Dispersion of polyethylene imine-coated chopped fibers in rubber matrix.
Recently, many research activities have been focused primarily on developing alternative treatments that are formaldehyde-free systems that deliver acceptable coatings on fibers. For practitioners to readily accept such alternative technology, the goal must be to use the same or similar dipping equipment used in RFL dip coating programs. Louis et al. discussed formaldehyde-free dip technology introduced by Kordsa Group for reinforcing textile materials (ref. 6). This technology consists of epoxy, polyisocyanate and latex, which were proportionately blended into a dipping solution (ref. 7). The objective was to replace the traditional RFL dipping chemistry and still be able to use the traditional dipping equipment. Gomes et al. evaluated RFL-free coatings of tire textiles which they compared to an RFL system for various fiber materials; the RFL-free coating consisted of Ricobond 7004 from Cray Valley and other ingredients (ref. 8). Ricobond 7004 is a dispersion of a functionalized polymer. The results were reported to be comparable to the RFL system, but with lower peel forces. Bridgestone (ref. 9), as part of its broad and comprehensive Eco-Activities program, is involved in the development of RFL-free coating systems. Mehler Engineered Products has recently announced an RFL-free dip coating for various fibers (ref. 10).
Different types of fibers, coated with RFL, are generally used to reinforce technical rubber applications. In this instance, technical rubber implies all rubber applications other than tire. The objectives of this study were to: 1) evaluate polyethyleneimine-coated chopped fibers and compare to RFL-coated chopped fibers, and 2) determine if polyethyleneimine can be used as a safer and environmentally-friendly coating replacement for RFL for chopped fibers in technical rubber applications.
An EPDM v-belt recipe was used to evaluate the effects of RFL-coated and polyethyleneimine-coated chopped fibers. The physical properties of uncured and vulcanized samples were compared.
The physical properties evaluated were Mooney viscosity, MDR, durometer, tensile, elongation, modulus, die C tear, trouser tear, compression set, DeMattia flex crack growth, rotary drum abrasion, DMA strain sweep and fiber dispersions in the rubber matrix. Results show that the polyethyleneimine-coated fibers and RFL-coated fibers yield similar physical properties, but the polyethyleneimine coating seems to give slightly better dispersion of the chopped fibers in the rubber matrix.
Fiber preparation and coating
Polyethyleneimine from BASF Corporation, sold under the tradename "Lupasol," was formulated into an RFL-free formulation and used for coating the fibers. The general structure is shown in figure 1. Lupasol types are cationic molecules whose charge density depends on pH. The molecules consist of branched polymer structure with various degrees of branching. The molecular weights range from 800 g/mole to 2,000,000 g/ mole. Lupasol products are multifunctional polyethyleneimine with the typical branched polymer structure shown in figure 1.
This structure can be represented by the following formula:
- [(C[H.sub.2] - C[H.sub.2] - NH).sub.n] - where 10 < n < [10.sup.5]
These products have the largest amino group density of any commercial polyamine with a nitrogen to carbon ratio of approximately 2:1.
The polyester fibers used in this evaluation were 2,000 denier with 492 filaments. Three ends of the untreated polyester yams were twisted into one construction (2,000/1/3) and subsequently wound into a spool; two spools of the twisted fibers were prepared. The coating was done with a traditional cord dipping system. The polyethyleneimine-coated and RFL-coated fibers were chopped into three millimeter lengths. For this study, two spools of twisted fibers were treated; one with 5% Lupasol WF and 0.5% surfactant in aqueous solution, and the other spool was treated with a 20% concentration of RFL. The characteristic colors of both RFL and Lupasol coatings are distinctly exhibited in figures 2 and 3.
Table 1 shows the physical properties of Lupasol WF used in formulating the coating solution for the twisted fibers. A suitable surfactant was optionally included in the composition to lower the surface tension and improve wetting of the coating solution. The coating of tine fibers was accomplished in a typical dip coating unit with both the RFL and polyethyleneimine.
A model EPDM v-belt recipe was used to compare the effects of RFL and Lupasol coatings on chopped polyester fibers. Table 2 provides an overview of the key chemical ingredients used in the recipe.
