Peel Adhesion: Influence of Surface Energies and Adhesive Rheology

The adhesion properties of polymers are known to be influenced by both intermolecular forces operative at the interface and the rheological history of both bonding and unbonding. Recent adsorption and viscoelastic theories of adhesion and cohesion are implemented in a comprehensive examination of these phenomena. Eight peel force “master curves” extending over 14 decades of reduced rate and representing glassy state to flow region rheology are superimposed to provide a composite response envelope. Each master curve represents rate-temperature reduced adhesion of an alkyl acrylate adhesive (γ_c = 26 dyne/cm) to substrates ranging from low adhesion fluorinated polymers (γ_c = 15 to 17 dyne/cm) to polar poly-amide surfaces (γ_c = 45 dyne/cm) and glass. The rate dependent transition from interfacial to cohesive failure, a subject not treated by adsorption theory, is shown to be coincident with the onset of entanglement slippage within the polymeric adhesive. Thermodynamic criteria of polymer adhesion are shown to be applicable only to the flow region of polymeric response. This study indicates that measured surface tensions or calculated surface energies of polymeric solids do not properly account for the contributions of three dimensional network structure of the polymeric bulk phase to its total work of cohesion. Evidence of true interfacial failure of a polymer-polymer bond is supported by critical surface tension measurements.     

Physico-Chemical Criteria for Maximum Adhesion. Part II: A New Comprehensive Thermodynamic Analysis

In this paper, two parameters defined as the relative work of adhesion [W_A/γ_L] and the relative interfacial energy [γ_SL/γ_L] have been examined for their assumed usefulness in correlating the thermodynamic properties of the components of the system substrate/ adhesive with its practical performance (strength). It is shown that the minimum value of [γ_SL/γ_L] relevant to conditions for the maximum adhesion becomes zero only for those systems (relatively rare) for which interaction factor Φ₀ is equal to 1.0.

Several transition points were identified for boundary conditions acquired at θ = 0° and θ = 90° which can be used to predict the properties and performance of an adhesive joint. These transition points are: a_MIN—energy modulus of the system (E. M. S.), relevant to the minimum interfacial energy; a_S—E. M. S. where self-spreading of adhesive occurs; a_CRIT—E. M. S. relevant to conditions under which the thermodynamic work of adhesion becomes negative and the system exhibits a tendency for self-delaminating or has “zero-strength”; a_CF—E. M. S. beyond which the geometry of the interface at any interfacial void or boundary of the joint may be regarded as a crack tip.

It is shown that only in those systems for which Φ₀ = 1.0 can a minimum contact angle of 0° indicate a condition for the maximum strength. If Φ₀ is known, the optimum contact angle can be estimated and hence the optimum surface energy of the substrate (adjusted by surface treatment, etc.) for the maximum adhesion.     

A Reinterpretation of Organic Liquid-Polytetrafluoroethylene Surface Interactions

The wettability of polytetrafluoroethylene (PTFE) by organic liquids is reanalyzed in terms of dispersion-polar interactions across the liquid-solid interface. The analysis provides values of γ_s^d = 19.6 dyne/cm, and γ_S^D = 2.0 dyne/cm for the respective dispersion and polar parts of the surface tension γ_s = 21.6 for PTFE. The definition of a polar contribution to the surface tension of PTFE clarifies detailed aspects of the wettability of this polymer by different homologous liquid series. A modified analytical definition for work of adhesion is developed and applied to this discussion.     

The Rule of Interfacial Equilibrium

The detailed analysis of the well-known relation γ₁₂ = γ₁ -γ₂ has been done (γ₁₂, γ₁ and γ₂ are interfacial and surface tensions of phases 1 and 2). The analysis of equilibrium of a liquid drop at the interface with another liquid allowed us to prove that this relation should be modified by including in it the value γ₂* which is considered as the result of all interactions at the interface and is accepted depending on γ₁. The value γ₂* represents the surface tension of a phase in the ternary point solid-liquid-gas, i.e. in the zone of interfacial nonuniformity. The modified form of the relation, called the rule of interfacial equilibrium, allows us to show that thermodynamic work of adhesion is equal to the cohesion energy of the interphase formed by phase 2.     

Dispersion-Polar Surface Tension Properties of Organic Solids

A new definition for work of adhesion W_a is applied to computationally define the dispersion γ_s^d and polar γ_s^d components of the solid surface tension γ_s = γ_s^d + γ_s^d for twenty-five low energy substrates. These calculated surface properties are correlated with surface composition and structure. Surface dipole orientation and electron induction effects are respectively distinguished for chlorinated and partially fluorinated hydrocarbons. Published values for critical surface tension of wetting γ_c are correlated with both γ_s^d and γ_s.     

Surface-Chemical Criteria for Optimum Adhesion

According to Bikerman, who attributes failure in adhints to a weak boundary layer, it is almost impossible and meaningless to correlate adhesive strength to surface-chemical properties of adhints. Though his assertion seems to be confirmed by the recent studies of Schonhorn and his coworkers on the methods of CASING and TCR, not a few results have yet been accumulated, which show a close relation between them. In this paper surface-chemical criteria for the optimum adhesion are investigated and the minimum interfacial tension or the maximum wetting pressure is deduced from the published data and our own as a first approximation. It is emphasized that, when critical surface tension γ_c would be used as a measure of surface-chemical properties of solid, its variability according to liquid series (nonpolar, polar and hydrogen bonding liquids) should be carefully taken into consideration. The importance is shown for polyethylene and its fluorine substituted polymers, using newly measured contact angle data and Zisman's data. Results of Levine et al. and Schonhorn et al. on adhesive shear strength with epoxy adhesives are replotted against available values of γ_c obtained by the use of hydrogen bonding liquid (γ_c^c), which are thought to reflect wetting behaviors of epoxy adhesives quite well. Each curve shows a maximum around γ_c^c = 40 dyne/cm with few points falling off the curves.     

The relationship between the strength of an adhesive bond and the thermodynamic properties of its components

Maximum stability of any system is achieved when its free energy is minimum, in accordance with the second law of thermodynamics. Considering the adhesive bond as a thermodynamic system, it is proposed that the minimum interfacial energy coincides with (1) the maximum strength, and (2) the maximum durability, understood as bond resistance to degradation under environmental attack. The thermodynamic properties of bond components which play a key role in promoting conditions for maximum strength of adhesion have been identified. The general pattern of the relationship: STRENGTH = function (interfacial energy and related parameters), has been developed based on experimental data covering a variety of adhesives and substrates such as metals (steel and aluminium), plastics, ceramics and glass fibre composites. The influence of adhesion promoters (eg, silanes) has also been considered.

It is shown that conditions for maximum strength coincide with the minimum interfacial energy of the system, acquired when the ratio of the surface energy of the substrate, γ₁, to that of the cured adhesive, γ₂ (ie, a = γ₁/γ₂), has a specific value denoted a_MIN. Systems with energy ratios a a_MIN were found to have engineering utility, because the strength deficiency for a >a_MIN was found to be significantly less than for a MIN.     

On the Surface Tension and its Temperature Variation in Film-Forming Polymers

A thermal gradient bar has been used for convenient measurements of γ_c and dγ_c/dT in complex polymers used as film-formers. The technique yields both γ_c and its temperature variation in one experimental sequence well suited for rapid, routine applications. Surface tension data have been obtained for a styrene-acrylic terpolymer, and these have also been used to characterize the compatibility of external plasticizers for the polymer. The surface tension approach has shown that glyceryl dibenzoate, though compatible with the polymer at temperatures above ∼70°C becomes incompatible at use temperatures, and exudes to the polymer film surface. Measurements of moisture sensitivity in plasticized polymer samples have confirmed the incompatibility and illustrated one of the applications to which the gradient bar and its data generation potential may be put.     

Surface Energy Analysis of Treated Graphite Fibers

Wettability measurements and surface energy analysis are applied to isolate the (London-d) and (Keesom-p) polar contributions to solid-vapor surface tension γ_sv=γ^d_sv + γ^p_sv of surface treated graphite fibers. Surface treatments include metal coatings with Al, Cu, and Ni, chemically reducing heat treatments in H₂ and vacuum, and films of highly chlorinated polymers such as polyhexachlorobutadiene and polychloral. This study shows that the highly polar surface properties γ^p_sv|γ_sv ≃ γ^d_sv|γ_sv ≃ 0.50 of commercial graphite fibers can be modified by surface treatment to display dominant dispersion character with γ^d_sv|γ_sv ≃ 0.79 to 0.92 without substantial reduction in total surface energy γ_sv. For adsorption bonded fiber/matrix interfaces a new method of mapping the surface energy effects of an immersion phase upon the Griffith fracture energy γ_G is applied to define criteria for strong interfacial bonding under both air and water immersion.     

Interfacial Bonding and Environmental Stability of Polymer Matrix Composites

Recently developed adsorption-interdiffusion (A-I) theory of adhesion is employed to isolate the (London) dispersion γ_i,j^d and (Keesom) polar γ_i,j^p components of the excess interfacial free energy γ_i,j=γ_i,j^d+γ_i,j^p at the fiber-matrix interface in polymer matrix composites. For adsorption bonded interfaces the theory defines a new method of mapping the surface energy effects of an immersion phase upon the Griffith fracture energy γ_G. The stability of interfacial bonding between graphite fiber-epoxy matrix is defined in terms of the theoretical model and experimentally evaluated by accelerated aging studies which monitor changes in fracture energy for crack propagation perpendicular to the fiber axis. Applications of the model to control fiber surface treatments and select matrix components for optimized bond strength and environmental resistance is discussed.     

On Multi-valued Surface Properties of PMMA Films

An apparent link between the surface properties of polar group-containing polymers, such as PMMA and Styrene/Acrylic copolymers, and the thermodynamic quality of solvents used in solutions from which the polymers were cast, was described in earlier papers.^1,2 In these polymers, significant variations have been observed in critical surface tensions(γ_c), and in the thermodynamic interaction parameters for selected vapor-polymer pairs, when the configuration of the polymer in solution was varied through the suitable selection of solvents of differing thermodynamic quality. The “solvent history” effect on surface properties of solid film was not detected however for non-polar polymers such as polystyrene (PS).^1,2 Apparently the distinct chain configurations adopted in solution by PMMA are carried over into the solid and result in different proportions of non-polar (backbone) and polar (side chain) moieties being located in the surface layer of the solid. Since only one surface state can correspond to a thermodynamic equilibrium, it may be expected that the film surface properties will change with time, as the thermodynamically preferred state is attained. As a consequence, use properties of these films should also display (initially) the “solvent history” effect, and should vary similarly with time. The present communication is concerned with these points.     

On the importance of some surface and interface characteristics in the formation of the properties of adhesive joints

The importance of some relative surface characteristics which determines the strength of adhesive joints: specific surface of substrate , relative contact area β and specific contact area β in the adhesion interaction process were emphasised. Existing and potential methods of experimental evaluation of these characteristics were shortly analysed. The durability of the adhesive joints in water media significantly increases with growth of specific surface ^* of chemically treated substrate evaluated from the SEM micrographs. Specific surface calculated from the experimental data of hexane adsorption measurements for iron particles (particulate model of steel substrate) is more then ten times greater than respective ^* values. The relative contact area β of the Al₂O₃ particles (in wide range of ) with PE melt was in a roundabout way evaluated by experimental determination of the affect of on kinetic of peel strength formation of adhesive joints: Al₂O₃ filled PE-steel. The speculation was based on the ability of Al₂O₃ to adsorb low-molecular products of contact oxidation of PE which takes place in the process of formation of adhesive joints and determines their strength. The ability of sorption in its turn is proportional to efficient value of β. The availability of the Al₂O₃ surface was evaluated.     

Pigment inhomogeneity and void formation in organic coatings

The conventional theory of organic coatings assumes that the volume density of pigment particles is uniform throughout the sample. The coating is then described in terms of the pigment volume concentration Π, which equals the volume of pigment divided by the volume of pigment and polymer. Voids form in the coating when the pigment particles are randomly dense-packed, which implies that Π exceeds the critical pigment volume concentration (Φ_c or that Λ=Π/Φ_c > 1. Due to fluctuations in the local density of pigment however, some regions in the coating may become randomly dense-packed even below Φ_c. Hence, voids may form in the densely-packed islands even when Λ<1. Our model for void formation contains two fitting parameters: N_o is the smallest number of pigment particles in a densely-packed cluster that may contain a void; C_q is the coarseness of the polymer space-filling in the volume of the sample not occupied by pigment. When C_q=0 and Π > 1, the polymer completely fills the interstitial volume and the void concentration vanishes. But when C_q > 0, voids may form in the densely-packed regions of the coating even below Φ_c. The coarseness parameter C_q, depends on the sample preparation, on the properties of the pigment and polymer, and on the pigment volume concentration Π. For any nonzero C_q we conclude that optical measurements will systematically underestimate Φ_c. On the other hand, since the density of polymer is less than half the density of pigment, the peak in the mass density p(Π) of the coating will overestimate Φ_c. Unless C_q is abnormally large, the void percolation threshold value Φ_v is larger than Φ_c and is a decreasing function of the coarseness C_q. The predictions of this simple model are in good agreement with experiment, and are relevant to the general class of random concentrated composites which include organic coatings, ceramic polymer slips, composite solid polymer electrodes and some forms of battery separators.     

Interfacial and Bulk Contributions to Peeling Energy

Thin polyurethane films, having low adhesion to dried protein, were developed as candidate materials for non-adhesive surgical dressings. In order to model wound-adhesion, gelatine was cast from solution on to the film and allowed to dry. The film was peeled from the gelatine at 180° peel angle, and the peel force measured as a function of the temperature of test. The dynamic mechanical properties of the films were measured over the range -90°C to 110°C and values of tan δ were determined at the temperatures employed for peeling. Thus, a correlation was obtained between peeling energy and tan δ for each of eight films.

The generalised theory of fracture mechanics states that the adhesive failure energy is given by the product of an interfacial energy term and a “loss function” involving the hysteresis ratio of the material. If the strains are small the hysteresis ratio is proportional to tan δ. The experimental results show excellent agreement with the theory, but the interfacial term turns out to be much greater than the true interfacial energy (or thermo-dynamic work of adhesion). The reason for this result is discussed.     

Über den Einfluß der Grenzflächenspannung auf die Haftfestigkeit

The values of adhesion between four diffrent adhesive,and (i) steel substrates whose surface energy had been altered by adsorption, and (ii) several polymer having different surface energies, had been measured. The results show that the adhesion has a maximum value when the surface energy of the hardened adhesive is equals to that of the substrate, i.e. when the interfacial energy adhesive/substrate is a minimum. The adhesion of the adhesives to the polymer was much smaller than to the steel sprcimens and the dependence of the adhesion onm the interfacial energy was sharper in the case of the polymers. The decrease of the adhesion with increasing interfacial energy was fiund to be greater if the liquid adhesive wets the substrate badly than the steel speciman.     

Fracture toughness of interfaces between glassy polymers in a trilayer geometry

We studied the fracture behavior of trilayer A/B/A assemblies based on polystyrene (PS) and poly(methylmethacrylate) (PMMA) where the central layer of the A polymer was confined (0.5–200 μm) between two thick plates of the B polymer (1– 3 mm). Diblock and random P(S-MMA) copolymers were used to provide a good stress transfer across the interfaces. Fracture experiments were performed with the double-cantilever beam method and the fracture mechanisms were observed by optical microscopy on microtomed slices of the damaged zone. The measured _c of the A/B interface fractured during the test was dependent on the molecular structure at the interface (random copolymer, diblock copolymer or no copolymer), on the crazing stress of the bulk materials and on the interfacial shear stresses. When the phase angle of the loading was even slightly positive, oblique crazes were observed in the PS increasing greatly _c. If PS was the central layer, this resulted in a very marked dependence of _c on the thickness of the central layer for a thickness range 10–200 μm which was not observed when the PMMA was the central layer. Thermal treatments modifying the interfacial shear stresses were also found to have a very strong effect on _c.     

The Effect of Interfacial Tension on the Adhesion of Cathodic E-coat to Aluminum Alloys

Adhesion of a cathodically electrodeposited paint (E-coat) to aluminum alloys, Alclad 2024-T3, AA 2024-T3 and AA 7075-T6, was investigated to examine the influence of interfacial tension at the paint/metal interface. The surface energy of an aluminum plate was modified by depositing a plasma polymer of a mixture of trimethyl silane (TMS) and one of three diatomic gases (O₂, N₂, and H₂) by cathodic plasma polymerization. The contact angle (Φ) of water on a modified surface changes as a function of the mole fraction of the diatomic gas. The plot of cos Φ_PP of a plasma polymer as a function of the mole fraction of the gas crosses the plot of cosΦ_EC of the E-coat. The difference, ΔCosΦ = cos Φ_PP - cosΦ_EC, is a parameter which indicates the level of interfacial tension at the paint/metal interface. ΔCosΦ = 0 represents the minimum interfacial tension. The adhesion of a cured E-coat on a panel was evaluated by the N-methyl pyrrolidinone (NMP) paint delamination time test. The maximum peak of adhesion test values plotted as a function of ΔCosΦ occurred around the zero point. ΔCos Φ = 0, indicating that maximum adhesion is obtained with minimum interfacial tension. Mixtures of TMS and N₂ on all three aluminum alloys studied consistently displayed longer delamination times in the NMP test than mixtures of TMS and O₂ or H₂.     

Pressure Sensitive Adhesives Based on VectorR SIS Polymers I. Rheological Model and Adhesive Design Pathways

Dexco Polymers (a Dow/Exxon partnership) has been manufacturing Vector^R SIS polymers since 1990.¹ This paper describes experiments carried out to study Vector SIS polymers and model pressure sensitive adhesive (PSA) formulations based on Exxon Chemical's Escorez^R 1310LC tackifier. The adhesive behavior of tackified polymers was quantitatively analyzed by applying the rheological principle of time-temperature superposition² and the mapping approach,^3,4 and the pressure sensitive rheological model⁵ developed earlier. This model⁵ was developed by expanding and modifying an equation [adhesive fracture strength = (intrinsic adhesion) × (bulk energy dissipation)] proposed by Gent et al.^6,7 and Andrews et al.^8,9 This study delivers two key results. The first is that the fracture strength of the PSA/steel bond is the multiplication of three terms: the intrinsic (or interfacial) adhesion, the bonding and the debonding terms (Fig. 1). The debonding term is correlated with the logarithm of the loss modulus at the PSA debonding frequency or with the logarithm of the monomeric friction coefficient of the block copolymer/tackifier system. Both the loss modulus and the monomeric friction coefficient measure the energy dissipation in the bulk adhesive. The second is that PSA design pathways can be established by a mapping approach in the rheological space of the plateau modulus versus the loss modulus peak position in the frequency scale (Fig. 2). Plateau modulus is the bonding parameter because it measures the wetting capability of the adhesive with the substrate surface. The loss modulus peak position is the debonding parameter because it corresponds approximately to the time scale (or the frequency scale) in which one deforms the adhesive to maximize energy dissipation. Therefore, the tackifier and oil combination lowers the plateau modulus, but increases the T_g of the polyisoprene phase of the SIS polymer. This increase in T_g is equivalent to the lowering of the rate of local rearrangement (frequency of segment jumps) of the polyisoprene chains of the block copolymer. An optimal “tackification pathway” in this rheological space is achieved by tailoring the tackifier type and T_g, and the amount of oil used in the PSA.

In brief, the PSA rheological model and mapping approach described in this work for Vector SIS polymers give a comprehensive understanding and adhesive design pathways. This concept and approach not only allow raw material suppliers to improve and design better tackifier and polymer products, but also provide PSA formulators a quantitative tool to achieve PSA end property results.     

Measurement of the Surface Free Energy of Amorphous Cellulose by Alkane Adsorption: A Critical Evaluation of Inverse Gas Chromatography (IGC)

Surface energies of amorphous cellulose “beads” were measured by IGC at different temperatures (50 to 100°C) using n-alkane probes (pentane to undecane). The equation of Schultz and Lavielle was applied which relates the specific retention volume of the gas probe to the dispersive component of the surface energy of the solid and liquid, γ^d_s and γ^d_l, respectively, and a parameter (“a”) which represents the surface area of the gas probe in contact with the solids. At 50°C, γ^d_s was determined to be 71.5 mJ/m², and its temperature dependence was 0.36 mJ m^-2 K^-1. Compared with measurements obtained by contact angle, IGC results were found to yield higher values, and especially a higher temperature dependence, d(γ^d_s)/dT. Various potential explanations for these elevated values were examined. The surface energy, as determined by the Schultz and Lavielle equation, was found to depend mostly on the parameter “a”. Two experimental conditions are known to affect the values of “a”: the solid surface and the temperature. While the surface effect of the parameter “a” was ignored in this study, the dependence of the surface energy upon temperature and probe phase was demonstrated to be significant. Several optional treatments of the parameter “a” were modeled. It was observed that both experimental imprecision, but mostly the fundamental difference between the liquid-solid vs the gas-solid system (and the associated theoretical weakness of the model used), could explain the differences between γ^d_s and d(γ^d_s)/dT measured by contact angle and IGC. It was concluded that the exaggerated temperature dependence of the IGC results is a consequence of limitations inherent in the definition of parameter “a”.