当前位置：文档库 › 1-s2.0-S0008622306006099-main

1-s2.0-S0008622306006099-main

Synthesis of carbon–Fe

O coaxial nano?bres by pyrolysis

of ferrocene in Fangyu Cao a ,Changle Chen

a,*

Hefei National Laboratory for Physical Sciences at Microscale and Department of Materials Science and Engineering,

University of Science and Technology of China,Hefei 230026,China

Department of Chemistry,The University of Chicago,5735,South Ellis Avenue,Chicago,IL 60637,USA

Received 29May 2006;accepted 29November 2006

Available online 11January 2007

Abstract

Carbon–Fe 3O 4coaxial nano?bres with a diameter around 100nm were synthesized by pyrolysis of ferrocene in supercritical CO 2at 400°C.The single crystal Fe 3O 4cores of the coaxial nano?bres are continuous,and the carbon shell is rich in H.The M–H hysteresis loop for the nanorodes measured at room temperature shows a saturation magnetization of 27.5emu/g,much lower than 92emu/g of the corresponding bulk material.This is attributed the considerable mass of the carbon shell and the nanoscale and anisotropy of the mag-netite nanorods.A possible mechanism was proposed.ó2006Elsevier Ltd.All rights reserved.

1.Introduction

Ferrimagnetic oxides,especially those with nano size,have attracted considerable interests in magnetic informa-tion storage,ferro?uid,various biomedical medium,heter-ogeneous catalysts,active electrode material for lithium batteries and electrochemical capacitors because of the magnetic,structural,and transport properties,while its application always encounters the problem of structural variation due to erosion in air,especially when its size reaches nanoscale range.One solution is to cover the mag-netite nanoparticals or nanorods by inertial material such as carbon nanotubes (CNTs)giving a capsulate structure.Various materials including magnetic metals and iron oxi-des encapsulated in carbon nanotubes [1–5]have been pre-pared for drug delivery,magnetic data storage,toners and inks for xerography and magnetic resonance imaging [6,7].Ferrocene can serve as carbon and iron sources,whose pyrolysis gives CNTs and iron sources [3–5].However,high reaction temperature,the toxicity and cost of raw

materials or solvents prevent these methods from being widely used,and the loading of prefab CNTs is often dis-continuous.It is the purpose of this research to carry out a simple and secure way to produce well continuous car-bon-magnet coaxial nano?bre materials by a one-pot method.

Supercritical carbon dioxide (Sc–CO 2)has emerged as an important solvent in industrial applications because it is inexpensive,non?ammable,nontoxic,and environmen-tally friendly [8].Recently,many carbon-based materials such as cubic diamond,carbon nanotube,nanosheet and nanosphere were synthesized by reduction of Sc–CO 2with alkali metals [9–11].Herein,we report a one-pot preparing of a novel structure of carbon–Fe 3O 4coaxial nano?bres by pyrolysis of ferrocene in Sc–CO 2at 400°C.

2.Experimental

Ferrocene (98%),CO 2(99.9%),toluene,ethanol and hydrochloric acid (analytically pure)were used as purchased.The reaction was carried out by heating 2.0g of ferrocene and 8.0g of dry ice in a 20ml steel autoclave at 400°C for 800min.After cooling down to room temperature naturally and venting the remnant CO 2,the dark brown spongy product was washed with toluene and ethanol respectively,and dried at 110°C for

Corresponding author.Fax:+865513631760.E-mail address:cqw@https://www.wendangku.net/doc/1115992069.html, (Q.Chen).

https://www.wendangku.net/doc/1115992069.html,/locate/carbon

Carbon 45(2007)727–731

hours.Part of this sample was treated with hot diluted hydrochloric acid to obtain FeO x-free product.Some control reactions were also carried out. Samples used are summarized in Table1.If it is not speci?cally mentioned, the sample used is CM400L1.

