Abstract
There is a strong need to replace carbon black and silica as a reinforcing filler in rubber industry. Lignin is a promising option as a replacement, because of its wide availability, complex chemistry and sustainable source. Typically, the reinforcing effect of lignin is weak and one major cause is the poor dispersion of filler. In this study, typical dispersion parameters in mixing are tested in a systematic way to validate their effect on kraft lignin dispersion in polybutadiene rubber. Tested mixing variables are temperature, rotor speed, mixing time, fill ratio and shear forces. The effect of variables to dispersion are analyzed from electron microscopy images with an image analysis software. With this novel method exact number of particles and their properties could be calculated. Unlike for conventional fillers (carbon black and silica), the results show clearly that only the mixing temperature has an effect on lignin dispersion in rubber. With high mixing temperatures, dispersion and hence mechanical properties improved. The better dispersion is especially seen as decreasing number of large (>Ø 5 μm) particles. Other mixing variables did not have significant effect on dispersion and the number of large particles.
Introduction
Different kinds of fillers are used with elastomers to affect their mechanical properties and price of the compound. Mostly used reinforcing fillers are carbon black and silica.1–3 To gain full reinforcing potential from carbon black and silica, the dispersion of the filler to rubber matrix is crucial. Applied mixing method and mixing parameters have a major effect on the dispersion. Within the parameters, applied shear forces, mixing time and temperature, speed of the rotors, fill factor and ram pressure have the most significant effect on the dispersion. Especially with silica and carbon black high shear forces during the mixing improve the dispersion as filler agglomerates are broken to smaller aggregates. During the dispersive mixing stage too high mixing temperature (>130°C) will lower the viscosity of the mixture and lower the shear forces, while high speed of the rotors will maximize the shear effect. Correct fill factor and ram pressure will ensure that shear forces from rotors are transferred to the mixture. After the mixing, typical target diameter for a reinforcing filler is between 10 and 100 nm. For semi-reinforcing fillers, like hard clay, the critical diameter is typically between 100 and 1000 nm, while non-reinforcing fillers, like barium sulfate, the diameter is between 1000 and 10,000 nm.4–6
Carbon black is produced from heavy petroleum and production of silica consumes a lot of energy, therefore more sustainable replacements are widely studied.1,3 One of the sustainable alternatives to be used as a reinforcing filler is lignin and recent study shows that greenhouse gas emissions can be reduced with substitution of conventional fillers with it. 7 Lignin has a highly branched three-dimensional phenolic structure, which includes phenolic hydroxyl, carboxyl and methoxy groups.8,9 It is widely available, as over 15% of dry weight of most terrestrial plants is lignin. Lignin is produced by the pulp and paper industry as a side stream, and as there are not many suitable applications for its use, lignin is mostly burned to generate power in biomass refineries.8,10,11 First studies using lignin with rubbers are made in 1940s and 1950s, however the reinforcing properties of lignin were found moderate or poor.12–16 Polarity of the lignin molecules causes problems for lignin to integrate with non-polar rubber matrix. This can result in poor dispersion and poor properties of the compound, which is typical behavior of the lignin.
The use of lignin is typically studied with most common elastomers, natural rubber (NR), styrene-butadiene rubber (SBR) or butadiene rubber (BR).17–20 Recent studies show promising results with BR21,22 which is typically used with natural rubber on tyre sidewalls. 4 As the poor dispersion has noted as a hindrance, different methods are studied to improve it. Such methods are for example chemical23–25 or thermal 26 modification of lignin or using some coupling agent21,26,27 between lignin and elastomer interface. For examples benzoyl peroxide, 28 aldehydes 24 and silylation 25 are studied as a chemical modification to improve the chemical compatibility of lignin to rubber matrix. In addition silane coupling agents are used during mixing to enhance the interaction between lignin and polymer.22,26,29 Some studies show, that using epoxidized natural rubber (ENR) as a compatibilizer improves the dispersion of lignin and carbon black in NR/BR blend.24,30 Studies show that poor dispersion problem occurs especially when dry powder form of lignin is used instead of solutions.24,31 However available commercial grades of lignin are in powder or granular form.
