MILLS, MICROSCOPES AND MICROSURFACING G. Holleran, Vice President, Valley Slurry Seal Company, USA Back to VSS Technology Papers Library 1. INTRODUCTION The importance of particle size in emulsions has been discussed in many papers (1,2,3). It is a determinant of emulsion stability, coating, break rate and cure rate. This is of course not the whole story, as formulation, raw materials and aggregates are also critical. However particle size and particle size distribution are important variables and are controllable with formulation, raw materials and the equipment used to manufacture the emulsion. Many of the processes of breaking and curing are directly dependant on it (3). This study began as a technology transfer to a foreign customer to develop firstly a quick set slurry system using their asphalt and aggregates, progressing to a microsurfacing system. The obstacles that needed to be overcome included a difficulty in correlating their laboratory results to our own. The investigation highlighted the equipment as an important variable and the solutions involved some novel formulation methods to overcome this. In the meantime some interesting results were observed on the effect of particle size and distribution on microsurfacing and quick set slurry performance. The main conclusions that could be drawn are that particle size and particle size distribution have an effect on microsurfacing and quick set slurry break, coating of aggregate , cure time and traffic time. This finding was applied to adjusting the mix designs and later to equipment technology for other customers. The variables that affect emulsion and end use performance are many but an optimized balance can be achieved and even shortcomings of some systems overcome. Such as asphalt quality. (4). It may be concluded that the colloid mill is a large influence on emulsion performance and that this in turn has significant effects on 2. POTENTIAL EFFECTS OF PARTICLE SIZE AND DISTRIBUTION The work of Durand and others (1) shows some correlations between emulsion properties and particle size distribution and size. It needs to be pointed out that emulsion instability and controlled break are different effects. There is no doubt that coarse emulsion will break faster in the silica flour test or in straight aggregate mixing tests ( coagulation) but what we need to consider is break after adequate coating of aggregate particles and the formation of a cohesive mixture. 2.1 Emulsion Stability: The main mechanisms of stability are (3): Sedimentation Flocculation Coalescence Inversion ( water entrapment) Ostwald Ripening The potential effects of particle size and distribution (PS and PSD) on these properties are: Sedimentation: That is the settlement of the emulsion particles down in the emulsion. This is an inverse square relationship. That is ,as particle size ( of any of the particles in the emulsion) , increases the sedimentation rate increases as the square of that increase. E.g. double the particle size increase the sedimentation rate by a factor of 4. Flocculation: That is the creation of aggregates (floccs) of particles. This is a function of the surface charge and double layer thickness. Largely therefore an emulsifier effect. However , as particles are coarse they will form larger aggregates and exacerbate sedimentation. Larger particles may also trap more water in the interstitial space in the floccs. Coalescence: The creation of large particles from floccs. This is a function again of energy differences and largely an emulsifier effect. However larger floccs form larger coalesced particles and again effect sedimentation and water entrapment. Inversion: That is water entrapment. This is a function of the rate of flocculation and coalescence and is an emulsifier effect or reactivity effect. However, larger particles will entrap larger amounts of water due to the large interstitial spaces. Ostwald Ripening: That is coarsening over time by change in droplet solubility or dispersion. Clearly coarser emulsions will create even coarser emulsions as this effect occurs. It is controlled by emulsifier (interfacial tension). So the main effect of particle size on emulsion stability is to create a coarser emulsion sooner. 2.2 Emulsion Break and Cure: There are three main steps(3): Break Coating and film formation Cure Break: That is the flocculation, coalescence within the emulsion and the reaction with the aggregate surface. The PS and PSD could have several effects here. Firstly on coalescence and flocculation, If the emulsion is already coarse the floccs will be large and create large coalesced particles. Such particles will have the charge spread over a larger particle and the charge density at the point of contact with the particle will be reduced, hence the contact angle reduced and the rate of reaction reduced thus the thermodynamic break rate will be reduced. The kinetic effect of access to the aggregate surface could be significant. That is, larger particles interfering with each other and flocculating and coalescing within the emulsion rather than reacting with the surface. This may give the appearance of a fast set but be more likely to create false slurries. Coating and Film Formation: That is the binder evenly coating all particles and forming a coherent film. The above kinetic effects will affect the coating of the particles and, as we are dealing with graded fine aggregates in slurry surfacing aggregates, a tendency to break on the larger surface area fines and not coating large particles at all. Coherent films require even coalescence on the particle surface and entrapped water will interfere with this, leading to a less cohesive mix. PSD is an important factor in film formation and a range of sizes that fit together will assist this process. Cure: That is the steady loss of water from the system and the stiffening of the total mix as cohesion increases. Entrapped water will clearly inhibit cure. Rejection of water from the aggregate surface is both a thermodynamic and kinetic effect. The thermodynamics relate to the energy differences between the emulsifier and the aggregate charge, the kinetic to the diffusion controlled loss of water through the coalescing binder. So the potential effects on performance are that coarse emulsions with wide PSD will be more likely to give poor coating, false slurries, and poor cure rate expressed as cohesion build up or traffic time. 3. CONTROLLING PARTICLE SIZE: Several papers have been written describing methods to improve particle sizing of emulsions by methods of formulation and adjustment of asphalt chemistry. (3,4). The methods usually involve improvement of chemical systems, doping of asphalt with surface active agents, tailoring asphalt composition and optimization of manufacturing conditions. The approach is basically to improve dispersion of the asphalt in the mill and to stabilize the resulting emulsion against early flocculation and coalescence. However the physical act of milling is the main determinant of initial particle size for a given asphalt and emulsifier system. The particle size is determined by the shear in the mill and the residence time (1,5) This can be expressed by: Shear Rate = (2 pi R V / 60 E) .............(1) Where: R = Colloid Mill Radius V = Rotation Speed (rpm) E = Gap Dimension That is the particle size is a function of mill diameter , gap and perhipheral speed. (1) and a correlation between this shearing and the d50 value, A correlation between PSD and the initial particle sizes in the mill has been reported (6). However mill configurations internally are quite different, with different tooling and effective gap sizes created by this. For this reason the relationship is changed to: Shear Rate = (2 pi R V / 60e) Mf.............(2) Where Mf is the mill factor. Where the mill factor is the increase or decrease in shear created by the mill configuration and tooling. This is estimated as follows. A very stable emulsion is made using a SS-1 type formulation and vinsol resin. The particle sizing is done optically using a microscope on fresh emulsion. ( less than 1 hour old). The d 50 particle size is compared to the theoretical from the graph above and compared to the actual. The reduction or increase is expressed as a ratio and this is the factor. Table 1 shows the factors for three mills used in this study. It is appreciated that this is only an estimate but the figures show the mill effects. Table 1 Mill Factors Estimated and shear rates. The main mills used in this study on microsurfacing effects were lab mills 1 and 2 and production mill 1 and 2. 4. EXPERIMENTAL 4.1 Quick Set Slurry- Latex Modified 4.1.1 Formulation Quick set Slurry.(latex Modified). A standard emulsifer was used initially, Roadchem® CQS, now called Roadchem®300. This was optimized to the local asphalt. Adjustments were made to the asphalt to give the best results. This work was carried out on the lab 1 mill. The resulting emulsion met the LMCQS-1h Caltrans specification and microsurfacing emulsion specification. The foreign customer repeated this formulation on lab mill 2. The results were as shown in table 2. Table 2 Comparison of Emulsion Properties Initially. LMCQS-1h type The difference was initially puzzling and other optimization techniques were attempted such as surfactant doping and use of higher temperatures in the mill. Doping with kerosene was also tried. This improved the emulsion to sedimentation in 5 days of 7%. In both cases the latex was milled with the emulsion. The production on production 1 mill compared well with the lab mill 1 having the same mill factor , the production mill 2 owned by the customer was better than the lab mill and mirrored the results of the optimized lab system. 4.1.2 Particle Sizing: An optical method was used as described elsewhere (3,7).This showed significant differences in particle sizing as may be seen in figure 1. Figure 1a Mill 1 Lab Figure 1 b Mill 2 Lab Figure 1 c Mill Lab Modified emulsion Figure 1 d Production Mill 1 Figure 1 e Production Mill modified Production of the base formulation showed similar sizing and distribution for lab as for production but far superior to the base formulation for lab mill 2. Production mill 2 was inferior to lab mill 1 but better than lab mill 2. Specification results could be achieved with a modified system and the production mill 2. 4.1.3 Mix work: ISSA standard mix design was used. Wet Cohesion was measured at intervals and a plot made of cohesion build up. A local ( foreign) type II aggregate was used and the emulsion content optimized at 16%. 