Improved co-vulcanization of rubber blends.

Dithioates are liquid substances available in either a dry liquid concentrate (silica bound) or as granules (polymer bound). These products have very good solubility and can be uniformly dispersed in the compound due to their good dispersion characteristics. The improved dispersion causes a more homogeneous distribution of the chemical which gives a better co-vulcanization. Thus, more homogeneous vulcanizate properties are achieved, e.g., better dynamic properties demonstrated by lower heat built up.

When blending two different polymers with significantly different polarities, it is often difficult to achieve certain properties needed to meet the continuously improved performance demanded by the industry. Using blends of incompatible polymers (NBR and NR with EPDM), this work shows that a better co-vulcanization can be obtained with the proper curing system (ref. 1). These rubber blends cured using dithioate curing systems provided better properties than those of the control. Dithioate is a Rhein Chemie chemical family name for all phosphorous-sulfur based curing agents and mixtures thereof. Traditionally, rubber blends were cured with standard vulcanizing agents, e.g., thiurams and carbamates. Many of those accelerators form carcinogenic N-nitrosamines, which are now under public criticism (refs. 2 and 3).

The solubility and diffusion rate of the curing agents influence the cured state of the single blend components. For blends of polymers, the compatibility of the molecules of both polymers will have an effect on the homogeniety of the final blend. The miscibility of the two polymers can be divided into thermodynamic compatibility and technological compatibility. Two substances are thermodynamically compatible when the molecules completely intermingle. This means that the mixing enthalpy [Delta][G.sub.m] (equation 1) is negative ([Delta][G.sub.m] [is less than] 0) (ref. 4).

(1) [Delta][G.sub.m] = [Delta][H.sub.m] - T [multiplied by] [Delta][S.sub.m] at T = constant; p = constant.

Three cases can be distinguished; all with an increase of entropy.

Case 1: [Delta][S.sub.m] [is greater than] 0, [Delta][H.sub.m] [is less than] 0; exotherm mixing process.

Case 2: [Delta][S.sub.m] [is greater than] 0, [Delta][H.sub.m] = 0; adiabatic mixing process.

For both cases, at all temperatures, the two-polymer system is completely miscible. Thus, the values of the mixing enthalpy [Delta][G.sub.m] are negative. The effect is a homogeneous single-phase system with one glass transition point, [T.sub.G]. This ideal case is only achieved for blends of the same polymer, e.g., SBR types with different styrene content (ref. 4).

Case 3: [Delta][S.sub.m] [is greater than] 0, [Delta][H.sub.m] [is greater than] 0; endotherm mixing process.

a) [Delta][H.sub.m] = T [multiplied by] [Delta][S.sub.m] ?? [Delta][G.sub.m] = 0; the blend is before and after mixing in an energetic equilibrium.

b) [Delta][H.sub.m] [is less than] T [multiplied by] [Delta][S.sub.m] ?? [Delta][G.sub.m] [is less than] 0; as in cases 1 and 2, both polymer phases are completely mixable, but the dependence on temperature must be considered.

c) [Delta][H.sub.m] [is greater than] T [multiplied by] [Delta][S.sub.m] ?? [Delta][G.sub.m] [is greater than] 0; "practical case" of technological mixability

Technological mixability describes a different dispersion of a polymer phase in a rubber matrix, i.e., a microheterogeneous two-phase polymer system. The extent to which the polymer chains penetrate the phase boundary is dependent on the rubber polarity and the technological restraint during the blend production.

Figure 1 shows a classification of different rubber types through the solubility parameter [Delta]. The solubility parameter [Delta] is a direct measure of the interaction energy between similar molecules, thus, the solubility parameter [Delta] has the following unit: [J.sup.1/2]/[cm.sup.3/2] (ref. 5).

