Carbon Nanofibers Decorated with Poly(furfuryl alcohol)-Derived

Oct 8, 2011 - The present paper introduces a novel method to functionalize nanofiber surfaces with carbon or silica nanoparticles by dip coating. This...
0 downloads 0 Views 2MB Size
Download PDF
ARTICLE pubs.acs.org/Langmuir

Carbon Nanofibers Decorated with Poly(furfuryl alcohol)-Derived Carbon Nanoparticles and Tetraethylorthosilicate-Derived Silica Nanoparticles Y. Zhang† and A. L. Yarin*,†,‡ †

Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, 842 West Taylor Street, Chicago Illinois 60607-7022, United States ‡ Center for Smart Interfaces, Technische Universit€at Darmstadt, Petersenstrasse 32, 64287 Darmstadt, Germany ABSTRACT: The present paper introduces a novel method to functionalize nanofiber surfaces with carbon or silica nanoparticles by dip coating. This novel approach holds promise of significant benefits because dip coating of electrospun and carbonized nanofiber mats in poly(furfuryl alcohol) (abbreviated as PFA) is used to increase surface roughness by means of PFAderived carbon nanoparticles produced at the fiber surface. Also, dip coating in tetraethylorthosilicate (abbreviated as TEOS) is shown to be an effective method for decorating carbon nanofibers with TEOS-derived silica nanoparticles at their surface. Furthermore, dip coating is an inexpensive technique which is easier to implement than the existing methods of nanofiber decoration with silica nanoparticles and results in a higher loading capacity. Carbon nanofiber mats with PFA- or TEOS-decorated surfaces hold promise of becoming the effective electrodes in fuel cells, Li-ion batteries and storage devices.

1. INTRODUCTION Numerous three-dimensional electrospun nanofiber mats have been produced and studied for several years.1 One of the important advantages of nanofiber mats is their ultrahigh surfaceto-volume ratio. Providing an extra rough nanocoating on the surface of such nanomaterials can lead to even higher surface-tovolume ratio. This was previously demonstrated using electroplating of electrospun nanofiber mats and studying their effect on drop/ spray cooling.2 The value of such additional nanocoatings further increases because they can possess special thermal and/or electronic characteristics along with a three-dimensional architecture.3 6 Carbon nanofiber mats and microporous carbon membranes have been widely used in microelectronics and in gas separation. Polymers such as polyimide (PI),7,8 poly(furfuryl alcohol) (PFA),9,10 polyacrylonitrile (PAN),11 15 and phenolic resin16,17 have been used as carbon precursors to prepare carbon nanofiber mats as well as some other carbon nano- and micromaterials. Electrospinning followed by sol gel processes, carbonization or calcination was widely used for the production of carbon, silica and metal oxide nanofibers. Carbon precursors, such as PAN, pitch, poly(amic acid) (PAA), poly(p-xylenetetrahydrothiophenium chloride) (PXTC), polyimide, poly(benzimidazole) (PBI), and poly(vinyl alcohol) (PVA) have been electrospun and used to make carbon fibers.18 For making silica nanofibers, tetraethylorthosilicate [TEOS, Si(OC2H5)4] is one of the most popular silicon alkoxides used as a precursor.19 Some other metal oxide fibers were produced using electrospinning from polymer solutions with added metal acetates, such as copper acetate, nickel r 2011 American Chemical Society

acetate,20 palladium acetate21 and zinc acetate,22 followed by an appropriate postprocessing. Silica nanoparticles and some other combinations of organic/ inorganic materials with silica have been widely studied and used due to their thermal and electronic stability. The preparation and usage of silica nanoparticles arranged in different ways are described in refs 23 25. The present study aims at the preparation of carbon nanofiber mats with enhanced roughness by using dip coating in furfuryl alcohol, FFA, as a precursor to increase roughness. To achieve this goal, such parameters as FFA concentration, as well as FFA coating process and carbonization conditions were optimized to control and increase nanofiber roughness. A similar approach was taken to achieve rough carbon nanofibers decorated with silica nanoparticles. These decorated fibers were produced by means of dip coating electrospun nanofibers into TEOS which hydrolyzes and condenses as silica deposits at the carbon nanofiber surface.