The elastomer used in this study is Royalene 580-HT, an EPDM with Mooney viscosity of 60 (ML [1+4] @ 100[degrees]C) with a 53/47 ratio of ethylene to propy-lene and 2.7% ENB content, supplied by Lion Elastomers. Additional ingredients such as N650 carbon black, processing oil, antidegradants, zinc oxide, peroxide and co-agent are used in typical proportions.
The formulas were compounded in a Farrel Model 2.6 BR Banbury mixer using a 74% fill factor for the first pass, 73% fill factor for the final pass, and with ram pressure set to 0.28 MPa. A two-stage mixing process was used, as outlined in table 3, in which elastomers, fillers, processing oil, antidegradants and zinc oxide were added in the first pass. In the first mixing stage, the rotor speed was increased after the ingredients were incorporated in order to bring the batch temperature to 138[degrees]C. The peroxide and co-agent were mixed with the masterbatch in the final (productive) pass.
The rubber was sheeted out on a Farrel two-roll mill after each Banbury mixing stage. Cure rate information was determined according to ASTM D 5289-17 using a moving die rheometer (Tech Pro rheoTech MDR, 0.5[degrees] arc, 170[degrees]C) (ref. 11). Rubber samples were compression molded with curing temperature equal to 170[degrees]C and molding time equal to 15 minutes for test plaques and 20 minutes for compression set buttons, abrasion specimens and crack growth specimens. The samples were then post-cured in an air oven for two hours at 149[degrees]C. Processing properties, including Mooney viscosity, ML (1+4) at 100[degrees]C, were determined according to ASTM D 1646-17 in a Monsanto MV 2000 viscometer using the large rotor (ref. 12).
Physical properties of the compounds were tested: tensile strength, elongation, durometer, tear resistance, compression set, DeMattia flex (crack growth) and DIN abrasion (rotary drum).
Tensile properties were tested according to ASTM D 412, Test Method A, die C (ref. 13). Tear strength was tested according to ASTM D [624-00.sup.E1] (2012), die C and die T (trouser tear) (ref. 14). Five tensile and tear specimens per sample were die cut from 2 mm thick test plaques using a hydraulic die press. Tensile and tear properties were evaluated using an Instron dual column testing system equipped with a 5 kN load cell and a long travel extensometer. For tensile strength, the gage length was 25 mm and grip separation velocity was 500 mm/minute. For tear resistance, the grip separation velocity was 500 mm/ minute for die C and 50 mm/minute for die T.
Durometer was measured as directed in ASTM D [2240-15.sup.E1], type A (ref. 15).
Compression set was tested according to ASTM D [395-16.sup.E1], Method B (ref. 16). Button specimens were aged 70 hours at 125[degrees]C under 25% deflection, and measurements were taken after a half hour recovery at room temperature.
DIN abrasion (rotary dmm) was tested per DIN 53 516/ ASTM D 5963-04 (2015), Method A (ref. 17).
DeMattia crack growth was tested per ASTM D 813-07 (2014) (ref. 18) using grooved and pierced specimens tested at 300 cpm, from a 2 mm starting crack until the crack grew to 12.7 mm.
Viscoelastic properties were examined using dynamic mechanical analysis (DMA) according to ASTM D 5992-96 (2011) (ref. 19). Storage modulus (E1), loss modulus (E") and tan [delta] data were obtained through strain sweeps in tension at 30[degrees]C with frequency equal to 1 Hz using a Metravib DMA 150 dynamic mechanical analyzer.
A dispersion analysis was performed using a Nanotronics nSpec 3D. A topography scan was performed using a 10X Objective and scan settings of [DELTA]Z = 0.5 and Model = 0.4. The 3D model was flattened after the scan.
Results and discussion
The evaluation of key processing parameters provides information about required manufacturing times and constraints. The compounds were evaluated for Mooney viscosity and cure kinetics.
Mooney viscosity at 100[degrees]C is used to indicate the ease of processing compounds or the ability of the compounds to flow at processing temperatures. Figure 4 provides a comparison of the effect of Lupasol and RFL coatings on the polyester fibers' resulting viscosity. There was no significant difference in the viscosities of the two compounds.