0.25% aluminium sulphate retarder was required to get mixes with the coarsest emulsion from lab mill 2. Figures 2, 3, 4, 5 show respectively wet track abrasion, loaded wheel testing, mix time testing, cohesion testing. Samples were also examined for coating after manufacture. Figure 6 shows coating for lab 2 mix. The rest showed good coating. a) WTAT: The 1 hour tests were not sensitive to the particle sizing and similar results were achieved. The 6 day tests showed a significant improvement for finer particle sizing. The poorest coating leading to the worst result. ( Lab mill2). Figure 2: Wet Track Abrasion Data LMCQS-1h Type b) LWT: No sensitivity of deformation resistance was noted. However for the sand pick up test lab 2 was worst, perhaps due to the increased binder film thickness on finer aggregate and poor coating. Figure 3: Loaded Wheel Testing LMCQS-1h Type c) Mix time Testing: The coarser emulsions had shorter mixing times. This indicates poorer handling. As may be seen for elevated temperatures of 35C and 50C this effect got worse. Figure 4: Mixing Times d) Cohesion Testing: The coarser emulsions had a slower build up in cohesion value and tended to flatten out indicating trapped moisture. This was repeated at 35 and 50C ,and showed similar results. The coarsest gave worst results at elevated temperatures. Figure 5 Selected Cohesion Data 4.2 Microsurfacing Emulsion 4.2.1 Formulation of Microsurfacing Emulsion The same aggregate was used as it passed all ISSA requirements for microsurfacing. Roadchem ® 200 was used and SBR latex. The formulation was optimized and the same trends observed for the emulsion properties. Lab mill 1 produced an on spec material lab mill 2 did not. Modifications were made again and production mill 2 was able to make a spec product. ( although coarser and poorer properties than mill 1 lab ). One set of emulsions was made using SBS at 3% instead of the latex. The same trend was observed. 1.0% cement was used in formulations. The emulsifier was used as the retarder in a 5% solution. Table 3 Shows typical emulsion results. Table 3a Emulsion Results Table 3 b SBS Results 4.2.2 Particle Sizing Figure 7 shows a comparison of the particle sizing. Results were similar to those for the LMCQS-1h. Figure 7 a Lab Mill 1 Latex Microsurfacing Emulsion Figure 7 b Lab Mill 2 Latex Microsurfacing Figure 7c Lab Mill 1 S Figure 7d Lab Mill 2 S 4.2.3 Mix Results. Figures 9,10 11,12,13, 14 show respectively wet track abrasion, loaded wheel testing, mix time testing, cohesion testing and coating. SB was measured and only the Mill 1 type mixes and the modified mill2 production passed the requirement. a) WTAT: The 1 hour tests were not sensitive to the particle sizing and similar results were achieved. The 6 day tests showed a significant improvement for finer particle sizing. The poorest coating leading to the worst result. ( Lab mill2). Little difference was noted between latex and SBS. Figure 9: WTAT results for Microsurfacing b) LWT: No sensitivity of deformation resistance was noted. However for the sand pick up test ,lab 2 was worst, perhaps due to the increased binder film thickness on finer aggregate and poor coating. SBS and latex behaved very similarly. Figure 10: LWT test results Microsurfacing c) Mix time Testing: The coarser emulsions had much shorter mixing times. This indicates poorer handling. As may be seen for elevated temperatures of 35C and 50C this effect got worse. The SBS emulsion had worse mix times. Figure 11 Mix time testing d) Cohesion Testing. The coarser emulsions had a slower build up in cohesion value and tended to flatten out, indicating trapped moisture. The coarsest could not be stabilized sufficiently to give a good mix. This was repeated at 35 and 50C ands showed similar results. The coarsest gave worst results at elevated temperatures. Figure 12 Selected Cohesion Results e) Coating. It may be seen that the coarser emulsions gave poorer coating as did the SBS. But this may have been a compatibility issue for the SBS. Figure 13 Cured Coating Of Microsurfacing Mixes 4.3 Field Results The production mill 2 emulsions were compared in the field for both LMCQS-1h modified and unmodified and Microsurfacing modified and unmodified and contrasted with production mill 1. Table 4 shows the results for set times (walking) and traffic times as measured by a pick up truck and marking of the surface. Table 4 Field Results 5. CONCLUSIONS The results show that the mill type and configuration is a great influence on the particle size and particle size distribution of the emulsion. This is not new. It shows also that this can effect the performance of microsurfacings and slurries in the field and that this effect is temperature dependant. Emulsions must be optimized and the correct emulsifiers used but also high shear systems are needed. 6.0 REFERENCES Durand,G Piorier,J E (1996) AEMA International Symposium On Asphalt Emulsions Washington D.C. Booth,EH, Gaughan, G, Holleran, G (1994) Australian Road Research Board International Conf Perth. Holleran, G (1999) AEMA International Symposium on Asphalt Emulsions Washington D.C Holleran ,G (1999) International Slurry Surfacing Meeting Porta Villarta Province, R (1986) Workshop on Bitumen Emulsions Melbourne Australia Holleran US Patent 5,518,538 Holleran , G (1999) ISSA/AEMA Meeting Amelia Island Florida. Back to VSS Technology Papers Library Copyright © 2000 Valley Slurry Seal Co All Rights Reserved.