Question: Has your lab noticed that, when you conduct T59, Section 7, three hours in the oven is not enough time to remove all of the water from most emulsion samples? What has your lab been doing to address this issue? Have you increased the heating time or stuck with the three hours?
- Pamela Turner – Assistant Research Engineer, NCAT
Answer: I’m glad you asked this question, Pamela. When we are confronted with a portion of the published standards that doesn’t work, the tendency is to think about the intent of the author. If the intent of T59, Section 7 is to remove all of the water from an emulsion sample, then it makes sense we would want to leave the sample in the oven for a sufficient amount of time to accomplish that. Unfortunately, in the case of this standard, there doesn’t seem to be any wiggle room for adding the extra time. We stick with the three hours and hope for the best.
Your question brings up the broader issue of testing variability as it relates to strictly following a potentially flawed standard versus “how everybody does it.” Your question also led to another article in this edition.
Clean and shiny lab floors
We hear many comments on how our binder lab floors are so clean and shiny. The comments range from “Wow! Your floors sure are clean and shiny!” to an incredulous “Wow! You obviously don’t test asphalt binder in here.”
Yes, we do test asphalt binder in our binder lab; a whole lot of it, in fact. So, how do we keep our floors so clean that we don’t worry about tracking home bitumen bits? I’ll try to explain in a clear and concise manner – a Swiffer and paper towels. Tadaaa!
No need to buy those expensive Swiffer pads. Just use good quality paper towels instead. Use a small squirt of mineral spirits or a comparable solvent and a few swipes later and the binder is gone.
Consistency is key, as well. We have them positioned throughout our lab for easy access. But if you let your floors go long enough, a Swiffer will be powerless to attack the built-up grunge.
Kari Wagner – Binder and Emulsion Technician, Heartland Asphalt Materials
“I test binders and emulsions, but a big part of my winter is spent formulating PMAC for Alaska and emulsions for Heartland.
I really enjoy formulating and being able to work with different materials. This year, we had a big change to our asphalt so I also spent part of the spring running formulations. It is so rewarding to find that formulation that hits the specs, and hopefully, saves some money in the process!
So basically, I not only do testing but also get to formulate and solve problems when issues arise. I feel very fortunate that I get to work with so many companies and so many wonderful people – I definitely enjoy what I do.”
I became certified through the NBTC program in my first year of asphalt binder testing experience. The program helped reinforce what I learned in training, and I was also able to meet other technicians who shared some great tips and experiences. I truly believe this helped me gain my footing in this industry, and also allowed me to network with other technicians whom I still am in contact with. I will definitely be returning for re-certification in 2021, and look forward to working with the Asphalt Institute again.”
Sizzle, pop, bubble – when a standard lets you down
You scanned the list of required tests for the PSP emulsion sample and your eyes stopped at AASHTO T59, Section 7. Gulp. THAT one again – the test designed to slap you with a low Z score. Emulsified Asphalt Residue by Evaporation couldn’t be more straightforward; pour a little emulsion in a beaker and heat to remove the water. Easy peasy. If you can nail the BBR and DSR, this one ought to be a piece of cake, right?
No! Not a piece of cake. The standard requires the samples are heated for two hours, stirred and heated for an additional hour. They are then removed from the oven for residue testing. So, what’s the big deal? If you can grasp the concept of a phase angle and an m-value, surely you can’t screw this up.
Most of the time, we see variability accumulate in test results due to operator error, whether from improper sample handling, sample prep or misinterpretation of the standards (a polite way to say “not actually reading them”). But, once in a while, variability comes from the way a portion of the standard is written. In the case of T59, Section 7, it is common that water is still present in the samples after the three hours of heating. In fact, just for fun, we recently repeated section 7 on our PSP sample and continued heating until the water was gone. It took SEVEN hours.
T59, section 10 (demulsibility) states “Repeat the heating and mass determination until successive masses differ by no more than 0.1g.” This ensures that all water has been removed from the sample before completion of the test. Section 7, on the other hand, has no provision for additional heating time if water is still present. This impacts section 18 as well because Storage Stability of Emulsified Asphalt includes section 7 as the means for removing water from the top and bottom sections. So, what do you do if water is still present after the prescribed heating time?
Many technicians nationwide use the ‘sizzle-pop-bubble (SPB)’ method (I just made up that name). If, after the three hours of heating, they can still hear sizzling and popping sounds and see bubbling, they stir again and increase the heating time until the sample shuts up and relaxes. A quiet, still sample usually means a dry sample.
So, back to those low Z scores. Nobody likes them. All that extra work explaining your testing sins to AASHTO can be a drag. So, what do you do when you know that, by following the standard as-written, your PSP results will probably suffer? If you end the heating time at three hours as stated in T59, Section 7 and many other labs are extending the time to ensure all of the water is removed from the asphalt residue, it stands to reason that you will not generate reproducible data with those labs.
There have been similar instances over the years of “follow-the-standard-at-your-peril” such as the heating time for Direct Tension molds in AASHTO 314 (ASTM D6723). Heating them for the stated three-minute maximum would ensure that your stress at failure averages would be lower than those of the labs who increased the heating time to five minutes.
In these cases, it seems it’s “pick your poison” – perform the test exactly as stated and increase variability while decreasing your Z score or ‘fix’ the problem and violate the published standard. There is no easy solution to this problem but this much is true: If testing procedures vary from lab to lab, the data generated between those labs also vary.
So, as in the question asked in this email, what do you do when you know there is still water in the sample? The only correct answer is to follow the test method as-written and then participate in the formal process of improving it.