Quantitative appraisal of the interfacial anchoring state of polyaromatic hydrocarbons during the formation of C/C composites

(1)

Quantitative appraisal of the interfacial anchoring state

of polyaromatic hydrocarbons during the formation

of C/C composites

Cheng-Te Lin

a

, Min-Chiao Tsai

a

, Chi-Young Lee

a,b

, Hsin-Tien Chiu

c

,

Tsung-Shune Chin

d,*

a_{Department of Materials Science and Engineering, National Tsing Hua University, Hsinchu 30013, Taiwan}

b_{Center of Nanotechnology, Materials Science, and Microsystems, National Tsing Hua University, Hsinchu 30013, Taiwan} c_{Department of Applied Chemistry, National Chiao Tung University, Hsinchu 30050, Taiwan}

d_{Department of Materials Science and Engineering, Feng-Chia University, Taichung 40724, Taiwan}

A R T I C L E I N F O Article history:

Received 29 August 2009 Accepted 11 November 2009 Available online 15 November 2009

A B S T R A C T

The preferred anchoring state for polycyclic aromatic hydrocarbons (PAHs) pyrolyzed onto carbon substrates has been studied by semi-quantitative methods, examining their inter-facial lattice arrangements. The samples were prepared by decomposing petroleum pitch inside carbon nanotubes resulting in a variety of crystallinities forming one-dimensional C/C composites. Studies indicate that the preferred anchoring state of PAH molecules depends on the nature of the substrate. Accordingly the PAHs in mesophase pitch should exhibit a face-on orientation on the carbonaceous substrates, including a graphite sheet, glassy carbon, and pyrolytic carbon. However, it showed that the anchoring state (face-on or edge-(face-on) of PAH units can be altered even (face-on carb(face-onaceous substrates. The results demonstrated that in C/C composites the anchoring state is predominantly determined by the relative degree of crystallinity of the pitch/carbon substrate, and can be semi-quan-titatively estimated using the ID/IGratio from Raman spectra. Face-on anchoring is

pre-ferred when the ID/IGratio of substrate is smaller (higher crystallinity) than that of the

pyrolyzed precursor, whereas edge-on anchoring is favored when it is larger. Such an esti-mation method is useful in tailoring microstructures in the fabrication of C/C composites using PAH precursors.

1. Introduction

The investigation of the surface anchoring effect at the boundaries between polycyclic aromatic hydrocarbon (PAH) molecules/substrates attracts academic interest. There are two categories of surface anchoring states, edge-on and face-on, expounded by Hurt et al. [1–3]. The preference of the anchoring state is dependent on the competition between intermolecular forces within PAH molecules and between PAH

units/substrate. Usually the sample is prepared by spreading and pyrolyzing the PAH precursor on the substrate for obser-vation under a polarized optical microscope. However, the conventional method cannot provide direct evidence of the anchoring state in the vicinity of the boundary between PAHs/matrices. This is because the ordered anchoring region is sometimes on the scale of sub-micrometers and therefore beyond the optical resolution limit [2]. Another effective way to obtain information of interfacial anchoring states is

* Corresponding author: Fax: +886 3 5719868.

E-mail address:tschin@mx.nthu.edu.tw(T.-S. Chin).

a v a i l a b l e a t w w w . s c i e n c e d i r e c t . c o m

(2)

to examine the arrangement of graphene layers within the sample using high-resolution transition electron microscopy (TEM). Since the penetration depth of electron beams at 200 kV is only tens of nanometers, the sample prepared for optical microscope analysis is not appropriate for TEM observation.

Recently, the template synthesis method using nanopor-ous anodic alumina (AAO) has been extensively applied to produce various one-dimensional carbon nanomaterials, like nanofibers [4–6] or nanotubes [7–9]. The microstructure of such nano-sized materials can be easily studied under TEM. In addition, it is known that the alignment of PAHs, which is affected by interaction with the surface of the substrate (the surface anchoring effect), can be extended into the sam-ple body in the range of sub-micrometers[1,6]. Although the anchored molecules are restricted to the superficial region of a bulk material, the arrangement of the PAH molecules can dominate the entire microstructure as the dimension of the products approaches nanoscale. Therefore, such one-dimensional carbon nanomaterials are ideal intermediates to study the interfacial microstructure because of the pro-nounced anchoring behavior.

Present-day studies imply that the anchoring states of mesophase pitches depend on the type of substrate[1–3,10]. Edge-on anchoring is found on most substrates, whereas face-on anchoring is only observed on carbon, mica, polyim-ide, and silver. These reports also indicate that PAH molecules always exhibit a face-on orientation on a graphite-based sur-face and carbonaceous substrates, such as glassy carbon, car-bon fibers, and pyrolytic carcar-bon. The results inspired our interest in investigating whether PAHs can exhibit alternative anchoring states on a certain substrate and, if so, what inter-facial anchoring patterns will be shown.

In this study, a variety of tubular carbon substrates were prepared and investigated for their preferred orientations of the graphene layers at the interface of pitch/carbon ites. The anchoring states of carbonized pitch in C/C compos-ites were identified by high-resolution TEM and quantitatively characterized by Raman spectroscopy for their crystallinity.

2. Experimental

Thin carbon layers (5–50 nm) with different degrees of crystal-linity were coated on a channel surface of AAO (Whatman Ltd.) becoming a carbonaceous substrate. Because each chemical vapor deposition (CVD) process has its own limita-tions, several kinds of deposilimita-tions, such as acetylene decom-position (Samples B and D)[11,12], ethanol pyrolysis (Sample C)[13,14], and Wurtz-type reaction (Sample E) [15–17]were employed to coat the AAO channels. The details of prepara-tion can be found in Supporting Informaprepara-tion. The adopted PAH precursor is isotropic petroleum pitch (A-240, Ashland Inc.), which is composed of hetero-PAH molecules. The pre-cursor was spread on top of the templates with or without carbon coating for pyrolyzing the PAHs inside the channels during subsequent thermal treatment. In a tube furnace the temperature was held at 300 °C for half an hour under flowing Ar to soften the precursor, and then gradually raised to 700 °C for 2 h for the pyrolysis. The oxide template was removed by

immersing in hydrofluoric acid. For the study of interfacial microstructure, all the products were annealed at 2500 °C to enhance their lattice orientation. The samples were charac-terized by high-resolution TEM (HRTEM, JEOL JEM-2010) and Micro-Raman scattering measurements (CHROMEX 501is: 488 nm).

3. Results and discussion

The petroleum pitch carbonized at 700 °C in the bare channels of AAO template without carbon coating (Sample A) was first examined. After leaching off the template, the product was found to be composed of uniform nanofilaments, as shown in Fig. 1a. Their diameter and length were evaluated to be 300 ± 50 nm and 60 lm, respectively, which accurately corre-spond to the channel size of the template. This implies that during the softening process, the channels of the template were fully impregnated with liquid-phase pitch, in which the alignment of PAH molecules is strongly influenced by the oxide inner surface of the channels. In the lattice image (Fig. 1b) it can be seen that the orientation of graphene layers is normal or oblique to the longitudinal axis of the nanofila-ments. This suggests that the interaction between PAHs and alumina would result in an edge-on anchoring state. The re-sult is in agreement with that of previous observations [5,18,19].

Carbon substrates made by coating carbon on AAO chan-nels with different methods (Samples B–E) were obtained after the removal of template. Their appearance in scanning electron micrographs (SEM) is similar to that inFig. 1a. How-ever, in TEM (see Fig. S2) these substrates exhibit a tubular structure, and their hollow cores will be used in pitch-filling to form C/C composites with a core–shell structure. The dif-fraction spots inFig. S2acan be assigned to the (0 0 2) lattice plane of graphite. It indicates that after graphitization the car-bon substrate of Sample B comprises of graphite basal planes, whereas other samples show nanocrystalline carbon. In order to determine their interfacial anchoring state, the lattice images at the boundary of C/C composites (Samples B–E) were observed by high-resolution TEM. These core–shell materials were also treated at >2500 °C to develop crystallinity, thus enhancing the contrast of the lattice framework. Such a high-temperature treatment does not change the graphene orientation of the sample post-carbonization[2,20].

Fig. 2a (Sample B) shows that the arrangement of carbon-ized PAH units on the right side prefer to align parallel to the basal planes of the substrate on the left side, representing a face-on anchoring behavior. A similar result is seen with the pyrolysis of naphthalene polymer into multi-walled carbon nanotubes [21]. The lattice images become slightly clearer after thermal treatment, and the alignment tendency does not change (seeFig. 2b). As liquid-phase pitch carbonized in the channel coated with nanocrystalline carbon layer, the C/ C interface becomes ambiguous and difficult to distinguish, as shown inFig. S3a and c. After high-temperature treatment, the lattice alignment of PAH units in both Samples C and D can be discriminated as having a face-on anchoring prefer-ence (seeFig. S3b and d), in agreement with previous observa-tions as well[1,2].

(3)

An entirely different result was obtained when the PAHs were carbonized on a carbon substrate with very low crystal-linity (Sample E). After carbonization, an ambiguous C/C boundary with different lattice orientations can be roughly observed, as shown in Fig. 2c. In lattice images of Fig. 2d and e, the inclined edge-on anchoring from graphitized

Sam-ple E is shown. The semi-circular loops on the edge of the pitch filler imply that the dangling sites, rather than basal planes of PAH units, are exposed at the C/C interface, con-firming edge-on alignment. Such loops on the surface of graphene edge are usually found in graphitized carbon mate-rials[5,20]. Two sets of (0 0 2) diffraction spots exhibited in the Fig. 2 – Interfacial lattice images of PAHs pyrolyzed on graphite (Sample B) and nanocrystalline carbon (Sample E) substrates: Sample B after (a) carbonization, and (b) graphitization; Sample E after (c) carbonization, and (d) (e) graphitization. A dot line indicates where the C/C boundary is.

Fig. 1 – (a) SEM and (b) high-resolution TEM images of carbon nanofilaments (Sample A) produced by carbonizing the pitch in the channels of nanoporous AAO. The lattice image exhibits an edge-on anchoring state.

(4)

inset ofFig. 2d also signifies that at the C/C boundary it is composed of two groups of graphene stackings. The signifi-cant influence of surface anchoring effect in nanoscale can be manifest inFig. S4. Even though the thickness of carbon shell (substrate) is only 2% of the diameter of the nanofila-ment (seeFig. 2e), the appearance of the C/C composite is still substantially altered as the anchoring state changes. In nano-confined environments, the enlarged elastic strain can domi-nate the anchoring arrangement as well. Four possible face-on cface-onfiguratiface-ons cface-onsidering surface anchoring strength and elastic strain were proposed by Jian et al.[21]. The gra-phitic microstructure of Sample E is disagreement with any proposed face-on configuration, indicating that two anchor-ing states can not be interchanged by internal elastic strain. In addition, edge-on state on disordered carbons might be af-fected by oxygen-containing functional groups on the surface of substrates. Although during thermal treatment most of ad-sorbed molecules should be expelled by reaction with carbons.

The degree of graphitization of all samples was deter-mined by Raman scattering spectroscopy to systematically investigate the correlation between anchoring states and sub-strate crystallinity. Although the crystallinity of samples can be identified by the diffraction patterns of TEM as well, it is difficult to do quantitative comparison. Additionally, average grain size of samples can be estimated by peak width col-lected from X-ray diffractometer using Scherrer’s equation. However, as the crystallinity of carbon samples is low, only tiny (0 0 2) peak can be found in the pattern and others are hidden in the background. Therefore, the ID/IGratio estimated

from Raman pattern is suitable to semi-quantitatively denote the degree of graphitization of samples in a low-crystalline state. In this study, ID/IG ratios of samples were calculated

as the ratio of peak heights using Breit–Wigner–Fano (BWF) + Lorentzian fits for comparison with Ferrari’s work [22]. The BWF function that has an asymmetric line shape, is given by: IðxÞ ¼I0½1 þ 2ðx x0Þ QC 2 1 þ ½2ðx x0Þ C 2 ð1Þ

where I0is the peak intensity, x0is the peak position, C is the

full width at half-maximum (FWHM), and Q is the skewness coefficient. The symmetrical Lorentzian shape is recovered in the limit as Q1_!₀_[22–24]_{. Such a curve-fitting method}

is expounded to be satisfactory for the case of nanocrystalline and amorphous carbons[25]. All the fitting results were sum-marized inTable 1, including the ID/IGratios obtained using a

Gaussian fitting routine[26,27]. The correlation of the ID/IG

ra-tios between using two kinds of fitting procedures is shown in Fig. S5.

In Raman spectra (Fig. 3a and b), the G-band (at 1550– 1605 cm1_{) is attributed to the stretching-vibration mode of}

sp2 _{sites, such as in aromatic compounds or olefins (C}_@C

chains), whereas D-band (near 1350 cm1) is caused by the breathing mode of those sp2sites solely in aromatic rings. Fer-rari et al. indicated that ranging from graphite (100% sp2_sites)

to tetrahedral amorphous carbon (100% sp3_{phase) visible}

Ra-man spectra can be elucidated using a three-stage model[22]. In our case while the G-band moves to the lower frequency, the ID/IGratio decreases, and the FWHM of peak decreases as

well. Therefore it can be concluded that all studied samples are composed of from micro- to nanocrystalline graphite with-out sp3_{hybridization. The average I}

D/IGratios of Sample A after

pyrolysis (at 300 °C) and carbonization (at 700 °C) evaluated fromFig. 3a are 0.74 and 0.75, respectively. At the pitch/AAO interface the dispersive interaction between pitch/oxide is much weaker than the internal p–p bonding within PAH units, resulting in strong edge-on anchoring (seeFig. 1b). Typical Ra-man spectra of as-prepared carbon substrates before pitch-fill-ing are shown inFig. 3b. The average ID/IGratio of Sample B is

0.15, indicating a well-developed graphite structure. The ratios of other samples with lower crystallinity were estimated to be in the range 0.60–1.09. Sample E has the highest ID/IGratio,

rep-resenting a very low-crystalline phase.

It was noted from the average ID/IGratios, that if the

crys-tallinity of carbon substrate (ID/IGratio: 1.09, Sample E) is

infe-rior to that of pyrolyzed pitch (ID/IGratio: 0.74), the edge-on

anchoring state is created. On the contrary, when the crystal-linity of substrate is close to or better than that of pyrolyzed pitch, a face-on anchoring arrangement will be formed. Fig. 3c interprets that the interfacial microstructure is related to the ID/IGratios (degrees of graphitization) of filler/substrate.

This suggests that as the liquid-phase pitch pyrolyzes inside the carbon-coated channels, a competition in degree of order between self-assembling PAH molecules and the interfacial bonding with substrate will determine its ultimate anchoring state. Edge-on anchoring occurs on nanocrystalline carbon substrates because the PAH molecules assemble with them-selves having higher surface energy than align along the sub-strate surface. As the degree of graphitization of subsub-strate increases, the intermolecular attraction between pitch/sub-strate is enhanced, resulting in face-on anchoring to reach a lower energy conformation. An illustration of the two anchor-ing states at the boundary of pitch/carbon substrate is shown inFig. 4. Such a semi-quantitative estimation method using

Table 1 – Fitting parameters using BWF + Lorentzian and double-Gaussian fits for all samples.

Sample A (300 °C) Sample A (700 °C) Sample B Sample C Sample D Sample E

G-band positiona_(cm1₎ _1598.5 _1594.8 _1587.0 _1589.5 _1602.2 _1581.3

FWHMa(cm1) 75.1 72.1 119.9 76.0 53.1 20.0

Peak ratioa(ID/IG) 0.74 0.75 1.09 0.69 0.60 0.15

Peak ratiob(ID/IG) 2.60 2.73 4.11 2.63 2.45 0.42

a _{The parameters were obtained using BWF + Lorentzian fits; the I}

D/IGratio was evaluated from the ratio of peak heights. b_{The I}

D/IGratio was evaluated from the ratio of integrated peak areas using two Gaussians.

(5)

Fig. 3 – Raman spectra measured from (a) pitch carbonized inside the channels of AAO, and (b) various carbon substrates before pitch-filling, (c) the ID/IGratios of samples versus their FWHM showing a linear relationship.

Fig. 4 – Comparative schemes: (a) face-on anchoring is preferred as the crystallinity of carbon substrate is better than that of pyrolyzed PAHs. (b) On the contrary edge-on anchoring will be created.

(6)

average ID/IGratios provides a simple and useful way to

pre-program the microstructure of pitch/carbon composites. The interfacial texture in C/C composites deeply influences their physical and mechanical behavior[28,29].

4. Conclusions

The preferred anchoring state at C/C boundary in composites has been studied by directly observing their lattice orientation in high-resolution TEM. After examining the results of car-bonized pitch on various carbon substrates, it has been dem-onstrated that the anchoring state is not only dominated by the nature of the substrates, but also the relative degree of crystallinity of pitch/carbon substrate. In addition, the resul-tant anchoring state can be determined by comparing the average ID/IGratio of pyrolyzed filler with that of carbon

sub-strates from Raman spectra. When the carbon substrate has higher crystallinity than that of pyrolyzed pitch, the interfa-cial graphene orientation shows face-on anchoring, other-wise edge-on anchoring is preferred. This is helpful in understanding the microstructural patterns in C/C compos-ites using PAHs as precursor.

Acknowledgements

The authors would like to thank the National Science Council of the Republic of China, Taiwan, for a partial financially sup-porting this research under Contract Nos. NSC 95-2120-M007-008, NSC 94-2213-M-007-035, and NSC 94-2213-M-009-003. Miss Jacqueline Kao is acknowledged for the correction of grammatical and writing style errors.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, atdoi:10.1016/j.carbon.2009.11.024.

R E F E R E N C E S

[1] Hurt R, Krammer G, Crawford G, Jian KQ, Rulison C. Polyaromatic assembly mechanisms and structure selection in carbon materials. Chem Mater 2002;14(11):4558–65. [2] Jian KQ, Shim HS, Tuhus-Dubrow D, Bernstein S, Woodward

C, Pfeffer M, et al. Liquid crystal surface anchoring of mesophase pitch. Carbon 2003;41(11):2073–83.

[3] Han KQ, Xianyu HQ, Eakin J, Gao YM, Crawford GP, Hurt RH. Orientationally ordered and patterned discotic films and carbon films from liquid crystal precursors. Carbon 2005;43(2):407–15.

[4] Jian KQ, Shim HS, Schwartzman A, Crawford GP, Hurt RH. Orthogonal carbon nanofibers by template-mediated assembly of discotic mesophase pitch. Adv Mater 2003;15(2):164–7.

[5] Lin CT, Chen WC, Yen MY, Wang LS, Lee CY, Chin TS, et al. Cone-stacked carbon nanofibers with cone angle increasing along the longitudinal axis. Carbon 2007;45(2):411–5.

[6] Chan C, Crawford G, Gao YM, Hurt R, Jian KQ, Li H, et al. Liquid crystal engineering of carbon nanofibers and nanotubes. Carbon 2005;43(12):2431–40.

[7] Bae EJ, Choi WB, Jeong KS, Chu JU, Park GS, Song S, et al. Selective growth of carbon nanotubes on pre-patterned porous anodic aluminum oxide. Adv Mater 2002;14(4):277–9. [8] Zhi LJ, Gorelik T, Wu JS, Kolb U, Mu¨llen K. Nanotubes

fabricated from Ni-naphthalocyanine by a template method. J Am Chem Soc 2005;127(37):12792–3.

[9] Liu QY, Li Y, Liu HG, Chen YL, Wang XY, Zhang YX, et al. Nanotubes fabricated from sandwich-type mixed (porphryinato)(phthalocyaninato)europium complex by template technique. J Phys Chem C 2007;111(20):7298–301. [10] Brooks JD, Taylor GH. Formation of graphitizing carbons from

the liquid phase. Nature 1965;206:697–9.

[11] Sui YC, Acosta DR, Gonzalez-Leon JA, Bermudez A, Feuchtwanger J, Cui BZ, et al. Structure, thermal stability, and deformation of multibranched carbon nanotubes synthesized by CVD in the AAO template. J Phys Chem B 2001;105(8):1523–7.

[12] Jeong SH, Lee KH. Fabrication of the aligned and patterned carbon nanotube field emitters using the anodic aluminum oxide nano-template on a Si wafer. Synth Met

2003;139(2):385–90.

[13] Yu GJ, Wang S, Gong JL, Zhu DZ, He SX, Li YL, et al. Synthesis of carbon nanotube arrays using ethanol in porous anodic aluminum oxide template. Chin Sci Bull 2005;50(11):1097–100. [14] Lin C-T, Chen T-H, Chin T-S, Lee C-Y, Chiu H-T. Quasi

two-dimensional carbon nanobelts synthesized using a template method. Carbon 2008;46(5):741–6.

[15] Wurtz A. Ueber eince neue Klasse organischer Radicale. Justus Liebigs Ann Chem 1855;96(3):364–75.

[16] Wang LS, Lee CY, Chiu HT. New nanotube synthesis strategy application of sodium nanotubes formed inside anodic aluminum oxide as a reactive template. Chem Commun 2003;15(15):1964–5.

[17] Huang CH, Chang YH, Wang HW, Cheng S, Lee CY, Chiu HT. Formation of porous carbon materials with in situ generated NaF nanotemplate. J Phys Chem B 2006;110(24):11818–22. [18] Kim TW, Park IS, Ryoo R. A synthetic route to ordered

mesoporous carbon materials with graphitic pore walls. Angew Chem Int Ed 2003;42(36):4375–9.

[19] Zhi LJ, Wu JS, Li JX, Kolb U, Mu¨llen K. Carbonization of disclike molecules in porous alumina membranes: toward carbon nanotubes with controlled graphene-layer orientation. Angew Chem Int Ed 2005;44(14):2120–3.

[20] Jian K, Yan A, Kulaots I, Crawford GP, Hurt RH. Reconstruction and hydrophobicity of nanocarbon surfaces composed solely of graphene edges. Carbon 2006;44(10):2102–6.

[21] Jian KQ, Hurt RH, Sheldon BW, Crawford GP. Visualization of liquid crystal director fields within carbon nanotube cavities. Appl Phys Lett 2006;88(16):163110.

[22] Ferrari AC, Robertson J. Interpretation of Raman spectra of disordered and amorphous carbon. Phys Rev B

2000;61(20):14095–107.

[23] McCulloch DG, Prawer S, Hoffman A. Structural investigation of xenon-ion-beam-irradiated glassy carbon. Phys Rev B 1994;50(9):5905–17.

[24] McCulloch DG, Prawer S. The effect of annealing and implantation temperature on the structure of C ion-beam-irradiated glassy carbon. J Appl Phys 1995;78(5):3040–7. [25] Nakazawa H, Mikami T, Enta Y, Suemitsu M, Mashita M.

Structure chemical bonding and these thermal stabilities of diamond-like carbon (DLC) films by RF magnetron sputtering. Jpn J Appl Phys 2003;42(6B):676–9.

[26] Tamor MA, Vassell WC. Raman ‘‘fingerprinting’’ of amorphous carbon films. J Appl Phys 1994;76(6):3823–30.

(7)

[27] Wang Y, Serrano S, Santiago-Aviles JJ. Raman characterization of carbon nanofibers prepared using electrospinning. Synth Met 2003;138(3):423–7.

[28] Trouvat B, Bourrat X, Naslain R. Interest of adding a carbon interphase on mechanical properties of C/C microstructures. EuroCarbon 98 1998;2:651–2.

[29] Dwivedi H, Mathur RB, Dhami TL, Bahl OP, Monthioux M, Sharma SP. Evidence for the benefit of adding a carbon interphase in an all-carbon composite. Carbon 2006;44(4):699–709.