The powder X-ray di?raction(XRD)analyses were performed on a Rigaku D/MAX-c A X-ray di?ractometer equipped with Cu K a radiation (k=1.542A?)over the2h range of10–70°.X-ray electron spectroscopy (XPS)was performed on a VG ESCALAB MKII Electron Spectrometer with the X-ray source of Mg K a.Transmission electron microscopy (TEM)analyses were performed on a Hitachi H-800transmission electron microscope with electron di?raction and the accelerating potential is 200kV.For TEM observations,the products were separated in ethanol by ultrasonic dispersion then transferred onto carbon-coated copper grids. Field emission scanning electron microanalysis(FESEM)was taken on a JEOL JSM-6700F Field emission scanning electron microanalyser.High-resolution transmission electron microscope(HRTEM)images were taken on JEOL-2010with an accelerating voltage of200kV.Elemental analysis was carried out on an Elementar VARIO ELIII Elemental Analyzer, giving information of C:H ratio in the samples.Fourier transform infrared spectrum(FT-IR)of the samples was recorded at the ambient temperature on a Nicolet MAGNA-IR750Fourier Transform Infrared Spectrometer with the resolution factor of±0.1cmà1.Raman spectrum was recorded on a JY LABRAM-HR Confocal Laser Micro Raman spectrometer with an argon-ion laser at the excitation wavelength of514.5nm and the reso-lution factor of±1cmà1.Magnetic measurements were carried out on a Riken BHV-55Vibrating Sample Magnetometer(VSM).

3.Results and discussion

Fig.1a shows XRD patterns of the sample,which was indexed to a mixture of amorphous carbon(the broad peak

with2h value of around23.0°)and iron oxides(JCPDS card85-1436,magnetite).It is hard to identify the iron oxi-des as Fe3O4or c-Fe2O3(JCPDS cards39-1346,maghe-mite)because of their similarity in XRD patterns.The peaks at16.8°and34.3°(labeled by‘‘*’’)in Fig.1a were indexed as hydrous carbonate or basic carbonate of iron, considering its chemical properties shown in our experi-ment.After removing iron oxide,only a broad peak around23.0°was found(CT400L1,Fig.1b),away from the graphite(111)di?raction(JCPDS card75–2078, 26.6°),revealing the amorphous property of these carbon materials.The XRD pattern of CM500(Fig.1c)obtained at500°C,shows an increase in intensity of the broad peak and shift towards26.6°,showing the improvement in crys-tallinity of carbon under higher temperatures.

From the distribution of the iron oxide and carbon in the sample indicated by XPS(Fig.2),combined with images of electron microscopy,the core-shell structure was indicated.The average diameter of the nanorods is about100nm,and that of iron oxide core is about40nm (Fig.3).This kind of FeO x core is grown continuously and uprightly along an orientation as a single crystal.After removing the iron oxide,hollow nanotubes were observed in sample CT400L1,shown in Fig.3b.A broad ring exhib-its in the electron di?raction pattern of the sample(Fig.3b, insert),further supporting the amorphous nature of the carbon shell.HRTEM image(Fig.3h)shows the bubble like amorphous carbon shell and lattice pattern of iron oxide core with growth orientation of[110].As shown in FESEM image(Fig.3g),the nanorods are produced with high purity,uniformity and considerable length.In com-parison,only agglomerated encapsulated nanoparticles were produced by pyrolysis of ferrocene without Sc–CO2 as solvent(Fig.3f),showing that the homogeneous Sc–CO2is one of the necessary conditions for the formation of carbon–Fe3O4coaxial nano?bres.

Table1

Synthesis conditions of the samples

Name of the sample Reactant and

amount

Reaction conditions Puri?ed

by HCl Ferrocene/

CO2/

Temperature/

°C

Time/

minute

CM400L1 2.008.0400800No

CM400S1 2.008.0400200No

CM400L20.108.0400800No

CT400L1 2.008.0400800Yes

CM500 2.008.0500800No

728 F.Cao et al./Carbon45(2007)727–731

FT-IR spectrum of (Fig.4)reveals the coexistence of sp 2and sp 3hybridized C–C,C @C and C–H bonds,in which 2925,2857,and 1447cm à1were assigned to hydrogen-bonded sp 3carbon,and 1630,1100,820,and 515cm à1to C @C and C @C–H bonds.Raman spectra were shown in Fig.4,insert.The only iron oxide identi?ed was hematite (peaks at 220cm à1,280cm à1and 390cm à1),which might come from magnetite or maghemite under laser exposure [12].Nasrazadani and Raman [13]show that magnetite exhibits two strong bands at around 570(m 1)and 390(m 2)cm à1,di?erent from the absorbing peaks of maghemite at 630and 430cm à1.On the basis of their experiments,the broad peak at around 580cm à1in Fig.4was assigned to magnetite.No visible peak was found at around 630cm à1,implying the small amount of maghemite.More convincing evidence is needed to con?rm this conclusion.The peak at around 1580cm à1involves the in-plane bond stretching motion of pairs of carbon sp 2atoms,which also exists as the G mode with the E 2g symmetry [14].The peak around 1430cm à1was assigned to sp 3-CH n [15].The D

mode

Fig.3.TEM,FESEM and HRTEM images of di?erent samples.TEM image of (a)CM400L1,(b)CT400L1,(c)CM500,(d)CM400L2,(e)CM400S1,(f)product of pyrolysis of ferrocene without addition of CO 2,(g)FESEM images of sample CM400L and (h)HRTEM image of sample CM400L1.

F.Cao et al./Carbon 45(2007)727–731729

of graphite around1350cmà1is related to the degree of disorder in carbon sp2bonded clusters in graphite crystal. The absence of this peak in Fig.4,insert(a)and its pres-ence in Fig.4,insert(b)imply better carbonization and dehydrogenization of carbon of CM500than that of CM400L1.Elemental analysis(not shown)shows the con-tent of element C and H is51.92wt%and3.039wt%in sample CM400L1and53.79%and 1.591%in CM500 respectively,with the rest mass as iron oxide.The C:H ratios(atom number)of sample CM400L1and CM500 are1.434and2.837,conforming that carbon in the samples is more hydrogenised under lower temperatures.Based on the data above,it is reasonable to conclude that the carbon shell is formed with a typical amorphous C:H structure [16].

M–H hysteresis loop(Fig.5)measured at room temper-ature shows that the powder sample exhibits ferrimagne-tism with saturation magnetization M s=27.5emu/g, remnant magnetization M r=12.7emu/g,and coercive ?eld H c=324.5Oe.The reasons M s is much lower than 92emu/g of the corresponding bulk magnetite[17],which could be due to the considerable mass of carbon in the sam-ple and anisotropy of magnetite nanorods[18].Magnetite nanorods show superparamagnetism when its diameter is near or below the critical size(54nm)of single domains. The high shape anisotropy of the nanorods prevents them from magnetizing in directions other than along easy mag-netic axes.With nanorods randomly oriented,the projec-tion of the magnetization vectors along the?eld direction will be lower than that of nanoparticles without the large shape anisotropy e?ect.

The well continuous growth of iron oxide nanorod core, the encapsulated nanoparticles adhere to nanotubes at low ferrocene concentration,and the high H capacity of the carbon shell provide some clues for the growth mechanism of the coaxial nanostructures.Because ferrocene is well soluble in Sc–CO2[19],it is hypothesized that ferrocene is evenly dispersed by Sc–CO2under reaction conditions. At a temperature lower than that of pyrolysis,the low valence iron cations in ferrocene were slowly oxidized by CO2.Our former research shows that,the trace amount of hydrogen in hot Sc–CO2system?nally changes into long-chain-alkanes and polycyclic aromatic hydrocar-bons.[20]Also,Fe2+could be oxidized to Fe3+by CO2, [21]and Fe3O4is more stable than Fe2O3at the reaction temperature.Based on these,it is inferred that,in this experiment,the CO2molecules react with ferrocene,substi-tuting the cyclopentadienyl-group and oxidizing Fe2+.A complex reaction happens among the CO2ligand,the cen-tral Fe2+cations,and the cyclopentadienyl anions,produc-ing Fe3O4,amorphous carbon and hydrocarbon:

FeCp

tCO2!

