Improving thermal conductivity and service life in bladder compounds.

The use of carbon black dates back approximately 5,000 years. Initially lamp blacks were used as pigments for inks and in cosmetics in India and China. In 1892, the process of channel black was developed. The resultant product was finer in nature than lamp black, and its use expanded to rubber goods. The next big discovery was in 1904, when S.C. Mote working for the India Rubber, Gutta Percha and Telegraph Works in Silverton, England, discovered the reinforcing nature of carbon black. This discovery, along with the growth of the pneumatic tire industry, started the growth of carbon black consumption. To meet this increasing demand for carbon blacks, in the 1920s two new carbon black production processes were introduced, the thermal process and the gas furnace process. These two processes all but eliminated the use of lamp and channel blacks by the 1940s. The next breakthrough in manufacturing processes occurred in 1943 with the introduction of the oil furnace process. Even though other fillers and processes have been introduced since then, oil furnace carbon blacks have remained the predominant filler. This is due to carbon blacks' unique properties and cost-performance ratio. In 2002, a new type of high-performance, thermally-modified carbon was jointly developed by Columbian Chemicals and Superior Graphite, and the objective of this article is to elucidate the performance benefits of these new high performance carbons.

Carbon blacks are one of the most versatile and unique nanomaterials available on such a large commercial scale, and they are used for a wide variety of applications. The versatility of carbon blacks comes from their turbostratic, carbon-based nature, organophilic surface and broad range of particle and aggregate size and shape distributional properties (ref. 1).

The high performance carbons described in this work undergo thermal modification in a continuous process that allows control of the degree of thermal modification of the internal microstructure. Additionally, the carbon black type or morphology (fineness and structure) can be tailored to meet the performance requirements of the end product. This unparalleled degree of control allows the rubber and plastics compounder flexibility in the design of articles to optimize performance and processability. Additionally, these carbons are very pure and provide many properties favorable for very critical applications such as nil sulfur, low polynuclear aromatic compound content (these carbons are FDA compliant; 21CFR 1783297 for indirect food contact applications), oxidation resistance, negligible moisture pickup, and reduced ash and metals.

The internal microstructure of the high performance carbons can be controlled via the thermal modification process. The differences in the internal microstructure of the high performance carbon (tradename Pureblack Carbon and SCD grade) compared to a typical furnace carbon black and an acetylene black are shown in figure 1. Note that the graphitic layers become more ordered or graphitic in going from the "as-produced" furnace black to the high performance carbon, which improves the electrical and thermal conductivity of the high performance carbon versus the as-produced furnace black. Acetylene carbon blacks are also partially graphitized carbon blacks as-produced, which, in addition to its purity (acetylene is the feedstock), has made them one of the carbon blacks of choice for conductive plastics and battery applications. However, the new class of high performance carbons, while comparable in their internal microstructural properties to acetylene blacks, offers significant advantages since they can be produced with a wider range of morphological properties and graphitization to optimize compound processability and performance.

[FIGURE 1 OMITTED]

One application where these new high performance carbons offer performance improvements is in bladder applications. Bladders, in general, are used in various applications such as tire curing. Here, they require long service life with other special properties such as low permeability and good thermal conductivity. Good bladder life can be achieved through the use of the appropriate materials, including polymers, curing resins (ref. 2) and performance fillers such as carbon blacks. Typically, butyl rubber is used for good service life and reduced permeability to gases, with tire innerliners being a classic example where butyl rubber provides the necessary physical properties. For tire curing bladders, butyl rubber is again used; however, in this case, the additional requirement of good thermal conductivity is important for efficient transfer of heat during the tire curing process. To this end, the new class of high performance carbons can be utilized to deliver higher thermal conductivity to the curing bladder to potentially increase tire throughput and increase bladder service life.

Experimental

Formulation and experimental design

The model curing bladder formulation (ref. 3) shown in table 1 was used in conjunction with an experimental design with varied filler type, loading and blend ratio for evaluation of the curing bladder compounds. A series of the new class of high performance carbons was also part of the experimental design (table 2). The compounds were mixed in one pass in a 1.5 liter internal mixer with tangential rotors. All polymers and zinc oxide were added at 0 seconds, followed by 75% of the carbon black added at 30 seconds, with the remaining carbon black and process oil added at 90 seconds, and the stearic acid added at 180 seconds. The compounds were then placed on a two roll mill, and the phenolic resin was added and cross blended six times before sheeting off for curing. Where applicable, all rubber physical properties were measured according to ASTM test protocols.

A variety of carbon blacks is typically used in curing bladders and may include, among others, N300 and N200 series grades, and acetylene black in varied blend ratios. The high performance carbons, which can be considered as acetylene black replacements, were mixed alone and in various ratios with N330. Table 2 shows the list of carbon black variations used in this study.

High performance carbons

The high performance carbons used in this study were prepared by thermally modifying the microstructure using the proprietary furnace process of Superior Graphite. Two SCD carbons were produced for this study which included a high structure, low-surface-area carbon (SCD205) and a medium structure, medium-surface-area carbon. The latter carbon was also thermally modified at three different conditions (SCD-545, 550 and 555) to examine the effect of microstructural changes on its performance in the curing bladder with regard to reinforcement and thermal conductivity.

Rubber properties and thermal conductivity testing

The dispersion level of all samples was characterized using the interferometric microscopy (IFM) technique recently developed by Columbian Chemicals (ref. 4). Static stress-strain properties were determined using a multistation tensometer. Capillary rheometry data were acquired using a capillary rheometer at 180[degrees]C at shear rates of 20[s.sup.-1], 100[s.sup.-1], 500[s.sup.-1] and 1,000[s.sup.-1]. Fatigue life data were determined using a United FTF-48 at constant peak-to-peak strain (0 to 100% elongation). Die C tear strength was determined using a Monsanto Tensometer 2000, and thermal conductivity was evaluated by Akron Rubber Development Labs (ARDL) using a thermal conductivity analyzer at room temperature (23[degrees]C) according to ASTM C-518.

Results and discussion

The curing bladder compounds mixed in this study were tested for filler dispersion, shear viscosity, tensile properties, failure properties (fatigue life and tear) and thermal conductivity. The primary areas of focus that can have the largest impact in improving curing bladder performance include theology, bladder life and thermal conductivity. The rheological properties are important because curing bladders can be pressure or injection molded, and any new compounds must be similar in their mold filling properties to compounds used today. The failure properties such as fatigue life and tear are also very important properties to consider. The fatigue life is important because the curing bladder lifetime in a plant has a direct impact on the number of cure cycles possible and the number of bladders needed to produce a given number of tires. Of course, longer fatigue life or bladder lifetime is desired as it has the potential to reduce operating costs. The tear properties, especially hot tear, are important to insure the bladder maintains its integrity upon release from the mold during its production. Lastly, thermal conductivity improvements may result in shorter cure cycles which can increase tire throughput. This last property is especially important as tires are getting larger, which increases curing time.

Dispersion and processability

Table 3 shows the dispersion index for the various compounds mixed in this study. As can be seen, the dispersion of all compounds was very good. The dispersion index (DI), which ranges form 0 to 100, was generally higher than 94 for most compounds. A few samples, mixed at 65 phr (SCD 550 and 555) had poorer dispersion, with DI values in the 83 to 88 range. However, this did not occur in the compounds where the filler blends were used, so overall the compounds and SCD grades dispersed well. It should be noted that the acetylene carbon black can be fluffy and powdery in nature, and occupies a larger volume for a given weight. This fluffy nature of acetylene carbon black makes it difficult to handle in a production environment. However, the new high performance carbons are in beaded form and provide better handling, less dusting and good mixing characteristics.

The rheological properties of the uncured compounds were determined on a capillary rheometer at shear rates of 20[s.sup.-1], 100[s.sup.-1], 500[s.sup.-1] and 1,000[s.sup.-1]. Table 3 also contains the shear viscosity data, and figures 2 and 3 show examples of the shear viscosity as a function of shear rate for the control compounds and various SCD grade compounds. As can be seen, graphically it is difficult to observe any significant differences in the shear viscosity of the control compounds versus the SCD grade compounds. However, further analysis shows that the SCD grades overall have equal to lower shear viscosity than the three control compounds, with only a few exceptions.

[FIGURE 2 and 3 OMITTED]

Tensile properties

The static stress-strain values are important as verification of the general reinforcement properties of the high performance carbons versus the controls. The 100% modulus, tensile strength and elongation to break are plotted in figures 4 and 5, respectively, for the controls and unblended SCD grades. As can be seen in figure 4, the acetylene (65 phr) and acetylene/ N330 (50/12 phr) blend compounds are similar in modulus, with a slight increase upon addition of N330, a more surface active carbon black. The N330 (60 phr) mix, however, shows a drop in modulus, possibly due to its lower structure and slightly lower loading. However, the 65 phr SCD compounds, compared to the control samples, show a relatively significant drop in 100% modulus, primarily due to their low surface activity as a result of the thermal treatment. Modulus development is a function of carbon black structure and surface activity, and these results show the importance of surface activity in modulus development. Figure 5 shows the tensile strength and elongation to break data for the control compounds and the SCD compounds at 65 phr. The results show comparable tensile strength values for all compounds, with the SCD compounds being directionally lower, but with a higher elongation at break due to their lower modulus.

[FIGURE 4 and 5 OMITTED]

Figures 6 and 7 are plots of the 100% modulus, tensile strength and elongation to break, respectively, showing the comparison of a range of blends of the high performance carbons with N330 versus the control compounds. As expected, as the loading of the SCD grades increases (while the N330 loading is held constant at 12 phr), the modulus increases. These compounds are very similar in modulus to the 65 phr SCD compounds, indicating that the 12 phr of N330 only provides a small modulus enhancement. As an outcome of these results, alternatives to N330 are currently being considered that may increase modulus, while maintaining the processability characteristics and other important performance indicators for curing bladders.

[FIGURE 6 and 7 OMITTED]

The tensile strength and elongation at break for the SCD grade blends with N330 (figure 7) again show the same pattern versus the control compounds, i.e., they (the SCD blends) have a similar tensile strength, but a higher elongation at break most likely due to their lower modulus. However, note that as the SCD grade loading is increased, a decrease occurs in the tensile strength and elongation at break, possibly as a result of the dilution effect, general to all fillers, where the volume fraction of polymer in the composite decreases with increasing filler loading. Note the contrast with the control compounds, where in going from all acetylene to the acetylene/ N330 blend to all N330, an increase in tensile strength is observed, since a more surface active carbon (N330) is added to the compound.

Failure properties: Fatigue life and tear

The characteristic fatigue life of the compounds was determined under constant peak-to-peak strain (0 to 100% elongation) and is plotted in figures 8 and 9. The data were analyzed with the Weibull procedure. As can be seen in figure 8, the SCD grades show higher fatigue life than the acetylene and acetylene/N330 control compounds. This is believed to be primarily due to the lower modulus of these compounds, which results in less energy input during the constant (0 to 100%) strain, even though one might surmise, as in the tensile properties, that the inert surface of the SCD grades would lead to a higher probability of crack initiation. In fact, the SCD550 (65 phr) compound had the highest characteristic fatigue life in this group.

[FIGURE 8 and 9 OMITTED]

Figure 9 shows the characteristic fatigue life data for the SCD/N330 blends versus the control compounds. The SCD/ N330 blends at 40/12 phr loading ratio had characteristic fatigue life superior to the controls. The rest of the blends had characteristic fatigue life equal to or lower compared to the controls. As expected, as the SCD grade loading is increased, the fatigue life decreases. This result is in line with the observed changes in modulus, where increasing SCD grade loading increased the modulus, thus increasing the energy input at constant deformation. Also of note is the fact that the tensile strength decreased with increasing loading. Higher loadings increase the strain amplification and can reduce the fatigue life.

Figures 10 and 11 show the Die C tear strength for the control compounds and the SCD grade compounds. As can be seen, overall the tear strength of the various SCD compounds was equal to or lower than the control compounds. It appears that in the case of tear strength, the inert surface of the SCD grades is detrimental to this failure property and may provide a higher probability for crack initiation. Note however, that as the SCD grade concentration increases (figure 11) the tear strength increases, due to higher filler loading.

[FIGURE 10 and 11 OMITTED]

Thermal conductivity

Thermal energy is transmitted through crystalline solids via phonons (lattice vibrations) and is controlled by the inter-atomic forces within the solid or crystal (ref. 5). In the case of carbons, namely graphite, the phonon transfer is anisotropic because of the different bonding mechanisms for the in-plane or basal plane carbon atoms (the <a> plane direction, covalent bonding) and the between-basal-plane bonding (the <c> plane, van der Waals bonding), with the most efficient conduction occurring in the basal plane direction (ref. 5). In fact, the in-plane (<a> direction) thermal conductivity (~1,000 W/mK) is 100 times greater than the between-plane (<c> direction) thermal conductivity (~10 W/mK) at room temperature (ref. 5). In considering the images in figure 1, the high performance carbons appear more graphitic and "'crystalline" compared to the as-produced carbon black, which is more amorphous or paracrystalline than the thermally treated carbons, with smaller graphene basal planes or parallel layers in the <a> dimension.

Indeed, the x-ray crystallographic properties shown in table 4 for the carbons listed in table 2 confirm the differences in crystallinity. As can be seen, the N330 carbon black has the smallest crystallites in the <a> dimension ([L.sub.a]) and the smallest stacking height in the <c> direction ([L.sub.c]), followed by the acetylene black and then the high performance carbons, which have the highest degree of crystalline order. Thus, one might expect that the more graphitic, high performance carbons would possess a higher thermal conductivity than the more paracrystalline carbon blacks.

The plot in figure 12 shows the thermal conductivity of the various curing bladder compounds containing the acetylene black, acetylene/N330 blend, N330 and SCD 205, 545, 550 and 555, respectively. As can be seen, the curing bladder compound with the highest thermal conductivity is the SCD 205, followed by the SCD 545-555, and then the acetylene carbon black. Note that the acetylene/N330 blend and the all N330 compound have the lowest thermal conductivity. These results, in general, confirm the higher thermal conductivity of the more graphitic carbon blacks.

[FIGURE 12 OMITTED]

Figures 13 and 14 show plots of the curing bladder compound thermal conductivity versus the x-ray diffraction properties of the individual carbon blacks ([L.sub.a], average graphene layer dimension, and Lc, average crystallite stacking height), again demonstrating the direct correlation of microstructural order with increasing compound thermal conductivity.

[FIGURES 13 AND 14 OMITTED]

It should be noted that morphological differences may play a role in affecting thermal conductivity. However, in a previous study (ref. 6), Atkins and Sullivan found that there was only approximately a 15% difference in thermal conductivity in a butyl compound containing carbon blacks ranging from a thermal black to a highly conductive carbon black. Excluding these two specialty carbon blacks, the range in thermal conductivity narrowed even further (~8% difference) between the ASTM rubber furnace blacks. They also showed that there was little or no change in thermal conductivity of a butyl curing bladder compound at 100[degrees]C versus the higher temperatures at which curing bladders are used.

Summary and conclusions

As discussed, the three important properties targeted for improvement of the curing bladder manufacturing and performance included maintaining viscosity, longer bladder life and higher thermal conductivity. The control compounds (acetylene 65 phr, acetylene/N330 50/12 phr and N330 60 phr) were used as benchmarks to determine the best SCD compounds and their relative advantages in meeting or exceeding targets in processability (viscosity), increased bladder life (fatigue life) and reduced cure cycle times (better compound thermal conductivity). The three control compounds listed above were chosen to represent the large variety of formulations and carbon blacks used in the marketplace. Thus, comparison of the high performance carbon data with those of the three control compounds will allow a wide range of compounders to develop formulations and to understand the benefits that the SCD grades can provide.

Evaluation of the data showed that the two compounds that met these criteria are the SCD550/N330 (50/12 phr) blend and the SCD555/N330 (50/12 phr) blend. However, for optimum thermal conductivity, the SCD205 65 phr compound gave the best results; although, it gave slightly lower tear strength relative to the control compounds. Tables 5-8 show the degree of improvements attainable with the SCD grades. Note in table 5 that significant improvement in fatigue life ranging from 30% to >100% is possible for any of the three SCD grade combinations relative to the control compounds. However, the N330 (65 phr) control compound showed good fatigue life, and was generally comparable to the SCD grade combinations.

