by the comparison ofthe electric conductivity between PS-CMK-3 (unsulphonated) and CMK-3 (Fig.4).These properties may offer a remarkable advantage for applications to functional electrode materials such as amperometric biosensors28,29.The ionexchanging polymer–carbon composite may also provide unique opportunities for the design offuel-cell electrode materials that can support high dispersions ofplatinum nanoparticles24and possess high ionic conductivity simultaneously.

The present strategy to support organic polymers on carbon frameworks may be extended to other nanoporous materials such as mesoporous silicas and colloidal crystals,as long as the supported polymer layer can be anchored to the nanopore walls.For example,the SBA-15 mesoporous silica can be synthesized with highly microporous frameworks.We have confirmed that various polymers,such as crosslinked poly(2-hydroxyethyl methacrylate) and poly(methyl methacrylate) can be supported onto such SBA-15 silicas (M.Choi and R.Ryoo,manuscript in preparation).The resultant polymer–silica composite materials possess uniform mesoporosity,and the XRD intensity increases,similar to the aforementioned case ofPS-CMK. In addition,the composite materials exhibit significantly enhanced hydrothermal stability in boiling water.However,it is noteworthy that microporosity ofthe mesoporous silica walls varies widely depending on the details ofthe synthesis conditions.In the case ofsilica pore walls with low microporosity,we have failed to form a film ofpolymers without causing pore blockage.

In conclusion,we propose that the present strategy should provide a versatile route to nanoporous polymer–carbon composites with various compositions and pore architectures.The polymer–carbon composite materials can be obtained with uniform porosity,various pore sizes (typically,in the range 2–10nm),shapes and connectivity, taking advantage ofthe structural variation ofthe carbon supports. High thermal and chemical stabilities ofthe polymer-like pore walls supported on the carbon frameworks is an advantage for functionalization.The polymer–carbon composite can have the surface

Direct blue 15 NaO ONa NaO

Fraction of d ye absorbed


Sulphonated PS-CMK-3

Figure3Elution of ‘Direct Blue 15’ (an organic dye with 3.3 nm rod-like molecular geometry) after adsorption on CMK-3 and sulphonated PS-CMK-3 composite.The adsorption was performed from a dye solution prepared by using ethanol–water mixture as a solvent(1 ×10M,0.1 ml),at the top of a column of 0.7cm diameter,and packed with 0.3 g sample.The adsorbed dye was eluted by flow of the ethanol–water mixture at the rate of 0.1 ml min.The dye concentration in the first 2ml eluent was analysed by the light absorption at 600 nm wavelength (JASCO V-530 ultraviolet/visible spectrophotometer).

Figure4The electric conductivities of CMK-3 mesoporous carbon and PSCMK-3,plotted as a function of compressive pressure at room temperature. The conductivities were measured with a home-built apparatus following the fourpoint probe method.The conductivity of CMK-3 does not decrease significantly, even after loading of 45wt% polystyrene.


© 2003 Nature Publishing Group

476 nature materials| VOL 2| JULY 2003| w.nature.com/naturematerials nature ofpolymers while the electric conductivity ofthe original carbon is maintained,which suggests new application possibilities as advanced electrode materials.Furthermore,the present synthesis strategy can be extended from carbon to porous silica materials.Owing to their properties such as high surface area,regular pore structure,high capacity for metal dispersion and facile chemical functionalizability,we believe that the nanoporous composite system could find broad applications in many areas ofmaterials science,including separation of biomaterials, removal of pollutants, selective ion e xchange, and manufacturing ofhigh-performance catalysts and sensors.

