Tribological and mechanical properties of lubricant filled microcapsules in thermoplastic composites

,


Introduction
Polymers and their compounds are often modified with functional fillers or additives to tailor the material properties to specific needs of an application. Due to their low cost, ease of processing and low density, polymers are widely used in engineering applications [1]. The use of plastics in tribologically stressed components is specifically limited, especially at high speeds or heavy duty. High frictional forces may exceed the thermal load capacity and result in high wear and short lifetime of the component [2]. Plastics usually offer sufficient friction properties under dry running conditions. Besides the modification of the material, surface and design can be adapted to reach a lower frictional loss [3]. To improve the tribological properties, solid or liquid lubricant materials can be added. Solid materials like carbon nanomaterials, nanosilica, MO2 or PTFE can be embedded in the polymeric matrix [4][5][6][7][8][9][10][11][12]. Today, the market range of solid additives with tribological effects is quite limited. The effect of friction reduction of solid materials is always lower than the one of an external lubrication [13]. The main advantage of selflubricating plastics compounds is their low need of maintenance and the dramatically reduced wear. Therefore, they are predestined for use in components that are difficult to access, such as gear wheels in encapsulated structures. The direct incorporation of lubricating liquids in a polymeric matrix is strongly limited and brings difficulties in processing.
Micro encapsulated lubricants (i.e. oils or greases) represent a relatively new possibility as pseudo-solid additive for tribological applications. In recent years, several micro encapsulation technologies have been investigated. Shell materials alike melamin formaldehyde resin, polyurethane or polysulfone have been used to encapsulate various oils. Schoch et al. used melamin resin and polyurethane to encapsulate different commercial oils. They also demonstrated the micro encapsulation for oils with different chemical compositions. Additional additives in oils did not hinder the encapsulation process [14]. Therefore, the micro encapsulation process is accessible to a wide range of oils, available on the market today. Various research groups synthesized microcapsules for the usage in thermoset and thermoplastic materials [15]. During compounding, the microcapsule structures need to endure crack-free the thermal and shear stress of the compounding process. The desired release of the encapsulated oil-droplets is initiated by mechanical wear-damage at the surface of the part in the later application. The continuous release of small amounts of oil during the lifetime of the products offers an effective way to reduce frictional forces in the interface of the tribo-system.
The use of oil-filled microcapsules is predominantly described in epoxy resin-systems [16][17][18][19][20][21][22]. Guo et al. have encapsulated lubricant oil with poly(melamine-formaldehyde) and the resulting microcapsules were mixed epoxy-resins using different amounts [16]. Reaching a microcapsule concentration of 10 wt.-%, the coefficient of friction and wear decreased by 75 % and 98 %, respectively in comparison to the unfilled material. The work also shows, that a rising amount of microcapsules decrease the mechanical characteristics [16]. This could limit the use of microcapsules in some applications. Khun et al. investigated epoxy-based composites containing wax filled microcapsules and short carbon fibres (SCF) [18]. The results indicated that by additionally added SCFs the mechanical properties increased. Composites containing 10 wt.-% of microcapsules lead to lower friction and wear from up a level of 8 wt.-% SCF. The authors explain this with the 2.7 increased hardness of the material and the additional friction-reducing effect of SCF. [18]. The incorporation of oil-filled microcapsules in thermoplastic matrices has been described for thermoplastic polyurethane (TPU), polypropylene (PP) and polybutylenterephthalat (PBT) [13,[23][24][25]. Producing thermoplastic composites requires often higher thermal stability of the capsule wall than incorporating them in thermosets. For all these polymeric matrices, modified with capsules, a distinct decrease in tribological coefficient-values was measured in several tribological tests. Schoch et al. investigated polyglycol filled polyurethane microcapsules in POM. Rotating ball on disc tests showed a decrease of 70 % in COF and 84 % in wear volume. In their work, a systematic variation of capsules concentration has not been conducted to investigate tribological and mechanical properties. The quantity of publications on POM/microcapsules-composites is very low and only studied by the Schoch working group. This is remarkable, because POM is a widely used semi crystalline thermoplastic material for tribological applications and it's also being modified with solid state lubricants alike PTFE powder [12].
The aim of this study is to show the effect of microencapsulated food-grade oil using two different wall materials in engineered thermoplastics. The microcapsules were mixed, using a laboratory-scale compounder for POM and PBT with varying capsules concentrations. Based on rotating tribological steel ball on plate tests, the coefficient of friction and wear rates were evaluated. Mechanical properties were analysed using tensile tests (DIN EN ISO 527 at room temperature).

Materials
In order to also enable the possibilities of use in areas of food technology, a low-viscosity, appropriately qualified oil was chosen: "Food Lube R-5555" (Riedel Schmierstoffe, Germany) was encapsulated and used in the wall materials, mentioned above. This lubricant is based on synthetic oils and modified with PTFE. It is certified for food and pharmaceutical applications. Two different thermoplastic materials were chosen as matrix materials: • Kepital F30-03 (POM copolymer, produced by Korea Engineering Plastics, Seoul, South Korea). This natural coloured and easy-flowing grade is developed for injection moulding applications • PBT resin Toraycon 1200M, by Toray Industries (Tokio, Japan), also suitable for injection moulding processing.
Material properties (based on the material data sheet) of both polymeric materials are listed in Table 1.

2.2.
Synthesis of melamine resin and polyurethane microcapsules containing lubricant oil MF microcapsules containing Food Lube were synthesized by in-situ polymerization ( Figure. 1a). The synthesis process involved the preparation of the aqueous phase containing 0.12 g polyvinylalcohol (MW = 31,000 g•mol -1 ) and 4 g MF-resin that were dissolved in 150 g deionized water. The organic phase (12 g Food Lube) was dispersed in the aqueous phase under mechanical stirring (rotor-stator type of a stirrer, Ultra-Turrax) at 5000 rpm and 60 °C. The pH was adjusted to pH 4 and the dispersion was stirred for 2 h at 60 °C and 350 rpm to complete polycondensation reaction. The capsules were separated from side products via filtration. The resulting wet cake was spread on the tray and dried on air for 48 hours.
PU Microcapsules containing Food Lube were prepared by interfacial polycondensation (Figure 1b). The oil phase (8 g Food Lube, 2 g PMDI and 6 g cosolvent) were dispersed in 50 g 1,5 wt.-% PVP solution containing 5 g glycerol under mechanical stirring (rotorstator type of a stirrer, Ultra-Turrax) at 5000 rpm at room temperature. The suspension was further stirred with a glass blade stirrer at 500 rpm. 0,5 g DABCO dissolved in 5 g water were added to the suspension, which was further heated to 75 °C and stirred for another 2 hours for the completion of the interfacial polymerisation. The resulting wet cake was spread on the tray and dried on air for 48 hours.

Preparation of self-lubrication thermoplastic composites
The microcapsule-loaded thermoplastic composites were prepared on a counter-rotating micro-conical twin screw extruder (HAAKE MiniLab II, Thermo Fisher Scientific, Germany). In this extruder, compounding with a low amount of material is possible. Nevertheless, in contrast to industrial co-rotating twin screw extruders, the material must be fed into the hopper as a premix. To enhance the mixing process, the polymer granules were milled on centrifugal mill (ZM 200, Retsch, Germany) to a polymeric powder. Before compounding, PBT was dried for 6 h at 80 °C in a convection oven to a residual moisture of 0.02 %. Subsequently, the polymeric and microcapsule powders were pre-mixed manually. The two capsule materials each have a specific oil content, so that although there are different proportions of capsules in the compound formulation, there is the same proportion of oil in the compound. The compositions of the samples are listed in Table 2. 5 g of each sample composition were produced per batch. Several batches per capsule concentration were required to produce tensile bars and discs. A heated melt reservoir was used to transport the polymer melt out of the nozzle to a pneumatic laboratory injection moulding mechanism. Two different tools were used: • Disc-form samples (diameter: d=25 mm; thickness: s=1.65 mm) were used for tribological test. • Tensile rods (type DIN EN ISO 527-2 "5A") were used for mechanical characterization.
The rotational speed of the screws was set to 100 rpm for both polymers. The machine settings for sample production were kept constant for each polymer type. Temperatures for each polymer are listed in Table 3. Although processing in the minilab is a laboratory method, experience has shown that it can be seen as an indicator for transfer to a larger scale such as the co-rotating twin screw extruder [14].

Thermal stability
The thermal stability of the lubricant and microcapsules was investigated using TGA2 LF/1100/885 from Mettler with STARe Software. The weight of the specimen varied from 5 to 10 mg. All the measurements were performed under nitrogen atmosphere in aluminium oxide pan. The samples were heated with 10K•min -1 to 550 °C.

Scanning electron microscopy
The microcapsule powder and the processed polymer-microcapsule compounds were analysed via scanning electron microscopy (SEM) (Supra 40 VP, Carl Zeiss, Germany). SEM images of the compounds were made on fracture surfaces of the tensile test specimens.

Capsule size distribution
Laser Diffraction particle size analysis was used for the determination of the average microcapsule diameter using LS 13 320 SW (Beckmann Coulter). Low angle forward light scattering with optional PIDS (polarization intensity differential scattering) technology was used.

Tribology
The tribological investigations were performed on a rotational ball-on-disc setup at norm conditions (23 °C, 50 % rel. humid.). As counter body a 100Cr6 steel ball with a radius of 3 mm and a roughness of Ra 0.032 µm was used. Normal load and linear speed of rotation was kept constant at 10 N and 50 cm•s -1 . Each compound (n=3) was tested over a distance of 5 km and a track radius of 8 mm. Microscopical analysis on the steel ball did show any wear phenomena. The wear loss of the analysed tribological system is only described by the width of the wear track of the polymer part. The wear track width of the polymer discs was measured at 5 positions via light-microscopic images for each of the 3 samples. The calculation of the wear volume has been done according to ASTM G99-17. Equation (1) was used to determine the specific wear rate given by equation (2).

Mechanical properties
The influence of the microcapsules on the mechanical properties were characterised based on tensile test (Z010, Zwick/Roell, Ulm, Germany) according to DIN EN ISO 527. The strain was measured using a multiXtens-system. The specimens were tested with a speed of 50 mm•min -1 (Young's modulus at 1 mm•min -1 ) and a pre-force of 1 N. Young's modulus, stress at break, and elongation at break were tested and analysed.

Microcapsules
In figure  The particle size analysis ( Figure 3) confirms the results of the SEM analysis and demonstrates that SEM images can be considered representative. Thus, both types of capsules show rather broad particle size distributions with the small fraction of particles with diameters below 5 µm, which were also observed in SEM. The particle size distribution maxima are around 20 µm and 30 µm for MF and PU microcapsules, correspondingly.  Thermogravimetric analysis (TGA) was used to determine the thermal stability of the microcapsules under inert atmosphere. It can be seen ( Figure 4) that Food Lube itself has an outstanding thermal stability. Almost no thermal decay can be observed until ca. 340 °C. The slight weight loss (ca. 2 wt.-%) below 120 °C can be attributed to the loss of moisture, that is probably the result of storage of samples in open pans at ambient humidity before characterisation. Due to their excellent thermal stability both types of microcapsules were considered promising candidates for the integration into the chosen polymer matrixes by extrusion.

Sample preparation
Up to a lubricant concentration of 10 wt.-%, the sample preparation was done with constant process parameters using POM and PBT.

PBT
Mechanical properties of microcapsule filled PBT are shown in Figure 13. The incorporation of microcapsules leads to decreased stress at break (Figure 13 a). With increasing lubricant concentration, tensile strength decreases. The reduction of tensile strength is less, when MF-microcapsules are used. By incorporation of 10 wt.-% of lubricant tensile strength is 23 % (MF) and 31 % (PU) lower compared to unfilled PBT. The results for elongation at break are displayed in Figure 13 b. The unfilled PBT shows a high deviation in the elongation at break value. Lubricant containing PBT show a tendency of breaking at lower elongation rates. The amount of lubricant and the type of capsule wall material used have no influence on the elongation at break level.

Discussion
The investigations showed that it is possible to encapsulate additivated food grade oil with MF and PU microcapsule wall materials by in-situ polymerization (MF-capsules) and interfacial polymerization (PU capsules). MF-capsules have a nearly spherical shape and the morphology of the PU shells showed concavities. The concavities are believed to be the result of uneven cross-linking of the polymer on the still liquid droplet interface, which leads to the tensions and shape deformation before and during the capsule solidification process.
MF microcapsules show nearly gradual loss of mass until 300 °C (about 8 % in total), after which complete thermal decomposition of the microcapsules takes place in a 2step process between 300 °C and 500 °C. The weight loss below 300 °C is attributed to the evaporation of residual water (about 2 %) and post-condensation of melamine resin [26]. The post-condensation of the MF capsule walls at the temperatures between 120 °C and 200 °C is believed not to lead to the mechanical disruption of the microcapsules, but to their further cross-linking and densification. The first decomposition step starting at ca. 310 °C is assigned to the MF shell decomposition, which is further accompanied by the evaporation of the capsule core material. Interestingly, the decomposition of the Food Lube in this case happens at lower temperatures than of the Food Lube in nonencapsulated state. This can be attributed to the higher available surface area in case of the microencapsulated material and corresponding faster kinetics of evaporation.
PU microcapsules showed no weight loss up to ca. 300 °C. The stepwise weight loss above 300 °C is assigned to the decomposition of PU shell followed and further accompanied by the evaporation of Food Lube. The PU microcapsules appear to be slightly more thermally stable than MF microcapsules, which can be attributed to different moisture content, different thermal decomposition patterns for polymers used as the shell materials and different microcapsule size [27]. Besides the thermal properties, mechanical properties of the microcapsules are of great interest of the different shell materials. These values can be used to better predict the fracture behaviour in processing and later application.
The leakage of oil in the composite preparation at higher capsule concentrations could have been initiated by higher shear forces on the capsules. The observed phenomena of higher stability and less oil-release during the sample preparation of PU capsules at higher concentrations can be explained, based upon a higher stability of the PU capsules. Keller and Sottos investigated the mechanical stability of microcapsules used for selfhealing polymers [28]. They showed that the capsule size has a significant effect on failure strength and that smaller capsules can withstand higher loads. Even if PU capsules have a larger diameter, they showed higher stability in the compounding process. This could be explained by the higher wall thickness of the PU-capsules. At scale-up production conditions, higher capsules concentrations can be added through enhanced process engineering. Schoch et al. compounded lubricant filled microcapsules by adding the capsules to the molten polymer via side-feeder close to the nozzle of a twin-screw extruder [14]. Compared to the claimed mechanical properties of the plastic producers (Table 1), the samples produced in this work show slightly smaller values for POM and PBT. This can be attributed to the smaller standard sample geometries and the production on a laboratory scale. As the lubricant content increases, the influence of the matrix plastic used decreases. With a lubricant concentration of 5 and 10 wt.-% stress at break is at a comparable level for the plastic matrices used. The elongation at break values for both plastic types show very similar results. The addition of microcapsules leads to an elongation at break of between 10-20 %, regardless of the type and proportion of microcapsules. In further studies the effects of additional reinforcing fillers are of great interest in POM and PBT. Several studies have shown that mechanical values increase and the COF decreases when carbon fibres or inorganic fillers are added [13,17,18,29,30].
Under the chosen tribological conditions, unfilled POM shows lower tribological values than unfilled PBT. In contrast to POM, even a small proportion of microcapsules has a high impact on the tribological properties of PBT. With increasing capsule concentration, the differences in tribological values between POM and PBT compounds become smaller. However, the values of the POM composites remain lower than those of PBT for all concentrations. Under these sliding conditions, POM/microcapsule composites are to be preferred.
Further tribological tests are mandatory to evaluate the effectiveness of lubricant filled microcapsules in thermoplastic engineered materials. Investigations in different tribological systems and test conditions are of great importance to get deeper insights in the tribological behaviour of the composites. Comparative tribological tests with different internal lubricants can show under which conditions the respective use of lubrication additives is most suitable. Scale-up production are of importance to generate a higher sample quality and to investigate the injection moulding behaviour of microcapsule filled composites.

Conclusion
In this study, two capsule shell materials (melamin-formaldehyde resin and polyurthane resin) were used to generate lubricant filled microcapsules. Sample production of POM and PBT composites were performed by micro-twin-screw extruder. Up to an oil concentration of 10 wt.-%, no leakage of oil was detected. Mechanical properties revealed decreasing mechanical properties with increased lubricant concentration. Tribological characterization showed that lubricant filled microcapsules can effectively decrease the friction (-70 %) and wear values (-58 %). Taking mechanical properties into account, a capsule concentration of 5 wt.-% offered sufficient friction and wear reduction in POM and PBT. Test results of the two different types of microcapsules showed only small deviations in their tribological and mechanical behaviour and are both suitable for tribological applications. Further tribological investigations are required to extend the understanding of the friction and wear behaviour of the rarely investigated lubricant-filled microcapsules in thermoplastic materials.