Elastomer plasma polymerization Review - Characterization of … · 2018. 9. 11. · nished elastomer composites [11-15], elastomer surfaces modified by gas-phase fluorination [15-16]

PRÜFEN UND MESSEN TESTING AND MEASURING

19KGK · 09 2018www.kgk-rubberpoint.de

Elastomer plasma polymerization contact angle modified Wilhelmy ba-lance technique chemical hetero-geneity surface roughness contact angle hysteresis

This article features the current state of research with respect to plasma and varnished elastomer composites, which have been characterized by the sessile drop and the modified Wilhelmy balance technique. The sessile drop measurement is well suited to assess smooth and plasma modified surfaces of elastomers qualitatively. However, rough samples should be analyzed by other methods such as modified Wil-helmy balance technique having several advantages compared to the sessile drop method, e. g. high accuracy and reproducibility. It is shown that contact angle hysteresis values correlate well with roughness parameters that are ob-tained by white light interferometry.

Übersicht - Charakterisierung von modifizierten Elastomer-oberflächen mittels Benet-zung Elastomer Plasmapolymerisation Kontaktwinkel modifizierte Wilhelmy-Technik chemische Heterogenität Rauheit Kontaktwinkelhysterese

In diesem Artikel wird der aktuelle Stand der Forschung zu plasmabehan-delten und beschichteten Elastomer-kompositen beschrieben, die durch die Methode des liegenden Tropfens und durch die modifizierte Wilhelmy-Tech-nik charakterisiert wurden. Die Metho-de des liegenden Tropfens ist gut geeig-net, um glatte und plasmabehandelte Oberflächen von Elastomeren zu unter-suchen. Dennoch sollten raue Proben mittels anderer Methoden, wie zum Beispiel der modifizierten Wilhelmy-Technik untersucht werden, die einige Vorteile gegenüber der Methode des liegenden Tropfens aufweist, z. B. hohe Genauigkeit und Reproduzierbarkeit. Es konnte gezeigt werden, dass Kon-taktwinkelhysteresewerte mit Rau-heitsparametern gut korrelieren, die aus weisslichtinterferometrischen Un-tersuchungen gewonnen wurden.

Figures and Tables: By a kind approval of the authors.

1. Introduction: importance of wetting An important example for wetting of superhydrophobic surfaces is the lotus effect that can be attributed to the sur-face microstructuring and hydrophobic properties of epicuticular waxes.[1-2] This effect is also found in other plants, e.g. lady‘s mantle (Alchemilla mollis, see Fig. 1) and nasturtium (Nasturtium of-ficinale). [3] The design and surface structures are an important factor in the development of dirt-repellent surfaces. Important technical applications of hy-drophobic surfaces are self-cleaning roofing tiles, textile surfaces, paintings [4-5], hoses, seals and profiles [6] as well as icephobic polydimethylsiloxane (PDMS) coatings for the prevention of ice accumulation on wind turbines, aircrafts and heat pumps. [7-9] Unlike artificially produced surfaces, the lotus leaf surface has the property of self-regeneration. Its functionality remains after wetting. This serves as a protection from microorgan-isms or pathogens such as fungus spores or algae growth. [10]

2. Modified elastomer surfacesMany elastomer products, for instance seals, profiles and windscreen wipers, often have undesirable surface proper-ties with respect to their application or further processing [6]. Therefore, after-treatment or modification is required in order to clean and functionalize the sur-face. In this way, it is possible to influ-ence the adhesion behavior, to change the wetting properties of the polymer surface and finally to reduce the sliding friction value, stick-slip-behaviour and wear. Avoiding wear leads to a higher longevity of the elastomer products.

It has been shown that especially var-nished elastomer composites [11-15], elastomer surfaces modified by gas-phase fluorination [15-16] and nano-particle modified elastomers [15-17] are well-suited to influence wetting and to reduce friction. There has been very little research on the plasma treatment and plasma coating of elastomer surfaces in the literature so far. In particular, the ac-tivation of the elastomer surface and the

usage of precursors (e.g. HMDSO and TEOS) to form a plasma polymer layer might be promising for reducing friction and stick-slip behavior. [18] Advantages are offered by those plasma processes, which operate at atmospheric pressure and thus require no complex and expen-sive vacuum technology. Furthermore, such atmospheric pressure plasma reac-tors can be operated continuously. This enables simple process control and en-sures an efficient surface modification process. [6, 16] There are examples for the plasma treatment of elastomer sur-faces, which are described briefly in the following.

High friction coefficients can be evi-dently reduced by plasma polymeriza-tion. Furthermore, it was shown that stick-slip behavior has vanished at high velocities. [18] Plasma polymerization with vinyl compounds on nitrile rubber alters the surface and reduces friction compared to untreated elastomers. [19] Martinez et al. have investigated the in-fluence of different plasma treatments on the wettability of NBR and HNBR by means of sessile drop and XPS experi-ments. [20] Targeted chemical function-alization by plasma treatment on elasto-

Review - Characterization of modified Elastomer Surfaces by wetting

AuthorsC. W. Karl, W. Rahimi, A. Lang, U. Giese, M. Klüppel, H. Geisler, Hannover, Germany

Corresponding Author: Dr. Christian KarlInstitute of Structural AnalysisFaculty of Civil Engineering and Geodetic ScienceLeibniz Universität Hannover/ForWindAppelstraße 9A30167 HanoverE-Mail: [email protected]


20 KGK · 09 2018 www.kgk-rubberpoint.de

mers leads to an improvement of the wettability and the bonding by polar substances implies increased adhesive-ness. [6] For instance, the modification of natural rubber by plasma treatment leads to a concomitant decrease in sur-face roughness. [21] Deposited acetylene films on natural rubber by plasma po-lymerization are suitable primers to al-low a rubber-to-steel compound. [22] Romero-Sanchez et al. have shown that the ethylene glycol contact angles on SBR surfaces were reduced by corona treatment improving the wettability of the elastomer surface. [23] The release force using a PUR adhesive could be in-creased concerning the bonding of two EPDM sheets. [6] The permeation proper-ties of coated elastomers can be opti-mized, whereby the passage time could be delayed by 50 %. [6]

Plasma-coated silicone rubbers over-come the disadvantages of contact lens-es made of silicone rubber and do not affect the properties, such as flexibility of the lens; in this case the hydrophobic rubber is hydrophilized. [24] Recently, it has been shown that highly functional-

ized black silicone/elastomer composite surfaces exhibit either the lotus effect or the rose petal effect as a function of the mechanical strain in order to manipulate water droplets. The stretched or relaxed surface reveals extreme superhydropho-bicity having a low sliding angle (see lo-tus effect). Thus, the apparent contact angle of a droplet on the composite sur-face can be adjusted between 170° and 110° upon mechanical stretching. This enables a wide range of hydrophobic surface wettabilities. [25]

3. Wetting: fundamentals and terminologyContact angle measurements are carried out in order to characterize the wettabil-ity of solid technical surfaces. Due to its simplicity, relatively low-cost instru-ments and a long history, this method appears to be easy in its application. De-spite the long experience of contact an-gle investigations, the measurement and interpretation suffers from insufficient experimental and theoretical protocols. Furthermore, the misuses of wetting ter-minology lead to misinterpretation of

Fig. 1: Picture of the leaf surface of lady’s mantle (by C. W. Karl).

1 experimental contact angles. Contact an-gle measurements are the most surface sensitive of any common analysis tech-nique having an analysis depth of ca. 0.5-1 nm. [26-28] Hence, contact angles are sensitive to many factors, e. g. sur-face geometry, roughness, contamina-tion and deformation of polymer surfac-es. [29-30] As a result, contact angle data are lacking for common materials which can serve as reference source. Contact angles which are measured for an ob-served system cannot be regarded as unique values. [30-31] The use of the contact angle concept in different fields (e. g. mechanical engineering, material science and electrical engineering) and the miniaturisation of various systems have led to confusing results. [29-31]

In the following common terms for different types of surfaces and contact angles used in wetting science are de-scribed. [29, 32]

Ideal surface – A flat, perfectly smooth, rigid and chemically homogeneous sur-face which does not chemically interact with the probe liquid. It shows zero con-tact angle hysteresis. In practice, an ideal surface is rarely encountered.

