Printer Friendly

A Novel Poly(vinylidene fluoride) Composite Membrane for Catalysis and Separation.


Noble metal nanoparticles (NMNPs) have excellent catalytic reactivity due to their high surface area-to-volume ratios and excellent surface properties [1, 2]. However, their surfaces are often covered due to the agglomerate caused by the high surface energy [3]. This defect results in the significant decline of thencatalytic reactivity. In order to prevent the aggregation and/or decrease the coverage for the surface of the metal nanocatalysts, some researchers immobilize the metal NPs into silica nanosphere [4], grapheme and dendrimer [5, 6], block copolymer [7], or into the materials with a porous shell such as silica [8, 9], carbon [10], and metal oxides [11], However, the drawbacks have not yet been eradicated since the supports often aggregate in applications, leading to the deteriorative performance and the decreased continuous reuse [1, 3-6, 8-11].

In order to obtain a better benefit of the nanocatalyst strategy, the porous polymeric membranes have been strongly suggested as the supports for loading noble nanocatalysts [12-14], First, polymer membranes can provide enough surfaces for the immobilization of metal NPs [12]. Second, the unique operating model of membranes can promote the reaction by rapidly carrying the reactants to and the products off the active surface of catalysts [12, 13]. In addition, the operations for use and recovery are very convenient compared with other supports [12-14]. Because of the above advantages, several researchers have prepared different composite membranes with noble NPs for the catalytic applications [12-18]. Even though the catalytic reactivity and the convenient operation have obtained great progress, the formed products still need additional steps to be separated from the reactants. This not only increases the operational costs but also possibly causes the catalysts poisoning and the unexpected side reactions due to the long-time coexistence of the reactants and products in reaction process. Therefore, it is particularly urgent to provide a new composite membrane which can immediately expel the formed products from the reaction system.

Poly(vinylidene fluoride) (PVDF) membrane is the preferred support for loading noble NPs due to its excellent stability, chemical resistance, good mechanical strength, and membrane forming property [18, 19]. In general, preparation of composite membrane includes the introducing functional groups to PVDF membrane for adsorbing metal ions, and reducing the absorbed metal ions into metal NPs [18], Although several composite membranes have been reported [13, 17, 18], these composite membranes cannot separate the product from reactants because the formed metal NPs distributed both on membrane surface and in membrane pores results in the product always companied by the reactant in the whole process.

In this study, we designed and prepared a novel PVDF composite membrane to efficiently convert reactants to products and then timely separate the formed products from the reaction system. The composite membrane was prepared by the first formation of PVDF/poly(methacrylic acid) (PMAA) blend membrane with PMAA microspheres in membrane pores and then the followed noble NPs formation on the surfaces of the PMAA microspheres. The catalytic reactivity and separation property of the composite membrane were examined by the catalytic reduction of p-nitrophenol to p-aminophenol. Results indicated that the prepared composite membrane showed very high catalytic reactivity under a cross-flow operation model, and could timely separate the highly purified product from the reactant in reaction process. Thus, the prepared composite membrane shows great prospect in catalytic applications, with the merits of weakening or even avoiding the catalysts' poisoning and side reactions.



Methacrylic acid (MAA), acetonitrile, N,N-dimethyl formamide (DMF), and sodium borohydride (Na[BH.sub.4]) were purchased from Tianjin Kemiou Chemical Reagent Co, Ltd. Azodiisobutyronitrile (AIBN), chlorauride (Au[Cl.sub.3] x HCl x 4[H.sub.2]O), and silver nitrate (AgN[O.sub.3]) were obtained from Tianjin Yingda Rare Chemical Reagents Factory. Ethyleneglycol dimethacrylate (EGDMA) was obtained from J&K Scientific. Poly(vinylidene fluoride) (PVDF) powders ([M.sub.w] = 3.52 X [10.sup.5] g/mol, [M.sub.w]/[M.sub.n] = 2.3, Solvay Company, Belgium, Solef1010) were used as received. Ethyl alcohol was obtained from Tianjin Bodi Chemical Engineering Limited Corporation. The p-nitrophenol was purchased from Tianjin Guangfu Fine Chemical Engineering Institute. All the chemicals were analytical grades.

Preparation of PVDF/PMAA Blend Membrane

Preparation of PMAA Microsphere. PMAA microspheres were prepared by the method of distillation precipitation polymerization [20]. A mixture of MAA (2.0 g), EGDMA (2.1 g), AIBN (0.08 g), and acetonitrile (160.0 mL) was added into a 250 mL round-bottom flask. After reacted for 1.5 h at 85[degrees]C, the suspension was centrifuged. PMAA microspheres were obtained by washing with ethyl alcohol and deionized water for several times and then drying the sediment.