Figure 1 - [Delta] polarity parameter of rubber (ref. 7)

EPDM NR BR   NBR               NBR       NBR
             18% ACN           34% ACN   41% ACN

15,5                                     21,5
             [(J/[cm.sup.3])
             .sup.1/2]

[Delta] is composed of the cohesion energy [E.sub.cohesion] and the molecular volume v as shown in equation 2 (ref. 6).

(2) [Delta] = [([E.sub.cohesion/[bar]v]).sup.1/2]

with [E.sub.cohesion] [J] = cohesion energy, [bar]v [[cm.sup.3]] = molecular volume.

The cohesion energy is a unit of measurement for the strength of intermolecular forces that hold together a fluid and [E.sub.cohesion] [approximately equals] [Delta] [[bar]H.sub.v] (molar evaporation enthalpy). If both quantities are related to the molar volume, then it can be obtained following equation 3 (ref. 6):

(3) [E.sub.cohesion]   [approximately   [Delta] [[bar]H.sub.v]
    /[bar]v            equals]          /[bar]v

    cohesion energy                    evaporation enthalpy
    density                            density

The evaporation enthalpy [Delta] [[bar]H.sub.m] / [bar]v can be described as follows in equation 4.

(4) [Delta] [[bar]H.sub.v]/[bar]v = [v.sub.1]/([v.sub.1] + [v.sub.2]) [multiplied by] [v.sub.2]/([v.sub.1] + [v.sub.2]) [multiplied by] [([[Delta].sub.1] - [[Delta].sub.2]).sup.2]

[v.sub.1] [[cm.sup.3]] = volume of polymer phase 1

[v.sub.2] [[cm.sup.3]] = volume of polymer phase 2

[Delta] [J/[cm.sup.3] = solubility parameter

By correspondence of both solubility parameters, [[Delta].sub.1] and [[Delta].sub.2], the difference becomes zero and consequently the evaporation enthalpy density and the cohesion energy density likewise become zero.

The cohesion energy is a sum of dispersive, polar and hydrogen bonding contributions.

(5) [E.sub.cohesion] [J] = [Sigma] "minor valence forces" = [E.sub.d] + [E.sub.p] + [E.sub.h]

Also, the solubility parameter [Delta] is composed of three parts:

(6) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

It follows that two substances are ideally miscible if the polar contributions of the solubility parameter correspond, i.e., [[Delta].sub.p1] = [[Delta].sub.p2]. These parameters are often not known and are difficult to determine, so the solubility parameter [Delta] is usually little more than a guess for classifying the different polymer types through their polarity. Even so, the difference between the solubility parameters of the different polymers in blends is an indication of the mixability or miscibility of the two rubbers. As the polarity difference between the two polymers increases, the surface tension between the two polymers likewise increases with a subsequent decrease in miscibility. Consequently, the thickness of the interphase layer decreases.

The surface tension can be considered the sum of a dispersive and a polar contribution. The polar contribution is based on dipole-dipole interactions; whereas, the dispersive contribution is a function of Van der Waals forces. An optimal copenetration or intermingling of both phases is reached as the polar

contributions of each polymer phase become equal (ref. 8).

Dispersion and solubility of rubber chemicals in polymers

The dispersion of rubber chemicals is dependent on the panicle size and physical state. If liquid chemicals are adsorbed on fine materials like silica, then dry liquid concentrates can be obtained that are easily dispersed into rubber. On the other hand, after powdered chemicals are dispersed throughout the rubber matrix, they must then dissolve into the rubber before they can contribute to the curing reaction. The quality of chemical dispersion influences the tendency toward blooming. It is for this reason that much information is given in the literature concerning the solubility of rubber chemicals in various polymers (refs. 9 and 10). Table 1 clearly shows that many rubber chemicals have a low solubility in different polymers, whereas, liquid dithioates are very soluble (ref. 5).

