Ce

MD

a

ARR2AA

KCRCLA

1

bwcgAhidthaciia3i

h0

Journal of Analytical and Applied Pyrolysis 117 (2016) 317–324

Contents lists available at ScienceDirect

Journal of Analytical and Applied Pyrolysis

journa l h om epage: www.elsev ier .com/ locate / jaap

atalytic graphitization of resorcinol-formaldehyde xerogel and itsffect on lithium ion intercalation

anohar Kakunuri, Suresh Kali, Chandra S. Sharma ∗

epartment of Chemical Engineering, Indian Institute of Technology, Yeddumailaram, Hyderabad, Telangana502205, India

r t i c l e i n f o

rticle history:eceived 26 May 2015eceived in revised form8 September 2015ccepted 31 October 2015vailable online 7 November 2015

eywords:arbon xerogelesorcinol-formaldehyde

a b s t r a c t

Resorcinol formaldehyde (RF) derived xerogels were synthesized by polymerization of resorcinol andformaldehyde using sol–gel method followed by sub-critical drying. Thus prepared RF xerogel sampleswere then pyrolyzed at different temperatures ranging from 900 to 1800 ◦C with and without catalyst tounderstand the role of catalyst in graphitization while varying the temperature. These RF derived carbonxerogels were structurally characterized extensively by Raman spectroscopy, X-ray diffraction, high-resolution transmission electron microscopy and small angle X-ray scattering. Structural characterizationstudies confirmed that RF xerogels pyrolyzed at moderate temperatures in the presence of a catalyst weregraphitized however the degree of graphitization achieved was only 10% in case of no catalyst even atelevated temperatures (2500 ◦C). Later, these RF derived catalytically graphitized carbon xerogels were

atalytic graphitizationithium-ion batterynode material

tested for their electrochemical performance as potential anode materials for rechargeable batteries.Galvanostatic charge discharge experiments were performed at 0.1C rate (37.2 mA/g current density)to examine the effect of degree of graphitization on lithium ion intercalation. It was found that withincreased degree of graphitization, specific capacity increased with decrease in irreversible capacity alongwith excellent cyclic stability and coulombic efficiency more than 97%.

© 2015 Elsevier B.V. All rights reserved.

. Introduction

Carbon materials are widely used as anodes for lithium ionatteries due to their low cost and availability in various formsith reasonably good reversible lithium insertion/intercalation

apacity. They are mainly classified as graphite, hard carbon andraphitizable soft carbon based on their crystalline structure.mong them, graphite is a natural allotrope of carbon with sp2

ybridization (trivalent carbon atoms). Graphite is formed by stack-ng of graphene layers. These graphene layers consists of twoimensional network of regular hexagons with an interlayer dis-ance of 3.354 A◦ between these sheets. This interlayer distance isigher than carbon–carbon distance (1.42 A◦) of hexagonal ringss the sheets are weakly bonded in c-axis compared to strongarbon–carbon bonding in basal plane [1]. This interlayer bond-ng may further weaken thus losing the stacking and increasing thenterlayer distance further. Thus in disordered carbons like hard

nd soft carbons, the interatomic spacing varies from 3.354 A◦ to.7 A◦ while basal plane carbon–carbon bonding maintains same

nteratomic distance [1,2]. Hard carbons contains few graphene

∗ Corresponding author. Fax: +91 40 23016032.E-mail address: cssharma@iith.ac.in (C.S. Sharma).

ttp://dx.doi.org/10.1016/j.jaap.2015.10.017165-2370/© 2015 Elsevier B.V. All rights reserved.

layers arranged in a disordered manner like house of cards, andthey cannot be graphitized even by heating as high as 3000 ◦Cwhile soft carbons show slightly higher degree of order and thesecarbons can be graphitized on heating at elevated temperaturesover 2000 ◦C [3]. Among these different types of carbons, graphiteis most commercially used anode because of its flat charge dis-charge plateau below 0.2 V, good cyclic life and high coulombicefficiency [4]. Theoretical reversible capacity of perfectly struc-tured graphite is 372 mAh/g with LiC6 stoichiometry howevermost common graphitic materials that have been used extensivelyshowed the capacity in the range of 215–370 mAh/g with goodcoulombic efficiency [5–8]. On the other hand, hard carbons syn-thesized form pyrolysis of organic compounds like epoxy novolacresins, phenolic resins, synthetic isotropic pitches, polymers andoxidized pitches [9–12] show higher capacity (400–700 mAh/g)than graphitic materials but find limited use due to high irre-versible capacity that consumes more lithium in initial cycles[4,12].

Considering the limitations of both graphite as well as hardcarbons, efforts have been made in literature to tune the proper-

ties of graphite. Synthetic graphite can be prepared commerciallyfrom graphitizable carbons using resistive heating above 3000 ◦C[13]. However as compared to the cost of raw materials heatingat elevated temperatures take more share in process cost [13].

