Synthesis and Characterization of Nanostructured Cobalt

J. Phys. Chem. C 2010, 114, 6401–6407

6401

Synthesis and Characterization of Nanostructured Cobalt Hexacyanoferrate Mario Berrettoni,*,† Marco Giorgetti,† Silvia Zamponi,‡ Paolo Conti,‡ David Ranganathan,‡ Antonio Zanotto,§ Maria Luisa Saladino,§ and Eugenio Caponetti§ Dipartimento di Chimica Fisica ed Inorganica, UniVersita` di Bologna e Unita` di Ricerca INSTM di Bologna, Viale del Risorgimento 4, 40136 Bologna, Italy, Dipartimento Scienze Chimiche, UniVersita` di Camerino, Via S. Agostino 1s62032 Camerino, Italy, and Dipartimento di Chimica Fisica “F. Accascina”, UniVersita` di Palermo, Parco d’Orleans II, Viale delle Scienze pad.17 e Unita` di Ricerca INSTM di Palermo, Palermo 90128, Italy ReceiVed: January 14, 2010; ReVised Manuscript ReceiVed: February 22, 2010

Cobalt hexacyanoferrate (CoHCF) nanoparticles have been synthesized by mixing aqueous solutions of K3Fe(CN)6 and CoCl2 under vigorous stirring at different temperatures and following two different procedures, drop-by-drop or immediate mixing. The resulting CoHCF nanoparticles, with dimensions of several tens of nanometers, were characterized using TEM, SEM-EDX, IR, and XRD. Their electrochemical behavior was investigated in comparison with the CoHCF powder bulk compound. The CoHCF nanoparticles exhibit an electrochemically driven conversion to the bulk one that has been investigated by a chemometric approach in order to establish the best synthetic parameters. The rate and the degree of conversion depend on the synthesis temperature. Introduction In the last years, the need for materials with tunable characteristics has addressed the research of the synthesis of nanostructured compounds whose properties depend on the morphology, size, chemical composition, and crystal structure.1 Nanoparticles with a defined geometry present some advantages in comparison to bulk materials due to their high surface-tovolume ratio.2 Among the nanomaterials that show interesting properties, Prussian blue analogues (PBAs) have been synthesized for application in nanomagnetic, biosensing, biomedical, and electrochromic devices.3 The physicochemical properties of hexacyanometalate-like materials are related to the structure, namely, the possibility to be flexible and modulated by a change in the local environment driven by an external input. PBAs are usually described having the general formula XkM’ [M”(CN)6]mH2O in which M’ and M” are both transition metals in the high spin and the low spin configuration, respectively, and X is an inserted cation necessary to ensure the neutrality.4 Ludi,5 in his pioneering study, explained the structure of PBAs by XRD as a face-centered cubic lattice in which the high-spin metal and the low-spin metal sites are both octrahedral and surrounded by -NC and -CN units, with the potassium as counterions X and water molecules that occupy the interstitial sites. A distortion to a rhombohedral lattice can be observed when the structure is stressed, that is, in the electrochemical reduction process.6 These PBA nanoparticles have been prepared by different techniques in sol-gel or using surfactants and capping agents, which stabilize the nanoparticles, such as sodium bis(2ethylhexil)sulfosuccinate, sodium hexametaphosphate, apoferritin, and stearylamine, or polymers, such as poly(vinylpirrolydone), or in the presence of mesostructured silica, porous alumina, and nafion, as reported in ref 3 and references therein. * To whom correspondence should be addressed. E-mail: mario.berrettoni@ unibo.it. † Universita` di Bologna e Unita` di Ricerca INSTM di Bologna. ‡ Universita` di Camerino. § Universita` di Palermo e Unita` di Ricerca INSTM di Palermo.

