Open Chem., 2015; 13: 893–900 Invited Paper Open Access Adam Voelkel*, Beata Strzemiecka, Kasylda Milczewska, Zuzanna Okulus Inverse Gas Chromatographic Examination of Polymer Composites DOI: 10.1515/chem-2015-0104 received December 30, 2014; accepted April 1, 2015. of composite components and/or interactions between them, and behavior during technological processes. This paper reviews is the examination of Abstract: Inverse gas chromatographic characterization various polymer-containing systems by inverse gas of resins and resin based abrasive materials, polymer- chromatography. polymer and polymer-filler systems, as well as dental restoratives is reviewed. Keywords: surface activity, polymer-polymer interactions, 2 Discussion adhesion, dental restoratives, inverse gas chromatography 2.1 Surface energy and adhesion 1 Introduction IGC is useful for surface energy determination of solid polymers and fillers. Solid surface energy of consists of Inverse gas chromatography (IGC) was introduced in 1967 dispersive ( , from van der Waals forces) and specific by Kiselev [1], developed by Smidsrød and Guillet [2], and ( , from acid-base interactions) components: is still being improved. Its popularity is due to its simplicity D and user friendliness. Only a standard gas chromatograph = + (1) is necessary [3] although more sophisticated equipment has been advised. “Inverse” relates to the aim of the experiment. for solids can be calculated according to several It is not separation as in classical GC, but examination of methods [7]; one of the most often used is that of Schultz- the stationary phase properties. Test compounds with Lavielle [8-12]: known properties are injected onto the column containing dd the material to be examined. Retention times and peak RT××ln VN =× 2 N ×× aggsl × + C (2) profiles determine parameters describing the column filling. IGC makes possible polymer and composite study where: at different temperatures and humidities [4-6]. Polymer R = gas constant, 8.314 [J mol-1 K-1]; examination below the glass transition temperature (Tg) T = temperature [K]; 3 allows surface characterization, which can then be used VN = net retention volume [m ]; to monitor surface changes occurring during chemical N = Avogadro’s number, 6.023 × 1023 [1 mol-1]; modification. For liquids, IGC can determine the Flory- a = adsorbate cross sectional area [m2]; -2 Huggins interaction ( and/or ), solubility (δ2), and = surface free energy dispersive component [mJ m ]; three-dimensional Hansen solubility parameters. These = liquid probe surface tension dispersive component allow prediction of polymer mutual solubility, miscibility [mJ m‑2]; C = constant. Retention data for polar and non-polar test compounds *Corresponding author: Adam Voelkel: Chemical Technology and Engineering, Poznań University of Technology, ul. Berdychowo 4 are necessary to quantify the surface acidic and basic properties ( ) according to the Good-van Oss 60-965 Poznań, Poland, E-mail: [email protected] Beata Strzemiecka, Kasylda Milczewska, Zuzanna Okulus: concept [13]: Institute of Chemical Technology and Engineering, Poznań University of Technology, ul. Berdychowo 4 60-965 Poznań, Poland (3) © 2015 Adam Voelkel et al., licensee De Gruyter Open. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License. 894 Adam Voelkel et al. are the electron acceptor and donor parameters from difficulty in homogeneously packing the powder sp of the probe molecules. ∆G is the specific component of into tubes [17]. the polar compound’s Gibbs free energy of adsorption. IGC seems a much better method for studying solid Its determination has been described [7-9]. powders, especially for such solids as very hard 1–0.25 mm Dichloromethane (DM) and ethyl acetate (EA) can be abrasive grains [22]. It is impossible to form a disk and used as test compounds to determine and . DM the Washburn method might be inaccurate. Indirect is a monopolar acid with of 0.0 mJ m-2. Eq. (3) then estimation of the contact angle is impossible due to the reduces to: grains’ irregular shape and small size. IGC made it possible to determine their surface free energy. The authors of this (4) review have also used IGC to characterize carbon black powders [23]. Very often insufficient material is available has been reported to be 5.2 mJ m-2 [14]. Similarly, EA for contact angle measurement. IGC requires only several is a monopolar base and is 0.0 mJ m-2. Thus, for the mg. solid can be calculated: The surface energy of connected solids determines their strength (work) of adhesion, , resulting from (5) dispersive (van der Waals) and acid-base interactions. is 19.2 mJ m-2 [14]. However, in the literature there are (6) different values for test compound surface free energy components and parameters [15]. Van Oss gave of is the dispersive component and is the acid-base 6.2 mJ m-2 for EA, which differs from that in [16]. Moreover, component. Eq. (6) can be used to determine the work of for dichloromethane (DM) is not given therein, but only adhesion in solid polymer-filler system such as abrasive- that for chloroform (CH; 1.5 mJ m-2) [16]. hardened resin binder [22,24] or polyurethane-carbon We have calculated for the materials black [25]. can be calculated by [26-29]: studied here using from both these sources. The sensitivity of and to the (7) assumed has been discussed [17]. The values calculated from van Oss’ data [16] denote the dispersive components of the filler (f) were three times higher than those calculated according and resin (p) free surface energies. is the component to data from [14]. The values calculated from van due to acid-base interactions: Oss’ data [16] were from 1/3 to 1/5 those calculated from data in [14]. However, the trends in both parameters were (8) independent of which was used. The were similar as well as the and . Parameters estimated by IGC should not be treated as absolute, but they are very useful are the acidic and basic parameters for materials comparisons. of the polymer surface (p) describing its electron acceptor There are several methods for solid surface energy and electron donor abilities. denote the same determination; contact angle measurement is most often characteristics of the filler (f). used [17]. The use of different polar and nonpolar Eq. (6) does not include the polar component of liquids enables determination of the Lifshitz-van der the work of adhesion due to dipole-dipole and induced Waals ( ) and the Lewis acid-base ( ) components dipole-dipole interactions. Fowkes has demonstrated that of . Moreover, the electron-acceptor ( ) and electron- this contribution is negligible [26,27]. donor ( ) parameters of the acid-base component can be IGC also allows assessment of the Wa/Wcoh ratio for a determined if the components of the liquid surface free filler dispersion in a polymer matrix [24]. Wcoh, the filler energy are known. work of cohesion, is calculated as a sum of dispersive However, this method suffers from some limitations, ( ) and specific ( ) components in the same way as e. g. the diameter of the liquid drop affects the results the work of adhesion [29]: [17] and the solid surface must be smooth [18,19]. The solids are frequently powders and it is impossible to (9) prepare a flat and smooth surface [20]. The Washburn method [21] of powder surface energy estimation suffers (10) Inverse Gas Chromatographic Examination of Polymer Composites 895 If the ratio of Wa/Wcoh is close to 1 the filler cohesion forces polymer, polymer-filler, filler-filler), expressed by the and filler-polymer adhesion forces are in balance. Flory-Huggins parameter [33,34]: (13) 2.2 Polymer-polymer interactions Here, the second subscript of Vg identifies the nature of The usefulness of IGC for determining polymer–small the column. molecule interactions is well established. The Flory- To obtain for a polymer blend or composition Huggins interaction parameter is a measure of the utilizing IGC, values for all components must be free energy of interaction between the probe and the known. Therefore, three columns are usually prepared: material tested. Experimental retention parameters (e.g. two containing the single components and the third Vg – specific retention volume) can be converted to Flory– containing their composite. A further three columns Huggins parameters [30]: containing different composites can also be prepared if the effect of composite proportions is to be examined. These (11) columns should be tested using identical temperature, flow rate, inlet pressure, and test solutes [35]. The usual where: 1 denotes the solute and 2 or 3 denotes polymer sign convention is assumed; i.e., a large positive value o or filler, M1 the molecular weight of the solute, p1 the indicates unfavourable interaction, a low value indicates saturated solute vapour pressure, B11 the solute second favourable interaction, while a negative value indicates a virial coefficient, Vi the molar volume, ρi the density, R the rather strong specific interaction. gas constant, and Vg the specific retention volume. The effects of three variables on the polymer/filler For a filled polymer, Eq. (11) can be rearranged interaction parameter were examined: [31,32]: – the type of filler – the amount of filler – the test solute. (12) Although not predicted by the theory, values commonly where φ2 and φ3 are the polymer and filler volume fractions depend on the solute chemical structure [36]. This has and m refers to the composite. been interpreted as due to preferential interactions of the Inverse gas chromatography can also be used to test solute with one of the components. characterize composite component interactions (polymer- Figure 1: Filler influence on χ23 (PE2 – polyethylene; B – fillers: silica modified with: N-2-aminoethyl-3-aminopropyltrimethoxysilane (B2), 3-aminopropyltriethoxysilane (B3), 3-mercaptopropyl-trimethoxysilane (B4), n-octyltriethoxysilane (B5). 896 Adam Voelkel et al. Figure 2: Dependence of (chloroform) on the amount of modifier in filler at 363K (I, II, III, IV, V denotes 1, 2, 3, 5, 10% of modifier in filler B2).