2. EXPERIMENTAL SECTION: MATERIALS AND METHODS 2.1. Materials. PAN (Mw = 150 kDa) was obtained from Polymer Inc. N-Dimethylformamide (DMF), 99.8% anhydrous, FFA, TEOS, 37% hydrochloric acid, and ethanol were obtained from Sigma-Aldrich. Phenolic resin (PhR, novolac resin, Mw = 4000 5000) was donated by Gunei Chemical Industry Co., Gunma, Japan. Poly(vinyl butyral) Received: September 22, 2011 Published: October 08, 2011 14627

dx.doi.org/10.1021/la203728k | Langmuir 2011, 27, 14627–14631

Langmuir

ARTICLE

Figure 1. SEM images of carbonized nanofiber mats: (a) overall view of the carbonized mat; (b and c) zoomed-in views.

(PVB, Butvar B-98) and methanol (MeOH) were purchased from Sigma-Aldrich.

2.2. Electrospinning and Carbonization of Polymer Nanofiber Mats. Polyacrylnitrile (PAN, 12%) solutions containing 15 wt % of carbon black (CB) nanoparticles were electrospun using the electrospinning setup described elsewhere.26,27 The fiber mats produced had a very loose three-dimensional structure with large interfiber pores purposely created to use them in the future as electrodes in microbial fuel cells,6,28,29 Li-ion batteries and storage devices. The PAN nanofiber mats were placed in a furnace and stabilized in air for 20 min at 270 °C, then carbonized in nitrogen at 1050 °C (the ramp rate was 4 °C/min between room temperature and 270 °C and between 270 and 1050 °C). The presence of carbon black nanoparticles in polymer solutions in the electrospun jet increased charge density and enhanced the electrostatic self-repulsion of the jet loops during the bending instability. This resulted in nanofiber mats with a very loose structure with large pores of the order of 10 μm (Figure 1). These are the attractive features for the electrodes of microbial fuel cells.6 Overall, the resulting carbon nanofiber mats possessed a rigid structure with large pores which makes them ideal template for dip coating.

2.3. Dip Coating in FFA and Additional Heat Treatment. Individual carbon nanofiber mats were immersed into pure FFA for 30 s to increase roughness. A fiber of cross-sectional radius, a, after dip coating withdraws a liquid film of thickness of the order of h ≈ aCa2/3, with Ca = μU/σ being the capillary number, μ and σ being the viscosity and surface tension of the liquid, respectively, and U being the withdrawal velocity.30 This estimate follows the Landau Levich theory of withdrawal process when the capillary number Ca , 1. 31 For withdrawal from FFA, taking for the estimate μ ≈ 0.005 Pa 3 s, σ ≈ 40  10 3 N/m, and U ≈ 0.1 m/s, the value of the capillary number is Ca = 0.0125. Therefore, for the fiber cross-sectional radius a ≈ 500 nm, the film thickness is approxinately 27 nm. However, more FFA could be entrapped in the pores of the nanofiber mats. Hence, the excess of FFA was absorbed with a cotton stick to prevent formation of fiber flocks. Then, carbon nanofibers coated with FFA were inserted into a

Figure 2. SEM images of the extremely rough carbon fiber mats: (a) overall view of a carbon fiber mat; (b d) zoomed-in views of the carbonized fiber mat (d, individual carbon fiber). Grainlike structures at the fiber surfaces resulted from the carbonized FFA film withdrawn in the dip coating process. glass vessel and dried in a chemical hood for 6 h. After drying, a droplet of 5% HCl solution was dripped on top of the nanofiber mat as a polymerization catalyst and left to dry for 1 h. After that, the nanofiber mat was put into a furnace to dry and stabilize in air for 20 min at 60 °C and then polymerized for 2 h at a temperature of 90 °C in a nitrogen atmosphere. Next, temperature was increased to 250 °C to stabilize the fibers in nitrogen for 20 min and after that the fibers were carbonized in nitrogen at 1000 °C. The ramp rates were 4 °C/min between room temperature and 60 °C, between 60 and 90 °C, and between 90 and 250 °C; the ramp rate was 3 °C/min between the 250 and 1000 °C. The images of the resulting carbon nanofiber mats are shown in Figure 2. It is 14628

dx.doi.org/10.1021/la203728k |Langmuir 2011, 27, 14627–14631

Langmuir

ARTICLE

Figure 3. SEM images of carbon nanofiber mat decorated with silica nanoparticles at the surface: (a) overall view of a carbon nanofiber mat with silica nanoparticles at the surface; (b) zoomed-in view of the top layers of a carbon nanofiber mat decorated with silica nanoparticles. clearly seen that the roughness of the individual nanofibers (and as a result, of the whole mat) was dramatically increased. It is emphasized that the added volume V due to an attached sphere is of the order of the cube of sphere diameter, i.e., D3, while the added surface area S is of the order of D2. The ratio S/V ∼ D2/D3 ∼ D 1. For large sphere diameters D f ∞ and the ratio S/V f 0. However, for the small diameters D f 0 (exactly as in the case of nanometer-sized spheres), and the ratio S/V f ∞. Therefore, the effect of the added surface area of our spheres always overbears that of the added volume (mass), hence, the nanofiber mats became rougher and acquire a higher surface area (cf. Figure 2).

