Template Synthesis of Nanostructured ... - IngentaConnect

4 downloads 0 Views 488KB Size Report
Electrochemical Devices. Rocío Fernández-Saavedra1, Pilar Aranda1, Kathleen A. Carrado2, Giselle Sandí2, Soenke Seifert2, and Eduardo Ruiz-. Hitzky1*.
Current Nanoscience, 2009, 5, 506-513

506

Template Synthesis of Nanostructured Carbonaceous Materials for Application in Electrochemical Devices Rocío Fernández-Saavedra1, Pilar Aranda1, Kathleen A. Carrado2, Giselle Sandí2, Soenke Seifert2, and Eduardo RuizHitzky1* 1 2

Instituto de Ciencia de Materiales de Madrid, CSIC, Cantoblanco, 28049-Madrid, Spain Argonne National Laboratory, 9700 South Cass Ave., Argonne, IL 60439, USA Abstract: Novel conducting nanostructured carbonaceous materials of various microstructures have been successfully prepared from nanocomposites containing polyacrylonitrile (PAN) inside the nanosized pores of imogolite, sepiolite and porous alumina membrane templates. The synthesis and electrochemical characterization of the PAN-inorganic porous solid nanocomposites, as well as their carbonaceous derivatives produced after removal of the matrix, have been studied by CHN elemental chemical analysis, thermal analysis (TGDTA), specific surface area and porosity determinations (N2 isotherms), X-ray diffractometry, FTIR, Small-Angle X-ray Scattering (SAXS), SEM, TEM and Electrochemical Impedance Spectroscopy (EIS). The properties of the carbonaceous materials as electroactive materials in electrochemical devices such as rechargeable Li-ion batteries and Electrochemical Double-Layer Capacitors (EDLC) are reported. The main objective of this work is to study the influence of the template used for the preparation of different carbonaceous materials on their physical-chemical characteristics as well as their electrochemical properties, including their behaviour as electrode materials for Li-ion batteries and EDLC capacitors.

Keywords: Imogolite, sepiolite, porous alumina membranes, polyacrylonitirile, nanostructured carbons, electrochemical devices. 1. INTRODUCTION The design and synthesis of organic-inorganic hybrid materials constitutes a new strategy for preparing materials with synergistic or complementary behavior between organic species and inorganic solids interacting at the molecular level. These interactions have attracted increasing interest for structural and functional applications [1, 2]. Among inorganic solids, clay minerals have been widely studied as host matrices of a large variety of organic species [3-6]. In this context, intercalation of electroactive polymers into porous inorganic materials is an excellent way of constructing novel polymer-inorganic nanoassemblies showing peculiar electrical properties [7-12].

carbon precursors such as propylene, ethylene, pyrene, furfuryl alcohol and acetylene [28-32]. In situ polymerization of carbon precursors such as PAN, polypyrrole, or poly(furfuryl alcohol), previously incorporated into the pores of alumina membranes have been used for obtaining novel carbonaceous materials as well [33]. However, relatively few of these carbonaceous derivatives have been tested as electroactive materials for electrochemical systems such as rechargeable Li-ion batteries or EDLC capacitors [34-36].

Polymer-clay nanocomposites have been employed as intermediate phases to produce carbonaceous conducting materials. Different carbon precursors (sucrose, acrylonitrile, furfuryl alcohol, vinyl acetate, pyrene, ethylene, propylene) are easily intercalated in clays, giving rise to the formation of carbon-clay nanocomposites [10, 11, 13, 14]. In addition, the resulting carbonaceous materials have been tested as electrodes for Li insertion [15-18]. We have previously reported the ability of sepiolite to act as template substrate for the preparation of carbon nanofibers from in situ formed PAN-sepiolite nanocomposites [19]. Other inorganic porous solids such as mesoporous silicas (i.e., MCM-41, MCM-48, SBA-15) and zeolites have been used as templates to prepare carbonaceous materials from PAN [20-24]. Porous alumina membranes obtained by anodization of aluminum with regular pore size (Fig. (1A)) have been used for the preparation of diverse organic-inorganic systems with potential applications in nanoelectronics, magneto-optics, biotechnology, and optoelectronics [25]. Martin et al. have been pioneers on using porous alumina membranes as template for the preparation of nanostructured materials that could be conformed as nanowires, nanofibers, nanotubes, or other kind of nanostructures [26, 27]. Porous alumina membranes have been ideally suited as templates for the preparation of carbonaceous materials by chemical vapor deposition (CVD) of

*Address correspondence to this author at the Instituto de Ciencia de Materiales de Madrid, CSIC, Cantoblanco, 28049-Madrid, Spain; Tel: +34913349039; Fax: +34 91 3720623; E-mail: [email protected]

1573-4137/09 $55.00+.00

Fig. (1). Schematic structures of (A) a porous alumina membrane and the clay minerals (B) sepiolite and (C) imogolite.

In this paper we discuss the characteristics of carbonaceous materials prepared from the carbonization of PAN in different porous systems: two clay minerals, the natural fibrous silicate sepiolite and a synthetic imogolite, and porous alumina membranes with nominal diameters of 200 and 100 nm. The in situ generation of PAN from © 2009 Bentham Science Publishers Ltd.

