The Rolling Thin-Film Oven (RTFO) procedure (Figure 202) provides simulated short term aged asphalt binder for physical property testing. Asphalt binder is exposed to elevated temperatures to simulate manufacturing and placement aging. The RTFO also provides a quantitative measure of the volatiles lost during the aging process.
The basic RTFO procedure takes unaged asphalt binder samples in cylindrical glass bottles and places these bottles in a rotating carriage within an oven. The carriage rotates within the oven while the 325°F (163°C) temperature ages the samples for 85 minutes. Samples are then stored for use in physical properties tests or the PAV.
The standard Rolling Thin-Film Oven test is:
One of the basic tenets of the Superpave PG binder specification is that tests should be as closely tied with field performance as possible. Seeing as the constituent asphalt binder in HMA undergoes significant aging during the manufacturing and placement processes (Figure 203), a method to simulate the aging is important in investigating and predicting early age HMA pavement behavior and distresses. Specifically, the Superpave PG binder specification calls for short term aged asphalt binder to be tested at high temperatures to determine fatigue and rut resistance.
Although many different factors contribute to asphalt binder aging, the key component of concern for the RTFO is the loss of volatiles. The loss of smaller molecules from the asphalt binder, often termed “volatiles” increases an asphalt’s viscosity.
Asphalt binders typically lose volatiles during the manufacturing and placemen processes. The elevated temperature of these processes ages the asphalt binder by driving off a substantial amount of volatiles. Field tests have shown that in-place asphalt binder does not lose a significant amount of volatiles over its life (Corbett and Merz, 1975[1]).
The RTFO aging procedure is used to simulate aging during mixing and placement, while the PAV aging procedure is used to simulate aging during in-service life. Therefore, asphalt binder tests concerned with mix and placement properties (such as the DSR) are conducted on RTFO aged samples, while asphalt binder tests concerned with in-service performance (such as the DSR, BBR and DTT) are performed on samples first aged in the RTFO and then in the PAV.
The RTFO has problems with highly viscous binders (e.g., some polymer modified asphalt binders and PG 70-XX and higher) because they do not flow properly in the bottles as they are rotated.
The RTFO was developed as an improvement to the Thin-Film Oven Test (TFOT) for short term asphalt binder aging. The TFOT placed asphalt binder samples in shallow pans (of the same dimensions as those used for the PAV) and then heated them in an oven for an extended period of time to accomplish simulated aging. The RTFO is an improvement over the TFOT because:
The following description is a brief summary of the test. It is not a complete procedure and should not be used to perform the test. The complete procedure can be found in:
Unaged asphalt binder is placed in a cylindrical jar, which is then placed in a carousel inside a specially designed oven. The oven is heated to 325°F (163°C) and the carrousel is rotated at 15 RPM for 85 minutes. The carousel rotation continuously exposes new asphalt binder to the heat and air flow and slowly mixes each sample. Figure 205 shows major RTFO equipment.
3 hours from sample preparation to scraping of final bottle.
Any scraping tool may be used but an average of 90% of the residue must be removed from the sample bottles. It has been found that circumferential scraping is usually more effective than lengthwise scraping.
Mass change of a sample as a percent of initial mass.
The RTFO is primarily used to simulate short term asphalt binder aging for use in other tests.
Table 19: Performance Graded Asphalt Binder RTFO Specification
Material | Value | Specification | Property of Concern |
---|---|---|---|
Unaged binder | Mass loss1 | ≤ 1.0% | None |
Although some samples can gain weight due to the oxidative products formed during the test (Roberts et al., 1996[2]), there is currently no limit on mass gain.
Typical mass loss is in the range of 0.05 to 0.5 percent.
Calculations can be done for the mass change bottles in the following form:
Where: