HUNGARIAN JOURNAL OF INDUSTRY AND CHEMISTRY Vol. 45(1) pp. 61–66 (2017) hjic.mk.uni-pannon.hu DOI: 10.1515/hjic-2017-0009 SURFACE ENERGY HETEROGENEITY PROFILES OF CARBON NANOTUBES WITH A COPOLYMER-MODIFIED SURFACE USING SURFACE ENERGY MAPPING BY INVERSE GAS CHROMATOGRAPHY FRUZSINA GERENCSÉR, 1 NORBERT RIEDER, 1 CSILLA VARGA, 2 JENŐ HANCSÓK, 2 AND ANDRÁS DALLOS 1* 1 Department of Physical Chemistry, University of Pannonia, 10 Egyetem str., Veszprém, H-8200, HUNGARY 2 MOL Department of Hydrocarbon and Coal Processing, University of Pannonia, 10 Egyetem str., Veszprém, H-8200, HUNGARY The effectiveness and quantitative control of the surface transition of multi-walled carbon nanotubes (MWCNTs) was characterized by inverse gas chromatography (iGC). The surface energy profile of carbon nanotubes com- patibilized with an olefin-maleic-anhydride-ester-amide (OMAEA)-type coupling agent was determined by a sur- face energy analyzer (SEA). The surface energetic heterogeneity with energy distributions of dispersive and specific (acid-base) components of the surface energy of the MWCNTs were determined at various surface cov- erages. The results of the surface energy mapping showed that surface treatment significantly reduced the dis- persive surface energy of MWCNTs and increased the specific surface energy. Furthermore, the surface modifi- cation enhanced its Lewis basic character and simultaneously decreased the acidic character of MWCNTs. It has been demonstrated that the surface treatment modified the heterogeneity profiles of the energetic surface of the carbonaceous nanomaterials. Keywords: carbon nanotubes, surface treatment, inverse gas chromatography, surface energy analysis 1. Introduction Carbon nanotubes (CNTs) can serve as excellent candi- date materials for uses in numerous industrial applica- tions because of their considerable advantages. CNTs are one of the best reinforcing constituents for nano- composites [1] and hopefully catalytic metal–support in heterogeneous catalysis [2]. CNTs could replace the common catalyst supports of Ni/Mo-catalysts used in the production of fuel components of engine fuels with high hydrogen contents in their molecular structures [3]. Carbon nanotube-supported Co/Mo-catalysts with different Co/Mo atomic ratios were successfully used in the hydrocracking reaction of the vacuum resi- due of crude oil from Gudao oil field [4]. In the Fisch- er–Tropsch process (FTP), CNTs that supported transi- tion metal catalysts are used to increase catalytic activi- ties. An excellent study of FTP on Co catalysts support- ed by CNTs was reported by Tavasoli et al. [5]. Chen et al. [6] demonstrated that Fe nanoparticles encapsulated in CNTs are promising catalyst in FTP to synthesize light olefins. The catalytic consequence of hydrothermal liquefaction of microalgae to produce bio-oil over CNT- *Correspondence: dallos@almos.vein.hu supported transition metal (Co, Ni, Pt) catalysts was reported by Chen et al. [7]. To change the wettability and chemical character of the CNTs or to avoid agglomeration in nanocompo- sites, the CNT surfaces are often exposed to surface functionalization [2] and modification processes using polyfunctional anchoring, capping, and coupling agents [8]. Research has shown that metal–support bindings can be strengthened by functional groups that are cova- lently bonded (grafted) to the support. Functionalized carbon nanotube-supported Pt nanoparticles were ap- plied with favourable results in terms of selective olefin hydrogenation [9]. Because CNTs adsorb molecules well, functionalized CNTs are attractive chromatograph- ic stationary phases for separation of normal and isoal- kanes and aromatic compounds in the development of alternative fuels with high hydrogen/carbon ratios [10]. However, non-covalent functionalization using coupling agents or compatibilizers does not perturb the structure of the carbon nanotubes, establishes proper interactions between carbon nanotubes and the polymer matrix, and prevents the formation of nanotube agglom- erates [11]. In terms of the properties of the reinforced composites of CNTs, the couplings between the nano- tubes and the matrix are important beside the mechani- cal properties of the building parts [12]. These interac- tions depend on the surface properties and energies of the two materials. The surfaces of chemically derivati- zated CNTs were investigated by means of various ana- GERENCSÉR, RIEDER, VARGA, HANCSÓK AND DALLOS Hungarian Journal of Industry and Chemistry 62 lytical methods, e.g. thermal analysis [9-10,13-14], in- frared spectroscopy (IR) [10,13], transmission electron microscopy (TEM) [13,16], Raman [16] and atomic force microscopy [10], and inverse gas chromatography (iGC). iGC is a precise analytical method which is suita- ble for determining the surface energetic characteristics of the CNTs [13,15-17]. iGC was used for the character- ization of the chemical character of the surface and was utilized to measure dispersive and specific surface ener- gies, of numerous CNT substances [18]. The quantita- tive characterization of surface functionalization by surface energy mapping is of great importance. Howev- er, previous papers have presented surface energy val- ues for functionalized CNTs over unclear surface cover- ages without energetic profiles and surface energy dis- tribution functions, which, therefore, could not give correct information on the surface of the CNTs. In this study, the dispersive, specific (acid-base) components of the surface energy with their heterogene- ity charts and energy probability density functions of untreated and compatibilized MWCNTs are presented. A comparative quantitative characterization of the effec- tiveness and quantitative control of surface treatment is given. The exclusive energy scaling of the surfaces of the MWCNTs by energy heterogeneity charts with sur- face energy probability density functions over wide sur- face coverages is the new approach and main novelty of this paper. 2. Experimental and Methods 2.1. Samples and Measurements Multi-walled carbon nanotubes (MWCNTs) were manu- factured at 973 K by the chemical vapour deposition (CVD) process over a Fe/Co bimetallic catalyst at the Department of Chemical Engineering Science (Univer- sity of Pannonia, Veszprém, Hungary) [19]. Their di- ameter was between 10 and 20 nm and their average length was above 30 μm. An olefin-maleic-anhydride-ester-amide (Fig.1) copolymer (OMAEA) was used as a compatibilizer. The coupling agent was synthesized at the Department of MOL Hydrocarbon and Coal Processing (University of Pannonia, Veszprém, Hungary). The surface of MWCNTs was covered by the compatibilizer from a hydrocarbon solution of the coupling agent while the mixture was stirred for 1 hour at 333 K. The solvent was subsequently evaporated and the treated MWCNTs were dried at 383 K for 2 hours in air [11]. The surface energies of as-received and compati- bilized samples of MWCNTs were measured by a Sur- face Energy Analyzer (iGC-SEA, Surface Measurement Systems Ltd., Alperton, UK) over a series of surface coverages from (n/nm) = 0.005 to (n/nm) = 0.030. iGC samples were produced by filling 20-25 mg of CNTs into silanized Pyrex glass tubes of I.D. = 3 mm under a vacuum and moderate vibration. The samples of MWCNTs were stabilized in the column with plugs of silanized glass wool. The samples were preconditioned in the column at the actually measured temperature for 60 minutes before each measurement. The iGC experi- ments were carried out at a column temperature of 353 K, with a Helium carrier gas flow of 10 cm 3 /min. Me- thane gas was used as a dead-time marker using a flame ionization detector; and n-hexane, n-heptane, n-octane, n-nonane, chloroform and toluene as test compounds. The surface energy values were estimated using the specific retention volumes of the test compounds [20]. The specific retention volumes were obtained from the adjusted retention times: spw mVtV /c ' R  (1) The mean flow rate of the carrier gas in the column, cV  , was evaluated as given in Ref. [20]. 