3 l and

Gpptcg[orl[F[

fpcmdeatcscrxmfmdt

Rsewcmhdopetgifib[

2

2

aeadwfcw

18 M. Kakunuri et al. / Journal of Analytica

raphitization occurs due to displacement of layer of graphiticlanes and grouping of such layers at elevated temperatures amongarallel crystallites [14,15]. To reduce the energy penalty, graphi-ization can be achieved at lower temperatures by pyrolyzingarbon under high pressure or by introducing a metal or inor-anic/organic additives to serve as a catalyst for graphitization16–18]. Later is also known as catalytic graphitization. A numberf carbon precursor materials such as polyacrylonitrile, phenolicesins, furan resin, organic gels in the form of fibers and mono-iths have been successfully graphitized in the presence of catalyst19–26]. From these studies, it was observed that use of Ni ande as metal catalyst was more effective during graphitization19,21–24].

Since its introduction by Pekala in 1989, [27] resorcinol-ormaldehyde (RF) based organic gels have been widely used asrecursor to carbon aerogels, cryogels and xerogels. RF derivedarbon aerogels and cryogels have been widely used as anodeaterials for energy storage devices, mainly for supercapacitors

ue to high capacitance and interfacial resistance [28,29]. How-ver the major limitation of RF derived aerogels and cryogels asnode materials for lithium ion battery is higher surface area andhus large irreversible capacity losses in addition to higher pro-essing cost in synthesis. Non-porous carbon xerogels possess lessurface area and hence shows comparatively less initial irreversibleapacity compared to carbon aerogels when used as anode mate-ial in lithium ion battery [30,31]. In recent times, RF derived carbonerogels have been successfully demonstrated as potential anodeaterials for lithium ion battery but reversible capacity reported

or powder sample was only 145 mAh/g, which was lower thanost of graphitic and hard carbon materials [31]. To improve this,

oping with nitrogen is suggested as one of the approach in one ofhe recent literature [32].

However in this work, we focus on catalytic graphitization ofF derived carbon xerogel as an alternative approach to tailor theirtructural characteristics. Role of pyrolysis temperature and pres-nce of catalyst on microstructures of RF derived carbon xerogelas studied in detail which further improved the electrochemi-

al performance of these materials for their potential use of anodeaterials in lithium ion battery. Although RF derived carbon gels

ave been widely tested as anode materials for energy storageevices, to the best of our knowledge, this is first detailed studyn the catalytic graphitization of RF derived carbon xeorgels andaving the way to enhance their electrochemical performance byngineering their microstructure and thus crystallinity. In litera-ure, there are only a few studies available on effect of catalyticraphitization on lithium ion intercalation moreover these stud-es are not only limited to PAN based electrospun fibers, carbonbers grown by chemical vapor deposition and phenolic resinut also pyrolysis temperature is limited to less than 1000 ◦C33–35].

. Experimental

.1. Materials

Resorcinol (99% purity), formaldehyde (37% w/v; stabilized withbout 10% methanol), potassium carbonate (98% purity), LP-30lectrolyte (1 M LiPF6 in 1:1 v/v mixture of ethylene carbonatend diethyl carbonate) were purchased from Merck, India and usedirectly. Lithium foil (99.9% trace metal basis) and Iron (II) acetate

as purchased from Sigma–Aldrich, India while stainless steel (SS)

oil (polycrystalline SS 316, 100 �m thick) used as a substrate forell testing was purchased from MTI corp., USA. Deionized wateras used as solvent while making RF aqueous sol.

Applied Pyrolysis 117 (2016) 317–324

2.2. Preparation of RF sol

To prepare the RF sol, resorcinol and formaldehyde were firstadded in a molar ratio of 0.25 (Resorcinol to formaldehyde) fol-lowed by mixing using magnetic stirrer for 15 min. Thus preparedsolution was then added to aqueous solution of potassium carbon-ate with the resorcinol-to-water and resorcinol-to-catalyst molarratios 0.037 and 25 respectively. The graphitizing catalyst iron (II)acetate (1% w/w) was then added to as prepared RF sol and stirredfor 15 min in an inert atmosphere inside the glove box. RF sol mixedwith catalyst was then air dried at room temperature for 24 h fol-lowed by oven drying at 80 ◦C to yield RF xerogel. The RF sol withoutcatalyst was also stirred for 15 min followed by similar drying pro-cedure as in previous case of with catalyst.

2.3. Pyrolysis

Dried RF xerogel blocks were pyrolyzed in graphitizing tubularfurnace (Materials Research Furnace, USA). After introducing thesample into the furnace tube, the tube was evacuated to 100 mTorr pressure before purging with argon (Ar) gas. After purgingthe chamber twice, Ar gas flow was maintained throughout thepyrolysis process at the rate of 2 l/min. Initially the temperaturewas raised to 500 ◦C at a ramp of 10 ◦C/min to prevent cracks in RFxerogel blocks due to rapid mass loss. Then the furnace tempera-ture was raised to final pyrolysis temperature with 20 ◦C/min rampfollowed by a dwell time for 1 h at respective final pyrolysis tem-perature. The samples were then cooled to room temperature inthe presence of Ar flow. Hereafter, RF derived carbon xerogel sam-ples without catalyst were denoted as RFC-T, while the RF derivedcarbon xerogel samples with 1% (w/w) graphitization catalyst weredenoted as RFCC-T where T denotes final pyrolysis temperature inboth the cases in degree Celsius.