Cobalt hexacyanoferrate (CoHCF), among the PBAs, is one of the more interesting inorganic polymers and shows switching properties caused by the CoII(HS)-FeIII T CoIII(LS)-FeII (in the above expression, HS and LS stand for high and low spin) electron transfer accompanied by the spin state change of the Co ion.7 This material is very attractive because of the peculiar physicochemicalproperties:electrochromism,8,9 thermochromism,8-10 photochemical magnetism,7 and electrocatalytic and sensing properties.11 The redox process in CoHCF involves two different metal transition ions (Fe and Co), unlike other materials of the same class; the phenomenon is driven by a metal-to-metal charge transfer. This electronic change occurs with a significant shortening of the Co-N bond length from 2.08 Å (CoII(HS)) to 1.91 Å (CoIII(LS))12 and with a significant shortening of the cell parameters from 10.30 to 9.96 Å, when the electronic switch is spread in a cooperative way in the solid. The contemporary occurrence of FeIII and CoIII makes it possible to observe both the redox couples, FeIII/FeII and CoIII/CoII, in the same material.13 Many papers report on the preparation of CoHCF nanoparticles by electrochemical modification of the electrode surface, such as carbon nanotubes11 and nanowires;14 by synthesis in reversed micelles15 or water-in-oil microemulsions;16 and on the formation of the CoHCF-silica nanocomposites.17 Recently, our group synthesized CoHCF in a PAMAM-doped silica matrix,18,19 showing a direct influence of the silica network on the physicochemical properties. Zanotto et al. proposed a synthetic route of CoHCF nanoparticles embedded in poly(methyl methacrylate) (CoHCFPMMA), obtaining a 50 nm average particle size, often arranged in 300 nm aggregates, where apolar matrix properties reduce the reversibility of water-assisted redox processes.20 On the contrary, only few papers report on the direct synthesis of PBA nanoparticles without any templating and/or added substances.21 In this paper, we present a study on CoHCF nanoparticles, synthesized directly in solution without any added surfactant. The characterization was performed by using cyclic voltammetry, IR spectroscopy, XRD, SEM-EDX, TEM, and

10.1021/jp100367p  2010 American Chemical Society Published on Web 03/11/2010

6402

J. Phys. Chem. C, Vol. 114, No. 14, 2010

XRF techniques. Continuously repeated cyclic voltammograms were analyzed through multivariate curve resolution alternate least squares (MCR-ALS) to disclose the structure modifications driven by the electrochemical redox process. The application of chemometric deconvolution methods to the voltammetric electroanalytical data was widely reviewed by Esteban and others.22 These methods, usually, work under the hypothesis that the measured current is linear with respect to species concentration, which is not always true in electrochemical data. However the deviation from the linearity is not so big to impede the use of the multivariate curve resolution method on voltammetric data to study complex equilibrium.23 The MCR application in several fields is widely reviewed;24-26 we showed its utility in voltabsorptometry.27 The MCR-ALS acts on a decomposed matrix where the number of species F is well-known; in the case of evolving systems, this information is unknown and has to be extracted from the data in the case of evolving systems.

Berrettoni et al. experiments. ATR spectra of the samples, prepared by evaporation of the synthetic solution on graphite foil (see Electrode Preparation below), were recorded with a PerkinElmer Spectrum 100 Series FT-IR spectrometer. XRD measurements were performed on a Rigaku Mo KR X-ray source equipped with a Huber goniometer. TEM experiments were performed with a Philips CM10 microscope and XRF with a Shimadzu 800HS. Electrode Preparation. Samples were investigated with a glassy carbon electrode. The glassy carbon (0.7 mm diameter) working electrode was polished with 0.05 µm alumina before each experiment and carefully rinsed in an ultrasonic bath with distilled water. The glassy carbon surface was cleaned following the procedure reported in ref 29 to enhance the reproducibility of the measurements. After the cleaning procedure, 10 µL of CoHCF synthetic solution was dropped on the electrode surface and let to dry at room temperature for 0.5 h. A water dispersion of the W sample, prepared at the moment, was used in place of the synthetic solution in the case of the bulk compound. This procedure allowed reproducible results to be obtained.

