Polymer nanocomposites have drawn a great deal of interest in recent years because these materials possess high potential to achieve great property improvement by adding a small amount of nanoparticles in the polymer matrices. Plastic foams, on the other hand, represent a group of lightweight materials that have been widely used in a variety of industries with a market value of US $2 billion in 2000. However, the foam applications are limited by their inferior mechanical strength, poor surface quality, and low thermal and dimensional stability. Furthermore, the most widely used chlorofluorocarbon (CFC) blowing agents have been found to cause ozone depletion in the upper atmosphere and will be banned by 2010, according to the Montreal Protocol. Environmentally benign gases such as supercritical carbon dioxide (ScCO2) are attractive alternatives for CFCs as blowing agents. But low solubility and high diffusivity of CO2 in polymers make it more difficult to control the foam morphology. A small amount of well-dispersed nanoparticles in the polymer may serve as nucleation sites to facilitate the bubble nucleation process. Plate-like nanoparticles can also reduce gas diffusivity in the polymer matrix. In addition, the presence of nanoparticles may enhance mechanical and physical properties, the heat distortion temperature, and fire resistance of polymer foams. Novel nanocomposite foams based on the combination of functional nanoparticles and supercritical fluid foaming technology may lead to a new class of materials that are ligh weight, high strength and multifunctional. In this article, we review the recent progress in this area. Polymer nanocomposites cover a vast array of different polymer matrices and nanoparticles.Adetailed survey on this topic is beyond the scope of this paper. The readers are referred to dedicated reviews for details . We will first briefly reviewthe synthesis and processing techniques of nanocomposites based on polymers that are important in the foam industry. Both thermoplastic and thermoset nanocomposite foams will be addressed. This is followed by an introduction of foaming processing methods. The effect of nanoparticles on the foam morphology and properties is then discussed. We conclude with current and future trends of nanocomposite foams in both industrial and biomedical applications.
Polymer nanocompositesPolymer composites are widely used in automotive, aerospace, construction, and electronic industries because they provide improved mechanical properties (e.g., stiffness, strength) and physical properties over pure polymers. Micron-sized particulates and long fibers are most widely used in traditional polymer composites. Nanocomposites are a new class of materials providing superior properties when compared to their microcomposite counterparts. An addition of a small amount of nanoparticles can significantly improve a variety of properties without sacrificing the lightweight of polymer matrices.
Nanocomposites usually refer to composites in which at least one phase (the filler phase) possesses ultrafine dimensions (on the order of a few nanometers). They include the use of three different types of nanoparticles as shown in Fig. 1. The first type of nanoparticles only has one dimension in the nanometer scale. They possess a platelet-like structure. The lateral dimension may be in the range of several hundred nanometers to microns, while the thickness is usually less than a few nanometers. Clay is a good example of this type of nanoparticle. Layered nanographites are another example. If two dimensions of the nanoparticles are at the nanometer scale while the third is larger, these particles possess an elongated structure. Nanotubes and nanofibers belong to this group. The third type of nanoparticle has all three dimensions at the nanometer scale; for example, spherical silica particles, nanocrystals, gold and other metal nanoparticles, and block copolymers. A variant of this type of particle is the nanoporous microparticles. While the diameter of the particle may be in the order of microns, the pore sizes are in the order of nanometers. While all three types of nanoparticles have been used in nanocomposite synthesis and processing, plate-like clay nanoparticles and fiber/tube like carbon nanofibers (CNFs) and carbon nanotubes (CNTs) have attracted the most attention. The nanocomposites and nanocomposite foams discussed in this article are mainly based on
these nanoparticles.
Polymer foams
Foams are defined as materials containing gaseous voids surrounded by a denser matrix, which is usually a liquid or solid. Foams have been widely used in a variety of applications: e.g., insulation, cushion, absorbents and weight-bearing structures. Foams of high porosity with interconnected pores have also been used as tissue engineering scaffolds for cell attachment and growth . Various polymers have been used for foam applications, e.g., polyurethane (PU), polystyrene (PS), polyolefin (polyethylene (PE) and polypropylene (PP)), poly(vinyl chloride) (PVC), polycarbonate (PC), just name a few. Table 1 displays the US market for polymer foams by resin family in 2001 and 2006, and the projected growth rate for each resin family. In 2001, the US use of polymer foam products was 7.42 billion pounds and that transfers into a $16.2 billion market. PU occupies the largest market share (53%) in terms of the amount consumed, while PS is the second (26%).
Depending on the composition, cell morphology and physical properties, polymer foams can be categorized as rigid or flexible foams. Rigid foams are widely used in applications such as building insulation, appliances, transportation, packaging, furniture, flotation and cush-ion, and food and drink containers, whereas flexible foams are used as furniture, transportation, bedding, carpet underlay, textile, gaskets, sports applications, shock and sound attenuation, and shoes. According to the size of the foam cells, polymer foams can be classified as macrocellular (>100 lm), microcellular (1–100 lm), ultramicrocellular (0.1–1 lm) and nanocellular (0.1–100 nm).
Polymer foams can also be defined as either closed cell or open cell foams. In closed cell foams, the foam cells are isolated from each other and cavities are surrounded by complete cell walls. In open cell foams, cell walls are broken and the structure consists of mainly ribs and struts. Generally, closed cell foams have lower permeability, leading to better insulation properties. Open cell foams, on the other hand, provide better absorptive capability.
Synthesis of nanocomposite foamsOne-step reactive foaming is typical for thermoset polymers. A good example is PU/clay nanocomposite foams, where a physical blowing agent such as pentane is mixed with monomers and clay nanoparticles. Reaction exotherm leads to a temperature jump and foaming. Most thermoplastic nanocomposite foams to date are synthesized via a two-step process: the nanocomposite is synthesized first and followed by foaming. We shall briefly review the two steps separately.
Nanocomposite synthesis
The goal of the synthesis of nanocomposites is to achieve controlled nanoparticle dispersion and distribution in a polymer matrix. This is not a trivial task and is still a largely unaccomplished task. For example, each individual clay platelet offers a large surface area and high aspect ratio, and both are critical for improving mechanical properties, thermal stability and barrier properties. However, typical clay particles contain a large number of crystallites. The crystallite itself contains hundreds of individual layers stacking together. A typical clay particle may contain several thousand individual layers. Separating and distributing these layers throughout polymer matrices require the development of special surface chemistry, and dedicated synthesis and processing technologies. The most widely used strategy is to bind surfactants onto the clay surface in order to increase the hydrophobicity and compatibility through ion exchange reactions with the polymer matrices. Studies on polymer-CNTs and polymer-CNFs composites also show that CNTs/CNFs are inclined to hold together as bundles and ropes in the polymer matrix due to the intrinsic van der Waals attraction. Strategies proposed to accomplish good dispersion include the use of ultrasonication, high shear mixing, surfactants, and functionalization of the carbon surface. Typically, three approaches have been adopted to synthesize polymer nanocomposites: solution blending, melt blending, and in situ polymerization. In solution blending, a solvent or solvent mixture is used to disperse the nanoparticles and dissolve the polymer matrix.
Depending on the interactions of the solvent and nanoparticles, the nanoparticle aggregates can be disintegrated in a good solvent due to the weak van der Waals force that stacks the layers together. Polymer chains can then be adsorbed onto the nanoparticles. However, upon solvent removal, the nanoparticles tend to re-agglomerate. Few exfoliated nanocomposites were prepared via this method. Another disadvantage of this method is the large amount of solvent needed, resulting in a high product cost. The types of polymers that can be used to sythesize nanocomposites ultimately depend on the selection of proper solvent, limiting the applicability of this method. Nevertheless, this is an attractive route to prepare nanocomposites based on water soluble polymer and layered silicate nanocomposites because most water soluble polymers are polar and hydrophilic enough to interact with the silicate surface without the need of cation exchange modification on the silicate surface.
It is well known that inorganic layered silicates are able to exfoliate in water and form colloidal particles. Several polymer nanocomposites, including polyethylene oxide (PEO), polyvinyl alcohol (PVA), and polyacrylic acid (PAA) were prepared via this method. Some polymer/CNFs nanocomposites are also synthesized by this method. Instead of using solvent as the medium, nanoparticles can be directly mixed with a molten polymer. This process eliminates the use of solvent and is compatible with industrial polymer extrusion and blending processes. It offers an economically attractive route in fabricating polymer nanocomposites. A wide variety of polymer/clay nanocomposites have been prepared via this route,i.e., nylon 6, PS, and PP. Melt intercalation offers a 'simple' way of preparing nanocomposites. However, care has to be taken to 'fine tune' the layered silicates surface chemistry in order to increase the silicate compatibility with the polymer matrix. Many studies have shown that the polar interactions of polymer and clay surface play a critical role in achieving particle delamination/dispersion . For non-polar polymers, e.g., PP, a polar compatibilizer such as maleic anhydride modified PP (PP-MA) is commonly added to improve the compatibility of PP and clay and thus the clay nanoparticle dispersion. All reported studies on PP nanocomposite foams were synthesized in this manner. Processing conditions such as shear rate and mixing have profound effects on the structure evolution of polymer nanocomposites by melt intercalation and these effects are still not well understood. Polymer/CNFs nanocomposites have also been synthesized via this method. Shear stress is extremely important to disintegrate and disperse nanoparticles, therefore it needs to be controlled at an appropriate level. Too strong a shear force tends to break single fibers into many shorter pieces reducing the reinforcement efficiency and deteriorating properties in which high aspect ratio of the nanoparticle is essential.
Unlike melt intercalation, layered silicates is mixed with monomer before polymerization takes place in situ polymerization. Because of the low monomer viscosity (comparing to melt viscosity), it is much easier to achieve uniform mixing of particles in the monomer using a high shear mixer. In addition, the low viscosity and high diffusivity result in a higher rate of monomer diffusion into the interlayer region. It is also possible to control nanocomposite morphology through the combination of reaction conditions and clay surface modification. For most thermoset polymers, in situ polymerization is the only viable method to prepare nanocomposites. By tailoring the interactions between the monomer, the surfactant, and the clay surface, exfoliated nanocomposites (e.g., nylon-6, poly(e-caprolactone), epoxy and polycarbonate) have been successfully synthesized via the ring-opening polymerization. The functional group in the organic cation can catalyze the intralayer polymerization and facilitate layer separation. Free radical polymerization has also been employed to synthesize many thermoplastic nanocomposites. Efforts have been made to anchor initiators in the interlayer region to improve the intralayer polymerization rate for exfoliated nanocomposites. Reactive groups containing carbon-carbon double bonds were introduced to the clay surface via several approaches. Clay exfoliation/delamination has been dramatically enhanced this way. In our laboratory, a nanoclay was prepared by the ion exchange of a reactive cationic surfactant, 2-methacryloyloxyethylhexadecyldimethyl- ammonium bromide (MHAB, structure shown in Fig. 2) with cations on the montmorillonite surface. Compared to a commercial nanoclay, Closite 20A from Southern Clay, which contains a non-polar aliphatic chain with a similar length, the anchored organic surfactant with polymerizable groups on MHABS provides an additional kinetic driving force for layer separation. The TEM micrograph (Fig. 2(a)) of the intercalated PS/20A nanocomposite demonstrates that large clay aggregates are still present in the matrix. Face-to-face layer stacking and low angle intergrowth of tactoids is still observable. A TEM micrograph of PS/MHABS is shown in Fig. 2(b). The tactoids have been completely delaminated and uniformly dispersed in the matrix. Most clay layers are present as single layers, while stacks of a few layers are also observable in some region. Near complete exfoliation was reported for PS nanocomposites synthesized with this reactive nanoclay at a clay concentration of 20 wt%.
CNTs/CNFs nanocomposites have also been synthesized via in situ polymerization . The system viscosity needs to be high enough in order to fix the fibers in the monomer phase in the early stage of polymerization. Otherwise, CNFs would be inclined to bind with each other, causing a reduction of particle dispersion and other physical properties. Fig. 3 shows a complete dispersion of CNFs (the dark lines) in the PS matrix. Here, 10 wt% of PS was added into the mixture of styrene/CNFs to achieve a higher initial viscosity and consequently a more stable fiber suspension. In the case of PS/CNTs nanocomposites, the dispersion of carbon nanotubes is not easy. The same shear mixing and ultrasonication used to disperse 1%wt CNFs was not enough to separate 0.1% CNTs aggregates in styrene. PS, PVC and polyolefins are the three primary thermoplastics used in polymer foams. Synthesis of nanocompositesbased on these three polymers is briefly introduced next.
Synthesis of PS nanocomposites
PS/clay nanocomposites were synthesized in both intercalated and exfoliated structures. To prepare the nanocomposites, organo-nanoclay particles are pre-mixed with PS and then mechanically blended in single or twin-screw extruders. The formation of nanocomposites relies on the penetration of polymer chains into the interlayer region to separate the layers. The layer separation depends on the establishment of favorable interactions between the polymer and the clay surface and the subsequent system energy reduction. Limited interactions resulted in limited polymer chain penetration, leading to intercalated nanocomposites. In situ polymerization has also been carried out to prepare PS nanocomposites. By using reactive surfactants, the copolymerization of the interlayer surfactant and styrene monomer provides the driving force for delamination of clay crystallite. Highly exfoliated PS/ clay nanocomposites have been formed this way.
