Printer Friendly

Performance differences between carbon blacks and CB blends for critical IR applications.

The blending of carbon blacks has been a standard practice throughout the rubber industry for many years. One of the main rationales for this practice has been the need by rubber product manufacturers to keep inventories of raw materials to a minimum. Because carbon black comes in a wide variety of morphologies and, even in pelletized form, is a highly space-intensive material in terms of storage, it is not practical for users to maintain an inventory of every type. Instead, they select a few grades to meet their immediate needs.

When a new application or change in performance requirements calls for a different morphology of carbon black, the user often tries to compensate by developing a blend of the carbon blacks that are on hand. The logic is that the relation of the carbon black morphology to the resulting performance properties is a simple issue of particle size and structure. Therefore, blending carbon blacks should be able to give all of the benefits of the components while diluting their disadvantages. However, it is often found that compounds containing carbon black blends have relatively poor dispersion, inconsistent processing performance and lower than expected physical properties. These undesirable characteristics stem from, what is by nature, a highly complex interaction between carbon black and the polymer matrix. As with all complex environments, the interaction between carbon blacks and the polymer matrix cannot be completely defined by one or two parameters or a few simple equations. Considerable experimentation has resulted in a vast amount of literature showing that many factors govern the performance of carbon black in this environment.

Some of the factors that are usually overlooked when deciding whether or not to use a carbon black blend are the relationships between dispersion and performance properties of the following:

* Particle-size distribution of the blend compared to a single carbon black;

* pellet properties of the components versus those of a single carbon black; and

* the resulting incorporation of a bimodal aggregate mixture versus the incorporation of a single homogeneous distribution of aggregates.

In this article we examine some of the main processing and performance differences between rubber compounds containing carbon black blends and analogous rubber compounds containing individual carbon blacks.

Comparison of particle size distributions

Figure 1 shows the particle-size histograms for four common oil-furnace carbon blacks: N110, N351, N650 and N762. The charts represent the distribution of particle sizes in an average sample of each carbon black and are configured with a frequency percentage on the Y axis and diameter size increments along the X axis. Each bar represents the percentage of the particle-size distribution in that diameter range. Note that the two reinforcing carbon blacks (N110 and N351) have such a tight distribution that their histograms are shown in 5 nm increments over a relatively narrow range of diameters; the histograms for the semi-reinforcing carbon blacks (N650 and N762) are shown in 20 nm increments over a much wider range of diameters. Also, note that all of the charts are skewed toward the smaller particle diameters (ref. 2). These characteristics are even true when we compare the N110 distribution to the N351 and the N650 distribution to the N762.


Remembering that the purpose of blending should be to enhance the best properties of the components while diluting the least desirable properties, let us examine some hypothetical blends of N110 and N762 and compare them to each individual component (i.e., N110 alone and N762 alone) and to another single black, N351, using the data from the histograms. Figure 2 shows the calculated histogram of two hypothetical blends (percentage by weight) of N110/N762: 50/50 and 35/65 (calculated to approximately match the iodine number of 68 for N351). For the purpose of this example we have converted the N351 distribution chart from increments of 5 nm to increments of 20 nm. In doing so, we see how tight the N351 distribution is compared to the 20 nm increment chart of the N650 or the N762.


Next, let us examine the calculated histograms of the 50/50 and 35/65 blends of N110/N762 and compare them to the histogram of the single N351 carbon black. The normal N351 particle-size distribution is tight and homogeneous, whereas the skew of the hypothetical carbon black blends are exaggerated toward the particle size distribution of the two component grades. This means the small-particle fraction and the large-particle fraction of the carbon black blend are both significantly larger than that of the single black. This example demonstrates that even though the composite surface area measurement for a carbon black blend may approximate that of a single carbon black, the actual particle-size distribution is quite different.

This bimodal distribution will be even more distinct in an actual compound, since the N110/N762 blend will not be a homogeneous mixture as it enters the mixing operation. The N110 component will be much more difficult to incorporate into the rubber matrix due to its significant smaller particle-size distribution. The larger particle size N762 component will incorporate much quicker during the mixing operation. This can result in a compound with domains of undispersed N110. By comparison, the N351 with its homogeneous particle size distribution will incorporate at a much more consistent rate and will generally give a more uniform dispersion throughout the polymer matrix.

Comparison of pellet properties of CB blend components vs. single CB and the resultant effects on dispersion

Carbon black produced for industrial rubber applications is usually pelletized to facilitate bulk handling and transfer to the mixing operation. The mixing process then breaks down these pellets to achieve a uniform distribution of aggregates within the polymer matrix. Two of the main carbon black properties that determine ease of dispersion into a polymer matrix are pellet hardness and aggregate morphology. Studies have shown that the strength of carbon black pellets of comparable size rises with increasing carbon black surface area (iodine number) and with decreasing aggregate structure (DBPA) (ref. 3).