2.4. Carbon Nanofiber Mats Decorated with Silica Nanoparticles. To decorate carbon nanofiber mats with silica nanoparticles at the surface, two kinds of solutions were prepared: solution A, TEOS (41.7 g) mixed with ethanol (0.45 g); solution B, DI water (3.47 g) and 37% hydrochloric acid (0.034 g) mixed with ethanol (0.45 g). Solutions A and B were sonicated for 30 min. Then, solution A was stirred harshly and solution B was added drop by drop. After 20 min a uniformly dispersed solution of B in A was obtained. A small piece of carbonized nanofiber mat was dipped into the obtained reactive mixture, which was continually stirred for another 10 min. The latter was done to guarantee that small dispersed gel particles uniformly deposit onto carbon nanofibers and adhere to them. After that, the nanofiber mat was withdrawn and put into a chemical hood for 12 h to allow hydrolysis and polycondensation as much as possible. The final product consisted of SiO2 nanoparticles attached to the nanofiber surface, i.e. silica nanoparticles precipitated at the fiber surfaces. The spherical silica particles formed via the acid-catalyzed hydrolysis and polycondensation of tetraethylorthosilicate in an aqueous mixture containing ethanol and hydrochloric acid. The images of the resulting carbon nanofiber mat decorated with silica nanoparticles at the surface are shown in Figure 3. 2.5. Carbon Nanofibers Derived from Phenolic Resin. For electrospinning, 4.36 g of phenolic resin and 0.14 g of poly(vinyl butyral) (PVB) were dissolved in 5.5 g of MeOH to prepare a solution similar to that of ref 16. The electrospun resin nanofiber mats were stabilized by immersing them into 15% hydrochloride solution followed by adding 10% formaldehyde. Then the fiber mats were withdrawn and put in the chemical hood overnight. After that, the stabilized resin fiber mats were placed in a furnace and stabilized in air for 30 min at 60 °C, then carbonized in nitrogen at 1050 °C (the ramp rate was 1.5 °C min 1 between room temperature and 60 °C, as well as between 60 and 300 °C, while the rate was 4 °C min 1 between 300 and 1050 °C).

3. RESULTS AND DISCUSSION In the present work an enhanced roughness of carbon nanofiber mats, as well as their decoration with silica nanoparticles at the surface, was achieved by very simple means of dip coating of electrospun nanofibers into FFA or TEOS.

Figure 4. SEM images of resin and PAN layer-by-layer fiber mats: (a) overall view of a resin and PAN layer-by-layer fiber mat; (b) zoomedin view of the fiber mat; (c) overall view of a carbonized resin and PAN layerby-layer nanofiber mat; (d) zoomed-in view of the carbonized nanofiber mat.

The individual carbonized nanofibers were roughened using dip coating in FFA as an extra carbon coating precursor. The roughness and final diameter of the carbon nanofibers can be easily controlled by changing the concentration of the FFA solution or repeating the dip coating process with the subsequent heat treatment several times. Comparison of Figures 1 and 2 shows that the roughness of the individual nanofibers, their diameters, and the pore sizes in the carbon nanofiber mats have all been changed after dip coating in FFA followed by the additional heat treatment. By using this method, the surfaces of the individual carbon nanofibers can be made extremely rough, which leads to a much higher surface to volume ratio compared to ordinary carbon nanofibers. The extremely rough carbon nanofiber mats produced in this work represent themselves an attractive candidate for the electrode material for microbial and other fuel cells, Li-ion batteries, and storage devices. Similarly, dip coating of carbon nanofibers in TEOS as a silica precursor resulted in a very simple method of decorating nanofiber surfaces with silica nanoparticles. Compared to other methods of decoration of nanofiber surfaces with silica nanoparticles, e.g., co-electrospinning,32 the present method is easier to implement and results in a higher loading capacity. In the present method the silica nanoparticle size at the carbon nanofiber surface is controlled by adjusting the conditions of sol gel process. In particular, controlling the viscosity of TEOS gel can be used to change the size of silica nanoparticles emerging at the fiber surface. Moreover, in Figure 3 one can see that the nanofiber mats are heavily decorated with silica nanoparticles. The resin fiber mats produced by the method proposed in ref 16 were very crispy which practically prevented their further handling. Therefore, we introduced a different approach wherein the resin and PAN nanofiber layers are deposited layer-by-layer (such mats are shown in Figure 4). As a first step, a thin layer of PAN nanofibers was electrospun onto the collector; then a thick layer of resin nanofibers was electrospun on top of it. The fiber mats produced by this method have a much higher mechanical strength than the pure resin nanofiber mats. This is illustrated by the experimental results depicted in Figure 5. 14629

