1

NANDI et al.: FORMATION OF MESOPORES IN RESORCINOL-FORMALDEHYDE COMPOSITE RESIN

Formation of mesopores in resorcinol-formaldehyde composite resin

Mahasweta Nandi, Krishanu Sarkar & Asim Bhaumik*

Department of Materials Science and Centre for Advanced Materials
Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700 032, India

Email:

Received 10 January 2008; revised 12 May 2008

A new nanostructured resorcinol-formaldehyde material has been synthesized by hydrothermal condensation of resorcinol and formaldehyde at 363 K under mild alkaline condition in the presence of supramolecular assembly of cationic surfactant, cetyltrimethylammonium bromide as structure directing agent. The material has been characterized by powder
X-ray diffraction, N2 sorption, transmission and scanning electron microscopy , thermogravimetric and differential thermal analysis and UV-vis spectroscopy. The X-ray and TEM image analyses reveal disordered wormhole-like mesostructure with pores of ca. 2.5 nm. These composite materials exhibit photoluminescence property at room temperature, which may be utilized for the fabrication of novel organic optical devices.

1

NANDI et al.: FORMATION OF MESOPORES IN RESORCINOL-FORMALDEHYDE COMPOSITE RESIN

Synthesis of mesoporous materials usually followstemplating route, involving 2D or 3D supramolecular assembly (true liquid crystalline phases)1 formed by surfactant molecules as the core, around which the desired solid matrix is formed. Subsequent removal of the templating molecules from this composite can generate mesoporosity, and the size of the pores generated depends on the dimension of the self-assemblies of the surfactant molecules. Since the first report on the synthesis of ordered mesoporous silica, MCM-41/48, by Mobil researchers2 in 1992, the synthesis of ordered nanostructured materials by soft templating approach has been widely practiced. Mesoporous materials synthesized so far usually have very high surface area and tunable pore diameters vis-à-vis related microporous materials3, and hence have attracted widespread attention for adsorption, exchanger, catalysis and application as hosts for the synthesis of nanomaterials. Functionalized mesoporous molecular sieves have also found enormous potential applications in the field of catalysis4-7, adsorption8, electronics9 and sensors10. In recent times mesoporous carbons11 synthesized by pyrolyzing the mesophase pitch and resins have attracted significant interest in this context, whereas very little work has been done on purely organic porous nanostructured materials12. On the other hand considerable attention has been given to the synthesis of mixed
cresol-formaldehyde13-15 or resorcinol-formaldehyde
resins16-18, organic aerogels and their corresponding pyrolyzed carbon aerogels. Owing to their controllable pore size distribution, high surface area, low electrical resistivity and low thermal conductivity, these are efficient materials as electrodes for super capacitors and fuel cells19, catalyst supports and high temperature insulation materials20. Aerogels derived from sol-gel polycondensation of cresol or rescorcinol with formaldehyde under basic condition have been studied in detail16. In general, preparation of carbon aerogel proceeds through three steps: sol-gel polycondensation of an aqueous solution of cresol or rescorcinol with formaldehyde to form wet organic aerogel, supercritical or ambient pressure drying of the wet organic aerogel and finally pyrolysis of the organic aerogel to the carbon aerogel. Carbon aerogels prepared from resorcinol-formaldehyde organic aerogels synthesized in the presence of a neutral block copolymeric template, (Pluronic F127) has been reported21. The various carbon aerogels obtained have been found to have appreciable surface areas, which can be varied by changing the conditions.Also, the variation of the pore size was large, from 2-100 nm. In the present work, we have not attempted to prepare the carbon aerogel of the resorcinol-formaldehyde composite by pressure drying and pyrolysis. Rather, we have prepared the mesostructured resol material using the supramolecular assembly of cationic surfactant in situ in the synthesis of gel and thereafter hydrothermal treatment of the same. This was followed by the removal of the structure directing agent (SDA) by solvent extraction to make the pores free from the surfactants and obtain the corresponding all-organic porous framework material. Recently, we have synthesized mesoporous organic frameworks using ionic surfactant and mixture of non-ionic and ionic surfactants22. The resorcinol-formaldehyde nanostructured materials thus synthesized were found to have photoluminescence property, and hence, can be applied to design novel optical devices, which could open up new potential applications for these all-organic porous materials.

Materials and Methods

Cationic surfactant, cetyltrimethylammonium bromide, (CTAB, Loba Chemie) was used as the SDA. Organic precursor gels were synthesized via polycondensation of resorcinol (Loba Chemie) with formaldehyde, HCHO (E-Merck, 37%) in an aqueous alkaline solution of NaOH (E-Merck). In a typical synthesis, 3.64 g of CTAB (0.01 mole) was first dissolved in 20 ml of water. To it 1.6 g of sodium hydroxide (0.04 mole) dissolved in 10 ml of water was added under stirring condition. When the mixture became homogeneous, 3.24 g of HCHO (0.04 mole) was added to this solution. Finally, 2.2 g of resorcinol (0.02 mole) dissolved in 15 ml of water was added dropwise to it under constant vigorous stirring. After stirring for 2 h at room temperature the gel was transferred to an airtight polypropylene bottle and kept at 363 K for 3 days. The solid product was then obtained through filtration, repeated washing with water and drying under vacuum at ambient temperature. Surfactant was removed from the
as-synthesized sample by extracting twice with HCl/water solution for 4 h at room temperature. When the as-synthesized sample was extracted it showed a considerable amount of weight loss, (ca. 41%), which could be attributed to the loss of the surfactant molecules from the pores of the nanostructured
as-synthesized sample. In order to study the effect of time and temperature of the reaction as well as the concentration of the surfactants, a few more samples were synthesized. The detail has been discussed later in this article.

For characterization, X-ray diffraction patterns of the as-synthesized and the acid-extracted samples were recorded on a Bruker-AXS D8 Advance diffractometer using Cu-K(=1.5406Å) radiation. TEM images were recorded in a JEOL JEM 2010 transmission electron microscope and SEM images in a JEOL JEM 6700F field emission scanning electron microscope. Thermogravimetry (TG) and differential thermal analysis (DTA) of the samples were carried out on a TA instrument Q600 DSC/TGA thermal analyzer. N2 adsorption/desorption isotherms of the sample was obtained using a Quantachrome Autosorb 1-C instrument, at 77 K. Prior to the measurement, the sample was degassed at 363 K for 8 h under high vacuum. UV-visible diffuse reflectance spectra were obtained on a Shimadzu UV 2401PC spectrophotometer with an integrating sphere attachment. BaSO4 pellet was used as background standard. The excitation and emission spectra were recorded on a Fluoromax-P Horiba Jobin-Yvon luminescence spectrometer, using a solid sample holder at room temperature. The powder samples were pressed to form a smooth, opaque flat disk for the optical study. The band pass for the excitation and emission monochromators was set at 2.5 nm.

Results and Discussion

The low angle XRD patterns of the as-synthesized (a) and extracted (b) samples are shown in Fig. 1. A single low angle peak was observed for both the samples with no detectable peak at high angle, which indicates that the samples are nanostructured containing disordered mesophases, with no short or long range ordering. The extracted sample showed relatively weaker intensity and broader peak width (Fig. 1b) over the as-synthesized sample (Fig. 1a). This result suggests that the nanostructure has been restored after the re moval of the surfactant. However,
the arrangement of the pores became more disordered after the removal of SDA. In order to study the effect of surfactant concentration, time and temperature dependence of the reaction, a series of samples were synthesized which varied in their surfactant concentration, time and temperature of reaction. The results have been given in Fig. 1. The sample synthesized with the same composition and at same temperature but hydrothermally treated only for 1 day (c) showed much lower intensity of the peak. No improvement in XRD was obtained for the samples synthesized at higher and lower temperatures, viz., 393 K (d) and 348 K (e) for 3 days. Two other batches were also prepared by varying the concentration of surfactant in the reaction mixtures; one with twice the amount of CTAB (f) and other containing half the amount of CTAB with respect to sample (a). Here also no marked improvement was noticed as far as sample (b) was concerned. On the other hand, for sample (f) there was no XRD peak at all in the mesoporous region. Thus, the conditions chosen for the reaction are by far the most optimized conditions. To study the nature of the surfactant molecules, it is important to have an insight into the reaction mechanism. The –OH groups of resorcinol moiety in basic aqueous condition remain as Oˉ and this can interact with the cation derived from CTAB, i.e., CTA+. Thus, the ionic interaction directs the growth of the mesostructure. If instead an anionic template had been used there would be no ionic interactions between two negatively charged species. As far as the use of non-ionic templates is concerned, examples where non-ionic template Pluronic F127 has been used as a template for the synthesis of mesostucture are known21. Thus, we see that this kind of framework can be derived very easily from cationic and non-ionic templates.

The TEM image of the as-synthesized sample is shown in Fig. 2. The image confirms the formation of low electron density spherical spots of 2.0-2.5 nm diameter, corresponding to the small to medium size mesopores and their disordered arrangements. Thus, from the XRD pattern and TEM image analysis it may be concluded that these nanostructured resorcinol-formaldehyde composites have disordered wormhole-like mesostructure. Supramolecular-templated mesoporous silicas, e.g., MSU-123 or KIT-124, also exhibit similar common pore center-to-center correlation length, which is characteristic of a wormhole-like structure. The SEM image of the as-synthesized sample exhibits granular/spherical morphology
(Fig. 3). In these samples very tiny spheres of dimension 30-40 nm were found, which assembled together to form large spherical aggregates.