Table 1 - solubility limit of accelerators in
different polymers

Chemical    Melting pt.     EPDM      NR     SBR     NBR
           [[degrees] F]   [phr]   [phr]   [phr]   [phr]

TMTD                 278    <0.5    <0.5    <0.5    <3.0
TBzTD                266    <0.5    <2.0    <0.5    >5.0
DTDM                 266    <1.5    <2.5    >5.0    <5.0
CLD                  248    <1.0    <0.5    <1.0    >0.5
CBS                  204    <2.0    >5.0    <1.0    >5.0
TBBS                 220    <2.0    <0.5    <0.5    >5.0
MBTS                 354    <1.0    <0.5    <0.5    <1.0
ZBEC                 348    <0.5    <2.5    <0.5    <3.5
ZDBC                 226    <2.5    <3.0    <2.5    >5.0
TP                    RT    <6.0    <1.5    <6.0    >7.0
ZADT              liquid    >5.0    >5.0    >5.0    >5.0
ZDT               liquid    <6.0    <5.0    <6.0    >7.0
SDT               liquid    >7.0    >6.0    <6.0    <6.0

Co-vulcanization

In the co-vulcanization of two different polymers (A and B), three different combinations are possible (A-A, A-B and B-B); all of which are quite difficult to distinguish experimentally. Most of the polymer blends consist of a heterogeneous system, where the linkage A-B is quite improbable due to the difference of polarity and the lack of miscibility. In this case, the technological co-vulcanization is defined as an ideal co-vulcanization if the properties of the blend are a sum of the properties of the single components (ref. 11). If there is no "ideal" co-vulcanization, then the properties decrease proportionally with increasing mix ratio (figure 2). The degree of co-vulcanization depends mainly tin the cure rate of the individual polymers and the distribution of the accelerators between the two polymers as a consequence of the differences in their solubility in the two polymers (ref. 12).

[ILLUSTRATION OMITTED]

NBR/EPDM blends

NBR and EPDM greatly differ in polarity (ref. 7). This fact complicates the technological compatibility of the polar NBR with the nonpolar EPDM. A 34% ACN content NBR was used in this study. The 34% ACN content NBR was used to maintain good swelling properties even after dilution with the EPDM. A.J. Tinker examined several blend combinations, differentiating them by polarity and saturation grades. In order to determine the crosslink density, he used an FTNMR (ref. 15) analysis. First he analyzed NR/NBR blends containing 0.6 phr TMTM and 1.5 phr sulfur. He discovered that crosslink density increased with decreasing ACN content. Tinker theorized different diffusion rates of the chemicals, including sulfur, during vulcanization from the nonpolar NR phase to the polar NBR phase. In the case of low ACN content, the domain size within these blends is about 1-2 [micro]m. in this case, diffusion plays an important role. Further examination of blends with increasing ACN content showed the domain size to be about 100 [micro]m and very little dependence on diffusion rate. For this case, the crosslink density is influenced by the dispersion of the chemicals during mixing. Based on those results, he examined several accelerator combinations and found that for similar crosslink densities of both polymer phases, tear strength was clearly higher (refs. 13 and 14).

A.Y. Coran also worked on "Blends of dissimilar rubbers - cure rate incompatibility," emphasizing that differences in the cure rate are influenced by the differences in saturation between the polymers in the blend (ref. 15); e.g., in the present work, NBR has approximately 14 times more double bonds relative to the EPDM (ref. 16). Thus, not only is there a difference in polarity, but also a difference in the degree of unsaturation.

NR/EPDM blends

In the case of NR/EPDM blends, the polarity difference is not so great, but the differences in unsaturation strongly influence the crosslink density of the blend (ref. 13). The EPDM rubber used in this article is Buna EP G 5450 that contains 9% ethylidenenorbornene, so that the ratio of double bonds in the NR/EPDM blend is about 10 (ref. 16).