dx.doi.org/10.1021/la203728k |Langmuir 2011, 27, 14627–14631

Langmuir

ARTICLE

5. CONCLUSION Carbon nanofiber mats with enhanced surface roughness can be fabricated using a combination of the following inexpensive techniques: electrospinning, dip coating in FFA, and pyrolysis. Similarly, carbon nanofiber mats decorated with silica nanoparticles can be obtained using electrospinning, dip coating in TEOS, carbonization, hydrolysis, and polycondensation. These novel materials hold a great potential of applications in microbial and other fuel cells, Li-ion batteries, and storage devices. The nanotextured carbon mats developed here possess the electric conductivity of the bulk carbon pyrolyzed at the same temperature. It is emphasized that the previous generation of such carbonized electrospun electrodes revealed in our work6 has a bioelectrocatalytic anode current density of up to 30 A m 2, which represents the highest reported values for electroactive microbial biofilms. In addition, they are currently tested in Li-ion batteries. Figure 5. Comparison of mechanical properties of pure resin nanofiber mats with those of the nanofiber mat with the resin and PAN nanofiber layers deposited layer-by-layer: (a c) folding/unfolding test; (d and e) squeezing test.

Panel a in Figure 5 on the left shows a macroscopic sample of the pure resin nanofiber mat. On the right-hand side of Figure 5a is the nanofiber mat with the resin and PAN nanofiber layers deposited layer-by-layer. Panel b in Figure 5 shows both nanofiber mat samples folded, whereas panel c shows the same samples after they were unfolded. It is clear that the pure resinmade sample has cracked and broke. On the other hand, the nanofiber mat made of resin and PAN nanofibers deposited layer-by-layer is intact. In panel d in Figure 5, a third sample is added on the right. This additional sample was also made of the pure resin nanofibers. Panel e in Figure 5 shows a glass slide which was put on top of the three nanofiber mat samples and pressed by a heavy metal piece several times from the top. It is seen that both pure resin nanofiber mats were crumbled, whereas the nanofiber mat made of the resin and PAN nanofibers deposited layer-by-layer stays intact. It is emphasized that we were also able to make tensile tests of the carbonized rectangular samples of nanofiber mats made of the resin and PAN nanofibers deposited layer-by-layer and measure the entire stress strain curve and the corresponding mechanical parameters (Young’s modulus, yield stress, and stress and strain at failure).

4. MEASUREMENTS OF ELECTRIC CONDUCTIVITY OF CARBON NANOFIBER MATS The electric resistivity/conductivity of different fiber mats was measured by a digital multimeter. The corresponding electric conductivity results are as follows: the conductivity of carbonized PAN + CB nanofiber mat is 126 S/m; for PAN + CB with single FFA treatment the conductivity is 290 S/m; for PAN with CB with double FFA treatment the conductivity is 286 S/m. The conductivity of carbon nanofiber mats derived from phenolic resin is 243 S/m. These are the bulk carbon conductivity levels expected for the pyrolysis temperatures in the range of 1000 1050 °C employed in the present work.33 According to ref 33, an increase of 2 orders of magnitude is expected for the pyrolysis temperature of about 1300 °C.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Phone: +1(312) 996-3472. Fax: +1(312) 413-0447.

