Mixed recycling with emulsion and cement of asphalt pavements : design procedure and improvements achieved

MIXED RECYCLING WITH EMULSION AND CEMENT OF ASPHALT PAVEMENTS. DESIGN PROCEDURE AND IMPROVEMENTS ACHIEVED R. Miró Recasens, F.E. Pérez Jiménez and S. Castillo Aguilar Technical University of Catalonia, Spain ABSTRACT Cold in place recycling, as an alternative to milling and standard replacement, shows a series of technical, economic and environmental advantages, which have done its use to grow in Spain, especially from 1990 onwards. Nevertheless, these actions have revealed certain aspects and limitations that prevent cold recycling from being considered a fully consolidated technique nowadays. The use of vaguely defined criteria for the design of the recycled mix, the lack of criteria for construction control as well as of specifications about the end product, and the ignorance of the evolution of the material performance with time have given cold recycling some technical insecurity. In this article, the problems arising from the standard methodology used for the design of cold recycled mixes have been analysed firstly, as well as the effect that certain variables, compaction and curing especially, have on their properties. Secondly, the effect that the use of both emulsion and cement during recycling has on the mix properties has been studied, comparing the results with those obtained by means of the standard procedures where only emulsion or cement are used. Finally, a new methodology for the design and control of the recycled mix has been suggested. RÉSUMÉ Le recyclage à froid en place comme alternative au fraisage et réposition conventionnel présente une suite d'avantages techniques, économiques et environnementaux, qui ont fait que leur utilisation en Espagne ait augmenté progressivement, surtout depuis 1990. Cependant, ces mêmes activités ont mis en évidence certains aspects et limitations qui empêchent le recyclage à froid d'être consideré, à présent, comme une technique de réhabilitation consolidée. L'utilisation de critères de dessin de l’enrobé recyclé peu définis, la manque de critères de controle d'exécution ainsi que de spécifications sur le produit final, et la méconnaissance de l'évolution du comportement du matériel avec le temps, ont entouré le recyclage à froid d'une certaine insecurité technique. Dans cet article on analyse, premièrement, les problèmes que pose la méthodologie que l'on utilise d'habitude dans le dessin des enrobés recyclés à froid et l'influence que certaines variables, compacité et endurciment spécialement, ont sur leur propriétés; en deuxième lieu, on a étudié l'effet sur les propriétés de l’enrobé de l'utilisation conjointe d'émulsion et ciment pendant le recyclage, que l'on a comparé avec celles obtenues lorsque l’on n'utilise que de l'émulsion ou que du ciment et, finalement, on a proposé une nouvelle méthodologie de dessin et controle de l’enrobé recyclé. 1 MIXED RECYCLING WITH EMULSION AND CEMENT OF ASPHALT PAVEMENTS. DESIGN PROCEDURE AND IMPROVEMENTS ACHIEVED R. Miró Recasens, F.E. Pérez Jiménez and S. Castillo Aguilar Technical University of Catalonia, Spain 1. INTRODUCTION The reuse of bituminous materials removed from aged pavements cannot be considered to be a new technique as far as maintenance and rehabilitation of pavements are concerned since its use, more or less widespread, began in 1973 as a consequence of the oil crisis. Cold in place recycling, as an alternative to milling and standard replacement, shows a series of well known technical, economic and environmental advantages: improvement of structural capacity and of pavement regularity, decrease in the use of aggregates and bitumen, lesser energy consumption, quick execution, waste disposal, does not cause smoke pollution, etc., while being a relatively inexpensive operation that provides a longer pavement service life, which is also environmentally friendly. Encouraged by these advantages, the use of cold recycling in Spain has been gradually growing, especially from 1990 onwards, emulsion recycling being the most widely used, accounting for a 2.4 x 10 6 m 2 of the recycled surface up to now, followed by cement recycling, with about 0.8 x 10 6 m 2 , Ruiz [1]. Nevertheless, these actions have revealed certain aspects and limitations that prevent cold recycling from being considered a fully developed and consolidated technique nowadays. The use of vaguely defined criteria for the design of the recycled mix, the lack of criteria for construction control as well as of specifications about the end product, and the ignorance of the evolution of the material performance with time have given cold recycling some technical insecurity, causing road administrations to distrust the use of this technique. 