Experimental Section Chemicals and Solutions. Potassium hexacyanoferrate K3[Fe(CN)6] (Aldrich, 99%), potassium chloride KCl (Aldrich, 99%), and cobalt chloride hexahydrate CoCl2 · 6H2O (Fluka, 98.5%) were all used as received. All solutions were prepared with Millipore Milli-Q nanopure water with a resistivity of ≈17 MΩ cm. CoHCF Nanoparticle Synthesis and Characterization. CoHCF nanoparticles were obtained by precipitation using two different procedures: drop-by-drop (DbD) or immediate mixing (IM) at four different fixed temperatures, namely, 0, 5, 20, and 40 °C (batches A, C, E, and G and B, D, F, and H, respectively, for DbD and IM). For the drop-by-drop procedure, 10 mL of an aqueous solution of 0.01 M K3Fe(CN)6 at RT was taken in a dropping funnel and the solution was added drop by drop to 10 mL of an aqueous solution of 0.02 M CoCl2, which was thermostatted at the fixed temperature under stirring. The color changed from red to brown, and the solution became turbid. For the immediate mixing procedure, 10 mL of an aqueous solution of 0.01 M K3Fe(CN)6 at the fixed temperature was immediately added to 10 mL of an aqueous solution of 0.02 M CoCl2, which was at the same temperature. The resulting solution was stirred and maintained at the fixed temperature for 1 h. The color changed from red to brown, and the solution became turbid.28 A reference bulk compound was prepared at room temperature by dropping an equivalent amount of a K3[Fe(CN)6] 0.01 M solution in a CoCl2 0.02 M solution, obtaining CoHCF with a composition similar to that of the CoHCF nanoparticles. The resulting dark red suspension was filtered and carefully washed with deionized water to remove the initial unreacted salts; then, it was let to dry at room temperature. Hereafter, we refer to this CoHCF sample as (W). All the A-H samples were left in the synthetic solution and stored at room temperature before the experiments. Methods and Instruments. Cyclic voltammetry (CV) was performed with a CHI model 630 Electrochemical Workstation (CH Instruments, Austin, TX, U.S.A.) in a standard threeelectrode cell. A glassy carbon (GC) was the working electrode, a Pt wire was the counter electrode, and an Ag/AgCl saturated (KCl) acted as the reference electrode. All potentials were expressed versus Ag/AgCl saturated (KCl) reference electrode. All solutions were deoxygenated by bubbling nitrogen for 15 min and maintained under a nitrogen atmosphere during the

Results and Discussion Sample Characterization. All the investigated samples were characterized by IR spectroscopy, X-ray diffraction, and transmission electron microscopy. Their elemental formulas were determined by the XRF analysis. The synoptic Table 1 reports the main features of the samples. TEM images showed aggregates of particles with dimensions of around tens of nanometers for all A-H compounds; Figure 1 reports a TEM image of sample B. SEM-EDX analyses have confirmed the simultaneous presence of Fe and Co in the particles. Raw formulas are reported regardless of the potassium stoichiometric coefficient because of the poor sensitivity of the techniques used versus this element. The accurate quantification of the potassium content was prevented because the XRF analyses were executed on liquid samples that precluded the use of vacuum. The XRF analyses were performed on solid pellet under vacuum only in the case of the W sample. The calculated stoichiometric formula of the W sample agrees very well with that of the insoluble form of CoHCF, generally accepted in literature. Nevertheless, in all the other compounds, the Co/Fe ratio is still in agreement with the optimal one. These values still fit with the insoluble form of CoHCF, with the additional presence of Fe(CN)6 vacancies.30 The XRD spectrum of the bulk sample (W) shows the typical peaks for the CoHCF fcc cubic system with a cell parameter a ) 10.298 Å. All the spectra of the other samples presented the splits of reflections (220), (420), (440), and (620) with an intensity ratio of 1:1, indicating a rhombohedral distortion of the fcc lattice.6 All the evaluated cell parameters are ranging from 10.288 to 10.547 Å. Table 1 also reports the characteristic stretching frequencies of the CN group and their assignments based on the data available in the literature.13 All the IR measurements were performed in ATR mode taking care to record several spectra, moving the tip, in order to be sure that the spectra are not due to a local effect. An IR spectrum representative of each sample is reported in Figure 2. All spectra are characterized by the presence of three bands, more or less pronounced, at approximately 2085, 2120, and 2158 cm-1. The band at 2158 cm-1 can be attributed to the CN stretching in the chain FeIII-CN-CoII, the band at 2120 cm-1 to the chain FeII-CN-CoIII, and the band at 2085 cm-1 to the chain FeII-CN-CoII.13,31 It is noteworthy to highlight the contemporary presence of FeIII and CoIII species in all samples. Both the

