Using powdered factory scraps in EPDM compounds. (Tech Service).

Waste rubber recycling continues to be of economic as well as ecological interest. Utilization of ground rubber vulcanizate in virgin polymer matrices is an environmentally friendly technique, which if made cost-effective, call solve the global problem of waste disposal to a great extent. The cost-effectiveness of this approach mainly depends upon the method of grinding the waste rubber. Also, the size and topography of the particles depend upon the grinding technique, and these characteristics play a vital role in controlling the composite properties (refs. 1-4).

It has been reported by Jacob et al, that EPDM powdered vulcanizate prepared by an abrasion technique can be used as a low reinforcing filler ill EPDM and SBR matrices (refs. 5-7). In an earlier communication, the authors have reported that abraded NR vulcanizate powder can also be used as a filler at a loading of 50 phr in NR gum compound without deleterious effects on the properties (ref 8). This present article describes the studies on utilization of abraded factory scraps, based on EPDM window seals, in gum EPDM compound.

Experimental

Materials used

The EPDM rubber used has the following characteristics: An ethylene/propylene ratio of 52/48, the third monomer ethylidene norbornene (ENB) content is 4.9%; Mooney viscosity ML(1+4) at 100[degrees]C is 45 and at 125[degrees]C is 29. Other ingredients were rubber grade chemicals procured from the local market. Factory waste (EPDM vulcanizate) was the scrap obtained from Pulsar Rubber Manufacturing, Kolkata, India.

Thermogravimetric analyses of the factory scraps

For determining the approximate composition of the factory scraps, TGA analyses were carried out by using a thermogravimetric analyzer in a temperature range of 30-1,000[degrees]C and at a heating rate of 20[degrees]C. The percentage of polymer, filler and oil were calculated from the thermograms.

Preparation of powdered rubber

The window seal scraps of different hardness (60, 70 and 80 durometer A) were abraded against a silicon carbide abrasive wheel of a mechanical grinder, rotating at 2,950 rpm. The powders were collected in a holder kept beneath the wheel. [P.sub.60], [P.sub.70] and [P.sub.80] refer to the powders of the precursor scrap of hardness 60, 70 and 80 durometer A, respectively.

Characterization of the powdered rubber

The particle size and shape were determined by scanning electron microscope (SEM) using gold-coated samples spread over the SEM mount.

The particle size distribution was studied by using a light optical microscope. The particles were suspended in hexane, subjected to ultrasonic dispersion and examined under a light optical microscope at a magnification of 200x. Images of representative areas were transmitted to an on-line automated image analysis system (IAS). The individual particles were identified and their respective sizes measured. The data were subsequently transferred to a spreadsheet and respective histograms obtained.

XPS analyses were done for studying the surface characteristics of the powdered rubber. This system has a monochromatic Al K-[alpha] x-ray source (1,486.6 eV) that can be focused to a spot size of 0.6 mm x 0.3 mm. Operating pressure was 2 x [10.sup.-9] mbar. Experiments were done for the powdered samples as well as the original vulcanizates before grinding. All spectra were referenced to the C1s peak for carbon, which was assigned a value of 285.0 eV. Peaks in the spectra were deconvoluted using a standard Guassian multiple peak analyzer program, and the concentrations of the elements were calculated from their peak areas and standard photo-ionization cross-section values.

The specific gravity of the powders was determined as per ASTM D1817-97 by the pyknometer method using acetone as the non-interacting liquid media.

The percentage free sulfur in the powder was determined as per ASTM D297-93 by the quantitative conversion of free sulfur to thiosulfate and its estimation with standard iodine solution.

The presence of unreacted curatives in the powder was detected from the curing curve of the revulcanization of the powdered rubber using a rheometer. The powder was first made into the sheet form by passing through a two-roll mixing mill, and then the rheographs were taken at 150[degrees]C.

Mixing of rubber powder in EPDM rubber compound

Formulations are given in table 1. [H.sub.60] contains 100 phr of the powder prepared from the window seal scrap of hardness 60 durometer A, while [H.sub.70] contains 100 phr of the powder prepared from the window seal scrap of 70 durometer A and [H.sub.80] contains 100 phr of the powder prepared from the scrap of 80 durometer A hardness. Incorporation of rubber vulcanizate powder in gum EPDM rubber matrix was done on a two-roll rubber mixing mill at room temperature as per ASTM D3182.

Determination of Mooney viscosity and scorch time

Mooney viscosity of the rubber compounds, ML(1+4) 120[degrees]C, was determined by using a Mooney viscometer as per ASTM D1646-96a. The time for five units rise from the minimum value was taken as the Mooney scorch time.

