![Sample our Engineering & Technology journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5933/TOC-Subject-Sample-2021-Engineering-Tech.jpg"})
Abstract
Cold in-place recycling (CIR) is an asphalt pavement rehabilitation technology which is rapidly gaining popularity. This technique consists in milling the top 75–100 mm of an existing asphalt pavement, mixing the crushed material with a recycling agent, and reusing it to repave the roadway. Hence, there is a need to identify laboratory tests for developing CIR mixtures with improved resistance to cracking. This study investigates the applicability of semi-circular bend (SCB) and disk-shaped compact tension (DCT) tests to CIR mixtures. Tests were performed on five similar laboratory and field prepared mixtures. Since the CIR mixtures were relatively softer and more brittle than typical hot mix asphalt, minor modifications to the testing parameters and data analysis approach were found necessary. Fracture energies computed from SCB and DCT data resulted in being statistically similar and detected considerable differences among the five mixtures. Overall, both fracture tests showed good repeatability and potential applications in mix design engineering and optimisation of CIR mixtures, quality control and field validation usages.
Acknowledgements
The authors gratefully acknowledge Brown, Blue Earth and St. Louis counties for providing the material used for this study.
Disclosure statement
No potential conflict of interest was reported by the authors.
Discussion
BILL BUTTLAR: Thanks for a great presentation. It’s challenging to try to test these materials in any sort of test machine. Just a couple of comments on a statement that you made that fracture energy is a fundamental parameter. I would say that this total fracture energy is an engineering parameter, though if you model the test, whether it’s the SCB or DCT, and do an inverse analysis, you can attract local parameters for a failure model like a cohesive zone fracture model. We’ve done that. And if you do that, you’ll find that the local attracted size independent property for the two tests will often be the same. And then that shows that that’s a fundamental property that one could use in modeling. But from these practitioner tests, the fracture energy is just another engineering parameter, just useful. And that would be size dependent, so as you make a larger diameter specimen or a larger thickness, it will change the fracture energy. Even though you’re dividing by the fractured area there’s still a size effect. The temperature, the testing mode, and the size and the loading rate will all affect the fracture energy, and so you have to be careful and sometimes correction factors are needed to compare specimens of different sizes and rates, as you pointed out. But the results look great. And what we found for any one given test is that the size effect is really, if you plot it out, a function of the fractured area. And so you’d have similar results whether you had a different diameter or a different thickness as long as the area fractured is the same and you could compare specimens side to side. So nice work and very promising results.
EYOAB ZEGEYE: Thank you very much, Bill. I am indeed familiar with your excellent work on size effect of fracture energy. I have also done some work on the size effect of fracture energy measured from notched and un-notched SCB specimens of different sizes. In that study, we showed that the fracture energies measured from SCB test follow Type I or Type II size laws (depending on the presence or absence of a deep-notch ahead of the crack). We also showed that fracture energies determined from these tests could be analyzed using appropriate size effect laws to derive size-independent fracture energy and the effective size of the fracture process zone. These approaches are summarized in a previous paper, actually presented in a previous AAPT meeting: “Investigation of size effect in asphalt mixture fracture testing at low temperature” Zegeye, Le, Turos, and Marasteanu (Citation2012). So you are correct, in my sentence, I should have clarified that the fracture energies measured from these tests are size dependent. But, that ideally, fracture energy is a fundamental material property that is independent of stress-state, loading conditions and size. So, thank you for the clarification and for elaborating.
ILIYA YUT: I have a couple of questions on statistical inference in your paper. Why do you think the fracture energy computed from the DCT test produced better correlation with air voids in the specimens than that from SCB did?
EYOAB ZEGEYE: I can give you my opinion, but it is just that. I think the DCT is, first of all, a much more solid test. As I described in the presentation, the DCT is run at about 40 times faster than the SCB and uses thicker specimens. Furthermore, in DCT testing, the fracture energy is directly calculated from the load-CMOD curve obtained from the experiment. Whereas, in the SCB testing the tail part is predicted using the fitting-prediction models discussed in the presentation. The variance in the in the area under the tail of the SCB results may contribute to the high coefficient of variance observed in SCB test results. These could be some of the reasons. Although, I have heard arguments for the SCB due to the slow loading rate that may be more appropriate to capture the non-elastic behaviors of the tested material.
ILIYA YUT: Just to follow up, you cut your specimens from the same core, right?
