MICROCOMPOUNDING OF SMALL SAMPLES OF NATURAL RUBBER
ABSTRACT
We describe a microcompounding method that can be used to characterize and compare cured small natural rubber samples. First, commercial Hevea and dried guayule (Parthenium argentatum) latex samples were microcompounded to validate the method. Latex was then extracted from the ground branches of six greenhouse shrubs of different ages and genotypes of wild type and transgenic guayule and coagulated into rubber samples. Size exclusion chromatography showed that the samples had different molecular weights and oligomer content. Little variation in physical properties was found between rubber extracted from shrubs of different age within a genotype, but a larger variation among genotypes was found. The new method allows testing of cured rubber samples from experimental variants and early screening out of genotypes to assess the impact of variations in cultivation practices or lab-scale processing conditions on rubber quality.
INTRODUCTION
Natural rubber (NR) is an outstanding elastomer with properties that no synthetic rubber can replicate at the current time. Hevea brasiliensis trees have long been used as the sole commercial source of NR, but they only grow in specific tropical climates (mostly Southeast Asia) and cannot be grown in the United States.1–3 NR has been sourced from Southeast Asian rubber plantations almost exclusively in modern times because this region has both the proper environment and the available workforce to extract and process NR relatively cheaply. Intensive production of rubber (and oil palm) trees has led to mass deforestation, leaving only clones of a few highly productive rubber trees to dominate the land.4 The planet’s rubber supply is in a precarious position because one major disease outbreak could cripple the supply of NR worldwide.5,6 Diversifying the world’s source of NR with alternative rubber-producing species that can be grown in temperate climates will increase the overall supply of NR and safeguard critical supplies while also protecting our rainforests from further damage.1
The principal alternative rubber crops under development are the guayule shrub (Parthenium argentatum)7–12 and the rubber dandelion [Taraxacum kok-saghyz (TK)].13–18 The main advantage of using these plants as alternative sources of NR is that they can grow in temperate environments; rubber dandelions can even be grown in areas with severe winters.19
Guayule has been identified as an alternative to Hevea rubber since the 1890s, and it was the focus of extensive development as the United States began a search for domestic sources of NR during World War II.3 Guayule can be farmed in the semiarid, southwestern US, reducing our reliance on rubber from Southeast Asia and increasing both the geographical and biological diversity of this critical material. NR from guayule (GNR) lacks the proteins that can induce severe allergic reactions from products using H. brasiliensis NR and latex.20–22 Several attempts to increase NR yield in guayule through transgene expression have been completed with limited success.23,24 While authors of these studies focused on NR yield, the possibility of unintended modification of the NR macromolecular structure or the performance of cured materials has not been investigated, partially due to the limited quantities of guayule NR extracted from transgenic plants. High material performance requires high molecular weight (MW), which can be determined using very small samples, as reported,25–28 but many other factors that contribute to the quality of cured rubber must be assessed. In this paper, we describe a microcompounding method that can be used to compare properties of cured materials made from small amounts of NR from experimental materials, practices, and processes.
EXPERIMENTAL DETAILS
materials
Pale crepe Hevea rubber (Technical Specified Rubber TSR-10, Centrotrade, Chesapeake, Virginia) was selected to be the control NR. Control guayule latex was sourced from EnergyEne and dried at 50 °C. Guayule plants of two ages, approximately 10 months old (“young”) and 3 years old (“old”) were grown in an OSU greenhouse in Wooster, OH. Six different plants were selected for experimentation. AOO denotes a wild-type plant of the triploid interspecific hybrid line AZ101, whereas G1-2 and G1-4 are transgenic versions of the tetraploid wildtype line G7-11 (details will be discussed in a separate paper). Chloroform, acetic acid (ACS reagent, ≥99.7%), both from Sigma-Aldrich, St. Louis, and hexane (VWR, Radnor, PA) were used as received. Tetrahydrofuran (THF) was purchased from Fisher Chemicals. Teflon sheets were purchased from Polymershapes LLC, Cleveland. Chemicals for compounding and curing were a generous gift of HB Chemicals (Twinsburg, OH): N-330 carbon black (reinforcing filler), zinc oxide (accelerator), stearic acid (activator), N-(1,3-dimethylbutyl)-N′-phenyl-p-phenylenediamine (6PPD, antioxidant), sulfur (vulcanizing agent), and N-tert-butyl-2-benzothiazyl sulfonamide (TBBS, accelerator). Certified ACS toluene was obtained from Fisher Chemical, Waltham, MA (Lot #180525) for crosslink density measurements.