Synthesis and Characterization of Nanostructured CoHCF

J. Phys. Chem. C, Vol. 114, No. 14, 2010 6403

TABLE 1: Synoptic Information for All Investigated Compounds sample ID

a

T/°C

synthesis

a/Å

elemental analysis

A

0

DbD

10.401

K1.26Co1.76Fe(CN)6

B

0

IM

10.321

K0.32Co1.61Fe(CN)6

C

5

DbD

10.416

K0.00Co1.91Fe(CN)6

D

5

IM

10.547

K0.59Co1.43Fe(CN)6

E

20

DbD

10.068

K0.54Co1.53Fe(CN)6

F

20

IM

10.355

K1.08Co1.68Fe(CN)6

G

40

DbD

10.288

K0.00Co2.09Fe(CN)6

H

40

IM

10.335

K1.07Co1.73Fe(CN)6

W

RT

bulka

10.298

K0.08Co1.44Fe(CN)6

strongest IR frequencies

band attribution

2124 2159 2122 2159 2090 2119 2159 2089 2121 2158 2084 2123 2158 2084 2121 2157 2083 2123 2161 2158 2122 2083 2102 2158

FeII-CN-CoIII FeIII-CN-CoII FeII-CN-CoIII FeIII-CN-CoII FeII-CN-CoII FeII-CN-CoIII FeIII-CN-CoII FeII-CN-CoII FeII-CN-CoIII FeIII-CN-CoII FeII-CN-CoII FeII-CN-CoIII FeIII-CN-CoII FeII-CN-CoII FeII-CN-CoIII FeIII-CN-CoII FeII-CN-CoII FeII-CN-CoIII FeIII-CN-CoII FeIII-CN-CoII FeII-CN-CoIII FeII-CN-CoII FeII-CN-CoIII FeIII-CN-CoII

See text for details.

bands due to the FeIII-CN-CoII and FeII-CN-CoIII are present in all samples, but, as expected, the relative percentage cannot be evaluated due to the intrinsic limitation of the IR technique used. Voltammetric Behavior. The electrochemical behavior of CoHCF nanoparticles was characterized by CV. Figure 3 shows the CV of sample E in comparison with the CV of compound W. A similar behavior for all the nanosamples (A-H) was observed, as reported in Figure 4, for CVs recorded at 50 mV/ s. The small differences in the composition, namely, the Co/Fe ratio, do not influence the electrochemical response. The current intensity of sample E is, at least, 15 times greater than the one of the “bulk” sample, even if the two electrodes have been loaded, roughly, with the same amount of compound.11 The CV of sample E was characterized by a reversible redox process (ired/iox) at Epc ) 0.40 V and Epa ) 0.54 V, with a ratio of Ipc/Ipa very close to the unity. This behavior was already observed for CoHCF nanoparticles supported on carbon nanotubes.11 The electrochemical process associated with sample W

was more complex than that of sample E involving several redox couples. The main processes, labeled as II and I in Figure 3, are assigned, in the literature, to the reduction of FeIII(II) and CoIII (I).21,13 Further investigations on the dependence of the peak parameters, namely, the peak current (Ip) and charge (Q), associated with the redox process, were performed to better understand the involved electrochemical processes. The intensity of peaks II and I (bulk) is linearly dependent on the square root of the scan rate (ν1/2) if the scan rate is greater than 20 mV/s, whereas it is linearly dependent on the scan rate (ν) when the scan rate is below that value, indicating a mixed regime of diffusion with a surface confined process at a very low scan rate. The intensity of peak i in the nanocompound is linear with the square root of ν in the whole range of explored scan rates (0.5-500 mV s-1). The charge Q associated with each redox peak increases as the scan rate decreases.