’ ACKNOWLEDGMENT Partial support of this work by NSF (Grant CBET 0966764) and NASA (Grant NNX10AR99G) is greatly appreciated. ’ REFERENCES (1) Reneker, D. H.; Yarin, A. L. Polymer 2011, 49, 2387. (2) Sinha-Ray, S.; Zhang, Y.; Yarin, A. L. Langmuir 2011, 27, 215. (3) Kataphinan, W.; Teye-Mensah, R.; Evans, E. A.; Ramsier, R. D.; Reneker, D. H.; Smith, D. J. J. Vac. Sci. Technol. 2003, A21, 1574. (4) Choi, S.; Lee, S.; Im, S.; Kim, S.; Joo, Y. J. Mater. Sci. Lett. 2003, 22, 891. (5) Sakai, S.; Yamada, Y.; Yamaguchi, T.; Kawakami, K. Biotechnol. J. 2006, 1, 958. (6) Chen, S.; Hou, H.; Harnisch, F.; Patil, S.; Carmona-Martinez, A. A.; Agarwal, S.; Zhang, Y.; Sinha-Ray, S.; Yarin, A. L.; Schroder, U.; Greiner, A. Energy Environ. Sci. 2011, 4, 1417. (7) Yamamoto, M.; Kusakabe, K.; Hayashi, J.; Morooka, S. J. Membr. Sci. 1997, 133, 195. (8) Jones, C. W.; Koros, W. J. Carbon 1994, 32, 1419. (9) Shiflett, M. B.; Foley, H. C. Science 1999, 285, 1902. (10) Sedigh, M. G.; Onstot, W. J.; Xu, L.; Peng, W. L.; Tsotsis, T. T.; Sahimi, M. J. Phys. Chem. A 1998, 102, 8580. (11) Zussman, E.; Yarin, A. L.; Bazilevsky, A. V.; Avrahami, R.; Feldman, M. Adv. Mater. 2006, 18, 348. (12) Yarin, A. L.; Zussman, E.; Wendorff, J. H.; Greiner, A. J. Mater. Chem. 2007, 17, 2585. (13) Bazilevsky, A. V.; Yarin, A. L.; Megaridis, C. M. Langmuir 2007, 23, 2311. (14) Woo, S. W.; Dokko, K.; Nakano, H.; Kanamura, K. J. Power Sources 2009, 190, 596. (15) Sinha-Ray, S.; Yarin, A. L.; Pourdeyhimi, B. Carbon 2010, 48, 3575. (16) Imaizumi, S.; Matsumoto, H.; Suzuki, K.; Minagawa, M.; Kimura, M.; Tanioka, A. Polym. J. 2009, 41, 1124. (17) Hiralal, P.; Imaizumi, S.; Unalan, H. E.; Matsumoto, H.; Minagawa, M.; Rouval, M.; Tanioka, A.; Amaratunga, G. A. J. ACS Nano 2010, 4, 2730. (18) Cheng-Kun, L.; Kan, L.; Wei, L.; Mu, Y.; Run-Jun, S. Polym. Int. 2009, 12, 1341. 14630

dx.doi.org/10.1021/la203728k |Langmuir 2011, 27, 14627–14631

Langmuir

ARTICLE

(19) Kenji, I.; Toshiyuki, O.; Michitaka, S.; Mitsuaki, H. Mater. Lett. 2010, 1, 64. (20) Guan, H.; Shao, C.; Chen, B.; Gong, J.; Yang, X. Inorg. Chem. Commun. 2003, 6, 1409. (21) Viswanathamurthia, P.; Bhattarai, N.; Kim, H. Y.; Cha, D. I.; Lee, D. R. Mater. Lett. 2004, 58, 3368. (22) Yang, X.; Shao, C.; Guan, H.; Li, X.; Gong, J. Inorg. Chem. Commun. 2004, 7, 176. (23) Pu, Z.; Mark, J. E.; Jethmalani, J. M.; Ford, W. T. Chem. Mater. 1997, 9, 2442. (24) Feng, Y.; Yuchun, O.; Zhongzhen, Y. J. Appl. Polym. Sci. 1998, 69, 355. (25) Zaleski, R.; Stefaniak, W.; Maciejewska, M.; Goworek, J. J. Colloid Interface Sci. 2010, 343, 134. (26) Reneker, D. H.; Yarin, A. L.; Fong, H.; Koombhongse, S. J. Appl. Phys. 2000, 87, 4531. (27) Reneker, D. H.; Yarin, A. L.; Zussman, E.; Xu, H. Adv. Appl. Mech. 2007, 41, 43. (28) Torres, C. I.; Marcus, A. K.; Rittmann, B. E. Biotechnol. Bioeng. 2008, 100, 872. (29) Torres, C. I.; Marcus, A. K.; Lee, H. S.; Parameswaran, P.; Krajmalnik-Brown, R.; Rittmann, B. E. FEMS Microbiol. Rev. 2010, 34, 3. (30) Quere, D. Annu. Rev. Fluid Mech. 1999, 31, 347. (31) Levich, V. G. Physicochemical Hydrodynamics; Prentice Hall: Englewood Cliffs, NJ, 1962. (32) Li, D.; McCann, J. T.; Xia, Y. Small 2005, 1, 83. (33) Wang, Y.; Santiago-Aviles, J. J.; Furlan, R.; Ramos, I. IEEE Trans. Nanotechnol. 2003, 2, 39.

14631

dx.doi.org/10.1021/la203728k |Langmuir 2011, 27, 14627–14631