Mixing techniques and parameters can be optimized to improve the dispersion of lignin. If mixing is done with two roll mill, the reinforcing effect of lignin is very limited, mostly caused by poor dispersion.27,31,32 Using latex as co-precipitation method in compounding results better dispersion, however it limits the suitable elastomers which can be used.33–35 In mixing parameters, high temperature (180°C) during mixing has been shown to lead to good dispersion of sulfate lignin with ENR. The addition of lignin was made with two roll mill and the high-temperature dynamic heat treatment was made with internal mixer. The heat treatment time was considerably long, 30 min 36 In other study high temperature mixing process above the glass transition temperature Tg of lignin (approx. 150°C for kraft lignin) was found to improve the reinforcing properties of lignin to higher level than with carbon black in BR. Likewise, the dispersion of the lignin in rubber matrix was improved. The four stage mixing process, which was used in that study, took over 40 min to mix the sample. 21 Systematic research about effects of mixing parameters to lignin dispersion is not done. The morphology of lignin is totally different from carbon black, so the optimal mixing conditions may be also different. 37
Carbon black and silica dispersion in rubber have been studied widely, and ASTM standards explain the methods and result analysis.38–41 However, these methods are not directly applicable for lignin and there is no relevant study to measure the actual diameter of lignin particles in rubber matrix and determine what is the critical size for lignin to gain reinforcing properties. Typically reported values for lignin particle diameter in rubber matrix are between several nanometers to greater than 100 μm36,37,42 The average particle size of kraft lignin in dry powder form is measured around 80 nm, but they can form aggregates larger than 100 μm.21,42–45 Particle size of the dry lignin can be affected with chemical modification and milling.42,44 Generally in different publications dispersion analysis is based on few SEM pictures without systematic analysis about particle size and size distribution. In different studies lignin is mixed to elastomer with internal mixer and the dispersion of the lignin is analyzed visually from SEM images.31,37,42,46–48 In this study, a more advanced methodology to measure lignin dispersion with advanced sample preparation and quantitative particle analysis is introduced. The effect of typical mixing parameters of internal mixer to dispersion of kraft lignin were studied and the effect of large particle amount to mechanical properties were measured.
Experimental section
Polybutadiene rubber (BR, Buna CB 25, density 0.91 g/cm³) provided by Arlanxeo was used as the polymer in the rubber compounds. Biopiva 395 kraft lignin produced by UPM (Tg 150°C) was dried for 3 h at 60°C before mixing with rubber. Rubber additives, namely stearic acid, zinc oxide (ZnO), sulphur, N-cyclohexyl-2- benzothiazole sulfenamide (CBS) and N,N′-Diphenylguanidine (DPG) were industrially available products and used as received.
Compound formulation.

Mixing procedure schema for different samples.
Mixing parameters for different samples.
For the mixing temperature, rotor speed and mixing time series (Te, Sp and Ti), a masterbatch was first mixed with Krupp Elastomertechnik GK 1,5 E laboratory rubber mixer. Four sets of masterbatch were made and then each of them was divided to four subsets for Stage 2 mixing. The Stage 1 mixing for the fill ratio and shear force series (Fi and Sh), and the Stage 2 and 3 mixing for all samples were made with Brabender N 350 E twin rotor mixer with roller blades.
Excluding the mixing temperature series, the dump temperature was kept constant. In the case of mixing speed (Sp) series this was achieved by varying the start temperature at Stage 2. To control the dump temperature of the mixing time (Ti) series, the Stage 2 was divided in 5 min cycles letting the machine and the material to cool down between the cycles.
After mixing, the samples were roll milled to sheets and cured to 2 mm thick sheets at 160°C temperature for t90 + 3min. The curing time was determined with Advanced Polymer Analyzer (APA 2000, Alpha Technologies) at 160°C.
Samples for scanning electron microscopy (SEM) imagining were prepared with JEOL Cooling cross section polisher IB-19,520CCP. Approximately 7 × 7 mm sample was cut from vulcanized sheet and one of the cut edges was polished by argon gas with 6 kV for 180 min and with 3 kV for 60 min, both at −50°C. The sample and polished area can be seen in Figure 2(a). With polishing the topography of the measured area was removed, which enables the analysis of the lignin particles from the SEM image. Cross section of the sample with polished area(a), SEM image from polished area (b) and analyzed image after lignin particle separation with software (c).