Figure 1. TEM image of compound B.

Figure 2. Representative spectra of all investigated samples.

6404


Berrettoni et al.

Figure 3. Comparison of the cyclic voltammetry of compounds E (dotted line) and W (solid line). Scan rate ) 0.05 V/s. Figure 5. Selected sweeps of continuously repeated cyclic voltammograms (rate ) 0.05 V/s) on compound A.

Figure 4. Cyclic voltammetries normalized to the maximum cathodic current. Scan rate ) 0.05 V/s.

The overall reduction process can be written as

K[CoFe]V(CN)6 + e- + K+ S K2[CoFe]IV(CN)6

(1) regardless of which metal atom is in the oxidized form. The notation (CoFe)ox number used in eq 1 represents the actual situation of the cobalt hexacyanoferrate compounds, in which both CoIII and FeIII metal ions with a different ratio can coexist, depending on the synthetic route. In this class of compounds, the FeIII center is always electrochemically active, while the redox process related to the couple CoIII/CoII becomes allowed only for compounds crystallized in the so-called insoluble form. The insoluble form presents defects caused by the absence of 1 /4 of the Fe(CN)6 groups, replaced by water molecules, which allows the swelling of the structure caused by the contraction of the Co-N bond length while reducing to CoII. Real samples are usually constituted by a solid solution in which the insoluble and soluble forms coexist, giving rise to CVs that show a more complex morphology due to both the redox processes of FeIII/ FeII and CoIII/CoII couples.21 From the kinetic point of view, the limiting process usually is the intercalation/deintercalation

of the K+ ions necessary for the charge balance of the compounds. This leads to a process limited by the diffusion of K+ ions inside the structure with a linear dependence of Ip versus scan rate, at least for very low scan rates, as actually observed in the case of the “bulk” species. It can be assumed that the nanoparticles are stacked in several layers so that a very large surface with respect to their volumes is exposed to the electrolyte. This large electrochemically active surface is responsible for the most part of the current. This finding was also confirmed by the current intensity value that is larger for nano than for bulk samples. In this case, the kinetic limiting factor becomes the availability of K+ ions at the active electrode surface that has to be fed by diffusion from the bulk of the solution. This mechanistic explanation conforms to the linear dependence of Ip versus ν1/2, actually observed for samples A-H. Under this hypothesis, only elementary cells on the particle surface are involved in the redox process. The boundary surface can be considered rich in defects, resulting in the higher concentration of CoIII with respect to FeIII; this condition can lead to a CV that shows only one redox couple at the potential attributed to the reduction of CoIII. These findings do not contradict the contemporary presence of two bands in the IR spectra because the IR readout takes into account some layers in the bulk of the particle where the iron(III) is prevalent. Some cycles showing the behavior of the representative sample A, selected from a continuously cycling experiment, are reported in Figure 5. During the CV scans, a progressive modification of the curve morphology has been observed. This finding can be ascribed to the conversion of the starting nanocompound to a bulklike species, as can be inferred from Figure 3. The electrochemical processes can be schematically written

{K[CoFe]V(CN)6}nano + e- + K+ S K2[CoFe]IV(CN)6

referred to as R + e ) γ

(2a)

Synthesis and Characterization of Nanostructured CoHCF

J. Phys. Chem. C, Vol. 114, No. 14, 2010 6405

{K[CoFe]V(CN)6}bulk + e- + K+ S IV

K2[CoFe] (CN)6

referred to as β + e ) γ

(2b)

Kc

RSβ

(3)

The conversion process schematically described in eq 3, takes place with an isopotential point at about 0.5 V. The redox process for either A and B species are described by eq 1 and

hence are both monoelectronic. In such a case, the presence of an isopotential point ensures us on the electrode stability. The presence of an isopotential point in a repeated cyclic voltammogram experiment was studied by Fitch et al.32 when a chemical cross-reaction was present between two electrochemical processes. In our case, the cross-reaction is a progressive substitution during the oxidation process of some CoII by FeII. With reference to eq 1, it can be hypothesized that CoIII/CoII is mainly the starting electroactive couple. The continuous reduction/oxidation cycles continuously change the relative ratio

Figure 6. MCR-ALS of samples B and G. Panels a and a’ show the raw experimental CVs of samples B and G, respectively. Panels b and b’ show the computed concentration evolution of B and G, respectively. The comparison between some experimental baseline-subtracted sweeps and the corresponding computed ones are reported in panels c and c’.

6406


Berrettoni et al.

TABLE 2: Main Features of the Samples and Results Obtained with MCR-ALS Analysis

sample name

synthesis temperature °C

synthesis type

no. cycle at 50% of conversion

A C E G B D F H

0 5 20 40 0 5 20 40

DD DD DD DD IM IM IM IM

135 144 31 32 113 126 37 54

% composition after 500 cycles nano

bulk

40 42 0 0 35 30 0 0

60 58 100 100 65 70 100 100

between the CoIII/CoII and FeIII/FeII redox couples, increasing the percentage of the FeIII/FeII couple available for the next voltammogram. In the long run, the CV morphology approaches the form of the bulk sample. In such a case, the ratio of conversion of the electroactive couples can be assimilated to a chemical cross-reaction. The existence of the isopotential point shown in Figure 5 also ensures us that there is no leakage of active compound during the experiment. The current I, at a well-defined potential, for a mixture of two electroactive species (R and β), is the sum of the current due to the species R and β. Named IR and Iβ, respectively, the current due to R and β, we can write I ) CRIR + CβIβ that can be written in a matrix form as XnV,nE ) CnV,FVF,nE in which XnV,nE is the matrix of nV voltammograms (one per row) sampled at nE (columns in X) potentials. The matrix CnV,F is a matrix of concentrations of every column of which store the concentration values of one of the F species for every nV voltammograms; VF,nE have in each row a voltammogram characteristic of a species. The two matrices can be unraveled with a chemometric approach. In this study, the MCR-ALS was performed on a matrix of linear sweep voltammograms (the cathodic sweeps of cyclic voltammetric experiments) that were iterated to study the evolution of the electrochemical system. The XnV,nE matrix consists of baseline-subtracted voltammograms. The considered baseline is a straight line obtained by fitting, with the leastsquares method, the first and last five points of the voltammogram under consideration. Fixed size window evolving factor analysis (FSWEFA)33 was used to point out the number of species, F, and their range of existence in the system. FSWEFA is a one-step method based on the PCA decomposition of a fixed size moving window matrix. The algorithm of the data treatment is based on the method developed by Gampp et al.34 and optimized by some of us for the spectroelectrochemistry.27 The above method was applied to all A-H samples. Figure 6 reports the typical procedure followed in the elaboration of the data acquired for all samples and shows the results for samples B and G. Some selected continuously repeated cyclic voltammograms (every ten cycles) are plotted in panels a and a’. The evaluated concentration profiles as a function of cycle number are plotted in panels b and b’, and the comparison of the experimental and the calculated cyclic voltammograms are plotted in panels c and c’. Two species only were found in each analyzed sample, regardless of the synthetic procedure and temperature. On the contrary, the rate and completeness of the conversion, reported in Table 2, seem to depend from the synthetic temperature. Looking at the half conversion cycle shown in Table 2, it can