Template Synthesis of Nanostructured Carbonaceous Materials

Current Nanoscience, 2009, Vol. 5, No. 4

507

Scheme 1. Schematic illustration of the procedures followed for the preparation of PAN-inorganic nanocomposites and their templated carbonaceous derivatives.

acrylonitrile was chosen to assure the accessibility of the polymer to the small size pores of sepiolite and imogolite. The effect of the template on the carbon materials’ structure and their influence on the electrical and electrochemical properties is reported here. Clay minerals have either a layered or porous structure that is able to insert a large variety of organic species in their structure [5, 6]. The clay mineral sepiolite is a hydrated magnesium silicate with the theoretical half unit-cell formula Si12O30Mg8 (OH, F)4 (OH2)4·8H2O [37], structurally formed by blocks similar to talc (i.e., an anhydrous magnesium 2:1 layer silicate) (Fig. (1B)). Each block is constituted by an octahedral sheet of magnesium oxidehydroxide, packed between two tetrahedral silica layers. Each of the Mg2+ cations located at the edges of the octahedral sheets, i.e., those acceding to the tunnels, completes their octahedral coordination by bonding to two molecules of water (coordinated water). The periodic inversion of the SiO4 tetrahedra is the origin of structural cavities (i.e., tunnels) extended along the c axis, i.e., the axis of the microfibers, with openings of ca. 1.08 nm x 0.4 nm. Inside the tunnels are located the coordinated water, a small number of exchangeable cations and zeolitic water. These tunnels are accessible to a large variety of species, the only restriction being related to the dimensions of the guest species, where the mineral acts as a molecular sieve. The external surface of sepiolite microfibers is formed by channels (external tunnels) and structural steps defined by the external blocks distribution. Because of the discontinuity of the silica sheets, silanol groups are present on the external surface of the silicate particles and are directly accessible to reagents, allowing the organic groups functionalization of the mineral [38, 39]. Imogolite is a hydrated aluminosilicate considered as a clay mineral by the AIPEA [40]. It has a net composition of (HO)3Al2SiOH and an unique one dimensional microporous tubular structure constituted by hollow tubes with an external diameter of ca. 2 nm and lengths ranging from about 400 nm to several micrometers (Fig. (1C)). The tube walls consist of curved gibbsite-like sheets with Si-OH (silanol) groups on the inside and Al-OH groups on the outside surface, respectively [41]. Three different pore types are postulated for this clay which include: the space within the tubes, the space between three aligned tubes in a regular packing and mesoporosity associated with the packing of tube bundles (Fig. (1C)). Under ambient atmospheric and temperature conditions imogolite tubes are filled with water, though a large number of studies have been performed on the adsorption of neutral molecules on imogolite [42-45].

2. EXPERIMENTAL SECTION 2.1. Starting Materials and Reagents Synthetic imogolite was obtained following a method already reported by Koenderink et al. [46]. Sepiolite from the VallecasVicálvaro clay deposits (Madrid, Spain) was kindly provided by TOLSA S.A. This is a commercial grade, named Pangel, which is prepared by a wet grinding, containing > 95 % of pure sepiolite. It was used as-received. Commercially porous alumina filtration membranes (ANODISC®) were purchased from Whatman. Acrylonitrile monomer (AN; > 99 %) and PAN (average Mw = 150.000) were obtained from Aldrich. Azo-bis-isobutyronitrile (AIBN; Fluka) was used as the initiator in the radical polymerization of AN. Dimethyl sulfoxide (DMSO; (Fluka,  99 %) and methanol (Fluka,  99.8 %) were used for the PAN excess elimination. HCl (37 % conc., Fluka) and HF (40 % conc., Panreac) were used for the clay removal. NaOH concentrated solutions (Fluka, pellets 98 %) were used for the porous alumina membranes elimination. Super P carbon (MMM carbon), EPDM (methylene-propylene-diene monomer) and cyclohexane (Normasolv) were used for the electrodes preparation. LiPF6 in a mixture of ethylene carbonate: dimethylcarbonate (50 wt %) (Merck, LP30), was used as the electrolyte in both Li-insertion and supercapacitor cells and ribbon of Li metal (Aldrich, 0.38 mm thick, 0.23 mm wide) was used as the negative electrode for the Li-insertion cells. 2.2. Synthesis Procedures Nanostructured carbons prepared using clays as template were obtained by using PAN-clay nanocomposites as intermediate phases. The PAN-clay nanocomposites were prepared by in situ polymerization of AN previously inserted in the pores of the clay template (Scheme 1). To prepare the PAN-imogolite and PANsepiolite nanocomposites, a procedure described by Kurt et al. was followed [47]. In our method [19], clay powder was previously dehydrated at 140 ºC in vacuum for two hours in order to eliminate free water molecules before mixing with the monomer. Then, acrylonitrile was slowly sprayed out on the clay sample using a syringe and adding AIBN (0.7 wt %) as the radical initiator. This process was carried out inside a glove box to avoid contact of air water molecules with the clay mineral. After soaking for 24 hours, the mixture was then heated at 60 ºC for 24 hours in sealed bottles to induce the polymerization, leading to the PAN-clay nanocompo-