2.2. Theoretical Methodologies The dispersive component of the surface energy ( ds ) and its heterogeneity profile of samples of MWCNTs were calculated using the Dorris-Gray method [21] over different surface coverages:   2 CH C,1C, CH d s 22 /ln 4 1             aN VVRT nwnw   (2) When plotting RTln(Vw,nC) against carbon number, nC, for the n-alkane probes, a straight line is generated from the gradient from which the dispersive free energy of the sample surfaces of the MWCNTs, d s , can be calcu- lated. The specific (Lewis acid-base) surface energy ab s of samples of MWCNTs was calculated from the basic component (  s ) and the acidic component (  s ) of the surface energy:  ss ab s 2  (3) The basic and acidic components of the surface energy were obtained from the specific parts of free enthalpy changes of adsorption ab ,ads iG of polar probes i: Figure 1. Structure of the olefin-maleic-anhydride- ester-amide copolymer (OMAEA) coupling agent, where R1: alkyl chain with length of the olefinic mon- omer; R2: alkyl chain with R1–2 carbon number; a, b: 2–21; k: 0.2–2; l: 1–7; m: 1–7 and n: 0.3–2 [11] SURFACE ENERGY HETEROGENEITY PROFILES OF CARBON NANOTUBES 45(1) pp. 61–66 (2017) 63        ss ab ,ads 2  iiii aNG (4) applying the van Oss-Chaudhury-Good theory [22] with the Della Volpe scale [23]. The specific free energy changes of adsorption of the polar probes were obtained as suggested by Donnet et al. [24]. 3. Results and Analysis 3.1. Experiments The dispersive surface energy profiles of the untreated samples of MWCNTs and those treated with the olefin- maleic-anhydride-ester-amide copolymer (OMAEA) coupling agent compatibilized at 353 K and over low surface coverages (n/nm) are presented in Fig.2. The energy profiles show reasonable devaluation of disper- sive surface energy for MWCNTs after surface modifi- cation detected by n-alkane molecular probes: the dis- persive surface energy ( d s ) of the MWCNTs decreased to half of its initial value. The untreated samples of MWCNTs exhibited dis- persive surface energies of ~110 mJ/m 2 at 353 K, which is comparable to the values reported by other research- ers studying carbon nanotubes [13-15] and graphitic carbon materials [25]. The relatively high values of the dispersive surface energy of untreated MWCNTs can be attributed to a strong nonpolar interaction potential to build physical long-range Keesom, Debye, and London attractions, which explains their high tendency to ag- glomerate [14]. However, the anchoring of olefin- maleic-anhydride-ester-amide (OMAEA) up on the MWCNTs surface caused a marked decrement in dis- persive part of surface energy from ~110 mJ/m 2 to ~48 mJ/m 2 at 353 K. The large drop in the value of d s of surface-treated MWCNTs shows that the dispersive surface energy of MWCNTs has been obviously altered by the coupling agent. The surface treatment also affected the dispersive surface energy heterogeneity profile of the MWCNTs. The surface energy mapping of the samples of MWCNTs indicated that the dispersive components of surface energies of untreated samples of MWCNTs are almost constant within the region of low surface cover- age. Consequently, the surface of the untreated MWCNTs can be considered quasi-homogeneous. However, the dispersive surface energy heterogeneity profiles of the treated MWCNTs prove that the copoly- mer-modified MWCNT surface is energetically slightly heterogeneous, because the dependence of d s on sur- face coverage is relatively strong within the region of low surface coverage. In addition, the distributions of the dispersive sur- face energies (Fig.3) obtained by point-by-point integra- tion of dispersive surface energy profiles over the inves- tigated range of the surface coverage support in a more illustrative manner also results in an increase in the dis- persive surface energy heterogeneity. The dispersive surface energy probability function of the MWCNTs became more spread out after modification of the sur- face indicated a greater degree of energetic surface in- homogeneity. The specific surface energy ( ab s ) profiles of the untreated samples of MWCNTs and those treated with OMAEA compatibilized at 353 K and over various sur- face coverages (n/nm) are presented in Fig. 4. The un- treated samples of MWCNTs possess specific surface energy of ~10 mJ/m 2 at 353 K, which value is near to that given by Lou et al. (8.84 mJ/m 2 ) for pristine carbon nanotubes at 373 K and over undefined degrees of sur- face coverage [14]. The quantitative surface energy analysis obtained by iGC-SEA methodology demon- strated that surface treatment of MWCNTs resulted in Figure 3. Dispersive surface energy probability func- tions of untreated samples of MWCNTs and those treated with an olefin-maleic-anhydride-ester-amide copolymer (OMAEA) coupling agent compatibilized at 353 K (the solid correlation lines are only to im- prove visualization). Figure 2. Dispersive surface energy profiles of un- treated samples of MWCNTs and those treated with an olefin-maleic-anhydride-ester-amide copolymer (OMAEA) coupling agent compatibilized at 353 K and over various surface coverages (the dotted correla- tion lines are only to improve visualization). GERENCSÉR, RIEDER, VARGA, HANCSÓK AND DALLOS Hungarian Journal of Industry and Chemistry 64 significant changes in surface energies: the specific sur- face energy of CNT surfaces increased more than four- fold, from ~10 mJ/m 2 to ~41 mJ/m 2 . Furthermore, the dependence of ab s on surface coverage is pronounced for the compatibilized samples of MWCNTs which indicates that energetic heteroge- neity attributed to chemical heterogeneity and the exist- ence of electron donor-acceptor atomic groups on the surface. However, the quasi-constant specific compo- nent of surface energy for untreated MWCNTs suggests an energetically homogeneous surface and the absence of high specific energy surface sites. The specific surface energy values of compatibil- ized MWCNTs are much higher than those of the un- treated surfaces, declared the enhanced connection be- tween compatibilized MWCNTs and polar analytes. The observed diversity in specific surface energies is ac- complished from the adsorbed polar atomic clusters: namely the moderately electron-withdrawing maleic anhydride groups; and the electron-donating ester and amide groups with different nucleophilic or electro- philic characteristics. The specific surface energy distributions in Fig. 5 represent the heterogeneity of the samples of MWCNTs and reveal that the untreated MWCNTs exhibited ab s values which varied from 9.9 to 10.3 mJ/m 2 . This small variation in the specific surface energy demonstrated a fairly energetically homogeneous surface for untreated MWCNTs. However, as surface treatment increased the concentrations of polar clusters on the surface of MWCNTs, the modified surface exhibited great varia- tions in ab s  (from 37.3 to 42.9 mJ/m2), implying that the compatibilized MWCNTs are surface energetic het- erogeneous. The surface treatment also modified the chemical characteristics of the MWCNTs. The acid-base surface energy mapping of the samples of MWCNTs (Figs.6 and 7) indicate that both the electron-accepting and do- nating abilities of the MWCNTs were raised apprecia- bly after compatibilization of untreated MWCNTs con- sistent with the adsorption of electron-withdrawing and electron-donating atomic groups on the surface. The untreated samples of MWCNTs possess base surface energy of ~16 mJ/m 2 at 353 K, which value is near to that given by Lou et al. (12.97 mJ/m 2 ) for pristine car- bon nanotubes at 373 K and over undefined surface coverage [14]. The surface energy mapping using the iGC-SEA methodology confirms that the surface treatment of MWCNTs raised the basic component (  s ) of surface energy of MWCNTs: it resulted in a sevenfold increase from ~16 mJ/m 2 to ~112 mJ/m 2 . The acidic component of surface energy  s of untreated samples of MWCNTs was measured as ~1.7 mJ/m 2 at 353 K, which is in good agreement with that reported by Lou et al. (1.51 mJ/m 2 ) for pristine carbon nanotubes at 373 K and over unde- fined surface coverages [14]. The surface modification of MWCNTs resulted in a more than twofold increase in the value of the acidic component (  s ) of the surface energy of MWCNTs from ~1.7 mJ/m 2 to ~3.7 mJ/m 2 . However, the profiles of acid-base surface energy heterogeneity of the treated MWCNTs also prove that the copolymer-modified MWCNT surface became en- ergetically more heterogeneous, because the dependen- cies of  s and  s on surface coverage are relatively stronger than those of the untreated MWCNTs. The presence of non equi-energetic active surface centers exposes that the surfaces of the compatibilized MWCNTs are not energetically uniform for specific acid-base interactions and the surface treatment consid- erably altered the ability of MWCNTs to connect with molecular species by specific interactions. Figure 5. Specific surface energy probability functions of untreated samples of MWCNTs and those treated with an olefin-maleic-anhydride-ester-amide copoly- mer (OMAEA) coupling agent compatibilized at 353 K (the solid correlation lines only improve visualiza- tion). Figure 4. Specific surface energy profiles of untreated samples of MWCNTs and those treated with an olefin- maleic-anhydride-ester-amide copolymer (OMAEA) coupling agent compatibilized at 353 K and over vari- ous surface coverages (the dotted correlation lines on- ly improve visualization). SURFACE ENERGY HETEROGENEITY PROFILES OF CARBON NANOTUBES 45(1) pp. 61–66 (2017) 65 The larger change in the base component of the surface energy ( s ) after compatibilization with a pol- yalkenyl-poly-maleic-anhydride-ester-amide additive indicates a higher concentration of electron-donating ester and amide groups on the surface. The large value of s (relative to  s ) implies a more basic characteris- tic and donor properties of the surface of the treated MWCNTs. 4. Conclusion The experimental data demonstrated that iGC is a useful methodology of characterizing the variation in the sur- face characteristics of MWCNTs after non-covalent functionalization. The exclusive energy scaling of the SEA methodology using energy heterogeneity charts with surface energy probability density functions over wide surface coverages presents profitable additional information on the differences in terms of the nature, homogeneity and heterogeneity of surface energies re- sulting from surface transformations. The multilateral surface energy analysis of SEA presents a quantitative control of the effectiveness of surface treatment and demonstrates the importance of the dependence of sur- face energy analysis on coverage. Acknowledgement The present work was published within the framework of the project GINOP-2.3.2-15-2016-00053. SYMBOLS 2CH a , ai cross sectional area of an adsorbed methylene group and of probe i ab ,ads iG specific free enthalpy of adsorption msp mass of the adsorbent in the column N Avogadro’s number R gas constant T temperature ' Rt adjusted retention time cV  mean flow rate of the carrier gas Vw specific retention volume ab s , d s specific and dispersive parts of the surface energy of solid sample materi- al 2CH  surface energy of a methylene group  s ,  s acid-base components of the surface energy of solid sample material  i  ,  i  acid-base components of the surface tension of polar liquid probe i Θ=n/nm surface coverage REFERENCES [1] Dresselhaus, M.S.; Dresselhaus, G.; Avouris, P.: Carbon nanotubes: synthesis, structure, properties, and applications (Springer-Verlag GmbH, Berlin, Germany) 2001 DOI: 10.1007/3-540-39947-X Figure 6. Basic component profiles of surface energy of untreated samples of MWCNTs and those treated with an olefin-maleic-anhydride-ester-amide copoly- mer (OMAEA) coupling agent compatibilized at 353 K and over various surface coverages (the dotted cor- relation lines only improve visualization). Figure 7. Acidic component profiles of surface energy of untreated samples of MWCNTs and those treated with an olefin-maleic-anhydride-ester-amide copoly- mer (OMAEA) coupling agent compatibilized at 353 K and over various surface coverages (the dotted cor- relation lines only improve visualization). GERENCSÉR, RIEDER, VARGA, HANCSÓK AND DALLOS Hungarian Journal of Industry and Chemistry 66 [2] Yan, Y.; Miao, J.; Yang, Z.; Xiao, F.X.; Yang, H.B.; Liu, B.; Yang, Y.: Carbon nanotube cata- lysts: recent advances in synthesis, characterization and applications, Chem. Soc. Rev., 2015 44, 3295- 3346 DOI: 10.1039/C4CS00492B [3] Sági, D.; Holló, A.; Varga, G.; Hancsók, J.: Co- hydrogenation of fatty acid by-products and differ- ent gas oil fractions, Journal of Cleaner Produc- tion, 2017 161, 1352-1359 DOI: 10.1016/j.jclepro.2017.05.081 [4] Li, C.; Shi, B.; Cui, M.; Shang, H.J.; Que, G.H.: Application of Co-Mo/CNT catalyst in hydro- cracking of Gudao vacuum residue, J. Fuel. Chem. Technol., 2007 35(4), 407−411 DOI: 10.1016/S1872-5813(07)60026-7 [5] Tavasoli, A.; Sadagiani, K.; Khorashe, F.; Seifkor- di, A.A.; Rohani, A.A.; Nakhaeipour, A.: Cobalt supported on carbon nanotubes — A promising novel Fischer–Tropsch synthesis catalyst, Fuel Process. Technol., 2008 89(5), 491–498 DOI: 10.1016/j.fuproc.2007.09.008 [6] Chen, X.; Deng, D.; Pan, X.; Bao, X.: Iron catalyst encapsulated in carbon nanotubes for CO hydro- genation to light olefins, Chinese J. Catal., 2015 36(9), 1631–1637 DOI: 10.1016/S1872-2067(15)60882-8 [7] Chen, Y.; Mu, R.; Yang, M.; Fang, L.; Wu, Y.; Wu, K.; Liu, Y.; Gong, J.: Catalytic hydrothermal liquefaction for bio-oil production over CNTs sup- ported metal catalysts, Chem. Eng. Sci., 2017 161, 299–307 DOI: 10.1016/j.ces.2016.12.010 [8] Varga, Cs.; Bartha, L.: A novel route for injection moulding of long carbon fibre-reinforced LLDPE, J. Reinf. Plast. Comp., 2014 33(20), 1902-1910 DOI: 10.1177/0731684414549443 [9] Chen, P.; Chew, L.M.; Xia, W.: The influence of the residual growth catalyst in functionalized car- bon nanotubes on supported Pt nanoparticles ap- plied in selective olefin hydrogenation, J. Catal., 2013 307, 84–93 DOI: 10.1016/j.jcat.2013.06.030 [10] Speltini, A.; Merli, D.; Quartarone, E.; Profumo, A.: Separation of alkanes and aromatic compounds by packed column gas chromatography using func- tionalized multi-walled carbon nanotubes as sta- tionary phases, J. Chromatogr. A, 2010 1217(17), 2918–2924 DOI: 10.1016/j.chroma.2010.02.052 [11] Szentes, A.; Varga, Cs.; Horváth, G.; Bartha, L.; Kónya, Z.; Haspel, H.; Szél, J.; Kukovecz, Á.: Electrical resistivity and thermal properties of compatibilized multi-walled carbon nano- tube/polypropylene composites, Express Polym. Lett., 2012 6(6), 494–502 DOI: 10.3144/expresspolymlett.2012.52 [12] Varga, Cs.; Miskolczi, N.; Bartha, L.; Lipóczi, G.; Falussy, L.: Improving the compatibility of man- made fibre reinforced composites, Hun. J. Ind. Chem., 2008 36(1-2), 137-142 [13] Zhang, X.; Yang, D.; Xu, P.; Wang, C.; Du, Q.: Characterizing the surface properties of carbon nanotubes by inverse gas chromatography, J. Ma- ter. Sci., 2007 42(17), 7069–7075 DOI: 10.1007/s10853-007-1536-7 [14] Luo, Y.; Zhao, Y.; Cai, J.; Duan, Y.; Du, S.: Effect of amino-functionalization on the interfacial adhe- sion of multi-walled carbon nanotubes/epoxy nanocomposites, Materials and Design, 2012 33, 405–412 DOI: 10.1016/j.matdes.2011.04.033 [15] Menzel, R.; Tran, M.Q.; Menner, A.; Kay, C.W.M.; Bismarck, A.; Shaffer, M.S.P.: A versa- tile, solvent-free methodology for the functionali- sation of carbon nanotubes, Chem. Sci., 2010 1(5), 603-608 DOI: 10.1039/C0SC00287A [16] Menzel, R.; Lee, A.; Bismarck, A.; Shaffer, M.S.P.: Inverse gas chromatography of as-received and modified carbon nanotubes, Langmuir, 2009 25(14), 8340–8348 DOI: 10.1021/la9000607s [17] Díaz, E.; Ordóñez, S.; Vega, A.: Characterization of nanocarbons (nanotubes and nanofibers) by in- verse gas chromatography, J. Phys. Conf. Ser., 2017 61(1), 904-908 DOI: 10.1088/1742-6596/61/1/180 [18] Menzel, R.; Lee, A.; Bismarck, A.; Shaffer, M.S.P.: Deconvolution of the structural and chem- ical surface properties of carbon nanotubes by in- verse gas chromatography, Carbon, 2012 50(10), 3416–3421 DOI: 10.1016/j.carbon.2012.02.094 [19] Szentes, A.; Horváth, G.: Role of catalyst support in the growth of multi-walled carbon nanotubes, Hun. J. Ind. Chem., 2008 36(1-2), 113–117 [20] Kondor, A.; Dallos, A.: Adsorption isotherms of some alkyl aromatic hydrocarbons and surface en- ergies on partially dealuminated Y faujasite zeolite by inverse gas chromatography, J. Chromatogr. A, 2014 1362, 250-261 DOI: 10.1016/j.chroma.2014.08.047 [21] Dorris, G.M.; Gray, D.G.: Adsorption of n-alkanes at zero surface coverage on cellulose paper and wood fibers, J. Colloid Interf. Sci., 1980 77(2), 353-362 DOI: 10.1016/0021-9797(80)90304-5 [22] van Oss, C.J.; Good, R.; Chaudhury, M.K.: Addi- tive and nonadditive surface tension components and the interpretation of contact angles, Langmuir, 1988 4(4), 884-891 DOI: 10.1021/la00082a018 [23] Della Volpe, C.; Sibioni, S.: Some reflection on acid-base solid surface free energy theories, J. Col- loid Interf. Sci., 1997 195(1), 121-136 DOI: 10.1006/jcis.1997.5124 [24] Donnet, J.B.; Park, S.J.; Balard, H.: Evaluation of specific interactions of solid surfaces by inverse gas chromatography. A new approach based on po- larizability of the probes, Chromatographia, 1991 31, 435-440 DOI: 10.1007/BF02262385 [25] Donnet, J.B.; Park, S.J.: Surface characteristics of pitch-based carbon fibers by inverse gas chroma- tography method, Carbon, 1991 29(7), 955-961 DOI: 10.1016/0008-6223(91)90174-H