be hypothesized that both synthetic ways produce similar compounds, whereas the temperature at which the synthesis is done seems to influence the stability of the compound. It was observed that, at low temperatures, the conversion of the compounds needs a larger number of cycles to occur, whereas the samples synthesized at room or higher temperatures rearrange rapidly and completely. Conclusion The proposed synthetic routes for CoHCF seem to generate compounds in the nano size range that show physicochemical properties very similar to those of the bulk one. In particular, the IR spectra show a set of three bands, with different intensities, assigned to the CN stretching related to the chain Co-NC-Fe in which Co and Fe can be in both II or III oxidation states. The XRD spectra present the characteristic reflections for the CoHCF compound. In turn, the cyclic voltammograms of the nanocompounds present only one reversible redox process centered at 0.45 V, whereas the corresponding bulk compound presents a more complex reduction process involving two species that are attributed to the FeIII/FeII and CoIII/CoII couples, as usually reported in the literature. In the nanocompounds, the only electroactive species is the CoIII/CoII couple. The continuously repeated CVs on each nanocompound show a progressive transition to the characteristic morphology of the bulk compound. This feature is strongly dependent on the synthetic temperature. The compounds synthesized at 20 and 40 °C show a conversion rate greater than those synthesized at 0 and 5 °C. In the investigated time scale, only the compounds synthesized at high temperatures undergo a complete conversion to the bulk one. In this work, it was demonstrated that the synthetic pathway used to produce nanosized CoHCF particles is reliable, fast, and cheap. The nanoparticles left in the synthetic solution are stable for a long time. The electrochemical behavior of nanocompounds converts to that of the bulk when an external electrochemical driving force is applied. The rate and the degree of conversion depend on the synthesis temperature. Further studies are needed to clarify the role of temperature in the synthetic pathway. We then suggest the use of the notation [CoFe]ox number to indicate the mixed valence state in cobalt iron cyanide;35 this finds application also in some spin crossover based on Co/Fe cyanides due to charge-transfer phenomena.7,12 The notation represents the actual situation of the cobalt hexacyanoferrate compounds where both CoIII and FeIII metal ions coexist with a different ratio, depending on several factors. Acknowledgment. M.B. and M.G. acknowledge the Rimini branch of the University of Bologna for financial support. E.C., M.L.S., and A.Z. would like to thank the MIUR for supporting this research through the PRIN 2007 prot. 20077R3PXF_002 “New nanocomposites preparation for optical, electric and magnetic applications”. Supporting Information Available: Selected XRD patterns and TEM photographs. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Shi, Y.; Zhou, B.; Wu, P.; Wang, K.; Cai, C. J. Electroanal. Chem. 2007, 611, 1–9. (2) Baioni, A. P.; Vidotti, M.; Fiorito, P. A.; Ponzio, E. A.; Cordoba de Torresi, S. I. Langmuir 2007, 23, 6796–6800. (3) Vo, V.; Van Minh, N.; Lee, H. I.; Kim, J. M.; Kim, Y.; Kim, S. J. Mater. Res. Bull. 2009, 44, 78–81.

Synthesis and Characterization of Nanostructured CoHCF (4) Gimenez-Romero, D.; Agrisuelas, J.; Garcia-Jareno, J. J.; Gregori, J.; Gabrielli, C.; Perrot, H.; Vicente, F. J. Am. Chem. Soc. 2007, 129, 7121– 7126. (5) Ludi, A.; Gudel, H. U. Struct. Bonding (Berlin, Ger.) 1973, 14, 1–21. (6) Kaplun, M. M.; Smirnov, Yu. E.; Mikli, V.; Malev, V. V. Russ. J. Electrochem. 2001, 37, 914–923. (7) Sato, O.; Hayami, S.; Einaga, Y.; Gu, Z. Bull. Chem. Soc. Jpn. 2003, 76, 443–470. (8) Kulesza, P. J.; Malik, M. A.; Zamponi, S.; Berrettoni, M.; Marassi, R. J. Electroanal. Chem. 1995, 397, 287–292. (9) Kulesza, P. J.; Malik, M. A.; Miekzmikowski, K.; Wolkiewicz, A.; Zamponi, S.; Berrettoni, M.; Marassi, R. J. Electrochem. Soc. 1996, 143, L10–L12. (10) Kulesza, P. J.; Malik, M. A.; Berrettoni, M.; Giorgetti, M.; Zamponi, S.; Schmidt, R.; Marassi, R. J. Phys. Chem. B 1998, 102, 1870–1876. (11) Yang, M.; Jiang, J.; Yang, Y.; Chen, X.; Shen, G.; Yu, R. Biosens. Bioelectron. 2006, 21, 1791–1797. (12) Escax, V.; Bleuzen, A.; Cartier dit Moulin, C.; Villain, F.; Goujon, A.; Varret, F.; Verdaguer, M. J. Am. Chem. Soc. 2001, 123, 12536–12543. (13) Lezna, R. O.; Romagnoli, R.; de Tacconi, N. R.; Rajeshwar, K. J. Phys. Chem. B 2002, 106, 3612–3621. (14) Qian, L.; Yanf, X. Talanta 2006, 69, 957–962. (15) Yamada, M.; Sato, T.; Miyake, M.; Kobayashi, Y. J. Colloid Interface Sci. 2007, 315, 369–375. (16) Vaucher, S.; Fielden, J.; Li, M.; Dujardin, E.; Mann, S. Nano Lett. 2002, 3, 225–229. (17) Moore, J. G.; Lochner, E. J.; Ramsey, C.; Dalal, N. S.; Stiegman, A. E. Angew. Chem., Int. Ed. 2003, 42, 2741–2743. (18) Zamponi, S.; Giorgetti, M.; Berrettoni, M.; Kulesza, P. J.; Cox, J. A.; Kijak, A. Electrochim. Acta 2005, 51, 118–124. (19) Giorgetti, M.; Berrettoni, M.; Zamponi, S.; Kulesza, P. J.; Cox, J. A. Electrochim. Acta 2005, 51, 511–516.