Hence, a blend of carbon blacks with significantly different surface area and/or structure will also have an inherent difference in pellet hardness for each component and an expected inherent difference in dispersibility during the rubber mixing operation just from their pellet properties alone. The smaller particle-size or lower structure component will usually have a harder pellet, and this harder pellet will be more difficult to break down during mixing (ref. 4).

Comparison of morphology of CB blend components vs. single CB and the resultant effects on dispersion

A second major influence on the dispersion of carbon black into a rubber compound is the morphology of the carbon black. Smaller particle size (higher iodine number) carbon black disperses slower than larger particle-size carbon black of the same structure. Also, lower structure (lower DBPA) carbon black of the same particle size disperses much slower than higher structure carbon black of similar particle size (ref. 5). Hence, incorporation of the small particle size carbon black (high iodine number) or low structure carbon black (low DBPA) would be expected to require longer mixing time than a carbon black with large primary particles or with high structure to obtain the same dispersion level.

The result expected from mixing a blend of larger particle-size and smaller particle-size carbon blacks into a rubber compound would be ready dispersion of the larger particle-size carbon black and slower dispersion of the smaller particle-size carbon black. Under marginal mixing conditions, the high-surface-area carbon black could be poorly dispersed, thereby giving the rubber compound potentially lower physical and performance properties than expected (refs. 6-8). In similar fashion, carbon black with high structure would be expected to disperse more completely than one with lower structure. In blends of carbon black with both high and low structure fractions, a similar disparity in dispersion quality would be anticipated.

Effect of poor dispersion on performance properties

Several studies have shown that many of the critical performance properties of rubber compounds are affected by the level of dispersion of the carbon black in those compounds (refs. 6-8). Tensile strength, fatigue life and abrasion resistance are just some of the properties that can be affected by poor dispersion. Also, applications that require smooth, cured surfaces to perform their specified function (e.g., seals, extrusion profiles and hoses) are significantly affected by the dispersion quality of the rubber compounds (ref. 9). Therefore, a carbon black blend that exhibits poor dispersion could also exhibit lower than expected performance properties and poor surface appearance.


Following are two examples of a carbon black blend and its performance versus a single carbon black. This carbon black blend (and blends similar to it) attempts to utilize the reinforcement properties of a smaller particle size component to enhance the physical properties of a larger particle-size carbon black. The combination chosen for both examples in this evaluation is a 50/50 blend by weight of N650 carbon black and N330 carbon black. This blend was chosen because it has a relatively large difference between the larger particle component (N650) and the smaller particle black (N330). Also, this blend (or a variation of this blend) is actually used for many industrial rubber product applications.

The two examples compare a 50/50 blend by weight of N650 and N330 with a single black (experimental carbon black A). Experimental carbon black A was chosen because its particle size is near the average of the two components of the blend. (Experimental carbon black A has an iodine number of 55 versus a calculated mean iodine number for the blend of 58). In each case we will examine four compounds - one for each of the components of the blend, one for the carbon black blend itself and one for the single carbon black, ECB-A (experimental carbon black A). In each example, care was taken to employ the same mixing procedure so that differences in dispersion could be determined. Each example will be viewed from the perspective of attempting to improve the performance properties of the N650 or the dispersion of the N330.

Example 1: Nitrile compound for molded and hose products applications

As noted previously, carbon black blends are often used in an attempt to improve physical properties by combining a large particle size black with a more reinforcing carbon black. For this example we selected a simple nitrile compound designed for a general molded or hose application. We compared the physical properties and dispersion data of a compound with the 50/50 blend of N330 and N650 carbon blacks with those of a single black selected to be near the center of the particle size distribution of the blend (ECB-A). The compound uses a 34% ACN nitrile, 15 phr DOP and a standard sulfur cure system. Compounds were mixed for five minutes in a single-pass, right-side-up mix in a laboratory BR internal mixer with the curatives being added on the mill. A loading of 80 phr carbon black was selected to obtain a hardness level between 75 and 80 Shore A. Each batch was cured for 15 minutes at 175 [degrees] C for the physical testing (see table 1). The physical properties from the testing appear in table 2.
Table 1 - formulations for example 1

Ingredients Batch 1 Batch 2 Batch 3 Batch 4

NBR 100 100 100 100
Zinc oxide 5 5 5 5
Stearic oxide 1 1 1 1
DOP 15 15 15 15
ODPA 1.5 1.5 1.5 1.5
Sulfur 1.5 1.5 1.5 1.5
MBTS 1.5 15 1.5 1.5
N330 80 0 40 0
N650 0 80 40 0
Experimental carbon
 black A 0 0 0 80
Table 2 - physical properties and dispersion results for example 1

Properties Batch 1 Batch 2 Batch 3 Batch 4

N650, phr 80 0 40 0
N330, phr 0 80 40 0
Experimental carbon
 black A, phr 0 0 0 80
Hardness, Shore A 76 79 78 78
Tensile, MPa 16.8 17.3 16.2 18.2
30% modulus, MPa 2.0 2.3 2.1 2.3
100% modulus, MPa 4.5 4.8 4.5 5.3
200% modulus, MPa 10.4 11.4 10.3 11.8
Elongation, % 340 302 310 329
Tear, die C, Kn/M 44.8 48.2 48.9 50.6
Dispersion rated
 10 = best 9.0 6.8 7.5 9.3

Results of example 1


The N330 carbon black showed the worst dispersion of the four compounds. The blend of 50/50 N330 and N650 exhibited only slight improvement over the N330, indicating that the smaller particle size component acted independently of the better dispersing larging particle size N650. The best dispersion results were achieved by the single carbon black, ECB-A. As in the hypothetical case described above, the carbon black blend was not incorporated into the compound as a homogeneous mixture. Rather, it was mixed into the compound as two distinct components, with the larger particle-size, higher structure, softer pellet N650 carbon black being dispersed better than the smaller particle-size, lower structure, harder pellet N330 carbon black. This condition apparently left undispersed N330 carbon black domains that prevented the blend from achieving the intended performance (table 3). If this compound was intended for an application where surface smoothness was important, the poorer dispersion of the N330 carbon black in the blended compound would also have resulted in more surface defects than the better dispersing single component, ECB-A.
Table 3 - formulations for example 2

Ingredients, phr Batch 1 Batch 2 Batch 3 Batch 4

EPDM 100 100 100 100
Zinc oxide 5 5 5 5
Stearic acid 2 2 2 2
Paraffinic oil 90 90 90 90
Solidified ethlyene
 glycol 2 2 2 2
Sulfur 1.25 1.25 1.25 1.25
TBBS 0.80 0.80 0.80 0.80
TMTD 0.60 0.60 0.60 0.60
DPTT 0.80 0.80 0.80 0.80
ZDBC 0.60 0.60 0.60 0.60
N330 150 0 75 0
N650 0 150 75 0
Experimental carbon
 black A 0 0 0 150

Physical properties

As with the dispersion results, the physical properties of the carbon black blend compound did not achieve the expected results. In fact, the tensile value of the blend compound was the lowest of the four compounds. Also note the lower than expected tensile results of the N330 compound. Neither the blend nor N330 compound reached optimum dispersion during the single-pass mix, and the result is lower than expected physical properties. These results are contrasted with the tensile and modulus results of the compound containing the single ECB-A. This compound did achieve the expected physical properties because the single carbon black incorporated as a homogeneous component and was able to reach optimum dispersion with this single-pass mix. These results indicate that significant processing changes - including longer mixing cycles or additional mixing passes - would be needed to compensate for the dispersion deficiencies of the blend and to achieve the desired performance properties.

Example 2: EPDM compound for extrusion product applications

Blends are also used to improve a major performance deficiency in the original compound. The next example involved an EPDM extrusion profile compound that originally used N650, for which a 50/50 blend of N330/N650 was proposed to improve tensile strength. This example illustrates that even though a carbon black blend may give the desired physical properties of the combined morphologies, other problems can occur that prevent the carbon black blend from successfully fulfilling its purpose. The compound is a basic sulfur-cured EPDM formulation with 90 phr of paraffinic oil and a small amount of solidified ethylene glycol. It uses a standard non-bloom cure system and is mixed for 3.5 minutes using a single-pass, upside-down procedure in a standard laboratory BR internal mixer. For this example, 150 phr of carbon black was selected to achieve a hardness between 70 and 75 Shore A (table 3). Again four compounds were mixed, with the N330 and N650 being mixed separately to better evaluate the performance of the carbon black blend. As with the first example, a 50/50 blend of N330/N650 will be compared to single ECB-A with a particle size near the average of the particle size range of the two components. Each batch was cured for 15 minutes at 175 [degrees] C for the physical testing. Results from the testing are shown in table 4.
Table 4 - physical properties and dispersion results for example 2

Properties Batch 1 Batch 2 Batch 3 Batch 4

N650, phr 150 0 75 0
N330, phr 0 150 75 0
Experimental carbon
black A, phr 0 0 0 150
Hardness, Shore A 70 76 73 73
Tensile, MPa 13.4 15.2 14.8 14.4
30% modulus, MPa 1.7 2.0 1.8 1.8
100% modulus, MPa 5.5 4.8 4.8 5.2
200% modulus, MPa 10.9 10.3 10.0 10.2
Elongation, % 380 313 320 316
Tear, die C, Kn/M 30.8 31.9 32.2 36.8
Dispersion rated
 10 = best 8.2 6.9 7.4 8.4

Results of example 2


As with the first example, the dispersion of the N330 and the blend of N330 and N650 are both significantly poorer than either the N650 or ECB-A. Likewise, ECB-A showed the best dispersion of the four compounds tested. Previous work has shown a correlation between dispersion and surface defects for an extruded surface or profile (ref. 9). If the N330 or the N330/N650 blend had been used in a profile extrusion, their poor dispersion performance would have caused a large number of visual defects. This, in turn, would have resulted in higher scrap rates and/or additional processing. Again, the single-pass mix was not sufficient to achieve the optimum dispersion for the carbon black blend or the N330 but was quite sufficient for the N650 and the single black, ECB-A.

Physical properties

The physical properties for example 2 do not appear to be as adversely affected by the poorer dispersion of the carbon black blend. Comparing the four tensile results, the values correspond to the expected progression - with the N650 being lowest, followed by ECB-A, the 50/50 blend of N650/N330 carbon black and the N330. Similarly, other physical properties do not show any unusual variation that could be attributed to poor dispersion or other abnormalities that could be caused by the carbon black blend.


This article has demonstrated that carbon black blends do not act as a single, homogenous ingredient in rubber compounds; rather, they act as a group of separate ingredients. Each component of the blend incorporates into the rubber compound based on its own characteristics, and the resulting dispersion is a function of the morphology and pellet properties of each individual carbon. A carbon black blend of two distinctly different particle sizes results in a bimodal distribution of particles sizes and not a homogeneous distribution such as a single carbon black would exhibit.

These and other considerations suggest that development of carbon black blends to successfully achieve desired performance requirements can be a very complicated process. All factors of each component must be considered and evaluated to determine their effect on the processing and performance of the compound. The pellet properties and morphological differences must be examined with great care since these will significantly affect the dispersion and performance of the final compound.

An alternative approach to blending carbon blacks is to use a single carbon black that has a morphology designed to meet the desired performance criteria. This single-black approach can, in many cases, deliver the desired effect without encountering the dispersion problems and end-product performance loss associated with the carbon black blend. Depending on the individual application and the tightness of its specifications, the benefits of using a single black may offset other considerations related to the inventory and storage of raw materials.


(1.) B.G. Vaidya, D.J. Bharucha and A. Roy, Rubber News, 28 (August, 1971).

(2.) J.B. Donnet, R.C. Bansal and M.J. Wang, Carbon Black, Science and Technology, 2nd edition, p. 111, (1993).

(3.) J.B. Donnet, R.C. Bansal and M.J. Wang, Carbon Black, Science and Technology, 2nd edition, pp. 159-160, (1993).

(4.) D.T. Norman, The Vanderbilt Rubber Handbook, 13th Ed., pp. 413-414, (1990).

(5.) G.R. Cotton, Plastics and Rubber Processing and Applications, Vol. 7, No. 3, pp. 173-178, (1987).

(6.) E.M. Dannenberg, European Rubber Journal, 167 (1), 25 (1985).

(7.) A.Y. Coran and J.B. Donnet, Rubber Chemical Technology, 65, 973 (1992).

(8.) W.M. Hess, V.Z. Chirico and P.C. Vegvari, Elastomerics, 112, No. 1 (1980).

(9.) B. Chung, J. Medashi, B. Mackey and D.J. Curtis, ACS Paper No. 55, (May 1995).


"Filler-filler and filler-polymer interactions as a function of in-rubber carbon black dispersion" is based on a paper presented at the May, 1998 Rubber Division meeting. "Peroxide curing with precipitated silica" is based on a paper presented at the May, 1998 Rubber Division meeting. "Performance differences between carbon blacks and CB blends for critical IR applications" is based on a paper presented at the May, 1998 Rubber Division meeting.
COPYRIGHT 1999 Lippincott & Peto, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Reed, Thomas
Publication:Rubber World
Geographic Code:1USA
Date:Apr 1, 1999
Previous Article:Peroxide curing with precipitated silica.
Next Article:People in the News.

Related Articles
Use of reinforcing silica in model sidewall compounds: effects of carbon black type, polymer type and filler level.
Performance-driven black selection system.
The use of microwave ashing technology for the determination of carbon black in SBR.
The effect of carbon-black filling on the pyrolysis behavior of natural and synthetic rubber.
World CB consumption to reach nearly 8 million tons by 2000.
Role of carbon black in lowering overall compound costs.
Carbon black's effects on vibration isolation.
Increase in fines content and adhesion behavior in the pneumatic conveying of CB.
Cure system and carbon black effects on NR compound performance in truck tires.
New reinforcing materials for rising tire performance demands.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters