Performance evaluation of high structure carbon black in different polymer blends.

In the rubber industry, carbon black is commonly used as filler to give reinforcing effects, such as altering the modulus or tensile strength of the rubber (refs. 1 and 2). However, it is not clear what structural differences exist between the rubber before and after the reinforcement. Nor is it clear how structural disorder is introduced when different types of carbon black are compounded into the rubber. In the simplest terms, carbon black is essentially elemental carbon in the form of extremely fine particles having a partially amorphous molecular structure, among microcrystalline arrays of condensed aromatic rings (ref. 3). The orientation of the arrays within the amorphous mass appears to be random, consequently a large percentage of arrays has open edges of their layer planes at the surface of the particle. Associated with these open edges are large numbers of unsaturated carbon bonds providing sites for chemical activity. In use, particles remain largely intact and many of the useful properties of carbon black stem from their morphological characteristics, while others result from surface activity. Carbon black is composed of clusters of fused, primary particles called aggregates. If the primary aggregate is composed of many prime particles with considerable branching and chaining, it is referred to as a high-structure black, where the structure is expressed in terms of DBP adsorption value. It was reported that the tear strength increases with increasing both carbon black nitrogen surface area and DBP absorption values (ref. 4). The increases are more pronounced and correlated with increasing DBP absorption values more than they are with increasing nitrogen surface areas. The properties of both the primary aggregates and the spherical particles comprising them are important controlling factors in carbon-black performance (ref. 5). Higher surface areas, as imparted by finer prime particle blacks, require more energy for wetting and dispersion. It was observed from FTIR spectra that the calculated ratio of 1,450 [cm.sup.-1] band to 1,433 [cm.sup.-1] in the spectra slightly increases with increasing carbon black DBP absorption values (ref. 6). From the infrared data, it was noted that a high-structure carbon black could be absorbed on the elastomer surface and promotes mostly trans-conformation for the rubber structure. It is likely that carbon black primary aggregates or high-structure aggregates reinforce the rubber structure by entanglement and mechanical interlocking forces (ref. 7). Carbon blacks contain chemical complexes on their surfaces, either resulting from oxidation during manufacture or from a deliberate after-treatment of the surface. These complexes are chemisorbed oxygen groups in the form of carboxy, quinone, lactone and phenolic groups (ref. 8). On the other hand, the reinforcing carbon black will not participate in any crosslinks within the network of polymer and carbon black. This conclusion is especially consistent with the recently reported nuclear magnetic resonance (NMR) relaxation results for natural rubber filled with carbon black (ref. 9). The relaxation results in the study indicated that in the process of curing natural rubber, active carbon black led to a higher reduction of crosslink densities than inactive carbon black.

In this article, we have taken three high structure carbon blacks and tried to evaluate their performance with different polymer blends to identify the impact of different carbon black and/or polymer properties on the compound behavior.

Experimental

The formulation details have been explained in table 1. It can be seen from the table that we have taken three carbon black grades named ADIT 93 (N339), ADIT 74 (N347) and ADIT 57 (N375). The four polymers selected are NR (natural rubber ISNR 3 CV), solution SBR, PBR-1220 and SBR-1712. The selections of the polymers have been done on the basis of the common tire formulations being practiced nationally and internationally. The addition of different ingredients was also done using the general practice of tire manufacturers.

Natural rubber is a very high molecular weight polymer: therefore, to mix NR with other ingredients, the NR was masticated beforehand. Mastication shortens rubber molecular chains, resulting in a reduced molecular weight and homogenization in NR. Only then ingredients were distributed uniformly into the rubber matrix. We have used a two roll open mill for this purpose. Other polymers were not masticated and were used directly.

The details of the mixing and processing schedule followed in this study are explained in table 1. We have used a 1.5 L internal mixer for the initial mixing, and final compounding and sheeting was done on an open 6" diameter and 13" width, two-roll mill.

Curing of the compounds was done after eight hours cooling at room temperature in a 180 Mt. curing press using the hard chrome plated molds of 152 x 152 x 1.90 mm size as per ASTM D 412.

The three carbon blacks were characterized physicochemically following the ASTM standard methods. Table 2 shows the major properties of all three carbon blacks, along with the test methods adopted for the analysis.

The following are the equipment used for physico-chemical and compound property characterization: DBP (Brabender DBP machine model E, with DADS software from Hitec); [N.sub.2]SA (Quantachrome); aggregate size (Bi-DCP, Brookhaven Instruments); tint (Erichsen tint tester); toluene discoloration (Shimadzu UV spectrophotometer); pellet hardness (IPHT, Hitec); Mooney viscometer (MV 2000, Alpha Technologies); rheology (MDR 2000, Alpha Technologies); hardness (IRHD, Wallace); tensile (Zwick UTM); dispersion (Dispergrader 1000 GT, Optigrade AB); specific gravity (Wallace); abrasion (Zwick Abrader 6102); rebound (Zwick 5109); tangent delta and heat buildup at different temperatures (Goodrich Flexometer, Model II); crack growth and initiation (DeMattia Flexometer); and infrared study (Perkin-Elmer, Spectrum GX).

Using the statistical analytical tools like cluster analysis, the data have been analyzed to create a system of organizing observations into groups where members of the groups share properties in common.

Results and discussion

The compound properties are shown in table 3. The two very important physico-chemical properties of carbon black that control the ultimate compound properties are size and structure.

Size

The size of the particles in carbon black can be understood from many measurements like iodine adsorption (ASTM D 1510), nitrogen surface area (ASTM D 6556), CTAB surface area measurement (ASTM D 3765), aggregate size analysis (laser aggregate size analyzer) and tint strength (ASTM D 3265).