Preparation of PVDF/PMAA Blend Membrane. In previous studies, we found that the prepared membrane could show good stability in filtration process when a (polymeric microspheres/ [PVDF + polymeric microspheres]) weight ration is less than 0.4 [21,22], In this study, the similar conditions are used to prepare PVDF/PMAA blend membrane with a (PMAA/ [PVDF + PMAA]) ratio of 34% for the subsequent loading of metallic NPs. A mixture of PMAA microspheres (2.45 g), PVDF powders (4.75 g), and DMF (30.0 mL) was continuously stirred for 5 h at 60[degrees]C. After a further 0.5 h vacuum degassing, the casting solution was formed. The casting solution was poured to a dry and smooth glass plate and then scraped into a liquid film by a steel knife. Subsequently, PVDF/PMAA blend membrane was prepared by immersing the glass plate into deionized water. Pure PVDF membrane was also prepared by the same method.

Preparation of Composite Membrane. PVDF/PMAA blend membrane was immersed in 250.0 mL of sodium hydroxide aqueous solution (0.01 mol/L) for 24 h to transform the carboxylic acid (--COOH) of PMAA into carboxylate (--COONa), and then washed by water to remove the excess sodium hydroxide. Subsequently, the membrane was immersed in AgN[O.sub.3] aqueous solution (0.004 mol/L, 250 mL) for 24 h in a dark environment. After water rinse, the blend membrane was placed in a supersaturated aqueous solution of NaBH4 to reduce [Ag.sup.+] to Ag NPs, forming the PVDF/PMAA-Ag membrane. The PVDF/PMAA-Pd membrane and PVDF/PMAA-Au membrane had been prepared by the same method by just replacing the AgN[O.sub.3] solution with Pd[Cl.sub.2] solution (0.004 mol/L, 250 mL) and Au[Cl.sub.3] solution (0.004 mol/L, 250 mL). All the composite membranes and PVDF/PMAA blend membrane have a same thickness of about 0.17 mm.


A Bruker TENSOR37 instrument had been utilized to test the Fourier transform infrared spectra (FTIR) of the membranes. Thermal behavior of membrane was studied by using DSC (200 F3 Netzsch Co., Germany). A thermal scan rate of 10[degrees]C/min was applied to increase temperature from -60[degrees]C to 220[degrees]C. The thermal stability of composite membrane was evaluated by thermogravimetric (TG) analysis using NETZSCH STA 409 PC/PG. Under nitrogen atmosphere, a membrane sample was heated from room temperature to 800[degrees]C at a rate of 10[degrees]C/min. The mechanical stability of composite membrane was evaluated using an YG065NL Electronic testing machine under room temperature. A piece of dried membrane (length: 10 mm, thickness: 0.17 mm and width: 5 mm) was drew at a loading velocity of 10 mm/min. The morphologies of the membranes were studied by scanning electron microscope (SEM, Hitachi S-4800, Japan) equipped with an energy dispersive X-ray spectrograph (EDX). X-ray photoelectron spectroscopy (XPS) was performed to investigate the surface chemical compositions of membranes using Al-Ka X-ray source (hv = 1486.6 eV). The binding energy scale was calibrated from the carbon hydrocarbon using the C1s peak at 284.6 eV.

Pore sizes were determined by the reported method [19]. Porosities ([epsilon]) of the membranes had been calculated by the following equation according to dry-wet weight method [19]:

[epsilon] = [[W.sub.w] - [W.sub.d]]/[[[rho].sub.w] x V] (1)

where [W.sub.w] (g) was the weight of the wet membrane at 25[degrees]C, [W.sub.d] (g) was the weight of the dry membrane, [[rho].sub.w] (g/mL) was the density of deionized water, V ([cm.sup.3]) was the apparent volume of the membrane.

The metal NPs contents of the membranes were determined by firstly dissolving the metal NPs with aqua regia (HN[O.sub.3]: HCl = 1:3) and then measuring the concentration of the metal ions with the inductively coupled plasma atomic emission spectroscopy (ICP-AES, 715-ES, VARIAN, America) [20].

Catalytic Ability of Composite Membrane

The reduction of p-nitrophenol by NaB[H.sub.4] was widely used to evaluate the catalytic performance of a catalyst because the model reaction had mild operation conditions and a single product of p-aminophenol [23]. The same reaction was utilized to characterize the catalytic ability of the membrane. A p-nitrophenol aqueous (feed solution, 500 mL, 20 mg/L, pH = 10) was firstly prepared. After added excess NaB[H.sub.4] (13.43 g/L), the feed solution in feed tank was continuously circulated through a device equipped with a membrane (membrane area: 17.9 [cm.sup.2]) under a cross flow model (Fig. 1). The catalytic property was studied by recording the absorbance change of feed solution al 400 nm with an ultraviolet spectrograph. Unless stated, the catalysis was conducted at 20[degrees]C and under 0.1 MPa. In addition the static catalysis was also studied by directly immersing the composite membrane into the feed solution which was strongly stirred and then recording the absorbance at 400 nm.


Preparation and Characterization of Composite Membrane

Figure 2 shows the FT-IR spectra of the membranes. The absorptions of all the membranes are almost uniform at wave-numbers less than 1,500 [cm.sup.-1], ascribed to PVDF [22]. Different from PVDF membrane, an additional C=O stretching vibration absorption at 1,730 [cm.sup.-1] appears in the PVDF/PMAA blend membrane and composite membranes, confirming that PMAA is incorporated [20]. In addition, all the membranes show strong characteristic peaks at 840 and 1,280 [cm.sup.-1], indicating the PVDF matrixes have [beta]-crystalline phase [24, 25]. This crystalline phase is considered to the main form crystal of PVDF matrixes when membrane is formed from solutions in DMF solvent at a temperature lower than 30[degrees]C [26, 27]. Supporting Information Fig. S1 shows the DSC curves of different membranes. As shown, all the membranes display a same single melting peak at 173[degrees]C. The same peak position and shape of blend membrane and composite membrane compared with PVDF membrane indicates the PVDF crystal phase is hardly affected by PMAA microsphere, i.e., PVDF phase is separated from PMAA microsphere in blend membrane and composite membrane [27, 28]. The crystallinity calculated according to the melting peak of Supporting Information Fig. S1 is shown in Supporting Information Table S1. As seen, the crystallinities of PVDF membrane, blend membrane and composite membrane are respectively 62.04%, 53.03%, and 43.59% (Supporting Information Table S1). The crystalline decrease in crystalline from pure PVDF membrane to blend membrane and composite membrane is due to the disturbance signals from the glass transition of PMAA, the release of bound water and the anhydride reaction [29].

Figure 3 shows the SEM images of the membranes. The pure PVDF membrane displays a compact surface and a porous cross section with a top skin layer, a middle finger-like layer and a spongy bottom layer (Fig. 3a and c). This is the general membrane structure, formed by the casting solution of pure PVDF polymer undergoing nonsolvent induced phase separation [19, 30], Different from the pure PVDF membrane, the PVDF/ PMAA blend membrane has a porous surface (Fig. 3b).

These surface pores are formed from the phase separation of the hydrophilic PMAA microspheres with the freezable PVDF when DMF of the casting solution is rapidly exchanged with deionized water in the membrane forming process [19, 30], Since the poor interaction between the PMAA microspheres and PVDF, no PMAA microspheres are left on membrane surface after the preparation and water rinse processes. The cross-section of the PVDF/PMAA blend membrane shows a similar porous structure with that of the pure PVDF membrane, except of PMAA microspheres found in membrane pores (Fig. 3d). Obviously, the phase separation of PMAA microspheres and the PVDF from the inners of the casting solution leads to PMAA microspheres finally remaining in the membrane pores. Owing to PMAA microspheres other than PVDF absorbing metal ions, the PMAA microspheres thus can be further used to introduce Au, Ag, and Pd NPs to the membrane pores by in site reduction of the adsorbed noble metal ions [31]. In order to make more metallic NPs introduced into the membrane pore, the high PMAA/(PVDF + PVDF) ratio is used to prepare PVDF/PMAA membrane (see Experimental).

Figure 3e-g confirmed the successful introduction of metal NPs into the membrane pores by adsorbing and then reducing noble metal ions with reductants. From the partial enlargement of cross-sections (Fig. 3e-g), massive NPs are found to be coated on the surface of PMAA microspheres in the membrane pore. The Ag, Au and Pd elements are confirmed by the EDX results (Supporting Information Fig. S2). The zero valents of the metallic NPs are respectively confirmed by Ag 3d signal appearing at about 370 eV (Ag [3d.sub.5/2] at 368.6 eV and Ag [3d.sub.3/2] at 374.6 eV, Fig. 4a and b) for the PVDF/PMAA-Ag membranes [32], Au 4f peaks (Au [4f.sub.7/2] at 84.2 eV and Au [4f.sub.5/2] at 87.9 eV, Fig. 4a and c) for the PVDF/PMAA-Au membranes [33], and Pd 3d (Pd [3d.sub.5/2] at 335.1 eV and Pd [3d.sub.3/2] at 340.8 eV, Fig. 4a and d) for the PVDF/PMAA-Pd membranes [34]. The diameters of Ag, Au, and Pd NPs are, respectively, calculated to be about 28.8, 43.4, and 26.3 nm from the SEM images. The metal NPs contents measured by I CP are respectively 0.94 (Ag NPs), 1.45 (Au NPs), and 0.54 (Pd NPs) mg per gram of membrane (Table l). The composite membranes have similar pores and porosities, larger than that of PVDF membrane owing to the phase separation of PVDF and PMAA microspheres during membrane forming process. The improved pore and porosity are mainly ascribed to the high PMAA content. In our previous studies, we proved that the increase of polymer spheres could lead to the increases of porosity and pore size [21, 22]. Thus, besides the introduction of metallic NPs into membrane pores, the other purpose of the high PMAA micrspheres' content in the present study is to obtain a membrane matrix with an improved pore and porosity, which are beneficial to the mass transfer and are thus good for the catalysis applications of the composite membranes.