Determination of cure characteristics

Cure characteristics of the rubber compounds were determined by using a rheometer at 160[degrees]C. The minimum and maximum rheometric torques ([M.sub.L] and [M.sub.H], respectively) and optimum cure time ([t.sub.90]) were recorded automatically from the cure curve.

Molding of test specimens

Molding of the rubber compounds was done in a hydraulic press as per the optimum cure time obtained at 160[degrees]C. Samples used for abrasion were molded in standard molds, giving additional curing time for the extra thickness of the sample (i.e., five minutes for each increment of 6 mm).

Determination of mechanical properties

Dumbbell shaped specimens for tensile testing were punched out from the molded sheets with specific dies (C type) along the grain direction, while un-nicked 90[degrees] angle specimens for tear testing were cut across the grain direction. Tensile testing was done as per ASTM D412-98a at a crosshead speed of 500 mm/min. Tear strength of the vulcanizates was determined on angle specimens (die C) as per ASTM D624-98 at room temperature. Tension set at 100% elongation was determined on dumbbell specimens as per ASTM D412-98a. Hardness of the samples was determined by using a durometer as per ASTM D2240-97. Hysteresis was determined on dumbbells by elongating up to 100% at a crosshead speed of 500 mm/min. The area of the stress-strain loop was recorded for the extension-retraction cycle. Abrasion loss was determined and expressed as abrasion loss (ASTM D394-59), which is the volume in [cm.sup.3] abraded from a test sample per hour.

Solvent swelling studies of rubber vulcanizates

Samples of approximately 1 cm diameter and 0.25 cm thickness were allowed to swell in the solvent (n-heptane for EPDM compounds; toluene for SBR). Swollen samples taken out after specific intervals were blotted with filter paper and weighed quickly. The swelling index, defined as the grams of solvent per gram of rubber hydrocarbon, was calculated as (ref. 9):

(1) Swelling index (Q) = Swollen weight - dried weight/ Initial weight x [Rubber hydrocarbon weight/Formulation weight]

The 1/Q value for the vulcanizates was taken as the apparent crosslink density. For the rubber powder filled compounds, the rubber hydrocarbon (RH) weight includes RH from rubber powder also.

Sample preparation for AFM studies

Cryomicrotoming of the specimens for AFM studies was done using glass knives, after freezing the specimens below their glass transition temperature using liquid nitrogen. Average sample thickness was 20 [micro]m.

AFM measurements

The AFM measurements were carried out in air at ambient conditions (25[degrees]C) using an atomic force microscope. Height and phase images were recorded simultaneously in the tapping mode. Height images were taken in a data scale of 3 [micro]m and phase images were recorded in a phase angle of 60[degrees]. The scanning was done using an etched silicon tip (TESP probe), square pyramid in shape, with a base dimension of approximately 4 [micro]m x 4 [micro]m. The characteristics of the probes were: spring constant (K), 20-100 N/m; resonance frequency, 200-400 kHz; nominal tip radius of curvature, 5-10 nm; cantilever length, 125 [micro]m; cantilever configuration, single beam; tip half angle, 18[degrees] side, 25[degrees] front and 10[degrees] back. The set point ratio of the cantilever, which governs the tapping force, was between 0.8 and 0.9 for all the scans. All the images were recorded using a free oscillation amplitude of 140 [+ or -]10 nm. The scan area was 30 [micro]m x 30 [micro]m, and all the images contained 256 data points.

The images obtained after scanning the surfaces were analyzed. The bows in the raw AFM data were removed using the operation called flattening, which would eliminate the unwanted features from the images by calculating a second order least square fit of the selected segment, then subtracting it from the scan fine. For each sample, a minimum of five images were analyzed.

While doing the roughness analysis, the flattened images were subjected to a second order plane fit in order to remove the tilt and distortions in the images. A comparison of the surface roughness can be obtained from the values of Rq (i.e., the root mean square average of the height deviations taken from the mean data plane), [R.sub.max] (i.e., the maximum vertical distance between the highest and the lowest data points on the surface) and the surface area.

Results and discussion

Composition of factory scrap

Table 2 shows the results of thermogravimetric analyses of the factory scrap in air and nitrogen atmospheres.

Characterization of powdered rubber

The SEM photomicrographs of the particles show that the particles exist in chain-like aggregated structures (figure 1a) as it occurs in the case of wear debris in conventional abrasion processes (refs. 10 and 11). The finer the particle, the greater is its tendency to aggregate. Therefore, it is difficult to conduct the sieve analysis as is normally done in the case of cryoground rubber. A typical SEM photomicrograph of a single particle is given in figure 1(b). The surface of the particles is rough and convoluted as is normally observed in the case of conventional ambient grinding (ref. 12). The shape of the particles is irregular.