MATERIALS SYNTHESIS The ordered mesoporous carbon referred to as CMK-3 was synthesized by pyrolysis ofsucrose inside the SBA-15 mesoporous silica,following the procedure reported elsewhere.The silica template was subsequently removed by washing with HF solution twice,and evacuated for the removal ofmoisture at 300°C.Divinylbenzene (DVB,Aldrich,80%) and styrene (Janssen,9%) were purified through an alumina column to remove polymerization inhibitors.1 g CMK-3 was impregnated with a mixture of 0.356 g DVB/0.112 g styrene/0.026 g 2,2′-azobisisobutyronitrile/1.09 g methylene chloride.After the methylene chloride was evaporated in a drying oven for 3 h at 60°C,the CMK-3 sample containing the organic monomers and the polymerization initiator was sealed under argon atmosphere.The sample was subsequently heated as sealed for two days at 150°C for polymerization.The sample was thoroughly washed with chloroform and dried in a vacuum oven.The polymer content in the organic-carbon composite thus obtained was analysed from the increase in sample weight due to the organic incorporation.This analysis revealed that the sample weight increased by 45 wt% ofthe carbon weight, which indicates that over 90% ofthe organic monomers were polymerized to form an inseparable composite with carbon frameworks.

It should be noted that the use ofmethylene chloride and its subsequent evaporation was an effective way to achieve polymerization in the desired locations.The solvent was useful to achieve uniform distribution ofthe small amount ofmonomers and initiator that amounted to only 40% ofthe pore volume.When the solvent was not removed,the polymerization resulted in mesopore blockage.The XRD intensity decreased as a result ofthe polymer location in the mesopores,and no dye adsorption occurred at any solvent compositions between water and ethanol due to mesopore blockage.

MEASUREMENTS XRD patterns were recorded by using a Rigaku multiflex diffractometer equipped with CuKαradiation (40 kV,50 mA).XRD scanning was performed under ambient conditions over the 2θregion 0.7–5°at steps of0.01°and accumulation time 5 s per step.TEM images were obained with a JEOL JEM-4000EX operated at 400 kV.Samples for TEM measurements were suspended in ethanol and supported on a carbon-coated copper grid.Nadsorption isotherms were measured at 7 K using a Quantachrome AS-1MP volumetric adsorption analyser.Before the adsorption measurements,all samples were outgassed for 24h at 333K in the degas port ofthe adsorption analyser.

Received 10 January 2003; accepted 23 April 2003; published 2 June 2003.

References 1.Muthukumar,M.,Ober,C.K.& Thomas,E.L.Competing interactions and levels ofordering in selforganizing polymeric materials.Science277,1225–1232 (1997). 2.Langley,P.J.& Hulliger,J.Nanoporous and mesoporous organic structures:new openings for materials research.Chem. Soc. Rev. 28, 279–291 (1999). 3.Wulff,G.Molecular imprinting in crosslinked materials with the aid ofmolecular templates-a way towards artificial antibodies.Angew.Chem.Int.Edn Engl.34,1812–1832 (1995). 4.Buchmeiser,M.R.New ways to porous monolithic materials with uniform pore size distribution.

Angew.Chem.Int.Edn Engl. 40, 3795–3797 (2001). 5. Peters,E. C.,Svec,F.& Fréchet,J. M. J.Rigid macroporous polymer monoliths.Adv. Mater. 1, 1169–1181 (1999).

6. Widawski,G.,Rawiso,M.& François,B.Self-organized honeycomb morphology ofstar-polymer polystyrene films.Nature 369, 387–389 (1994). 7.Chen,Y.Y.,Ford,W.T.,Materer,N.F.& Teeters,D.Facile conversion ofcolloidal crystals to ordered porous polymer nets.J.Am.Chem.Soc.122,10472–10473 (2000). 8. Smith,R. C.,Fischer,W. M.& Gin,D. L.Ordered poly(p-phenylenevinylene) matrix nanocomposites via lyotropic liquid-crystalline monomers.J.Am.Chem.Soc.119,4092–4093 (1997). 9.Pindzola,B.A.,Hoag,B.P.& Gin,D.L.Polymerization ofa phosphonium diene amphiphile in the regular hexagonal phase with retention ofmesostructure.J.Am.Chem.Soc.123,4617–4618 (2001). 10.Zalusky,A. S.,Olayo-Valles,R.,Taylor,C. J.& Hillmyer,M. A.Mesoporous polystyrene monoliths.

