Bituminous Binders – Ageing Profile Test

A study on the ageing of bituminous binders

By Corina Hill, Martin Heslop, Iswandaru Widyatmoko, Richard Elliott and Donna James

As one of the main factors affecting the durability of asphalt pavements, the process of bitumen ageing in service has raised major interest over the years. Various laboratory techniques have been developed to artificially simulate the short-term ageing (occurring during the mixing and construction stage, mainly due to the loss of volatile components) and long-term ageing (occurring over several years in service, mainly due to oxidation) of bitumen.

Conventional ageing methods, such as the Rolling Thin Film Oven Test (RTFOT)² and Pressure Ageing Vessel (PAV)³, have been most frequently adopted to approximate in the laboratory the various stages of binder ageing/hardening. Currently, the empirical tests of penetration and softening point are specified for asphalt binders after RTFOT and data are being collected in order to specify both: rheological properties, such as complex shear (stiffness) modulus G* and phase angle ∂; and low-temperature characteristics, using the bending beam rheometer (BBR) and Fraass on binder as supplied and after ageing protocols.

Although RTFOT and PAV have been employed over the years, many concerns have been raised with regard to their suitability, especially for ageing of polymer-modified binders (PMBs) in the Superpave specifications⁴:

• During the RTFOT, the PMB sample tends to form a ‘surface skin’ that prevents the specification requirement for a ‘moving film’ of binder and, therefore, homogeneous ageing of the sample; one of the solutions being considered is to increase the temperature of test to 200°C, which will probably degrade modified binders.
• The PAV is a static test that can generate heterogeneous aged samples, and, for ageing of PMBs, the testing conditions may result in separation of the polymer. Oxidation only takes place at the surface of the sample as oxygen does not diffuse easily into bitumen (a few microns at normal pressure), hence the need for 2.1Mpa of pressure and high temperature. As the thickness of the sample is 3,200 microns, careful mixing of the residue prior to subsequent testing is critical, especially if a polymer skin has formed.

As a result, other ageing protocols have been investigated. Two candidate protocols with good potential due to their multi-purpose dynamic nature are:

• The Rotating Cylinder Ageing Test (RCAT), developed by the Belgian Road Research Centre, (EN 15323)⁵. Verhasselt⁶ suggested that adopting similar conditions to the RTFOT (163°C), but with an ageing duration of 235 ± 5min in the RCAT, can simulate the conventional RTFOT ageing (average value reported to be suitable for both paving grade and PMBs). The BitVal report7 states that 20h of PAV at 100°C was found to be equivalent to 176 ± 16h RCAT at 85°C and 125 ± 11h RCAT at 90°C. For some PMBs, however, a larger variability in the test duration has been observed for equivalent ageing, possibly due to polymer separation during PAV. The main advantages of the RCAT are the fact that only one apparatus is used, eliminating unnecessary handling/heating operations associated with conventional ageing methods, and a reasonable amount of homogeneous aged binder can be recovered for subsequent testing. However, the duration of the test, the cost and lack of laboratories with RCAT equipment represent the main disadvantages.
• The Ageing Profile Test, as specified in Clause 955¹, for binders used in asphalt manufacture. This dynamic laboratory ageing procedure using conventional equipment, currently used by most binder laboratories, provides short-, medium- and long-term, homogeneously aged samples in less than a day. Clause 955 also provides a rapid stabilizing and ageing protocol for polymer-modified emulsion binders⁸.

This paper compares the conventional ageing procedures RTFOT and PAV with Clause 955, by assessing the rheological properties of five binder residues, removed at various stages of ageing. The testing of both complex shear (stiffness) modulus and phase angle provides greater assurance that the aged residues can be compared. Measuring stiffness by only using penetration or softening point values may not pick up differences in elastic behaviour that could be the result of different levels of oxidation. The bitumen source, colloidal state and grade also determine the rate of oxidation, so it is likely that ageing profiles rather than end values will be important to characterize binders for the future. It is accepted that, at the temperatures of test, to accelerate ageing, chemical changes occur that would not happen in the road, but the balance of temperature vs time is a practical consideration.