Real surface – A solid surface that is not ideal, regardless of type(s) of its de-viation from the ideality of surface. The majority of surfaces that are used and tested are real surfaces.

Contact angle – Experimentally ob-served angle on the liquid side between the tangent to the solid surface and the tangent to the liquid-gaseous interface at the contact line among the three phases. The point of intersection be-tween tangent lines is denoted as the contact point (see Fig. 2).

Young contact angle – This contact angle is a thermodynamic property of the three-phase system that corresponds to the lowest state of the energy for the system. In the year 1805 Young was able to show that the contour of a liquid drop-let on an ideal solid, if one neglects its rheological properties, is determined only by the gravitational force and the interfacial tension. [33] If the droplet in-teracts with a solid, surface energy of the solid determines the contact angle of the drop. In the presence of a fluid phase in the form of a droplet, a pressure differ-ence exists between the droplet interior and the gas phase, which can be quanti-fied by Young-Laplace‘s equation. [34] The Young equation is (see Eq. 1): 𝑐𝑐𝑐𝑐𝑐𝑐 𝜃𝜃𝑌𝑌 =

𝛾𝛾𝑆𝑆𝑆𝑆 − 𝛾𝛾𝑆𝑆𝑆𝑆𝛾𝛾𝑆𝑆𝑆𝑆

(1)

Fig. 2: Water drop on a steel surface with the apparent contact angle θ. The letters S, L and V indicate the phases of the solid, the liquid and the gas phase.

2



The letters S, L and V indicate the phases of the solid, the liquid and the gas phase, respectively. However, it is assumed that the drop has an ideal size [36]. In the case of small drops, the line tension may have to be considered since the contact angle increases as the ratio of length to surface of the droplet increases. This effect is negligible for drops larger than 1 μm. [37]

The interfacial tension can be consi-dered as the reversible work, which is re-quired under isothermal and isobaric conditions to increase the surface bound-ary by one cm2. For following unit is used: [𝛾𝛾] =

𝐽𝐽𝑚𝑚2

= 𝑁𝑁𝑚𝑚

.

For liquids, N/m whereas for solids or polymers J/m2 is used. The order of mag-nitude of the surface tension of liquids is in the range between 10 and 100 mN/m. [38] The surface energies for untreated polymers have values ranging from ca. 16 J/m2 (PTFE) to ca. 41 J/m2 (polymaide 66). Higher values can be achieved by chemi-cal modification. For high-energy surfaces such as metals, values in the range of 100 to 5000 J/m2 are observed. [38]

Ideal contact angle – The ideal contact angle is the contact angle on an ideal surface. For drops with radii of curvature larger than nanometric size, this contact angle equals the Young contact angle and represents the single equilibrium state that a drop may have on an ideal solid. For drops with smaller radii of cur-vature, a correction term that is related to the concept of ‘line tension’ needs to be applied. [29] The ideal contact angle is not accessible experimentally due to dif-ficulties associated with the preparation of an ideal solid surface. Furthermore it is difficult to maintain equilibrium condi-tions during measurements. [39-42]

Actual or local contact angle – The contact angle existing locally at any point along the contact line is denoted as actual contact angle. It was theoretically proven that the actual contact angle is equal to the ideal contact angle at equili-brium. [43] However, the actual or local contact angle cannot be easily measu-red. Bhattarai et al. have introduced the surface roughness of a composite inter-face via an array of spherical nanopartic-les with controlled wettability. They have found that the liquid interface is flat and its location relative to the solid substrate is determined by the particle size and the local contact angle. [38]

Apparent or global contact angle – The contact angle measured experimentally

on the macroscopic scale is the only one that can be routinely measured. It is ter-med the apparent contact angle. This contact angle describes an ‘average’ con-tact angle for the entire three-phase contact line. There are a few types of ap-parent contact angles that are related in different ways to the Young contact ang-le, depending on the specific nature of the real solid surface. Apparent contact angles may represent either metastable equilibrium states of the system (local minimum in the Gibbs energy curve) or a stable equilibrium state (lowest mini-mum in the Gibbs energy curve, see Fig. 3). [36] The various kinds of the apparent contact angles are described in the fol-lowing.Most stable contact angle – The apparent contact angle associated with the state of the lowest Gibbs energy for a system representing the global minimum (see Figure 3) is denoted as most stable con-tact angle. In a few studies, the vibration of the solid surface has been applied by Della Volpe et al. [44] However, this con-tact angle has not been measured ex-actly so far. The most stable apparent (measured) contact angle on a rough but chemically homogeneous solid surface is predicted by the Wenzel equation. [45] This equation is valid if the liquid com-pletely penetrates into the grooves of a rough surface and if the drop is suffi-ciently large compared to the roughness scale of the surface. Additionally, the drop must be axissymmetric. [46] The most stable contact angle on a chemi-cally heterogeneous but smooth solid surface is predicted by the Cassie equa-tion [47], if the drop is sufficiently large compared to the heterogeneity scale [48]. An extended version of the Wenzel

and Cassie equations exist for cases that deal with wetting of a rough surface, when the liquid incompletely penetrates (or does not penetrate at all) into the roughness grooves. As such, the surface is ‘seen’ by the liquid as a rough and het-erogeneous surface that is composed of the solid and ‘air’. This case is covered by the Cassie-Baxter equation. [47]

Advancing contact angle – The highest metastable apparent contact angle that can be measured (e.g. by increasing the volume of the drop, see Fig. 3).

Receding contact angle – The lowest metastable contact angle that can be measured (e.g. by decreasing the volume of the drop, see Fig. 3). The values of ad-vancing and receding contact angles may vary depending on the method of mea-surement [49] and on the parameters of the system. Their measurements with a sessile-drop method are described in de-tail [50]. For example, in the case of a system that consists of a drop on a solid surface, these contact angles may de-pend on the drop volume. [51-52] Re-gardless of the experimental method, the values of advancing and receding contact angles may be affected by ran-dom noise – for example, vibrations of a building. [53]

Contact angle hysteresis – The diffe-rence of advancing and receding contact angle is the contact angle hysteresis which cannot be correlated with the ide-al contact angle based on current know-ledge. For example, using the Wenzel equation for the advancing or receding angle is misleading. However, the mea-surement of these angles is extremely important since contact angle hysteresis is a measure of the non-ideality of the solid surface. Also, contact angle hystere-

Fig. 3: Simplified schematic repre-sentation of the Gibbs energy of a liquid droplet on an ideal and real (rough) sur-face. [29] Energy barriers are loca-ted between the multiple local minima.

3



sis may be thought of as a rough indica-tion of the degree of drop adhesion onto a solid surface, since in many cases the larger the contact angle hysteresis, the stronger is drop adhesion. It was de-monstrated that nanometric holes in a monolayer can be detected through measurements of contact angle hystere-sis. [54] In addition, a common miscon-ception needs to be clarified. It is quite common to call the advancing and rece-ding contact angles ‘dynamic’. It is true that the transition between metastable contact angles is indeed a dynamic pro-cess; however, these contact angles by themselves represent equilibrium states (although they are metastable).

The dynamic contact angle is a con-tact angle measured under dynamic flow conditions that is affected by the velocity of the flow due to viscous effects. Thus, it cannot characterise the equilibrium wettability. The deviation of the dynamic contact angle from the equilibrium con-tact angles depends on the capillary number. [55]

Static (sessile) contact angle – This is a term that refers to the measurement of the contact angle when the drop lands on the solid surface. To determine con-tact angles, the elastomer surface has to be prepared properly and pure liquids have to be used. There are also other factors which are important. The results depend on the experience of the opera-tor. Furthermore, the measurement of a single static contact angle is not really adequate, because there is no single equilibrium contact angle. [55] As far as the static contact angle measurement is concerned, inhomogeneities and chan-ges in wetting over time can be detected. [37]

Smooth surfaceHydrophilic surface – This surface is char-acterised by the Young contact angle for water being smaller than 90 ° . [44]

Hydrophobic surface – This surface is characterised by the Young contact angle for water that is larger than or equal to 90 °. [44]