2.4. Ball milling

RF derived carbon xerogel blocks were first crushed in a mortarfor 15 min before milling it to get fine powder in a ball mill. Crushedsamples were then filled into stainless steel bowl with 3 mm diam-eter SS balls. Sample to balls weight ratio was maintained to be 1:12for all the samples. The samples were milled using planetary ballmill pulversette7 (FRITSCH).

2.5. Characterization

Powder X-ray measurements were recorded in scattering anglerange of 10◦–50◦ on PANalytical X-ray diffractometer with CuK�radiation while Raman spectroscopy measurements were per-formed with a 532 nm laser on Bruker Raman microscope (Model:Senterra). Small angle X-ray scattering (SAXS) (0.01–10 nm−1)curves were recorded on point collimation SAXess (Anton-Paar)instrument in transmission mode to estimate the radius of gyration.The carbon xerogel powder samples were placed in between X-raytransparent Kapton films and packed between two copper win-dows with the help of a sample holder. X-ray path including sampleholder maintained in vacuum of 1 mbar to prevent the absorp-tion and scattering of X-rays (� = 0.154 nm) with air. Further highresolution transmission electron microscopy (HRTEM) (FEI TecnaiG2S-Twin) operated at 200 kV was used to study the morphologyof microstructure.

2.6. Electrode preparation

For electrochemical studies, SS foil was used as current collector.A slurry was prepared by dissolving 10% polyvinylidene difluoride(PVDF) and 90% carbon xerogel powder in N-methylpyrrolidone

M. Kakunuri et al. / Journal of Analytical and

Table 1Effect of pyrolysis temperature and catalyst on structural parameters of RF derivedcarbon xerogels.

Sample Degree ofgraphitizationg = (0.344-d002)/0.0086

Empiricalparameter R

Radius ofgyration Rg

(nm)

RFC-900 – – 2.12RFC-1200 – 1.78 2.19RFC-1500 – 1.46 2.44RFC-1800 – 1.90 2.70RFCC-900 – 2.54 2.95

(tl(sgtsu

3

3

tilobcrsnpcoDiissaiaciibmDtt1[fopeD(

RFCC-1200 0.45 3.51 2.97RFCC-1500 0.66 4.00 3.16RFCC-1800 0.69 4.83 3.21

NMP) solvent. The working electrode was prepared by coatinghis slurry on SS foil followed by vacuum drying at 120 ◦C andithium foil was used as a counter electrode. Glass microfiber filterWhatman, Grade GF/D) soaked with LP-30 electrolyte was used aseparator between two electrodes and packed inside the Ar filledlove box using Swagelok cell assembly. Before electrochemicalesting, the electrochemical cells were soaked for 12 h. Potentio-tat/Galvanostat (Bio-Logic Science Instruments, Model VSP) wassed to study the electrochemical performance of packed half-cells.

. Results and discussions

.1. Structural characterization

Raman spectrum of RF xerogel samples pyrolyzed at variousemperatures with and without catalyst is summarized belown Fig. 1a and b. For RF derived carbon xerogels without cata-yst (Fig. 1a), D-band at about 1345 cm−1 corresponding to edgesf small crystallite edge motions and vibrations of tetrahedrallyonded (sp3) carbon atoms, which are not located in graphiterystallite is dominating over G-band at about 1585 cm−1 that cor-esponds to two dimensional in-plane motion of strongly coupledp2 bonded carbons (E2g symmetry) in the hexagonal honeycombetwork of graphite lattice [1]. This trend is same for all samplesyrolyzed from 900 to 1800 ◦C. However in case of RF derivedarbon xerogels pyroyzed in the presence of catalyst (RFCC), webserved significant changes in the intensities of G-band and-band with varying temperature (Fig. 1b). Intensity of G-band

ncreased significantly compared to D band intensity with increasen pyrolysis temperature. Beyond 1200 ◦C, G-band peak intensitytarted dominating over D-band. Further we observed that for RFCamples (without catalyst), intensity ratio (ID/IG) of these two char-cteristics bands decreased marginally from 1.8 to 1.2 and similarlyn-plane crystallite width (in plane diameter) (La = 4.4/(ID/IG) nm)lso increased slightly from 2.3 to 3.4 nm as shown in Fig. 2a. Inontrast, RFCC samples (with catalyst) showed significant changen intensity ratio from 1.6 to 0.3 and in-plane crystallite widthncreased from 2.6 to 12.2 nm as shown in Fig. 2b. Graphitic G-ands of samples pyrolyzed above 1200 ◦C were observed to beore intense than D-band indicating initiation of graphitization.

band and G band position also varied with heat treatmentemperature, as temperature increases G band and D band posi-ion shifted towards lower wave number and G band approached582 cm−1 corresponding to E2g mode of single crystal graphite2]. In addition to D band and G band we can observe D’ bandor the samples RFCC-15000 and RFCC-1800. Further narrowingf G band and splitting in G band and D’ band for the samples

yrolyzed above 1200 ◦C clearly shows that the graphitization isnhanced in presence of graphitization catalyst [19]. In addition to

and G band, Raman spectrum also shows the second order peak2D) near 2700 cm−1, which appears most commonly in graphene

Applied Pyrolysis 117 (2016) 317–324 319

and graphitic carbon and often called as G’ band. This band is fre-quently used to distinguish single layer, bilayer graphene, frommultilayer graphite based on the shift towards smaller wavenum-ber as number of layers decreases [36]. In our study it can beobserved that the second most prominent peak (G’) of graphitematerials is not appeared in samples without metal catalyst asshown in Fig. 1(a). A broad 2D band can be seen for the RFCC-1200 sample, which is indication of beginning of graphitization.Further, intensity of this peak increased as graphitization increaseswith pyrolysis temperature. However, the shift in 2D band is verylittle towards higher wave number with increasing graphitiza-tion.

Along with in-planar crystallite width calculated from Ramanspectroscopy, crystallite thickness and interlayer spacing are twoother important parameters to study the extent of graphitization.

XRD patterns for RFC and RFCC samples as shown in Fig. 3were used to study these parameters. Fig. 3a and b shows thediffraction patterns of RFC and RFCC samples prepared at differ-ent pyrolysis temperatures. Diffraction peaks at 2� value of ∼25◦

and ∼43.8◦ corresponds to (0 0 2) and (1 0 l) reflections respectively[22]. No clear signature of diffraction peak at 2� = 25◦ for RFC sam-ple pyrolyzed at 900 ◦C reveals complete amorphous nature of thesecarbon xerogels. Even RFC samples pyrolyzed at higher tempera-ture (equal or more than 1200 ◦C), only a broad peak at 2� = 24◦

was observed which still indicates the amorphous nature of carbonxerogels, although within short range. Interestingly for RFCC sam-ples, the diffraction peak at about 26◦ was found to be very sharpas compared to RFC samples. This observation clearly suggest theordering of graphite layers at higher temperature in the presenceof catalyst. Further using XRD data, we have calculated interlayerspacing and crystallite thickness using Scherrer formula as summa-rized in Fig. 4 [37]. The angle at which (0 0 2) plane diffracted wasconsidered in this formula. For RFC samples, as shown in Fig. 4a,the interlayer spacing decreased from 3.61 to 3.49 A◦ while thecrystallite thickness increased marginally from 7.7 to 11.4 A◦ onincreasing the pyrolysis temperature from 900 to 1800 ◦C respec-tively.

However in case of RFCC samples as shown in Fig. 4b, interlayerspacing was reduced from 3.48 A◦ at 900 ◦C to 3.38 A◦ at 1500 ◦Cwhich remained nearly same even at 1800 ◦C. At the same time,the crystallite thickness increased significantly from 31 to 55 A◦ asshown in Fig. 4b. These observation suggested that in the presenceof catalyst at higher temperature, there was long range of orderedgraphitic layers within RF derived carbon xerogel samples. To sup-port this further, we recorded XRD patterns for RFC at very hightemperature 2500 ◦C (given in the supplementary material) how-ever a broad peak at 2� = 25.9◦ with 3.43 A◦ interlayer spacing and17 A◦ crystallite thickness suggests that RF derived carbon xerogelis non-graphitizable carbon without the use of catalyst [13].

In order to further understand the arrangement of graphene lay-ers in RF derived carbon xerogels, we conducted small angle X-rayscattering measurements. The intensity versus scattering vector (q)plots obtained for RFC and RFCC samples are shown in Fig 5a and b.Scattering vector q is defined as q = 4� sin

(�/2

)/� . As we observe

in case of RFC and RFCC samples both, there is a rise in scatteringintensity of the carbon xerogel samples with pyrolysis tempera-ture. This may be due to separation of graphite crystallites edgesand voids at higher temperatures [38]. Further we estimated theradius of gyration form the slope ln I(q) versus q2 plot of the scatter-ing data recorded for the carbon xerogel samples in Guinier regionusing following simplified equation [39].

2 2

ln I (q) = lnI (0) − RG q

3

Alignment of graphene layers at elevated temperature (graphiti-zation) resulted in reduction in the number of voids/pores between

320 M. Kakunuri et al. / Journal of Analytical and Applied Pyrolysis 117 (2016) 317–324

Fig. 1. Raman spectrum of carbon xerogel pyrolyzed at different temperature (a) without graphitization catalyst and (b) with 1 wt.% graphitization catalyst.

Fig. 2. Effect of pyrolysis temperature on in-plane crystallite width and ID/IG of RF derived carbon xerogel (a) without graphitization catalyst and (b) with 1 wt.% graphitizationcatalyst.

Fig. 3. XRD spectrum of RF derived carbon xerogel (a) without graphitization catalyst and (b) with 1 wt.% graphitization catalyst.

M. Kakunuri et al. / Journal of Analytical and Applied Pyrolysis 117 (2016) 317–324 321

Fig. 4. Effect of pyrolysis temperature on interlayer spacing and crystallite thickness of RF derived carbon xerogel (a) without graphitization catalyst and (b) with 1 wt.%graphitization catalyst.

Fig. 5. Small angle X-ray scattering pattern of RF derived carbon xerogel pyrolyzed at different temperatures (a) without graphitization catalyst and (b) with 1 wt.%graphitization catalyst.

Table 2A comparison of specific capacity for various organic gel derived carbón materials.

Sample type/Ref. No. 1st cycle reversiblecapacity(mAh/g)/coulombic

efficiency (%) Number of cyclesreported (n)

Specific capacityafter ‘n’ cycles(mAh/g)/Currentdensity

Carbon xerogel(powder)/[31]

171/43 30 145/100 mA/g

Carbon cryogel hollow microspheres/[42]

164/33 5 155/0.15 mA/cm2

Carbon cryogel microcapsules/[42] 360/34 5 340/0.15 mA/cm2

Carbon cryogel microspheres/[42] 140/32 5 130/0.15 mA/cm2

Carbon aerogel/[30] 312/20 50 242/50 mA/gNitrogen doped carbon xerogel/[32] 275/43 50 212/37.2 mA/g

tw[iiFttwog

Carbon xerogel thin films/[43] 195/47

RFC-1800 (present work) 300/40

RFCC-1800 (present work) 385/45

he graphene layers and therefore increase in pore diameter whichas correlated to radius of gyration as summarized in Table 1

40]. In case of graphitization without catalyst, radius of gyrationncreased from 2.12 to 2.7 nm while in the presence of catalyst, itncreased from 2.95 to 3.21 nm from 900 to 1800 ◦C respectively.or any given temperature, RFC sample has less radius of gyrationhan corresponding RFCC samples. Not only that, even at the highestemperature (1800 ◦C), radius of gyration for RFC samples (2.7 nm)

as less than RFCC samples at lowest temperature (900 ◦C). This

bservation was more profound while calculating the degree ofraphitization.

20 190/76.4 �A/cm2(257 mA/g)60 223/37.2 mA/g60 280/37.2 mA/g

We reported degree of graphitization based on a correlationbetween d-spacing (measured from XRD data), radius of gyrationfrom SAXS analysis and an empirical parameter (R) that corre-sponds to number of single graphene layers. Empirical parameter(R) is the ratio of the height of (0 0 2) diffraction peak and back-ground of XRD pattern. It is inversely proportional to numberof single graphene layers [12]. This empirical parameter valueincreased with pyrolysis temperature as summarized in Table 1.

Finally the degree of graphitization values revealed that graphi-tization occurred only beyond 1200 ◦C that too in the presence of

322 M. Kakunuri et al. / Journal of Analytical and Applied Pyrolysis 117 (2016) 317–324

F agnification image of RFC-1800, (c) low magnification image of RFCC-1800 and (d) highm

cf

apcltrctsaSc1opc

ocim

ig. 6. HRTEM micrographs of (a) low magnification image of RFC-1800, (b) high magnification image of RFCC-1800.

atalyst. At 1500 ◦C, degree of graphitization achieved was 0.66 thaturther increased marginally to 0.69 at 1800 ◦C.

To further examine the microstructure of carbon xerogels withnd without catalyst HRTEM micrographs were recorded for sam-les pyrolyzed at 1800 ◦C. Fig 6a and b reveals that RFC-1800 sampleontains disordered carbon with very few layer of short grapheneayers stacked without any well-ordered graphite domain. In con-rast, graphitized RFCC-1800 sample is made up of graphitic carbonibbon like structures. The shape of these ribbon like structuresan be attributed to precipitation of graphitic carbon aroundhe metal nanoparticles while cooling [23]. These ribbon liketructures of size about 4–6 nm (40–60 A◦) from HRTEM imagesre close to the crystallite thickness calculated from XRD usingcherrer equation. Improved graphitization observed in case ofarbon samples at higher temperatures with metal catalyst (RFCC-500 and RFCC-1800) can be ascribed to enhanced dissolutionf amorphous carbon into metal nanoparticles at elevated tem-erature followed by precipitation of graphitic carbon during theooling [15].

Once we studied the microstructure and crystallinityf RF derived carbon xerogel prepared with and without

atalyst, we then examined their electrochemical behav-or and explained it in terms of their structural para-

eters.

Fig. 7. Cyclic voltammogram of RFCC -1800.

3.2. Electrochemical performance

3.2.1. Cyclic voltammetryFig. 7 shows the cyclic voltammogram of RFCC-1800 sample

recorded at a scan rate of 0.1 mV/s in the voltage range of 0–3 V.It exhibited two cathodic current peaks, a broad peak at 0.6 V and

M. Kakunuri et al. / Journal of Analytical and Applied Pyrolysis 117 (2016) 317–324 323

ures (a

aSnplAap

3

a3sFft4Fbccc1wtniti

scfi9bcishtdsu9tc2

Fig. 8. Cyclic performance of carbon xerogel pyrolyzed at different temperat

nother sharp peak near 0 V. Broad peak at 0.6 V can be ascribed toEI formation due to electrolyte decomposition and sharp peak atear 0 V attributed to lithium insertion into carbon xerogel sam-les [41]. Anodic broad peaks at about 0.2 V and 1 V corresponds to

ithium extraction from carbon xerogel samples during charging.s there was no major change in chemical composition and surfacerea, all other samples (RFCC 1200 and RFCC 1500) show similareaks in cyclic voltammetry.

.2.2. Galvanostatic charge-discharge experimentsGalvanostatic charge discharge experiments were carried out

t 0.1C rate (37.2 mA/g current density) in between 0.01 V and V. Fig 8a and b shows the cyclic performance of RFC and RFCCamples pyrolyzed at different temperatures up to sixty cycles.or RFC samples (Fig. 8a), first cycle discharge capacity increasedrom 494 to 754 mAh/g with increase in temperature from 900o 1800 ◦C. Initial cyclic efficiency was also increased from 32 to0% for RFC samples pyrolyzed at 900 and 1800 ◦C respectively.or all RFC samples pyrolyzed at different temperatures, coulom-ic efficiency was observed to be more than 90% by third cycle ofharge/discharge which further increased to nearly 97 % after tenycles. After sixty cycles of continuous charge and discharge, spe-ific reversible capacities for RFC samples pyrolyzed from 900 to500 ◦C were stabilized in similar range from 187 to 195 mAh/gith no significant increase with temperature. However at fur-

her higher temperature (1800 ◦C), reversible capacity increasedearly 15% to 224 mAh/g. This electrochemical behavior was also

n line with structural characterization for these samples in whichhere was no significant change observed in their crystallinity andnternal microstructure.

Further more importantly in case of RF derived carbon xerogelamples prepared in presence of catalyst, we observed signifi-ant larger values of reversible capacities with much improvedrst cycle efficiency at higher temperature (Fig. 8b). For RFCC-00 sample, first discharge/intercalation capacity was found toe significantly higher (641 mAh/g) with 34% initial cycle effi-iency. Interestingly, for RFCC-1800 sample, first discharge capacityncreased further to 857 mAh/g while charge capacity was mea-ured to be 385 mAh/g, larger than that of graphite, with sufficientlyigher initial efficiency of 45%. For carbon materials in litera-ure, lower first cycle efficiency (20–35%) can always be explainedue to formation of solid electrolyte interface however for RFCCamples, we have shown a sufficiently high first cycle efficiencyp to 45%. Further coulombic efficiency was increased up to

8% for RFCC samples at higher temperature, a further confirma-ion of excellent cycling behavior of these samples. After sixtyycles of charge discharge, reversible capacity was stabilized to80 mAh/g for RFCC-1800 sample, a significant 25% increase to

) without graphitization catalyst and (b) with 1 wt.% graphitization catalyst.

their corresponding RFC-1800 sample. Among all samples, RFCC-1800 sample showed best electrochemical performance with itsspecific capacity in the similar range of graphitic carbon mate-rials. This behavior can easily be ascribed to higher degree ofgraphitization achieved that resulted in increased electronic con-ductivity of the sample. At the same time, we found an interestingobservation. RFCC-900 sample showed higher capacity than RFCC-1200 and RFCC-1500 samples. This could be due to adsorptionof lithium ions on surface of nanopores of disordered carbonof RFCC-900 sample compared to moderate graphitized sampleswith less number of nanopores of adsorption of lithium ions[12].

Reversible capacity and first cycle efficiency values reportedhere in this work are superior to what has been reported in litera-ture for RF derived carbon xerogel powder samples [31]. Further acomparison of electrochemical performance of various other formsof carbon gel materials is summarized in Table 2. This confirms theenhanced reversible capacity and much improved cyclic stability inRF derived carbon xerogels. This enhancement can be attributed tocatalytic graphitization which has improved the electronic conduc-tivity in basal plane of stacked graphene layers. The electrochemicalperformance of thus prepared carbon xerogel samples is much bet-ter than cryogels and aerogels in terms of initial cycle coulombicefficiency and cyclic stability.

4. Conclusions

RF derived carbon xerogels with and without catalyst were pre-pared by pyrolyzing RF xerogel at different temperatures from900 to 1800 ◦C. A detailed structural characterization using Ramanspectroscopy, XRD, SAXS and HRTEM suggested that RF derived car-bon xerogels were non-graphitizable (hard) carbon which couldbe graphitized in the presence of metal catalyst at elevated tem-perature of 1200 ◦C and beyond. On the other hand, there was noobserved change in crystallinity and arrangement of graphene lay-ers even at temperature as high as 2500 ◦C without catalyst. Degreeof graphitization and therefore crystallinity was found to be largelydependent on final pyrolysis temperature. These findings were fur-ther supported while examining the electrochemical performanceof these materials. Galvanostat charge/discharge experiments sug-gested the use of RF derived graphitized carbon xerogel as efficientanode materials for lithium ion battery with excellent cyclingbehavior and cyclic stability. Graphitization not only improved

first cycle efficiency and therefore reduced the irreversible capacitylosses associated with SEI layer formation in carbon materials butalso a significantly large first charge capacity was obtained in thesematerials.

3 l and

A

atU

A

t

R

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

microspheres and applications in lithium-ion batteries, Mater. Chem. Phys.133 (2012) 429–436.

[43] C.S. Sharma, H. Katepalli, A. Sharma, G.T. Teixidor, M.J. Madou, Fabrication of

24 M. Kakunuri et al. / Journal of Analytica

cknowledgments

CSS acknowledges the Indian Institute of Technology, Hyder-bad for providing necessary research infrastructure to carry outhis work. Authors acknowledge the use of HRTEM facility at Centralniversity of Hyderabad.

ppendix A. Supplementary data

Supplementary data associated with this article can be found, inhe online version, at http://dx.doi.org/10.1016/j.jaap.2015.10.017.

eferences

[1] F. Tuinstra, J.L. Koenig, Raman spectrum of graphite, J. Chem. Phys. 53 (1970)1126–1130.

[2] T. Jawhari, A. Roid, J. Casado, Raman spectroscopic characterization of somecommercially available carbon black materials, Carbon 33 (1995) 1561–1565.

[3] H. Azuma, H. Imoto, S. Yamada, K. Sekai, Advanced carbon anode materials forlithium ion cells, J. Power Sources 81–82 (1999) 1–7.

[4] M. Endo, C. Kim, K. Nishimura, T. Fujino, K. Miyashita, Recent development ofcarbon materials for Li ion batteries, Carbon 38 (2000) 183–197.

[5] H. Shi, J. Barker, M.Y. Saydi, R. Koksbang, L. Morris, Graphite structure andlithium intercalation, J. Power Sources 68 (1997) 291–295.

[6] K. Zaghib, G. Nadeau, K. Kinoshita, Effect of graphite particle size onirreversible capacity loss, J. Electrochem. Soc. 147 (2000) 2110–2115.

[7] P. Ridgway, H. Zheng, A.F. Bello, X. Song, S. Xun, J. Chong, Comparison ofcycling performance of lithium ion cell anode graphites, J. Electrochem. Soc.159 (2012) A520–A524.

[8] H. Huang, W. Liu, X. Huang, L. Chen, E.M. Kelder, J. Schoonman, Effect of arhombohedral phase on lithium intercalation capacity in graphite, Solid StateIonics 110 (1998) 173–178.

[9] A. Piotrowska, K. Kierzek, P. Rutkowski, J. Machnikowski, Properties andlithium insertion behavior of hard carbons produced by pyrolysis of variouspolymers at 1000 ◦C, J. Anal. Appl. Pyrolysis 102 (2013) 1–6.

10] W. Xing, J.S. Xue, T. Zheng, A. Gibaud, J.R. Dahn, Correlation between lithiumintercalation capacity and microstructure in hard carbons, J. Electrochem. Soc.143 (1996) 3482–3491.

11] I. Mochida, C. Ku, Y. Korai, Anodic performance and insertion mechanism ofhard carbons prepared from synthetic isotropic pitches, Carbon 39 (2001)399–410.

12] Y. Liu, J.S. Xue, T. Zheng, J.R. Dahn, Mechanism of lithium insertion in hardcarbons prepared by pyrolysis of epoxy resins, Carbon 34 (1996) 193–200.

13] A. Oya, H. Marsh, Review: phenomena of catalytic graphitization, J. Mater. Sci.17 (1982) 309–322.

14] R.E. Franklin, Crystallite growth in graphitizing and non-graphitizing carbons,Proc. R. Soc. Lond. A 209 (1951) 196–218.

15] R.E. Franklin, The structure of graphitic carbons, Acta Cryst. 4 (1981) 253–261.16] J.G. Zhao, F.Y. Li, C.Q. Jin, Graphitization of activated carbon under high

pressures and high temperatures, Solid State Commun. 149 (2009) 818–821.17] H. Marsh, A.P. Warburton, Catalysis of graphitization, J. Appl. Chem. 20 (1970)

133–142.18] A. Oya, S. Otani, Catalytic graphitization of carbons by various metals, Carbon

17 (1979) 131–137.19] S. Tzeng, Catalytic graphitization of electroless Ni-P coated PAN-based carbon

fibers, Carbon 44 (2006) 1986–1993.

20] G. Hasegawa, K. Kanamori, K. Nakanishi, Facile preparation of macroporous

graphitized carbon monoliths from iron-containing resorcinol-formaldehydegels, Mater. Lett. 76 (2012) 1–4.

21] Z. Ni, Q. Li, L. Yan, J. Gong, D. Zhu, Catalytic graphitization of furan resincarbon by yttrium, Carbon 46 (2008) 378–379.

Applied Pyrolysis 117 (2016) 317–324

22] F.J. Maldonado-Hodar, C. Moreno-Castilla, J. Rivera-Utrilla, Y. Hanzawa, Y.Yamada, Catalytic graphitization of carbon aerogels by transition metals,Langmuir 16 (2000) 4367–4373.

23] M. Sevilla, A.B. Fuertes, Catalytic graphitization of templated mesoporouscarbon, Carbon 44 (2006) 468–474.

24] H. Zhen-hua, P. Qi-ling, H. Zhen-hua, Y.U. Qiang, C. Jen-hua, K. Ya-fei, Catalyticgraphitization of PAN-based carbon fibers with electrodeposited Ni–Fe alloy,Trans. Nonferrous Met. Soc. China 21 (2011) 581–587.

25] W. Kicinski, J. Lasota, Carbon xerogel-supported iron as a catalyst incombustion synthesis of carbon fibrous nanostructures, J. Mater. Sci. Technol.28 (2012) 294–302.

26] W. Kicinski, M. Norek, M. Bystrzejewski, Monolithic porous graphitic carbonsobtained through catalytic graphitization of carbon xerogels, J. Phys. Chem.Solids 74 (2013) 101–109.

27] R.W. Pekala, Organic aerogels from the polycondensation of resorcinol withformaldehyde, J. Mater. Sci. 24 (1989) 3221–3227.

28] R.W. Pekala, J.C. Farmer, C.T. Alviso, T.D. Tran, S.T. Mayer, J.M. Miller, Carbonaerogels for electrochemical applications, J. Non Cryst. Solids 225 (1998)74–80.

29] B. Babic, B. Kaluderovic, L. Vracar, N. Krstajic, Characterization of carboncryogel synthesized by sol–gel polycondensation and freeze–drying, Carbon42 (2004) 2617–2624.

30] L. Nian-Ping, S. Jun, G. Da-yung, L. Dong, Z. Xiao-Wei, L. Ya-Jie, Effect of carbonaerogel activation on electrode lithium insertion performance, Acta Phys.Chim. Sin. 29 (2013) 966–972.

31] X. Yuan, Y.J. Chao, Z.F. Ma, X. Deng, Preparation and characterization of carbonxerogel (CX) and CX-SiO composite as anode material for lithium-ion battery,Electrochem. Commun. 9 (2007) 2591–2595.

32] X. Liu, S. Li, J. Mei, W. Lau, R. Mi, Y. Li, H. Liu, L. Liu, Frommelamine–resorcinol–formaldehyde to nitrogen-doped carbon xerogels withmicro- and meso-pores for lithium batteries, J. Mater. Chem. A 2 (2014)14429–14438.

33] S. Yoon, C. Park, H. Yang, Y. Korai, I. Mochida, R.T.K. Baker, Novel carbonnanofibers of high graphitization as anodic materials for lithium ionsecondary batteries, Carbon 42 (2004) 21–32.

34] Y. Chen, X. Li, X. Zhou, H. Yao, H. Huang, Y. Mai, Hollow-tunneled graphiticcarbon nanofibers through Ni-diffusion-induced graphitization ashigh-performance anode materials, Energy Environ. Sci. 7 (2014) 2689–2696.

35] J.M. Skowronski, K. Knofczynski, New anode material for lithium-ion cellsproduced by catalytic graphitization of glassy carbon at 1000 ◦C, J. New Mat.Electrochem. Syst. 9 (2006) 359–365.

36] L.M. Malard, M.A. Pimenta, G. Dresselhaus, M.S. Dresselhaus, Ramanspectroscopy in graphene, Phys. Rep. 473 (2009) 51–87.

37] M.A. Short, P.L. Walker, Measurement of interlayer spacing and crystal sizes inturbostratic carbons, Carbon 1 (1963) 3–9.

38] K. Fukuyama, T. Nishizawa, K. Nishikawa, Investigation of the pore structurein glass-like carbon prepared from furan resin, Carbon 39 (2001) 2017–2021.

39] H. Tamon, H. Ishizaka, SAXS study on gelation process in preparation ofresorcinol-formaldehyde aerogel, J. Colloid Interface Sci. 206 (1998) 577–582.

40] E. Buiel, A.E. George, J.R. Dahn, On the reduction of lithium insertion capacityin hard-carbon anode materials with increasing heat-treatment temperature,J. Electrochem. Soc. 145 (1998) 2252–2257.

41] S. Zhang, M.S. Ding, K. Xu, J. Allen, J.R. Jow, Understanding solid electrolyteinterface film formation on graphite electrodes, Electrochem. Solid-State Lett.12 (2001) A206–A208.

42] H. Zhang, H. Xu, C. Zhao, Synthesis of morphology-controlled carbon hollowparticles by carbonization of resorcinol—formaldehyde precursor

resorcinol-formaldehyde xerogel based high aspect ratio 3-D hierarchicalC-MEMS structures, ECS Trans. 61 (2014) 45–54.