Polymer-stabilized palladium nanoparticles for catalytic membranes: ad hoc polymer fabrication
© Domènech et al; licensee Springer. 2011
Received: 5 November 2010
Accepted: 1 June 2011
Published: 1 June 2011
Skip to main content
© Domènech et al; licensee Springer. 2011
Received: 5 November 2010
Accepted: 1 June 2011
Published: 1 June 2011
Metal nanoparticles are known as highly effective catalysts although their immobilization on solid supports is frequently required to prevent aggregation and to facilitate the catalyst application, recovery, and reuse. This paper reports the intermatrix synthesis of Pd0 nanoparticles in sulfonated polyethersulfone with Cardo group membranes and their use as nanocomposite catalytic membrane reactors. The synthesized polymer and the corresponding nanocomposite were characterized by spectroscopic and microscopic techniques. The catalytic efficiency of catalytic membranes was evaluated by following the reduction of p-nitrophenol in the presence of NaBH4.
The unusual physical, chemical, and catalytic properties of metal colloids (better known in nowadays as metal nanoparticles, MNPs) have attracted increasing interest of scientists and technologists during the last decade . A characteristic high percentage of surface atoms and the associated quantum effects make MNPs efficient and selective catalysts for several types of catalytic reactions [2, 3]. However, for most practical catalytic applications, MNPs must be immobilized on solid supports to prevent their aggregation and to facilitate the catalyst recovery . In this sense, encapsulation of MNPs in polymers seems advantageous because, in addition to stabilizing and protecting effects towards MNPs, polymers offer unique possibilities for enhancing the access of reactants to the catalytic sites . The in situ synthesis of MNPs into a polymeric membrane matrix permits to combine catalysis and membrane processes, which may result in process intensification [6, 7]: destruction and separation of pollutants in a single step [7, 8]. These polymer stabilized MNPs (PSMNPs) can be synthesized by various methods [1, 9, 10], including the intermatrix synthesis (IMS) technique [11, 12]. Some common polymers such as cellulose acetate, hydrophilized polysulfone, polyacrylic acid-modified polyethersulfone, polyvinylidene fluoride, and many others, which are widely used as membrane materials, have been successfully employed for the synthesis of PSMNPs . These polymers can be used as a support of metal catalysts [14, 15] after a simple casting procedure. However, it is imperative to adequate the polymeric matrix for both the IMS of PSMNPs and the final application of the nanocomposites. Therefore, the development of catalytic membranes requires the use of polymers with good chemical, thermal, and mechanical stability, and these polymers also have to be easily fabricated in a suitable form for their further practical application .
The main goal of this work is the development of palladium nanoparticles (Pd0-NPs) by IMS technique inside SPES-C membranes for their use as catalytic membrane reactors. The characterization of the catalytic effect was performed by following the reduction of p-nitrophenol (4-np) in presence of NaBH4 .
Commercial PES-C was kindly supplied by Dr. Trong Nguyen from Université de Rouen, France. H2SO4, HCl, N-methyl-2-pyrrolidone (NMP), and N,N-dimethyl-formamide (DMF) (all from Panreac, S.A., Castellar del Vallès, Spain) organics and metal salts (4-np, NaCl, NaOH, NaBH4, CuSO4, Na2SO4, and [Pd(NH3)4]Cl2 all from Aldrich, Munich, Germany) were used as received.
Sulfonation of PES-C was carried out following a procedure similar to those published elsewhere [16, 17]. The polymer powder was dried at 70°C for 48 h prior to use. Then, it was dissolved (11% w/v) in concentrated sulfuric acid and mixed at constant temperature (60°C). After 5 h, the reaction medium was precipitated in a cold-water bath under strong stirring to obtain the SPES-C in its acid form. Afterward, it was neutralized with NaOH 1 M solution, filtered, washed with deionized water, and dried at 70°C (polymer in Na form).
Membranes were prepared by wet phase inversion method  by using a polymer solution in NMP (25 wt.%). After casting, membranes were stored in deionized water.
Attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectra were recorded with a Perkin Elmer Spectrum GX spectrometer (Norwalk, CT, USA).
Indirect titration: samples were equilibrated with HCl 2 M for 24 h to convert the functional groups of the polymer into protonated form. Then, samples were equilibrated with an excess of NaCl 1 M solution for 24 h, and the resulting solutions were titrated with NaOH 4.3 mM using phenolphthalein as indicator.
Cu2+ displacement and determination: a sample of membrane in the H form (see above) was equilibrated with 20 ml of CuSO4 0.1 M solution for 20 h, washed with water several times and finally equilibrated with 20 ml of Na2SO4 0.5 M. The exchanged Cu2+ ions were analyzed by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using an Iris Intrepid II XSP spectrometer (Thermo Electron Corporation, Milford, MA, USA).
In order to determine the exact metal content in the prepared nanocomposites, the membrane samples of known weight were treated with aqua regia to completely dissolve all MNPs. The resulting solutions were diluted and analyzed by ICP-AES. The average uncertainty was <2% in all cases.
With the aim of characterizing the size and structure of Pd0-NPs, transmission electron microscope (TEM) images were obtained by using a JEOL JEM-2011 Microscope (Jeol Ltd., Tokyo, Japan). For this purpose, a drop of samples diluted in DMF was deposited onto a Cu TEM grid, followed by solvent evaporation at room temperature.
The catalytic performance of modified SPES-C membranes was evaluated by using the reduction of p-nitrophenol (4-np) by NaBH4 to p-aminophenol as a model reaction , which was considered to follow the pseudo-first-order kinetics. In a typical run, a given amount of NaBH4 was added to 5 ml of 4-np solution (5 mM) to achieve a NaBH4 concentration of 55 mM. After mixing, a piece of nanocomposite (1 cm2) was added to the vessel. The process was monitored at 390 nm by a Pharmacia LKB Novaspec II spectrometer (Biochrom Ltd., Cambridge, UK).
The qualitative confirmation of polymer sulfonation was also supported by the quantitative determination of the IEC and the corresponding SD values. The IEC value obtained by titration of three samples (uncertainty 95%, t = 2.92) was 0.48 ± 0.12 meq/g whereas that obtained by Cu2+ exchange was found to be equal to 0.31 ± 0.07 meq/g. The higher value obtained by the first method can be due to incomplete displacement of Cu2+ from the polymer and to the partial retention of some metal ions in the matrix when preparing samples for ICP-AES. The obtained SD values (0.27 ± 0.09 and 0.17 ± 0.05, respectively) are far from the practical limitation of 0.9, which has been reported to cause the polymer dissolution in water . These results prove that a moderate SD can be achieved under mild reaction conditions corresponding to reasonable reaction temperature and time (60°C, 5 h) just by using concentrated H2SO4. This procedure appears to be far simpler than, for instance, the polymer modification by the metalation route .
The measured values of MWU for the sulfonated polymer and the corresponding nanocomposite were found to be 2.73 ± 0.13 and 3.21 ± 0.19 g H2O/g dry membrane, respectively (three replicates in each case). These values prove that (1) even if hydrophilicity of the material is high, it does not result in the membrane dissolution, and (2) the IMS of Pd0-NPs does not substantially affect the membrane morphology and the accessibility of water to the sulfonic groups.
The content of Pd-MNPs in the nanocomposites was found to be 156.7 ± 0.8 μmol of Pd0 per gram of dry nanocomposite (16.68 ± 0.08 mg Pd0/g) by analyzing eight different samples by ICP-AES. Similarly to the other studied parameters, it is worth noting the high homogeneity of samples, probably due the IMS methodology.
The possibility to reuse nanocomposite catalytic membranes was also evaluated. Figure 4 (right) shows the decrease of the normalized absorbance vs. time for four consecutive cycles, which testifies to the decrease of the reaction rate and the increase of the activation time after each cycle. This can be explained either by possible poisoning of the catalyst due to the generation of by-products of the 4-np reduction, which can interfere with the catalyst surface , or (and most probably) by a stronger competition between hydrogen absorption and catalytic reactions, after the initial discharge in the first cycle. It must be pointed out that the loss of catalytic activity cannot be explained by the leaching of Pd0-NPs from the matrix to the aqueous solution as no change of Pd content in nanocomposite membranes before and after catalytic cycles has been detected. Moreover, the absorbance of reacting solutions did not keep decreasing after the catalytic membrane was removed, even when stored overnight.
It has been reported that some polymer-encapsulated catalytic MNPs do aggregate after catalytic reactions . However, due to the presence of the polymeric matrix, this mechanism might not be applicable for Pd0-SPES-C nanocomposites.
As it can be seen in Figure 4 (right), the decrease in catalytic activity diminishes after each catalytic cycle if one takes into account the slope of the absorbance vs. time plot. The difference between the first and second cycles is much higher than that corresponding to third and fourth cycles. This fact may suggest that after few cycles, the catalytic efficiency could be preserved, although at a lower value when compared with the first run. However, this point requires a stricter experimental confirmation if membranes operate under a pressure gradient.
Polyethersulfone with Cardo group was easily sulfonated and used as a suitable matrix for IMS of palladium nanoparticles. Pd0-NPs were well separated from each other, which indicates the high stabilizing efficiency of the polymer. Synthesis of Pd-MNPs by IMS resulted not only in the formation of catalyst NPs but also in their additional activation due to dissolution of H2 inside Pd-MNPs. The membrane morphology is not affected by IMS of Pd-MNPs inside the polymeric matrix; however, further investigations are required to determine if the membrane transport processes are affected by the presence of MNPs. The obtained nanocomposite material exhibits a high catalytic activity, although an induction period is needed before reaction starts and the catalytic efficiency is not maintained with time. Further experiments are required in order to explain the deactivation effect observed when running consecutive catalytic cycles.
We thank ACC1O for VALTEC09-02-0058 grant within the "Programa Operatiu de Catalunya" (FEDER). The Ministry of Science and Innovation of Spain is acknowledged for the financial support to J. Macanás, and D.N. Muraviev. Prof. Trong Nguyen is acknowledged for supplying the polythersulfone Cardo. Special thanks are given to Servei de Microscopia from Universitat Autònoma de Barcelona.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.