Rough surfaceIn practice, however, only very rarely smooth and chemically homogeneous surfaces are encountered, which are nec-essary in order to determine the contact angle according to Young for ideal sur-faces. Contact angles of real surfaces are described on the one hand by the rough-ness of the surface, which can be ex-

plained by means of a Wenzel model (see Eq. 2) or the Wenzel factor. [45] The Wen-zel factor r denotes the roughness of the surface. It indicates the ratio of the ac-tual surface to the corresponding geo-metrically projected surface. For r = 1 the Young equation (ideal surface) is ob-tained. It represents the apparent, i.e., the macroscopically visible and meas-ured contact angle, which is referred to the actual contact angle, i.e. the local contact angle due to the roughness, but which is not accessible experimentally. [56]

𝑟𝑟 =𝑐𝑐𝑐𝑐𝑐𝑐 𝜃𝜃𝑊𝑊𝑐𝑐𝑐𝑐𝑐𝑐 𝜃𝜃𝑌𝑌

(2)

Kamusewitz et al. derived a relationship between the contact angle hysteresis and Wenzel’s contact angle [67, 98]. If the droplet is larger than the roughness scale by two to three orders of magni-tude, the Wenzel equation applies, as shown by Marmur. [57]

Whereas for the Wenzel equation the state of homogeneous wetting, in which the wetting liquid penetrates completely into the depressions of the rough sur-face, the state in which air cavities are present in the depressions of the topo-graphy is referred to heterogeneous wet-ting. In this case, due to the surface ten-sion of the solvent and the roughness, the surface cannot be fully wetted. In the case of heterogeneous wetting of a rough and chemically homogeneous sur-face, the following equation was formu-lated by Cassie and Baxter (see Eq. 3), [47]:

𝑐𝑐𝑐𝑐𝑐𝑐 𝜃𝜃𝐶𝐶𝐶𝐶 = 𝑓𝑓𝑏𝑏𝑟𝑟𝑏𝑏𝑐𝑐𝑐𝑐𝑐𝑐 𝜃𝜃𝑌𝑌 − (1 − 𝑓𝑓𝑏𝑏) (3)

Considering fb is the proportion of the projected wetted surface with respect to the total projected area, then rb repre-sents the roughness factor of the wetted portion of the surface. Assuming the values for fb become very small, the most stable contact angle tends to 180 °. The resulting contact angle is greater than in the case of the Wenzel model since the interface between the two substances is smaller. These types of rough surfaces have the property of trapping air in their deeper layers, which is not displaced by the liquid during wetting showing the lotus effect. [2] Up to now, it is not fully understood how physical roughness and chemical heterogeneity affect the con-tact angle which is the decisive quantity to calculate the surface energy. [58] As shown, the knowledge of surface energy

allows estimating the compatibility of fillers with rubbers. [59] To obtain the interfacial energy per unit area (work of adhesion) of two solid surfaces in con-tact, the substrate and the coating ma-terial, the surface energies of both are needed [60]. In contrast to ideal surfac-es, real surfaces cannot be characterized by a single stable macroscopic contact angle, which is denoted as apparent con-tact angle. As a result there are different macroscopic contact angles. [32] These angles, described by metastable states (see Fig. 3), are due to the locally varying slope of the topography and thus corre-spond to multiple local minima of the free enthalpy of a liquid droplet on a solid surface (see Fig. 3). Energy barriers are present between these minima. In energetic equilibrium, where the Gibbs energy has the lowest value, the system is in the most stable state. The corre-sponding most stable macroscopic con-tact angle is called θeq. (see also Fig. 3). It is calculated from the arithmetic mean of the advance and receding contact angles. [32, 61] As a prerequisite for measuring the contact angle according to Marmur, a ratio between drop diam-eter and lateral extent of the roughness structures of at least three orders of magnitude is necessary. [32] As men-tioned earlier, two regimes of wetting exist on rough surfaces: the liquid ei-ther completely or partially penetrates into the grooves of a rough surface. In the former case, the Wenzel equation predicts that the most stable contact angle on a roughened hygrophilic sur-face is lower than the corresponding ideal contact angle. On roughened hy-grophobic surfaces, this equation pre-dicts that the most stable contact angle is higher than on the ideal one. In the case of partial penetration, the most stable contact angle may further in-crease, since part of the surface that is in contact with the liquid consists of air. This, however, may reduce hysteresis and enable easy roll-off of a drop from the solid surface. The following defini-tions help to distinguish clearly be-tween these states. More details are described in [29, 62].

Only little work has been devoted to correlate the contact angle hysteresis of a surface with roughness parameters, such as fractal dimension and height profile data. [56, 63] Only the mean roughness index has been used to cha-racterize the surface roughness of a material with respect to changing the



contact angle. [64] Different lacquers coated on elastomers were considered in order to investigate the influence of roughness and surface composition on friction and surface energy. [61] Recent studies indicate that, in addition to the-se two models, there are so-called mixed-roughness-heterogeneous sur-faces, based on 2-dimensional simulati-on calculations, from which the two borderline cases can be derived. [65, 66] Recent studies suggest that in addition to these two models of homogeneous and heterogeneous wetting, so-called mixed-roughness-heterogeneous sur-faces based on 2-dimensional simulati-on calculations exist, from which the two limiting cases can be derived. [65, 67] In contrast to ideal surfaces, real surfaces cannot be characterized by a single stable macroscopic contact angle, which is also referred to as the apparent contact angle (measured contact angle). Instead, there are different macroscopic contact angles. [32] These angles, which are described by metastable states are due to the locally varying slope of the topography and thus correspond to multiple local minima of the Gibbs energy of a liquid droplet on a solid sur-face (see Fig. 3). Energy barriers are pre-sent between these minima. In the energetic equilibrium (global mini-mum), where the Gibbs energy has the lowest value, the system is in the most stable state. The corresponding most stable macroscopic contact angle θeq is calculated from the arithmetic mean value of the advancing and receding contact angles which are obtaind from experimental investigations. [32, 59]

4. Work of adhesionThe contact angle between a liquid and a solid is a measure of the energetic inter-action between the solid and the liquid and hence a measure of the wetting be-havior. In order to measure the contact angle, the surface tension of the liquids used must be greater than the surface energy of the solid or the sample, respec-tively. If the surface tension of the liquids and the surface energy of the sample are the same, complete wetting takes place. This means that the contact angle be-comes equal to 0°. According to the Young equation (see Eq. 1), the interfa-cial tension is also equal to 0. In the case of complete spreading, this means com-plete wetting of the sample surface, con-sequently cos θ = 1 and θ = 0°. Spreading occurs when γsv ≥ γlv. This behavior is a

characteristic of the spreading pressure pSp (see Eq. 4):

γ𝑐𝑐𝑠𝑠 − (γ𝑐𝑐𝑠𝑠 + γ𝑠𝑠𝑠𝑠) ≡ 𝑝𝑝𝑆𝑆𝑝𝑝 > 0 (4) For contact angles θ < 30°, good to suffi-cient wetting conditions are present. Complete non-wettability is θ = 180°; but this value is not achieved in practice. The highest reached value is about 160° (superhydrophobic surfaces). [68] Adhe-sion is caused by the contact between the components. The adhesion is charac-terized by the so-called adhesion work, which is described by the Dupré equa-tion (s. Eq. 5):

W𝐴𝐴𝐴𝐴 = γ𝑐𝑐𝑠𝑠 + γ𝑠𝑠𝑠𝑠 − γ𝑐𝑐𝑠𝑠 (5)

The work of adhesion assumes the high-est value when the surface tension of both components is the same; it is neces-sary to separate two partners. The higher WAd, the higher is the adhesion. This is due to intermolecular forces along the interface. [55]

5. Contact angle hysteresis of elastomer surfacesThe contact angle hysteresis ∆θ is the dif-ference between the advancing and re-ceding contact angle. Usually the values for ∆θ vary between 5° and 20°, but there can be also significantly larger values. [34]. In the case of peroxide crosslinked natural rubber surfaces and water as wetting medium, values between 37° and 66° are found depending on the de-gree of crosslinking. The differences are even larger for water on BR (butyl rub-ber). The range is from 22° to 69°. The removal of peroxide residues implies a decrease in the contact angle hysteresis. With the same chemical environment and roughness of the elastomer surface, the following changes can be observed: as the degree of crosslinking increases, the modulus of elasticity increases, the elastomers become harder and the ad-vancing contact angle decreases, where-as the receding contact angle increases. Consequently, a decrease in the contact angle hysteresis is observed in the case of NR. Regarding the influence of the polarity of different solvents, no clear trend was observed. [69] On the basis of another model system, which consists of thin NBR films (raw rubber) on glass plates, it was shown that in the case of water as a solvent for the very smooth samples with increasing ACN content on the surface and inside (bulk) The hyster-

esis increases significantly [70]. The con-tact angle hysteresis of polymer surfaces, in particular with respect to the influ-ence of different solvents on the hyster-esis of elastomers, has not been investi-gated in detail. [69-71]