Figure 5 compares the [T.sub.s]2 scorch time of the rubber compounds at 170[degrees]C. The [T.sub.s]2 is the time it takes for the torque to rise two points above the minimum torque (ML). Figure 6 compares [T.sub.c]90 at 170[degrees]C, the time it takes for the rubber compounds to reach 90% of the maximum torque. Figure 7 shows the full rheometer curves for the Lupasol and RFL compounds. As shown by figures 5-7, the Lupasol and RFL compounds exhibit similar cure kinetics.
In addition to the processing characteristics, physical properties such as durometer, tensile, elongation, modulus, tear, compression set and abrasion were also evaluated for the Lupasol and RFL compounds. Tensile strength and tear strength were both tested with the grain and against the grain. The grain is imparted on the rubber compound during milling prior to molding. With grain means the fibers are oriented with the direction of strain, which means that the stress increases rapidly as the fibers take the load. In theory, the yield point is where the bonds between the fibers and the rubber begin to fail. The rubber then continues to stretch until the rubber fails (figure 8). Against grain means the fibers are oriented perpendicular to the direction of strain, which means that the rubber is stretching as stress is applied. These tensile curves look more typical for a rubber compound because the rubber is primarily taking the load of the stress (figure 9).
Table 4 shows a summary of the physical property data, including tensile, elongation, modulus, tear strength, durometer hardness, compression set and DIN abrasion. Most of the physical properties are similar between the Lupasol and RFL compounds. The slight differences favoring Lupasol in some cases suggest that the fibers with the Lupasol coating may be more thoroughly dispersed than the RFL coated fibers.
Dynamic properties, including DeMattia crack growth and dynamic mechanical analysis (DMA), were also evaluated for the Lupasol and RFL compounds. Figures 10 and 11 show that the Lupasol and the RFL had similar performance for DeMattia crack growth.
Storage modulus (E'), loss modulus (E") and tan [delta] data were obtained through strain sweeps in tension at 30[degrees]C with frequency equal to 1 Hz. The range of dynamic strain was selected to ensure that the bonds between the rubber and the polyester fibers were not broken. Figures 12-14 show the storage modulus (E'), loss modulus (E") and tan [delta] data of the compounds, respectively. The Lupasol compound has slightly higher storage modulus values than the RFL, which suggests that the Lupasol compound is more reinforcing, possibly due to better dispersion of the fibers or better bonding of the fibers to the rubber. The Lupasol compound has slightly higher loss modulus values than the RFL, which also suggests that the Lupasol compound is more reinforcing than the RFL compound. The tan deltas of the Lupasol and the RFL are very similar, despite the differences in the storage and loss modulus profiles.
Figures 15 and 16 show the Payne effect and Mullins effect of the Lupasol and RFL compounds. The Payne effect is the drop in E' as the dynamic strain is increased. The Payne effect is attributed to the filler-filler interaction, the breaking and recovery of weak physical bonds linking adjacent filler particles. Better filler dispersion gives lower Payne effect. The Mullins effect is a measure of the dynamic stress softening (the drop in E') that is observed between the first and second strain sweeps due to the polymer-filler matrix being pulled apart during the first strain sweep and not having time to reagglomerate. A lower Mullins effect would indicate stronger polymer to filler interaction. The Lupasol compound has a slightly lower Payne effect than the RFL, which suggests that the Lupasol compound has better dispersion of the fibers. The Lupasol compound has a lower Mullins effect than the RFL compound, which suggests that the Lupasol coated fibers have better bonding to the rubber matrix than the RFL coated fibers.
A 3D topography scan was performed on the Lupasol and RFL compounds looking at cuts that were made with the grain (looking at the sides of the fibers) and against the grain (looking at the ends of the fibers). Table 5 shows a summary of the data collected from the topography scan. In this analysis, a peak or a valley was identified as a fiber (either the end of it or the side of it, depending on the view).
Sa is the arithmetical mean roughness value (area): the arithmetical average of the absolute values of the profile height deviations from the mean surface plane, recorded within the evaluation area. Sq is the root mean square deviation (area): the root mean square average of the profile height deviations from the mean surface plane, recorded within the evaluation area. It is equivalent to the standard deviation of heights.
Lupasol and RFL, with grain
The with grain data from table 5, along with the images in figures 17-20, suggest that the fibers with the RFL have a larger volume than the Lupasol coated fibers. Since the same polyester fibers were used (only different coatings) in both compounds, this may indicate that the RFL fibers are clumped together, instead of individually dispersed.
Lupasol and RFL, against grain
The against grain data from table 5, along with the images in figures 21-24, also suggest that the fibers with the RFL have a larger volume than the Lupasol coated fibers. The camera images in figures 21 and 22, as well as the number of peaks and valleys shown in the table, show that there are many individual fibers visible in the compound with the Lupasol coated fibers, while the RFL coated fibers seem to be grouped together in a lower number of clumps. This dispersion analysis supports other evidence (such as the physical property data and the DMA data) that suggests that the Lupasol coated fibers achieved better dispersion in the polymer matrix than RFL coated fibers.
Most of the physical properties of the Lupasol and RFL compounds are very similar. The slight differences in properties such as elongation and tear resistance suggest that the fibers with the Lupasol coating may be more thoroughly dispersed than the RFL coated fibers.
The dynamic mechanical analysis showed that the Lupasol compound has slightly higher storage modulus and loss modulus values than the RFL, while maintaining similar tangent deltas. This also suggests that the Lupasol compound is more reinforcing, possibly due to better dispersion of the fibers or better bonding of the fibers to the rubber.
The dispersion analysis of the against grain compound showed both visually and numerically that there are more individual fibers in the compound with the Lupasol coated fibers, while the RFL coated fibers seem to be grouped together in a lower number of clumps.
All the data indicate that the Lupasol coated fibers achieved better dispersion in the polymer matrix than RFL coated fibers, while maintaining the physical properties of the compound.
This article is based on a paper presented at the 194th Technical Meeting of the Rubber Division, ACS, October, 2018.
by Charles O. Kerobo, Prabodh Varanasi and Dennis Berry, BASF, and Kylie Knipp, Akron Rubber Development Laboratory
(1.) U.S. Environmental Protection Agency, Office of Air and Radiation, Report to Congress on Indoor Air Quality, Volume II: Assessment and Control of Indoor Air Pollution (1989).
(2.) World Health Organization, International Agency for Research on Cancer, IARC Monographs on The Evaluation of Carcinogenic Risks to Humans, Volume 88 (2006).
(3.) National Toxicology Program, U.S. Department of Health and Human Services, 14th Report on Carcinogens, Known to Be Human Carcinogens (2016).
(4.) RAC Adopts 17 Scientific Opinions, European Chemicals Agency (2012).
(5.) James A. Swenberg, Benjamin C. Moeller, Kun Lu, Julia E. Rager, Rebecca C. Fry and Thomas B. Starr, "Formaldehyde carcinogenicity research: 30 years and counting for mode of action, epidemiology and cancer risk assessment, " Toxicologic Pathology, 41: 181-189, 2013.
(6.) A. Louis, J.W.M. Noordermeer, W.K. Dierkes and A. Blume, University of Twente, Department of Elastomer Technology and Engineering, P.O. Box 217, 7500 AE Enschede (The Netherlands).
(7.) U.S. Patent 2017130396 Dipping Solution for Cord Fabrics to Kordsa Global, Nacide Nurcin Cevahir, Ali Ersin Acar and Mustafa Yasin Sen, May 11, 2017.
(8.) Alexandre Gomes, Nermeen Nabih and Thomas Kramer, Rubber World, pp. 24-26, March 2016.
(9.) Bridgestone Group, Eco-Activities, Bridgestone Corporation, Looking Ahead to the World in 2050, Bridgestone Group Environmental Report, 2014.
(10.) Mehler Engineered Products, Edelzeller Str. 44, 36043 Fulda, Germany.
(11.) ASTM D 5289-17 Standard Test Method for Rubber Property, Vulcanization Using Rotorless Cure Meters, 2017.
(12.) ASTM D 1646-17 Standard Test Methods for Rubber, Viscosity, Stress Relaxation and Pre-Vulcanization Characteristics (Mooney Viscometer), 2017.
(13.) ASTM D 412-16 Standard Test Methods for Vulcanized Rubber and Thermoplastic Elastomers, Tension, 2016.
(14.) ASTM D 624-00E1 (2012) Standard Test Method for Tear Strength of Conventional Vulcanized Rubber and Thermoplastic Elastomers, 2012.
(15.) ASTM D 2240-15E1 Standard Test Methods for Rubber Property, Durometer Hardness, 2015.
(16.) ASTM D 395-16E1 Standard Test Methods for Rubber Property, Compression Set, 2016.
(17.) DIN 53 516/ASTM D 5963-04 (2015) Standard Test Method for Rubber Property, Abrasion Resistance (Rotary Drum Abrader), 2015.
(18.) ASTM D 813-07 (2014) Standard Test Method for Rubber Deterioration, Crack Growth, 2014.
(19.) ASTMD 5992-96 (2011) Standard Guide for Dynamic Testing of Vulcanized Rubber and Rubber-Like Materials Using Vibratory Methods, 2011.
Caption: Figure 1--structure of polyethyleneimine
Caption: Figure 2--RFL-coated polyester chopped fibers
Caption: Figure 3--Lupasol-coated polyester chopped fibers
Caption: Figure 7--measured rheometer torque from MDR at 170[degrees]C
Caption: Figure 8--stress-strain curve for Lupasol and RFL tested with the grain
Caption: Figure 9--stress-strain curve for Lupasol and RFL tested against the grain
Caption: Figure 11--DeMattia crack growth versus number of cycles
Caption: Figure 12--storage modulus from strain sweep at 30[degrees]C
Caption: Figure 13--loss modulus from strain sweep at 30[degrees]C
Caption: Figure 14--tangent delta from strain sweep at 30[degrees]C
Caption: Figure 17--image at 10x magnification of Lupasol with grain
Caption: Figure 18--image at 10x magnification of RFL with grain
Caption: Figure 19--3D model of surface of Lupasol with grain
Caption: Figure 20--3D model of surface of RFL with grain
Caption: Figure 21--image at 10x magnification of Lupasol against grain
Caption: Figure 22--image at 10x magnification of RFL against grain
Caption: Figure 23--3D model of surface of Lupasol against grain
Caption: Figure 24--3D model of surface of RFL against grain
Table 1--physical properties of Lupasol WF used for coating Typical physical characteristics Lupasol WF Average molecular weight (Mw) (g/mol) 25,000 Viscosity at 20[degrees]C (mPa.s) 100,000 Concentration in (wt.%) >99 Water concentration (wt.%) ~1 Pour point ([degrees]C) -1 Boiling point ([degrees]C) >>200 Density at 20[degrees]C (g/[cm.sup.3]) 1.10 pH (1% in water) 10-12 pKa value 7-10 Charge density 17 Ratio of 1[degrees]: 1:1.1:0.7 2[degrees]: 3[degrees] amine Table 2--recipe formulation (parts per hundred rubber, by weight) First pass Lupasol RFL Material phr phr Royalene 580-FIT 100.00 100.00 N-650 (carbon black) 50.00 50.00 Sunpar 2280 (paraffinic oil) 15.00 15.00 Zinc oxide 5.00 5.00 Vanox CDPA 1.00 1.00 Vanox ZMTI 1.50 1.50 Polyester fibers with Lupasol 15.00 -- Polyester fibers with RFL -- 15.00 Total 187.50 187.50 Final pass Lupasol RFL Material phr phr Masterbatch, first pass 187.50 187.50 Vanax MBM 1.00 1.00 Varox DCP-40KE 8.00 8.00 Total 196.50 196.50 Table 3--mixing protocol of EPDM v-belt compounds First pass masterbatch Starting temperature 38[degrees]C Starting rotor speed: 65-70 rpm 0 minutes Add black, oil, remaining chemicals and polymer 82[degrees]C Sweep 93[degrees]C Sweep 110[degrees]C Sweep 138[degrees]C Dump Final pass Starting temperature: 38[degrees]C Starting rotor speed: 65-70 rpm 0 minutes Sandwich in cures 82[degrees]C Sweep 99[degrees]C Dump Table 4--summary of physical property data Lupasol RFL with grain with grain Tensile strength at break, MPa 9.63 9.15 Standard deviation 0.35 0.95 