Iodine adsorption number

Iodine adsorption (expressed in mg/kg of carbon) measures the amount of iodine, which can be absorbed from a potassium iodide solution, on the surface of a given mass of carbon black. The iodine adsorption number is a primary indication of surface area for defining different carbon blacks. In the present experimentation, all three grades are in a similar iodine value of 87 to 89 mg/gm.

Nitrogen surface area ([N.sub.2]SA)

Nitrogen surface area is a measurement of the amount of nitrogen that can be adsorbed on a given mass of carbon black. High surface area is associated with a high level of reinforcement and high surface energy, but at the expense of more difficult dispersion, processing and increased hysteresis (ref. 10). It was found that N375 has a higher [N.sub.2]SA in comparison to the other two. In theoretical terms, the ionic radius of iodine is 0.216 nm and nitrogen is 0.171 nm in their normal oxidation states (ref. 11). Being smaller in ionic size, nitrogen has access to smaller pores and voids on the surface of carbon black. The difference in surface area measured values between nitrogen and iodine gives the roughness of the carbon black; this is termed, in other words, as surface activity. It was observed that this property has a direct bearing on the abrasion of the compound through average aggregate size and tint. For the sample N375, the difference between [N.sub.2]SA and iodine is a positive number. This indicates its highly reactive surface (ref. 12). Similar observation was seen from toluene discoloration results at 338 nm, and details will be discussed later.

Aggregate size

Clusters of fused particles of carbon black make up aggregates. The distribution of aggregates and the average aggregate size play major roles in determining the utility of a given carbon black grade in a rubber formulation. Smaller aggregates contribute to higher rubber reinforcement, but also to more difficult dispensability and lower resilience. Since aggregate size or surface area is the primary determinant of reinforcement, perhaps the logical starting point in choosing a carbon black is to assess the level of reinforcement which will be required and, based on these considerations, pick a surface area range with which to start. Compound requirements such as tensile strength, abrasion resistance, tread wear and tear resistance must be considered, beating in mind that these properties are improved by choosing a higher surface area (ref. 10). In the present study, we have found that the average aggregate size ([d.sub.50]) of the three samples remains between 72 to 93 nm. This property was found to have no correlation with the iodine number. The carbon black N375 was found to have a very narrow distribution pattern of aggregates compared to the others (figure 1). Higher surface areas, as imparted by finer primary particle black N375, require more energy for wetting and dispersion. They also tend to tie up more rubber per unit weight of carbon, resulting in stiffer compounds than with coarser blacks like N347 and N339.

[FIGURE 1 OMITTED]

Tinting strength

In this test, a small amount of carbon black is mixed into a paste with zinc oxide and oil (Paraplex G62). The reflectance of this paste is compared with a similar paste based on an internationally accepted standard tint reference black (ITRB #3).

Modern carbon black theory relates tint strength inversely to aggregate diameter (ref. 13). In an unpublished study, comparison of measured tint strength to mean aggregate diameter values across a broad range of carbon black parameters suggested that most (about 90%) of the variation in tint strength was accounted for in the variation in mean aggregate diameter (ref. 12). Still, tint strength is very highly indicative of ultimate reinforcement potential. In this study, we have found that tint strength is closely related to surface area, and decreases with increasing aggregate size. Tint of the carbon black increases, and the aggregate size distribution becomes narrower with increasing velocity of combustion gas (ref. 4). Among the three carbon black samples, we have measured highest tint values for the N375 grade.

CTAB surface area

The CTAB Surface Area test was designed to overcome both the problems mentioned for the iodine number and nitrogen surface area tests; in that the molecule to be adsorbed, cetyltrimethylammonium bromide, is far larger than the nitrogen molecule and is also a very effective wetting agent. In the present study, we have seen a good agreement of CTAB values with the aggregate size distribution, as observed by S.S. You and S.K. Choi (ref. 4)

Surface activity

No direct means of surface activity measurement exists. However, two relatively simple indicators are available for oil furnace grades, particularly tread grades like the present samples.

When the surface of the carbon particles is oxidized in the manufacturing process, surface pitting occurs, which disturbs the relationship among the iodine value, particle size and nitrogen surface area. The interior surface of the pits or micropores can significantly increase the nitrogen surface area, leaving the particle size and iodine value relatively unchanged. One way of understanding is to subtract the iodine number from the nitrogen surface area. This will be a positive value for pitted carbon blacks, which are called very highly active carbon blacks. As [N.sub.2]SA minus iodine becomes negative, rubber reinforcement can be expected to progressively deteriorate. In the present study, we have found that N347 has a negative value for the difference between [N.sub.2]SA and iodine. This might be due to the closely packed dense structure of the carbon black. It is also seen that both iodine and CTAB are similar for this grade, indicating its closely packed structure.

The second indicator for surface activity of carbon black is the toluene discoloration test. From the time of the very earliest carbon black production, it was noted that the depth of color of the solvent could judge completeness of reaction after performing an extraction on a sample of carbon black. The commonly chosen colorimeter filter produced a wavelength of 425 nm.

Studies of extractables with scanning spectrophotometers indicated that absorption peaks also occur at shorter wavelengths. A significant absorption occurs at 338 nm. When manufacturing a carbon black, if the reaction time is short for a given surface area, a low reading (approaching zero) can be obtained at this wavelength, while still meeting 425 nm specifications. Carbon blacks that exhibit low percent transmission values at 338 nm usually are found to be highly active (ref. 12), having been subjected to minimum heat treatment and oxidation. Similar results have been observed for the present samples. N375 was found to have significantly lower TD values at 338 nm, indicating more surface activity.

Surface activity is more likely a function of the number of open edged layer planes exposed at the surface along with the associated unsaturated carbon bonds than it is a function of the chemical groups that might exist there. This theory fits empirical observations with the process. The carbon blacks that provide the highest reinforcement to surface area ratio are those that have been produced using the shortest reaction time, i.e., N375. The evidence is that the immediate precursor of finished carbon black includes a very rough-surfaced particle composed of fused microcrystalline arrays. These arrays should first appear joined in a totally random fashion, with a large percent of them oriented in such a way that the ends of planes are exposed, along with a large number of unsatisfied carbon bonds. This is the more active surface that was obtained for N375 compared to the other two.