After the polishing the samples were coated with a thin carbon layer to ensure their sufficient conductivity. Zeiss ULTRAplus microscope was used for SEM imaging with 10 kV acceleration voltage. The SEM images were analyzed with Dragonfly software, by Objects Research System, version 2022.1.0.1259. In the image analysis, the SEM image (Figure 2(b)) was separated to three regions of interest: dark area as rubber matrix, grey particles as lignin and white dots as zinc oxide. Then the software was used to separate the lignin particles each as their own (Figure 2(c)), so they could be analyzed. The analysis gives the total number of particles and measures mean feret diameter and area of each particle on the cutting plane. For the macro dispersion analysis, the used imaging magnification was 500x and the total observed area was 0.15 mm2. The number of measured particles was between 5000 and 21,000 per sample, enabling quantitative result of dispersion. For the micro dispersion analysis the magnification was 2000x and the total observed area 0.01 mm2. The number of measured particles was between 2000 and 6000 per sample.
Tensile properties were measured for each sample with Zwick/Roell Robotest L/F testing machine. Tests were made according to ISO 37 using dumbbell specimen and with speed of 500 mm/min. The results are reported as average values for at least nine parallel samples together with respective standard deviations.
Results and discussion
Curing times
Sample curing time.
Dispersion analysis
There are certain standards and commercially available equipment to analyze the dispersion of carbon black or silica in rubber matrix. Unfortunately, existing methodology is unable to analyze lignin dispersion, as the method is based on optical microscopy and the light refection of lignin differs from carbon black and silica. However, the basic idea of their analysis method could be utilized to lignin. The SEM images from well polished surface contains no topography contrast, which enables the separation of filler particles from rubber matrix solely based on the compositional contrast and further quantitative analysis with image analysis software. With this quantitative data, similar calculations can be performed as with existing methodology for carbon black and silica.
For typical fillers, the rubber compound dispersion analysis is based on ASTM D7723 standard (Standard Test Method for Rubber Property—Macro-Dispersion of Fillers in Compounds), that uses optical microscopy to measure the macro dispersion of the filler based on the roughness of the cut rubber surface. Based on the standard ASTM D3053, the macro dispersion is defined as the particles sized between 2 µm and 100 μm. In addition, the ASTM D2663 standard (Standard Test Methods for Carbon Black—Dispersion in Rubber) explains different methods to evaluate the degree of dispersion of carbon black in rubbers. With one of those methods the degree of dispersion can be calculated from cut section by measuring the surface area of particles larger than a chosen diameter. The digital image of the measured cut section is processed so that filler agglomerates can be isolated from the rubber matrix. Then the surface area of chosen particles can be calculated from the image and agglomerate area fraction is divided by the volume percentage of the carbon black in compound. The standard gives classifications of dispersion to certain dispersion percentages for carbon black. These classifications cannot be directly used with lignin, as the filler properties are different, but it still gives some starting point for lignin dispersion evaluation. According to the ASTM D7723 standard, the percentage of filler that has been dispersed below certain particle size can be calculated by the following equation (1),
In this study the number of larger than 5 μm particles was found to have a strong relationship with the mechanical properties of the rubber (see next Section), hence that particle diameter was chosen as the limit value between macro (>Ø5 μm) and micro (<Ø5 μm) dispersion. To gain sufficient measured area per image for representative and reliable results to measure sufficient number of large (>Ø5 μm) particles, a relatively low SEM image magnification (500x) was chosen. However, this caused uncertainty on micro dispersion analysis, meaning the identification of smaller (<Ø5 μm) particles. Therefore, a higher magnification (2000x) was chosen for the micro dispersion study so that almost all dispersed particles could be measured, including the smallest ones. Further, in the micro dispersion study the observed area was chosen to avoid large particles filling the image area and impairing the results. Representative SEM image of each sample can be found in supplementary data.
The effect of mixing temperatures on compound properties
The most significant effect of the studied mixing parameters on the lignin dispersion was found to be on the mixing temperature. First the macro dispersion of lignin was analyzed from the samples with different mixing temperatures. From Figure 3(a) it can be clearly seen that the number of particles having diameter larger than 5 μm and 10 μm decreases as a function of maximum mixing temperature. Same improvement of dispersion by the elevated mixing temperature was seen in previous studies.21,36 At the same time the degree of dispersion increases, as can be seen in Table 4. When the number of large particles decreases, the total number of particles increases as large particles are broke to smaller ones. The total area covered by lignin was between 16%–24% of the cross sections of the measured samples, lignin volume fraction in compound is 25%. Number of particles as a function of mixing temperature in samples Te1…8 (a), maximum particle size as a function of mixing temperature (b) and average particle size in micro dispersion with different mixing temperatures (c). Dispersion % in samples te1…8 with chosen diameter 5 µm and 2 µm.
As seen in Figure 3(a), the number of >10 μm particles reach zero as the mixing temperature is above 200°C. Also, the maximum size of the particles decreases within increasing temperature but stabilizes above 180°C to just below 10 μm (Figure 3(b)). In Figure 4, the one hundred largest particles on imaged area on samples Te1 and Te8 are highlighted. The 100 largest particles on samples Te1 (a) and Te8 (b).
In the micro dispersion analysis, the average particle diameter of all measured particles was between 400 and 600 nm, with elevated mixing temperature average particle diameter increased. This can be caused by the analyzing technique, the large >10 μm particles were not visible in micro dispersion images, and while with higher mixing temperatures those large particles are breaking down to smaller ones, those will fit to measured area and increased the average diameter. From Figure 3(c) can be seen, that the average diameter of smallest particles (first decile) remained between 100 and 150 nm within the range of studied mixing temperatures. Thus, increasing the mixing temperature seems to help to decrease the diameter of the largest particles but to have less effect on the smallest ones. However, the dispersion analysis of the smallest particles should be improved with more accurate imaging methods. The applied magnification and the selected electron detector (angle selective backscatter detector) do not provide the highest resolution and accuracy for the analysis; the selection was based on the best possible contrast between lignin and rubber. Therefore, the smallest particles close to each other can be merged in one in the image analysis resulting in distorted results.
The effect of improved dispersion could also be seen with mechanical testing. The average stress – strain curves for samples with different mixing temperatures can be seen in Figure 5(a). The tensile strength increased with higher mixing temperature, while at the same time the elongation at break decreased (Figure 5(b)). Same effect of mixing temperature is visible on stress at 100 % strain, where highest mixing temperature sample Te7 and Te8 had the highest values (Figure 5(c)). In Figure 5(d) can be seen, that there is a strong correlation between the number of > Ø5 μm particles and 100 % modulus. In general, better dispersion of fillers improves the mechanical properties of rubber5,49: even if there are no connections between a polymer and a filler, the tensile properties of the compound are improved compared to non-filled rubber, for example by hydrodynamic effect.2,50,51 Also, large filler particles act as a stress concentrators decreasing the tensile properties of the compound.
49
The decrease in elongation is assumed to be caused by the higher mixing temperature, as the polymer chains start to break during mixing. Stress – Strain curves for samples with different mixing temperatures (a), tensile strength and elongation at break as a function of mixing temperature (b), stress at 100% strain of samples with different mixing temperatures (c) and 100% modulus as a function of the number of > Ø5 μm particles.
The effect of other mixing parameters on compound properties
The other test series included rotor speed, mixing time, fill ratio and shear forces. The macro dispersion analysis of other test series indicated, that these mixing parameters do not cause any major improvement to lignin dispersion. Mixing with different rotor speed, mixing time, fill ratio or shear forces caused some variations on the number of lignin particles larger than 5 μm and 10 μm, but the total amounts were so large that the degree of dispersion was very low in all of the cases. The number of >Ø5 μm particles was between 76 and 182 pieces, which correlates the number of particles in the samples from the temperature series which were mixed below 155°C. The individual macro dispersion analysis results can be seen is Figure 6. Also, no variation on the micro dispersion level of lignin could be detected, the average diameter of particles was around 400 nm and the average of smallest particles (first decile) remained between 100 and 150 nm. Number of particles as a function of mixing time (a), number of particles as a function of rotor speed (b), number of particles as a function of fill ratio (c) and number of particles with different shear forces (d).
The effect of very low macro dispersion of the lignin could be seen also in the tensile results. The tensile strength of these test series were between 2.0 and 2.3 MPa, which is the same low level as the low temperature mixing samples Te1 and Te2 mixed below 140°C. In addition, the stress at 100% strain were around 1.6 MPa, which was comparable to samples Te1 and Te2.
