Surface energetics, dispersion, and nanotribomechanical behavior of POSS/PP hybrid nanocomposites

Abstract

Hybrid organic–inorganic polymer nanocomposites incorporating polyhedral oligomeric silsesquioxane (POSS) nanoparticles are of increasing interest for high performance materials applications. Octaisobutyl POSS/polypropylene nanocomposites were prepared at varying POSS concentrations via melt blending. The interplay of POSS molecular geometry, composition, and concentration in relation to the tribological, nanomechanical, surface energy, and bulk properties of the nanocomposites were investigated. Ultra-low friction and enhanced hardness, modulus, and hydrophobicity were observed for the nanocomposite surfaces, with minimal changes in the bulk thermomechanical properties. Parallel AFM, SEM, TEM, and spectroscopic analyses demonstrated significant differences in POSS distribution and aggregation in the surface and the bulk, with preferential segregation of POSS to the surface. Additionally, contact angle studies reveal significant reduction in surface energy and increase in hysteresis with incorporation of POSS nanoparticles. The differences in bulk and surface properties are largely explained by the gradient concentration of POSS in the polymer matrix, driven by POSS/POSS and POSS/polymer interactions. © 2007 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 45: 2441–2455, 2007

INTRODUCTION

High performance hybrid polymer nanocomposites (HPNC's) with improved surface properties, specifically improved tribological performance, such as reduced friction and wear, have significant potential applications ranging from microelectronic devices and aeronautic applications to low friction fibers for bandages and other prosthetic applications.1 Conventionally, fluorinated materials such as polytetrafluoroethylene are used for low friction applications; however, these materials are immiscible with most polymers, difficult to process, costly, and subject to environmental concerns. These practical challenges motivate the development of novel strategies for developing cost effective, processable, and non-halogenated low friction surfaces. In an attempt to address these challenges, we have studied HPNC's of polyhedral oligomeric silsesquioxane (POSS) nanoparticles in a polymeric matrix prepared via melt blending techniques and evaluated their surface and bulk properties.

A fully condensed POSS molecule has a nanostructured core of an inorganic silicon–oxygen–silicon (Si-O-Si) network frame surrounded by a corona of organic moieties (R) attached to the corner silicon atoms. The diameter of a fully extended POSS molecule varies from one to several nanometers, depending on the composition of the substitutents.2, 3 POSS based HPNC's inherit their merits from the robust inorganic POSS cages combined with processable organic polymers. Over the last decade POSS has been copolymerized with multiple monomer systems to produce a wide range of thermoplastics and thermosets with POSS molecules incorporated as an integral part of the polymer chain.1–9 The majority of these research efforts are focused on improving thermomechanical performance of the polymer matrix through uniform molecular level POSS nanoparticle dispersion achieved by directly incorporating POSS moieties into the polymer chain.9, 10 Thermomechanical properties are reported to be strongly influenced by the structure and concentration of the POSS moieties, POSS-POSS interactions and POSS–polymer interactions. In general, hybridization of an organic polymer matrix with inorganic POSS results in improved thermal and mechanical properties along with increase in oxidative and flame resistance.3–8, 11 Recently Nanda et al.12 synthesized well dispersed polyurethane/POSS hybrid materials utilizing a solution polymerization process. Ordered morphologies with homogenously distributed POSS domains (∼100–150 nm) were observed. The authors reported significant increases in physical properties, including tensile strength, storage modulus, complex viscosity, surface hydrophobicity, and glass transition temperature. Nanostructure self assembly, extent of crosslinking during assembly, control of aggregation, spatial distribution of nanoscopic POSS building blocks, molecular dynamics and Monte Carlo simulation studies related to the interchain dynamics of POSS copolymers have also been reported.6, 13–15

Fewer studies have been reported, in which POSS nanoparticles are physically dispersed in the polymer matrix utilizing high shear melt mixing processes.16–19 As in the case of hybrid materials prepared via chemical incorporation of POSS, improvements in thermomechanical properties are reported for melt-blended systems. Again, the changes in the properties are highly dependent on the POSS structure and its interactions with the polymer matrix. In most of the studies, it was reported that POSS incorporation does not affect the crystallization behavior of the polymer matrix.20 However, in some studies it was reported that incorporation of certain types of POSS molecules induce polymorphism in selected polymer systems.18 Morphology studies of hybrid composites prepared via melt blending also showed mixed results regarding dispersion of POSS. Most of the studies reveal non-homogenously dispersed POSS crystalline aggregates ranging from a few hundred nanometers to several microns.6, 18

Recently, studies on surface properties of POSS HPNC's have appeared in the literature, and these are focused mainly on the surface hydrophobicity of POSS based copolymers,21, 22 fluorinated POSS,23 and fluorinated POSS/ fluorinated polymer composites.24 All of these studies report increased hydrophobicity on incorporation of POSS nanoparticles. Takahara and coworkers25 evaluated surface dewetting characteristics of polystyrene-POSS hybrid films prepared via solution dispersion of POSS and PS in a common solvent. They report segregation of POSS to the film surface. Fukuda and coworkers26 also reported a higher concentration of POSS moieties on the film surface for a poly (methylmethacrylate) (PMMA)/POSS system. The authors synthesized a tadpole shaped hybrid polymer with an inorganic head of fluorinated POSS and an organic tail of PMMA, blended the hybrid polymer with PMMA in solution, and prepared a film via spin coating. While surface property investigations reported to date have focused primarily on hydrophobicity of POSS HPNC surfaces, the tribological and nanomechanical properties of POSS/polymer nanocomposite surfaces have not been explored in detail. Additionally, little has been reported for surface properties of melt-blended systems, and most studies have focused on solution-blended systems.

As the size of engineering devices moves towards miniaturization, understanding of surface properties is critical, and control of tribological performance such as friction and wear is of particular importance for devices containing moving parts. Friction is an aggregate effect arising from physical phenomena, such as adhesion, viscosity, capillary forces, surface chemistry, and electrostatic interactions; and any one of these properties can dominate friction performance depending upon the operating conditions. Friction behavior of a multicomponent system is influenced by complex interactions between the components, their geometry and their relative contribution to surface roughness. To develop a model of friction for thin films andsystems containing nanoparticles, friction measurements at nanoscale are invaluable in developing a fundamental understanding of these complex interactions. In this study we have utilized nanoprobe lateral force microscopy (LFM) and nanoindentation to investigate the nanoscale surface tribological and mechanical behavior of Octaisobutyl (Oib)-POSS/Polypropylene (PP) HPNC's.

Analytically Amonton's law describes the friction at macro scale (eq 1). According to this law, friction coefficient (μ) is the ratio of the friction force (F_f) and the total normal force (F_n).27, 28

• (1)

Polymeric materials deviate from this law because apart from the surface roughness and chemical nature of the sliding surface, frictional force also depends on the adhesion properties of the sliding surface. Surface tension as well as the viscoelastic nature of the polymeric materials also contribute significantly to the friction forces.29 The adhesive force component (F_a) when added to the external applied load (F_l) gives the total normal force. The overall relative coefficient of friction (COF) can be best described by eq 2.

• (2)

Relative COF is obtained from the slope of a plot of friction force as a function of applied load.30

Nanomechanical property measurements are based on the contact mechanics of an axisymmetric indenter with an elastically isotropic half space, utilizing the method developed by Oliver and Pharr.31 Hardness values (H) are calculated as:

• (3)

P_max = maximum applied load

A = contact area between the probe and the specimen

Reduced modulus (E_r) values are taken from the slope (dh/dP) of the unloading portion of the force curve and are dependant upon the contact area by the relation:

• (4)

h = depth of penetration

P = applied load

In this investigation, nanocomposites of octaisobutyl-POSS (Oib-POSS) with PP were prepared at varying concentrations via melt blending. Complimentary microscopy and spectroscopy techniques were employed to evaluate nanoscale dispersion of POSS particles in thin polymer films. Surface and bulk properties of the HPNC's were analyzed with specific focus on nanotribological behavior. The interplay of POSS molecular geometry, composition, and concentration in relation to the nanotribomechanical properties and surface energy was investigated to develop an understanding of the nanotribological behavior of these HPNC's, with the ultimate goal of developing non-halogenated low friction surfaces with controlled POSS dispersion and maintained performance properties in the bulk.

EXPERIMENTAL

Materials

Isotactic PP (Hival® 2420) was purchased from General Polymers (Cincinnati, OH) and Octaisobutyl POSS (MSO825) (Oib-POSS) was provided by Hybrid Plastics (Hattiesburg, MS). Oib-POSS was received as a crystalline white powder. All materials were used as received unless mentioned specifically. The chemical structure of the Oib-POSS molecule is shown in Figure 1.

Figure 1. Schematic structure of octaisobutyl POSS.

Sample Preparation

Melt blends of Oib-POSS with PP were prepared utilizing a CT-25 corotating twin screw extruder (B&P Processing, screw diameter = 25 mm and L:D ratio of 44:1). Blends were prepared at 0, 5, and 10 wt % Oib-POSS in PP. Samples were extruded at 225 °C and 300 rpm. To obtain smooth films for evaluation of the surface, samples were prepared by melt pressing the extrudate between two clean silicon wafers. For evaluations of the bulk, samples were prepared by cryomicrotoming pellets at −90 °C, using a diamond knife.

Atomic Force Microscopy Surface Topography and Friction

Atomic force microscopy (AFM) surface topography and relative friction studies were conducted on a MultiMode™ scanning probe microscope from Veeco Instruments, (Santa Barbara, CA). Probes were purchased from Veeco Probes, (Santa Barbara, CA). A silicon probe with 125-μm long silicon cantilever, nominal force constant of 40 N/m and resonance frequency of 275 KHz was used for tapping mode surface topography studies. A triangular silicon nitride (Si₃N₄) probe, with a nominal cantilever spring constant of 0.58 N/m, was used for relative surface friction studies. AFM studies were conducted under ambient conditions in a temperature (27 °C) and humidity (40–45%) controlled room. All samples were stored in a humidity controlled chamber and measurements were conducted in the same day to minimize environmental effects. Surface topographies of the melt pressed surfaces and the cryomicrotomed bulk regions were studied on 5 μm ×5 μm scan areas with an image resolution of 512 × 512 pixels at a scan rate of 1 Hz. Multiple areas were imaged and figures show representative morphology. Nanoscale relative surface friction studies were performed via AFM in lateral force mode on 5 μm × 5 μm scan areas. Force-distance curves and friction loops were obtained at different set points to obtain statistical data for analysis. Surface roughness analysis was performed using Nanoscope version 5.30 r2 image analysis software. To verify the reproducibility, two sets of readings were taken for each sample and an average value is reported. The differences between the two readings were less than 1%.

Scanning Electron Microscopy-Energy Dispersive X-Ray

To qualitatively confirm the presence of POSS molecules on the surface and in the bulk of HPNC's samples, scanning electron microscopy, and elemental mapping was performed using an FEI Quanta 200 Scanning Electron Microscope (SEM) in environmental scanning mode, coupled with a Thermo-Noran Vantage light element energy dispersive X-ray detector. Characterization of all the elements except hydrogen was obtained by X-ray spectroscopy under electron flux. The SEM images were obtained at a voltage 20 kV and pressure of 1 torr.

Transmission Electron Microscopy-Energy Dispersive X-Ray

Bulk morphology of Oib-POSS/PP nanocomposites was investigated using a JOEL-2100 ultra high resolution transmission electron microscope (Joel, Tokyo, Japan) at an accelerating voltage of 200 kV. Ultrathin sections, about 90-nm thick, were prepared by cryoultamicrotoming the samples with a diamond knife using a Leica EM UC6 cryomicrotome at −90 °C. Subsequently, the ultrathin sections were collected on a 600-mesh copper grid. Elemental composition of the ultrathin sections was analyzed by mapping the carbon, oxygen and silicon elements using energy dispersive X-ray analysis. Elemental maps were acquired using EDAX Genesis software.

Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy

To analyze surface composition of the nanocomposites, Micro ATR-FTIR spectroscopy measurements were conducted on the melt-pressed surfaces using a Bio-Rad FTS-6000 FTIR single-beam spectrometer with 4 cm⁻¹ resolution. The surfaces of each specimen were analyzed using a 2-mm germanium (Ge) crystal with a 45° angle maintaining constant contact pressure between the crystal and the specimens.

Nanoindentation Studies

Nanoindentation was performed on the melt pressed film surfaces to determine the nanoscale relative hardness and reduced modulus of the Oib-POSS/PP nanocomposites. The Triboindenter (Hysitron, Minneapolis, MN) was operated with a three-sided diamond (Berkovich type) tip, calibrated on fused silica. Nanoindentation was performed under closed loop with load control using a compliance method, in which the force-displacement curves were obtained during loading and unloading cycles. As the indenter presses into the surface, the displacement is recorded continuously as a function of the applied load. A single indent was made on every surface by applying a normal load with a loading rate of 50 μN/s. Unloading rate was maintained similar to loading rate and a 5 s. hold time was provided at the maximum force of 2000 μN. Total cycle time for the load control indentation was 85 s. Indentation tests were conducted at different regions across the surface to confirm the homogeneity of POSS distribution on the sample surface.

Contact Angle and Surface Energy

Static and dynamic contact angle measurements were conducted using the sessile drop technique by a ramè-hart goniometer coupled with DROPimage® data analysis software. Small drops of water (10 μL) were dropped onto a flat surface and the image of the drop was captured. Contact angle hysteresis was calculated by measuring advancing and receding contact angles using the tilting plate technique. Five measurements were taken and averaged. Surface energy was calculated by measuring the contact angle between the sample surface, deionized water, and diidomethane (CH₂I₂) as a second test fluid, utilizing the Fowkes et al.32 and Owens–Wendt method.33

Dynamic Mechanical Analysis

DMA was used to investigate the thermal transitions, relaxation behavior, and dynamic storage modulus of Oib-POSS/PP nanocomposites. Thermal transitions were recorded using a TA Q800 dynamic mechanical analyzer over a temperature range of −50 to 175 °C at a heating rate of 2 °C/min and an oscillation frequency of 1 Hz.

Modulated Differential Scanning Calorimetry

Modulated DSC, in which a modulation of ±1 °C was performed every 60 s, was utilized to obtain precise thermal data. A TA Instruments Q Series DSC Q100 was used to determine the melting (T_m) and crystallization temperature (T_c) of the samples. Each sample, ∼10-mg weight, was analyzed under a nitrogen blanket. Samples were heated initially from −50 to 250 °C at a rate of 10 °C/min to erase their thermal history. This heating cycle was followed by cooling the samples from 250 to −50 °C at the rate of 5 °C/min to study their crystallization behavior. Finally, melting behavior was studied by further reheating the samples from −50 to 250 °C at a programmed rate of 10 °C/min.

RESULTS AND DISCUSSION

AFM images of the Oib-POSS/PP nanocomposites are shown in Figures 2 and 3. Incorporation of POSS in the PP matrix leads to dramatic modification of the polymer surface, as revealed by tapping mode AFM images of the Oib-POSS/PP blends (Fig. 2). While the surface of the neat PP sample, shown in Figure 2(A), is smooth with no apparent surface features (root mean square roughness, RMS, of 1.1 nm), the Oib-POSS/PP blends exhibit raised features with increased surface roughness [Fig. 2(B): 5% Oib-POSS blend, RMS = 7.2 nm, Fig. 2(C): 10% Oib-POSS blend, RMS = 12.1 nm]. These raised features are attributed to the presence of POSS aggregates, whose presence is further substantiated by EDAX and ATR-FTIR analyses presented later in this section. Similar features are observed in bulk Oib-POSS/PP, which are absent in neat PP samples (Fig. 3). Furthermore, analysis of the morphology of the surface in comparison to that of the bulk indicates that there is preferential segregation of the POSS aggregates to the surface. AFM phase imaging provides insight about the distribution of nanostructured POSS domains based on the differences in localized stiffness and modulus. In Figure 3 AFM phase images of melt pressed surfaces of neat PP [Fig. 3(A)] and the 10% Oib-POSS/PP blend [Fig. 3(B)] are presented in comparison to phase images of bulk samples prepared via cryomicrotoming [neat PP bulk in Fig. 3(C), 10% Oib-POSS/PP bulk in Fig. 3(D)]. The neat PP surfaces are featureless, whereas the POSS blends show raised spherical/oblong features that are presumed to arise from the presence of POSS aggregates. Image analysis (Table 1) indicates that the features have a mean length of 38 nm and mean width of 21 nm on the surface, whereas the features in the bulk show a mean length of 61 nm and mean width of 23 nm. The greater elongation and the greater variability of particle size in the bulk may be a result of microtoming effects. While these oblong features are present both on the surface and in the bulk of the melt-pressed sample, the features are more widely distributed and lower in concentration in the bulk microtomed sample, indicating preferential segregation to the surface of the composite. As the POSS aggregates are of the size of several to less than 100 nm, we will refer to the systems as nanocomposites in the remainder of the discussion.

Figure 2. Three-dimensional AFM tapping mode height images and surface roughness analysis of melt pressed (A) neat PP, (B) 5% Oib-POSS/PP, and (C) 10% Oib-POSS/PP (Z scale: 150 nm, R_q: root mean square roughness, R_a: mean roughness, H_max: maximum height).

Figure 3. AFM phase images of melt pressed (A) neat PP surface, (B) 10% Oib-POSS/PP surface, and microtomed (C) neat PP bulk, and (D) 10% Oib-POSS/PP bulk.

Table 1. AFM Particle Size Analysis of Melt Pressed 10 wt % Oib-POSS/PP Surface and Bulk Area

Dimension	Mean	Minimum	Maximum	Sigma
Surface
Length (nm)	37.9	13.8	161.7	22.8
Width (nm)	20.8	13.8	86.5	9.3
Bulk
Length (nm)	60.9	13.8	773.7	70.8
Width (nm)	22.8	13.8	209.7	19.5

Similar oblong features are observed in SEM images of Oib-POSS/PP nanocomposites, and SEM/EDAX analysis provides further evidence of POSS enrichment at the surface in the nanocomposites. SEM images with EDAX evaluation of melt-pressed surfaces and microtomed samples of the bulk are shown in Figure 4. As observed in the AFM analysis, the PP melt-pressed surface appears smooth in the SEM image [Fig. 4(A)], whereas the 10% Oib-POSS/PP nanocomposite exhibits raised spherical features covering the surface. The POSS molecules appear to form a network assembly with interconnected spherical beads on the surface. We attribute this network morphology to the POSS intermolecular attractions. Elemental mapping via EDAX provides further insight into the chemical identity of these spherical features present on the surface and in the bulk. This analysis reveals a strong signal attributed to silicon for the POSS-containing samples that is absent in the neat PP sample. For the microtomed bulk sample, the spherical features are more widely distributed and at lower concentration, with concomitant reduction in the silicon signal in the EDAX analysis. Consistent with the information obtained from AFM phase imaging, SEM micrographs and differences in the silicon peak intensity observed in the EDAX analysis indicate that POSS molecules have an affinity for the surface. In recognition of the fact that EDAX probes the sample from the surface to several microns depth, complimentary tools such as ATR-FTIR were utilized to validate these findings.

Figure 4. SEM/EDAX mapping of melt pressed (A) neat PP surface, (B) 10% Oib-POSS/PP surface, and microtomed (C) 10% Oib-POSS/PP bulk nanocomposite.

TEM imaging coupled with EDAX analysis of the POSS nanocomposites reveals further information about the POSS dispersion in the nanocomposite. Widely dispersed oblong particles ranging from 10 to 100 nm in length are observed in a microtomed sample of the 10% Oib-POSS/PP nanocomposite (Fig. 5). Elemental mapping reveals that the areas where the particles appear are rich in silicon and oxygen, and poor in carbon. Areas away from the particles exhibit high carbon concentration with low silicon and oxygen concentrations. These findings provide further confirmation that the observed particles are indeed POSS aggregates.

Figure 5. TEM-EDAX elemental surface mapping of a microtomed 10% Oib-POSS/PP sample (A) carbon, (B) oxygen, (C) silicon, and (D) silicon overlap.

ATR-FTIR analysis provides further evidence of POSS surface enrichment (Fig. 6). For reference purposes, IR spectra of Oib-POSS and neat PP are shown in Traces A and B, respectively. Characteristic absorbances are observed in the pure Oib-POSS trace at 1109 cm⁻¹, attributed to Si-O-Si stretching vibrations, and at 1230 cm⁻¹, corresponding to CH₂ symmetric stretching vibrations and SiC symmetric vibrations. These absorbances are absent in the neat PP spectrum. The characteristic 1109 and 1230 cm⁻¹ absorbances are observed in the 5% Oib-POSS/PP (Trace C) and 10% Oib-POSS/PP composites, with strong intensity signals indicating the presence of POSS on the surface. Absorbances at 2916 to 2950 cm⁻¹ corresponding to CH₂ asymmetric stretching vibrations are observed in all samples, as would be expected based on the structure of the materials.34, 35

Figure 6. ATR-FTIR spectra of (A) Oib-POSS, (B) neat PP, (C) 5% Oib-POSS/PP, and (D) 10% Oib-POSS/PP.

In parallel to the morphological and compositional changes observed for the POSS/polymer nanocomposites, dramatic changes in surface mechanical and physical properties were observed. LFM was employed to evaluate nanoscale relative friction and adhesion of the nanocomposite materials. As described in detail in the introduction section, the probe is scanned across the surface and the frictional force it experiences in contact with the surface is measured in LFM. By evaluating the frictional force as a function of the applied normal force, a relative COF is obtained. Real time friction loops for neat PP and Oib-POSS/PP nanocomposites at equivalent loading force are shown in Figure 7. The distance between the extending and retracting friction curves is a qualitative measure of the friction between the probe and the surface.36 Surface friction measurements were conducted at a fixed scan rate of 1 Hz, consistent with methods commonly reported in the literature for polymer samples.30, 36 POSS nanocomposites exhibit reduced relative friction in comparison to the neat polymer, with friction decreasing as a function of increasing POSS concentration. In Figure 8 plots of friction force as a function of applied normal force are shown, and high correlation coefficients are achieved for all systems. Relative COF decreases with increasing POSS concentration, from a value of 0.17 for neat PP to 0.07 for the 10% Oib-POSS/PP nanocomposite [The LFM COF of Teflon is reported as 0.03. (ref.37)]. Note that relative values of COF are reported, using the nominal force constant for the probe as described in the experimental section. Although this will not provide absolute values of force of adhesion and friction force, comparison of the relative values is valid as the same cantilever is employed to obtain the measurements.30, 36, 38 The samples were imaged in tapping mode after LFM measurements to ensure that surfaces were not damaged during the friction studies, and no artifacts were observed on the images. Additionally, AFM probes were imaged via SEM after surface friction measurements to ensure that there were no visible changes to the geometry of the probe during scanning. Sliding friction is determined by multiple structural and mechanical factors, including adhesion, roughness, plowing, capillary forces, heterogeneity on the surface, shear stress, surface hardness, and true area of contact.39–41 We attribute the observed reduction in nanoscale surface friction for the POSS nanocomposites to the interplay of a number of these factors, particularly increase in surface hardness and modulus, discussed in later sections. Another important parameter is the observed increase in surface roughness for the nanocomposite surfaces. Increase in the surface roughness reduces the real area of contact between the AFM probe and the surface, thereby reducing the observed friction. Some authors have suggested a self-lubricating “nano ball bearing” mechanism for low friction nanomaterials, including POSS,42 diamond nanoparticles,43 and fullerene-like inorganic nanoparticles.44

Figure 7. Friction loops obtained via lateral force microscopy at applied normal force of 28 nN for (A) neat PP, (B) 5% Oib-POSS/PP, and (C) 10% Oib-POSS/PP.

Figure 8. Friction force as a function of applied normal force measured via LFM. COF (μ) is obtained from the slope, adhesive force (F_a) is obtained from the intercept. (A) neat PP, (B) 5% Oib-POSS/PP, and (C) 10% Oib-POSS/PP.

Nanoscale friction characteristics are also significantly influenced by the magnitude of the adhesive force between the AFM probe and the surface. The relative force of adhesion (F_a) between the probe and the surface decreases for the POSS-containing nanocomposites in comparison to the neat PP (Fig. 8). This reduced adhesion may in part explain the reduced friction for the POSS nanocomposites. The adhesive force between the hydrophobic PP and the relatively hydrophilic AFM probe is small. Incorporation of the hydrophobic Oib-POSS nanoparticles increases the surface hydrophobicity and further reduces the adhesive interaction between the surface and the AFM probe, with a 50% reduction in measured relative F_a for the 10%Oib-POSS/PP nanocomposite. The reduction in relative adhesive force is indicative of reduced sticking and sliding friction between the surface and the AFM probe.

Friction behavior depends not only on the chemical nature and the adhesion between the surfaces in contact, but also on the mechanical properties, such as the relative hardness and modulus of the materials. According to contact mechanics theories, surface friction is directly proportional to the shear stress and the true area of contact.39, 40 Study of the surface nanomechanical properties of the Oib-POSS/PP nanocomposites provides mechanistic insights for the improved friction properties of these HPNC's. Nanoindentation evaluation revealed increased surface hardness and modulus for the POSS nanocomposites. Force/displacement curves for the materials are shown in Figure 9, while maximum penetration, reduced modulus and hardness values are given in Table 2. Surface hardness and reduced modulus increase as a function of increasing POSS concentration, whereas maximum penetration depth decreases. At 5 wt % POSS concentration, surface hardness and modulus increase by 40%, and further increase by 100% at 10 wt % POSS concentration. Thus, incorporation of low percentages of Oib-POSS results in dramatic surface hardening of the PP nanocomposite. POSS nanoparticles, because of their robust inorganic silicon oxygen structure, provide nanoscale reinforcement to the PP matrix. These findings can be further correlated to the reduced surface friction exhibited by the POSS HPNC's in LFM studies. Surface friction is related to the hardness of the surface and the ability to resist formation of wear particles. Because of their high surface hardness and modulus, these HPNC's exhibit greater resistance to plastic deformation, scratching, and wear compared with the neat PP matrix. These combined mechanical property factors help to explain the observed low friction coefficients for the POSS nanocomposites.

Figure 9. Nanoindentation force as a function of displacement measured using the Hysitron. (A) neat PP, (B) 5% Oib-POSS/PP, and (C) 10% Oib-POSS/PP.

Table 2. Maximum Penetration Depth, Reduced Modulus, and Hardness of Oib-POSS/PP HPNC's

Sample	Max. Penetration (nm)	Reduced Modulus (GPa)	Hardness (MPa)
Neat PP	1000	1.9	109
PP+ 5% Oib-POSS	840	2.7	157
PP+ 10% Oib-POSS	699	3.9	225

Surface modification as a function of POSS nanoparticle concentration was also observed in contact angle studies. Contact angle measurement provides insight into the hydrophobicity as well as the surface energy of the film. Surfaces with water contact angle less than 90° are considered wetting whereas those with water contact angle greater than 90° are considered nonwetting.45 When a liquid drop is placed on a flat and smooth surface, it spreads over the surface until the mechanical and thermodynamic forces are balanced. Work of adhesion (W_A) is expressed by Young's eq 5.

• (5)

where γ_LV, γ_SV are liquid–vapor, solid–vapor interfacial tension, respectively, and θ is contact angle. Work of adhesion is used to calculate the polar (γ^p_SV) and dispersive components (γ^d_SV) of surface energy using Fowkes et al.32 and Owens–Wendt33 geometric mean formula, eq 6.

• (6)

Static contact angles obtained with water and diiodomethane, along with surface energies and polar and dispersive components calculated by the Owens–Wendt method,33 are given in Table 3. Surface hydrophobicity increases and surface energy decreases with increasing POSS concentration. Incorporation of 10% Oib-POSS yields a 43% reduction in the surface energy, to a value of 24 mN/m, and a 27% increase in the water contact angle, to 99°, approaching reported values of 16 mN/m surface energy and 120° contact angle for Teflon.46 Increase in the water contact angle demonstrates the hydrophobic nature of the Oib-POSS/PP nanocomposite surface. This can be explained in part by the cumulative effect of the eight hydrophobic isobutyl groups attached to the corner silicon atoms of the POSS cage. The effect is pronounced because of the higher concentration of Oib-POSS moieties on the film surface. Increments in surface hydrophobicity may also be related to the surface roughness,47, 48 which was observed by AFM roughness analysis to increase with increasing POSS concentration. The ultra low friction coefficients observed for these materials may also be related to their low surface energy. Low surface energy of POSS nanoparticles causes them to preferentially migrate towards the surface. This factor coupled with weak interactions between the Oib-POSS and the PP matrix partly explains the observed gradient in POSS distribution from bulk to the surface. These findings are consistent with previously reported studies showing increased surface hydrophobicity for other POSS/polymer systems and POSS molecules with fluorinated alkyl chains attached to the POSS cage.12, 21–24

Table 3. Static Contact Angle and Surface Energy of Oib-POSS/PP HPNC's

Sample	Contact Angle (°)		Surface Energy, γ (mN/m)	Polar, γ^p (mN/m)	Dispersive, γ^d (mN/m)
	Water	Diiodomethane
Neat PP	78.1	43.3	42.2	4.1	38.1
PP+ 5% Oib-POSS	89.0	51.6	35.1	2.0	33.1
PP+ 10% Oib-POSS	99.0	69.7	24.2	1.4	22.8

Surface dynamics were further studied via dynamic water contact angle analysis. The advancing (θ_a) and receding (θ_r) water contact angles were measured and their difference, the contact angle hysteresis (θ_h), was calculated (Table 4). Contact angle hysteresis increases with increasing POSS incorporation. Hysteresis is influenced by many factors, including the surface roughness, chemical heterogeneity of the surface, presence of low molecular weight species, and molecular orientation.49 In a multiphase system advancing contact angle is more sensitive to the low surface energy or hydrophobic domains whereas receding contact angle is more sensitive to the high surface energy or hydrophilic domains.50 We attribute the increased hysteresis in the POSS containing systems to the surface roughness as well as the chemical heterogeneity imparted by the enrichment of Oib-POSS nanoparticles on the PP surface. These factors coupled with the low surface energy characteristic of Oib-POSS, which promotes its migration towards the surface, explains the hysteresis behavior exhibited by Oib-POSS/PP nanocomposites.

Table 4. Advancing (θ_a), Receding Water Contact Angle (θ_r), and Hysteresis (θ_h) of Oib-POSS/PP HPNC's

Sample	θa (°)	θr (°)	θh (°)
Neat PP	77.25	72.95	4.3
PP+ 5% Oib-POSS	86.30	78.20	8.1
PP+ 10% Oib-POSS	96.70	86.10	10.6

While microscopy, spectroscopy, nanotribology, nanomechanical, and contact angle analyses all indicate dramatic modification of surface properties for these nanocomposites, bulk properties are only minimally affected by incorporation of POSS in the PP matrix. DSC analysis showed only minimal changes in T_c and T_m for POSS nanocomposites in comparison to neat PP, with a 2° reduction in the thermal transition temperatures for 10% incorporation of Oib-POSS (Table 5). Similarly, DMA evaluation showed minimal changes in thermomechanical behavior. Loss factor (tan δ) is shown in Figure 10. Two relaxation processes are observed, the β-transition at low temperature (−10 to 35 °C) and the α-transition at higher temperature (50–140 °C). The β-transition is generally attributed to the glass transition temperature, whereas the α-relaxation process is associated with the relaxation of restricted amorphous chains in the crystalline phase of the polymer.51 Incorporation of POSS leads to a broadening of the β-transition peak and a small reduction in the calculated transition temperature from 15 °C for neat PP to 12 °C for 10% Oib-POSS/PP. The α-transition peak, on the other hand, is sharper for the POSS-containing nanocomposites and the transition temperature increases slightly from 81 °C for the neat PP to 85 °C for 10% Oib-POSS/PP.

Figure 10. DMA thermograms of Oib-POSS/PP nanocomposites.

Table 5. DSC Crystallization and Melting Temperatures of Oib-POSS/PP Hybrid Nanocomposites

Sample	Tc (°C)	Tm (°C)
Neat PP	115	167
PP+ 5% Oib-POSS	114	166
PP+ 10% Oib-POSS	113	165

Storage modulus obtained via DMA is shown in Figure 11. A slight increase in the storage modulus of the nanocomposites is observed in comparison to that of the neat PP. An ∼ 10% increase in storage modulus is observed in the glassy regime for the 10% Oib-POSS/PP composite, while the observed increase in reduced modulus of the surface measured by nanoindentation was 100% for the same nanocomposite (Table 2). These findings again illustrate the dramatic effect of surface segregation of the Oib-POSS in the PP matrix. Similar preferential migration is reported for incompatible polymer blends.52–55 The mechanism of self-stratification for incompatible blends can be explained on the basis of the degree of incompatibility between the two phases and surface energy.^56–58 In principle, a similar mechanism of surface segregation is expected for the Oib-POSS/PP blend. POSS moieties preferentially segregate towards the film-air surface as a consequence of the strong thermodynamic driving force to minimize the surface energy. The nanoscale size of the POSS molecules, as well as the degree of control and opportunity to tune the interactions through substituents on the POSS cage, differentiate these materials from conventional additives.

Figure 11. DMA storage modulus of Oib-POSS/PP nanocomposites.

In this article, melt mixing of a commercial POSS molecule in a commercial PP matrix resulted in nanocomposites displaying dramatic increase in surface hardness and modulus, with parallel increase in hydrophobicity, decrease in surface energy and decrease in measured COF. In fact, measured friction, hydrophobicity, and surface energy levels approach that of Teflon, in a completely non-halogen system. Bulk properties, on the other hand, were minimally affected. These findings indicate that the incorporation of a small amount of POSS of the correct composition allows desired modification of surface properties, without dramatic alteration of bulk polymer performance. Conversely, altering the chemical composition of the POSS substituents and adjusting processing parameters will provide further modification of bulk properties of the nanocomposite.

Overall dispersion and domain size of POSS in the polymer matrix is a function of various inter- and intramolecular forces acting between POSS-POSS moieties as well as between POSS and the polymer matrix. Dispersion and compatibility of POSS in the polymer matrix can be controlled by the proper selection of organic (R) groups attached to the corner silicon atom. Furthermore, various processing parameters, including shear forces and temperature profile, during the melt extrusion process also play crucial roles in improving dispersion of POSS particles in the polymer matrix. In this system, octaisobutyl substituted POSS demonstrates intermolecular forces that result in small aggregates, while providing enough compatibility with the PP matrix to generate well-dispersed nanoparticle aggregates in the range of 10–100 nm, schematically illustrated in Figure 12. As observed from the microscopic studies, this schematic represents the differences in the POSS distribution and aggregation in bulk versus surface, thereby manifesting POSS concentration gradient between the bulk and surface. Additionally, the hydrophobicity and low surface energy characteristics of the POSS aggregates drive their segregation towards the surface of the nanocomposite. This research work contributes to the existing understanding of surface properties of POSS/polymer HPNC's through correlation of tribological and nanomechanical behavior with POSS structure, composition, and dispersion behavior. These findings to date indicate that POSS gradient concentration can be precisely controlled through correct control of the POSS chemical structure, polymer matrix, and melt processing conditions, to produce nanocomposites in a cost effective process with desired surface and bulk properties for a wide range of important technical applications.

Figure 12. Schematic representation of bulk versus surface distribution of POSS nanoparticles in Oib-POSS/PP nanocomposite.

CONCLUSIONS

POSS/PP HPNC's were successfully prepared via high shear melt mixing and their bulk and surface properties investigated. POSS aggregates ranging in size from 10 to 100 nm were observed dispersed in the PP matrix via AFM, SEM/EDAX and TEM/EDAX analysis. Dramatic modification of surface properties was demonstrated. Incorporation of 10% Oib-POSS yielded a 60% reduction in relative COF, from 0.17 to 0.07, a doubling of hardness (109–225 MPa) and reduced modulus (1.9–3.9 GPa) measured by nanoindentation, and an increase in water contact angle from 78° to 99°. Bulk property evaluations, on the other hand, showed only minimal changes on incorporation of POSS, with a 10% increase in modulus measured by DMA for 10% POSS/PP nanocomposites. Combined microscopy and spectroscopy analysis demonstrated preferential segregation of the POSS aggregates to the surface in comparison to the bulk region. The POSS concentration gradient helps explain observed differences in surface and bulk properties of the nanocomposites.

The ultra low nanoscale friction demonstrated by the POSS nanocomposites is a function of both the structural features of the Oib-POSS nanoparticles and the surface nanomechanical properties. The low friction is attributed to the interplay of a number of factors, with the enhanced surface hardness and modulus reducing surface friction by providing resistance to plastic deformation, surface damage and production of wear particles. Further, the increased surface roughness, resulting in reduced contact area between the AFM probe and the surface, in combination with the demonstrated high hydrophobicity and reduced adhesion contribute significantly towards reducing surface friction. The enhanced tribological performance of these materials, combined with their high surface hardness and hydrophobicity indicate their potential utility for applications such as micro/nanoelectronic devices requiring ultra low friction and wear performance. Precise control of surface properties is indicated through optimization of POSS structure and POSS/polymer interactions.

Acknowledgements

This work was supported primarily by the STTR program of the National Science Foundation under Award Number OII-0539295. This work was also supported by the major research instrumentation program of the National Science Foundation under Award Numbers MRI0421406, DMR0421403, and DMR0215873.