2 The aim of this article is to contribute to the clarification of some of these issues. It shows the results of a study about cold recycling of asphalt pavements, carried out at the Road Research Laboratory in the Transportation Department at the Technical University of Catalonia, following the monitoring of some of the actions undertaken in Catalonia (Spain). From this starting point, the problems arising from the standard methodology used for the design of cold recycled mixes have been analysed firstly, as well as the effect that certain variables, compaction and curing especially, have on their properties. Secondly, the effect that the use of both emulsion and cement during recycling has on the mix properties has been studied, comparing the results with those obtained by means of the standard procedures where only emulsion or cement are used. Finally, a new methodology for the design and control of the recycled mix has been suggested, which is compatible with and complementary to the current methodology. 2. ANALYSIS OF PROPORTIONING AND CONTROL PROCEDURES FOR THE RECYCLED MIX The standard procedure for the proportioning of the recycled mix is based on the Proctor test for determining the optimal compaction fluids content and in the standard test method for effect of water on compressive strength -immersion testfor the mechanic characterisation of the mix and the selection of emulsion content. After characterising the milled material removed from the aged pavement (gradation of milled material and extracted aggregate, percentage and characteristics of the original bitumen), the optimal percentage of compaction humidity is determined, by means of the Modified Proctor compaction test. Then, a coating test is performed with an initial percentage of emulsion and water, the addition of both being equal to the Proctor humidity content. 3 If the coating is good, specimens for the immersion test are produced -as appears in the Spanish norm NLT-162/84 [2]-, while increasing and decreasing the emulsion percentage in respect to the one initially fixed, but keeping the compaction fluids constant. After keeping the specimens in an oven at 60 ºC for 72 hours, they are tested at simple compression determining the dry and after water immersion (1 day at 60 ºC) resistance. The minimum emulsion percentage is established according to the simple compressive strength, which has to be over 1,500 kg, and to the retained strength, equal or over 75%, recommending not to have less than 2% by mass of residual binder even though the retained strength is over 75%, because the possible homogeneity differences in milling during the road works could provoke superficial stripping in those areas with a higher quantity of fines. Although these criteria, backed up by experience, have been used there is no specification about them, but in the last years different values according to traffic category have appeared as well as requirements on the value of strength after immersion, Fernández del Campo [3]. However, this procedure poses several problems. The first and most important one is that the optimal emulsion percentage is selected on a more compact material than the one achieved during works. The real density obtained from cores, extracted within 30 and 60 days after the execution of several asphalt layers in place recycled with emulsion, is about 93% of the value obtained with the corresponding specimens produced in laboratory, using the static compaction of 17,000 kg standard specified. Consequently, the properties of the mix produced in laboratory can be slightly different from the ones that the compacted mix will actually have. Fig. 1 shows the changes in compressive strengths of the recycled mix specimens in respect to the compaction energy used, noting that the most similar densities to the ones obtained during works (2.27 g/cm 3 on cores) are achieved with a 6,000 kg compaction load. The second problem that this proportioning procedure poses is that compressive strength is evaluated for a pre-established period of time and curing conditions, ignoring the strength degree that will be achieved immediately after the execution -which is a very 4 important value since the recycled mix layer is open to traffic once the compaction is overas well as at the end of the maturation period, once the water excess coming from the emulsion breakage is suppressed. Figs. 2 and 3 reveal that the strengths obtained on specimens cured in ovens at 60 ºC for 3 days are quite higher than those obtained at a very short term and fairly lower than the ones obtained at a longer term when specimens are cured at room temperature. Thirdly, the immersion test is useful for revealing the effect of water on the mix, adhesiveness, but it is rather ineffective in order to characterise it mechanically. The test is by no means the most appropriate for determining the ductile and tough cohesion provided to the milling by the bituminous emulsion; an increase in the emulsion content usually brings about a decrease in compressive strength, Pérez Jiménez [4]. On the other hand, the technique of cold in place recycling has in itself an additional disadvantage: the difficulty of controlling the density during works. The nuclear device usually controls the density during works, which should be aimed at being the highest possible. Nevertheless, with this procedure, the humidity measurement could not be accurate, given the fact that the bitumen existing in the old mix as well as the added one through the emulsion can be measured as water. Therefore, the recorded humidity is not real and could be about twice the real one. Consequently, the dry densities obtained through the nuclear device do not correspond to the real ones, being around 86 and 90%, depending on the different cases, of the ones obtained from cores. Besides, the densities measured in situ with the nuclear equipment evolve with time, as water coming from the emulsion breakage is lost, through evaporation or filtration, a more or less difficult process depending on weather conditions, thickness of the compacted layer and condition of the road base. By way of illustration, Figs. 4 and 5 represent the evolution with time of average dry densities and humidities, measured in situ on road C-1410, together with those obtained from cores extracted out of the recycled layer. Therefore, the dry density obtained in situ varies 5 from 1,860 g/cm 3 , after the layer compaction is complete, to 1,942 g/cm 3 after 40 days, a value which is still distant from 2,119 g/cm 3 obtained on cores extracted between 1 or 2 months after its construction, or from 2,280 g/cm 3 corresponding to the job mix formula in this case. The rest of parameters that are usually controlled during construction do not show so many problems. These parameters are: gradation of the milled material in order to check the milling process homogeneity, aggregate gradation after extraction and bitumen content to detect any variations in the existing mix in the aged pavement, and compressive strength, dry and after immersion, of the recycled mix in order to compare it with the one obtained in laboratory during the proportioning process. In view of these considerations, improving the mechanical characteristics of the mix is considered, especially in the first moments after its construction, by the incorporation of both emulsion and cement in the recycling, adding new tests and production procedures that allow for a better correlation between laboratory results and work results. This double treatment in place with both emulsion and cement for recycling asphalt pavement was used in 1992 by Sainton et al. [5, 6], with the purpose of simultaneously giving a higher rigidity to the typical solution of emulsion recycling and a higher flexibility to the treatment with cement, thus minimising the fissure processes of the latter. Engbers et al. [7] confirm this aspect and, moreover, reveal the good fatigue resistance of this mixed mix. Our study is mainly aimed at the analysis of the effect of this kind of treatment in the first stages of curing. 3. ANALYSIS OF THE MIXED MIX PROPERTIES The experimental study has basically consisted in determining the properties, and their evolution with time, of the cold recycled mix using both emulsion and cement, comparing them with the ones obtained when only emulsion or cement is used. 6 The compressive strength, the indirect tensile strength, and the dynamic moduli of five recycled mixes have been analysed: one with emulsion, two with cement and the other two with both emulsion and cement, made from the same milling material, removed from the asphalt pavement of road C-147, which was recycled with emulsion during the spring of 1996, Pérez Jiménez et al. [8] This material has been used with the same gradation employed during its construction in order to compare the laboratory results with those of the cores extracted. Table 1 shows the proportioning of each mix. The percentages of water added to mixes RE, RC-1, and RC-2 are the required ones to obtain the optimal compaction fluids in the Modified Proctor test. The aim has been to maintain these optimal compacting fluids for remixed RM-1 and RM-2, modifying them slightly in order to obtain a similar water/cement ratio. The specimens have been produced by static compaction, using a 6,000 kg load in order to obtain densities similar to those obtained during works. The curing of the specimens containing emulsion has been carried out at room temperature, whilst the ones containing only cement have been kept in a humid chamber for its curing. Simple compressive strength The evolution of the compressive strength of the mixed mix during the curing period, determined at 20 ºC according to NLT-161/84 standard [9], is quite similar to that of the emulsion recycled mix, although the higher the cement addition is, the higher the strength values are, as shown in Fig. 6. In the short term -seven daysthe compressive strengths of the emulsion and cement formulations become quite similar to those of the cement formulations for whatever cement content, and after 28 days the effect of a higher or lower cement content starts to be 7 perceptible. In the long run, from 60 days onwards, the cement recycled mixes are the ones showing higher strengths. Compressive dynamic modulus The compressive dynamic modulus, NLT-349/90 [10], has been determined at 20 ºC