There are several reasons for the oc-currence of hysteresis. [35, 72-73] A dis-tinction is made between kinetic and thermodynamic hysteresis. [74-75] In ki-netic hysteresis, changes occur, such as swelling, on the surface of the solid. [75] Roughness and chemical heterogeneity influence thermodynamic hysteresis. [74]

The physical surface roughness has the most important influence. Whenever a drop hits an elevation, it will jump to a position where the same contact angle is reached as before. In this way, the drop is prevented from further spreading until it is so great that it can overcome the eleva-tion and spread further. The increase in contact angle hysteresis with increasing roughness is called a barrier effect. [71] Another reason is the chemical hetero-geneity of the surface.[76] When a drop-let spreads on a surface with a different environment, the so-called three-phase contact line is fixed („pinning“) by lyo-phobic, i.e. liquid-repellent areas, and the contact line of lyophilic areas is retai-ned when the droplets become smaller. A cause for the occurrence of hysteresis is the potential impurities of the measu-ring solution, which accumulate on the three-phase contact line. In this way, these substances hinder or facilitate the spreading or withdrawal of the liquid. This can be substantially reduced with the aid of solvents which do not swell the surface of the polymer. Soft surfaces, e. g. polymers, have the property that forces occur on the three-phase contact line which are so strong that the underly-ing layer is mechanically deformed. This also leads to a measurable contact angle hysteresis. [77-78] Furthermore, the ad-sorption and desorption of molecules during dissemination and withdrawal of the liquid is associated with the dissipa-tion of energy, which manifests itself in the occurrence of hysteresis. In many ca-ses combinations of these effects occur. A heterogeneity of the roughness of sur-faces can lead to the pinning effect at certain points in the so-called three-phase contact line. This results in a con-tact line which runs in a zigzag manner. The order of magnitude and topography of the roughness is particularly impor-tant because it affects the wetting pro-



perties most strongly. On the basis of si-mulations, it was recently confirmed that the hysteresis on randomly rough, self-affine surfaces depends mainly on the value of the Wenzel factor r. In addi-tion to this, it was shown that only the roughness in the μm range exhibits a hysteresis, but not in the nanoscale ran-ge. [79]

6. Methods for investigating the wetting of surfacesThe surface tension of liquids is deter-mined by direct measurements. [34, 80] The well-known methods are the ring method according to de Nouy, the pen-dant-drop method and the Wilhelmy method. For solids, on the other hand, only indirect measuring methods can be used to determine the surface energy. The majority of available measuring methods are dynamic, such as the pow-der contact angle method, the single-fiber Wilhelmy method, the needle-in method, the tilting plate method, and the modified Wilhelmy balance tech-nique. One differentiates between static and dynamic measurement methods in the investigation of solid surfaces. The method of the sessile drop can be used in both static and dynamic mode. In dy-namic mode, it is also referred to as a „needle-in method“. In this article, the sessile drop method for static (see Chap-ter 6.1) and the modified Wilhelmy bal-ance technique (see Chapter 6.2) were used for dynamic investigations.

6.1 Determination of contact angle using (static) sessile drop method Static contact angle investigations can be used to characterize smooth surfaces [6, 74]. The static sessile-drop method

was used in this review to determine the contact angle at room temperature (23°C). The procedure is shown in Fig. 4. A droplet of defined volume (2 μl) is care-fully deposited on the surface to be ex-amined and a drop contour image („snapshot“) is recorded within 2 sec-onds by the CCD camera. [17, 64] In the scope of these contact angle investiga-tions, 8 drops of each testing liquid (wa-ter, ethylene glycol and diiodomethane) on five different specimens of each sam-ple were considered. The OCA 20 appara-tus (Data Physics Instruments GmbH) was used for performing sessile drop ex-periments.

The static contact angle is determi-ned by the drop contour analysis. In the first step, the drop image is subjected to a gray scale analysis, in which the calcu-lation of the difference of the bright-ness of one image point to the neighbo-ring image point is performed for the entire image. The drop outline and the baseline, i.e. the contact line with the surface, are obtained from the position of the maximum differences in bright-ness. The maximum contrast is calcula-ted from the zero point of the 2nd deriva-tive of the brightness profile from which the strongest change in brightness is obtained. To calculate the contact angle, the contour of the drop is adapted to a mathematical model. For this adjust-ment, there are different approaches depending on accuracy and computatio-nal complexity, including elliptical, cir-cular and Young Laplace fitting. Depen-ding on the method used, the entire drop contour, a part of the drop con-tours or only the area of the three-pha-se contact point, is evaluated. In the course of this work, the drop contour is

automatically detected. Here, the con-tact angle determination was carried out according to the ellipse-fitting me-thod, which proved to be particularly suitable for asymmetrical drop con-tours. [17-18] In the ellipse method, a line of the shape of an ellipse is fitted to the drop outline. This method is well-suited for contact angles between 10° an 100°. [81] The contact angle is deter-mined as the slope of the contour line in the three-phase contact point on the left and right sides of the drop. [82]

According to Marmur, contact angle investigations using the sessile drop me-thod on rough surfaces are only useful if the axissymmetry of the drop is ensured and the droplet diameter is 3 orders of magnitude larger than the scale of the arithmetic average roughness value (Ra) of the samples. [83-84]

In previous studies, the surface tensi-on was measured using the static sessile drop method with water, diiodomethane and ethylene glycol as liquids on samples cleaned with isopropanol, which was found to have no effect on surface rough-ness. [11, 17-18, 61] The arithmetic ave-rage roughness value (Ra) of the samples was lower than 0.5 μm [18]. In the con-text of these studies, acetone and isopro-panol were used in order to remove bloo-ming residues such as accelerators (e. g. CBS and MBT) and their fission products from the elastomer surface [18] before the contact angle measurements. This plays an important role with respect to the technical application of polymers, since they contain not only the fission products of accelerators, but also nonpo-lar Anti-Sun Check Waxes and anti-oxidants.

6.2 Determination of dynamic contact angles (advancing and receding contact angles) using the modified Wilhelmy balance technique Concerning the (dynamic) modified Wil-helmy balance technique or method, ad-vancing and receding contact angles can be determined on solid samples with defined geometry. This method has sev-eral advantages compared to conven-tional optical methods like the sessile drop method. The measurement of a contact angle is reduced to measuring weight and length of a specimen, which is achieved by high accuracy and repro-ducibility. [74, 85-87] Furthermore the measured force (see Eq. 6 and 7) concern-ing any given depth of immersion is al-ready an averaged value. This does auto-

Fig. 4: Procedure for the determination of contact angles with the sessile drop method [68].

4



matically give a more accurate contact angle value showing the property of the entire sample [74, 83, 88], the roughness and chemical heterogeneity, respectively. Additionally the graphs of the Wilhelmy balance method are useful to study con-tact angle hysteresis at different wetting speeds. [74, 85] Stick-slip effects, which are shown for example in the needle-in method, are avoided [89]. In previous studies, only a few examples of surface modified polymers and elastomer com-posites have been investigated so far. [13, 18, 59, 74, 80, 90, 91]

The conventional Wilhelmy method, by which the surface tension of the li-quids is determined, has been modified (see Fig. 5) in a way that a test specimen with defined dimensions was used instead of a DIN-standardized and roug-hened platinum plate. [55] Before dip-ping the sample in a test liquid, the sur-face tension of the solvents had to be determined by the Wilhelmy Pt-Ir-Plate. The DCAT 11 (Data Physics Instruments GmbH) was used for performing wetting experiments.

After balancing the weight force (mg = 0), the test specimen is immersed in the particular test liquid and emerged. The forces Fadv and Frec (s. Eqs. 6 and 7) are measured as a function of the im-mersion depth h. V = hbd is the volume and l = 2 (b + d) is the wetted length of the test specimen, and γlv designates the surface tension and ρlv the density of the solvent.

𝐹𝐹𝑎𝑎𝐴𝐴𝑠𝑠 = 𝑠𝑠𝛾𝛾𝑠𝑠𝑠𝑠𝑐𝑐𝑐𝑐𝑐𝑐𝜃𝜃𝑎𝑎𝐴𝐴𝑠𝑠 − 𝑆𝑆𝜌𝜌𝑠𝑠𝑠𝑠𝑔𝑔 + 𝑚𝑚𝑔𝑔 (6) 𝐹𝐹𝑟𝑟𝑟𝑟𝑐𝑐 = 𝑠𝑠𝛾𝛾𝑠𝑠𝑠𝑠𝑐𝑐𝑐𝑐𝑐𝑐𝜃𝜃𝑟𝑟𝑟𝑟𝑐𝑐 − 𝑆𝑆𝜌𝜌𝑠𝑠𝑠𝑠𝑔𝑔 + 𝑚𝑚𝑔𝑔 (7)

By linear regression to the immersion depth zero, the buoyancy force Fa = Vpg can be eliminated from the recorded force-distance diagrams. If the sum of buoyancy and weight force is equal to zero, the resulting force corresponds to the wetting force. Hence, the corre-sponding measured forces Fadv and Frec, from which the contact angles θadv and θrec can be calculated are obtained sepa-rately for the extrapolation of h to 0 (i.e. V=0). [93]

𝜃𝜃𝑎𝑎𝐴𝐴𝑠𝑠 /𝑟𝑟𝑟𝑟𝑐𝑐 = 𝑎𝑎𝑟𝑟𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 �𝐹𝐹𝑎𝑎𝐴𝐴𝑠𝑠 /𝑟𝑟𝑟𝑟𝑐𝑐 (ℎ = 0)

𝑠𝑠𝛾𝛾𝑠𝑠𝑠𝑠� (8)

Fig. 6 shows a plot from immersion and emerging a sample and the calculation of the mean contact angle. [11, 61] The samples have to be cleaned with isopro-panol before use. The preparation of the samples is explained elsewhere .[61]

7. Sample preparation

7.1 Fabrication of samples used for sessile drop experiments

Preparation of elastomers

Two elastomers were studied in this work for the wetting investigations: EPDM and NBR. The EPDM sample composition is made of 100 phr (“per hundred rub-ber”) Keltan 4450 D, 5 phr ZnO, 1 phr stearic acid, 1.8 phr sulphur, 0.6 phr N-cyclohexyl-2-benzthiazylsulfenamide (CBS), 0.6 phr Mercaptobenzthiazole (MBT), zinc dibenzyl dithiocarbamate (ZBEC) as accelerators, zinc dicyanato diamine as cross-linking activator, 50 phr carbon black N 550 and 40 phr of an ali-

phatic plasticizer. Filled samples were prepared in an industrial type mixer. As far as the NBR samples are concerned, the type Perbunan NT 3445 was used. The NBR sample composition is made of 100 phr rubber, 5 phr ZnO, 1 phr stearic acid, 2 phr sulphur, 1.5 phr N-cyclohexyl-2-benzthiazylsulfenamide (CBS), 20 phr plasticizer alkylsulfonic phenyl ester (Mesamoll II) and 50 phr carbon black N 550.

Activation of the elastomer surfaceFor the activation of the elastomer sur-face by the Atmospheric Plasma-Plasma-Enhanced Chemical Vapor Deposition (AP-PECVD), a plasma jet (or plasma noz-zle) was used (see Fig. 7), which is inex-pensive to purchase. To operate a plasma

Fig. 5: Schematic representation of the immersing and the emerging cycles for the deter-mination of advancing and receding contact angles of the modified Wilhelmy balance technique. [13]

5

Fig. 6: Modified Wilhelmy balance technique: first immersion and emerging cycles of a SBR-sample coated with polysiloxane varnish in polyethylene glycol 200. [61]

6

not used for itforce at immersion (Fadv)

immersion depth [mm]

0 2 4 6 8 10

force at emerging (Frec)

1st sample immersion cycle1st sample emerging cycle

forc

e [N

]

Frec

Fadv

0,0015

0,0010

0,0005

0,0000

-0,0005

-0,0010

-0,0015

-0,0020

-0,0025

-0,0030

-0,0035

d



nozzle, a high-voltage power supply is needed, which is essential for igniting the plasma. An ionization gas (com-pressed air was used here), which is swirled, passes through the nozzle at 2000 L / h and is thereby put into the plasma state by the high-energy input of the high voltage. The stainless steel housing and the nozzle head are ground-ed, so that the potential-free plasma meets the surface to be treated.

Plasma polymerization Additionally, a precursor is required for the coating, which is conveyed as a liquid via a pump in an evaporator, so that it

can enter the gas phase. The carrier gas (nitrogen) transports the gaseous pre-cursor into the plasma nozzle, where the ignited plasma fragments the precursor. Due to the flow of the ionization gas, the plasma flame with the fragmented pre-cursor flows out of the nozzle and reacts on the surface to be coated. On the sub-strate, a highly crosslinked nanoscale plasma polymer is formed. The plasma polymer influences the friction behavior, whereas the bulk properties of the coat-ed elastomer remain unchanged. In or-der to realize the goal of friction minimi-zation between elastomers and the glass surface, the organosilicon precursors

hexamethyldisiloxane (HMDSO) and tetraethyl orthosilicate (TEOS) were used to feed the plasma nozzle. The precursor density was configured manually and automatically checked by a balance through which the software recognizes deviations in precursor weight and ad-just the pumping power to obtain the desired precursor flow. A peristaltic pump delivered the liquid precursor au-tomatically into the evaporator where temperatures above the boiling point of the precursor prevail. Thus, the gas phase of the precursor is reached in the evapo-rator and passed through the carrier gas nitrogen into the nozzle. The process pa-rameters related to the plasma jet treat-ment were not varied.

Since deposits of steel plates, finger-prints, blooming etc. are on the surfaces after vulcanization, it is important to first clean them with a lint-free cloth, acetone and isopropanol to ensure a clean surface on each substrate. Subse-quently, the cleaned elastomers were placed at a distance of 10 mm below the plasma nozzle and activated with com-pressed air without inflow of the precur-sor, so that the last impurities of the surface are removed and the applied plasma polymer layer adheres to the substrate in the second step. The activa-tion causes an increase in the surface energy and hydrophilizes the elastomer surface, which otherwise has low surface energies.

The plasma nozzle scans the substra-te in the dimensions that are given to the system. This allows local surface treat-ment without masking. After the elasto-mer surfaces are activated, the actual coating process can take place in a se-cond step. A big advantage of this me-thod lies in its potential to be used as an inline process, which does not require interruption of production.

7.2 Samples used for the modified Wilhelmy balance technique

Preparation of elastomersThe samples were prepared according to the following procedure: S-SBR (VSL-2525-0) and EPDM (Keltan 512) either unfilled or filled with 50 phr of carbon black N 339 and N 550, respectively, were prepared via a conventional sulphur vulcanization sys-tem. During vulcanization, samples were produced with extremely smooth (through smooth vulcanization plates) as well as with defined rough (through rough vul-canization plates) surfaces. [13-14]

Fig. 7: Survey about the fabrication process of the plasma modification of elastomer surfaces using HMDSO (hexamethyldisiloxane) and TEOS tetraethyl orthosilicate as precursors [18].

7

Fig. 8: Contact angle measure-ments on EPDM samples: water on untreated surface (a), water on activated sur-face (b), diiodo-methane on un-treated surface (c), diiodometha-ne on activated surface (d).