For the processability, shear viscosity at 20[s.sup.-1] and 1,000[s.sup.-1] was used to compare the lowest and highest shear rates tested. As can be seen in tables 6 and 7, the SCD grade compounds had lower shear viscosity at both shear rates; on the order of 1% to 25% lower versus the control compounds, indicating better to equal mold filling properties for the SCD grades. However, again the SCD grade compounds were generally equal to slightly higher in shear viscosity versus the N330 (60 phr) control.

Lastly, table 8 shows the improvements in thermal conductivity that the SCD grades provide, and they range from 4% to 79% higher than the control compounds. The N330 control showed the lowest thermal conductivity for reasons explained previously.

In summary, the new high performance carbons are unique, and provide a step change in specialty carbon materials. The ability to control the internal microstructural crystalline characteristics over a wide range of morphology (fineness and structure) will provide materials manufacturers the opportunity to further optimize their compound processibility and performance.

In particular, this work has focused on one application, curing bladders for tires, and it has been shown that indeed these carbons give performance improvements in processability, bladder lifetime and thermal conductivity, providing the tire maker the potential to reduce costs and increase tire throughput.

References

(1.) J.B. Donner, R.C. Bansal and M.J. Wang, "'Carbon Black," 2nd Ed., Marcel Dekker, Inc., New York, 1993.

(2.) R.P. Lattimer, R.A. Kinsey, R. W. Layer and C.K. Rhee, Rubber Chem. Technol. 62, 107 (1989).

(3.) H.E. Rooney, "Polysar Butyl Handbook, " The Ryerson Press, Toronto, p. 358, 1996.

(4.) A.P. Smith, T.L. Aybar, R. W Magee and C.R. Herd, Rubber Chem. Technol. 77, 691 (2004).

(5.) "Graphite and precursors" in "World of Carbon, " Volume 1, edited by Pierre Delhaes, Overseas Publishers Association (opa) n.v 2001, published by license under the Gordon and Breach publishers imprint.

(6.) J.H. Atkins and J.E. Sullivan, Jr., Rubber Chem. Technol., 42, 1,314 (1969).

Table 1--model curing bladder formulation

Ingredient                          Phr

Butyl 301                           100
Neoprene W                           5
Carbon black                       Varied
Stearic acid                         2
Zinc oxide                           5
SP1045 (phenolic curing resin)       8
Processing oil                       6

Table 2--experimental design for carbon black and high performance
carbon (SCD grades) loadings in a model curing bladder

Carbon                     Phr

Acetylene   65   50
N330             12   60
SCD205                     65
SCD545                          65
SCD550                               65
SCD555                                    65

Carbon                 Phr

Acetylene
N330        12   12   12   12   12   12
SCD205      40   50   60
SCD545                     40   50   60
SCD550
SCD555

Carbon                  Phr

Acetylene   12   12   12   12   12   12
N330
SCD205
SCD545      40   50   60
SCD550                     40   50   60
SCD555

Table 3--dispersion index (DI) and rheological properties
(180[degrees]C, capillary rheometer) for butyl curing bladder compounds
containing high performance carbons and acetylene and N330 controls

                                                 Shear vicosity
                            Dispersion               Pa *s

Compound                 IFM DI   20[s.sup.-1]   100[s.sup.-1]

Acetylene, 65phr          97.7       11,986          4,837
Acet./N330, 50/12 phr     97.0       14,310          4,705
N330, 60phr               97.8       11,311          4,077
SCD205, 65 phr            96.6       10,790          4,952
SCD545, 65 phr            87.8       11,265          4,592
SCD550, 65 phr            96.2       11,693          4,170
SCD555, 65 phr            83.7       11,721          4,330
SCD205/N330, 40/12 phr    94.4       8,230           3,805
SCD205/N330, 50/12 phr    96.5       10,687          4,454
SCD205/N330, 60/12 phr    95.5       12,126          5,058
SCD545/N330, 40/12 phr    96.1       10,970          3,964
SCD545/N330, 50/12 phr    95.6       11,613          4,243
SCD545/N330, 60/12 phr    95.6       13,480          4,898
SCD550/N330, 40/12 phr    86.7       11,183          3,981
SCD550/N330, 50/12 phr    94.7       12,080          4,474
SCD550/N330, 60/12 phr    97.5       13,682          4,937
SCD555/N330, 40/12 phr    91.4       9,212           3,479
SCD555/N330, 50/12 phr    93.3       10,808          3,989
SCD555/N330, 60/12 phr    92.9       12,363          4,603

                            Shear vicosity Pa *s

Compound                 500[s.sup.-1]   1,000[s.sup.-1]

Acetylene, 65phr             1,326             595
Acet./N330, 50/12 phr        1,286             528
N330, 60phr                  1,156             442
SCD205, 65 phr               1,462             588
SCD545, 65 phr               1,374             519
SCD550, 65 phr               1,271             487
SCD555, 65 phr               1,316             515
SCD205/N330, 40/12 phr       1,078             373
SCD205/N330, 50/12 phr       1,293             502
SCD205/N330, 60/12 phr       1,476             634
SCD545/N330, 40/12 phr       1,097             377
SCD545/N330, 50/12 phr       1,244             459
SCD545/N330, 60/12 phr       1,486             609
SCD550/N330, 40/12 phr       1,095             374
SCD550/N330, 50/12 phr       1,292             476
SCD550/N330, 60/12 phr       1,468             584
SCD555/N330, 40/12 phr        988              346
SCD555/N330, 50/12 phr       1,185             442
SCD555/N330, 60/12 phr       1,421             582

Table 4--x-ray diffraction data for carbon black grades mixed in curing
bladder compounds

                     X-ray diffraction crystalline data

                  [L.sub.a],nm   [L.sub.c],nm   [d.sub.002],nm

N330                  1.9            1.6            0.358
Acetylene black       3.3            2.9            0.351
SCD545                5.6            4.9            0.346
SCD550                6.1            5.3            0.345
SCD555                6.6            5.7            0.346
SCD205                6.5            7.4            0.345

Table 5--characteristic fatigue life comparison in curing bladder
compounds

Characteristic fatigue life (kcycles)

                                     Control compounds
Percent improvement in
fatigue life versus      Acetylene   Acetylene/N330        N330
control                    65 phr        50/12 phr       60 phr

SCD550/N330, 50/12 phr       +79%             +57%          +7%
SCD555/N330, 50/12 phr      +120%             +94%         +32%
SCD205, 65 phr               +50%             +31%          -7%

Table 6--shear viscosity advantages of high performance carbons (SCD
grades) versus control compounds at shear rate of 20 [s.sup.-1]

                         Shear viscosity at 20[s.sup.-1]

                                     Control compounds
Percent reduction in
shear viscosity at       Acetylene   Acetylene/N330        N330
20 [s.sup.-1]              65 phr        50/12 phr       60 phr

SCD550/N330, 50/12 phr      Equal        15% lower    7% higher
SCD555/N330, 50/12 phr   7% lower        24% lower     4% lower
SCD205, 65 phr           10% lower       25% lower     5% lower

Table 7--shear viscosity comparison of high performance carbons (SCD
grades) versus control compounds at shear rate of

                         Shear viscosity at 1,000[s.sup.-1]

                                     Control compounds
Percent reduction in
shear viscosity at       Acetylene   Acetylene/N330        N330
[1,000.sup.-1]             65 phr        50/12 phr       60 phr

SCD550/N330, 50/12 phr   20% lower       10% lower    8% higher
SCD555/N330, 50/12 phr   26% lower       16% lower        Equal
SCD205, 65 phr           1% lower       11% higher    33% higher

Table 8--thermal conductivity comparison showing advantages of high
performance carbons (SCD grades) versus control compounds of acetylene,
acetylene/N330, and N330 in curing bladder compounds

                         Thermal conductivity (W/mK)

                                     Control compounds
Percent increase in
thermal conductivity     Acetylene   Acetylene/N330        N330
versus control             65 phr        50/12 phr       60 phr

SCD550/N330, 50/12 phr        +4%             +11%         +40%
SCD555/N330, 50/12 phr        +4%             +11%         +40%
SCD205, 65 phr               +33%             +42%         +79%

Weidong Wang, Rikki Lamba, Charles Herd, Deepak Tandon and Charles Edwards, Columbian Chemicals