A new approach to optimize cure cycle of a tire using DSC.

The extent of cure provided to a tire compound plays an important role in its performance. An attempt has been made to apply a differential scanning calorimetry (DSC) technique to understand the state of cure (including over cure) of tire tread compounds. When an uncured compounded rubber is subjected to a time-temperature program in DSC, an exotherm (AH, enthalpy) is observed based on the nature of the polymer, curatives and other additives in the matrix. When the same compound is step cured for different times at a fixed temperature and then subjected to a similar temperature scan in DSC, the [DELTA]H reduces and tends to zero with the extent of cure. When the matrix is fully cured, i.e., no more crosslinking takes place, then instead of exothermic, sometimes we get endothermic behavior, depending on the nature of polymer and its tendency to reversion. Based on this observation, optimizations of cure cycles of two different tire tread compounds, viz. one is with 100% NR and the other with a blend of NR and BR (50:50) were done and have been supplemented with data obtained from the measurement of chemical crosslink densities and various physical properties.

The crosslinking of polymer molecules, also known as vulcanization or curing, is essentially used to achieve desired end properties of elastomeric materials. Compounded rubber, which is plastic in nature, becomes a useful viscoelastic material after curing. Various types of vulcanizing agents can be used to cure natural and synthetic rubbers, but the most commonly used system for vulcanizing the unsaturated rubbers in the tire industry is sulfur (ref. 1).

In the process of manufacturing a finished product, the extent of cure provided to the rubber compound, with an aim to impart a certain crosslink density, plays an important role in its performance.

Mechanical properties of the material are strongly dependent on the crosslink density of rubber. Modulus and hardness increase monotonically with increasing crosslink density, and the material becomes more elastic; or stated alternatively, less hysteretic. Fracture properties, such as tear and tensile strength, pass through a maximum as crosslinking is increased. Elastomers have an optimum crosslink density range for practical use. Crosslink levels must be high enough to prevent failure by viscous flow, but low enough to avoid brittle failure (ref. 2).

It has long been known that most of the performance properties show a maximum at a certain degree of cure (ref. 3). Researchers are engaged in the process of optimizing the extent of cure with an objective to enhance productivity and quality of a product.

Different techniques (refs. 4-7) are used for measuring the extent of cure. The very simple and oldest method is to step cure the compound and measure its porosity. The cure cycle is optimized based on the time when porosity disappeared (ref. 8). Also, step cured compounds are used for the measurement of physical properties and, based on the data, optimization of the cure cycle is done. Another method, which is frequently used in thick and composite products like tires, is a thermocouple technique (refs. 9 and 10). By this method, a temperature profile of the materials is measured, and subsequently the state of cure is expressed in terms of cure equivalents (i.e., time with reference to a standard temperature, where 141.7[degrees]C for one minute is considered as one cure equivalent). The cure equivalent is calculated using the Arrhenius Equation based on the temperature profile from thermocouple studies. Various other standard ASTM test methods (ref. 11) of measuring state of cure require evaluation of properties such as tensile, modulus, elongation, hardness, compression set and, in some cases, cold temperature brittleness.

All tests require either ASTM slabs or buttons. Moreover, all these processes are very tedious, time consuming and involve huge cost. Sometimes, lots of approximations are also required to get the required results.

In the present study, an attempt has been made in developing and applying a DSC technique, which is very fast and precise, to understand the state of cure and optimize the cure cycle of a rubber product. The method is based on the value of the exotherm (measured as cure enthalpy) obtained in the temperature region of approximately 175-250[degrees]C from a sample specimen when it is heated in DSC from 150-300[degrees]C at a scan rate of 20[degrees]C under inert atmosphere.

The method

It is observed that when an uncured compounded rubber is subjected to a time-temperature program in DSC, we typically get an exotherm ([DELTA]H, enthalpy) based on the nature of the polymer, curatives and other additives in the matrix. When the same compound is step cured for different times at a fixed temperature and then subjected to a similar temperature scan in DSC, the [DELTA]H reduces and tends to zero with the extent of cure (figure 1). In other words, as the crosslinking increases with the extent of cure, the exotherm indicating the residual vulcanization reaction that is taking place during the DSC experiment decreases. When the matrix is fully cured, i.e., no more crosslinking takes place, then instead of exothermic, sometimes we get endothermic behavior, depending on the nature of polymer, curatives and tendency to reversion. This observation is utilized to optimize the cure cycle.

[FIGURE 1 OMITTED]

To support this DSC measurement, the state of cure has also been followed and measured by chemical means, or more precisely by measuring crosslink density (a direct measure of extent of cure) of rubber using an equilibrium swelling method (refs. 12 and 13). A plot of crosslink density vs. curing time reveals that the extent of cure passes through a maximum (figure 2).

[FIGURE 2 OMITTED]

In this work, two different tire tread compounds were used (table 1), one is with 100% NR (lug tire) and the other with a blend of 50% NR and 50% BR (rib tire). They were cured stepwise at 142[degrees]C from 30 minutes to 70 minutes with an interval of 10 minutes. Cure enthalpy was measured by DSC for all of these step-cured compounds. Two separate calibration curves were also developed between the DSC enthalpy and cure times of the two different compounds in order to judge the state of cure of an unknown component of similar formulation by simply positioning its enthalpy value on the specific calibration curve. Physical properties, including hardness, of all these step-cured compounds were measured with an objective to correlate the DSC findings.

Experimental

Raw material used

The raw materials used for conducting this project were collected from standard Indian and international sources.

Mixing

Two compounds (1 and 2) of table 1 were prepared using a 3.0 liter laboratory internal mixer, with a fill factor of 0.8. Compound mixing was carried out in two stages. In the first stage (masterbatch), rubber was mixed with all other ingredients except curatives. The mixing was done at 50 rpm rotor speed for six minutes and subsequently sheeted out on a two-roll mill. After maturation (24 hours), the batches were mixed with the curatives in the final stage. The final batch mixing was done at 25 rpm for four minutes, and batches were dumped at 100-110[degrees]C.

Molding

Compounds were molded and cured under pressure (~10 MPa) at 142[degrees]C to various states of cure by varying cure times, starting from 30 minutes to 70 minutes with an interval of 10 minutes. The molded samples after curing were immediately cooled by ice-water to quench any further cure.

Measurement of cure enthalpy

Cure enthalpy was measured by using a Perkin Elmer Pyrisl DSC. A certain mass of sample (in the range of 15-20 mg) was encapsulated in a sample pan and was scanned from 150-300[degrees]C at a scan rate of 20[degrees]C/minute under nitrogen atmosphere.

Measurement of crosslink density

The procedure for the determination of crosslink density (CD) involves the swelling of a weighed sample of rubber (in the range of 0.2-0.4 g) in toluene for approximately 48 hours. The rubber is removed, blotted quickly with filter paper and weighed in a tared weighing bottle. Swelling index (expressed in percentage) is defined as the ratio of swollen weight to initial weight of the test sample. After removal of the solvent in a vacuum oven, the weight of imbibed solvent is obtained as the difference between the weight of swollen sample and dried sample. The extent of swelling is given by the volume fraction of the rubber ([V.sub.r]) in the swollen gel and is calculated using the method of Ellis and Welding (ref. 14).

[V.sub.r] = (D - FA) [rho]r-1/(D - FA) [rho]r- 1 + A0 [rho]s-1 (1)

where, D = swollen weight, F = fraction insoluble, A = sample weight, A0 = weight of the absorbed solvent corrected for swelling increment, [rho]r = density of rubber and [rho]s = density of solvent.

The crosslink density (number of crosslinks per unit volume) is calculated from [V.sub.r] by means of the Flory-Rehner relationship (ref. 12):

CD = [-ln(1- [V.sub.r]) + [V.sub.r] + X [V.sub.r] 2]/(2pV) [V.sub.r] 3 (2)

where, X = polymer-solvent interaction parameter, [rho] = density of polymer and V = molar volume of solvent.

Measurement of physical properties

Stress-strain behavior of specimens was determined according to ASTM D412, using type D dumbbells tested on a Z010 Zwick UTM at a crosshead speed of 500 mm/min. Compound hardness was measured by durometer (A) in accordance with ASTM D2240.

Results and discussion

Sulfur vulcanization of natural and synthetic elastomers is known to be a first order exothermic process (refs. 15 and 16). Because of this exothermicity of the vulcanization reaction, true isothermal conditions are never established in a vulcanizing elastomer system. Moreover, vulcanization processes are extremely complex, involving numerous consecutive and simultaneous reactions (refs. 17-19).

Differential scanning calorimetry (DSC) is a thermal technique capable of detecting endothermic and exothermic processes characteristic of elastomer systems. The observed enthalpy as obtained by DSC is the mathematical sum of the enthalpies of all individual reactions occurring in the temperature range of measurement.

In a first order reaction like sulfur vulcanization of natural and synthetic robber, the following conversion is assumed for a peak observed on the DSC curve when the sample is subjected to a controlled temperature heating:

A (Reactant) k [right arrow] B (Product) + [DELTA]H (3)

where, k = velocity constant or specific reaction rate constant and [DELTA]H = heat of reaction (vulcanization) or cure enthalpy.

The fractional conversion ([X.sub.n]) is defined as: [X.sub.n] = [DELTA][H.sub.u] - [DELTA]H.sub.ct]/[DELTA][H.sub.u] (4)

where, [DELTA]Hu = enthalpy of uncured compound, [DELTA][H.sub.ct] = enthalpy of compound cured for time t.

The residual heat of vulcanization (1 - [X.sub.n]) is proportional to the unreacted cure system and is a function of time.

DSC cure enthalpy ([DELTA]H) obtained for all step-cured samples from compounds 1 and 2 are shown in table 2. It is observed that the absolute value of the enthalpy decreases as the state of cure increases (figure. 1).

Negative enthalpy values are indicative of exothermic (vulcanization) reactions. Only one positive value, found in the case of compound 1 cured for 70 minutes, may be due to the reversion (endothermic) process, a common phenomenon observed with natural rubber when it is over-cured.

The amount and/or type of ingredients affect the cure enthalpy, i.e., compositional knowledge of samples is a must for comparison purposes. This is evident when individual cure enthalpy from the set of compounds 1 and 2 are looked into separately. An appreciable high value of cure enthalpy is found with the 100% NR formulation (compound 1) as compared to that observed with the blend of NR and BR (50:50) formulation (compound 2).

This trend of [DELTA]H value when studied along with the trend of [V.sub.r] value, CD and other physical properties with cure time helps us to predict the extent of cure accurately.

In order to support and correlate DSC findings, determination of crosslink density by equilibrium swelling measurements was also carried out to judge the state or the degree of vulcanization. Table 3 represents the values of swell index, volume fraction and crosslink density of all stepcured samples from compounds 1 and 2. The swelling index value decreases with increasing degree of cure. As the crosslinking of rubber is increased more and more, three-dimensional network structures are formed that restrict swelling in the solvent. In this way, when the crosslinking is increased further, the gel point is eventually reached, and the whole composition no longer dissolves in the solvent. Volume fraction and crosslink density, which can be expressed as the reciprocal of swelling index value, pass through a maximum as crosslinking is increased (figure 2 represents a typical curve for compound 1).

When physical properties are considered (table 4), as anticipated, the tensile, modulus and elongation all changed with increased cure time. The general trend is: With increasing cure time, increased hardness as a function of increased vulcanization is observed in all cases.

Modulus of the compounds 1 and 2 increases with cure time up to a certain extent, and then (after 50 minutes) it is found to be decreasing with cure time (table 4 and figure 3). This is due to the fact that, with increasing cure time, more and more crosslinks are formed, attained a maximum and then started decreasing with time (over-cure). This reversion phenomenon is prominent with natural rubber when it is overcured, i.e., destruction of crosslinks predominates over the usual bond formation while curing. A similar trend is also observed in the case of hardness (figure 4). So, maximum physical properties are attained at 50-minute cure time. Our DSC study also points towards similar findings. This corresponds to about 90% achievement of state of cure. By comparing these two findings (physical properties and DSC results), technically one can predict how much time is to be provided in the process of curing in order to get a quality product.

[FIGURES 3-4 OMITTED]

This behavior of vulcanizate systems allows DSC to be useful in monitoring the state of cure of a particular system or its heat history.

In the process of curing a tire in the press, the cure cycle is designed in such a way that it is cured up to the maximum extent, in the range of 90% corresponding to rheometric data. This is further corroborated with the optimum physical properties achieved at that cure state.

This concept can be applied successfully to predict the state of cure of a freshly cured tire or a service return tire made out of similar compound from the calibration curve drawn subsequently.

Two calibration curves for compounds 1 and 2 were constructed separately (figure 5) by the plotting of ln(1 - [X.sub.n]) vs. time of cure (t). The calibration curve is compound-specific, i.e., different calibration curves are needed for compounds with different formulations. When predicting the state of cure of a tire component, it must be assured that the formulation of the component cut from the tire is the same as the formulation of the compound used to construct the individual calibration curve. Information regarding whether a certain tire component is cured above or below optimum cure is determined directly from the specific calibration curve.

[FIGURE 5 OMITTED]

In addition, a limited portion of the heat history of a tire in service can also be determined, since all of the sulfur vulcanization reactions (plus all other reactions both endothermic and exothermic which may occur within this temperature range) have not reached completion at the stage when the tire is normally considered completely cured based on rheometer data. These reactions continue, even at the relatively low temperatures generated by the tire in service, and the heat history can be determined as long as those reactions have not reached 100% completion.

Conclusion

DSC thermal analysis can be used very effectively for finalizing the cure cycle. In addition, it can be used in monitoring the states of cure of an unknown sample, including a service-failed tire, made out of the same formulation. This cannot be predicted by other techniques with such a high precision. The major advantage of the DSC technique is the speed with which information can be generated accurately from a small sample. In practice, the sample needs to be scanned over the temperature range 150-300[degrees]C, which, at 20[degrees]C/minute, requires only a few minutes. Compositional knowledge of samples is a must for comparison purposes, because changes in amount and/or type of ingredients can affect the vulcanization enthalpy. Moreover, the agreement between the determinations of degree of cure from the DSC technique and from a solvent swelling method was found to be very good. As a result, a sample of any size or shape may be used for determination of the state of cure of an unknown sample. This makes the method adaptable to routine measurements for new or aged products, and will help in adjusting the cure cycle to achieve an optimum performance of a product.

References

(1.) J.A. Brydson, Rubber Chemistry, Applied Science Publishers Ltd., London, 1978.

(2.) G.R. Hamed in "Engineering with Rubber," ed. A.N. Gent, Hanser Publishers, Munich, 2001.

(3.) M.S. Dozortsev, VA. Sapronov and M.M. Reznikovskii, Soviet Rubber Technology, 41, p.25 (1966).

(4.) W.H. Bodger, Rubber Chem. Technol. 09, p. 95 (1936).

(5.) R.L. Warley and R.J. Del Vecchio, Rubber World, 30, September 1987.

(6.) A.I. Isayev and J.S. Deng, Rubber Chem. Technol., 61, p. 340 (1988).

7. L. Little, Elastomerics, February 1989, p. 22.

(8.) A.I. Kasner and E.A. Meinecke, Rubber Chem. Technol., 69, p. 424 (1996).

(9.) K. Hada and T. Nakajima, Rubber Chem. Technol., 06, p. 56 (1933).

(10.) H.F. Church and H.A. Daynes, Rubber Chem. Technol., 17, p. 923 (1944).

(11.) ASTM Standard 09.01 (2005).

(12.) D. De and A.N. Gent, Rubber Chem. Technol., 69, p. 834 (1996).

(13.) A.K. Chandra, A. Biswas, R. Mukhopadhyay, B.R. Gupta and A.K. Bhowmick, Plastics, Rubber and Composites Processing and Applications, 22, p. 249 (1994).

(14.) B. Ellis and G.N. Welding, Rubber Chem. Technol., 37, p. 563 (1964).

(15.) D.W. Brazier, Rubber Chem. Technol., 53, p. 437 (1980).

(16.) A.K. Chandra, A.S. Deuri, R. Mukhopadhyay and A.K. Bhowmick, Kauts. + Gummi. Kunst., 50 (2), p. 106 (1997).

(17.) D. W. Brazier and G.H. Nickel, Rubber Chem. Technol., 48, p. 26 (1975).

(18.) A.Y. Coran in "Science and Technology of Rubber," ed. F.R. Eirich, Academic Press, New York, 1978.

(19.) A.K. Sircar in "Thermal Characterization of Polymeric Materials," ed. E. A. Turi, Academic Press, California, 1997.

by Arup K. Chandra, Tapas Mandal, Bijan Kumar Roy and P. K. Mohamed, Apollo Tires Ltd.

Table 1--formulations

Ingredient                Compound 1       Compound 2
                               (phr)            (phr)

Natural rubber                   100             50.0
Polybutadiene rubber              --             50.0
Peptizer                         0.1             0.05
Zinc oxide                       5.0              2.5
Stearic acid                     3.0              2.5
Silica                            --              5.0
Carbon black                    50.0             55.0
Process oil                      8.0             14.0
Antioxidant/antiozonant          3.2              2.5
Wax                               --              2.0
Cure system                CV system   Semi EV system

Table 2--DSC results

Compound      Cured         Cure           Fractional cure
time (min.)               enthalpy      [Xn =([DELTA][H.sub.u] -
                      ([DELTA]H, J/g)     [DELTA][H.sub.ct])/
                                          [DELTA][H.sub.u]]

Compound 1        0     (-) 13.465                  --
                 30      (-) 3.355              0.7508
                 40      (-) 1.863              0.8616
                 50      (-) 0.888              0.9341
                 60      (-) 0.454              0.9663
                 70      (+) 0.095              1.0070

Compound 2        0      (-) 6.461                  --
                 30      (-) 1.333              0.7937
                 40      (-) 0.686              0.8938
                 50      (-) 0.633              0.9020
                 60      (-) 0.429              0.9336
                 70      (-) 0.232              0.9641

Compound      Cured       Residual             In
time (min.)                 cure        (1 - [X.sub.n])
                      (1 - [X.sub.n])

Compound 1        0             -               --
                 30        0.2492            -1.39
                 40        0.1384            -1.98
                 50        0.0659            -2.72
                 60        0.0337            -3.39
                 70       -0.0070               --

Compound 2        0            --               --
                 30        0.2063            -1.58
                 40        0.1062            -2.24
                 50        0.0980            -2.32
                 60        0.0664            -2.71
                 70        0.0359            -3.33

Table 3--swelling data

Compound      Cured   Swell     Volume     Crosslink density,
time (min.)           index,   fraction,     CD(g molelg of
                        Si     [V.sub.r]         rubber)

Compound 1       30    2.79      0.2415         8.5 E-05
                 40    2.72      0.2498         9.3 E-05
                 50    2.69      0.2539         9.6 E-05
                 60    2.67      0.2557         9.8 E-05
                 70    2.71      0.2515         9.4 E-05

Compound 2       30    2.91      0.2173         8.4 E-05
                 40    2.80      0.2281         9.3 E-05
                 50    2.79      0.2301         9.5 E-05
                 60    2.78      0.2307         9.5 E-05
                 70    2.79      0.2294         9.4 E-05

Table 4--physical properties

Compound      Cured     100%       200%       300%      Tensile
time (Min.)           modulus,   modulus,   modulus,   strength,
                        MPa        MPa        MPa         MPa

Compound 1       30     2.5        6.6        12.0        26.7
                 40     2.8        7.4        13.1        25.9
                 50     2.8        7.5        13.3        25.8
                 60     2.7        7.2        12.9        25.9
                 70     2.7        7.2        12.8        25.6

Compound 2       30     1.7        4.3         8.1        20.0
                 40     1.9        5.0         9.1        19.6
                 50     2.0        5.1         9.4        20.5
                 60     2.0        5.0         9.3        20.2
                 70     1.9        5.0         9.3        19.6

Compound      Elongation   Hardness
time (Min.)    at break,   (duro A)
                   %

Compound 1        540          61
                  504          63
                  501          65
                  514          63
                  513          63

Compound 2        594          59
                  539          61
                  543          62
                  541          62
                  525          61