[FIGURE 1 OMITTED]

The particle size as determined by the light optical microscopic method, and the size distribution for the representative powdered rubber sample is given in figure 2. Average particle size is ~9.8 microns.

[FIGURE 2 OMITTED]

The XPS analyses of a representative factory scrap (60 durometer A, before grinding) and the corresponding powder ([P.sub.60]) reveal a mild oxidation of the surface of the particles. Deconvoluted peaks for C1s and O1s in the factory scrap vulcanizate mad the powder are given in figure 3. Two types of carbon atoms corresponding to the binding energies of 285.0 eV (84.5%) and 286.4 eV (6.0%) are found in the powder ([P.sub.60]), indicating the attachment of carbon to oxygen as in C=O and C-O-C groups. The oxygen binding energy of 533.5 eV in the powder also corresponds to that of C=O group (1.25%) which is absent in the spectrum of precursor vulcanizate.

[FIGURE 3 OMITTED]

The specific gravity of the powdered rubbers falls in the range of 1.03-1.18 due to the wide particle size distribution, the smaller particles being heavier due to higher carbon black content. Earlier, Naskar et al (ref. 13), while characterizing GRT, reported that smaller particles contain more carbon black and less polymer.

Free sulfur content of [P.sub.60], [P.sub.70] and [P.sub.80] are 0.098, 0.091 and 0.082% respectively.

The rheograph for the recuring of [P.sub.60] is given in figure 4. The increase in [DELTA]torque indicates the presence of unused curatives in the powdered rubber which leads to a second step curing of the particles. Similar results have been reported earlier in the case of cryoground tire rubber powder (refs. 14 and 15).

[FIGURE 4 OMITTED]

Mooney viscosity and curing characteristics of powdered rubber-filled rubber compounds

The Mooney viscosity of the compounds show a slight increase with an increase in hardness of the precursor window seal scrap (table 3), due to the increasing filler content. Scorch time is affected by the curative migration from the powdered rubber to the matrix rubber as has been reported earlier (refs. 16 and 17). The minimum rheometric torque also shows a slight increase with all increase in hardness of the precursor scrap, as is observed in the case of Mooney viscosity. In general, the effect of rubber powder on the curing characteristics of the EPDM compound is not significant.

Mechanical properties

The mechanical properties of the vulcanizates are given in table 4. Mechanical properties of the vulcanizates indicate the reinforcing nature of the powdered rubber. It is observed that as the hardness of the precursor scrap increases, the mechanical properties like tensile strength, modulus, tear strength and abrasion resistance of the vulcanizates improve, which can be ascribed to the reinforcement due to the increasing filler content in the powder. Although the 1/Q value (a measure of the crosslink density of the vulcanizate) decreases by the addition of the powdered rubber, the effect of variation of hardness of the precursor scrap is negligible.

AFM studies

The representative AFM phase images of the gum vulcanizate ([H.sub.0]) and the powdered rubber-filled vulcanizate ([H.sub.60]) are given in figure 5(a) and (b), respectively. The bright areas observed in the case of [H.sub.0] (figure 5(a)) can be ascribed to the crystals of unreacted compounding ingredients (ZnO, sulfur or accelerator). Similar observations were made in the scanning electron microscopic fractography of the gum vulcanizate (ref. 5).

[FIGURE 5 OMITTED]

Additional features in figure 5(b), as compared to figure 5(a), are due to the powdered rubber particles. It is evident that the aggregates of the powdered rubber break down under shear during mixing with the raw rubber, leading to a homogeneous dispersion. Similarly, comparison of the surface plots of the AFM height images (that is, figure 5(c) compared with 5(d)) shows an increased number of projections corresponding to the dispersed rubber particles.

Results of sectional analysis of the gum vulcanizate and the powdered rubber-filled vulcanizates are given in figures 6(a) and (b), respectively. The AFM surface roughness values, as summarized in table 5, indicate an increase in the surface roughness due to the incorporation of the powdered rubber into the gum rubber matrix.

[FIGURE 6 OMITTED]

Conclusions

* Powdered rubber obtained by the abrasion technique is highly aggregated, but the particles are ultra-fine with a rough, convoluted surface.

* Powdered EPDM rubber shows a reinforcing effect in the gum EPDM matrix and mechanical properties of the powdered rubber-filled vulcanizates improve with increasing filler content in the precursor scrap.

* AFM studies of the powdered rubber-filled vulcanizates reveal the breakdown of aggregates of powered rubber and a homogeneous dispersion of the particles in the rubber matrix.