508 Current Nanoscience, 2009, Vol. 5, No. 4

sites formation. Clay (g): AN (mL) ratios of 1:0.6 and 1:3 were used for the PAN-imogolite and PAN-sepiolite nanocomposites preparation, respectively. The PAN-sepiolite samples were washed with DMSO and methanol to eliminate excess polymer loading on the external surface of the nanocomposites. To produce the carbon-clay materials, the PAN-clay complexes were further treated in two steps. First, the nanocomposites were heated to 250 ºC for 24 hours under air atmosphere to induce the cyclization and conjugation of PAN. During a second pyrolysis step, the solids resulting from the first step were heated (heating rate 1.6 ºC/min) from 250 ºC to 750 ºC, keeping this temperature for 3 hours under N2 flux, in order to carbonize the polymer. Free carbon-like materials were formed by extracting the clay moiety treating the samples with concentrated acids. Elimination of the silicate network from the carbon-imogolite nanocomposites was achieved by treating the samples with conc. HF (40 %) for 1 hour, followed by refluxing with conc. HCl (37 %) for 1.5 hours and then rinsing until complete removal of the acid. In the case of carbonsepiolite nanocomposites, a two steps acid treatment was applied instead. The samples were refluxed with HCl/isopropyl alcohol mixture (50 %) at 80 ºC for 48 hours in order to obtain carbon-silica nanocomposites. Then, carbon-silica complexes were treated with conc. HF (40 %) and then rinsed to neutral pH, for obtaining the resulting carbonaceous material. Nanostructured carbons derived from porous alumina membranes were prepared by a templating method using PAN as carbon precursor (Scheme 1). This polymer was directly inserted within the pores of alumina membranes through a simple method consisting on dipping the alumina template into a 2.5 % (wt-wt) PAN-DMSO solution for ca. 15 hours. The PAN-alumina system was transformed in a carbon-alumina one applying a two steps thermal treatment. First, the sample was heated up to 250 ºC for 4 hours in air (stabilization step) and then, the resulting product was submitted to a pyrolysis treatment in N2 at 650 ºC for 1 hour (carbonisation step). To remove the alumina template, the carbon-alumina system was treated in a 6N NaOH solution during 48 hours and sonicated for 10 min. After filtering the formed suspension through a 0.45 Nm pore size (DURAPORE®) filter, an easy handling carbon nanotubules film was obtained. 2.3. Characterization The PAN-clay nanocomposites, the PAN-alumina systems and the resulting carbonaceous materials were characterized by XRD (Bruker D8 instrument with a Cu anode and Ni filter), FTIR (Bruker, ISS 66V-S spectrophotometer), elemental chemical analysis (Perkin Elmer 2400 CHN analyzer), thermal analysis (TG-DTA, Seiko SSC-5200 equipment), TEM (LEO-910 microscope, operating at an accelerating voltage of 80 kV) coupled to an EDAX 9100 analyzer and SEM (ZEISS DSM-960 microscope, operating at an acceleration voltage of 20 kV). BET specific surface area and pore volume measurements were performed using N2 at 77 K (Flowsorb II 2300 apparatus from Micromeritics and Coulter Omnisorp 100 apparatus). In situ small-angle X-ray scattering (SAXS) experiments were carried out at the Advance Photon Source (Basic Energy Sciences Synchroton Research Center CAT), Argonne National Laboratory. The SAXS intensity of the investigated material I(Q) is function of the angle of scattering (2) and the wavelength () of the applied radiation, which could be expressed as Q = 4/sin . Monochromatic X-rays (20 keV) were scattered off the sample and collected on a 15 x 15 cm2 CCD camera. The scattered intensity was corrected for absorption and instrument background. The differential scattering cross section is expressed as a function of the scattering vector Q and its value is proportional to the inverse of the length scale (Å-1). The instrument was operated with a sample-to-detector distance of 880 mm to obtain data at 0.4 < Q < 1.1 Å-1.

Fernández-Saavedra et al.

Impedance experiments were carried out using a frequency response analyzer (Solartron 1255) coupled to a potentiostat/galvanostat (Princeton Applied Research PAR 273A). The samples were prepared as thin pellets (adding 5 wt % of EPDM as a binder) or films and covered with Au sputtered on both faces for assuring a tight contact with the Pt electrodes. A 300 mV or 10 mV amplitude signal (pellets or films, respectively) was applied over the frequency range of 100 kHz to 100 mHz. In the case of carbon materials formed on clays, the electrodes for the electrochemical devices were conformed into films, from a suspension containing 90 wt % of the carbonaceous material together with 5 wt % of Super P carbon black and a binder solution made of EPDM dissolved in cyclohexane. In the case of carbon materials grown on alumina membranes, films of carbon nanotubules obtained in the synthesis were directly used as electrodes. All the electrodes were oven-dried at 120 ºC overnight. The electrochemical cells consist of carbon/1M LiPF6 (in EC:DMC, 50%)/Li. Charge-discharge cycles were performed using a multi-channel potentiostat/galvanostat Celltest system® (Solartron 1480A), at constant current rates of 20 mA/g, in a potential range between 3.0-0 V vs. Li. To measure the capacitance, supercapacitor cells were built by carbonaceous material/1M LiPF6 (EC:DMC, 50 wt %)/carbonaceous material. The non-aqueous LiPF6 (EC:DMC, 50 wt %) electrolyte was chosen to ensure the integrity of the electrode films. Three different electrochemical techniques were applied consecutively. First, ten cycles of cyclic voltamperometry were performed using a potentiostat/galvanostat (Princeton Applied Research PAR 273A) at a constant rate of 5 mV/s in a potential range of 2.5-0 V. Second, 10 charge-discharge galvanostatic cycles were performed applying the constant current rate obtained from the previous cyclovoltamperometry experiment in the same potential range. Finally, EIS was conducted using a potentiostat/galvanostat Celltest system® (Solartron 1480A), coupled to a frequency response analyzer (Solartron 1260A), applying a 10 mV amplitude signal over the frequency range of 100 kHz to 1 mHz. 3. RESULTS AND DISCUSSION Acrylonitrile (AN) is a polar monomer able to diffuse inside the cavities of porous materials such as clays; e.g. montmorillonite and sepiolite has been previously reported [10, 13, 19, 48]. Imogolite also shows the ability for inserting acrylonitrile in its pores. A novel method for preparing novel PAN-imogolite nanocomposites is discussed here. The XRD diagram of the PAN-imogolite nanocomposite (Fig. (2B)) shows the presence of the 0.51 nm characteristic reflection of bulk PAN [49], indicating that at least a part of this polymer is crystallized on the external surface of the silicate similarly to that observed in PAN-sepiolite nanocomposites with PAN excess [19]. In addition, TG-DTA analyses of PAN-imogolite nanocomposites prove the presence of PAN in the pores of the clay. The DTA curve of the PAN-imogolite sample (Fig. (3)) shows an exothermic peak around 318 ºC, related to the elimination of the external PAN and corresponds to the amount of polymer in excess. The appearance of another exothermic peak at a higher temperature (414 ºC) can be ascribed to the PAN inside the pores of the clay, which is supporting more hindrance in its combustion than the one located on the external part of the clay. In this figure, the broad peak centred at 538 ºC is associated with PAN combustion in air atmosphere. It is known that molecules of pyridine, methylene blue, indigo, and propylene showing a ladder-like structure, are accessible to the sepiolite tunnels, preventing the folding of cavities up to temperatures around 500-600 ºC [50, 51]. In situ SAXS performed during the thermal heating of a PAN-sepiolite sample with a small excess

Template Synthesis of Nanostructured Carbonaceous Materials

Current Nanoscience, 2009, Vol. 5, No. 4

509

1.20 nm, associated with the 110 reflection of the sepiolite (Fig. (4)). The major change in the intensity of this peak and its shift to higher Q values (1.16 nm) occurs at temperatures above 530 ºC, which corresponds to the temperature of the sepiolite folding in the nanocomposite. Interestingly, the same experiment carried out over pristine sepiolite gives rise to a folding temperature of 487 ºC [52]. This delay of the sepiolite folding temperature in the nanocomposite by as much as 40 ºC, corroborates the insertion of AN molecules into the tunnels of the mineral. In addition, this PAN-sepiolite sample presents a surface area of 73 m2/g (Table 1), which fits to a clay sample with PAN molecules filling its intracrystalline pores or at least capping them (partial penetration into the tunnels).

Fig. (2). XRD patterns of (A) imogolite, (B) PAN-imogolite, (C) PANimogolite after heat treatment in air at 250 ºC, 24 hours, (D) carbonimogolite and (E) templated carbon derived from carbon-imogolite. Fig. (4). SAXS curves of PAN-sepiolite nanocomposite with a small excess of PAN heated from 20 to 700 ºC in N2 flux at 2 ºC/min.

The insertion of PAN within the pores of alumina membranes and its graphitization has been studied by FTIR spectroscopy (Fig. (5)). The PAN-alumina spectrum shows a band at 2244 cm-1 (CN) (Fig. (5B)) ascribed to the nitrile group of the polymer. During the stabilization step, this band disappears progressively with the concomitant developing of two new bands centred at 1589 cm-1 (CC, C=N) and 1382 cm-1 (C-N), respectively (Fig. (5C)), both associated with a ladder-like aromatic structure due to PAN cyclization and conjugation processes [53]. After the graphitization treatment, the band at around 1589 cm-1 (CC, C=N) characteristic of carbon structures derived from PAN [54, 55] is observed together with a band at 2342 cm-1 (CO) (Fig. (5D-E)), which is probably resulting from the substitution of oxygen by carbon in the carbon nanotubules that are finally obtained [56].

Fig. (3). TG/DTA analysis of PAN-imogolite nanocomposites carried out in air at 5 ºC/min rate.

EIS technique was applied to measure the electrical conductivity of the carbonaceous materials derived after removal of the inorganic template. Values of the electrical conductivity were deduced

of PAN (washed with DMSO), from 20 to 700 ºC under N2 flux, shows the gradual diminishing of the intensity of the peak at about

Table 1. CHN Chemical Analyses, Estimated Inorganic Residue, Surface Area Data and Electrical Conductivity Values of the Carbonaceous Materials Sample

C (%)

H (%)

N (%)

Estimated Inorganic Residue (%)

Surface Area (m2/g)

Electrical Conductivity (S/cm)

Carbon derived from bulk PAN

75.3

1.4

14.9

-

53

2.42·10-4

Nanostructured carbon from carbon- imogolite

69.2

1.7

12.1

10

484

5.43·10-2

Nanostructured carbon from carbon- sepiolite

61.9

2.1

13.2

15

33

4.11·10-8

Carbon nanotubules derived from alumina membrane (pore = 100 nm)

51.4

2.2

5.9

32

-

1.65·10-7

Carbon nanotubules derived from alumina membrane (pore = 200 nm)

46.8

2.3

5.9

37

-

1.80·10-6

510 Current Nanoscience, 2009, Vol. 5, No. 4

Fig. (5). FTIR spectra of (A) alumina membrane, (B) PAN-alumina, (C) PAN-alumina after heat treatment in air at 250 ºC, 4 hours, (D) carbonalumina and (E) carbon nanotubules derived from alumina membrane.

from the Nyquist plots (Table 1). Nanostructured carbons obtained after removing the clay template present different electrical conductivity values, being the one corresponding to the carbon derived from carbon-imogolite (5.43·10-2 S/cm) remarkably higher than the one adscribed to the carbon derived from carbon-sepiolite (4.11·10-8 S/cm) and even more elevated than the one associated with the carbon derived from bulk PAN using the same carbonization experimental conditions (2.42·10-4 S/cm). This behaviour could be explained based on the different morphology showed by these materials as well as the different amount of inorganic residue associated with them. Electrical conductivity of the carbon nanotubules is low inversely correlates with the pore size of the alumina template. For example, carbon nanotubules grown in 100 nm pore size alumina membranes show values of 1.65·10-7 S/cm and carbon nanotubules grown in 200 nm pore size ones show values of 1.8·10-6 S/cm (Table 1). After removing the templates from the carbon-inorganic systems, nanostructured carbons with different morphologies were obtained. TEM and SEM images show that all the nanostructured carbons prepared in this study maintain the morphology defined by the template (Fig. (6)). Thus, nanostructured carbons derived from carbon-imogolite nanocomposites exhibit a fibrous morphology but less defined than the one corresponding to the pristine imogolite (Fig. (6A)). Nevertheless, nanostructured carbons derived from carbon-sepiolite samples present a fibrous morphology with nominal dimensions in the order of 20-30 nm diameters by 1 @m long (Fig. (6B)). In the case of nanostructured carbons grown on alumina membranes, SEM images clearly show the tubular morphology of the resulting materials (Fig. (6C-D)). Depending on the pore size of the pristine alumina membranes, carbon nanotubules with different external and internal diameter can be obtained. In this way, carbon nanotubules grown in the 100 nm pore diameter alumina membranes have an external diameter of 180-245 nm and an internal one of 80-105 nm, while carbon nano-

Fernández-Saavedra et al.

Fig. (6). TEM micrographs of (A) templated carbon derived from PANimogolite, (B) templated carbon derived from PAN-sepiolite and SEM micrographs of carbon nanotubules derived from alumina membranes of (C) 100 nm and (D) 200 nm pore diameter.

tubules grown in the 200 nm pore diameter alumina membranes have an external diameter of 250-285 nm and an internal one of 155-175 nm, according to the SEM images. From these data, it can be deduced that the external diameter does not correspond to the pore diameter of the original alumina membranes. This feature could be explained by considering that a certain amount of alumina did not dissolve and remains associated with the resulting carbon nanotubules, increasing their external diameter. This hypothesis agrees with the elemental chemical analyses, which show that carbon nanotubules contain about 55-60 % of carbon, hydrogen and nitrogen (Table 1), so the ca. 40 % remainder could be ascribed to the oxygen derived from PAN and the alumina associated with the carbon nanotubules. XRD diagrams of nanostructured carbons formed in clays present a broad peak at ca. 0.36 nm, corresponding to the (002) reflection of poorly crystallized graphite, similar to that reported elsewhere for different carbon precursors [57, 58]. These templated carbons exhibit a similar carbon content of about 66 % (Table 1), which means that about 10% silicate is incorporated into the framework of these solids even after treating the samples with acids (HF, HCl) twice. Duclaux et al. [17] suggest that this fact could be explained because the formed carbon is oxidized by the neighbor silicate during the pyrolysis treatment. However, it is difficult to understand how silica could act as a real oxidizing agent at these low temperatures, which are far from those typically reported (around 1400ºC) [59]. On the other hand, due to the presence of nitrogen in the carbonaceous precursor (PAN), carbonaceous materials exhibit a noteworthy nitrogen content of ca. 14 % (Table 1). Nanostructured carbons derived from imogolite exhibit surface area values of 484 m2/g, which are remarkably higher than those associated with the carbonaceous material derived from bulk PAN (53 m2/g) (Table 1). However, nanostructured carbons derived from sepiolite present a surface area value of 33 m2/g, which is lower than that corresponding to the carbon derived from bulk PAN (Table 1). Disordered carbons are capable of uptaking much more Li than graphite, due to several mechanisms explained elsewhere [60, 61].

Template Synthesis of Nanostructured Carbonaceous Materials

Current Nanoscience, 2009, Vol. 5, No. 4

511

derived from sepiolite. Several factors can be responsible for this difference, for example, a higher surface area (484 m2/g), microporosity (0.191 cm3/g), and electrical conductivity (5.43·10-2 S/cm) (Table 1). Reversible specific capacity values for the templated carbonaceous materials are remarkably improved with respect to the carbon derived from bulk PAN obtained under the same experimental conditions (ca. 371 mAh/g) (Fig. (7B)). The use of nanoporous inorganic solids as templates in the preparation of the carbons appears to impart improved microtextural and electrical properties that could also be of great interest for the development of supercapacitors among other electrochemical devices.

Fig. (7). Charge-discharge cycles of (A) carbon nanotubules derived from alumina membranes of 100 nm (—) and 200 nm (---) pore diameter and (B) nanostructured carbon from carbon-imogolite (—), nanostructured carbon from carbon-sepiolite (---) and carbon derived from bulk PAN (…).

That is why intensive research is focused on the development of new strategies to prepare new nanostructured carbonaceous materials that could substitute graphite in Li-ion batteries. The Li insertion capacity of the nanostructured carbons prepared was tested. From the charge-discharge cycles (Fig. (7)) it is deduced that all the materials obtained in this work are able to insert Li in a reversible way. Overall, galvanostatic experiments carried out at higher insertion rates give rise to smaller specific capacity values than those run at slower rates. Beside, a loss of capacity close to 50 % is observed between the first and the second cycle in all the materials (Fig. (7). This behavior is attributed to the development of a SEI film (solid electrolyte interphase) on the surface of the carbonaceous solids during the first discharge, which is associated with the electrolyte decomposition and the formation of lithium organic compounds [62, 63]. The specific capacity is almost constant in successive cycles (< 10). All first charge curves present a similar behavior, that is, a rapid drop of the potential to near 0.8 V vs. Li where SEI layer formation occurs, followed by a gradual decrease of the potential to 0 V vs. Li (Fig. (7)). This behavior is ascribed to Li insertion into a disordered structure providing electronically and geometrically non-equivalent sites [64]. Moreover, charge-discharge cycles exhibit polarization, i.e., Li uptake occurs at potentials lower than 0.25 V vs. Li, whereas Li deinsertion occurs at much more positive potentials (> 1.5 V). As suggested by different authors, this last behavior can be attributed to the presence of nitrogen and hydrogen in the carbonaceous structure [60, 65]. Carbon nanotubules grown in porous alumina membranes present higher reversible specific capacities than nanostructured carbons derived from clays. Furthermore, such values are directly related to the pore size of the alumina membrane, i.e., carbon nanotubules grown in 100 nm pore size alumina membranes have a specific capacity of 838 mAh/g, while carbon nanotubules grown in 200 nm pore size ones show a value of 537 mAh/g, both measured at a 20 mA/g constant rate (Fig. (7A)). On the other hand, nanostructured carbon derived from imogolite shows higher reversible Li insertion capacity than those

The nanostructured carbonaceous materials have been tested as electrodes for electrochemical double-layer capacitors (EDLC). For this purpose, supercapacitor cells were built using carbons as the electroactive materials and an organic electrolyte. Capacitance values were deduced by applying three different electrochemical techniques: CV, galvanostatic charge-discharge, and EIS (Table 2). Pseudorectangular shape exhibited by the cyclic voltamperometry curves (Fig. (8A)), lineal profile observed in the galvanostatic charge-discharge curves (Fig. (8B)), and an almost vertical dependence of the imaginary part vs. the real impedance part in the Nyquist plot obtained by EIS (Fig. (8C)), prove the faradic behavior of these systems. In general, specific capacitance values measured with different electrochemical techniques exhibit similar values (Table 2). In this case, carbon nanotubules grown in porous alumina membranes also present higher capacitance values than those associated with nanostructured carbons derived from clays. Moreover, binder-free electrochemical cells were made with these carbon nanotubules, therefore enhancing their specific capacitance values. Thus, nanostructured carbonaceous materials derived from imogolite and sepiolite showed capacitance values of 22.4 F/g and 50.2 F/g, respectively, while carbon nanotubules derived from porous alumina membranes exhibited specific capacitance values of 62.7 F/g (100 nm pore size alumina membranes) and 46.2 F/g (200 nm pore size ones) (Table 2). In addition, the carbonaceous material derived from bulk PAN presented a very low value of specific capacitance (2.1 F/g), thus confirming the advantage of using inorganic templates for the preparation of this type of compounds. 4. CONCLUSIONS Novel nanostructured carbons with different morphologies and properties (surface area, electrical conductivity) have been prepared from PAN as the carbon precursor within different porous inorganic templates (imogolite, sepiolite, and porous alumina membranes). The synthesis procedures employed intermediate polymer-template systems. PAN-clay nanocomposites have been successfully prepared by the in situ polymerization of AN molecules previously inserted in the pores of the clays. PAN-alumina systems have been obtained by direct insertion of PAN within the pores of the alumina membranes. After thermal treatment, nanostructured carbons showing variable microstructural and electrical conductivity properties have been obtained. These materials exhibited high capacities for Li insertion, as well as high specific capacitance values when used as electroactive materials in EDLC capacitors. In summary, this study reveals that the use of different inorganic solids as templates is the key for the preparation of nanostructured carbons with improved microstructural and electrochemical properties useful in electrochemical devices such as rechargeable Li-ion batteries and EDLC capacitors. The methodology here applied can be extended to other nanoporous solids for a tailoring-made preparation of conducting carbonaceous materials with predetermined morphology, which can drive in a further step to multifunctional materials based on the intrinsic properties of the inorganic solid used as template [66]. Furthermore, opportunities exist to employ more ecologicallyfriendly carbon precursors such as sucrose [67].