Tensile strength at yield, MPa 8.03 8.25 Standard deviation 0.60 0.52 Elongation strain at break, % 244 221 Standard deviation 7 23 Elongation strain at yield, % 31 32 Standard deviation 7 13 50% modulus, MPa 7.49 7.46 Standard deviation 0.46 0.28 100% modulus, MPa 7.14 7.26 Standard deviation 0.27 0.07 200% modulus, MPa 8.08 7.07 Standard deviation 0.13 2.81 Tear strength die C, kN/m 38.31 37.51 Standard deviation 3.65 3.15 Tear strength die T, kN/m 18.23 13.31 Standard deviation 1.44 1.20 Durometer A, points 75 72 Standard deviation 1.3 1.1 Compression set, % 25 25 Standard deviation 3.6 1.3 Abrasion average volume loss 195 191 ([mm.sup.3]) Standard deviation 5.9 8.7 Lupasol RFL against grain against grain Tensile strength at break, MPa 9.33 8.58 Standard deviation 0.12 0.43 Tensile strength at yield, MPa No yield No yield Standard deviation point point Elongation strain at break, % 291 263 Standard deviation 6 12 Elongation strain at yield, % No yield No yield Standard deviation point point 50% modulus, MPa 2.70 2.43 Standard deviation 0.43 0.20 100% modulus, MPa 3.69 3.55 Standard deviation 0.43 0.17 200% modulus, MPa 5.99 6.38 Standard deviation 0.33 0.17 Tear strength die C, kN/m 35.53 30.87 Standard deviation 2.06 3.09 Tear strength die T, kN/m 14.86 16.31 Standard deviation 0.42 1.96 Durometer A, points -- -- Standard deviation -- -- Compression set, % -- -- Standard deviation -- -- Abrasion average volume loss -- -- ([mm.sup.3]) Standard deviation -- -- Table 5--summary of data from nSpec 3D topography scan 3D topography scan summary Sample ID Orientation Lupasol RFL With grain With grain Average volume of peaks and 4,880.3 6,419.3 valleys, [micro][m.sup.3] Sa (surface roughness), [micro]m 3.65 6.92 Sq (roughness deviation), [micro]m 53.95 71.19 Number of peaks and valleys 105 189 3D topography scan summary Sample ID Orientation Lupasol RFL Against grain Against grain Average volume of peaks and 5,776.2 9,013.1 valleys, [micro][m.sup.3] Sa (surface roughness), [micro]m 6.61 6.56 Sq (roughness deviation), [micro]m 82.15 63.86 Number of peaks and valleys 264 164 Figure 4--Mooney viscosity at 100[degrees]C Mooney viscosity 100[degrees]C Mooney viscosity MU Lupasol 71.82 RFL 73.10 Note: Table made from bar graph. Figure 5--scorch time ([T.sub.s]2) by MDR MDR 0.5[degrees] arc/170[degrees]C TS2, minutes Lupasol 0.42 RFL 0.42 Note: Table made from bar graph. Figure 6--[T.sub.c]90 (time to 90% cure) by MDR MDR 0.5[degrees] arc/170[degrees]C T90, minutes Lupasol 4.85 RFL 5.35 Note: Table made from bar graph. Figure 10--DeMattia crack growth after 1,000 cycles DeMattia crack growth (after 1,000 cycles) Crack growth (mm) Lupasol 23.62 RFL 22.35 Note: Table made from bar graph. Figure 15--Payne effect from strain sweep at 30[degrees]C Payne effect Payne effect (Pa) Lupasol 3.26E+06 RFL 3.93E+06 Note: Table made from bar graph. Figure 16--Mullins effect from strain sweep at 30[degrees]C Mullins effect Mullins effect (Pa) Lupasol 4.34E+05 RFL 1.13E+06 Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Author:||Kerobo, Charles O.; Varanasi, Prabodh; Berry, Dennis; Knipp, Kylie|
|Date:||Feb 1, 2019|
|Previous Article:||Understanding devulcanization, the path to a circular economy and potential market impact.|
|Next Article:||Improved off-the-road tire formulations with discrete surface functionalized multiwall carbon nanotubes: Correlating wear, abrasion and tear...|