and at a 10 Hz frequency. The emulsion recycled mix shows the lowest dynamic moduli, and the lowest modulus increase during the curing period, going from 3,540 to 4,878 MPa in a 7 to 90 day period, as seen in Fig. 7. The mixed mix reaches higher moduli, and the higher the cement percentage is, the higher its increase during the curing period will be. Therefore, a mix with 3.5% by mass of emulsion plus 2.5% by mass of cement goes from 4,071 to 6,006 MPa, and a mix with 3.5% by mass of emulsion plus 3.5% by mass of cement goes from 4,484 to 7,907 MPa in the specified period. The cement recycled mixes show the higher moduli; the higher the cement percentage is, the higher the moduli increase during the curing period will be. Resilient dynamic modulus The evolution of the instantaneous resilient modulus, at 20 ºC and 0.5 Hz, NLT-360/91 [11], for the set of analysed formulations shows a tendency similar to the one previously obtained for the compressive dynamic modulus, but with lower values. The emulsion recycled mix reaches the lowest values, the cement recycled mix the highest ones, and the mixed mix shows an intermediate performance. The higher the percentage of cement is, the higher the increase of the resilient modulus during the curing period for the cement formulations will be. 14 LIST OF TABLES AND FIGURES Table 1. Proportioning of each of the analysed mixes. Figure 1. Variation of compressive strength of laboratory specimens with the compaction load. Figure 2. Evolution of compressive strength of laboratory specimens with time and curing type. Compaction 6 t. Figure 3. Evolution of compressive strength of laboratory specimens with time and curing type. Compaction 17 t. Figure 4. Evolution of in situ and core densities with time. Figure 5. Evolution of in situ humidities with time. Figure 6. Evolution of the compressive strength of laboratory specimens of the analysed mixes. Figure 7. Evolution of the compressive dynamic modulus, at 10 Hz, of laboratory specimens. Figure 8. Evolution of the instantaneous resilient modulus, at 0.5 Hz, of laboratory specimens. Figure 9. Evolution of the indirect tensile strength, at 5 ºC, of laboratory specimens. Figure 10. Variation of the indirect tensile strength at 5 ºC with the used emulsion percentage. 14 Figure 1. Variation of compressive strength of laboratory specimens with the compaction load. Figure 2. Evolution of compressive strength of laboratory specimens with time and curing type. Compaction 6 t. 0 0.5 1 1.5 2 2.5 3 3.5 4 0 2 4 6 8 10 12 14 16 18 20 COMPACTION LOAD (t) COMPRESSIVE STRENGTH (MPa) Dry Wet d= 2.264 g/cm 3 d= 2.391g/cm 3 d= 2.435g/cm 3 0 1 2 3 4 5 6 0 20 40 60 80 100 120 140 160 180 200 CURING PERIOD (days) COMPRESSIVE STRENGTH (MPa) S dry (curing 60 ºC) S wet (curing 60 ºC) S dry (curing 20 ºC) S wet (curing 20 ºC) 15 Figure 3. Evolution of compressive strength of laboratory specimens with time and curing type. Compaction 17 t. Figure 4. Evolution of in situ and core densities with time. 0 1 2 3 4 5 6 0 20 40 60 80 100 120 140 160 180 200 CURING PERIOD (days) COMPRESSIVE STRENGTH (MPa) S dry (curing 60 ºC) S wet (curing 60 ºC) S dry (curing 20 ºC) S wet (curing 20 ºC) 1.840 1.880 1.920 1.960 2.000 2.040 2.080 2.120 2.160 2.200 2.240 2.280 2.320 0 5 10 15 20 25 30 35 40 45 50 55 60 TIME (days) DENSITY (g/cm 3 ) COLD RECYCLING WITH EMULSION C-1410, section CARDONA-SOLSONA IN PLACE NUCLEAR DENSITY (to 50 mm) CORES DENSITY DESIGN DENSITY (IMMERSION TEST) 16 Figure 5. Evolution of in situ humidities with time. Figure 6. Evolution of the compressive strength of laboratory specimens of the analysed mixes. 0.0 2.0 4.0 6.0 8.0 10.0 12.0 0 5 10 15 20 25 30 35 40 45 50 55 60 TIME (days) MOISTURE CONTENT (%) COLD RECYCLING WITH EMULSION C-1410, section CARDONA-SOLSONA IN PLACE NUCLEAR MOISTURE CONTENT(to 50 mm) 0 0.5 1 1.5 2 2.5 3 3.5 4 0 10 20 30 40 50 60 70 80 90 100 CURING PERIOD (days) COMPRESSIVE STRENGTH (MPa) 3.5 E 3.5 E, 2.5 C 3.5 E, 3.5 C 5.0 C 7.0 C 17 0 2000 4000 6000 8000 10000 12000 0 10 20 30 40 50 60 70 80 90 100 CURING PERIOD (days) DYNAMIC COMPRESSIVE MODULUS (MPa) 3.5 E 3.5 E, 2.5 C 3.5 E, 3.5 C 5.0 C 7.0 C Figure 7. Evolution of the compressive dynamic modulus, at 10 Hz, of laboratory specimens. Figure 8. Evolution of the instantaneous resilient modulus, at 0.5 Hz, of laboratory specimens. 0 2000 4000 6000 8000 10000 12000 0 10 20 30 40 50 60 70 80 90 100 CURING PERIOD (days) RESILIENT MODULUS (MPa) 3.5 E 3.5 E, 2.5 C 3.5 E, 3.5 C 5.0 C 7.0 C 18 Figure 9. Evolution of the indirect tensile strength, at 5 ºC, of laboratory specimens. Figure 10. Variation of the indirect tensile strength at 5 ºC with the used emulsion percentage. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 10 20 30 40 50 60 70 80 90 100 CURING PERIOD (days) INDIRECT TENSILE STRENGTH (MPa) 3.5 E 3.5 E, 2.5 C 3.5 E, 3.5 C 5.0 C 7.0 C 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 12345 EMULSION CONTENT (%) INDIRECT TENSILE STRENGTH (MPa) Dry (curing 60 ºC, 3 days) Wet (curing 60ºC, 3 days) Dry (curing 60ºC, 1 day) Wet (curing 60ºC, 1 day)