J. Phys. Chem. C, Vol. 114, No. 14, 2010 6407 (20) Zanotto, A.; Matassa, R.; Saladino, M. L.; Berrettoni, M.; Giorgetti, M.; Zamponi, S.; Caponetti, E. Mater. Chem. Phys. 2010, 120, 118-122. (21) Shi, L.; Wu, T.; Wang, M.; Li, D.; Zhang, Y.; Li, J. Chin. J. Chem. 2005, 23, 149–154. (22) Esteban, M.; Arinõ, C.; Dıáz-Cruz, J. M. TrAC, Trends Anal. Chem. 2006, 25, 86–92. (23) Fernańdez, M.; Arinõ, C.; Dıáz-Cruz, J. M.; Tauler, R.; Esteban, M. J. Electroanal. Chem. 2001, 505, 44–53. (24) De Juan, A.; Tauler, R. Crit. ReV. Anal. Chem. 2006, 36, 163–176. (25) Martıńez Galera, M.; Gil Garcıá, M. D.; Goicoechea, H. C. TrAC, Trends Anal. Chem. 2007, 26, 1032–1042. (26) Garrido, M.; Rius, F. X.; Larrechi, M. S. Anal. Bioanal. Chem. 2008, 390, 2059–2066. (27) Marassi, R.; Zamponi, S.; Conti, P.; Lanteri, S. Ann. Chim. 2002, 92, 261–270. (28) Berrettoni, M.; Giorgetti, M.; Magnani, A.; Zamponi, S.; Conti, P.; Ranganathan, D.; Caponetti, E.; Zanotto, A.; Saladino, M. L. Proceedings of the International Meeting about Nanotechnology for Industry - Nanotech 2008, Venezia, Italy, 2008. (29) Kiema, G. K.; Aktay, M.; McDermott, M. T. J. Electroanal. Chem. 2003, 540, 7–15. (30) Giorgetti, M.; Berrettoni, M. Inorg. Chem. 2008, 47, 6001–6008. (31) Sato, O.; Einaga, Y.; Fujishima, A.; Hashimoto, K. Inorg. Chem. 1999, 38, 4405–4412. (32) Fitch, A.; Edens, G. J. J. Electroanal. Chem. 1989, 267, 1–13. (33) Keller, H. R.; Massart, D. L. Anal. Chim. Acta 1991, 246, 379– 390. (34) Gampp, H.; Maeder, M.; Meyer, C. J.; Zuberbuhler, A. D. Talanta 1986, 33, 943–951. (35) Martinez-Garcia, R.; Knobel, M.; Balmaseda, J.; Yee-Madeira, H.; Reguera, E. J. Phys. Chem. Solids 2007, 68, 290–298.

JP100367P