Scope of work

For many years the conventional RTFOT method² has been adopted to approximate in the laboratory the short-term ageing of bitumen, which occurs during the manufacture, transport and laying of asphalt. As its first task, this paper compares the conventional RTFOT method against the Standard ‘Short Term Ageing Test’ described in Clause 955 of the MCHW¹, by assessing the rheological properties of the binder residues.

Considering the long-term changes (hardening in the road), the conventional PAV method³, developed during SHRP (from a UK test method for ageing road tar) is specified for paving grade bitumens and for PMBs. In the SHRP PAV, the sample, approximately 3.2mm in thickness, is statically exposed to the combined effects of elevated temperature (90°C, 100°C or 110°C) and pressure (2.1MPa) for 20h. The High Pressure Ageing Test (HiPAT or PAV85) was developed in the UK, based on the SHRP PAV, but adopted a lower testing temperature (85°C) and a longer exposure time (65h)⁹. This long-term ageing method is now included in the EN Standard test procedure³. A summary showing comparison between these test methods is presented in Table 1.

As the Modified Ageing RTFOT of Clause 955 has the capability to produce samples at various stages of ageing, it has the potential to provide an ‘Ageing Profile’ showing the changes during the life of the binder as well as at the end of life. This paper uses rheological data to examine the Ageing Profile and to establish a correlation between the different methods.

Samples for test

The study was carried out for five binders, specifically three Paving Grade (10/20, 40/60 and 100/150) and two PMBs (elastomeric and plastomeric).

Experimental programme

Conventional RTFOT (EN) compared with the Standard Short-Term Ageing Test (Clause 955)

The main differences between the two test protocols are that Clause 955 adopts polytetrafluoroethylene (PTFE) in place of the traditional glass bottles and introduces a stainless steel screw in each PTFE bottle, to stir and homogenize the sample during ageing (shown in photos 1 & 2). Consequently, due to the continuous exposure of fresh sample to air, the test duration is reduced (the rotating screws accelerate the ageing process and prevent the sample from creeping out of the bottle) and, in the case of testing PMBs, the problems related to ‘skinning’ are minimized.

For each binder, the following short-term ageing procedures were carried out:

• Conventional RTFOT (EN): eight glass bottles, each containing 35g of binder, were rotated in a vertically rotating shelf, while blowing hot air into each bottle, at 163°C for 75min.
• Standard ‘Short Term Ageing Test’ (Clause 955): eight PTFE bottles, each containing 19g of binder and a stainless steel screw, were subjected to similar testing conditions as above, at 163°C but for a shorter period of time (45min).

Sub-samples of short-term aged binders were collected for rheological characterization and long-term ageing was subsequently continued on the remaining samples.

Rheological characterization (temperature sweeps from 0°C to 80°C at 0.4Hz) was carried out using an advanced Gemini Dynamic Shear Rheometer (DSR), with 8mm flat-plate geometry and 1mm gap. Specimens were prepared and tested in accordance with the EN standard¹⁰. Table 2 presents a summary of selected data from the rheological assessment carried out for the short-term aged samples, specifically the complex shear (stiffness) modulus (G*) and phase angle (∂) at 25°C and 0.4Hz.

For the 40/60 and PMBs, the stiffness values of the aged residues obtained by the two methods were found to be comparable, while the 10/20 and 100/150 binders appeared to have aged more under the conventional RTFOT than under the shorter protocol. However, the differences between phase angle values obtained by the two protocols appear to be generally small.

Conventional PAV85 or HiPAT (EN) compared with the Modified Ageing Rolling Thin-Film Oven Test (Clause 955)

For each binder, the ageing was continued as follows:

• Conventional PAV85 was carried out by subjecting the sample post RTFOT (75min) to oxidation in a pressurized ageing vessel. The procedure involves ageing 50g of binder in a 140mm diameter tray (binder film approximately 3.2mm thick) in the heated vessel, pressurized with air at 2.1MPa, for 65h at 85°C.
• The Modified Ageing RTFOT (Ageing Profile Test) was carried out by subjecting the sample post Standard ‘Short Term Ageing’ (45min) to further ageing, at 135°C. Sub-samples were then removed at three stages of ageing: after 4h, 8h and 22h.

For simplicity, schematic diagrams of the ageing protocols are presented hereafter [see attached PDF].

The rheological characterization test results are presented in table 3 and figures 1 to 4.

Figures 1 to 4 show a progressive hardening of binder during the Modified Ageing RTFOT. The data presented in table 3 and figures 1 to 4 indicate a reduced test duration in the Modified Ageing RTFOT, estimated at 8h, in comparison to the PAV85 (65h).

Overall, the unmodified binders appear to have aged at a similar rate; however, the 10/20 binder displayed a relatively slower increase in stiffness during the long-term ageing. The rheological data for the unmodified binders demonstrate that the ageing/hardening obtained in the conventional PAV85 was slightly more severe than that following 8h in the Modified Ageing RTFOT of Clause 955, illustrated by the slight increase in the elastic response of the aged binders. Nevertheless, it can be concluded that, overall, the stiffness values of the PAV85 aged samples were found to be comparable to those after 8h in the Modified Ageing RTFOT.

For PMBs, it can be seen that during the intermediate ageing (4h), the plastomeric binder appeared to have a relatively lower resistance to ageing than the elastomeric binder, as shown by the steeper incline of the phase angle curve. However, during the long-term ageing, both PMBs appeared to have aged at a similar rate.

It was also observed that the conventional PAV85 method appeared to have hardened/aged the plastomeric binder more significantly than the 8h Modified Ageing RTFOT, while for the elastomeric binder the aged sample’s stiffness values were comparable. The differences between phase angle values obtained by the two protocols appear to be generally small.

EME2 binder specification

Hard paving-grade bitumen, 10/20, was included for test as it is currently of great interest for the manufacture of EME2 asphalt concrete, and at the present time there is a lack of data for this binder, caused by the large time period to produce aged residues.

The EME2 specification (Table 9/6 of Clause 930¹) requires the collection of rheological and low-temperature data after RTFOT and PAV. The Ageing Profile test (Clause 955), as described in this paper, is an alternative ageing protocol to RTFOT and PAV for EME2 hard paving grades 10/20 and 15/25, and it is hoped that this rapid protocol will allow data to be collected quickly and economically.

Viscous to Elastic Transition (VET) temperature has been defined as the temperature at a phase angle value of 45°, at which the elastic component of the complex shear (stiffness) modulus, G’, of a bituminous material equates to the viscous component, G”, (hence G’ = G” at VET). Both VET and G*VET (that is G* at VET) are among the binder characteristics to be reported under Table 9/6 of Clause 930¹. These parameters have been considered as a useful tool to demonstrate changes in the properties of bituminous materials at different levels of age hardening and/or distress level (cracked or uncracked asphalt road sites)¹². An example of typical cracks throughout the asphalt layers is shown in photo 3. As the level of distress increases (eg cracked areas), the VET temperature increases but the complex modulus (at the VET temperature) decreases, and vice versa. A previous study¹² reported that, in order to reduce the propensity for cracking in the road pavement, a good-quality 15pen bitumen should have a VET temperature (0.4Hz) less than 35°C and G*VET
(0.4Hz) greater than 5MPa during its service life.

The results for EME2 10/20 hard grade bitumen presented in table 4 show that, after 8h ageing in the Ageing Profile Test and after 65h in the PAV 85 test, the results are similar for equivalent penetration and VET data. On the other hand, the short-term ageing comparison data confirm that the conventional RTFOT is more aggressive than the Standard ‘Short Term Ageing Test’ (Clause 955), resulting in lower penetration, lower G*VET and higher VET. However, the results did not meet the VET and G*VET criteria; EME2 asphalt concrete manufactured from this binder is likely to have a reduced life.

Developments and future research

The problem experienced with PMBs in the conventional RTFOT (ie binder creeping out of the bottles) has been overcome by the use of screws manufactured from high-quality stainless steel. The direction of the screw is such that the binder is drawn to the closed end of the bottle during the ageing protocol. This new screw-tip configuration (work by Jacobs Laboratory) prevents the screw itself from winding out of the bottle. The rotation moves the binder continuously around the screw maximizing the air-binder interface. The problem experienced by some laboratories of variable-quality PTFE bottles becoming slightly deformed might be overcome by the use of PTFE-coated cans. It is very important in respect of repeatability to have precise dimensions and symmetry; this is also difficult to achieve in a glass bottle. In future work, PTFE-coated cans will be used together with lower test temperatures to minimize activation reactions and to investigate low-temperature and healing characteristics.

Conclusions

The following conclusions can be drawn from the assessment of ageing characteristics:

• The conventional RTFOT (EN) and the Ageing Profile Test (Clause 955) at 45min produce relatively comparable ageing results in terms of phase angle and complex shear (stiffness) modulus.
• The conventional PAV85 (EN) appears to yield aged samples with similar characteristics to those obtained after 8h in the Ageing Profile Test (Clause 955); therefore, the Ageing Profile test should be shortened from 22h to a maximum of 8h.
• The Clause 955 test protocol uses only one apparatus that is currently available in many laboratories and has the ability to carry out both short- and long-term ageing processes, eliminating unnecessary handling/heating operations.
• Adopting the Modified Ageing RTFOT, the long-term ageing test can be carried out in a much shorter period of time than is the case with the PAV or RCAT and the Ageing Profile plot provides more information about the rate of oxidation.
• EME2 hard grade binders can be quickly assessed in terms of VET temperature and the other specified rheological tests in Clause 930.

Acknowledgements

The authors gratefully acknowledge the Highways Agency, for commissioning and financially supporting the study, and the staff involved in carrying out the laboratory testing. The authors’ views expressed in this paper are not necessarily those of the organizations they represent.

REFERENCES

1. UK Department of Transport, ‘Specification for Highway Works’ and ‘Notes for Guidance on the Specification for Highway Works’, Manual of Contract Documents for Highway Works, vols. 1 & 2 (MCHW 1 & 2).
2. British Standards Institution, Bitumen and bituminous binders. Determination of the resistance to hardening under the influence of heat and air, BS EN 12607-1: 2007.
3. British Standards Institution, Bitumen and bituminous binders. Accelerated long-term ageing conditioning by a Pressure Ageing Vessel (PAV), BS EN 14769: 2005.
4. BAHIA, H.U., HANSON, D.I., ZENG, M., ZHAI, H., KHATRI, M.A., and R.M. ANDERSON: Characterization of Modified Asphalt Binders in Superpave Mix Design, NCHRP Report 459, 2001, Transportation Research Board.
5. British Standards Institution, Bitumen and bituminous binders. Accelerated long-term ageing/conditioning by the rotating cylinder method (RCAT), BS EN 15323: 2007.
6. VERHASSELT, A.: ‘Short and long term ageing of bituminous binders – simulation with the RCAT method’, Proceedings of the 6th International RILEM Symposium, Zurich, 2003.
7. FEHRL. BitVal – Analysis of Available Data for Validation of Bitumen Tests, BitVal Phase 1 Report.
8. WALSH, I.D., SHRUBSOLE, P., and M.F.W. HESLOP: Comparison of Ageing Test Methods for Polymer Modified Road Binders and a Review of the Rapid Recovery Test for Emulsion and Cut-back Binders, 3rd Eurasphalt and Eurobitume Congress, 2004.
9. HILL, C.L., CATT, C.A., and R.C. ELLIOTT: Assessment of the Reproducibility of a Long Term Ageing Test for Bituminous Binders, 3rd Eurasphalt and Eurobitume Congress, 2004.
10. British Standards Institution, Bitumen and bituminous binders. Determination of complex shear modulus and phase angle. Dynamic Shear Rheometer (DSR), BS EN 14770: 2005.
11. GERSHKOFF, D.: Polymer-modified bitumens – performance in empirical and rheological tests, 1st European Workshop on the Rheology of Bituminous Binders, Eurobitume, 1995.
12. WIDYATMOKO, I., HESLOP, M.W, and R.C. ELLIOTT: ‘The Viscous to Elastic Transition Temperature and the In Situ Performance of Bituminous and Asphaltic Materials’, Journal of the Institute of Asphalt Technology, Asphalt Professional, April 2005, no. 14, pp 3-7 (ISSN 1479-6341).