Structure

The carbon black structure relates to the bulkiness of carbon black and is determined by the microscopy (refs. 14 and 15), oil absorption and small angle scattering XRD (ref. 16). Carbon black has two types of structures, primary and secondary.

Dibutyl phthalate absorption (DBPA)

The test, using an absorptometer, measures the quantity of DBP oil required to change the consistency of the black from a free-flowing powder to a coherent paste at a specified torque level. The results are expressed as milliliters of DBP per 100 grams of carbon black. Carbon black structure, as indicated by DBP absorption, inhibits the elasticity of a compound. Essentially, spherical aggregates loaded in a compound will inhibit the elasticity of a compound to a degree, but when these aggregates have a certain shape factor, that is have a long dimension, they act as if they were very short fibers and interfere with the elastic mobility of the polymer in which they are dispersed. This stiffening effect is, therefore, more pronounced with structure than with aggregate size. Increased structure manifests itself in many related ways. Viscosity, modulus, hardness and dispersibility are all increased (ref. 17).

CDBP

DBP values decrease upon application of mechanical force to carbon black samples. Apparently, such work progressively breaks the bridging occurring between particles, beginning with the weakest. It has been postulated that some of the bridges are so weak, in fact, as to be vulnerable to breakage or disruption in the process of being dispersed into a rubber matrix. These weak elements of structure lead to what has been called transient structure, that is, structure measured with the DBP test, but non-existent in the final rubber compound. In order to more closely approximate the structure remaining after mixing, a test was devised wherein the carbon black sample is compressed four times at 24,000 psi prior to the normal DBP test procedure. These repeated compressions break agglomerates and some weak associations of some aggregates.

It has been observed that a higher structure carbon black gave higher CDBP, but the difference between DBP and CDBP was also higher, indicating the growth of secondary structure.

Compound properties

The compound properties in different formulations have been shown in table 3.

Crosslink density

The properties of the vulcanized rubber depend on the nature and distribution of the crosslink and the crosslink density. The volumetric swell ratio between the original solid and the gel is calculated correcting for temperature and density.

For measuring crosslink density, the cured sample is carefully weighed (W1), then immersed in reagent grade toluene (EMerck) at 25 [+ or -] 2[degrees]C for 72 hours with intermittent stirring. At the end of this period, the sample is carefully weighed (Wg) and the final weight (Wd) was taken alter it was dried in an electric oven at a temperature of 60[degrees]C until a constant weight was obtained. The formula used is: Crosslink density = [(Wg - Wd) / Wd] K+ 1, where K is the ratio of densities of the solvent (toluene = 0.865 gm/cc at 25[degrees]C) to polymer (NR = 0.92 and SBR = 0.94 gm/cc). Details are shown in figure 2. The swell ratio is proportional to the sample's ability to change size, which is restricted by crosslinks and polymer chain entanglement. High values show low levels of crosslinking, and a low swell ratio shows increased crosslinking. We have observed that the stiffness and resilience increase with crosslink density. It was found that N339 grade gave the minimum swell ratio, indicating its higher stiffness, which is correlated with its modulus (ref. 18). For most of the compounds, tensile strength does not have any linear relation with crosslink density. All the carbon blacks in formulations 1 and 2 with NR behave almost identically. Formulation 3 was found to have the minimum swell ratio, and formulation 4 the highest. This was possibly due to the underloading of CB in formulation 4.

[FIGURE 2 OMITTED]

Mooney viscosity

Viscosity is the stiffness of an unvulcanized stock. This is an especially important processing indicator for stocks that are to be calendared, extruded or molded. High structure blacks produce high Mooney viscosity. As carbon black loading increases, the Mooney viscosity differences between high and low structure carbon blacks increase. When adding carbon black to a rubber compound, the rate of Mooney viscosity change varies for each elastomer. In the present study, we found that the N375 grade showed higher viscosity in three of the four formulations. For maximum torque, as observed from the theological study, N339 always had the highest MH in all the formulations.

Elongation

Elongation represents the increase in length of the sample at rupture, calculated as a percent of the original length. Its value is taken as a percentage of its original length. Usually, lower structure carbon blacks give higher elongation, and increased loading reduces the elongation. It can be seen (figure 3) that N347, while having the highest DBP value, gave good elongation irrespective of the formulation used.

[FIGURE 3 OMITTED]

Normally, carbon black particles try to form branched aggregates (hydrodynamic properties due to occluded rubber) or agglomerated aggregates (filler networking trapped rubber), and effective filler volume will be more when the structure is high. During the mixing of rubber compounds, secondary structure partially vanishes (effective filler volume is reduced). So, the polymer attached with secondary structure is now free (due to the breakdown of secondary structure) and contributes to more elongation. This happens in the case of N347, which has more secondary structure, as indicated by the difference between DBP and CDBP values.

Tensile strength

Because all rubber products undergo stress during use, most compounds are designed to meet tensile strength requirements. Tensile strength is the force needed to rapture a dumbbell specimen. The rate of tensile strength response to carbon black addition varies among carbon blacks and polymers. Tensile strength increases with decreasing aggregate size. The carbon black particles act as stress homogenizers. They allow slippage and a redistribution of stress among the polymer chains. This creates strong bonds between the carbon black and the rubber. Mechanical rapture of polymer chains also helps to bond the carbon black and robber together. This increases the tensile strength. Tensile properties of the three carbon black grades in all four formulations do not show a specific trend for any carbon black grade from the results shown in figure 4. It seems that both the base polymer, as well as the physico-chemical properties of the carbon black, in combination, are responsible for the tensile properties.

[FIGURE 4 OMITTED]

In the passenger tread formulations, the base polymer used was BR+SBR and the loading of CB varied from 65-70 phr. In this combination, N339 shows better tensile strength compared to the other two grades. From this we can understand that for a passenger tread formulation with BR+SBR as base polymer, a higher loading of medium structure carbon black like N339 gives better tensile strength values.

The high structure carbon black N347 did not give better tensile properties, perhaps because of more secondary structure as envisaged in the formulations used. This indicates that the proportion of high primary structure in combination with low secondary structure contributes to the tensile properties. But the same is not true in the case of a truck tread formulation.

As far as truck tread formulation is concerned, the base polymer used was NR and the loading of CB varied from 47-55 phr. In this combination, N375 shows better tensile strength properties compared to other grades. From this we can understand that for a truck tread formulation with NR as base polymer, less phr of low structure carbon black gave better tensile properties.

Tear strength

The tear strength results for the different carbon black compounds are shown in figure 5. Tear strength was found to be high in NR and low in formulations without NR. This indicates that NR contributes to tear strength. Among the three carbon blacks, N375 shows better tear strength properties, followed by N347 and N339. Apart from carbon black, the polymer type plays a major role in the tear strength. The trend observed in the present study was:

[FIGURE 5 OMITTED]

NR > NR+BR > SSBR+BR = OSBR+BR.

Modulus

Modulus is an expression of the force per cross sectional unit area required to stretch a test piece to a given elongation. It may also be considered as resistance to elongation or as stiffness in the vulcanizate. The modulus can be influenced by the rubber grade (viscosity) and by the vulcanization system.

It can be seen from figure 6 that, irrespective of the formulation used, N339 gives higher modulus values compared to other grades. Modulus mainly depends on reinforcement, which in turn is related to roughness of carbon black. Both structure and loading affect modulus. As the loading increases, the modulus increases in polymers. Higher structure carbon blacks usually provide higher modulus, but that was not observed in this study. It seems that alter a certain value, the impact of structure is not visualized for improvement in modulus. It is interesting to note that the MH values bad a direct correlation with the modulus of the compound.

[FIGURE 6 OMITTED]

Hardness

Hardness in rubber is defined as the ability to resist indentation. It is expressed as a number referring to the scale on the instrument used for the measurement. Hardness is determined by the amount and the type of filler, the degree of dispersion and the crosslink density. In the present study, we found that N339 gave the highest IRHD hardness, irrespec- tive of the formulations used.

Abrasion

As shown in figure 7, N375 gave minimum abrasion loss in all the formulations, while the abrasion loss for the other two grades were comparable. The reason is as follows: N375 has the highest tint and narrow aggregate size. The higher the tint strength, the higher will be the uniformity of aggregate size distribution, which leads to more reinforcement of the rubber compound, which in turn gives better abrasion resistance. The other property found in N375 is that the difference between DBP and CDBP is minimal. The lowest secondary structure with a narrow aggregate size distribution makes N375 more resistant to abrasion, irrespective of the formulation and carbon black loading.

[FIGURE 7 OMITTED]

Impact resilience

Impact resilience (rebound) represents the percent of returned energy when a pendulum is dropped against a sample. Resilience decreases as filler level is increased (formulations 7-9, with 70 phr carbon black loading, figure 8.). High structure shows a secondary effect of improving rebound. The rebound value decreases as carbon black loading increases, or with decreasing carbon black aggregate size.

[FIGURE 8 OMITTED]

The rolling resistance behavior of carbon black compounds is very complex, and almost all the properties have a bearing on it (ref. 19). Polymers with a low glass transition temperature have a high resilience, and polymers with a high Tg have a low resilience; which is also reflected in the compound resilience irrespective of carbon black properties. No significant effect of carbon black properties could be seen in different formulations for the rebound values. But high resilience is observed with NR as the base polymer.

Dynamic mechanical properties

Tan [sigma]

Tangent delta (tan [sigma]) is the ratio of energy lost and the energy stored during cyclic deformation, which indicate the amount of hysteresis present during deformation. The energy is related to the storage modulus and the loss modulus that corresponds to the elastic and viscous property of the compound respectively. In the present study, we have observed the following (figure 9):

[FIGURE 9 OMITTED]

* Formulations play a major role for tan [sigma] values in the compounds under investigation.

* Formulation 1 with natural rubber was found to have lower tan [sigma] values at all testing temperatures from 0-100 [degrees]C.

* The impact of carbon black was visible in formulations 2 and 4, where N375 was found to give the lowest values.

* In formulation 3, N375 gave the highest tangent delta at 0 [degrees]C and lowest at 70 [degrees]C. This is peculiar behavior found for the N375 grade.

Cut initiation and growth

The following observations, as shown in figure 10, have been made from the DeMattia study:

[FIGURE 10 OMITTED]

* Carbon black grade N339 showed the best performance with respect to crack initiation in almost all formulations.

* Formulation 1, with natural rubber, was found to have poor performance for crack initiation compared to others; which relates to the polymer rather than the properties of carbon black (ref. 20).

The crack growth rates were found to have a different aspect of combination of formulation and carbon black, as seen in figure 11.

[FIGURE 11 OMITTED]

* N339 was found to have the lowest growth in natural robber in formulation I.

* Formulation 4 (OSBR+BR) shows higher cut growth.

* In case of the N375 grade, the cut growth was found to be lowest in formulation 4, perhaps due to high tint and high surface area with low average aggregate size that contribute to more effective volume under identical loading.

Heat buildup

Following are the observations (figure 12) for the heat buildup study:

[FIGURE 12 OMITTED]

* At 60 [degrees]C, formulation 1 (NR) gave the lowest heat buildup. A similar trend was seen for tan [sigma] also.

* At 100 [degrees]C, formulation 3 gave the lowest heat buildup.

* Formulation 2 gave the highest buildup at both temperatures.

* At 60 [degrees]C, N339 grade was found to have the lowest heat buildup in all formulations, except in formulation 1, where N375 had the lowest value.

* At 100 [degrees]C also, N339 gave the lowest value, except in formulation 4, where N375 had the lowest value.

Statistical approach

Cluster analysis

Cluster analysis is a multivariate statistical technique which assesses the similarities between units or assemblages, based on the occurrence or non-occurrence of specific artifact types or other components within them. In order to classify a mountain of information into manageable piles, cluster analysis is of great utility to organize observed data into meaningful structures, that is, to develop taxonomies (ref. 22). The purpose of this algorithm is to join together objects (e.g., test parameters) into successively larger clusters, using some measure of similarity or distance. A typical result of this type of clustering is the hierarchical tree. Cluster analysis classifies a set of observations into two or more mutually exclusive unknown groups based on combinations of interval variables.

Hierarchical clustering methods do not require preset knowledge of the number of groups. Two general methods of hierarchical clustering methods are available, including divisive and agglomerative. The divisive techniques start by assuming a single group, partitioning that group into subgroups, partitioning these subgroups further into subgroups and so on until each object forms its own subgroup (ref. 23). In either case, the results of the application of the clustering technique are best described using a dendogram or binary tree. The objects are represented as nodes in the dendogram and the branches illustrate when the cluster method joins subgroups containing that object. The length of the branch indicates the distance between the subgroups when they are joined. A dendogram that clearly differentiates groups of objects will have small differences in the far branches of the tree and large differences in the near branches

It has been tried to understand the inter-correlation of different properties of carbon black and its compound properties through the statistical approach using the cluster analysis. For this purpose, we have used Minitab Software (ref. 24). The cluster figure is shown in figure 13. It can be seen from figure 13 that there are three numbers of major clusters as identified 1-3, in which they have sub-clusters like 2a, 2b, etc. The different parameters belonging to each cluster have unique combinations and indicate their specific importance as will be discussed.

[FIGURE 13 OMITTED]

Cluster 1

Cluster 1 consists of the basic properties of carbon black like iodine, DBP, toluene discoloration and CTAB, along with the difference of DBP-CDBP. This cluster shows the maximum control over all the clusters. It has been clearly noticed that the difference between DBP and CDBP, i.e., the secondary structure of carbon black, plays a very important role, along with the CTAB, for controlling the compound properties in the high structure carbon blacks studied presently.

Cluster 2

This has three sub groups, 2a, 2b and 2c, as shown in figure 13. The cluster subgroup 2a consists of t5, t35, t90, cut growth and crosslink density, where crosslink density plays a major role, being controlled by the cure time. Cut growth was controlled by the crosslink density. In cluster 2b, heat buildup at 60 [degrees]C, tan [sigma] 8 at 100 [degrees]C and heat buildup at 100 [degrees]C are interlinked. The crosslink density and heat buildup at 100 [degrees]C were found to have a broad link with each other. The subgroup 2c has two further sub-groups--consisting of tensile strength and tear strength in one, and elongation break and rebound resilience in the other. All four properties are linked with abrasion loss. So, the group 2 cluster has three major contributors, including crosslink density, heat buildup and abrasion loss.

Cluster 3

This constitutes five sub groups denoted as 3a-3e. The sub group 3a has nitrogen surface area, tint and ash. The sub group 3c consists of MH, IRHD hardness, 300% modulus and tan [sigma] at 23 [degrees]C. Cut initiation has a very good link with tan [sigma] at 70 [degrees]C belonging to sub group 3d. The sub group 3b consists of only Mooney viscosity that has a very good relation with the sub groups 3c and 3d. Tint was found to have a better control among all the sub groups. This has further links with tan [sigma] at 0 [degrees]C. In cluster 3, nitrogen surface area and tint are the two major primary properties of carbon black found to have control over the Mooney viscosity, 300% modulus, cut initiation, hardness and tan [sigma] at 0 [degrees], 23 [degrees] and 70[degrees] the compound.

Conclusion

Reinforcement of rubber

Rubber can be reinforced by carbon black in two main ways:

* Adsorption of polymer molecule segments: This takes place at the disordered and somewhat porous surface of the particles. For this N2SA, aggregate size distribution and tint of carbon black play a key role. A similar observation was also recorded from the cluster analysis, as discussed earlier.

* Chemical bonding between the polymer and the residual hydrogen and oxygen-based chemical groups at the surfaces of the particles: Polymer chains are attracted to these sites and create strong bonds with the carbon black instead of with each other. This is the role played possibly by surface activity of the carbon black, inferred from toluene discoloration at 338 nm.

Effect on rubber

Carbon blacks greatly modify the physical performance of rubber compounds. The effects are dominated by the two properties already discussed, surface area or 'aggregate size' and structure. A narrow aggregate size distribution with a high effective surface area gives high levels of reinforcement. This not only leads to a high tensile strength, but also a high resistance to abrasion.

Structures with large aggregates have a higher compound viscosity. These structures also have a higher modulus, and hence the products are much stiffer. Rheological properties (MH) of the compound have a very good correlation with the modulus. The carbon black N339 was found to satisfy this observation. Similar observations are also made in cluster analysis that high crosslink density contributes to the modulus, and that is true for the N339 grade.

Abrasion resistance is achieved by the improvement in tinting strength and primary structure content of carbon black. This study indicates that, irrespective of the formulation used, N375 gives the least abrasion loss.

The carbon black structure, both primary and secondary, plays an important role for the elongation properties. It has been observed that, irrespective of the formulation used,

N347 gave better elongation properties.

The combination of base polymer, as well as the physicochemical properties of the carbon black particles, affect tensile and tear strength properties such that a high proportion of primary structure in combination with low secondary structure gives better tensile properties. Formulations with NR and the N375 grade show the best combination for these two properties. Even in cluster analysis, both these properties are in one cluster only, which was found to be controlled by crosslink density. In order to supplement our observations for the better surface activity of the carbon black grade N375, we have carried out an FTIR study, which shows more surface groups are available with N375 compared to the other two grades.

Toluene discoloration measured at 338 nm, in comparison to 425 nm values, gives a very good insight for the surface activity of carbon black. This may be possible due to the short residence time in the reaction zone in the reactor for a given surface area. Although this may be affected by the nature of oil used for the manufacturing of the carbon black, in our present study, all three carbon black grades have been produced from the same carbon black feedstock, hence the effect of oil is ignored.

For the tan [sigma] at different test conditions, the impact of polymer was seen. Natural rubber formulations were found to have lower tan [sigma] and the N375 grade was found to give the lowest value. From the statistical analysis, it was found that tint and nitrogen surface area are the two parameters that have a bearing on tan [sigma] values up to 70 [degrees]C.

As referred from the cluster analysis, the cut initiation is correlated with the Mooney viscosity and 300% modulus. The N339 grade was found to have late crack initiation due to its high modulus (ref. 21). NR based formulations were found to have fast initiation, but slowest growth.

Table 1--formulations and processing procedures

Formulation        1       2       3       4       5       6
Ingredients       Phr     Phr     Phr     Phr     Phr     Phr
NR                 100     100     100      75      75     75
Solution SBR        -       -       -        -       -      -
PBR-1220            -       -       -       25      25     25
SBR-1712            -       -       -        -       -      -
N339 (ADIT 93)     47       -       -       55       -      -
N347 (ADIT 74)      -      47       -        -      55      -
N375 (ADIT 57)      -       -      47        -       -     55
Aromatic oil        7       7       7       12      12     12
Zinc oxide          5       5       5        5       5      5
Stearic acid        2       2       2        2       2      2
6-PPD               3       3       3        3       3      3
MC wax              -       -       -        -       -      -
PVI              0.15   0.15     0.15    0.20   0.20     0.20
Sulfur            2.5     2.5     2.5     2.0     2.0     2.0
NOBS              0.6     0.6     0.6     0.8     0.8     0.8
Total           166.8   166.8   166.8   179.5   179.5   179.5

Formulation        7       8       9      10      11      12
Ingredients       Phr     Phr     Phr     Phr     Phr     Phr
NR                  -       -       -       -       -       -
Solution SBR       70      70      70       -       -       -
PBR-1220           30      30      30      50      50      50
SBR-1712            -       -       -      69      69      69
N339 (ADIT 93)     70       -       -      65       -       -
N347 (ADIT 74)      -      70       -       -      65       -
N375 (ADIT 57)      -       -      70       -       -      65
Aromatic oil       30      30      30      20      20      20
Zinc oxide          4       4       4       4       4       4
Stearic acid        1       1       1       2       2       2
6-PPD               2       2       2       2       2       2
MC wax              2       2       2       2       2       2
PVI                 -       -       -       -       -       -
Sulfur            2.0     2.0     2.0     1.5     1.5     1.5
NOBS              1.5     1.5     1.5     2.0     2.0     2.0
Total           212.0   212.0   212.0   216.8   216.8   216.8

Start temp. 50[degree]C
First stage: 1.5 L internal mixer
Time (min.) Load
  0-1.5 Add rubber
  1.5-3.0 Add 1/2 CB + ZnO
         + S. acid
3.0-4.5 Add remaining CB +
        oil
4.5-4.8 Sweep
4.8-5.0 Dump
  Dump at 110 -115[degree]C
  Maturation period: 2 hrs.
Final stage
Time (min.)
  0-2 Warm masterbatch
  2-4 Add curatives
  5-6 Sheet it out and cool
      in room temp.

Table 2--physico-chemical properties

Physicochemical                  UOM     ASTM     ADIT     ADIT
properties                              metho       57       93
Iodine adsorption              mg/kg        d    (N375)   (N339)
CTAB                      [m.sup.2]/g   D 1510    87.1     88.4
[N.sub.2]SA               [m.sup.2]/g   D 3765    80.9     87.1
[N.sub.2]SA--iodine                     D 6556    93.1     88.0
Tint strength                % ITRB3               6.0     -0.4
Average aggregate size            nm    D 3265   115.3    104.4
DBPA                        cc/100 g              72.0     82.0
CDBPA                       cc/100 g    D 2414   113.1    120.3
DBP--CDBP                   cc/100 g    D 3493    94.3     97.8
Ash                                %              18.8     22.5
Toluene discol. -425 nm            %    D 1056    0.44     0.30
Toluene discol. -338 nm            %    D 1618    96.0     98.0
Pellet hardness avg.              gf         -    40.6     65.8

Physicochemical            ADIT
properties                   74
Iodine adsorption         (N347)
CTAB                       89.6
[N.sub.2]SA                88.7
[N.sub.2]SA--iodine        82.0
Tint strength              -6.4
Average aggregate size     98.4
DBPA                       93.0
CDBPA                     127.2
DBP--CDBP                 100.3
Ash                        27.0
Toluene discol. -425 nm    0.20
Toluene discol. -338 nm    99.4
Pellet hardness avg.       85.0

Table 3--compound properties

Formulation No.

Rubber properties                          Units    Method
Compound specific gravity                  gm/cc     ASTM
M[L.sub.1+4] @ 100[degree]C,                  mu    D 1646
  final viscosity
Mooney scorch @ 135[degree]C                        D 1646
  Minimum
  [t.sub.5]                                  m:s
  [t.sub.35]                                 m:s
Rheological properties @
  145[degree]C                                      D 6204
  MH                                      lb.in.
  [t.sub.90]                                 m:s
Hardness
  Durometer-A                                       D2240
  IRHD                                              D1415
Stress-strain properties                            D 412
  300% modulus                               MPa    D 412
  Tensile strength                           MPa    D 412
  Tear strength                              MPa    D 624
  Elongation at break                          %    D 412
Abrasion loss                                 mg    D 5963
Rebound DIN 53512, N = 5                            D 1054
Crosslink density                          gm/cc    D 2765
Heat buildup at 60[degree]C     [DELTA]T[degree]C   D 623
Heat buildup at 100[degree]C    [DELTA]T[degree]C   D 623
Tan [delta], at 0[degree]C,                         D 623
  10 Hz, 10% strain
Tan [delta], at 23[degree]C,                        D 623
  10 Hz, 10% strain
Tan [delta], at 70[degree]C,                        D 623
  10 Hz, 10% strain
Tan [delta], at 100[degree]C,                       D 623
  10 Hz, 10% strain
Cut initiation, K cycles,                           D 813
  at 50[degree]C
Cut growth after 10 K cycles,                 mm    D 813
  50[degree]C

Formulation No.                     1       2          3

Rubber properties               N339    N347       N375
Compound specific gravity        1.114   1.116      1.116
M[L.sub.1+4] @ 100[degree]C,    47.7    47.6       52.1
  final viscosity
Mooney scorch @ 135[degree]C
  Minimum                       35.1    34.8       40.8
  [t.sub.5]                     15:49   17:24     14:04
  [t.sub.35]                       20   22:16     18:55
Rheological properties @
  145[degree]C
  MH                            14.36   13.66      14.02
  [t.sub.90]                    20:34   21:43     21:03
Hardness                        145[degree]C/30 min.
  Durometer-A                      65      63         63
  IRHD                             67      65         66
Stress-strain properties        145[degree]C/23 min.
  300% modulus                   9.7     9.1       10.0
  Tensile strength                21      22         23
  Tear strength                   83      92        107
  Elongation at break            519     543        532
Abrasion loss                    109     109         91
Rebound DIN 53512, N = 5        50.6    50.3       48.7
Crosslink density                3.22    3.26       3.23
Heat buildup at 60[degree]C       30      28         26
Heat buildup at 100[degree]C      29      29         31
Tan [delta], at 0[degree]C,     0.236   0.240      0.232
  10 Hz, 10% strain
Tan [delta], at 23[degree]C,    0.218   0.224      0.175
  10 Hz, 10% strain
Tan [delta], at 70[degree]C,    0.179   0.183      0.155
  10 Hz, 10% strain
Tan [delta], at 100[degree]C,   0.138   0.132      0.144
  10 Hz, 10% strain
Cut initiation, K cycles,         453     310        453
  at 50[degree]C
Cut growth after 10 K cycles,    2.8     3.3        3.3
  50[degree]C

Formulation No.                     4       5          6

Rubber properties               N339    N347       N375
Compound specific gravity        1.133   1.131      1.127
M[L.sub.1+4] @ 100[degree]C,    54.3    55.90      56.8
  final viscosity
Mooney scorch @ 135[degree]C
  Minimum                       42.6    43.7       46.1
  [t.sub.5]                     18:22   20:53     17:11
  [t.sub.35]                    24:57   28:27     25:03
Rheological properties @
  145[degree]C
  MH                            15.28   14.46      14.12
  [t.sub.90]                    23:02   25:41     23:47
Hardness                        145[degree]C/35 min.
  Durometer-A                      60      60         59
  IRHD                             64      62         62
Stress-strain properties        145[degree]C/26 min.
  300% modulus                    9.5     9.0        9.2
  Tensile strength                 19      20         20
  Tear strength                    79      83         81
  Elongation at break             513     524        522
Abrasion loss                      92      93         82
Rebound DIN 53512, N = 5         46.1    47.0       47.7
Crosslink density                 3.18    3.21       3.2
Heat buildup at 60[degree]C        40      47         41
Heat buildup at 100[degree]C       36      39         39
Tan [delta], at 0[degree]C,     0.352   0.289      0.287
  10 Hz, 10% strain
Tan [delta], at 23[degree]C,    0.262   0.253      0.243
  10 Hz, 10% strain
Tan [delta], at 70[degree]C,    0.219   0.199      0.201
  10 Hz, 10% strain
Tan [delta], at 100[degree]C,   0.175   0.170      0.167
  10 Hz, 10% strain
Cut initiation, K cycles,       NI *      520        520
  at 50[degree]C
Cut growth after 10 K cycles,    3.9     3.2        3.3
  50[degree]C

Formulation No.                     7       8          9

Rubber properties               N339    N347       N375
Compound specific gravity        1.153   1.153      1.151
M[L.sub.1+4] @ 100[degree]C,    62.9    57.7       54.7
  final viscosity
Mooney scorch @ 135[degree]C
  Minimum                       37.7    34.3       32.9
  [t.sub.5]                     24:03   25:42     25:23
  [t.sub.35]                    29:29   30:45     31:04
Rheological properties @
  145[degree]C
  MH                            18.82   18.28      17.65
  [t.sub.90]                    25:45   26:30     27:41
Hardness                        145[degree]C/40 min.
  Durometer-A                      69      68         68
  IRHD                             75      74         73
Stress-strain properties        145[degree]C/29 min.
  300% modulus                  13.1    11.6       12.2
  Tensile strength                17      15         14
  Tear strength                   57      62         65
  Elongation at break            373     377        334
Abrasion loss                     85      85         83
Rebound DIN 53512, N = 5          39    39.4       38.2
Crosslink density                2.79    2.82       2.85
Heat buildup at 60[degree]C       32      36         35
Heat buildup at 100[degree]C      27      29         27
Tan [delta], at 0[degree]C,     0.244   0.402      0.437
  10 Hz, 10% strain
Tan [delta], at 23[degree]C,    0.331   0.355      0.289
  10 Hz, 10% strain
Tan [delta], at 70[degree]C,    0.207   0.209      0.206
  10 Hz, 10% strain
Tan [delta], at 100[degree]C,   0.139   0.147      0.151
  10 Hz, 10% strain
Cut initiation, K cycles,       NI *    NI *         465
  at 50[degree]C
Cut growth after 10 K cycles,    3.6     3.7        5.5
  50[degree]C

Formulation No.                    10      11         12

Rubber properties               N339    N347       N375
Compound specific gravity        1.129   1.131      1.129
M[L.sub.1+4] @ 100[degree]C,    53.2    52.4       54.8
  final viscosity
Mooney scorch @ 135[degree]C
  Minimum                       36.3    36.8       37.4
  [t.sub.5]                     28:3    28:57     26:45
  [t.sub.35]                    33:58   34:03     32:11
Rheological properties @
  145[degree]C
  MH                            13.17   12.68      12.52
  [t.sub.90]                    32:39   32:58     29:31
Hardness                        145[degree]C/45 min.
  Durometer-A                      54      54         54
  IRHD                             56      58         57
Stress-strain properties        145[degree][C/34 min.
  300% modulus                   8.1     7.6        7.8
  Tensile strength                16      14         14
  Tear strength                   62      61         57
  Elongation at break            510     511        461
Abrasion loss                     76      65         59
Rebound DIN 53512, N = 5        42.6    43.2       43.1
Crosslink density                4.02    3.99       3.76
Heat buildup at 60[degree]C        36     36         39
Heat buildup at 100[degree]C       34     31         29
Tan [delta], at 0[degree]C,     0.243   0.307      0.283
  10 Hz, 10% strain
Tan [delta], at 23[degree]C,    0.228   0.215      0.173
  10 Hz, 10% strain
Tan [delta], at 70[degree]C,    0.282   0.171      0.164
  10 Hz, 10% strain
Tan [delta], at 100[degree]C,   0.175   0.175      0.166
  10 Hz, 10% strain
Cut initiation, K cycles,       NI *      420        432
  at 50[degree]C
Cut growth after 10 K cycles,   15.4   15.6       10.3
  50[degree]C

* NI = Cracks not initiated after 1,000 K cycles

References

(1.) J.B. Donnet and A. Voet; Carbon Black: Physics, Chemistry and Elastomer Reinforcement, Marcel Dekker, NY (1976).

(2.) ASTM Standard D 3849; "Primary aggregate dimensions for electron microscopy image analysis," Ann. Book ASTM Std, 09.01, P. 619 (2002).

(3.) Y. Schwob, "Reinforcing properties of carbon black," Rev. G., Caout. Plast., 71, 733, 70-74, July 1994.

(4.) S.S. You and S.K. Choi, A new characterization method of tread carbon black by statistical regression treatment, DC Chemical Co. Ltd., Korea.

(5.) Jay Janzen and G. Kraus; "Effects' of aggregate size and structure aggregate size distributions in carbon black reinforcement of elastomers," Proc. Int. Rubber Conf., G7- P1-7 (1972).

(6.) J. Leisen, J. Breidet and J. Kelm, Rubber Chem. & Tech, 72, 1 (1999).

(7.) M.K. Khromov, et al, "Nature of the effect of fillers on the dynamic properties of rubbers," Int. Polym. Sci. Technol., 21, 1, 22-24 (1994).

(8.) J.B. Donnet, et. al, "Study of surface activity of carbon black by invert gas chromatography II: Effect of carbon black thermal treatment on its surface characteristics and reinforcement," Kaut. U. Gummi. Kunst. 46, 11 866 (1993).

(9.) K. Sperling-Ischinsky and W.S. Veeman, "129Xe-NMR of carbon black filled elastomers," J. Braz. Chem. Soc., 10, 4, 299 (1999).

(10.) M. Wang; Rubber Chem. Tech, 71, 521 (1998).

(11.) Web reference: www.lenntech.com/periodic-chart-element.

(12.) D.T. Norman, "Rubber grade carbon blacks," Product Development Witco, Concarb Division, web literature.

(13.) Walther W. Tscharnuter, Lily Zu and Bruce B. Weiner, "ASTM carbon black reference materials: Particle sizing using a Brookhaven Instruments BI-DCP, disc centrifuge photosedimentometer," Brookhaven Instruments.

(14.) A. Medalia; J. Colloid interface Sc., 32, 115 (1970).

(15.) T. Gruber, et al., Rubber Chem. Technol., 67, 280 (1994).

(16.) G. Beaucage, et al., J. Poly. Sc., Part B: Polym Phys., 37, 1,105 (1999).

(17.) D.J. Kohls and G. Beaucage, "Rational design of reinforced rubber," Current Opinion on Solid State & Material Science, 6.

(18.) A.N. Gent, et. al., "Effect of carbon black on crosslinking," Rubber Chem. Technol., 76, 0517 (2003).

(19.) E.H. David and J.C. Moreland. "Fundamentals of rolling resistance," Rubber Chem. Technol., 74, 0525 (2001).

(20.) H.J. Kim and G.R. Hamed, "On the reasons passenger tire sidewalls are based on blends of natural rubber and cispolybutadine, "Rubber Chem. Technol., 73, 0743 (2000).

(21.) G.R. Hamed and N. Rattanasom, "Effect of crosslink density on cut growth in gum natural rubber vulcanizates"; Rubber Chem. Technol., 75, 0323 (2002).

(22.) Andre Hardy, "On the number of clusters, " Computational Statistics & Data Analysis: volume 23, p. 83-96, 1996.

(23.) R. Duda, P. Hart and D. Stork, Pattern Classification: 2nd edition. John Wiley & Sons, Inc: California, USA, 1998.

(24.) Minitab Statistical Software, Release 13, Minitab Inc.