8



Preparation of the coated elastomer composites Varnished elastomers of EPDM and SBR have been investigated with respect to their friction properties on model sub-strates of steel, car glass and varnished sheets in previous studies. It has been demonstrated that the friction coeffi-cient of elastomers coated with certain sliding lacquer systems can be reduced significantly compared to the uncoated equivalents. [13-14] To improve the adhesion of the lacquers to the corre-sponding elastomers, a suitable pretreatment of the elastomer surface had to be taken into account. Cleaned with isopropanol and fluorinated, the smooth and rough SBR samples (2 mm thickness) were coated entirely with the four different varnish systems and heated to 120-140°C for 2-5 minutes. [13-14, 61]

■ 1-component PUR lacquer with PTFE particles (PTFE)

■ 1-component polyurethane varnish without particles (TPU)

■ 2-component PUR varnish (PU) with crosslinker

■ PUR lacquer with polysiloxane, PTFE particles and crosslinker (PDMS)

The different lacquer layer thicknesses of 5 to 40 μm were achieved with varying the solids content of the lacquers. [13-14]

Characterization of the coated compositesDigital microscopy and SEM were applied to determine the coating thickness and the topography of the elastomer com-posites. The composition of the lacquers was analyzed by ATR-IR and EDX map-ping. [13]

8. Results and discussion

8.1 Wetting of elastomer composites using the sessile drop methodIn the following EPDM and NBR com-posites have been investigated by the static sessile drop method. Two out of three liquids were considered for the investigations (water and diiodometh-ane). The fabrication of the samples is described in section 7.1. The samples were always cleaned with acetone and isopropanol to remove blooming resi-dues (e. g. accelerators) from the sur-face before the measurement (see sec-tion 6.1). The results of the untreated surfaces are revealed in Fig. 9 and 11a

and c. In the next step the elastomer surface was activated with air (see Fig. 9 and 11 b and d). To reduce friction and stick-slip, the activated surfaces were treated with two precursors (HMDSO and TEOS) on both elastomers to form a plasma polymer layer (results see Fig. 9 and 11 e-h). The results of the sessile drop measurements of EPDM are shown in Fig. 8 and 9 and the results of NBR are reflected in Fig. 10 and 11.

As far as the EPDM samples are con-cerned, the arithmetic mean water con-tact angle on the activated surface de-creases significantly compared to the untreated sample. This is due to the fact that the number of polar groups are in-creased by the incorporation of oxygen and nitrogen in the elastomer surface (see Fig. 8 and Fig. 9 a and b). The activa-ted EPDM surface which was plasma po-lymerized in the presence of the precur-

Fig. 9: Arithmetic mean contact angles of EPDM samples: water on untreated surface (a), water on activated surface (b), diiodomethane on untreated surface (c), diiodomethane on activated surface (d), water on activated surface with precursor HMDSO (e), diiodomethane on activated surface with precursor HMDSO (f), water on activated surface with precursor TEOS (g), diiodomethane on activated surface with precursor TEOS (h).

9

Fig. 10: Contact angle measure-ments on NBR samples: water on untreated surface (a), water on activated sur-face (b), diiodo-methane on unt-reated surface (c), diiodometha-ne on activated surface (d).

10



sor HMDSO exhibits a higher contact angle compared to the activated surface. The reason for this lies in the fact that this surface is less polar. Using the pre-cursor TEOS, the EPDM surface shows a

contact angle of ca. 60° (see Fig. 9 e-h). The surface treated with HMDSO is less polar compared to one with TEOS.

The arithmetic mean contact angle of the NBR sample, which was activated by

atmospheric plasma reveals also a signi-ficant decrease in comparison with the untreated NBR surface (see Fig. 10 and Fig. 11 a and b). This means that there are more polar groups in the surface of the activated NBR sample. Using HMDSO as precursor, a higher arithmetic mean contact angle compared to the activated surface is obtained. In contrast to this, the activated NBR surface which was plasma polymerized in the presence of the precursor TEOS reflects a higher con-tact angle in comparison with the sur-face which was only activated (see Fig. 11 b, e-h).

The samples of both elastomers which were plasma polymerized in the presence of the precursor HMDSO exhi-bit a higher contact angle than the sam-ples which were processed using TEOS as precursor.

8.2 Wetting of elastomer composites using the modified Wilhelmy balance technique It has been shown in previous studies that rough elastomer surfaces can be characterized very well by means of con-tact angle hysteresis, which is calculated from the difference of advancing and re-ceding contact angle. Dynamically meas-ured contact angles are relevant for the calculation of equilibrium contact angle values and surface energies. [11, 59, 61, 94] Previous studies have shown that contact angle hysteresis values, which represent a measure of the roughness and chemical heterogeneity of a surface [35, 65, 59], correlate well with rough-ness factors using differently smooth and rough coated elastomers. Ra, Rq, and Rz values are mainly used for the quanti-tative characterization of the roughness of smooth surfaces [64]. For many sur-faces, particularly for rough specimens, it is necessary to use contact angle hys-teresis values or fractal surface descrip-tors to describe surfaces and their mor-phology. [11, 96-97] For this purpose, contact angle hysteresis values of sur-face modified elastomers were com-pared for the first time with roughness values obtained from white light inter-ferometry and stylus instrument meas-urements which correlate well with vari-ous roughness parameters. [11, 13, 90] We used the equilibrium contact angle as an arithmetic means of advancing and receding contact angles since no accept-ed analytical method is known [59, 98]. Three different samples for each solvent with respect to one varnish were consid-

Fig. 11: Arithmetic mean contact angles of NBR samples: water on untreated surface (a), water on activated surface (b), diiodomethane on untreated surface (c), diiodomethane on activated surface (d), water on activated surface with precursor HMDSO (e), diiodomethane on activated surface with precursor HMDSO (f), water on activated surface with precursor TEOS (g), diiodomethane on activated surface with precursor TEOS (h).

11

Fig. 12: The first immersion and immersion cycles of three smooth (a) and rough SBR test specimens (b) of the modified Wilhelmy balance technique are revealed. The topography of the SBR surfaces was measured by the white light interferometry method [11, 13].

12



ered. By using the modified Wilhelmy balance method, the advancing and re-ceding contact angles could be repro-duced well for solvents of different polar-ity. A set of different test liquids over a wide range of polarity was used to ob-tain the contact angles. Three samples with the same coating were needed to calculate advancing and receding con-tact angles for each of the seven sol-vents. [11, 13] The advantage of using the modified Wilhelmy balance tech-nique for the determination of dynamic contact angles compared to the sessile drop method is that a large area is taken into account during the measurement [19], a good reproducibility is achieved and a method with very strong surface sensitivities is used. [11, 61, 74, 76, 88, 99] As an example, the following immer-sion and emerging cycles of the smooth and rough SBR samples are shown (see Fig. 13).

Water is used as probe liquid. It is a preferred liquid for contact angle mea-surements, because it has the highest surface tension (ca. 72.5 mJ/m2 at 298 K). Hence, it has measurable contact angles on most polymorphic materials. [63, 74] As it is accepted, water has both non-dispersive and dispersive character. In contrast to this, hexadecane for in-stance has the lowest surface tension of ca. 27.3 mJ/m2 resulting in near-zero contact angles on many polymers, but is useful for either fluorinated surfaces or when oleophobicity needs to be exami-ned. Further details are explained else-where. [63, 74]

Roughening the SBR surface leads to a significant increase of the contact angle hysteresis from ca. 21° (smooth SBR sur-face) to ca. 41° (rough SBR surface). The smoothness of the curve indicates that the solid sample is heterogeneous.

By means of white light measure-ments, by which the surface is scanned without contact by a light beam, stan-dard deviations σ from histograms of the z height values of the samples are obtai-ned by fitting a Gaussian distribution function (see Eq. 9) to the measured data (s. Fig. 7). Here, A and B denote fit para-meters and z0 represents the mean value of the corresponding distribution.

20 )/)(()( σzzBeAz −−+=Φ (9)

For the evaluation of the white light inter-ferometry height data, a Gaussion fit was used. As shown in Fig. 13, the fit curve describes the data points quite well.

The parameter σ representing the standard deviation of the Gaussian fit (see Fig. 13) is regarded as a parameter for the roughness of a surface. Advan-cing and receding contact angles of the four different varnishes on smooth and rough SBR were measured by the modi-fied Wilhelmy balance technique [11, 13]. Contact angle hysteresis values cor-relate quite well with the σ-values as it is revealed in Fig. 14. In addition to this, σ values are also in good accordance with

values of the arithmetic average rough-ness value Ra measured by a stylus inst-rument. [11, 13] As expected, ∆θ-values for the very smooth SBR surface (refe-rence) are very low in contrast to the varnished samples and rough SBR (see Fig. 14).