Catalytic Abilities of the Composite Membranes

The catalytic abilities of the composite membranes were researched by the reduction reaction of p-nitrophenol to p-aminophenol under the cross flow model (Fig. 1). Under this model, feed solution flowed through the membrane surface (tangential flow) and penetrated through the membrane pores (penetrative flow) (Fig. 5a). Figure 5b shows the ultraviolet absorption spectra of the feed solution with reaction time. Initially (t = 0), the feed solution shows a powerful absorption peak at 400 nm, ascribed to p-nitrophenol [4, 20]. This absorption becomes lower and lower as the reaction progresses, meaning that p-nitrophenol in the feed solution becomes less and less. Meanwhile, the feed solution shows a new absorption peak at 310 nm, ascribed to p-aminophenol [6, 20], which becomes stronger with time. Finally, the absorption peak at 400 nm has almost disappeared totally, and simultaneously the growth of the absorption peak at 310 nm has almost stopped, indicating that catalytic reaction finishes. In the process of catalytic reaction, the color of feed solution becomes more and more shallow, finally becomes colorless.

In order to further explore the catalytic activity laws of three composite membranes, we researched the kinetics of the catalytic reaction. The used NaBH4 in the present reaction is overdosed so that the rate of catalytic reaction is not affected by the concentration of NaB[H.sub.4]. Thus, the reaction can be seen as a first-order reaction [4, 6, 20], Since the p-nitrophenol concentration of feed solution is proportional to the intensity of absorption peak at 400 nm, the rate constant [] can be calculated by the following equation:

ln ([C.sub.t]/[C.sub.0]) [varies] ln ([A.sub.t]/[A.sub.0]) = -[]t (2)

where [C.sub.t] is the concentration of p-nitrophenol in time t, [C.sub.0] is the initial concentration of p-nitrophenol, [A.sub.t] is the absorbance of ultraviolet absorption peak at 400 nm in time t, [A.sub.0] is the initial absorbance of ultraviolet absorption peak at 400 nm, and [] ([min.sup.-1]) is the apparent first-order rate constant. As shown in Fig. 5c, ln([A.sub.t]/[A.sub.0]) is linearly related to reaction time, indicating the reaction catalyzed by PVDF/PMAA-Ag composite membrane is first-order kinetics.

Effect of Initial Concentration of p-Nitrophenol on Reduction Reaction. Effects of different initial concentrations of p-nitrophenol on catalysis are shown in Fig. 6a. As seen, ln([A.sub.t]/[A.sub.0]) shows a linear relation to the reaction time for three composite membranes, indicating that the reaction is in accordance with first-order reaction kinetics. The apparent first-order rate constant [] is obtained from the slope of the fitting line in Fig. 6a by Eq. 2. From Fig. 6b, one can see that the values of [] decrease with the increase of the initial concentration of p-nitrophenol.

Effect of Reaction Temperature on Reduction Reaction. Effect of reaction temperature on reduction reaction is shown in Fig. 7a. We can see that ln([A.sub.t]/[A.sub.0]) shows a linear relation with the reaction time for different membranes. Furthermore, the fitting line is more and more steep with the increase of the reaction temperature, indicating that reaction becomes faster and faster. The apparent rate constant [] is calculated from the slope of the fitting line in Fig. 7a by Eq. 2. As shown in Fig. 7b, ln([]) is linearly related with the reciprocal of temperature (1 /T). Therefore, the relationship of ln([]) with 1/7 can be further treated by Arrhenius equation:

ln[] = -[E.sub.a]/RT + ln [A.sub.0] (3)

where [] ([min.sup.-1]) is apparent first-order rate constant, [E.sub.a] (kJ/ mol) is apparent activation energy, R (J [mol.sub.-1] K) is ideal gas constant, T (K) is Kelvin temperature, [A.sub.0] is pre-exponential factor ([min.sup.-1]). According to Eq. 3, we can calculate the [E.sub.a] values by the fitting line shown in Fig. 7b. As can be seen, the values are respectively 44.3 for PVDF/PMAA-Ag membrane, 84.8 for PVDF/PMAA-Au membrane, and 79.6 for PVDF/PMAA-Pd membrane. It has reported that the diffusion-controlled reactions in solution have low activation energies (8-21 kJ/mol), conversely, the surface-controlled reactions have larger activation energies (>29 kJ/mol) [20]. Thus, the present reactions catalyzed by the three composite membranes are thought to surface-controlled reactions. In this reaction, the active points of catalysts are fully utilized, therefore, the [] decrease with the increase of the initial concentration of p-nitrophenol instead (Fig. 6b). The [E.sub.a] value of PVDF/PMAA-Au membrane is similar with that of PVDF/ PMAA-Pd membrane but larger than that of PVDF/PMAA-Ag membrane, indicating the reaction catalyzed by PVDF/PMAA-Au membrane or PVDF/PMAA-Pd membrane is more easily accelerated by temperature due to raising temperature leading to the increase of the corresonding [] value.

Effect of Different Noble Metal NPs on Reduction Reaction. Figure 8 shows the effect of different noble metal NPs on the catalytic reaction. As shown, ln([A.sub.t]/[A.sub.0]) is linearly related with reaction time, indicating the reaction is first-order kinetics. The line described by ln([A.sub.t]/[A.sub.0]) versus 1/7 indicates the order of the apparent reaction rate: Ag>Au>Pd. The reaction rate is affected by the content, sizes, category of metal catalysts, and other factors [9, 14, 18, 35], In order to more accurately study the effect of different noble metal NPs on the catalytic reduction reaction, the surface area-based rate constant [k.sub.s] is acquired by the Eq. 4 [15, 18]:

ln ([A.sub.t]/[A.sub.0]) = - []t = -[k.sub.s][a.sub.s][[rho].sub.m]t (4)

where [k.sub.s] is the surface area-based rate constant (L [min.sup.-1] [m.sup.-2]); as is the specific surface area of the three metal NPs ([m.sup.2] [g.sup.-1]); and [[rho].sub.m] is the mass concentration of the metal NPs ([L.sub.-1]). For the present case, [[rho].sub.m] of Ag NPs, Au NPs, and Pd NPs are respectively 5.7650 X [10.sup.-4], 9.8275 x [10.sup.-4], and 1.3034 x [10.sup.-4] g/L, and thus the corresponding [k.sub.s] can be calculated to be 4.21, 7.55 and 16.03 L [min.sup.-1] [m.sup.-2] (Table 1). The surface area-based rate constant follows the sequence: Pd > Au > Ag. For the apparent reaction rate, the order of Ag > Au > Pd is ascribed to the contents and sizes of the metallic catalysts used. The content of Au NPs is larger than that of Pd NPs, therefore the apparent reactivity order is Au > Pd. Although the content of Ag NPs is slightly less than that of Au NPs, their much smaller sizes make the apparent reactivity of Ag NPs be actually higher. However, for catalysts with a same surface area, the rate constant sequence of Pd > Au > Ag means that the real reactivity order for the metallic catalysts of the composite membrane should be Pd > Au > Ag.

Comparing the Catalytic Reactivity of the Composite Membrane under the Cross-Flow and Static Mode. A comparison of the three composite membranes under static mode and cross flow is presented in Fig. 9a-c. As can be seen, the reactants show a much higher conversion under cross flow catalysis than static catalysis at every reaction time. In order to more intuitively compare the reactivity, we further tested the required time to achieve a conversion of 80% under cross-flow mode and static mode. Results indicate that it will take less than 16 min for PVDF/PMAA-Ag membrane, 17 min for PVDF/PMAA-Au membrane, and 23 min for PVDF/PMAA-Pd membrane to complete an 80% conversion under the cross-flow mode; however, it will take more than 4 h to complete the same conversion under the static mode. The present results indicate the composite membrane more favors the cross flow operating model when used as catalyst. The different reaction rates are due to the large difference of mass transfer under the two models. Under static catalysis, reactants should diffuse from solution to nanocatalysts of membrane and then react. In such case, diffusion of reactants is so slow that the reaction rate is small. In contrast, reactants can rapidly get to and products can quickly get away from the nanocatalysts under the cross flow, therefore, the reaction can quickly proceed.

Tangential and Penetrative Flow Catalysis of Cross-Flow Catalysis. Since cross flow catalysis proceeds by reactants tangentially flowing through the membrane surface (tangential flow catalysis) and penetrating through membrane pores (penetrative flow catalysis), it is thus necessary to further learn about the change of reactants passing through the two zones.

Figure 10a shows the flow rate of the penetrative flow catalysis and tangential flow catalysis under different operating pressures. From Fig. 10a, we can note that tangential flow has a very high flow rate of 1,200 to 1,600 mL/min, which decreases with the increase of operating pressure; On the contrary, penetrative flow has a very low rate of 0.5 to 4.5 mL/min, which increases with the increase of operating pressure. Base on this flow state, the conversion of reactants from penetrative flow can get to more than 95% while that from tangential flow can be only less than 10% (Fig. 10b). The conversion is largely influenced by the flow state. In general, a slow flow rate of reactants is beneficial to the conversion of reactants [14, 15, 17, 18]. Therefore, the slow rate of the penetrative flow leads to a long time contact between reactants and catalysts, resulting in a much higher conversion compared with that of the tangential flow. Based on the same reason, with the increase of operating pressure, the increased flow rate of the penetrative flow causes the decline of conversion while the conversion of tangential flow slightly increases conversely. Besides the flow rate, the metal catalysts also play an important role to the reactant conversion. For the present composite membrane, only the very few metal nanocatalysts exposed in mouth of the membrane pore make reactants be reacted in the tangential flow, also resulting in a low conversion. In contrast, many metal nanocatalysts in membrane pores will lead to more reactants reacted in the penetrative flow, contributing to a very high conversion. In a word, the novel membrane structure combined with the cross flow model results in the very high conversion of reactants from the penetrative flow. This allows us to directly obtain products from the penetrative flow under cross flow catalysis. In such a catalysis model, reactants are thoroughly changed to products when they penetrate through membrane pores while reactants still residues when they flow through membrane surfaces. Thus, the products can be timely separated from the reactants during reaction process. Obviously, using the present composite membrane avoids the additional costs of separating products and nanocatalysts from reactants and thus is convenient and economic. More importantly, products and reactants are isolated in two different zones (membrane surface and membrane pores), avoiding their long time contact and thus greatly preventing the possible side reactions and catalyst poisoning. These merits make the present composite membrane be superior to other catalysts [1, 8, 10, 11].

Stability of Three Composite Membranes. The reusability of the composite membrane was investigated by repeated testing under a same condition. After every experiment, the membrane was washed by deionized water for 1 h and was then used to the next experiment. The compared catalytic rate constant in every recycling experiment was shown in Fig. 11. As can be seen, the changes of [k.sub.n]/[k.sub.1] only show a very little decrease after seven recycles. This indicates that the composite membranes have fast recovery ability and high recycle stability. Supporting Information Figs. S3 and S4 show the thermal and mechanical stabilities. As seen, the degradation and stretching behavior show an almost same variation before and after repeated experiments, indicating the composite membrane also shows good structure stability. These good stabilities grant a good application prospect of the composite membrane.


A novel composite membrane with noble metal NPs in membrane pores are prepared to catalyze the reduction of p-nitrophenol. This composite membrane is prepared by firstly locating PMAA microspheres in the membrane pores of the PMAA/PVDF membrane and then anchoring noble metal NPs on the surface of the PMAA microspheres. The composite membranes exhibit excellent catalytic abilities when applied to the catalytic reduction reaction of p-nitrophenol to p-aminophenol under a cross flow model. Under this model, the membrane can rapidly accelerate the reduction reaction by promoting the mass transfer through the membrane. Since noble metal NPs are only in membrane pores, reactants are only slightly reacted on membrane surface but completely reacted in membrane pores due to the catalysis of the enriched metal NPs. Thus, products can be directly obtained only by collecting the penetrative flow fluid. Catalysis by the composite membrane avoids the additional costs of separating products and nanocatalysts, and is therefore convenient and economic. In addition, the prepared composite membrane also shows good thermal, mechanical, and catalytic stability in operation process. Thus, the prepared composite membrane is superior to other catalysts and can be widely applied in catalytic reaction. Based on the reaction and separation characteristic, the present composite membrane possibly shows a good prospect for the multicomponent system with one small molecule reacted into the other one product.


[1.] C. Wang, R. Ciganda, L. Salmon, D. Gregurec, J. Irigoyen, S. Moya, J. Ruiz, and D. Astruc, Angew. Chem. Int. Ed. Engl., 55, 3091 (2016).

[2.] H. Xiao and Y. Xia, Polym. Eng. Sci., 50, 1767 (2010).

[3.] M. Liang, R. Su, R. Huang, W. Qi, Y. Yu, L. Wang, and Z. He, ACS Appl. Mater. Interfaces, 6, 4638 (2014).

[4.] Z. Dong, X. Le, C. Dong, W. Zhang, X. Li, and J. Ma, Appl. Catal. B, 162, 372 (2015).

[5.] S.K. Movahed, M. Dabiri, and A. Bazgir, Appl. Catal. A, 488, 265 (2014).

[6.] J.H. Noh and R. Meijboom, Appl Surf. Sci., 320, 400 (2014).

[7.] H.M. Zhou, H. Cheng, and Z.H. Luo, Polym. Eng. Sci., 53, 1475 (2013).

[8.] A. Guerrero-Martinez, J. Perez-Juste, and L.M. Liz-Marzan, Adv. Mater., 22, 1182 (2010).

[9.] S. Kim, D.W. Lee, K.Y. Lee, and E.A. Cho, Catal. Lett, 144, 905 (2014).

[10.] C. Bao, L. Zhou, Y. Shao, Q. Wu, J. Ma, and H. Zhang, RSC Adv., 5, 82666 (2015).