The Rollbiege experiment shows a larger temperature increase for samples with insufficient co-vulcanization compared to those of greater co-vulcanization because it is the internal friction at the boundary surfaces created during bending that causes the heat buildup. The quality of the co-vulcanization can be determined by observing the development of heat. If the transfer of mechanical stress from one polymer domain to another is more effective (it means tan [Delta] [right arrow] 0), then it suggests a better connection between polymer phases. An example is given in figure 3. Good co-vulcanization is indicated when the rubber blend shows two Tgs that are closer together than those of the single polymers. A perfect blend with ideal co-vulcanization would be achieved when only one glass transition temperature can be measured.

[ILLUSTRATION OMITTED]

FT-IR spectroscopy (ref. 17)

Through the use of IR spectroscopy, chemical bonds of a molecule can be excited by infrared absorption. The Lambert-Beer Law gives the dependence between extinction and the frequency of chemical bonds in a molecule (concentration). Thus, it is possible to make a prediction of quantity based on equation 7.

(7) E = [Epsilon] [multiplied by] c [multiplied by] d

E = extinction;

[Epsilon] = logarithmic extinction coefficient;

c = concentration;

d = layer thickness.

In the present work, a variation of FT-IR spectroscopy was used, i.e., the ATR - technology (figure 4). If an IR ray penetrates a medium with a high refractive index, [n.sub.1], oblique to the boundary surface of a medium with a lower optical density, [n.sub.2], and if the critical angle of the total reflection of the angle of incidence is exceeded, then there is a reverse reflection of the IR ray into the optically more dense medium. One part of the electromagnetic energy escapes into the optically thinner medium (1-10 [micro]m). If this portion completely returns into the medium with the higher refraction index, then one has total refraction. If after reflection some wavelengths are missing, then these were absorbed by the optically thinner medium (ref. 18). With this measuring method, the IR ray is reflected about 50 times between the probes in a measuring crystal (here: 45 [degrees] TaJ - crystal). With it the ray penetrates about 10-15 [micro]m and makes possible an analysis of the chemical composition of the surface. The penetration depth and especially the intensity of the absorption are only influenced by the frequency of the IR ray. It is possible to describe this dependence with equation 8 (ref. 19):

[ILLUSTRATION OMITTED]

(8) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

[d.sub.p] [cm] = IR ray penetration depth;

v[[cm.sup.1]] = IR ray frequency;

[differential] [[degrees] = angle of incidence;

[n.sub.1], [n.sub.2] = refraction index of crystal (1) and sample (2).

The sine of the incidence angle on the boundary surface of the measure crystal should be lower than the quotient of the refractive index sample/crystal (figure 4):

(9) sin[Alpha] [is less than] n(sample)/n(crystal)

with sin [Alpha] = angle of incidence;

n = refraction index.

The amount of radiation reduced by absorption is picked up by a detector and transformed into a spectrum. With this methodology it is possible to determine the diffusion behavior of low molecular weight accelerators in several polymers.

Experimental

Materials

The EPDM used in this study was a EPDM G 5450 obtained from Bayer AG. The NBR used in this study was a Perbunan NT 3445 obtained from Bayer AG. The NR used in this study was a commercial grade of SMR CV 50. If not otherwise stated, then all chemicals used in this study were polymer bound dispersions from Rhein Chemie. Carbon blacks, softeners and stearic acid were commercial grade and obtained from various suppliers.

Formulation and mixing

Tables 2 through 5 show the general recipes used in this study for NBR/EPDM and NR/EPDM blends. Compounding was carded out in a laboratory size mixer with an interlocking rotor system. Rotor speed was adjusted to 50 rpm. Mixing was completed within five minutes. Recipes were adjusted to obtain equal mixing volumes of 1,200 [cm.sup.3] times the specific gravity for each compound. Mixing was done in two stages. All ingredients except for the curing agents were mixed in the internal mixer. Curatives were added to the master batch on the mill.

Table 2 - general formulation for NBR/EPDM
blends

Product             Dosage [phr]

Perbunan NT 3445    100.0 - 60.0
Buna EP G 5450        0.0 - 40.0
Carbon black N660           60.0
Dioctylphthalate            10.0
ZnO active                   5.0
Stearic acid                 1.0
Vulkanox BKF                 0.6
Table 3 - general curing systems
(standard/DTP)

Standard-system     [phr]   Dithioate-system    [phr]

Rhenogran CBS-80      1.3   Rhenocure ZDT/G      0.6
Rhenogran TMTM-80     0.3   Rhenocure SDT/S      3.0
Rhenogran S-80        2.2   Rhenogran MBTS-80    1.2
                            Rhenogran S-80       2.2
Table 4 - general formulation for NR/EPDM
blends

Product              Dosage [phr]

SMR CV 50            100.0 - 60.0
Buna EP G 5450         0.0 - 40.0
Carbon black N 330           55.0
Naphthenic oil                6.0
ZnO active                    5.0
Stearic acid                  1.5
Table 5 - general curing systems
(standard/DTP)

Standard-system     [phr]   Dithioate-system    [phr]

Vulkacit CZ          1.0    Rhenocure ZDT/G      3.0
Rhenocure IS 90/G    2.2    Rhenogran MBTS-80    0.8
                            Rhenocure IS 90/G    2.2

Cure characterization was conducted in a rheometer elastograph in accordance with DIN 53529. Samples were compression molded at 170 [degrees] C for 1.5 times t90. Mooney viscosity and Mooney scorch were determined in accordance with DIN 53523. Tensile properties were determined in accordance with DIN 53504. Thickness of the dumbbells was determined by averaging three measurements taken at the neck of the dumbbells. The Rollbiege experiment was performed by turning a sample 1,430 rpm at an angle of 90 [degrees]. Due to this bending and stretching at high speed, the surface temperature is increased as measured by an infrared camera after 5, 10, 15 and 20 minutes.

FT-IR spectroscopy

An FT-IR spectrometer equipped with an interferometer was employed. By using the ATR technique, the diffusion rate of the curing chemicals was determined. A thin elastomeric film was slightly cured with peroxide and placed on top of a rubber sheet having incorporated 10 phr of the curing chemical. Crosslink density analysis was conducted prior to this set of experiments to assure that all samples had the same low crosslink density (ref. 1). IR spectra were recorded in short intervals until the growth of the P-O-C-band at 1,010 cm-1 came to an end. The growth rate was correlated to a constant band (reference). The background correction was acquired in "interleaved" mode. The sample was placed on the ATR crystal of TaJ for the whole experiment. W. Kleemann and K. Weber determined that the diffusion coefficient D of sulfur in NR was 0.57 x [10.sup.-7] [cm.sup.2]/s (ref. 20). The resulting diffusion rate was in the range of approximately 4 [micor]m/h. Experiments at room temperature, 50 [degrees] C and 80 [degrees] C revealed that the experiment at 50 [degrees] C would allow results in a practical measurable time frame. In order to find the correct peak height, a baseline was established and the transmission with and without sample was measured according to equation 10.

(10) E = log [1.sub.0]/1

E = extinction (absorbance);

[1.sub.0] [%] = transmission without substance;

1 [%] = transmission with substance (ref. 19).

The peak height and the extinction coefficient are directly correlated to the product concentration. Thus, by building a correlation line with different amounts of substance, the polymer surface concentration of a chemical C([t.sub.x]) [phr] at a given time is determined. The calibration line of Rhenocure ZDT in EPDM can be seen in figure 5. The film thicknesses of the polymer films were 58 [+ or -] 3 gm for EPDM and 55 [+ or -] 3 [micro]m for NR.

[GRAPH OMITTED]

Results and discussion

NBR/EPDM blends

Table 6 summarizes the compound properties for the NBR/EPDM blends. The Mooney viscosity is hardly affected by the standard curing system, and only slightly reduced with the dithioate cure system. At the same time, Mooney scorch values are prolonged with increasing EPDM content with the dithioate cure system. The same is true for the rheometer cure characteristics for both curing systems. Overall, the dithioate cure system has a slightly higher curing efficiency.

Table 6 - compound properties of the NBR/EPDM blends

Mooney viscosity
    ML(1+4)100
    [degrees] C
Polymer ratio NBR/EPDM   100/0   90/10   80/20   70/30   60/40   0/100
                           NBR                                    EPDM
  Standard system:
    ML1+4 [ME]            36.7    34.5    34.6    35.5    36.0    66.2
  Dithioate system:
    ML1+4 [ME]            34.0    32.7    32.5    32.9    35.2    65.4
Mooney scorch - values
    (T=120 [degrees]
    C)
  Standard system:
    [t.sub.5] [min]      21.53   22.26   22.36   22.10   21.22   22.34
  Standard system:
    [t.sub.35] [min]     23.58   24.25   24.25   23.47   22.55   26.35
  Dithioate system:
    [t.sub.5] [min]      13.22   15.05   15.50   17.00   17.44   27.35
  Dithioate system:
    [t.sub.35] [min]     16.05   18.07   19.27   21.08   22.34   45.38
Rheometer results
    (T=170 [degrees]
    C)
  Standard system:
    [t.sub.10] [min]      1.32    1.36    1.36    1.36    1.32    2.16
  Standard system:
    [t.sub.90] [min]      2.04    1.92    2.12    2.12    2.36    5.88
  Dithioate system:
    [t.sub.10] [min]      0.88    0.96    0.96    1.04    1.08    2.04
  Dithioate system:
    [t.sub.90] [min]      1.40    1.44    1.68    2.16    2.16    5.32
  Standard system:
    Vmax [Nm/min.]         1.5     1.4     1.3     1.2     0.9     0.8
  Dithioate system:
    Vmax [Nm/min.]         1.8     1.7     1.4     1.2     1.0     0.3

Mechanical properties

Tensile properties in figures 6 and 7 show decreasing values with increasing EPDM content. Remarkably, this effect is smaller for the dithioate system. Since the same mixing procedure was applied for all compounds, this effect can not be attributed to smaller and more uniformly dispersed EPDM domains or to an improved filler distribution. Since all dithioates are highly soluble and easily dispersed in several rubbers, it is most likely that these chemicals were more uniformly dispersed in both polymers, resulting in an improved co-vulcanization.

[GRAPHS OMITTED]

This is in line with results of G. Kerruth et. al. of a SBR/EPDM blend (ref. 11). The accelerator system consisted of 0.75 phr sulfur, 0.37 phr MBTS and 0.75 phr TMTD. This study found decreasing physical properties with increasing EPDM content, even after more than 50 phr of EPDM was used. After extracting the vulcanizates, 90 to 95% of the uncured EPDM was recovered. Since this cure system is comparable to the present standard cure system, these results support the theory of improved co-vulcanization with the dithioate cure system.

The Rollbiege experiment

The Rollbiege experiment supports the assumption of improved co-vulcanization with the dithioate cure system, since the temperature increase is at least 10 [degrees] C lower compared to the standard cure system. Additionally, figures 8 and 9 reveal a smaller band with increasing EPDM content, indicating a more uniform vulcanizate.

[GRAPH OMITTED]

NR/EPDM blends

Table 7 summarizes compound properties for the NR/EPDM blends. As expected, the Mooney viscosity is insignificantly affected by the curing system. At the same time. Mooney scorch values are reduced with increasing EPDM content with the dithioate cure system. The same is true for the rheometer cure characteristics for both cure systems. Overall, the dithioate cure system has a slightly higher curing efficiency.

Table 7 - compound properties of the NR/EPDM blends

Mooney viscosiity
    ML(1+4) 100
    [degrees] C
Polymer ratio NR/EPDM   100/0   90/10   80/20   70/30   60/40   0/100
                           NR                                    EPDM
  Standard system:
    ML1+4 [ME]           56.8    65.8    72.4    78.3    86.1   130.7
  Dithioate system
    ML1+4 [ME]           48.4    60.9    68.9    75.4    81.3   125.0
Mooney scorch -
    values (T=120
    [degrees] C)
  Standard system:
    [t.sub.5] [min]     28.06   26.16   24.46   23.49   24.09   27.44
  Standard system:
    [t.sub.35] [min]    30.28   28.49   27.15   26.21   26.32   33.06
  Dithioate system:
    [t.sub.5] [min]     21.39   18.30   17.56   16.36   16.03   24.14
  Dithioate system:
    [t.sub.35] [min]    25.38   22.08   21.41   20.37   20.17   35.32
Rheometer results
    (T=170 [degrees]
    C)
  Standard system:
    [t.sub.10] [min]     0.88    0.80    0.80    0.80    0.80    1.80
  Standard system:
    [t.sub.90] [min]     2.48    2.48    2.44    2.48    2.64   17.70
  Dithioate system:
    [t.sub.10] [min]     0.76    0.68    0.68    0.66    0.68    1.30
  Dithioate system
    [t.sub.90] [min]     2.62    2.56    2.26    2.38    2.40   16.50
  Standard system
    Vmax [Nm/min]         0.8     0.7     0.7     0.7     0.6     0.4
  Dithioate system
    Vmax [Nm/min]         0.9     1.1     1.0     0.9     0.7     0.4

The Rollbiege experiment

When comparing figure 10 with figure 11, a substantially smaller temperature increase is observed with the dithioate cure system. Since both pure polymers have a lower temperature level after 20 minutes, an even dispersion of cure chemicals can be assumed. In contrast, the samples from the NR/EPDM blend system broke apart after approximately 10 minutes when 30 phr or more of EPDM was present. It is hypothesized that this is due to the higher reinforcing carbon black N 330, which restricts polymer chain slippage, more so than carbon black N 660 used in the NBR/ EPDM blend system.

[GRAPHS OMITTED]

Evaluation of diffusion rate of dithioate curing chemicals in rubber blends

The physical data of the blend vulcanizates indicate an improved co-vulcanization. To further elucidate this finding, ATR-FTIR spectroscopy was used to characterize the diffusion behavior of dithioates between pure polymers. Rhenocure ZDT was selected to illustrate this finding. Since there is no adsorption around 1,010 cm-1 for the polymers, diffusion rate can be correlated to an increasing IR band compared to a control IR band.

The diffusion coefficient is determined according to the first and the second Fick's Law:

(11) F = -D [differential]C/[differential]x

The diffusion coefficient depends on the concentration in a given volume. By selecting a change in concentration in only one direction the 2nd Fick's Law is obtained (ref. 21):

(12) [differential]C/[differential]t = D [[differential].sup.2]C/[differential][x.sup.2]

By differentiation according to time and distance one obtains:

(13) D = C([t.sub.2]) - C([t.sub.1])/[Delta]t [multiplied by] /2 [multiplied by] 1/[C.sub.2] - [C.sub.1]

D [[cm.sup.2]/s] = diffusion coefficient;

[C.sub.1] [phr] = chemical concentration in rubber compound;

[C.sub.2] - [C.sub.1] [phr] = difference in concentration between both polymer films (at t = 0 [C.sub.1] is 0, [C.sub.2] is starting concentration in rubber compound);

/ [cm] = film thickness;

C([t.sub.x]) [phr] = concentration at a given time x;

[Delta]t [s] = time of the experiment.

Table 8 summarizes experimental results of the present work and from the literature (refs. 20 and 22). The high solubility of the dithioate in NR corresponds with the slightly higher diffusion rate. Taking into account the experimental results of Gardiner and Kleemann, one can conclude a similar diffusion rate of the above accelerators in other polymers.

Table 8 - diffusion coefficient of various curing
chemicals in selected polymers obtained by
FT-IR spectroscopy

Polymer   Chemical    Diffusion coefficient    Temperature
                         [[cm.sup.2]/s]       [[degrees] C]

NR        Sulfur      0.57 x [10.sup.-7]          60(1)
NR        Dithioate   6.30 x [10.sup.-9]          50
EPDM      Dithioate   3.39 x [10.sup.-9]          50
SBR       MBTS        3.10 x [10.sup.-9]          60(2)

(1) - Experimental results of Kleemann (ref. 20).

(2) - Experimental results of Gardiener (ref. 20)

Acknowledgements

"Improved co-vulcanization of rubber blends" is based on a paper given at the September, 1999 Rubber Division meeting.

"Stabilization of tire compounds with QDI" is based on a paper given at the September, 1999 Rubber Division meeting.

"Improving flex fatigue and dynamic ozone crack resistance through antidegradants" is based on a paper given at the September, 1999 Rubber Division meeting.

References

(1.) Th. Hanel, diploma thesis at the Fachhochschule Wurzburg-Schweinfurt-Aschaffenburg (1996).

(2.) H. Lohwasser, Kautsch. Gummi Kunstst. 44, (2), 172-178 (1991).

(3.) H. Lohwasser, Gummi, Fasern, Kunstst. 50, (6), 473-477 (1997).

(4.) H.F. Struckmeyer and W. Hofmann, Gummi, Fasern, Kunstst. 42, (3), 102 (1989).

[5.] H.-J. Graf and H.-M. Issel, Kautsch. Gummi Kunstst. 48, (9) 600 (1995).

(6.) B. Schneier, J. Appl. Polym. Sci., 17, 3,175 (1973).

(7.) A.J. Tinker, The Malaysian Rubber Producers' Research Association, Tun Abdul Razak Laboratory, Brickendonbury, Hertford SG13 8NL, England, 1 (1993).

(8.) G.H. Weiss and R.J. Rubin. Adv. Chem. Phys., 52, 363 (1983).

(9.) R.P. Mastomateo, J.M. Michell and R.D. Allen, Rubber Age 2, 64 (1970).

(10.) W. Hoffmann, Gummi, Fasern, Kunstst. 40 ,(9), 422 (1986).

(11.) G. Kerruth, H. Blumel and H. Weber, Kautsch. Gummi Kunstst. 22, 413 (1969).

(12.) Frenkel, Rafail S. and Safonov, Alexandr V. Plaste und Kautschuk 39, 265 (1992).

(13.) A.J. Tinker, Rubber Chem. Technol. 68, 461 (1995).

(14.) P.S. Brown, M. Loadman, R. John, A.J and Tinker, Rubber Chem. Techn. 65, Z44 (1992).

(15.) Coran, A.Y., Rubber Chem. Technol. 62, 281 (1989).

(16.) T. Kempermann, Handbuch fur die Gummiindustrie, 2nd edition, Bayer AG Leverkusen, (1991).

(17.) M. Hesse, H. Meier and B. Zeh, "Spektroskopische methoden in der organischen chemie," 3rd edition, Georg Thieme Verlag Stuttgart, 61 (1987).

(18.) Gunzler, H. and Bock, H., IR - Spektroskopie, 2nd edition, Verlag Chemie Weinheim (1983).

(19.) J. Deiters, doctoral thesis at the Universitat of Hannover (1993).

(20.) W. Kleemann and K. Weber, Formeln und Tabellen flit die Elastomerverarbeitung, 2nd edition, Dr. Gupta Verlag, Dusseldorf, 133 (1994).

(21.) J. Crank and G.S. Park, Diffusion in Polymers, 1st edition, Academic Press London, 1 (1968).

(22.) J. Brooke Gardiner, Rubber Chem. Technol. 42, 1,058 (1969).