Usually, hysteresis values of about 10° or more were verified in experimental studies. Even values above 50° are mea-sured. [63, 74] In the case of the uncoa-ted rough sample, the value of ∆θ is

Fig. 13: Histogram of the height differences versus the z-height and the Gaussian fit curve for a rough SBR sample coated with PTFE varnish. [11, 13]

13

Fig. 14: Dependency of the contact angle hysteresis from the roughness parameter obtained from white light interferometry measurements. The contact angle measurements were car-ried out by the modified Wilhelmy balance technique.

14



about 20° higher compared to the smooth sample. This is due to air inclusi-ons between the rough sample and the water, as shown by numerical simulati-ons. [100]. The hysteresis values and the roughness parameters increase for al-most all SBR samples with the applied varnishes. Compared to the rough refe-rence sample, the hysteresis values of the rough varnished samples are 4 - 10° higher. The standard deviations of the advancing and retreating angles are bet-ween 0.5° and 2.3°. This is within the scope of measurement inaccuracy and can be classified as very well-reproducib-le. [63, 92] Therefore the varnishes level the unevenness of the rough surface on-ly partially. Since even unevenness in the nm range can contribute to the contact angle hysteresis, there are no general criteria how smooth a surface must be, so that the surface roughness has no in-fluence on the contact angle. [101-102]

As expected, the hysteresis values for the rough samples of the TPU and PU varnishes are significantly higher than for the two remaining varnishes, which both contain PTFE particles. The island-like structures of the PU coated sample are significantly rougher compared to the uncoated elastomer samples and it requires a larger area, as taken into ac-count in the Wilhelmy measurements to capture the few, but significantly larger areas. [11, 13] 9. Conclusion and outlookCoated and plasma modified elastomer composites were characterized by the sessile drop and the modified Wilhelmy balance method. The sessile drop meas-urement is well suited to qualitatively assess smooth and plasma modified sur-faces of elastomers easily. However, rough samples with irregularities, such as varnishes on rough elastomer surfac-es, have to be analyzed by methods such as the modified Wilhelmy balance meth-od. This method has several advantages compared to the sessile drop method. The measurement of a contact angle is reduced to measuring weight and length of a specimen, which is achieved by high accuracy and reproducibility. [74, 85-87] Furthermore the measured force con-cerning any given depth of immersion is already an averaged value. This does au-tomatically give a more accurate contact angle value showing the property of the entire sample [74, 83, 88, 103-105], the roughness and chemical heterogeneity, respectively. It was shown that contact

angle hysteresis values correlate with the roughness parameters obtained from white light measurements. Further investigations are necessary to charac-terize surfaces in order to compare, for example, surface descriptors (cut-off lengths and fractal dimensions) with hysteresis values and to obtain a better insight into the wettability of, in particu-lar, rough specimens. Furthermore the characterization of samples with varied roughness and chemical heterogeneity is of particular interest to to get a better understanding of wetting processes of elastomer surfaces. Also the investiga-tion of the polymer-liquid interaction is important.

AcknowledgementWe would like to thank the Federal Min-istry of Economics and Technology for fi-nancial support within the framework of the program for the promotion of indus-trial community research and develop-ment (IGF project No. 15810 BG and 18822 BG).

References[1] M. Yamamoto, N. Nishikawa, H. Mayama, Y.

Nonomura, S. Yokojima, S. Nakamura, K. Uchida, Langmuir 26 (2015) 7355.

[2] W. Barthlott, C. Neinhuis, Planta 202 (1997) 1. [3] N. J. Shirtcliffe, G. McHale, S. Atherton, M. I.

Newton, Advances in Colloid and Interface Science 161 (2010) 124.

[4] Produktdatenblatt Lotusan, Sto AG, Stühlin-gen.

[5] Textile Beschichtungen: Nanotechnologie macht Stoffe zu Hightech-Gewebe. Der Sieb-druck (2009).

[6] A. Wildberger, H. Geisler, R. H. Schuster, Kautschuk Gummi Kunststoffe 60 (2007) 24.

[7] K. Golovin, S. P. R. Kobaku, D. Hyun Lee, E. T. DiLoreto, J. M. Mabry, A. Tuteja, Science Ad-vances 2 (2016) 3.

[8] L. Zhu, J. Xue, Y. Wang, Q. Chen, J. Ding, Q. Wan, ACS Appl. Mater. Interfaces 5 (2013) 4053.

[9] H. Sojoudi, M. Wang, N. D. Boscher, G. H. Mc-Kinley, K. K. Gleason, Soft Matter, 12 (2016) 1938.

[10] Z. Cerman, W. Barthlott, J. Nieder, Bionik – “Was wir von Pflanzen und Tieren lernen können“, Rohwolt Verlag, Reinbek (2011).

[11] C. W. Karl, M. Klüppel, Chem. Listy 105 (2011) 233-CL-12.

[12] C. W. Karl, L. Busse, M. Klüppel, U. Giese, 9th Fall Rubber Colloquium (2010).

[13] C. W. Karl, A. Lang, A. Stoll, A. Weiße, M. Stoll, M. Klüppel, Kautschuk Gummi Kunst-stoffe 65/4 (2012) 44.

[14] C. W. Karl, A. Lang, L. Busse, A. Stoll, A.

Weiße, M. Stoll, M. Klüppel, Kautschuk Gummi Kunststoffe 65/5 (2012) 33.

[15] G. Stein, GAK 61 (2008) 306.[16] M. Lückmann, R. Hänsel, H. Geisler, M. Stoll,

Galvanotechnik 9 (2011) 650. [17] L. Busse, K. Peter, C. W. Karl, H. Geisler, M.

Klüppel, Wear 271 (2011) 1066.[18] W. Rahimi, M. Heim, H. Geisler, 12th Fall

Rubber Colloquium (2016).[19] R. J. Ratway, C. M. Balik, Polymer Preprints,

38 (1997) 247.[20] L. Martinez, L. Alvarez, Y. Huttel, J. Mendez,

E. Roman, A. Vanhulsel, B. Verheyde, R. Jacobs, Vacuum 81 (2007) 1489.

[21] P. Wang, K. L. Tan, C. C. Ho, M. C. Khew, Euro-pean Polymer Journal 36 (2000) 1323.

[22] Y. M. Tsai, F. J. Boerio, D. K. Kim, Journal of Adhesion 61 (1997) 247.

[23] M. D. Romero-Sanchez, M. Pastor-Blas, J. M. Martin-Martinez, P. A. Zhdan, J. F. Watts, Journal of Materials Science 36 (2001), 5789.

[24] C. P. Ho und H. Yasuda, Polymeric Material Science and Engineering 56 (1987) 705.

[25] J. K. Park, Z. Yang, S. Kim, ACS Appl. Mater. Interfaces 9 (2017) 33333.

[26] E. B. Troughton, C. D. Bain, G. M. Whitesides, R. G. Nuzzo, D. L. Allara, M. D. Porter, Langmuir 4 (1988) 365.

[27] C. D. Bain, G. M. Whitesides, Science 240 (1988) 62.

[28] W. A. Zisman, J. Paint Tech. 44 (1972) 42.[29] A. Marmur, C. Della Volpe, S. S. Alidad, A. Ja-

roslaw, W. Drelich, 5 (2017) 3.[30] B. Bhattarai, N. V. Priezjev, Computational

Materials Science 143 (2018) 497.[31] E.L.Decker, B.Frank, Y.Suo, S. Garoff, Colloids

and Surfaces A: Physicochemical and Engi-neering Aspects 156 (1999) 177.

[32] A. Marmur, Annu. Rev. Mater. Res 39 (2009) 473.

[33] T. Young, Miscellaneous works, Vol. 1, Eds.: G. Peacock, J. Murray (1855)

[34] H.-J. Butt, K. Graf, M. Kappl, “Physics and Chemistry of Interfaces”, Wiley-VCH pub-lisher, Weinheim (2003).

[35] P.-G. de Gennes, Reviews of Modern Physics 57 (1985) 827.

[36] D. Quéré, Physica A: Statistical Mechanics and Its Applications 313 (2002) 32.

[37] https://www.kruss-scientific.com/services/education-theory/glossary/static-contact-angle/.

[38] W. M. Haynes, “Handbook of Chemistry and Physics”, Boca Raton: CRC Press (1986).

[39] J. Drelich, Colloids and Surfaces A: Physico-chemical and Engineering Aspects 116 (1996) 43.

[40] A. Marmur, Journal of Colloid and Interface Science 2 (1997) 462.

[41] A. Amirfazli, A. W. Neumann, Advances in Colloid and Interface Science 3 (2004) 121.



[42] L. Schimmele, M. Napiorkowski, S. Dietrich, Journal of Chemical Physics 16 (2007) 164715.

[43] G. Wolansky, A. Marmur, Langmuir 14 (1998) 5292.

[44] C. Della Volpe, D. Maniglio, S. Siboni, M. Morra, Oil & Gas Science and Technology – Revue DIFP Energies Nouvelles 56 (2001) 9.

[45] R. N. Wenzel, Ind. Eng. Chem. 28 (1936) 988.[46] A. Marmur, E. Bittoun, Langmuir 25 (2009)

1277.[47] A. B. D. Cassie, S. Baxter, Wettability of po-

rous surfaces. In: Transactions of the Fara-day Society 40 (1944) 546.

[48] S. Brandon, N. Haimovich, E. Yeger, A. Mar-mur, Journal of Colloid and Interface Sci-ence 263 (2003) 237.

[49] E. Pierce, F. J. Carmona, A. Amirfazli, Colloids and Surfaces A: Physicochemical and Engi-neering Aspects 323 (2008) 73.

[50] J. Drelich, Surface Innovations 4 (2013) 248.[51] J. Drelich, Journal of Adhesion 63 (1997) 31.[52] A. Marmur, Colloids and Surfaces A: Phys-

icochemical and Engineering Aspects 136 (1998) 209.

[53] E. Bittoun, A. Marmur, Langmuir 26 (2010) 15933.

[54] E. Bittoun, A. Marmur, M. Ostblom, T. Ed-erth, B. Liedberg B, Langmuir 25 (2009) 12374.

[55] H. Y. Erbil, “Solid and Liquid Interfaces“, Ox-ford Blackwell Publishing (2006).

[56] R.-D. Schulze, W. Possart, H. Kamusewitz, C. Bischof, J. Adhesion Sci. Technol. 39 (1989) 41.

[57] A. Marmur, Soft Matter 2 (2006) 12.[58] V. Broje, A. A. Keller, J. Colloid Interf. Sci. 305

(2007) 286.[59] K. W. Stöckelhuber, A. Das, R. Jurk, G. Hein-

rich, Polymer 51 (2010) 1954.[60] A. Le Gal, X. Yang, M. Klüppel, J. of Chem.

Phys. 123 (2005) 014704.[61] C. W. Karl, L. Busse, M. Klüppel, Proc. 9th

Rubber Fall Colloquium (2010).[62] A. Marmur, Soft Matter 8 (2012) 6867.[63] R. J. Good, Contact Angle, Wettability and

Adhesion- A Critical Review in Contact An-gle, Wettability and Adhesion K. L. Mittal, Ed, VSP (1993).

[64] S. Palzer, C. Hiebl, K. Sommer, H. Lechner, Chemie Ing. Technik, 73 (2001) 1032.

[65] J. Long, M. N. Hyder, R. Y. M. Huang, Adv. in Colloid and Interf. Sci. 118 (2005) 173.

[66] A. Marmur, Colloids Surf. A Physicochem. Eng. Asp. 81 (1998) 136.

[67] H. Kamusewitz, W. Possart, Appl Phys A 76 (2003) 899.

[68] S. Dallmann, PhD Thesis, Technical Univer-sity of Dortmund (2011).

[69] C. W. Extrand, Y. Kumagai, Journal of Colloid and Interf. Sci. 184 (1996) 191.

[70] L. Li, L.T. Weng, Macromolecules 30 (1997) 3698.

[71] H. Kamusewitz, W. Possart, D. Paul, Colloids and Surfaces A 156 (1999) 271.

[72] R. Shuttleworth, G.L.J. Bailey, Transactions of the Faraday Society 44 (1948) 16.

[73] D. Q. Li, Colloids & Surfaces A-Physicochem-ical & Engineering Aspects 116 (1996) 1.

[74] M. Strobel, C. S. Lyons, Plasma Processes and Polymers, 8 (2011) 8.

[75] M. Strobel, V. Jones, C. S. Lyons, M. Ulsh, M. J. Kushner, R. Dorai, M. C. Branch, Plasmas Polym. 8 (2003) 61.

[76] E. B. Troughton, C. D. Bain, G. M. Whitesides et al., Langmuir 4 (1988) 365.

[77] M. E. R. Shanahan, A. Carre, Langmuir 11 (1995) 1396.

[78] A. Carre, J. C. Gastel M.E.R. Shanahan Na-ture 379 (1996) 432.

[79] R. David, A.W. Neumann, Langmuir 29 (2013) 4421.

[80] H.-D. Dörfler, “Grenzflächen- und Kolloidchemie“,VCH Verlagsgesellschaft mbH, Weinheim (1994).

[81] https://www.kruss-scientific.com/services/education-theory/glossary/drop-shape-analysis/.

[82] Application Note Nr. 8, Beschreibung der analytischen Methode zur Charakter-isierung der Oberflächenenergie und Polar-ität von flüssigen und festen Haftklebstof-fen, data physics: http://www.dataphysics.de/fileadmin/user_upload/pdf/application-notes/5_Surface_tension_of_molten_poly-mers_and_hot-melts.pdf .

[83] T. S. Meiron, A. Marmur, I. S. Saguy, Journal of Colloids and Interface Science 274 (2004) 637.

[84] G. Wolansky, A. Marmur, Colloids Surf. A Physicochem. Eng. Aspects 156 (1999) 381.

[85] Bracco, Gianangelo, Holst, Bodil, “Surface Science Techniques, Chapter 1: Contact an-gle and wetting properties“, Y. Yuan, T. R. Lee, Springer Series in Surface Science, Springer Publisher Berlin (2013)

[86] MINITAB Statistical Software Release 15.1, Minitab, Inc. (2007).

[87] V. Jones, M. Strobel, M. J. Prokosch, Plasma Process. Polym. 2 2005 547.

[88] L. M. Lander, L. S. Siewierski, W. J. Brittain, E. A. Vogler, Langmuir 9 (1993) 2237.

[89] L. Gao, T. J. McCarthy, Langmuir 25 (2009) 14105.

[90] A. Marmur, Colloids and Surfaces A: Phys-icochemical and Engineering Aspects 116 (1996) 55.

[91] M. Miyama, H. K. Yasuda, Langmuir 14 (1998) 960.

[92] R. H. Johnson, Jr., R. H. Dettre, Wetting of Low-energy Surfaces, in Wettability, J. C. Berg, Ed, Dekker (1993).

[93] C. Della Volpe, C. Cassinelli, M. Morra, Lang-muir 14 (1998) 4650.

[94] C. W. Karl, M. Klüppel, U. Giese, Proc. Elasto-

mers 52 (2011).[95] K. F. Grythe, F. K. Hansen, Langmuir 23

(2006) 6109.[96] B. N. J. Persson, O. Albohr, U. Tartaglino, A. I.

Volokitin, E. Tosatti, J. Phys.: Condens. Mat-ter 17 (2005).

[97] Y. Nonomura, E. Seino, S. Abe, H. Mayama, J Oleo Sci. 8 (2013) 587.

[98] H. Kamusewitz, W. Possart, D. Paul, Colloid Surf. A-Physicochem. Eng. Asp. 156 (1999) 271.

[99] J. Pack, C. S. Lyons, M. Strobel, J. Adhes. Sci Technol. 17 (2003) 643.

[100] S. Iliev, N. Pesheva, V. Nikolayev, 11th Na-tional Congress on Theoretical and Ap-plied Mechanics (2009).

[101] J. P. Youngblood, T. J. McCarthy, Macromol-ecules 32 (1999) 6800.

[102] D. Y. Kwok, A. W. Neumann, Adv. Colloid In-terface Sci. 81 (1999) 167.

[103] B. D. Cassie, Disc. Faraday Soc. 3 (1948) 11.[104] A. Marmur, Annual Review of Materials

Research 39 (2009) 473.[105] L. Gao T.J. McCarthy, Langmuir 25 (2009)

14105.

Documents

Elastomer plasma polymerization Review - Characterization of … · 2018. 9. 11. · nished elastomer composites [11-15], elastomer surfaces modified by gas-phase fluorination [15-16]