METHODS
latex extraction
For each of the six greenhouse samples, the bark was homogenized in a 1:2 w:v shrub:extraction buffer (0.2% potassium hydroxide, 0.1% sodium sulfite, pH 10) in a blender (Model LAR-25LMB, Floor Food Blender w/Metal Container, Tilting, Skyfood Equipment LLC, North Miami, FL) and filtered as previously described.24 The homogenate was readjusted to pH 10 with KOH then poured into 1–4 L glass bottles, sealed, and stored at 4 °C to allow the heavy (dense) phase to settle, which allowed the lighter (less dense) phase to float on top of the mixture. The lighter phase of each homogenate, which contained the guayule latex, was pipetted into a new bottle to concentrate the latex. The new bottles were sealed and stored at 4°C, and the process of allowing the latex to float and then pipetting the latex into new bottles was repeated 3–4 times. pH was checked and adjusted as necessary at each stage. From the final bottles, approximately 45 mL of latex and residual homogenate was poured into 50 mL centrifuge tubes (VWR, Radnor, PA) and centrifuged at 2100 × g for 15 min using a D-37520 Legend T Plus Benchtop Centrifuge (Thermo Electron Corporation, Waltham, MA). Next, 1 mL of acetic acid (ACS reagent, ≥99.7%, Sigma-Aldrich, St. Louis) was added to the surface of the latex layer at the top of each tube and the samples centrifuged an additional 15 min at 2100 × g to coagulate the latex into solid rubber. Coagulated rubber was removed using forceps from the top of the remaining liquid and placed in a beaker. Here, 50 mL of chloroform was added to the material to aid in extraction of the entrained water. After decanting excess liquid, the rubber was allowed to dry in a fume hood until stable weight was achieved. The efficiency of the rubber production method was calculated based on latex quantification of the amount of rubber in the original shrub sample, as previously described.29
SIZE EXCLUSION CHROMATOGRAPHY
Size exclusion chromatography (SEC) measurements were performed using a system consisting of an Agilent 1260 infinity isocratic pump, a Wyatt Eclipse DUALTEC separation system, an Agilent 1260 infinity variable wavelength detector (UV), a Wyatt OPTILAB T-rEX interferometric refractometer, a Wyatt DAWN HELOS-II multiangle static light scattering detector (MALS) with a built-in dynamic light scattering module, a Wyatt ViscoStar-II viscometer, an Agilent 1260 infinity standard autosampler, and 6 StyragelVR columns (HR6, HR5, HR4, HR3, HR1, and H0.5). The columns were equilibrated at 35 °C, and THF, continuously distilled from CaH2, was used as the mobile phase at a flow rate of 1 mL/min. Samples were first dissolved in THF from the distillation, then 1 mL was filtered into the SEC vials using a 0.45 µm PTFE syringe filter. In every case, 100 µL was injected into the SEC. The results were analyzed using the ASTRA 7 software (Wyatt Technology). Absolute MWs were obtained using dn/dc = 0.130 mL/g for polyisoprene.30 A polystyrene standard with Mw = 30 000 g/mol (Đ = 1.05, dn/dc = 0.185) was used with each batch of measurements for quality assurance.
RUBBER MICROCOMPOUNDING
Table I shows the rubber compounding recipe. Neither an internal mixer, such as a lab Banbury, nor a two-roll rubber mill could be used to mix the compounds due to the small quantity of guayule NR. To overcome this problem, an alternative process was developed based on one created earlier for polyisobutylene-based rubbers.31
Here, 5.466 g of dried rubber was cut into small pieces (approximately 1 cm × 1 cm) and was placed in an amber glass jar with a stir bar, to which 100 mL of hexane was added. The jar was closed, placed on a stirrer plate at 250 rpm, and stirred overnight to dissolve the rubber. A thick solution resulted with a viscosity similar to honey. Next, while the solution continued to be stirred, carbon black, zinc oxide, stearic acid, and 6PPD antioxidant were slowly added over 10 min, and the final mixture stirred for an additional 2 h. Next, this “masterbatch” was coagulated. A 600 mL beaker was filled with approximately 500 mL of distilled water, placed on a hot plate with the temperature control dial set at 250°F, and the water slowly heated to a gentle simmer (100 °C). A 10 mL glass pipette was used to withdraw and drizzle approximately 5 mL of the continuously stirred rubber masterbatch solution on top of the simmering water. A thin layer of solid rubber was created, which was allowed to heat for approximately 1–2 min, as small bubbles were observed on the surface as the hexane evaporated. Using a lab spoon, the rubber layer was carefully skimmed off the top of the water and inside walls of the beaker and placed in a Petri dish. This procedure was repeated until the complete masterbatch was coagulated. The rubber in the Petri dish was left in a fume hood overnight to evaporate any residual hexane. This method was repeated for the control guayule and Hevea NR and the transgenic guayule samples.
COMPOUND MIXING
A Carver Bench Top Standard Heated Press (Model 4386 Laboratory Manual Press with Electrically Heated Platens, Wabash, IN) was heated to 54.4 °C. Each masterbatch sample was mixed to homogeneity by placing the sample between Teflon sheets and pressing under 2 t of compressive force. The flattened rubber sample was removed from the press, then folded 3 or 4 times before being placed back between the Teflon sheets and repressed. This process was repeated 12 times. A thin sheet, approximately 2 mm thick, was produced from all samples. Then the press temperature was raised to 110 °C, and the sheets were pressed under 5.5 t for 5 min to remove any remaining water or hexane.
Dried sheets were then placed back on the lower platen of the press at 54.4 °C. TBBS accelerator (1/3 of the total) was carefully added to the center of the sheet, the sheet was folded four times to entrain the TBBS powder, then the sample was pressed under 5.5 t for several minutes until the force began to drop. This process was repeated until all the TBBS was added. Then the sulfur was added 1/3 at a time following the same procedure (the sulfur was added last to prevent premature vulcanization). The rubber sheet was pressed an additional six times under 5.5 t, folding four times between presses to thoroughly disperse the compounding chemicals. The rubber compound was placed into a stainless-steel mold with inside dimensions of 60 mm × 60 mm × 1.85 mm and cured at 160 °C under 5.5 t pressure for 15 min, resulting in a cured sheet of rubber, based on preliminary cure tests demonstrating T90 ∼ 12 min. The mold was sandwiched between Teflon sheets to ensure that the cured compound would not stick to the stainless-steel platens of the press, forming a uniform rubber sheet.
TENSILE TESTING
Microdumbbells (50 mm in length) were cut for testing according to DIN 53504, using test piece type S3A (Figure 1). The guayule and Hevea controls were tested using a Zwick® Z005 Universal Testing machine with a 5 kN load cell at a strain rate of 500 mm/min. The transgenic samples were tested using an Instron® instrument (Norwood, MA) with a 50 N load cell at a strain rate of 200 mm/min.
Citation: Rubber Chemistry and Technology 98, 1; 10.5254/rct.24.00033
CROSSLINK DENSITY MEASUREMENTS
Here, 40 mL toluene was poured into separate 100 mL beakers. Then 1.000 g rubber samples, weighed to 1 mg precision, were immersed into the toluene. The beakers were then covered and allowed to sit for 1 d. After 24 h, the toluene was decanted, the rubber samples were weighed, then put back into fresh toluene. This process was repeated four times to ensure equilibrium swelling. After the final decanting, the rubber samples were weighed to determine their swelled mass and placed into a vacuum oven until they reached constant weight. The rubber density, ρrubber, was 0.92 from the literature.32 The Flory-Huggins NR-toluene interaction parameter χ was 0.391 according to the ASTM D 6814-02 standard, while ρsolvent was 0.867 g/cm3 at room temperature, and f is the weight fraction of nonrubber components (filler, curing packages, processing aids, and antioxidants).
Crosslink density of the compounded elastomers was calculated from swelling data using the Flory-Rehner (Eq. 1) and modified Kraus (Eq. 2 and 3) equations33:
where vcross is the crosslink density, vr is the volume fraction of the rubber in the swollen sample, χ is the Flory-Huggins interaction parameter (0.391), ρ is the density of the rubber (0.92 g/cm3), Vs is the molar volume of toluene (106.27 cm3/mol), and Vrubber, Vtoluene, and Vfiller are the volumes of the rubber, toluene, and carbon black filler, respectively.
STATISTICAL ANALYSIS
JMP software (JMP Statistical Discovery LLC, Cary, NC) was used to conduct the statistical analysis of data. A one-way analysis of variance was performed for all trials to determine if any significant differences in the mean values for the samples existed. If a dataset was found to be significant at a level of 0.05, an additional Tukey test was carried out to determine which data were significantly different from others. Three separate analyses were carried out for tensile strength at break, modulus at 100% elongation, and elongation at break.
RESULTS AND DISCUSSION
control sec analysis
The SEC traces of the Hevea (HC) and guayule (GC) controls are shown in Figure 2. The MALS (red trace in Figure 2) measures absolute MWs. The blue trace is the output from the differential refractive index (dRI) detector that measures sample concentration. The dRI traces for both samples show a high MW population and low MW oligomers. The Hevea sample has two populations in the high MW fraction, like that reported previously.25,26Table II summarizes the results. The high MW shoulder (I, about 20%) had a peak value of over 1 million g/mol (1512.9 kg/mol), whereas the main fraction (II, about 70%) had a peak value of 223.1 kg/mol. The oligomer content was about 5%. Guayule, on the other hand, had a monomodal high MW trace with Mw = 1226.5 kg/mol and ∼16% oligomer content.
Citation: Rubber Chemistry and Technology 98, 1; 10.5254/rct.24.00033
CONTROL TENSILE TESTING
The microcompounding method created for polyisobutylene-based thermoplastic elastomers generated mechanical performance results comparable with larger samples.31 The same method was adapted to NR. The tensile data are shown in Table III, and representative plots are shown in Figure 3. Stress hardening is seen, but no rubbery plateau was observed.
The tensile plot shows the characteristic strain hardening of NR due to strain-induced crystallization.34–37 The tensile data agree well with literature reports.38,39
Citation: Rubber Chemistry and Technology 98, 1; 10.5254/rct.24.00033
CONTROL CROSSLINK DENSITY MEASUREMENTS
The crosslink density of the microcompounds obtained from swelling experiments of the two control samples can be found in Table IV. Crosslink densities were calculated without and with considering the contribution of the filler (Flory-Rehner equation, and Kraus modification, see Experimental). Based on the difference between these values, the filler contribution was determined to be 33%.
ANALYSIS OF GREENHOUSE GUAYULE
Rubber Extraction. —
The extraction efficiency (Table V) was ∼50% from young plants. AOO had 45% efficiency from old plants, but G1-2 and G1-4 had only 25 and 12% from old plants.
Transgenic Sec Analysis. —
Every sample measured had one large peak and multiple smaller peaks in the oligomer range. The SEC plots are shown in Figure 4. Table VI lists the Mw (weight-average MW) and Ð (dispersity) values for the high MW peaks and the oligomer fraction percentage of the sample. Authors usually neglect to show the oligomer range in SEC measurements, but they can be substantial as in the case here and published earlier.25,26 AOO and G1-4 young samples had similar Mw, while G1-2 old had a lower Mw. In general, young plants produced rubber with higher Mw than older plants, which is counter to previous reports of wild type plants.7
Citation: Rubber Chemistry and Technology 98, 1; 10.5254/rct.24.00033
The oligomer content of the transgenic guayule samples (21–28%) is higher than that of the control sample (16%, Table II).
Transgenic Tensile Testing. —
Figure 5 and Table VII show tensile data. The GNR microcompounded samples had higher 100% modulus and lower tensile strength and elongation at break than the controls in this work and those reported for similar GNR compounds prepared with conventional mixing and milling equipment (Table VII).40 This can be attributed to the lower Mw of the greenhouse samples.
Citation: Rubber Chemistry and Technology 98, 1; 10.5254/rct.24.00033
Transgenic Crosslink Density Measurements. —
The crosslinking densities were significantly higher than those of the control samples (Table VIII). This might be due to the lower Mw of the greenhouse samples, so the same curing conditions led to higher crosslink densities. This explains the higher 100% moduli and lower tensile strength and elongation measured for the greenhouse samples.
Statistical Analysis. —
While the number of transgenic samples was small and the main goal of this work was to demonstrate the application of microcompounding for small NR samples, some comparisons can be made from the data in Table VII and Figure 6.
Citation: Rubber Chemistry and Technology 98, 1; 10.5254/rct.24.00033
The 100% moduli values were not significantly different among the samples. The age of the plants did not significantly affect the tensile strength at break, modulus at 100% elongation, or elongation at break with all p values > 0.05, even though older plants yielded NR with somewhat lower MW. However, both old and young G1-2 plants had significantly lower tensile strength and elongation at break than the other genotypes.
CONCLUSIONS
The microcompounding procedure described in this paper can be used for small, cured NR samples. The control samples confirm that the method is sound and gives results comparable with macrocompounded samples. SEC analysis revealed that samples from younger plants had higher MWs than those extracted from older plants. The results of the tensile tests showed that transgenic samples had higher 100% modulus than the control specimens. In comparison, the tensile strength and elongation at break values of transgenic samples were measured to be lower, which is likely due to the lower Mw and higher crosslink densities of the greenhouse samples. Being able to test small, cured samples from experimental plant variants will assist researchers in assessing the impact of variations in new germplasm, cultivation practices, or lab-scale processing conditions on rubber performance.
Test dumbbell.
SEC MALS (red) and dRI (blue) traces of (a) guayule and (b) Hevea brasiliensis.
Stress-strain plots for guayule (solid) and Hevea (dashed) controls.
SEC MALS (red) and dRI (blue) traces of the rubber samples. (a) AOO young, (b) AOO old, (c) G1-2 young, (d) G1-2 old, (e) G1-4 young, and (f) G1-4 old.
Stress-strain plots of all transgenic samples.
(a) Tensile strength at break, (b) modulus at 100% elongation, and (c) elongation at break of samples from three different genotypes and two ages. Solid columns are the old, and striped columns are the young plants. The number of replicates is listed in Table III. Standard error bars are plotted.
Contributor Notes