400 C Fe

O4ta-CtC n H m

The concentrated iron oxide was encapsulated by amor-phous carbon,forming capsulated nanoparticles.With the dispersion of Sc–CO2as solvent,if the concentration of fer-rocene is high enough,with the cyclopentadienyl anions as carbon feed stock and oxidized iron as the iron oxide source,the partially capsulated nanoparticles grew longer to form the carbon–Fe3O4nanorods(CM400S1,Fig.3e), which was irregular at?rst.Kept at400°C for a longer time,the composites would grow longer and more regular, forming a novel cable-like structure of carbon–Fe3O4.In contrast,if the concentration of ferrocene is relatively low,a?ected by surface and interface energy,iron oxide nanoparticles formed at?rst tended to be fully capsulated by carbon shell already before the beginning of growth, which isolated the magnetite particles from iron oxide source and stopped them from growing longer and forming nanorods.Some nanoparticles that were not fully encapsu-lated grew longer,but the growth was far from continuous because of the lacking of iron oxide source.Thus,capsu-

730 F.Cao et al./Carbon45(2007)727–731

lated Fe3O4nanoparticles adhered onto carbon nanotube and encapsulated discontinuous Fe3O4nanorod segments can be found in corresponding sample(CM400L2, Fig.3d).The adherence may be caused by magnetic inter-action between the nanoparticles and the nanorods.

Because the decomposition temperature of ferrocene is between400and500°C,it is possible that the mechanism of the growth of carbon nanotubes at higher temperature is di?erent.In sample CM500(Fig.3c),only hollow nano-tubes and nanotube bundles without iron oxide core were https://www.wendangku.net/doc/1115992069.html,bined with XRD,Raman spectroscopy and Elemental analysis mentioned above,it can be found that sample CM500present a distinctly di?erent feature form and more carbonized and dehydrogenized than sample CM400L1.When the temperature is high enough,ferrocene was pyrolyzed into small hydrocarbon(C n H m)and iron particles[4]?rst,and the iron particles were soon oxidized to iron oxide particles in the atmosphere of Sc–CO2.Cata-lyzed by them[22],carbon nanotubes could be formed.

Since the starting materials are facile and cheap,and the process is simple,it is envisioned that this approach will be of interest to various?eld.The carbon shell can protect magnetite from been oxidized or eroded and strengthen the fragile nanorods,whose magnetic properties may lead to application as magnetic information storage and probe of magnetic force microscopy.

4.Conclusions

In this supercritical-?uid-based approach,carbon–Fe3O4coaxial nano?bres were produced by pyrolysis of ferrocene in Sc–CO2at about400°C.The carbon nano-tubes are amorphous and hydrogenated,and the Fe3O4 core is continuous as a single crystal.A lower temperature than the decomposition temperature of ferrocene is essen-tial for growing the cable-like structure.This kind of struc-ture is expected to prevent Fe3O4nanorods from being oxidized in external atmospheres and keep their magnetic performance unchanged.The carbon–Fe3O4coaxial nano-?bres can be produced in high yield and potentially used as magnetic information storage,probe of magnetic force microscopy,etc.The detailed study in growth mechanism and application is under progress.

Acknowledgement

This work was supported by the Natural Science Foun-dation of China(90206034).

References

[1]Bao JC,Tie CY,Xu Z,Suo ZY,Zhou QF,Hong JM.A facile method

for creating an array of metal-?lled carbon nanotubes.Adv Mater 2002;14(20):1483–6.

[2]Jiang LQ,Gao L.Carbon nanotubes-magnetite nanocomposites from

solvothermal processes:formation,characterization,and enhanced electrical properties.Chem Mater2003;15(14):2848–53.

[3]Hou HQ,Schaper AK,Weller F,Greiner A.Carbon nanotubes and

spheres produced by modi?ed ferrocene pyrolysis.Chem Mater 2002;14(9):3990–4.

[4]Lu Y,Zhu ZP,Liu ZY.Carbon-encapsulated Fe nanoparticles from

detonation-induced pyrolysis of ferrocene.Carbon2005;43(2): 369–74.

[5]Xu LQ,Zhang WQ,Ding YW,Peng YY,Zhang SY,Yu WC,et al.

Formation,characterization,and magnetic properties of Fe3O4 nanowires encapsulated in carbon microtubes.J Phys Chem B 2004;108(30):10859–62.

[6]Bianco A,Prato M.Can carbon nanotubes be considered useful tools

for biological applications?Adv Mater2003;15(20):1765–8.

[7]Majetich SA,Artman JO,McHenry ME,Nuhfer NT,Staley SW.

Preparation and properties of carbon-coated magnetic nanocrystal-lites.Phys Rev B1993;48(22):16845–8.

[8]Kendall JL,Canelas DA,Young JL,DeSimone JM.Polymerizations

in supercritical carbon dioxide.Chem Rev1999;99(2):543–63.

[9]Lou ZS,Chen QW,Zhang YF,Wang W,Qian YT.Diamond

formation by reduction of carbon dioxide at low temperatures.J Am Chem Soc2003;125(31):9302–3.

[10]Lou ZS,Chen QW,Wang W,Zhang YF.Synthesis of carbon

nanotubes by reduction of carbon dioxide with metallic lithium.

Carbon2003;41(15):3063–7.

[11]Lou ZS,Chen QW,Gao J,Zhang YF.Preparation of carbon spheres

consisting of amorphous carbon cores and graphene shells.Carbon 2004;42(1):229–32.

[12]de Faria DLA,Silva SV,de Oliveira MT.Raman microspectroscopy

of some iron oxides and oxyhydroxides.J Raman Spectrosc 1997;28(11):873–8.

[13]Nasrazadani S,Raman A.The application of infrared-spectroscopy

to the study of rust system-II.Study of carbon de?ciency in magnetite (Fe3O4)produced during its transformation to maghemite(c-Fe2O3) and hematite(a-Fe2O3).Corros Sci1993;34(8):1355–65.

[14]Ferrari AC,Robertson J.Interpretation of Raman spectra of

disordered and amorphous carbon.Phys Rev B2000;61(20): 14095–107.

[15]Ristein J,Stief RT,Ley L,Beyer W.A comparative analysis of a-C:

H by infrared spectroscopy and mass selected thermal e?usion.J Appl

Phys1998;84(7):3836–47.

[16]Casiraghi C,Ferrari AC,Robertson J.Raman spectroscopy of

hydrogenated amorphous carbons.Phys Rev B2005;72(8).Art.No.

085401.

[17]Wang J,Peng ZM,Huang YJ,Chen QW.Growth of magnetite

nanorods along its easy-magnetization axis of[110].J Cryst Growth 2004;263(1–4):616–9.

[18]Wang J,Chen QW,Zeng C,Hou BY.Magnetic-?eld-induced growth

of single-crystalline Fe3O4nanowires.Adv Mater2004;16(2): 137–40.

[19]Cowey CM,Bartle KD,Burford MD,Cli?ord AA,Zhu S.Solubility

of ferrocene and a nickel-complex in supercritical?uids.J Chem Eng Data1995;40(6):1217–21.

[20]Wang Q,Cao FY,Chen QW.Ocurrence of long-chain-alkanes and

polycyclic aromatic hydrocarbons(PAHs)in the thermal system of carbonate and sodium.Lett Org Chem2005;2(1):136–8.

[21]Chen QW,Bahnemann DW.Reduction of carbon dioxide by

magnetite:implications for primordial synthesis of organic molecules.

J Am Chem Soc2000;122(5):970–1.

[22]Choi HC,Kundaria S,Wang DW,Javey A,Wang Q,Rolandi M,

et al.E?cient formation of iron nanoparticle catalysts on silicon oxide by hydroxylamine for carbon nanotube synthesis and electron-ics.Nano Lett2003;3(2):157–61.

F.Cao et al./Carbon45(2007)727–731731