J.Am.Chem.Soc. 123, 1519–1520 (2001). 1.Zalusky,A. S.,Olayo-Valles,R.,Wolf,J. H.& Hillmyer,M. A.Ordered nanoporous polymers from polystyrene-polylactide block copolymers.J.Am.Chem.Soc. 124, 12761–12773 (2002). 12.Antonietti,M.,Caruso,R. A.,Göltner,C. G.& Weissenberger,M. C.Morphology variation of porous polymer gels by polymerization in lyotropic surfactant phases.Macromolecules32,1383–1389 (1999). 13.Lee,H.K.et al.Synthesis ofa nanoporous polymer with hexagonal channels from supramolecular discotic liquid crystals.Angew.Chem.Int.Edn Engl.40,2669–2671 (2001). 14.Johnson,S.A.,Ollivier,P.J.& Mallouk,T.E.Ordered mesoporous polymers oftunable pore size from colloidal silica templates.Science283,963–965 (1999). 15.Liu,L.,Li,P.S.& Asher,S.A.Entropic trapping ofmacromolecules by mesoscopic periodic voids in a polymer hydrogel.Nature 397, 141–144 (1999). 16.Göltner,C.G.& Weissenberger,M.C.Mesoporous organic polymers obtained by ‘twostep nanocasting’Acta Polym. 49, 704–709 (1998). 17.Kim,J.Y.,Yoon,S.B.,Kooli,F.&Yu,J.S.Synthesis ofhighly ordered mesoporous polymer networks.

J.Mater.Chem. 1, 2912–2914 (2001). 18.Corma,A.From microporous to mesoporous molecular sieve materials and their use in catalysis.

Chem.Rev. 97, 2373–2419 (1997). 19.Ying,J. Y.,Mehnert,C. P.& Wong,M. S.Synthesis and applications ofsupramolecular-templated mesoporous materials.Angew.Chem.Int.Edn Engl. 38, 56–7 (1999). 20.Barton,T.J.et al.Tailored porous materials.Chem.Mater.1,2633–2656 (1999). 21.Ryoo,R.,Joo,S.H.& Jun,S.Synthesis ofhighly ordered carbon molecular sieves via templatemediated structural transformation.J.Phys. Chem.B 103, 7743–7746 (1999). 2.Jun,S.et al.Synthesis ofnew,nanoporous carbon with hexagonally ordered mesostructure.J.Am.

Chem.Soc.122, 10712–10713 (2000). 23.Ryoo,R.,Joo S. H.,Kruk,M.& Jaroniec,M.Ordered mesoporous carbons.Adv.Mater. 13, 677–681 (2001). 24.Joo,S.H.et al.Ordered nanoporous arrays ofcarbon supporting high dispersions ofplatinum nanoparticles.Nature 412, 169–172 (2001). 25.Lee,J.S.,Joo,S.H.& Ryoo,R.Synthesis ofmesoporous silicas ofcontrolled pore wall thickness and their replication to ordered nanoporous carbons with various pore diameters.J.Am.Chem.Soc.124, 1156–1157 (2002). 26.Ramsay,J.D.F,Kallus,S.& Hoinkis,E.SANS characterisation ofmesoporous silicas having model structures.Stud. Surf. Sci. Catal. 128, 439–448 (2000). 27.Smarsly,B.,Goltner,C. G.,Antonietti,M.,Ruland,W.& Hoinkis,E.SANS investigation ofnitrogen sorption in porous silica.J.Phys.Chem.B105,831–840 (2001). 28.Wang,J.,Musameh,M.& Lin,Y.Solubilization ofcarbon nanotubes by nafion towards the preparation ofamperometric biosensors.J.Am.Chem.Soc. 125, 2408–2409 (2003). 29.Park,J.K.et al.In vivo nitric oxide sensor using non-conducting polymer-modified carbon fiber.

Biosens. Bioelectron. 13, 1187–1195 (1998). 30.Smits,F.M.Measurement ofsheet resistivities with the four-point probe.Bell System Tech.J.711 (May,1958).

Acknowledgements This work was supported in part by the Creative Research Initiative Program ofthe Korean Ministry of Science and Technology,and by the School ofMolecular Science through Brain Korea 21 project.R.R. thanks O.Terasaki at Stockholm university for TEM measurements and helpful discussions. Correspondence and requests for materials should be addressed to R.R. Supplementary Information accompanies the paper on w.nature.com/naturematerials

Competing financial interests The authors declare that they have no competing financial interests.

© 2003 Nature Publishing Group


(Parte 1 de 2)