Table 1--formulations used for the study

Ingredients                             Mix designation

                                         H   [H.sub.60]/
                                             [H.sub.70]/
                                              [H.sub.80]

R-EPDM                                 100           100
[P.sub.60]/[P.sub.70]/[P.sub.80] (a)    --   100/100/100
ZnO                                      5             5
Stearic acid                             1             1
TMTD                                     1             1
MBT                                    0.5           0.5
Sulfur                                 1.5           1.5

(a) Rubber powder prepared from scraps of durometer hardness
of 60A, 70A and 80A

Table 2--results of the factory of thermogravimetric anlysis
of the factory scraps of different hardness

Designation of the    Polymer     Filler        Oil
  rubber powder      content %   content %   content %

    [P.sub.60]          37          46          17
    [P.sub.70]          44          48           8
    [P.sub.80]          36          54          10

Table 3--Mooney scorch time (at 120[degrees]C) and
curing characteristics (at 150[degrees]C) of the
compounds

Property                   Mix designation

                           H (a)   [H.sub.60]   [H.sub.70]   [H.sub.80]

ML(1+4) 120[degrees]C         35           52           54           66
Mooney scorch
  time, min.                  33            8           10            7
[M.sub.L], dN.m             0.35         0.82         1.01         1.31
[M.sub.H],dN.m              9.39         7.95         8.96        10.10
[M.sub.H]-[M.sub.L],dN.m    9.04         7.13         7.95         8.79
Optimum cure
  time (t90), min.           7.1          8.0          8.3          9.5

(a) Without rubber powder

Table 4--mechanical properties of the
vulcanizates

Property                                 Mix designation

                             H   [H.sub.60]   [H.sub.70]   [H.sub.80]

Tensile strength, MPa      1.5         5.71         6.63         7.04
Elongation at break, %     157          410          405          345
Modulus at 100%
  elongation, MPa         1.17         1.35         1.53         1.89
Modulus at 300%
  elongation, MPa           --         3.86         4.54         5.92
Tear strength, kN/m        8.9         18.8         23.0         29.8
Hardness, durometer A       45           47           50           53
Tension set (at 100%
  elongation), %           (b)            2            2            2
Hysteresis loss,
  (x[10.sup.-6]),
  J/[m.sup.2]            0.003        0.007        0.010        0.015
Abrasion loss, cc/hr      0.35         0.95         0.86         0.78
1/Q                       0.44         0.36         0.34         0.35

(b)-Sample broke

Table 5--results of AFM roughness analysis
Roughness parameters Mix designation

Roughness parameters               Mix designation

                                    H   [H.sub.60]

[R.sub.q] (nm)                     44           96
[R.sub.max] (nm)                  759          934
Surface area ([micro][m.sup.2])   554          700

References

(1.) S.K. De, Prog. Rub. Plast. Technol. 17(2), 113 (2001).

(2.) B. Adhikari, D. De and S. Maiti, Prog. Polym. Sci. 25, 909 (2000).

(3.) F.G. Smith, "Rubber recycling--ground rubber particulate process and materials of the 1990s," presented at a meeting of the Rubber Division, ACS, Oct. 26-29 (1993).

(4.) J. Scheirs, in Polymer Recycling, Chapter 12, John Wiley and Sons, Chichester, England (1998).

(5.) Ceni Jacob, Bhowmick, A.K., De, P.P. and De, S.K., J. Appl. Polym. Sci. 82, 3,293, (2001).

(6.) Ceni, Jacob, Bhowmick, A.K., De, P.P. and De, S.K., Plast. Rubber Comp. 31(5), 212 (2002).

(7.) Ceni, Jacob, De, P.P. and De, S.K., Kautschuk Gummi Kunstst (communicated).

(8.) Ceni, Jacob, De, P.P. and De, S.K., Rubber World 227(2), 43, 2002.

(9.) C.R. Parks and R.J. Brown, Rubber Chem. Technol. 49, 233 (1976).

(10.) A.N. Gent and C.T.R. Pulford, J. Appl. Polym. Sci. 28, 943 (1983).

(11.) J. Padovan, N. Prasad, D. Gerrard, S.W Park and N. Lindsley, Rubber Chem. Technol. 72, 343 (1999).

(12.) R.P. Burford and M. Pittolo, Rubber Chem. Technol. 55, 1,233 (1982).

(13.) A.K. Naskar, P.K. Pramanik, R. Mukhopadyay, S.K. De and A.K. Bhowmick, Rubber Chem. Technol. 73, 902 (2000).

(14.) A.A. Phadke, A.K. Bhattacharya, S.K. Chakraborty and S.K. De, Rubber Chem. Technol. 56, 726 (1983).

(15.) A.A. Phadke and S.K. De, Kautschuk Gummi Kunst. 37, 776 (1984).

(16.) D. Gibala and G.R. Hamed, Rubber Chem. Technol. 67, 636 (1994).

(17.) E. Morita, Rubber Chem. Technol. 53, 393 (1980).