512 Current Nanoscience, 2009, Vol. 5, No. 4

Fernández-Saavedra et al.

Table 2. Specific Electrode Capacitance Values of the Carbonaceous Materials Measured by Cyclic Voltamperometry, Galvanostatic ChargeDischarge and Electrochemical Impedance Spectroscopy Techniques

Specific Electrode Capacitance (F/g) Sample

Carbon derived from bulk PAN Nanostructured carbon from carbon-imogolite Nanostructured carbon from carbon-sepiolite Carbon nanotubules derived from alumina membrane (poro=100 nm ) Carbon nanotubules derived from alumina membrane (poro=200 nm)

CV

Galvanostatic Charge-Discharge

EIS

Average Value

2.3

2.6

1.3

2.1

9.4

11.5

12.0

11.0

51.6

50.4

48.7

50.2

62.2

61.8

64.2

62.7

47.5

52.1

39.0

46.2

Fig. (8). (A) Cyclic voltammogram, (B) charge discharge curves and (C) Nyquist plot (EIS) of a supercapacitor cell having as electrode material the templated carbon derived from carbon-imogolite nanocomposite and using 1 M LiPF6 in EC/DMC (50 wt %) as electrolyte.

ACKNOWLEDGEMENTS

[7]

This work was supported by the CICYT (Spain; project MAT2006-03356 and MAT2009- ) and by the Comunidad de Madrid (Spain; project S-0505/MAT/000227). R.F.-S. acknowledges a fellowship from the CICYT. We also acknowledge Dr. M.A. Martín-Luengo and Mr. T. García for specific surface area measurements, Mr. F. Pinto and Ms. S. Paniagua for SEM and TEM images acquisition and F. Fernandes for figure 1 development. This work, including use of the Advanced Photon Source at Argonne National Laboratory, was also supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.

[8] [9]

[10]

[11]

[12] [13]

REFERENCES [1] [2] [3]

[4] [5]

[6]

Ruiz-Hitzky, E. In Functional Hybrid Materials; Gómez-Romero, P., Sánchez, C. Eds.; Wiley-VCH: Weinheim, 2004; Vol. 2, pp. 15-49. Gómez-Romero, P. Hybrid organic-inorganic materials-in search of synergic activity. Adv. Mater, 2001, 13, 163-174. Rausell-Colom, J.A.; Serratosa, J.M. In Chemistry of Clays and Clay Minerals; Newman, A.C.D., Ed. Wiley-Interscience: New York, 1987; Vol. 8, pp. 371-422. Theng, B.K.G. Clay-polymer interactions; summary and perspectives. Clays Clay Miner., 1982, 30, 1-10. Ruiz-Hitzky, E.; Aranda, P.; Serratosa, J.M. In Handbook of Layered Materials; Auerbach, S.M., Carrado, K.A., Dutta, P., Eds.; Marcel-Dekker Inc.: New York, 2004; Vol. 3, pp. 91-154. Ruiz-Hitzky, E.; Van Meerbeeck A. In Handbook of Clay Science. Developments in Clay Science, Chp. 10.3, Bergaya, F., Theng, B.K.G., Lagaly, G., Eds.; Elsevier: London, 2006; Vol. 1, pp. 583-621.

[14]

[15] [16]

[17] [18]

[19]

Ruiz-Hitzky, E. Conducting polymers intercalated in layered solids. Adv. Mater., 1993, 5, 334-340. Ruiz-Hitzky, E.; Aranda, P. Confinement of conducting polymers into inorganic solids. An. Quim. Int. Ed., 1997, 93, 197-212. Ruiz-Hitzky, E., Aranda, P. In Polymer-Clay Nanocomposites; Pinnavaia, T.J., Beall, G.W., Eds.; John Wiley & Sons: West Sussex, 2000; Vol. 2, pp. 19-46. Aranda, P.; Darder, M.; Fernández-Saavedra, R.; López-Blanco, M.; RuizHitzky, E. Relevance of polymer-and biopolymer-clay nanocomposites in electrochemical and electroanalytical applications. Thin Solid Films, 2006, 495, 104-112. Aranda, P. In Clay Minerals Society Workshop Lectures Series, 15: ClayBased Polymer Nanocomposites; Carrado, K.A., Bergaya F., Eds.; The Clay Minerals Society: Chantilly, 2007; Vol. 6, pp. 171-196. Thomas, A.; Goettmann, F.; Antonietti, M. Hard templates for soft materials: creating nanostructured organic materials. Chem. Mater., 2008, 20, 738-755. Sonobe, N.; Kyotani, T.; Tomita, A. Carbonization of polyacrylonitrile in a two-dimensional space between montmorillonite lamellae. Carbon, 1988, 26, 573-578. Sonobe, N.; Kyotani, T.; Tomita, A. Carbonization of polyfurfuryl alcohol and polyvinyl acetate between the lamellae of montmorillonite. Carbon, 1990, 28, 483-488. Sandí, G.; Winans, R.E.; Carrado, K.A. New carbon electrodes for secondary lithium batteries. J. Electrochem. Soc., 1996, 143, L95-L98. Sandí, G.; Carrado, K.A.; Winans, R.E.; Johnson, C.S.; Csencsits, R. Carbons for lithium applications prepared using sepiolite as an inorganic template. J. Electrochem. Soc., 1999, 146, 3644-3648. Duclaux, L.; Frackowiak, E.; Béguin, F.J. Novel carbons from nanocomposites for high lithium storage. Power Sources, 1999, 81-82, 323-327. Duclaux, L.; Frackowiak, E.; Gibinski, T.; Benoit, R.; Béguin, F. Clay/carbon nanocomposites as precursors of electrode materials for lithiumion batteries and supercapacitors. Mol. Cryst. Lyq. Cryst., 2000, 340, 449454. Fernández-Saavedra, R.; Aranda, P.; Ruiz-Hitzky, E. Templated synthesis of carbon nanofibers from polyacrylonitrile using sepiolite. Adv. Funct. Mater.,

Template Synthesis of Nanostructured Carbonaceous Materials

[20] [21] [22]

[23]

[24]

[25] [26] [27] [28]

[29]

[30]

[31]

[32] [33] [34]

[35] [36] [37]

[38] [39] [40]

[41]

[42] [43]

[44]

Current Nanoscience, 2009, Vol. 5, No. 4

2004, 14, 77-82. Enzel, P.; Bein, T. Poly(acrylonitrile) chains in zeolite channelspolymerization and pyrolysis. Chem. Mater., 1992, 4, 819-824. Wu, C.G.; Bein, T. Conducting carbon wires in ordered, nanometer-sized channels. Science, 1994, 266, 1013-1015. Kyotani, T.; Nagai, T.; Inoue, S.; Tomita, A. Formation of new type of porous carbon by carbonization in zeolite nanochannels. Chem. Mater., 1997, 9, 609-615. Lu, A.; Kiefer, A.; Schmidt, W.; Schüth, F. Synthesis of polyacrylonitrilebased ordered mesoporous carbon with tunable pore structures. Chem. Mater., 2004, 16, 100-103. Jang, J.; Lim, B.; Choi, M. A simple synthesis of mesoporous carbons with tunable mesoporous using a colloidal template-mediated vapor deposition polymerization. Chem. Commun., 2005, 33, 4214-4216. Hernández-Vélez, M. Nanowires and 1 D arrays fabrication: an overview. Thin Solid Films, 2006, 495, 51-63. Martin, C.R.; Van Dyke, L.S.; Cai, Z.; Liang, W.J. Template synthesis of organic microtubules. Am. Chem. Soc., 1990, 112, 8976-8977. Martin, C.R. Template synthesis of polymeric and metal microtubules. Adv. Mater., 1991, 3, 457-459. Che, G.; Lakshmi, B.B.; Martin, C.R.; Fisher, E.R. Metal-nanocluster-filled carbon nanotubes: catalytic properties and possible applications in electrochemical energy storage and production. Langmuir, 1999, 15, 750-758. Wang, H.; Zhang, L.; Gavalas, G.R. Preparation of supported carbon membranes from furfuryl alcohol by vapor deposition polymerization. J. Membr. Sci., 2000, 177, 25-31. Lee, J.S.; Gu, G.H.; Kim, H.; Jeong, K.S.; Bae, J.; Suh, J.S. Growth of carbon nanotubes on anodic aluminum oxide templates: fabrication of a tube-intube and linearly joined tube. Chem. Mater., 2001, 13, 2387-2391. Koloczek, J.; Brodka, A.; Burian, A.; Dore, J.C.; Honkimaki, V.; Kyotani, T. Structural studies of carbon nanotubes obtained by template deposition using high-energy X-ray scattering. Diamond Relat. Mater., 2006, 15, 1036-1040. Schneider, J. J.; Engstler, J. Carbon and polymer filaments in nanoporous alumina. Eur. J. Inorg. Chem., 2006, 9, 1723-1736. Parthasarathy, R.V.; Phani K.L.N., Martin C.R. Template synthesis of graphitic nanotubules. Adv. Mater., 1995, 7, 896-897. Jung, M.; Kim, H.G.; Lee, J.K.; Joo, O.S.; Mho, S. EDLC characteristics of CNTs grown on nanoporous alumina templates. Electrochim. Acta, 2004, 50, 857-862. Frackowiak, E.; Jurewicz, K.; Delpeux, S.; Béguin, F. Nanotubular materials for supercapacitors. J. Power Sources, 2001, 97-98, 822-825. Frackowiak, E.; Béguin, F. Electrochemical storage of energy in carbon nanotubes and nanostructured carbons. Carbon, 2002, 40, 1775-1787. Ruiz-Hitzky, E.; Fripiat, J.J. Organomineral derivatives obtained by reacting organochlorosilanes with the surface of silicates in organic solvents. Clays Clay Miner, 1976, 24, 25-30. Ruiz-Hitzky, E. Molecular access to intracrystalline tunnels of sepiolite. J. Mater. Chem., 2001, 11, 86-91. Santarén, J.; Sanz, J.; Ruiz-Hitzky, E. Structural fluorine in sepiolite. Clay Clays Miner, 1990, 38, 63-68. Guggenheim, S.; Adams, J. M.; Bain, D.C.; Bergaya, F.; Brigatti, M.F.; Drits, V.A.; Formoso, M.L.L.; Galan, E.; Kogure, T.; Stanjek, H. Summary of recommendations of nomenclature committees relevant to clay mineralogy: report of the Association Internationale Pour l’Etude des Argiles (AIPEA) Nomenclature Committee for 2006. Clay Miner., 2006, 41, 863877. Cradwick, P.D.; Wada, K.; Rusell, J.D.; Yoshinaga, N.; Masson, C.R.; Farmer, V.C. Imogolite, a hydrated aluminum silicate of tubular structure. Nat. Phys. Sci., 1972, 240, 187-189. Adams, M. J. Gas chromatographic adsorption studies on synthetic imogolite. J. Chromatogr., 1980, 188, 97-106. Ackerman, W.C.; Smith, D.M.; Huling, J.C.; Kim, Y.W.; Bailey, J.K.; Brinker, C.J. Gas/vapor adsorption in imogolite: a microporous tubular aluminosilicate, Langmuir, 1993, 9, 1051-1057. Imamura, S.; Hayashi, Y.; Kajiwara, K.; Hoshino, H.; Kaito, C. Imogolite: a

Received: March 09, 2009

[45] [46] [47] [48]

[49]

[50]

[51]

[52]

[53] [54]

[55]

[56]

[57]

[58]

[59]

[60] [61] [62] [63] [64]

[65] [66]

[67]

Revised: April 14, 2009

513

possible new type of shape-selective catalyst. Ind. Eng. Chem. Res., 1993, 32, 600-603. Wilson, M.A.; Lee, G.S.H.; Taylor, R.C. Benzene displacement on imogolite. Clays Clay Miner., 2002, 50, 348-351. Koenderink, G.H.; Kluijtmans, S.G.J.M.; Philipse, A.P. On the synthesis of colloidal imogolite fibers. J. Colloid Interface Sci., 1999, 216, 429-431. Kurt, A.O.; Davies, T.J. Sepiolite-PAN intercalation used as Si3N4 forming precursor. J. Mater. Sci., 2001, 36, 957-962. Fernández-Saavedra, R.; Darder, M.; Gómez-Avilés, A.; Aranda, P.; RuizHitzky, E. Polymer-clay nanocomposites as precursors of nanostructured carbon materials for electrochemical devices: templating effect of clays. J. Nanosci. Nanotech., 2008, 8, 1741-1750. Sánchez-Soto, P.J.; Avilés, M.A.; del Río, J.C.; Ginés, J.M.; Pascual, J.; Pérez-Rodríguez, J.L. Thermal study of the effect of several solvents on polymerization of acrylonitrile and their subsequent pyrolysis. J. Anal. Appl. Pyrolysis, 2001, 58-59, 155-172. Kuang, W.; Facey, G.A.; Detellier, C.; Casal, B.; Serratosa, J.M.; RuizHitzky, E. Nanostructured hybrid materials formed by sequestration of pyridine molecules in the tunnels os sepiolite. Chem. Mater., 2003, 15, 49564967. Sandí, G.; Joachin, H.; Lub, W.Q.; Prakash, J.; Tassara, G. Comparison of the electrochemichal performance of carbon produced from sepiolite with different surface characteristics. J. New Mater. Electrochem. Syst., 2003, 6, 75-80. Sandí, G.; Winans, R.E.; Seifert, S.; Carrado, K.A. In situ SAXS studies of the structural changes of sepiolite clay and sepiolite-carbon composites with temperature. Chem. Mater., 2002, 14, 739-742. Bailey, J.E.; Clarke, A.J. Carbon-fibre formation-the oxidation treatment. Nature, 1971, 234, 529-531. Zhu, Y.; Wilding, M..; Mukhopadhyay, S.K. Estimation, using infrared spectroscopy, of the cyclization of poly(acrylonitrile) during the stabilization stage of carbon fibre production. J. Mater. Sci., 1996, 31, 3831-3837. Saufi, S.M.; Ismail, A.F. Development and characterization of polyacrylonitrile (PAN) based carbon hollow fiber membrane. Songklanakarin J. Sci. Technol., 2002, 24, 843-854. Lai, S.H.; Chen, Y.L.; Chan, L.H.; Pan, Y.M.; Liu, X.W.; Shih, H.C. The crystalline properties of carbon nitride nanotubes synthesized by electron cyclotron resonance plasma. Thin Solid Films, 2003, 444, 38-43. Sonobe, N.; Kyotani, T.; Tomita, A. Carbonization of polyfurfuryl alcohol and polyvinyl acetate between the lamellae of montmorillonite. Carbon, 1990, 28, 483-488. Jung, Y.; Suh, M.C.; Lee, H.; Kim, M.; Lee, S.; Shim, S.C.; Kwak, J. Electrochemical insertion of lithium into polyacrylonitrile-based disordered carbons. J. Electrochem. Soc., 1997, 144, 4279-4284. Sahajwalla, V.; Wu, C.; Khanna, R.; Chaudhury, N.S.; Spink, J. Kinetic study of factors affecting in situ reduction of silica in carbon–silica mixtures for refractories. ISIJ Int., 2003, 43, 1309-1315. Dahn, J.R.; Zheng, T.; Liu, Y.; Xue, J.S. Mechanism for lithium insertion in carbonaceous materials. Science, 1995, 270, 590-593. Winter, M.; Besenhard, J.O.; Spahr, M.E.; Novák, P. Insertion electrode materials for rechargeable lithium batteries. Adv. Mater., 1998, 10, 725-763. Yazami, R. Surface chemistry and lithium storage capability of the graphitelithium electrode. Electrochim. Acta, 1999, 45, 87-97. Zane, D.; Antonini, A.; Pasqualini, M. A morphological study of SEI film on graphite electrodes. J. Power Sources, 2001, 97-98, 146-150. Winter, M.; Besenhard, J.O. In Lithium Ion Batteries: Fundamentals and Performance; Wakihara, M., Yamamoto, O., Eds; Wiley-VCH, Weinheim/Kodansha Ltd.: Tokio, 1998; Vol. 6, pp. 127-155. Wu, Y.; Fang, S.; Jiang, Y. Effects of nitrogen on the carbon anode of a lithium secondary battery. Solid State Ionics, 1999, 120, 117-123. Gómez-Avilés, A.; Darder, M.; Aranda, P.; Ruiz-Hitzky, E. Functionalized carbon-silicates from caramel-sepiolite nanocomposites. Angew. Chem. Int. Ed., 2007, 46, 923-925. Darder, M.; Ruiz-Hitzky, E. Caramel-clay nanocomposites. J. Mater. Chem., 2005, 35, 3913-3918.

Accepted: July 21, 2009