The results clearly indicate that the macro dispersion of lignin can only be improved with higher mixing temperature. In contrast mixing time, rotor speed or fill ratio do not have any significant impact on the macro dispersion. Variation in torque value did not cause difference on the dispersion, from which can be concluded that shear forces during mixing do not have a major role in the dispersion of lignin. These results are inconsistent with the dispersion mechanism results of carbon black and silica. The possible explanation is the differences in chemistry and morphology of these fillers. While carbon black and silica form agglomerates which can be broken to aggregates with high shear forces, lignin has a polymeric macromolecular structure, and it most probably acts like a thermoplastic polymer during mixing. Higher mixing temperature might cause the breakup of the polymeric network and then improve the apparent dispersion of lignin. Hait et al. suggest, that mixing above the glass transition temperature Tg of lignin (153°C) is the key to improved dispersion. 21 However in this study there is no clear improvement of dispersion on that particular temperature area, but the dispersion improves quite steadily as a function of temperature in the range of 120 °C–215°C. The decomposition of lignin starts between 240-280°C, but studies show that in lignin powder there are visible changes in the particle shape already around 190°C, which can cause the improvement in dispersion at high mixing temperatures.52,53 On the other hand micro dispersion (<Ø 5 μm) seems to stay at same level throughout the test series.
The novel technique based on SEM images from polished samples to measure particle amount and size in rubber can be used to analyze the dispersion of lignin. The method enables quantitative analysis of filler particles and traditional calculations of dispersion can be obtained from the data. The technique is fairly easy and quick to be used at least in scientific research, however future development to gain accurate results of micro dispersion of lignin must be done.
The dispersion of filler is one of the key attributes to gain high reinforcing properties. On the way of replacing carbon black and silica, the improvements in lignin dispersion are crucial. This study widens the understanding of the effect of mixing parameters to the dispersion and introduces novel technique to analyze it. While the results show that the dispersion mechanism during mixing differs from CB and silica, new methods of mixing have to be developed to enable the use of lignin as a reinforcing filler.
Conclusion
Carbon black and silica are the most used reinforcing fillers in rubber applications. There is a strong need to find more sustainable replacement for them and as a biobased material lignin is one potential option. In this study kraft lignin was mixed with polybutadiene rubber and the effect of mixing variables on the lignin dispersion were analyzed. The mixing temperature, the rotor speed, the mixing time, the fill ratio and the amount of shear forces were varied and the dispersion of lignin in rubber matrix was analyzed with a novel method based on scanning electron microscope image analysis from argon polished rubber samples. With this method, exact number of particles could be calculated, and their size analyzed. Particularly, the macro dispersion (lignin particles >Ø 5 μm) and micro dispersion (lignin particles <Ø 5 μm) were analyzed separately as a function of the varied process parameters. Variation in mixing temperature changed the macro dispersion of lignin, with higher mixing temperature the number of large particles decreased, and tensile strength increased. However, the micro dispersion seemed unaffected. Other mixing parameters did not affect the micro or macro dispersion and the degree of lignin dispersion was very low. The mixing parameters, which change the dispersion of conventional fillers like carbon black and silica, do not affect the dispersion of lignin in the same manner. The result of this study emphasizes the difference of lignin compounding compared to carbon black and silica. In future studies the effect of improved dispersion to other mechanical properties should be studied.
Supplemental Material
Supplemental Material - Lignin dispersion in polybutadiene rubber (BR) with different mixing parameters
Supplemental Material for Lignin dispersion in polybutadiene rubber (BR) with different mixing parameters by Jukka Koskinen, Noora Kemppainen and Essi Sarlin in Progress in Rubber, Plastics and Recycling Technology
Footnotes
Acknowledgements
This work made use of Tampere Microscopy Center facilities at Tampere University. Some parts of the processing and testing were done in Nokian Tyres’ laboratory.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Jukka Koskinen works in Nokian Tyres group, but the work is separate from the research.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Tampere University and Nokian Tyres Plc through the Doctor School of Industry Innovation (DSII), and did not receive any specific grants from funding agencies in the public or not-for-profit sectors.
Correction (December 2024):
The article type has been changed from Review article to Original article since its original publication.
Supplemental material
SEM example images of all samples for macro dispersion analysis (PDF).
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