Fig. 1 –Low angle XRD pattern of resorcinol-formaldehyde samples. [As-synthesized samples with (a) rescorcinol:CTAB=2:1 at 363 K for 3 days; (b) extracted sample of (a);
(c) rescorcinol:CTAB=2:1 at 363 K for 1 day;
(d) rescorcinol:CTAB=2:1 at 393 K for 3 days;
(e) rescorcinol:CTAB=2:1 at 348 K for 3 days;
(f) rescorcinol:CTAB=1:1 at 363 K for 3 days;
(g) rescorcinol:CTAB=4:1 at 363 K for 3 days].

Fig. 2 –TEM image of as-synthesized sample.

Fig. 3 –SEM image of as-synthesized sample.

The quantitative determination of the content of resorcinol-formaldehyde in the surfactant-free sample was estimated by using TG and DTA in the presence of N2 flow. The TG and DTA plots of the resorcinol-formaldehyde sample show the first weight loss up to 373 K due to desorption of physisorbed water (about 3.3 wt %; Fig. 4). This is followed by a sharp decrease in the weight at temperatures between 473 and 573 K, which may be attributed to the loss of resorcinol-formaldehyde fragments present in the material. A considerable endothermic peak in the DTA plot centered at 518 K suggests that most of the resorcinol-formaldehyde fragment was decomposed at this point. The total weight loss for the resorcinol-formaldehyde fragment was ca. 29.9 wt % in the temperature range 473-573 K. However, from 573-873 K a gradual weight loss of ca. 25.2 wt % occurred with an exothermic peak at ca. 723 K, which could be due to the complete conversion of the material into carbon.

Fig. 4 – TG and DTA plot for the extracted sample.

We have estimated the surface area from the N2 adsorption-desorption isotherms of the extracted sample (not shown here). No sharp increase for N2 uptake corresponding to the capillary condensation was observed in the mesopore region. The BET surface area for this sample was 4.9 m2g-1. Pore size distribution of this sample estimated by employing the BJH model suggests a broad distribution with maxima at 2.3 nm. Since the framework of this porous resin composite is completely organic containing aromatic fragments, bond twisting or bond distortion may have occurred during degassing leading to lower surface area in this material. Moreover, since the resin surface is highly hydrophilic, removal of water molecules during degassing may also result in collapse of the mesostructure. Porous carbon synthesized from phenol and formaldehyde also exhibits similar low surface area25 (ca. 5 m2g-1). For purely organic mesoporous polyaniline22a samples a similar low surface area was observed for most of the batches and the highest value of surface area exhibited was
ca. 45 m2g-1. It is relevant to mention here that this is a preliminary report on purely organic mesoporous material and further study to improve adsorption properties is in progress.

In the UV-visible spectra (Fig. 5) strong absorption band at ca. 278 nm with a weaker one at 225 nm and two shoulders at 325 and 490 nm for resorcinol were observed (Fig. 5c), attributed to the various chromophoric functionalities in the molecule, whereas, the as-synthesized (Fig. 5a) and extracted samples (Fig. 5b), showed no absorption near 278 nm; rather a broad band was observed at higher wavelengths centered around 325 and 375 nm respectively. The hump at 490 nm that can be observed for resorcinol is somewhat shifted to higher wavelength, at 527 and 525 nm respectively for the
as-synthesized and extracted samples. Absence of any absorption band from 220-290 nm thus suggests that all the resorcinol moieties have been fully poly-condensed with formaldehyde in the material.

Fig. 5 – UV-visible diffuse reflectance spectra of
(a) as-synthesized sample, (b) extracted sample, and,
(c) resorcinol.

The photoluminescence spectra (both excitation and emission) of the as-synthesized sample have been given in Fig. 6. The emission spectrum was obtained by excitation of the polymer at the maximum absorption wavelength (λex= 349 nm). The maxima in the emission spectra were observed at 468 nm (weak intensity) and 565 nm (strong intensity), corresponding to deep blue and greenish light,

Fig. 6 – Photoluminescence spectra of as-synthesized sample.
[(a), excitation, and, (b) emission].

Reaction scheme and proposed model (partial) for the formation of resorcinol-formaldehyde matrix in presence of template.

Scheme 1

respectively. The existence of some small peaks around 470 nm peak region may be attributed to the nano-scale particle size distribution26 of the sample. Thus, upon irradiation with light of 349 nm, the material excites an electron from the HOMO to LUMO to generate the singlet excited state and subsequently the excited polymer resin relaxes to the ground state with the emission of blue and greenish light. The formation of the stable resin framework may tighten the entire skeleton, resulting in much weaker vibrations and more relaxation27, which may be responsible for its photoluminescence at room temperature.

Based on the above, we propose a reaction scheme (Scheme 1) for the formation of mesoporous resorcinol-formaldehyde framework. This scheme is constructed on the basis of sol-gel framework structures formed when resorcinol condenses with formaldehyde in the presence of a base. Here the condensation can take place randomly in all direction of the activated aromatic ring, giving rise to a stable cross-linked resin framework structure. The polar head groups of the supramolecular assembly of the cationic surfactant can interact with phenoxide anions under alkaline pH conditions leading to this novel nanostructured material.

Conclusions

A new nanostructured resorcinol-formaldehyde composite has been synthesized by using supramolecular assembly of a cationic surfactant CTAB through in situ aqueous polycondensation of resorcinol with formaldehyde under alkaline condition. Powder XRD and TEM studies suggests no long range ordering in these samples and wormhole-like disordered mesopores of dimension ca. 2.3 nm extending in all direction. UV-visible absorption data suggests the formation of resorcinol-formaldehyde polymeric network in these samples. This composite material exhibits photoluminescence at room temperature. We expect this novel, all organic porous resin to find potential applications in photoresponsive materials and in fabrication of organic optical devices etc.

Acknowledgement

MN and KS thank CSIR for Senior Research Fellowships. This work was partly funded by the Ramanna Fellowship and by the NanoScience and Technology Initiative project grants of Department of Science and Technology, New Delhi.

References

1Firouzi A, Atef F, Oertli A G, Stucky G D & Chmelka B F, J Am Chem Soc,119 (1997) 3596.

2Kresge C T, Leonowicz Roth W J, Vartuli J C Beck J S, Nature, 359 (1992) 710.

3(a) Szostak R, Molecular Sieves: Principles of Synthesis and Identification (Van Nostrand Reinhold, New York, 1989); (b) Corma A, Chem Rev,97 (1997) 2373.

4Goa Y, Wu P & Tatsumi T, J Phys Chem B,108 (2004) 8401.

5Bhaumik A & Tatsumi T, J Catal,189 (2000) 31.

6Bhaumik A, Samanta S & Mal N K, Microporous Mesoporous Mater,68 (2004) 29.

7Katou T, Lu D, Kondoa J N & Domen K, J Mater Chem12 (2002) 1480.

8Kimura T, Microporous Mesoporous Mater,77 (2005) 97.

9Frindell K L, Tang J, Harreld J H & Stucky G D, Chem Mater,16 (2004) 3524.

10(a) Métivier R, Leray I, Lebeau B & Valeur B,J Mater Chem,15 (2005) 2965; (b) Chandra D, Yokoi T, Tatsumi T & Bhaumik A, Chem Mater,19 (2007) 5347.

11Gierszal K P, Yoon S B, Yu J –S & Jaroniec M,J Mater Chem,16 (2006) 2819.

12(a) Mao H & Hillmyer M A, Soft Matter,2(2006) 57;
(b)Qiao W M, Song Y, Hong S H, Lim S Y, Yoon S H, Korai Y & Mochida I, Langmuir,22 (2006) 3791.

13Zhu Y, Hu H., Li W & Zhao H, J Non-Cryst Solids,352 (2006) 3358.

14Li W–C, Lu A–H & Guo S–C, J Coll Interface Sci,254 (2002) 153.

15Li W–C, Lu A–H & Guo S–C, Carbon, 39 (2001) 1989.
(b) Roy D, Basu P K, Raghunathan P & Eswaran S V, Bull Mater Sci,27 (2004) 303.

16Li W–C, Reichenauer G & Fricke J, Carbon,40 (2002) 2955.

17Liang C, Sha G & Guo S, J Non-Cryst Solids,271 (2000) 167.

18Schaefer D W, Pekala R & Beaucage G, J Non-Cryst Solids,186 (1995) 159.

19Anderson M L, Stroud R M & Rolison D R, Nano Lett,2 (2002) 235.

20Pierre A C & Pajonk G M, Chem Rev,102 (2002) 4243.

21Liang C & Dai S, J Am Chem Soc, 128(2006) 5316.

22(a) Nandi M, Gangopadhyay R & Bhaumik A, Microporous Mesoporous Mater, 109 (2008) 239; (b) Chandra D, Jena B, Raj C R & Bhaumik A, Chem Mater,19 (2007) 6290.

23Bagshaw S A, Prouzet E & Pinnavaia T J, Science,269 (1995) 1242.

24Ryoo R, Kim J M, Ko C H & Shin C H, J Phys Chem,100 (1996) 17718.

25Liang C & Dai S, J Am Chem Soc, 128 (2006) 5316.

26Ritchie J, Crayston J A, Markham J P J & Samuel I D W,
J Mater Chem,16 (2006) 1651.

27Huang Q R, Kim H –C, Huang E, Mecerreyes D, Hedrick J L, Volksen W, Frank C W & Miller R D,Macromolecules,36(2003) 7661.