[11.] J. Li, X. Liang, J.B. Joo, I. Lee, Y. Yin, and F. Zaera, J.Phys. Chem. C, 117, 20043 (2013).

[12.] Y. Dai, W. Zheng, X. Li, B. Chen, L. Wang, and G. He, Chem. Eng. J., 293, 252 (2016).

[13.] S. Bolisetty, M. Arcari, J. Adamcik, and R. Mezzenga, Langmuir, 31, 13867 (2015).

[14.] C. Emin, J.C. Remigy, and J.F. Lahitte, J. Memh. Sci, 455, 55 (2014).

[15.] L. Ouyang, D.M. Dotzauer, S.R. Hogg, J. Macanas, J.F. Lahitte, and M.L. Bruening, Catal. Today, 156, 100 (2010).

[16.] P. Ruiz, M. Munoz, J. Macanas, C. Turta, D. Prodius, and D.N. Muraviev, J. Chem. Soe. Dalton Trans., 39, 1751 (2010).

[17.] B. Domenech, M. Munoz, D. Muraviev, and J. Macanas, Catal. Today, 193, 158 (2012).

[18.] V. Smuleac, L. Bachas, and D. Bhattacharyya, J. Memhr. Sci, 346, 310 (2010).

[19.] X. Chen, B. Zhao, L. Zhao, S. Bi, P. Han, X. Feng, and L. Chen, RSC Adv., 4, 29933 (2014).

[20.] S. Bi, K. Li, X. Chen, W. Fu, L. Chen, H. Sheng, and Q. Yang, Polym. Compos, 35, 2251 (2014).

[21.] X. Chen, S. Bi, C. Shi, Y. He, L. Zhao, and L. Chen, Colloid Polym. Sci., 291, 2419 (2013).

[22.] Y. He, X. Chen, S. Bi, C. Shi, L. Chen, and L. Li, Polym. Adv. Technol., 24. 934 (2013).

[23.] P. Herves, M. Perez-Lorenzo, L.M. Liz-Marzan, J. Dzubiella, Y. Lu, and M. Ballauff, Chem. Soc. Rev., 41, 5577 (2012).

[24.] P.M. Martins, V. Gomez, A.C. Lopes, C.J. Tavares, G. Botelho, S. Irusta, and S. Lanceros-Mendez, J. Phys. Chem. C, 118, 27944 (2014).

[25.] P.M. Martins, R. Miranda, J. Marques, C.J. Tavares, G. Botelho, and S. Lancerosmendez, RSC Adv., 6, 12708 (2016).

[26.] Y. Chen, D. Yang, and D. Song, Chem. J. Chin. U, 14, 1326 (1993).

[27.] T. Tsuboi, H. Katayama, and T. Itoh, Polym. Eng. Sci., 53, 843 (2013).

[28.] J.F. Hester, P. Banerjee, Y.Y. Won, A. Akthakul, A. M. H. Acar, and A.M. Mayes, Macromolecules, 35, 7652 (2002).

[29.] C. Sung, A. Vidyasagar, K. Hearn, and J.L. Lutkenhaus, Langmuir, 28, 8100 (2012).

[30.] X. Chen, Y. He, C. Shi, W. Fu, S. Bi, Z. Wang, and L. Chen, J. Membr. Sci, 469, 447 (2014).

[31.] G. Cardenas, C. Munoz, M. Rodriguez, J. Morales, and H. Soto, Eur. Polym. J, 35, 1017 (1999).

[32.] Z.J. Jiang, C.Y. Liu, and L.W. Sun, J. Phys. Chem. B, 109, 1730 (2005).

[33.] S.G. Babu, M. Gopiraman, D. Deng, K. Wei, R. Karvembu, and I.S. Kim, Chem. Eng. J., 300, 146 (2016).

[34.] P. Wang, G.Y. Wang, W.L. Qiao, and Y.S. Feng, Catai. Lett., 146, 1792 (2016).

[35.] M.Q. Yang, X. Pan, N. Zhang, and Y.J. Xu, CrystEngComm, 15, 6819 (2013).

Liangliang Xu, (1) Shengkui Ma, (1,2) Xi Chen, (1,2) Chu Zhao, (1,2) Yiping Zhao, (1,2) Li Chen (1,2)

(1) Department of Polymer Science, School of Materials Science and Engineering, Tianjin Polytechnic University, Tianjin 300387, People's Republic of China

(2) State Key Laboratory of Separation Membrane and Membrane Process, Tianjin Polytechnic University, Tianjin 300387, People's Republic of China

Additional Supporting Information may be found in the online version of this article.

Liangliang Xu and Shengkui Ma equally contributed to the work. Correspondence to: X. Chen; e-mail: or Y. Zhao; e-mail:

Contract grant sponsor: National Natural Science Foundation of China; contract grant numbers: 51003076 and 51303129; contract grant sponsor: Science and Technology Commission Foundation of Tianjin; contract grant numbers: 14JCZDJC38100 and 14JCQNJC03200; contract grant sponsor: Science and Technology Plants of Tianjin; contract grant number: 15PTSYJC00250.

DOI 10.1002/pen.24542

Published online in Wiley Online Library (

Caption: FIG. 1. Schematic diagram of cross flow catalysis measurement. [Color figure can be viewed at]

Caption: FIG. 2. FTIR spectra of pure PVDF membrane (a), PVDF/PMAA blend membrane (b), and composite membranes: PVDF/PMAA-Ag membrane (c), PVDF/PMAA-Pd membrane (d), and PVDF/PMAA-Au membrane (e). [Color figure can be viewed at]

Caption: FIG. 3. Morphologies and structures of membranes: SEM images of the top surface (a, b) and cross-section (c-g) of various membranes: pure PVDF membrane (a, c); PVDF/PMAA blend membrane (b, d); PVDF/PMAA-Au composite membrane (e); PVDF/PMAA-Pd composite membrane (f); PVDF/PMAA-Ag composite membrane (g). TColor figure can be viewed at]

Caption: FIG. 4. XPS survey spectra of PVDF/PMAA-Ag membrane, PVDF/PMAA-Au membrane, PVDF/PMAA-Pd membrane (a); Ag 3d core-level spectrum of PVDF/PMAA-Ag membrane (b); Au 4f core-level spectrum of PVDF/ PMAA-Au membrane (c); Pd 3d core-level spectrum of PVDF/PMAA-Pd membrane (d). [Color figure can be viewed at]

Caption: FIG. 5. Feed solution flowing through a membrane under a cross flow model (a), ultraviolet absorption spectra of feed solution at different times (b), and change of ln([A.sub.t]/[A.sub.0]) with reaction time (c). PVDF/PMAA-Ag composite membrane was used. [Color figure can be viewed at]

Caption: FIG. 6. Effect of initial concentration of p-nitrophenol on reduction reaction, (a) ln([A.sub.t]/[A.sub.0]) versus t; (b): [] versus concentration. [Color figure can be viewed at]

Caption: FIG. 7. Effect of reaction temperature on reduction reaction, (a) ln([A.sub.t]/[A.sub.0]) versus t; (b) ln([]) versus 1/T. [Color figure can be viewed at]

Caption: FIG. 8. Effect of different composite membranes on reduction reaction. [Color figure can be viewed at]

Caption: FIG. 9. The conversions (%) for the hydrogenation reduction of p-nitrophenol of the three composite membranes in cross-flow mode and static mode. [Color figure can be viewed at]

Caption: FIG. 10. The reactant solution flow rates of composite membranes in penetrative flow catalysis and tangential flow catalysis under different operating pressures (a), and the conversions of three composite membranes for the hydrogenation reduction of p-nitrophenol under different operating pressures (b). [Color figure can be viewed at]

Caption: FIG. 11. Reusability of three composite membranes for the hydrogenation reduction of p-nitrophenol. [Color figure can be viewed at]
TABLE 1. Structures and compositions of various membranes.

                               Pure         PVDF/
Membrane                       PVDF        PMAA-Ag

Porosity (%)                   61.0          72.0
Average pore size (nm)          --          22.45
Contents of metal               0            0.94
  NPs (mg/g)
Average size of                 0     28.8 [+ or -] 1.76
  metal NPs (nm)
Specific surface area of        0           19.86
  metal NPs ([a.sub.s],
Surface area-based rate         0            4.21
  constant of metal NPs
  ([k.sub.s], L [min.sup.-1]

                                     PVDF/                PVDF/
Membrane                            PMAA-Au              PMAA-Pd

Porosity (%)                          71.8                 72.1
Average pore size (nm)               24.89                24.25
Contents of metal                     1.45                 0.54
  NPs (mg/g)
Average size of                43.4 [+ or -] 1.92   26.3 [+ or -] 1.73
  metal NPs (nm)
Specific surface area of              7.16                18.98
  metal NPs ([a.sub.s],
Surface area-based rate               7.55                16.03
  constant of metal NPs
  ([k.sub.s], L [min.sup.-1]
COPYRIGHT 2018 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Xu, Liangliang; Ma, Shengkui; Chen, Xi; Zhao, Chu; Zhao, Yiping; Chen, Li
Publication:Polymer Engineering and Science
Article Type:Report
Date:Feb 1, 2018
Previous Article:Clarification of Linear Low-Density Polyethylene Using Bis-Oxalamide Compounds.
Next Article:Study on Structure and Property Relations of [alpha]-iPP During Uniaxial Deformation via In Situ Synchrotron SAXS/WAXS and POM Investigations.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |