Filler = Something cheap to take up space or not?
Of the ingredients used to modify the properties of rubber and plastic products, the filler often plays a significant role. The term"filler" is misleading, implying a material intended primarily to occupy space and act as a cheap diluent of the more costly elastomer. Most of the fillers used today offer some functional benefit that contributes to the processability or utility of the rubber or plastic product.
The characteristics which determine the properties that a filler will impart to a composite are particle shape, particle size, surface area and particle-matrix compatibility. Particle-matrix compatibility relates to the ability of the polymer to coat and adhere to the filler.
The shape of most mineral filler particles can be approximated as a sphere, cube, block, plate, needle or fiber. Some fillers contain a mixture of shapes. Mineral particles resembling plates, needles and fibers are further characterized by their aspect ratio.
For fibers and acicular particles, the aspect ratio is the ratio of mean length to mean diameter. For platy particles, it is the ratio of the mean diameter of a circle of the same area as the face of the plate to the mean thickness of the plate.
In composites, applied stress is transferred from the polymer matrix to the strong and stiff mineral. It seems reasonable then that this stress transfer will be better effected if the mineral particles are smaller, because greater surface is thereby exposed for a given mineral concentration. Moreover, if these particles have a high aspect ratio (are needle-like, fibrous or platy in shape), they will better intercept the stress propagation through the matrix.
For elastomers, if the size of the filler particle greatly exceeds the polymer interchain distance, it introduces an area of localized stress. This can contribute to elastomer chain rupture on flexing or stretching. Fillers with particle size greater than 10,000 nm (10 um) are therefore generally avoided because they can reduce performance rather than extend or reinforce. Fillers with particle sizes between 1,000 and 10,000 nm (1 to 10 um) are used primarily as diluents and usually have no significant affect, positive or negative, on rubber properties. Semi-reinforcing fillers range from 100 to 1000 nm (0.1 to 1um). The truly reinforcing fillers, which range from about 10 nm to 100 nm (0.01 to 01 um) significantly improve rubber properties.
That said, there is probably no single property of fillers that is as misleading and misunderstood as particle size. A filler's technical description may contain a value, usually in micrometers, for median particle size or median diameter. This number is actually of limited practical value. The only three
dimensional geometric shape that can be unequivocally defined by a single number, such as diameter, is a sphere. The diameters of nodular and blocky particles are poorly related to that of a sphere. Platy and acicular minerals, particularly high aspect ratio types, cannot be defined in any literal sense solely by diameter, yet that is the way their particle sizes are routinely characterized.
The equipment used for particle size analysis, whether based on principles of x-ray sedimentation, light scattering, light extinction or electrical resistance/volume displacement, match the behavior of a particle, regardless of shape, to that of an ideal sphere of specific diameter. Particle size is
therefore conventionally reported as diameter, but the values reported are actually in terms of equivalent sepherical diameter (esd).
Particle size analysis is also complicated by different methodologies that base their calculations of esd on different physical properties. For a 3 micrometer median esd talc, for example, the equivalence may be to a 3 micrometer sphere having the same sedimentation rate, or the same volume, or the same mass, or the same surface area, and each will provide a somewhat different value.
Particle size analyzers designed on different principles of analysis should not be expected to necessarily provide comparable results even on the same sample.
For any given method of particle size analysis, particle size distribution data is somewhat useful for comparing the relative fineness of samples with similar particle shapes. However, particle size distribution, rather than the commonly reported median esd, must be compared. The median indicates only that half the particles have a larger esd and half a smaller one. It says nothing about the size of the largest or smallest particles, which can have a significant effect on filler performance.
Neither the median esd nor a particle size distribution can provide truly comparable results for two minerals with distinctly different shapes, such as calcium carbonate and acicular wollastonite.
Fillers are often sold according to"nominal" mesh size. This can mean anything from 95% to 99+% through the indicated mesh. Mesh size indicates nothing about the size distribution of the oversize particles or the undersize particles.
As with the data from automated particle size analyzers, mesh size of minerals with very different shapes will not necessarily be directly comparable. High aspect ratio particles, in particular, can be a challenge because a particle considerably longer than the mesh opening may pass through easily if its short dimension is smaller than the opening At best, mesh size provides a rough relative comparison.
Regardless of filler size and shape, intimate contact between the matrix and mineral particles is essential, since air gaps represent points of zero strength. Thus, compound strength is improved by good"wetting" of the mineral by the matrix and further enhanced when the matrix is adhered to the mineral surface via chemical bonding.
Surface coating is therefore often used to optimize filler-matrix compatibility and adhesion. Although a mineral to which any type of organic chemical has been added is commonly called surface-treated, a surface treatment may be differentiated from a surface modification on the basis of functionality.
A surface-treated filler is coated with a processing aid. A processing aid may or may not bond to the filler and does not bond to the matrix. It acts as a "wetting" agent to make the filler surface hydrophobic and more intimately coated by the polymer matrix.
Surface treatments improve filler particle deagglomeration and dispersion and allow higher filler loadings.
A surface-modified filler has a coupling agent durably attached to its surface by covalent bonds. The coupling agent is in turn bound to the matrix through chemical reaction or chain entanglement.
The coupling agents used as surface modifiers perform the same functions as surface treatments, and, in addition, improve matrix adhesion, thereby improving physical properties as well as the retention of physical properties with environmental exposure.
The most commonly used treatments and modifiers are the organosilanes. The general chemical structure of organosilanes is RSiX3, where X is a hydrolysable group, such as methoxy or ethoxy, and R is a nonhydrolysable organofunctional group. The organo group may be reactive toward the
polymer matrix or it may be unreactive and serve as a hydrophobe or wetting agent.
Modification with organosilane depends on the ability to form a bond with silanol groups (-Si-OH) on the silicate mineral surface. Since the silicates that are successfully silane-modified generally lack structural silanols, these are a result of the reaction of silica surfaces and adsorbed atmospheric moisture.
The hydrolysis of an alkoxysilane forms silanetriol and alcohol. The silanetriol slowly condenses to form oligomers and siloxane polymers. The - Si-OH groups of the hydrolyzed silane initially hydrogen bond with -OH groups on the mineral surface. As the reaction proceeds, water is lost and a covalent bond is formed. The reaction of hydrolyzed silane with mineral surface -OH can ultimately result in the condensation of siloxane polymer, encapsulating the mineral particle, if sufficient silane is used. Once the mineral is reacted with the silane it exposes an organophilic or organofunctional surface for interaction with the binder matrix. Silane treatment levels are typically 0.5 to 1.0% on mineral weight.
Because the silane-mineral reaction depends on the availability of silanol groups on the mineral surface, some fillers are more amenable to silane treatment than others. Glass, silica and wollastonite readily form surface silanols on exposure to atmospheric moisture and tend to be the most receptive to reaction with silanes.
These photomicrographs of glass beads in polyamide demonstrate the effectiveness of silane treatment in providing intimate filler-matrix contact. In addition to facilitating stress transfer, this minimizes or eliminates voids at the filler-matrix interface where moisture or gases can penetrate.
The oil absorption of a filler reflects the composite effect of all of the foregoing factors ¨C particle shape, particle size distribution, surface area and mineral-matrix interaction. Oil absorption values for mineral fillers are usually based on the spatula rub-out oil absorption test (ASTM D281) which
measures the amount of linseed oil that is just sufficient to coat the mineral particles and fill the inter-particle interstices.
There are two components of oil absorption: The first is the amount of oil required to wet and coat the mineral particles. This depends on the particles¡¯ surface area, which is affected by their particle size distribution; their hydrophilicity, which can be adjusted by surface coating; and their porosity, which is a result of their lattice structure. Surface coating, for example, can reduce oil absorption. After the particles are coated with oil, the second component of absorption is the additional oil that fills the interstices.
For a given mineral, oil absorption is increased by reducing particle size (increasing surface area), increasing aspect ratio (larger interstitial voids) and narrowing the particle size distribution (more interstitial volume). In practice, filled polymers generally benefit from a broader rather than narrower psd. A mix of particle shapes can be beneficial as well.
The mineral fillers most commonly used in rubber and plastics are calcium carbonate, kaolin clay, silica, talc, wollastonite and mica.
Calcium carbonate fillers can be differentiated as ground natural carbonate vs precipitated carbonate. Ground natural carbonates (GCC) are further characterized as dry-ground products, usually in grades from 200 mesh to 325 mesh, and wet-ground products. Of the wet-ground products, fine ground (FG) calcium carbonates range from about 3 to 12 micrometers in median esd, with a 44 micrometer top size. Ultrafine ground (UFG) grades range from about 0.7 to 2 micrometers in median esd, with a 10 micrometer top size.
Ground calcium carbonate is most often available with rhombohedral or prismatic particles. Needle-like calcium carbonate is collected as aragonitic sands in the Carribean, but this is not generally used as filler other than in cement.
Precipitated calcium carbonate (PCC) is produced for applications requiring any combination of higher brightnesss, smaller particle size, greater surface area, lower abrasivity, and higher purity than is generally available from ground natural products. Fine PCC has typically a 0.7 micrometer median, while ultrafine PCC has typically a 0.07 micrometer median. The shape of PCC crystals can be manipulated according to end use requirements. Both GCC and PCC are available with stearate surface treatments for better compatibility with polymer matrices.
Kaolin clay is available in a number of grades: Airfloat clay is dry-ground kaolin that has been air separated to remove impurities and control the particle size distribution. Airfloat clay is the variety most commonly used in the rubber industry in which it is further differentiated as hard clay and soft clay. Hard clay is semi-reinforcing or reinforcing with a median esd of about 0.25 micrometers. Soft clay, which gives softer rubber compounds, is more of a diluent than reinforcement with a median esd of about 1.25 micrometers.
Water-washed clay, usually soft clay, has been slurried in water and centrifuged or hydrocycloned to remove impurities and produce specific particle size fractions. Water-washed clays are often treated to improve brightness. This includes chemical bleaching and/or high-intensity magnetic separation to remove iron and titanium impurities.
Delaminated clay is made by attrition milling the coarse clay fraction from water washing. This breaks down the kaolin stacks into thin, wide individual plates for improved brightness and opacity.
Calcined clay is usually water-washed soft clay that is roasted to either partially or totally remove surface hydroxyl groups. Calcining increases brightness, opacity, oil absorption, and hardness (i.e., abrasivity). Kaolin clay is also available in surface coated grades (e.g., with stearates orsilanes).
Talc products are processed using various combinations of dry grinding, air separation and flotation depending upon the quality of the crude ore and the properties required for intended applications. The talc most often used as filler is commonly called platy talc. It is distinctly lamellar characteristically soft talc. Purity is typically >90% and filler grades are 325 mesh and finer.
The filler uses for wollastonite are dictated by the length of individual needles in the ore and the extent to which this shape is preserved during milling of the finished products.
Powder grades are milled to a low aspect ratio (3:1 - 5:1), either from naturally low-aspect ratio ores or from high-aspect ratio ores that have been ground in a way that breaks the needles widthwise. Despite their low average aspect ratioses, powder grades can retain a significant portion of acicular particles.
Acicular grades are produced from ore containing a suitably high percentage of long needles. The ore is milled in such a way that very fine, needle-like particles are preserved and recovered. Acicular grades typically have aspect ratioses of 15:1 to 20:1.Both powder and acicular forms of wollastonite are available with surface coatings, usually silane.
Most filler grade mica is first collected as flakes by flotation from ore that contains several minerals. Dry-ground products are air milled from flotation concentrate that has been partially of completely dried. Wet-ground products are ground in water using mills designed to delaminate the mica into flakes having a higher aspect ratio, sheen, and slip compared to dry-ground mica. Micronized mica is dry-ground to <20 or <10 micrometers in fluid energy mills using superheated steam.
Ground silica, also known as ground quartz or silica flour, is produced by grinding high-purity quartz to finer than 200 mesh. Air separation is used as required to remove mineral impurities. Ground silica for filler uses offers high brightness, low moisture, chemical inertness, relatively low surface area, and the low liquid absorption that allows high loading levels.
Novaculite is microcrystalline quartz that is milled to low-moisture, highpurity, platy particles. Brightness is generally lower than for other forms of ground silica, but novaculite offers higher aspect ratio, lower binder demand, lower abrasivity, and availability in a range of particle size distributions (to as small as 2 micrometers average).
Precipitated silica is produced by the controlled neutralization of sodium silicate solution by either concentrated sulfuric, hydrochloric or carbonic acids. Reaction conditions are manipulated according to the particle size required. Dilute solutions and controlled acid addition rate are used to minimize the formation of silica gel. The reinforcing properties of precipitated silica can usually be related to particle size; 10-30 nm particles are reinforcing, while 30+ nm particles are semi-reinforcing. Because of the difficulty in measuring the size of particles this small, surface area , rather than particle size, is usually used for classifying various grades. For example, silica in the range of 125-250 m2/g is generally reinforcing, while products in the range of 35-100 m2/g are semi-reinforcing. In rubber compounds, which are the primary application for precipitated silica,the silica particles exist mostly as small clustered agglomerates rather than discrete particles.
Precipitated silica is usually sold with about 6% adsorbed free water and a surface essentially saturated with silanol groups, the latter making it very receptive to in situ reaction with organosilane, most often mercaptosilane. Pretreated silica is also available.
Other mineral fillers used in rubber and plastics are barite, diatomite, feldspar, nepheline syenite, pyrophyllite and tremolitic talc.
Filler uses for barite generally require high brightness, high purity, and fine particle size, usually 325 mesh or finer. The highest quality filler grades are made by flotation, followed by wet grinding, bleaching with sulfuric acid, washing, drying, and milling.
Blanc fixe is precipitated barium sulfate for uses where higher brightness and purity and finer particle sizes are required than are generally available with barite. To make blanc fixe, crushed barite is first roasted with coke in a rotary kiln at about 1200oC to form barium sulfide. This is quenched in water and countercurrent leached to produce a barium sulfide solution. Blanc fixe isproduced by treating this solution with sodium sulfate to precipitate ultrafine barium sulfate.
Diatomite products are generally differentiated by process and particle size. Flux-calcined diatomite is calcined at about 1200oC with sodium carbonate or sodium chloride. The flux converts iron oxides to a colorless glassy phase, resulting in a white product. Flux-calcined diatomite is milled, screened, and air classified, with coarse fractions sold for filtration uses and fine fractions as white fillers. These products provide maximum void volume, which can exceed 90%, and consequent high absorptivity.
Calcined diatomite, also known as straight-calcined, is heated to between 870o and 1100oC. This burns off organic matter, converts some of the opaline silica to cristobalite, shrinks, hardens, and reduces the fine structure of individual particles. Calcination generally turns the white to off-whitenatural diatomite pink from iron oxidation, so it is used primarily for filtration applications and seldom as a filler.
Natural diatomite products are gently crushed and milled to retain the frustule shape and then screened and air classified to remove impurities and to segregate products into coarse fractions for filtration applications and fine fractions for filler uses.
Floated 325 mesh feldspar finds limited use as a filler in polymers despite its low vehicle demand, and resistance to abrasion and chemical degradation. Although the performance properties of feldspar and nepheline syenite are similar, the latter is more often used because it is available in a broader range of fine grinds (325 mesh to 1250 mesh) and it has generally higher brightness .
US pyrophyllite products are natural blends of quartz and pyrophyllite along with minor amounts of mica and kaolin. Quartz-induced abrasion generally restricts the use of pyrophyllite in polymers, although the fine platy nature of the pyrophyllite, mica and kaolin components can otherwise contribute to physical properties in a manner similar to kaolin or talc.
Tremolitic talc, aka New York talc, is a fine-grained"hard"talc that is a natural blend of mostly tremolite and talc. It lacks the distinctly lamellar, soft, hydrophobic characteristics normally associated with talc and is excluded from certain traditional talc applications. Its atypical properties (e.g., hydrophilicity, lower vehicle demand, durability) are used to advantage, nevertheless, in ceramics, paint, rubber and plastics.