Synthesis of PVC nanocomposites
PVC is a widely used matrix material in thermoplastic foams. Some efforts have been devoted to develop PVC nanocomposites for improved PVC properties. Particles used include clay, calcium carbonate, hydrosulfite, copper, and antimony trioxide The polar nature of the C–Cl bond makes it possible to form exfoliated nanocomposites of PVC in melt blending. However, clay surface modification is usually needed to achieve exfoliation. For example, exfoliated PVC nanocomposites have been realized by melt blending when the clay was modified by aromatic amine However, melt processing of PVC with alkyl-modified clays only produced partially exfoliated nanocomposites. A plasticizer like dioctylphalate (DOP) may serve as a co-intercalate to increase clay dispersion in PVC. In situ polymerization of PVC/clay nanocomposoites has been carried out by either emulsion polymerization or suspension polymerization. In general, in situ polymerization methods can achieve a much better clay dispersion. Highly exfoliated PVC/clay nanocomposites can also be produced by flocculating a mixture of polymer and clay mineral dispersions or by solution blending. However, organoclay tends to induce the degradation of PVC because of its low thermal stability. The allylic and tertiary chlorines of PVC chains are the main labile sites to release hydrogen chloride. When the mass loss due to dehydrochlorination reaches 0.1%, sequences of conjugated double bonds form, resulting in discoloration of PVC by the zipelimination mechanism. In PVC nanocomposites, the quaternary ammonium salts used for clay modification may accelerate the degradation of PVC. The organic ammonium cations act as Lewis acid, causing chlorine ions to separate from the PVC matrix and to release HCl. Several approaches can be used to reduce or eliminate PVC degradatio when processed with nanoclay. One approach is to cointercalate DOP into organoclay and then compound the mixture with PVC. DOP covers the quaternary amine groups preventing the contact between amine and active chloride atoms. Another approach is to intercalate/exfoliate nanoclay in a polymer, such as epoxy or poly caprolactone (PCL) which has good miscibility with PVC, by in situ polymerization. The clay surface is protected by a layer of epoxy or PCL to prevent the direct contact with PVC in melt blending, inhibiting its degradation.
Synthesis of polyolefin nanocomposites Polyolefin represents a large group of thermoplastic polymers. Melt intercalation usually results in immiscible microcomposites. This is because polyolefin is very nonpolar and hydrophobic while clay is highly hydrophilic. The enthalpic barrier prevents the intercalation of polymer chains. In melt intercalation, usually a third functionalized ingredient is added to serve as the compatibilizer in order to improve the compatibility of polyolefin and clay surface. In situ polymerization of polyolefins in the presence of clay has been conducted by several research groups. Tudor et al. used a synthetic fluorohectorite to prepare PP nanocomposites. In their study, a cationic metallocene catalyst was incorporated into the gallery region to catalyze propylene polymerization. Another cationic palladium based catalyst was incorporated in 1-tetradecylammonium cations modified synthetic fluorohectorite. This complex can catalyze ethylene polymerization at high yield, producing exfoliated nanocomposites . Exfoliated PE nanocomposites were also prepared using metallocene catalyst. However, the formed nanocomposite is not thermodynamically stable. During melt processing, the exfoliated structure tends to collapse into an intercalated structure.
Synthesis of thermoplastic nanocomposite foams
The synthesized nanocomposites can be used to produce nanocomposite foams. For large-scale production, the direct utilization of foaming (blowing) agents is the most commonly used method. Foaming agents are substances that form the gaseous phase in the foams. Two types of foaming agents are often used: physical or chemical blowing agents. Chemical foaming agents are usually reactive species that produce gases in the foaming process, while physical foaming agents are substances that gasify under foaming conditions. Typical physical foaming agents are volatile chemicals such as chlorofluorocarons (CFC), volatile hydrocarbons and alcohols, or inert gases such as nitrogen, carbon dioxide, argon, and water. Foams can be produced in either the liquid/melt state by extrusion, injection molding or compression molding, or the solid state where gas is forced into a solid polymer followed by depressurization. Both methods can use either physical or chemical blowing agents. In addition to utilizing a blowing agent, porous polymeric materials can also be prepared by other methods such as phase inversion, leaching and thermal decomposition. Most of these methods are only suitable for preparation of thin film products. We will focus on the foaming processes utilizing foaming/blowing agents. In physical foaming, a blowing gas is first dissolved in the polymer to form a homogeneous mixture. This is usually done through pressurization. Subsequent pressure release or temperature increase results in a supersaturation state, and gas starts to form nuclei and expand. The traditional CFC physical blowing agents have good solubility in the polymer matrix, low diffusivity and low thermal conductivity that result in foam products with good insulation and physical properties. However, their use has been greatly reduced globally because of their high ozone depletion effect. In recent years, the foam industry has switched to blowing agents containing less chlorine atoms by replacing them with hydrogen and fluorine atoms (i.e., HCFCs). The HCFCs are hydrogen-containing chlorofluorocarbons such as HCFC-141b (CCl2FCH3), HCFC-142b (CF2ClCH3), and HCFC-22 (CHF2Cl) that have lower ozone-depletion potential (ODP). North America (United States and Canada) is currently using HCFC-142b (CH3-CF2Cl2) for extruded polystyrene (XPS) foam production. HCFCs will be phased out by 2010. Other choices of foam blowing agents include hydrofluorocarbon HFCs such as HFC-134a (CH2F-CF3) and HFC-152a (CH3-CHF2); hydrocarbons (propane, butane, pentane, etc.); inert gases such as nitrogen, argon and CO2; and water. HFCs do not destroy the ozone, but they have a negative impact on global warming and their applications will most likely be regulated in the near future. Hydrocarbons present some serious problems, such as a greater fire hazard in closed-cell foams due to the entrapped blowing agent as well as adding to VOC emissions. Precaution must be taken to ensure safety of utilizing alkynes, which can be explosive in production. Among the rest, CO2 is the most favorable choice because it is inexpensive, non-toxic, and environmentally benign (zero Ozone Depletion Potential, and 100 year Global Warming Potential compared to 1300 years for HFC- 134a and 2000 years for HCFC-142b). Many challenges must be met to enable the use of CO2 as a blowing agent. The three primary issues are: (1) The low solubility of CO2 in most polymer melts – for instance, the solubility of CO2 in PS is only about 3.5% at 150 °C and 10 MPa pressure; however, a solubility of 5–6% is required to achieve the necessary cell growth. (2) CO2 has a high diffusivity in the polymer melt due to its small size – while this ensures a fast mixing process, it also results in quick escape of gas from the foam after processing. (3) CO2 has a higher gas thermal conductivity in comparison with that of HFC blowing agents. Blending CO2 with other hydrocarbons, such as CO2/2-ethyl hexane and CO2/ethanol leads to solubility similar to HCFCs, but this suffers from the same drawbacks of hydrocarbons. Nanocomposites offer several potential advantages in the foaming processes. The surface of nanoparticles can be modified to increase intermolecular interactions with CO2. The plate-like nanoparticles can improve the barrier property by slowing down the CO2 diffusion. Nanoparticles may also enhance the nucleation rate in the foaming process. Microcellular foams, which are characterized by foams with cell size less than 10 lm and cell density higher than109 cells/cm3, have shown many promising properties compared to conventional foams, which usually have a cell size around 100 lm and cell density less than 106 cells/cm3. But they require very stringent conditions to produce, e.g., extremely high pressure and high pressure drop rate. This greatly limits the processing window and the attainable size of the foam products. The presence of nanoparticles may overcome this bottleneck.
Non-continuous foaming
Foams can be produced by both non-continuous processes, e.g., batch foaming, injection molding foaming, and continuous processes such as extrusion foaming. In a batch foaming process, materials are first saturated with the foaming agent under certain temperatures and pressures. If the temperature is higher than the glass transition temperature, Tg, of the polymer matrix, the release of pressure would result in supersaturation and cell nucleation and growth. Cell structure is usually fixed by cooling the materials below its Tg. When the saturation temperature is lower than Tg, the cell is not able to nucleate and grow after the release of pressure even if gas is in the supersaturation state because of the glassy nature (high rigidity) of the matrix. Foaming may occur when temperature is raised above Tg. Cell structure is again fixed by cooling. The latter method allows independently manipulate saturation and the foaming condition, leading to higher process flexibility. However, during the transfer of gas-saturated materials to the high temperature environment, diffusion would inevitably occur, leading to a thick skin region. The major factors that determine the cell density are the saturation temperature, pressure, and pressure drop rate, and in the latter case, the temperature jump rate. Batch foaming is usually carried out at temperatures far below the polymer flowing temperature. The saturation time is very long (from hours to days depending on diffusivity). This greatly limits the productivity. Recently, a semi-continuous process was developed. In this process, a roll of polymer sheet with a gas channeling material (a flexible and highly permeable material) interleaved between the layers of polymers is saturated in a pressurized chamber with the foaming gas, usually under room temperature. Pressure is then released. The saturated polymer roll is then separated from the channeling material and pulled through a heating station to foam the polymers. This technique is similar to the second batch foaming method mentioned above (the saturation temperature is lower than Tg, but foaming occurs when temperature is raised above Tg) but can yield higher productivity. The majority of studies on nanocomposite foams were conducted via batch foaming. Polymers used include PS, PP, PLA, PC, and PVC. Batch foaming of expanded plastics were also investigated. Expanded polypropylene (EPP) is comprised of thermoplastic PP and an embedded low-boiling hydrocarbon compound as a blowing agent. When heated, there is an increase in the volume due to vaporization of the blowing agent. EPP is used for wings and fuselages for combat flying or in automotive applications (e.g., bumpers, dashes, etc.), that require effective impact energy absorption. It is also widely used in building,transport containers and packaging components. The foams made by EPP beads can bond together, essentially presenting a closed cellular structure. They exhibit greater strength and multi-cycle impact durability than other polymer foams. Because of its thermoplastic nature, EPP facilitates re-use of scrap and recycling after use. Its low density helps meet impact requirements using less material. Expandable polystyrene (EPS) foams are also studied and widely used as insulation materials in building. Most ordinary white coffee cups and packaging foams are examples of EPS. Expanded-bead foam is unsuitable for structural core material due to its poor mechanical properties. Compared with the extrusion foaming process, foam injection molding has its own advantages to produce parts with complex geometry. Currently, foam injection molding using CO2 as the foaming agent is applied to produce lightweight products with high mechanical strength. MuCell molding technology, invented by Suh et al., has been successfully commercialized by Trexel to injection mold microcellular foams after a series of modifications on certain components of a standard reciprocating-screw injection molding machine, such as the plasticizing unit, injection unit, hydraulic unit, clamping unit, and gas delivery unit. To produce microcellular foams, a new screw designed for better mixing and a new sealed barrel with gas injectors were used. The injection unit requires a fast injection speed to achieve the high pressure drop rate. It was found that a finer cell structure and more uniform cell size distribution can be achieved by controlling the pressure drop rate at the mold gate than at the injection nozzle. Of course, the injection speed should be controlled below the shear limit to prevent the melt fracture. Generally, foaming injection molding achieves increased melt flowability, smaller shot size, lower injection pressures, faster cycle times, and greater dimensional stability and weight savings in molded parts. Nylon 6/clay nanocomposite microcellular foams were foamed on an injection-molding machine equipped with a commercially available supercritical fluid (SCF) system. The results show that microcellular nanocomposite samples exhibit smaller cell size and uniform cell distribution as well as higher tensile strength compared to the corresponding base PA-6 microcellular samples. Among the molding parameters studied, shot size has the most significant effect on cell size, cell density, and tensile strength. The minimum cell size was achieved at the medium shot size. Conceivably, the larger the shot size, the slower the cooling rate, and thus the longer the coolingtime promotes more cell growth. On the other hand, with the smaller shot size, the cells have more space to grow, which is in favor of larger cell sizes. When the same shot size is used or the same amount of material is injection molded, the finer and denser microcells in samples usually lead to higher impact strength and less reduction in tensile strength, unless some defects such as coalescence and open cells occur in the cell formation. PP/clay nanocomposites were also foamed via the MuCell process.
Continuous foaming
Continuous extrusion foaming is the most commonly used technology in the foam industry. Both single- and twin-screw extruders can be used for plastic foaming. In a typical extrusion foaming process, the foaming gas is first injected into the barrel and mixed with the polymer to form a homogenous solution. When the homogenous polymer/gas mixture passes through a die, a rapid pressure drop induces phase separation and cell nucleation. Pressure drop, and especially the pressure drop rate, is the primary driving force for cell nucleation. An extra shaping die is used to control the product shape and foam expansion. The foamed materials continue to expand until the extrudate temperature is lower than Tg and the foam product is vitrified. Extrusion foaming of PS/clay was conducted with both intercalated and exfoliated nanocomposites using CO2 as the foaming agent. The effects of processing parameters, i.e., mass flow rate and pressure drop rate on the foam morphology, have been investigated. Due to the viscosity increase in the nanocomposites, a higher pressure drop rate is realized with the use of nanocomposites under the same processing conditions, thus greatly enhancing the nucleation rate. Extrusion foaming of PS/CNFs nanocomposite has also been reported recently.
Synthesis of thermoset nanocomposite foams
There are far fewer studies on thermoset nanocomposite foams than thermoplastics counterparts. Most literatures are concerned with PU nanocomposite foams. Some are related to polyisocyanurate and phenolic nanocomposite foams. Several patents on PU nanocomposite foams claim significant property improvement, such as improved compressive strength, thermal insulation, and fire retardance. However, few details were provided in the literature. To prepare thermoset nanocomposite foams, nanoparticles are first dispersed uniformly in one or more monomers. The mixture is then foamed by adding other monomers. Foaming agents could be either physical or chemical blowing agents. Similar to the synthesis of thermoplastic nanocomposites, the surface modification of nanoparticles is essential for nanoparticle dispersion. In most studies, the layered silicates were modified with functional surface modifiers that can react with one of the reactants to form an intermediate leading to a uniform nanoparticle distribution in the polymer matrix during foaming.
Morphology and propertiesTwo basic steps are involved in the foaming process: bubble nucleation and growth. Nucleation is the process in which a new phase (bubble phase) is generated from the initially homogeneous polymer-gas mixture. Growth is the process in which bubble nuclei grow into final bubbles. Both processes are affected by many physical properties such as viscosity, gas solubility, surface tension, and glass transition temperature. To make things more complicated, these properties are inter-related and many are complex functions of foaming conditions (temperature and pressure). Nucleation is a classical phenomenon and it exists in many processes, e.g., vapor condensation and crystallization. During nucleation, molecules overcome an energy barrier and gather together (via the local density and energy fluctuation) to form embryos of the new phase. When the sizes of the embryos are smaller than a critical size, an increase of embryo size is accompanied by an increase of free energy. On the other hand, if the size exceeds the critical size, further increase of embryo sizes leads to a reduction in free energy. Thus stable nuclei are generated.
Effect of nanoparticles on foam morphology
To obtain foam cells with a controlled structure and uniform distribution, a common practice is to add particles (nucleation agents) to reduce the nucleation free energy. Among them, the inorganic nucleation agents are most commonly used. A fine dispersion of these nucleation agents can facilitate the formation of nucleation centers for a gaseous phase. Although the nucleation mechanism is still under investigation, it is generally known that the size, shape and distribution of the particles, as well as the surface treatment, can affect the nucleation efficiency. Nucleation in PS microcellular foaming using zinc stearate additives was investigated. It was found that above the zinc stearate solubility limit, heterogeneous nucleation dominates and the nucleation rate increases with stearate concentration but is not affected by gas pressure. The presence of fillers was also shown to promote the accumulation of gas on the polymer–particle interface and creation of nucleation sites. Furthermore, foams with finer fillers show a higher cell density at a high saturation pressure. Generally, the particles used in these studies are of micron size. The amount and distribution of the nucleation agents are also important factors to determine the foam quality. The cell density is determined by the concentration of the foaming agent. A non-uniform distribution of the nucleation agents results in a foam that has more cells in the agent rich area and less cells in agent deficient areas, leading to a non-uniform cell size distribution in the foam product. Because the number and size of the bubbles are determined by the concentration of the foaming agent, the uniformity of the cell structure and the cell density are limited by the method used to mix the foaming agents and the polymer. In fact, it is difficult to obtain a uniform cell structure with a high cell density in the conventional foaming process. Compared to conventional micron-sized filler particles used in the foaming processes, nanoparticles offer unique advantages for enhanced nucleation. The improved nucleation efficiency is reported in many studies for different polymer/nanoparticle systems at very low particle concentrations, e.g., PP/clay, PS/clay, Nylon/clay, PLA/clay, PC/clay, PU/clay, PVC/clay and PS/CNFs and PS/CNTs. The extremely fine dimensions and large surface area of nanoparticles provide much more intimate contact between the particles, polymer matrix, and gas. Furthermore, a significantly higher effective particle concentration can be achieved at a low nominal particle concentration. Both could lead to improved nucleation efficiency. The effect of particle concentration on the foam nucleation was investigated. The cell density was found to increase linearly vs. clay concentration at low clay concentration, and starts to level off as clay concentration increases to 10% in some cases, while in other cases, an abrupt increase of cell density was observed as the clay concentration is increased. The effect of clay nanoparticle dispersion on the foam cell morphology was studied in detail. Both intercalated and exfoliated PS/clay nanocomposites were synthesized via in situ polymerization. The exfoliated nanocomposites yielded a much higher nucleation rate than the intercalated nanocomposites. This is attributed to the fact that even at the same nominal particle concentration, the effective particle concentration is higher once the particles are better dispersed. Consequently, more heterogeneous nucleation sites are available. The effect of nanoparticles on cell size was also studied. It was found that in the presence of nanoparticles, the cell size is reduced. As more bubbles start to nucleate concurrently, there is a less amount of gas available for bubble growth, leading to a reduction of cell size. Moreover, the nanoparticles can significantly increase the melt viscosity. Strain induced hardening was observed under elongation as a result of the nanoclay alignment. Both will hinder the cell growth and lead to a reduced cell size. The surface chemistry of clay nanoparticles not only affects the particle dispersion but also has a tremendous effect on the nucleation efficiency in a polymer–clay–foaming agent system. MHAB modified nanoclay can covalently bond either styrene or MMA chains on the clay surface via in situ polymerization as shown in Schematic1. The nature of the polymer tethered to the clay surface has a dramatic impact on the resulting foam morphology because of the interactions of CO2 with the interfacial region between clay and polymer. Here, surface polymerization of CO2-philic PMMA produced exfoliated nanocomposites. The strong interactions between CO2 and the surface-anchored PMMA also substantially reduced the nucleation free energy and enhanced the nucleation rate. Twentyfold increase in cell density was achieved via an addition of 2% clay in PS as shown in Fig. 4. Fig. 5 shows a comparison between several types of nanocomposite foams. The designation 20A is a composite produced by the physical mixing of a commercially available montmorillonite clay (Southern Clay) with polymer resulting in an intercalated structure. All MHABS composites are produced via in situ polymerization and have an exfoliated structure. The parenthesis denotes in Fig. 5 is the monomer used for polymerization with MHABS. The in situ polymerized composite was then physically mixed with
a second polymer, e.g. (PS/MHABS)/PMMA denotes MHABS in situ polymerized with PS after which the excess PS was washed out and the composite compounded again with pure PMMA. The difference in available nucleation sites is not responsible for the dramatic variation in cell density, as MHABS is well dispersed in both PS and PMMA matrices. The main difference between the copolymers tethered on the clay surface in PMMA/MHABS and that in PS/MHABS is that in PMMA/MHABS, the copolymer is essentially PMMA with a cationic ammonium head group bonded to the clay surface, whereas the copolymer in PS/MHABS is a PS polymer containing one methacrylic group. The former has a much higher affinity to CO2 due to the interaction between CO2 and the carbonyl groups in PMMA. More CO2 is likely to be attracted to the surface to form nuclei. Additionally, a strong affinity between CO2 and the carbonyl group of the tethered copolymers in PMMA/MHABS may reduce the gas-particle interfacial tension and consequently the contact angle. This would lead to the reduction in the work of nucleus formation and a large increase in nucleation rate. This hypothesis is verified by two foams shown in Figs.6(g) and (h). The (PS/MHABS)/PMMA has a cell size of 11.1 µm and a cell density of 6.25*10exp8 cells/cm3, whereas the PS/(MHABS/PMMA) has a smaller cell size (8.8 µm), and cell density almost two times higher (1.23 · 109 cells/cm3). The clay remains exfoliated in the former whereas the latter possesses an intercalated structure due to collapse of the exfoliated structure after the extraction of freePMMA. Subsequent blending with PS only yieldedintercalated nanocomposites. Although PS/(MHABS/PMMA) has fewer nucleation sites (being intercalated), it yields a higher cell density. This increase can be attributed to the reduction of nucleation free energy as a result of the presence ofPMMAat the polymer–clay–CO2 interface. In PMMA/MHABS, the exfoliated structure ensures the highest number of available nucleation sites, and the interfacial PMMA at clay surface leads to the reduction of the nucleation free energy. A combination of these two factors results in the extremely high nucleation efficiency as observed in Fig. 5(f). Fig. 6 shows CO2 desorption curves of three samples, PS, PS/5% 20A, and PS/5% MHABS. After releasing the pressure in the high-pressure vessel, the samples remained unfoamed at low temperatures, and thus the dimensions of the samples remained unchanged. By extrapolating the desorption data back to time zero, the solubility of CO2 in PS, PS/5% 20A, and PS/5% MHABS at 50°C and 10 MPa was found to be 8.5, 8.5, and 8.8 wt%, respectively. Carbon dioxide exhibits a slightly higher solubility in the exfoliated PS nanocomposite. The diffusivity at room temperature was calculated based on the three desorption curves by assuming a one-dimensional diffusion since the sample thickness is much smaller than the other two dimensions. The results are 5 · 10exp7, 3 · 10exp7, and 3 · 10exp7 cm2/sec for PS, PS/5% 20A, and PS/5% MHABS respectively. Obviously, the addition of nanoclay can slow down the CO2 diffusion out of the sample. In addition to nanoclay, CNFs and CNTs have also been utilized to prepare PS nanocomposite foams. In both cases, the incorporation of a small amount of nanoparticles leads to a microcellular cell structure, as shown in Fig. 7. Compared to the pure PS foam, the addition of 1 wt% of CNFs yields an increase of cell density by more than two orders, while the cell size decreases from 20 to 2.64 lm. CNTs also exhibit a good nucleation effect. Although with poor particle dispersion, the resultant PS foam with 0.1 wt% CNTs still displays a high cell density and a small cell size (Fig. 7(c)). The average cell density is 1.44*10exp 9 cells/cm3 and the average cell size is 7.11 lm. A similar effect of nanoclay on reducing the cell size and increasing the cell density has been observed in the thermoset PU nanocomposite foam system at 5% clay content. The efficiency of nanoclay on size reduction in this system, however, was not as strong as that in thermoplastics foams. Clay orientation and dispersion is somewhat affected by the foaming process.
Bubble nucleation in nanocomposites
The classical nucleation theory is the approach currently used to describe bubble nucleation in polymer foams, although its prediction of the nucleation rate can easily be off by several orders of magnitude. Within the polymer foaming community, the discrepancy between the classical theory and experiments is sometimes attributed to the intervening heterogeneous nucleation or has led to modifications of the classical theory by incorporating certain aspects specific to the polymer foaming process. The essential content of the classical theory is as follows. The steady state nucleation rate, N0, is given by:
where delta G crit, kB, and T denote, respectively, the free energy of critical nucleus formation, the Boltzmann factor, and the absolute temperature. C0 is the number of gas molecules dissolved per unit volume of the primary phase, and f0 is a kinetic preexponential factor that is
believed to be weakly dependent upon temperature. Since DGcrit appears in the exponent, Delta G crit has a strong impact on the foam quality. The classical theory draws from the following formally exact expression for Delta G crit:
where deltaP denotes the difference between two pressures, one pertaining to the nucleating phase if it existed in bulk at the same temperature and chemical potential as the metastable phase, and the other to the metastable phase (created by imposing a thermodynamic instability on a stable polymer/gas solution). If the polymer is fully saturated with CO2 and the partial molar volume of CO2 in the polymer is zero, deltaP can be taken as the difference between pressure inside and outside the nucleating bubble or as the pressure drop required to induce nucleation (Pinitial–Pfinal). However, our recent study showed that these assumptions may induce an overestimation of the energy requirement. Another difficulty in applying Eq. (2) lies in the fact that rdepends on the size of the critical bubble (nucleus) and usually is inaccessible by experiments. Thus, the classical nucleation theory introduces an approximation of replacing rby the surface tension of the macroscopic bubble/polymer interface, which can be measured. Nanoparticles undoubtedly serve as heterogeneous nucleation agents and their effect on cell density has been qualitatively described by the classical nucleation theory. In the case of heterogeneous nucleation, the nucleation rate is expressed as
where f1 is the frequency factor of gas molecules joining the nucleus and C1 is the concentration of heterogeneous nucleation sites. The work of forming a critical nucleus in a heterogeneous system, DGhet crit is considered proportional to the work in a homogeneous system (Eq. (2)) by a factor dependent on the contact angle h between the gas and polymer and particle surface
In addition to the contact angle, the surface curvature of particles also plays an important role in the critical nucleation energy. The dependency of DGhet crit on both the surface curvature and the contact angle can be described by
Here, f(m,w) is the energy reduction factor, which is a function of h and the relative curvature (w) of the nucleant surface (radius R) to the critical radius (rcrit) of the nucleated phase
In heterogeneous nucleation, the highest nucleation efficiency can only be achieved when the nucleation on the nucleant surface is energetically favored (relative to its homogeneous counterpart) and the nucleant is dispersed in the polymer matrix. In most cases, the observed cell density is much lower than the potential nucleant density, implying that either the nucleants are not energetically effective, or their effects have been compromised due to poor dispersion. The nucleation efficiencies of CNFs, CNTs and exfoliated nanoclay were compared in a recent study. The potential nucleant density in a heterogeneous nucleation system can be estimated by Eq. (12) :
where w is the weight fraction of the particle in the composite, qP is the density of the particle, qblend is the density of the polymer blend, and VP is the volume of the individual particle. In the case of CNFs, the potential nucleant density of the PS composite containing 1 wt% CNFs is 1.41*10exp12/cm3 according to Eq. (12). Experimentally, the cell density of the foam with the same fiber content is 2.78*10exp10 cells/cm3 (shown in Fig. 7(b)). The proximity of these two values indicates that most of the fibers served well as nucleants in the PS foaming. The nucleation efficiency, defined by the ratio of the measured cell density to the potential nucleant density, is 1.97% for CNFs. Similar calculations were conducted for PS/MHABS and PS/CNTs foams and the results are listed in Table 2. For both clay and CNTs systems, the potential nucleant densities are much higher than the final cell densities, leading to nucleation efficiencies that are orders of magnitude lower than that of CNFs. Fig. 8 illustrates how the reduction of critical energy is affected by the nucleants, in terms of surface property (contact angle) and particle geometry (nucleant curvature). Qualitatively, a small contact angle and a large surface curvature offer a higher reduction of critical energy, and consequently an increased nucleation rate. Under the foaming conditions (T = 120 °C, P = 13.8 MPa), sigma was calculated to be ~16.43 mJ/m2 from the known PS-CO2 surface tension value from the literature, r crit is 2.38 nm from Eq. (10). Thus the relative radius w is around 21 for individual CNF. With a typical contact angle of 20°, Eq.(7) yields a reduction factor f of 0.006, indicating that the energy required for the bubble nucleation [detlta G (het, crit)] on the surfaces of CNFs is only 0.003 (f/2) of that in the homogeneous case. In addition, since a complete dispersion of CNFs in the PS matrix was achieved (as shown in Fig. 3), the actual nucleant density is close to the calculated one. The combination of the low energy barrier and the high nucleant density results in a high nucleation rate and ultimately a high cell density. In the PS/CNTs system, if the CNTs are completely dispersed, the relative radius w is only 0.2 considering that the radius of an individual tube is 0.5 nm. In that case, f is 1.8 and the nucleation energy on any single tube surface would approach the homogeneous limit, completely diminishing the benefit of heterogeneous nucleation. However, experimentally, most CNTs were observed as spherical agglomerates with an average radius of approximately several dozen nanometers. These agglomerates with much larger radii can serve as lower nucleation energy sites, but the actual nucleant density is much lower than the theoretical value owing to poor dispersion, leading to the compromised nucleation efficiency. In the PS/MHABS system, the relatively low nucleation efficiency can be explained first by incomplete particle
dispersion. Although exfoliated, stacks of multiple layers are still observable in the polymer domain. A rough estimation from the TEM image of PS/5% MHANBS indicates an average stack thickness on the orders of tens of nanometers. Therefore, the actual nucleant density in the PS/5% MHABS system would be reduced by one order from the value shown in Table 2, i.e., from 5.45 · 10exp13 to 5.45 · 10exp12/cm3. This value, however, is still much higher than the measured cell density (4.02 · 108/cm3), suggesting that there must be other reasons accounting for the low nucleation efficiency. On the clay surface, the nucleation energy should approach the flat plate limit (R extend to the infinite) due to the layered structure of the nanoclay. The modified clay surface is more compatible with the PS matrix, and thus the interfacial tension of the PS melt and the clay is expected to be lower than that of PS and CNFs. Consequently, the contact angle h would be higher. This would lead to a significant increase of f, or much less reduction of nucleation free energy. The equilibrium interfacial tension data in the literature show the lower limit of h to be 105.5°. This leads to a minimum reduction factor f of 1.4 and a reduction of nucleation free energy by 30%. Although the PS/5% MHABS system has a much higher number of potential nucleants than both the PS/1%CNFs and the PS/0.1% CNTs systems, its nucleation efficiency is greatly compromised by the relative ineffectiveness of the energy reduction.
Effect of nanoparticles on foam properties
The high aspect ratio and large surface area of nanoparticles offer the potential for high reinforcing efficiency, good barrier properties, and improved dimensional and thermal stability. The nanometer dimension is especially beneficial for reinforcing foam materials, considering the thickness of cell walls is in the micron and submicron regime. It is therefore ideal to use nanoparticles to reinforce microstructures in order to achieve macroscale property improvement of the final products. Polymer nanocomposite foams exhibit substantially improved properties compared to their neat polymer foam counterparts. Blown by three parts by weight of CO2, the PVC/3% Cloisite 30B nanocomposite foam shows 17.9% increase of tensile strength, 25.9% increase of bending strength and 250% increase of the elongation ratio compared to the pristine PVC foam. The tensile modulus of PS/clay nanocomposite foams has been measured and compared to pure PS foam and PS/talc microcomposite foams. As shown in Fig. 9, the nanocomposite foams exhibit a higher reduced modulus (i.e., tensile modulus divided by the density of the sample), although it is still lower than that of non-foamed pure PS. Among the three foam samples, PS/talc, PS/5wt% 20A, and PS/5 wt% MHABS, the densities are close while the exfoliated nanocomposite foam has the highest reduced modulus. Compared to the non-foamed PS sample, the exfoliated nanocomposite foam sample has about 31% weight reduction with a sacrifice in the reduced modulus of about 19% from 2.6 to 2.1 GPa/g/cm3. In comparison, the PS/talc foam has about 29% weight reduction and a decrease of 43% in reduced modulus. The tensile properties of extruded PS/CNFs nanocomposite foams have been investigated as well. As shown in Fig. 10, all the foams exhibit a similar foam density of 0.6–0.7 g/cm3, indicating a similar weight reduction compared to the bulk PS. For neat PS foam, a weight reduction of 37% sacrifices the tensile modulus by 40% (1.26–0.74 GPa). In the presence of 1 wt% CNFs, the tensile modulus increases by 28% (0.74–0.94 GPa). Once the fiber content is increased to 5 wt%, the tensile modulus further increases to 1.07 GPa, which is comparable to that of the bulk PS (1.26 GPa). In order to normalize the impact of the foam density on the mechanical properties, reduced modulus was used to compare these samples. Due to the relatively lower densities, the reduced modulus of PS/CNFs foams is much higher than that of the bulk PS.
Using the batch foaming process, neat PS and PS/CNFs foams were generated with suitable dimensions for the tests of compressive properties. As shown in Fig. 11, the density of PS/CNFs foams falls in the range of 0.4–0.5 g/cm3. In the presence of CNFs (both 1% and 5%), PS foams show even a higher compressive modulus than that of the PS solid. For example, PS foam containing 5% CNFs exhibits a 12.4% increase in the modulus and a 136% increase in the reduced modulus over the PS solid. This result indicates that the integration of CNFs into the PS foams have a great potential to bridge the gap between the lightweight and high strength requirements. On one hand, CNFs can effectively induce the nucleation of a large amount of bubbles, which ultimately provides a comparable weight reduction to the neat PS foams. On the other hand, the strength deterioration resulted from the inclusion of cells can be compensated by the reinforcement of cell walls/junctions by CNFs. The fire retardance properties of nanocomposite foams have been demonstrated by Han et al.. After burning, the PS nanocomposite foams formed char and maintained structural integrity, while the pure foams quickly melted and dripped, causing fire spreading. Char also forms for PVC nanocomposite foams. The effects of nanoparticles on mechanical properties of thermoset PU/clay nanocomposite foam were also investigated. Two types of nanoclay bearing different functional groups were used to prepare PU nanocomposite foam. Montomorrilonite–OH (MMT-OH) has hydroxyl groups on clay surface and MMT-Tin has catalytic organotin for polyurethane reaction. In the case of semi-flexible/semi-rigid foams (polyol with equivalent weight of 180, polyol-180), a several fold increase in both compressive strength and modulus were observed, again demonstrating high reinforcement efficiency of nanoparticles. However, for the rigid foam system (polyol with equivalent weight of 100, polyol-100), the presence of clay nanoparticles can interfere with the hydrogen bond formation and network structure in the matrix, leading to reduced mechanical properties (i.e. compressive strength and modulus divided by the density of the foam sample) as shown in Fig. 12. It is well known that H-bond formation among urethane groups greatly contributes to the strength and modulus of PUs. For the organoclays used in this study, the PU molecules can be grafted onto the clay surface through the reaction between the –NCO groups and the –OH groups on the clay. The tethered clay may interfere with the H-bond formation in PU as shown in Schematic 2, causing a negative effect on the properties of PU nanocomposite foams. Furthermore, the involvement of organoclays in the reaction could also affect the network structure formation of PU. For PU foams prepared by poly-100, the urethane content is 0.87 mol/100 g, while the urethane content is only 0.65 mol/100 g for the foams prepared by polyol-180. The overall performance of PU nanocomposite foams depends on the competition between the positive effects of clay on polymer reinforcement and foam morphology, and the negative effects on H-bond formation and network structure. The positive effects are stronger for the less rigid polyol-180 foams, while the negative effects dominate for the more rigid polyol-100 foams. The same effect was observed for the rigid PU/nano-silica nanocomposite foam, in which the compressive strength decreases with the higher usage of nano-silica. Nanoclay has also been demonstrated to improve the thermal insulation and aging properties of PU foam (indicated by k-factors in Btu.in/ft2.h.F). The foam with 10% nanoclay gave lower initial and aged k-factors, indicating better thermal insulation.
Conclusions and future trendsDue to the high nucleation efficiency, nanoparticles provide a powerful way to increase cell density and reduce cell size. This is particularly beneficial for the production of microcellular foams. Microcellular foams have been considered as a lightweight and high strength material for structural applications. However, the narrow operation window and less than desirable cell morphology has limited the applications of this technology. Adding nanoparticles is possible to resolve this difficulty and may greatly enhance the industrial applications of microcellular foam.
Mass production of polymer nanocomposites and nanocomposite foams depends on reliable and affordable synthetic and processing methods. The literature is full of novel nanocomposite materials. But in order to move these materials into commercial products, several challenges must be overcome. There must be robust techniques to prepare exfoliated nanocomposites with required mechanical properties in large quantity and low cost. For foam products, various desirable cell morphologies (e.g., small vs. large cells, open vs. closed cells) must be attainable through the successful control of nucleation and growth of bubbles.
It has been found that the surfactant, introduced onto the clay surface to achieve good compatibility between the inorganic clay and the organic polymer or monomer for good clay exfoliation, is a fire hazard material. Natural clay without surface modification, however, can only disperse well in water-soluble polymers. Using water as a nanoclay carrier may achieve surfactant-free nanocomposites with good clay dispersion in hydrophobic polymers. We are currently developing a new method for nanocomposite synthesis by inverse emulsion/suspension polymerization. This technique has been used extensively to prepare water expandable polystyrene (WEPS).
Open cell polymer foams are one of the most commonly used scaffolds for tissue engineering. Nanoparticles may be helpful in generating open cell foams under external fields such as ultrasonic fields, because nanoparticles may behave like a stress concentrator. This may led to a new mass production technology for tissue engineering scaffolds. The high surface area and rich surface chemistry make nanoparticles potentially useful as carriers for desired biofunctionalities (e.g., adhesion sites and signaling molecules).
L. James Lee *, Changchun Zeng, Xia Cao, Xiangming Han, Jiong Shen, Guojun Xu
Composites Science and Technology 2005
The Ohio State University
C is the heat capacity and r is the mass of eachcomponent per unit mass of polymerizing mixture (mixture of un-reacted polyol and isocyanate). The subscript p means the polymerizing mixture, CO2 carbon dioxide, W water, BG physical blowing agent in the gas phase, and BL physical blowing agent in the liquid phase. X is the fractional conversion and (KDH) denotes the heat of reaction, whose subscripts OH and W mean polymerization reaction of the diol and blowing reaction of water molecules. [OH] is the number of moles of the hydroxyl end groups per unit volume of polymerizing mixture and [W] is the number of moles of water molecules. The subscript 0 means initial value. The mass of each component per unit mass of the polymerizing mixture, ri, can be calculated from the concentration and the fractional conversion as follows.
MW and MCO2 are the molecular weight of water and carbon dioxide, respectively. rCO2,D is the initial mass of carbon dioxide that is dissolved in the polymerizing mixture, which is listed in Table 1. rBL can be obtained from the evaporation model for physical blowing agent, which will be discussed later. l is the heat of vaporization for blowing agent per unit mass, which is listed in Table 2 for CFC-11 (CCl3F) and HCFC-141b (CH3CCl2F). The energy balance equation states that the sum of the energy needed for temperature increase of the whole system and the heat consumption by the evaporation of the physical blowing agent is equal to the sum of the heat generated by the exothermic polyurethane reaction and the blowing reaction of the chemical blowing agent.
AOH is the pre-exponential factor, EOH activation energy of the gelling reaction, Rg gas constant. The double underlined part of Eq. (8) denotes the effect of dilution of the reactive group due to the presence of the physical blowing agent and water in the reacting mixture [9]. The reaction parameters are listed in Table 3.
It has been generally accepted that the rate of water– isocyanate reaction is independent of the isocyanate concentration and follows the first order kinetics. However, Baser and Khakhar [9] reported that the second order kinetics predicted the water–isocyanate reaction more accurately than the first order kinetics. For this study, it is assumed that the reaction follows the second order kinetics as follows.
the mole fraction is a linear function of boiling temperature [8].
The relationship between mass ratio and mole fraction of the blowing agent in a liquid phase can be given as
where MB is the molecular weight of the blowing agent and Mno is the initial number average molecular weight of the polymerizing mixture listed in Table 1.
The material parameters of the CFC-11 and HCFC-141b for this model are listed in Table 2.
where p is the ambient pressure for the free rising foam. If the foam is pressurized during mold filling, the pressure should be changed to an appropriate value.
where Vg is the volume of gas phase per unit polymer mass and obtained as
In this study, nb is set to be 109 kgK1 according to the calculation by Niyogi et al. [18].
Here r is density, t is time, v is the velocity vector, g is the acceleration due to gravity, p is pressure, m is the shear viscosity, C is specific heat, and k is thermal conductivity.
Hg is heat generation during the polyurethane foam formation, which is the term on the right hand side of the Eq. (1). In addition to these equations, the mass conservation Eqs. (8), (9), and (15) are solved simultaneously. The time derivative v/vt in the mass conservation equations must be changed to the material derivative D/Dt. In order to track the interface between foam and air, the fractional volume function f(x,t) is defined such that
The overall numerical procedure for predicting mold filling of the polyurethane foam is as follows:
4.2. Free rising in a thin block
distance between the pressure contours decreases near the flow front. The temperature contours on the planes of xZ0.02, yZ 0.02, and zZ0.02 are shown in Fig. 10(a) when the filling time is 35 s. The highest temperature is 510 K and observed in the initially filled region. The polyurethane foam is blown by water and the temperature increases significantly as mentioned previously. Density contours on the planes of xZ 0.02, yZ0.02, and zZ0.02 at 35 s are more complicated than the temperature and pressure contours as shown in Fig. 10(b). Foam density is dependent on temperature, pressure, and conversion of the blowing agent. In general, the foam density is inversely proportional to conversion and temperature, while it is proportional to pressure. The lowest density is about 50 kg/m3 and observed in the initially filled region. On the other hand the highest density is about 300 kg/m3 and occurs at the flow front. The highest density is about 6 times larger than the lowest value. According to Marciano et al., thermal conductivity of the CFC-11 blown polyurethane foam depends on the foam density as follows if the density is higher than 50 kg/m3.
With the relationship, the thermal conductivity varies from 0.016 to 0.045 W/m K, where the foam density is between 50 and 300 kg/m3. The thermal conductivity at the flow front is about 3 times larger than that at the initially filled region. Since, uniform temperature distribution is important for energy efficient control of the refrigerator, thermal resistance of the polyurethane foam in the thickness direction must be uniform throughout the entire cavity. Prediction of the cavity filling and foam density will be utilized for design of the refrigerator geometry and determination of the processing conditions.