EYOAB ZEGEYE: Yes, we did.
ILIYA YUT: So some of them were used for DCT and some of them for SCB, correct?
EYOAB ZEGEYE: Some of them for DCT and some of them for SCB (see picture in the paper or presentation).
ILIYA YUT: It’s already proven that air voids are not really uniformly distributed throughout the specimen. So would you think that the location of specimen within a core was the factor?
EYOAB ZEGEYE: Yes. Precisely to address that, as you can see here in the plot, some of the SCB specimens were taken from the bottom and some others were taken from the top, and similarly the DCT specimens. Through this approach we tried to somehow minimize and randomize the effect of air voids due to the compaction. Furthermore, the specimens were cored from 150-mm Superpave Gyratory compacted cores, instead of the more traditional 170 mm. We believed the variation of the air voids in 150-mm SGC cores was less pronounced than in 170-mm SCG cores.
ILIYA YUT: Thank you.
EYOAB ZEGEYE: Thank you.
GABRIELE TEBALDI: My compliments; you did excellent work!
EYOAB ZEGEYE: Thank you.
GABRIELE TEBALDI: I have a comment more than a question. I think that what you showed about the relation between performance and air voids clearly shows one of the main properties of this kind of material. We are talking about not a fully bound material like an HMA mixture but something that is in the middle between a fully bound and non-bound material. Somebody also calls these kinds of material bituminous stabilized material. If you think about when we use foamed asphalt, we use 1% to maximum 2% of bitumen and when we use emulsion, we use 2 to 3% – that means 1.5 to 2% of bitumen in the mixture. So it’s important when we approach this kind of material to remember what you are showing in this figure; it is something that is in the middle between a granular material and a traditional asphalt mixture and it is not like an HMA mixture at low temperature. It is something different. It is more important to remember that when we are looking at performance and the level of the air voids. Again, good job.
EYOAB ZEGEYE: Thank you for making that note. To add to your comment, at the beginning, we included in this project some CIR mixtures that were prepared using foamed asphalt. However, the results showed mixed behaviors and we decided to do more investigation. As you said, mixes that use foamed asphalt are much more brittle and sometimes have much higher air voids.
GABRIELE TEBALDI: More than brittle, I would say less resistant.
EYOAB ZEGEYE: Yes. Relatively speaking. Thank you.
JOHN HARVEY: I didn’t jump up fast enough to get in front of Gabriele because I want have you go to that same slide. But I’ll add a little bit onto that. We do a lot of CIR in California, and one of the main concerns is both the cracking resistance, as you’ve shown here really well, which is great. And then as you showed the picture of how this is used in the structure, so this is used on top of existing pretty impermeable HMA. And then we put a pretty impermeable HMA layer on top of it, and we have a layer in between there with 11 to 15% air voids. We have mostly volcanic aggregate in this state, and we’ve spent 15 years dealing with our moisture sensitivity issues. And now we’ve got something in there that if we have any cracking on the surface of that thin overlay or when it happens and any problems from the side, we’ve now basically gone back to a reservoir layer in between the impermeable HMA and that overlay. So even if the water gets in there, it’s going to have a hard time evaporating out. So, for me, I think your work was fantastic because it really emphasized the cracking resistance, which is going to speed up the cracking in that thin overlay. Plus, now once the water gets in there, the whole thing is going to basically … We’re back to those old problems that we had with poorly compacted HMA layers 20 years ago. So really good. Any comments on that or thoughts about that, how to interpret the results that you’ve gotten here? And follow up on that foamed asphalt question because I think that bringing those water contents down in those air voids may be the key.
EYOAB ZEGEYE: Thank you very much for the comments. I agree with you, air voids have a serious impact on CIR mixtures, especially because we’re looking at the large air voids, similar to those you mentioned, 11 to 16%. We see that the higher the air voids, your CIR is more prone to rutting and cracking. In this paper, we see that there is a very good correlation between the fracture resistance and the air voids of the mixes. We didn’t test different mix designs with different emulsion contents, but we do know from experience and from our database, when you increase the emulsion content, you do decrease significantly the air voids. This, in turn, will have a positive effect on the material’s resistance to fracture. The effect of emulsion content and type on the fracture resistance of CIR mixtures is something we need to look into. We have not yet done that job for this presentation, but we are looking forward to doithat.
JOHN HARVEY: Thank you.
EYOAB ZEGEYE: Thank you.