U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

Public Roads
This magazine is an archived publication and may contain dated technical, contact, and link information.
Public Roads Home | Current Issue | Past Issues | Subscriptions | Article Reprints | Author's Instructions and Article Submissions | Search Public Roads
Publication Number:  FHWA-HRT-17-006    Date:  September 2017
Publication Number: FHWA-HRT-17-006
Issue No: Vol. 81 No. 2
Date: September 2017


What's in Your Asphalt?

by Terence S. Arnold

Recently, State highway agencies and FHWA were amazed by a discovery: The clandestine use of re-refined engine oil bottoms in asphalt is widespread.

Simon Hesp, Queen’s University (Kingston, Ontario)
Photo. A two-lane road with asphalt pavement.
The asphalt pavement on this segment of Route 655 in Ontario, Canada, does not contain re-refined engine oil bottoms and shows no cracking after 9 years of service. However, a similar stretch of the same highway that contains REOB is showing significant cracking. The Chemistry Laboratory at TFHRC is developing a test method to analyze for REOB in asphalt mixes.


Asphalt is the sticky black residue that is left over from the processing of crude oil. It has been used in paving for more than a hundred years. When asphalt first came into use, oil refiners would give it away. Today, however, it is a highly traded commodity that demands premium prices. These prices have increased dramatically. In 2002, asphalt sold for approximately $160 per ton. By the end of 2006, the cost had doubled to approximately $320 per ton, and then it almost doubled again in 2012 to approximately $610 per ton.

Asphalt is remarkably efficient, making up only 4 to 5 percent by weight of the pavement mixture. The asphalt, which serves as the pavement’s binder, is also the most expensive part of the cost of the material for paving roads. The weight of an asphalt pavement varies depending upon the aggregate type, the asphalt, and the air void content. Using an average example of 112 pounds per square yard per inch of thickness, a 1-mile (1.6-kilometer)-long, four-lane highway with a 4-inch (10-centimeter) lift and 12-foot (3.6-meter)-wide lanes weighs about 6,300 tons (5,700 metric tons). Of this, the approximately 6,000 tons (5,400 metric tons) of aggregate at about $7 per ton costs $42,000. The 300 tons of asphalt in 2002 would have cost around $48,000. By 2006 this would have increased to $96,000 and by 2012 to $183,000. That is an increase of about $135,000 for every mile of highway in just 10 years.

The rising price of asphalt had a major impact on the cost of constructing pavements, which increased interest in finding ways to reduce costs. Methods to reduce costs include minimizing the amount of asphalt in the mix, increasing the use of reclaimed asphalt pavement (RAP), and replacing part of the asphalt with lower cost additives. RAP already contains asphalt, albeit aged material that does not have the same properties of fresh asphalt.

During a hallway conversation at a 2010 technical meeting, Matt Mueller, then a State engineer of materials from Illinois, revealed that his department of transportation had found phosphorous in one of the asphalt binders it was purchasing. Illinois specifications do not allow binder modification by use of polyphosphoric acid (PPA). The vendor denied adding PPA, but declined to reveal what had been added to the binder. When pressed by the department of transportation, the vendor revealed that it was adding what it called an asphalt extender—now known to be re-refined engine oil bottoms (REOB). REOB contains a small amount of phosphorus, which is what department chemists initially identified.

Nobody, not at any of the State highway agencies nor at the Federal Highway Administration’s Turner-Fairbank Highway Research Center (TFHRC), had ever heard of REOB. “No one knew this material was being added to asphalt, had seen any research on how this might affect performance of hot-mix asphalt pavements,or knew for how long and how widely it was being used throughout the country,” says Mueller. After discussions at the technical meeting, he says, “It quickly went from being just an issue in Illinois to becoming a national and international concern.”

Part of the mission of the Chemistry Laboratory at TFHRC is to develop new test methods. Developing a test method to analyze for REOB became a research project.

Testing Asphalt

The properties of asphalt binders vary widely depending on the source of the crude oil and the refining process used. For low winter temperatures, softer asphalts are necessary to avoid cracking. To prevent rutting in hot weather, the asphalt must be stiffer. The original test for determining the stiffness of asphalt was for the tester to chew it. However, today’s methods are much more sophisticated.

A dynamic shear rheometer, shown here, measures the viscoelastic properties of paving asphalt. Operators place the material they are measuring between the two quarter-sized plates.
A dynamic shear rheometer, shown here, measures the viscoelastic properties of paving asphalt. Operators place the material they are measuring between the two quarter-sized plates.


A machine called a dynamic shear rheometer (DSR) was introduced to the industry during the Strategic Highway Research Program’s research project, which ran from 1987 to 1992. The DSR is now the industry standard for measuring the viscoelastic properties of paving asphalt. However, the machine was not developed for the paving industry. The application of DSR was adapted from the food, cosmetic, and pharmaceutical industries, which used it to measure the stiffness of materials at different rates of shear. For example, the DSR enables product developers to create toothpaste with the right consistency so that it can be squeezed from a tube but not fall off the toothbrush.

The DSR tests binder placed between two parallel plates about the size of a quarter. One of the plates moves and the machine measures the viscoelastic properties of the asphalt. The DSR is used to determine the maximum high temperature performance grade (PG) in degrees Celsius. These temperatures increase in steps of 6 degrees and are typically PG 52, 58, 64, 70, 76. They provide a maximum service temperature for the pavement. For example, a PG 70-28 binder would have a maximum service temperature of 70 degrees Celsius and a minimum service temperature of minus 28 degrees Celsius.

The addition of soft materials to asphalt will reduce the high temperature grade (for example, from a PG 76 to a PG 70). Several additives have been evaluated by industry and academia, including used frying oil, residues from corn stover, and even treated swine manure, for this purpose.

Similarly, the high temperature grade can be increased by adding something that makes the asphalt stiffer (typically polymers like styrene-butadiene-styrene polymers), but they are very expensive.

What Is REOB?

Companies collect the waste engine oil drained from cars, then process, or “re-refine,” it for reuse. In simplified terms, they remove the oil by vacuum distillation. The lubricating oil distills over in a vacuum tower and is reused. The recovered oil meets all the automotive industry specifications for fresh lubricating oil. The process, however, leaves behind a residue at the bottom of the vacuum tower that goes by a variety of names. For the purposes of this article, it is re-refined engine oil bottoms (REOB).

The oil in a car engine is not just oil. It contains a variety of additives to enhance the vehicle’s performance. These include polymers, viscosity modifiers, heat stabilizers, additional lubricants, and wear additives. The REOB contains all the additives that were in the waste oil as well as the wear metals from the engine (mainly iron and copper). These additives include zinc dialkyldithiophosphate, which contains zinc, sulfur, and phosphorous; calcium phenate, which contains calcium; and molybdenum disulfide, which contains molybdenum and sulfur.

During an oil change, the waste engine oil drains from a car, as shown here
During an oil change, the waste engine oil drains from a car, as shown here.


Analysis of liquid asphalt for the trace metals calcium, copper, zinc, and molybdenum provides a measure of the amount of REOB present. Sulfur and iron could also be analyzed, but because they occur naturally in asphalt, their use would confuse the analysis.

Testing at TFHRC

The FHWA researchers at TFHRC chose the method of x-ray fluorescence spectroscopy (XRF) for their analysis of REOB. They selected XRF because they already had the machine in-house and also because many State highway agencies already have XRF for analyzing cement. Other methods, such as inductively coupled plasma and atomic absorption spectroscopies, likely would work equally well. The basic principles of the XRF analytical method are available in the 2015 Transportation Research Board (TRB) paper titled “The Analysis of Asphalt Binders for Recycled Engine Oil Bottoms by X-Ray Fluorescence Spectroscopy.”

Because REOB is a waste product, its composition varies widely not only between producers but also between samples from the same producer on different days. The compositional analysis is also affected by the asphalt into which it is blended. However, by making many blends using different REOB samples and different asphalt binders, the variations largely can be averaged out.

Several States provided samples of known REOB composition to TFHRC researchers, who analyzed the samples to compare the percentage of added (known) REOB to the found (tested) amount. The analyses showed a comparable percentage of added and found REOB.

In addition, the researchers contacted State highway agencies to request samples of asphalt binders for testing. They received an overwhelming response. The TFHRC researchers analyzed 1,532 samples from 40 States, one Canadian province, and two Federal Lands Highway divisions. They analyzed each sample twice—amounting to more than 3,000 analyses. None of those States realized that the asphalt they were buying contained REOB. One State insisted its samples had no REOB. However, 38 of the first 90 samples from that State contained it.

Of the 1,532 samples tested, 12 percent contained REOB, and some contained appreciably high levels of it at 10–20 percent. The highest level was 34 percent in a sample from Texas, which TxDOT had used in a patching compound. This testing also revealed the presence of phosphoric acid in 11 percent of the samples, and 2 percent contained ground tire rubber.

The results of the study caused a high level of interest throughout the country. Two years ago at TRB’s annual meeting, the Federal researchers held an REOB workshop and presented the findings of their laboratory evaluations to a standing room-only crowd.

Round-Robin Testing

Although some agencies do not specifically ban REOB, they do impose physical tests that preclude its use—effectively a ban. Others do not ban it by specification, but have agreements with asphalt suppliers to avoid the use of REOB. Of the 50 States and Washington, DC, the 3 Federal Lands Highway divisions, plus Ontario, Canada, nearly half specifically or effectively ban the use of REOB, and the majority of the rest do not specify whether it is or is not allowed. A handful do allow REOB, some within certain limits. For example, Ohio and Texas limit levels to less than 5 percent of the asphalt.

Photo. A gloved hand pours the recycled engine oil bottoms, which is a liquid, from one cup to another.
At room temperature, REOB is a liquid.


To develop a reliable test method that all States can use, the TFHRC researchers set up a round-robin test plan. The participants are 11 State highway agencies (Illinois, Massachusetts, Minnesota, Mississippi, Montana, North Carolina, Oklahoma, South Carolina, Texas, Vermont, and Wyoming), 2 independent testing labs, the Ministry of Transport in Ontario, Queen’s University in Ontario, and an Ontario paving contractor.

To execute the plan, TFHRC provided the initial test method and 45blends of various REOB modified asphalt binders with REOB concentrations of 2, 5, 8, 10, and 20 percent. In total, the researchers prepared and shipped 720 blends.

The participants are testing the samples independently using the guidelines provided by the TFHRC researchers. The round-robin testing is nearly completed, and TFHRC is in the process of collecting the results. The output will be a proposed AASHTO test method that any State can adopt and use.

REOB and Pavement Life

The unanswered question that remains is whether REOB negatively influences pavement life. In the United States, very little evidence is available, perhaps because no State highway agencies knew their binders contained REOB until recently. However, research in Canada linked the premature failure of Highway 655 in Timmins, Ontario, with the presence of REOB.

Analysis of Samples from State Highway
Agencies with Known REOB Content
Source Percent REOB Added
Percent REOB Found
(Tested) In Two Analyses
Illinois DOT 5 5, 5
Illinois DOT 10 9, 9
North Carolina DOT Vendor 10–11 14, 13
North Carolina DOT Vendor 10–11 9, 9
New York State DOT 0 0, 0
New York State DOT 6 6, 5
New York State DOT 0 + SBS (styrenebutadiene-
styrene polymers)
0, 0
New York State DOT 2 + SBS 2, 2
New York State 6 + SBS 5, 5


Summary of REOB Analyses of 1,532
Samples of Binders Sent to TFHRC
State Number of
REOB Ground
Tire Rubber
Range (%)
AK 5 0 0 0   Banned1
AL 20 0 2 0 4–10 Banned
AR           Unspecified
AZ 35 9 23 10 1–10 Unspecified
CA 4 1 0 4   Unspecified
CFL 135 12 20 0 10–18 Unspecified
CO           Banned
CT 16 2 0 0   Banned
DC 9 0 3 0 2 Unspecified
DE           Unspecified
FL 11 0 0 4   Allowed
GA 38 0 0 0   Banned
IA           Banned
ID 9 0 3 0 6 Unspecified
IL           Banned
IN 10 3 3 0 9 Allowed
KS 21 0 0 0   Unspecified
KY           Unspecified
LA 6 0 0 4   Unspecified2
MA 16 10 2 0 9 Banned
MD 15 7 0 0   Banned
ME 8 5 0 0   Banned
MI 72 12 0 0   Banned
MN 11 4 1 0 1 Allowed
MO 48 2 1 0 3 Unspecified
MS 2 0 2 0 1 Unspecified
MT 23 0 0 0   Unspecified
NC 70 1 6 0 14 Banned
ND 10 2 0 0   Allowed
NE 30 0 2 8 2 Allowed
NH 19 6 1 0 1 Banned
NJ           Banned
NM           Banned
NV 5 3 1 0 2 Banned
NY 68 3 5 0 6 Banned
OH 13 1 0 0   Allowed3
OK 41 4 13 0 7–12 Unspecified
ONT 14 2 5 0 4–17 Banned
OR 8 3 3 0 1–3 Allowed
PA 73 10 4 0 3–15 Unspecified4
RI 5 2 0 0   Banned
SC 14 0 0 0   Allowed
SD 16 0 0 0   Unspecified
TN 12 0 0 0   Unspecified4
TX 86 6 18 4 2–34 Allowed3
UT 20 6 2 0   Unspecified5
VA 21 1 1 0 2 Unspecified
VT 15 0 1 0 0–6 Banned
WA 433 50 58 0 12 Unspecified1
WFL 12 0 0 0   Unspecified
WI           Unspecified
WV 33 0 1 0 3 Unspecified2
WY           Unspecified
Total 1,532 167 181 34    
%   11 12 2    
1 Physical test precludes use
2 Do not want it
3 5-percent maximum (only in base and intermediate courses in Ohio)
4 Considering a ban or max limit in future
5 Does not specify the use of PPA or REOB but has a SHRP Plus Specification including the DTT (direct tension test)
with a minimum 1.5-percent strain that helps to control these additives


The overnight temperature in the area can reach as low as -40 degrees F (-40 degrees C). The pavement without REOB on one segment of Highway 655 showed no distress after 9 years of service. The pavement with REOB, which is located 0.6 mile (1 kilometer) from the pavement without REOB, has identical subgrade, traffic density, and climate. However, the segment of Highway655 with 5 to 10 percent REOB showed significant cracking. In this example, the presence of REOB was the identified cause of cracking at a low temperatures.

“The performance of the various sections of test road in Timmins illustrates the effect it has had on the pavement life,” says Simon Hesp, professor of chemistry at Queen’s University in Kingston, Ontario. “In our experience in Canada, even small quantities of 2–3 percent can be a problem.”

Similarly, a section of test pavement in Minnesota (MN1-4) found to contain REOB also cracked prematurely. The pavement performed well for the first 3 to 4 years, but then started to crack. This pavement is also subject to low temperatures.

The TFHRC researchers carried out a few mix tests (mixing the binders with aggregate) in 2015. The tests were not extensive, but they showed that at levels of 6 percent or more, the tensile strength of the asphalt dropped significantly. At a level of 3.5 percent REOB, the variation in the physical test methods was greater than the effect of REOB. In fact, it was difficult for researchers to assess whether REOB was present.

Some evidence suggests that the presence of REOB may be detected using the bending beam rheometer. One binder parameter considered is the difference between the low temperature critical specification temperature for stiffness (S) in the bending beam rheometer and the bending beam rheometer creep slope (m-value) noted as ΔTcritical. ∆TC = TC (S) – TC (m-value). Evaluation of this parameter is still ongoing.

Two independent study teams, one from AASHTO and the other from the Asphalt Institute, concluded that more research is needed on the use of REOB in asphalt.

A New Perspective On Binders

At TFHRC, researchers are planning a different way of looking at asphalt binders. Previously, all asphalt testing measured engineering properties such as stiffness. These tests do not show what materials had been added to the asphalt.

One sample received during the TFHRC study had a very strange analysis. The sample had the following test results: Superpave® PG 64-28 with a high temperature grade of 67.3 ΔTcritical on the bending beam rheometer was 6.7 degrees Celsius. Chemical analysis indicated it contained approximately 1.7 percent phosphoric acid, 10 percent ground tire rubber, and 19 percent REOB. The addition of 1.7 percent phosphoric acid likely would make the asphalt very stiff. Ten percent ground tire rubber would make it even stiffer. Then 19percent REOB would soften it and bring it back within specification.

Although it passed the standardized AASHTO testing protocols, it failed the Hamburg physical rut testing “miserably” (in the researchers’ words). The results were not surprising because nearly 31 percent of the binder was not asphalt.

These results demonstrate there are weaknesses in the standardized engineering testing protocols that may be exploited. The producer may have an economic benefit and the product passes all the standardized tests, but the product may not be beneficial to ensuring long-term performance.

Photo. Shown here are several rows of small containers with various blends of recycled engine oil bottoms. FHWA sent out the blends to participants for round-robin testing.
This is the sample deck of the XRF spectrometer.


To address this issue and the expansion of new asphalt additives and extenders, TFHRC is starting a research program to use handheld spectroscopic devices, x-ray fluorescence spectroscopy, and Fourier transform infrared spectroscopy to enable analyses to be done in the field rather than having to take samples back to the lab. Fourier transform infrared spectroscopy can even find lime in the mix, as well as styrene-butadiene-styrene and styrene-butadiene rubber polymers. X-ray fluorescence spectroscopy can find REOB and phosphoric acid, and the handheld spectroscopy works for spot checks. These instruments can be preprogrammed and require no additional training or skills for operators. All of this testing can be done directly from the paving machine, or at the asphalt plant by an unskilled operator, saving time and associated costs. These methods are much more difficult to manipulate because they can almost always tell what materials have been added to the mix. They also enable the possibility of field spot checks and eliminate the possibility of sampling errors where the asphalt being used was not the same as received by the testing lab.

Simon Hesp, Queen’s University (Kingston, Ontario)
The asphalt pavement on this segment of Highway 655 in Ontario was made with a binder that contains REOB and styrene-butadiene-styrene polymer. After 9 years of service, it shows significant cracking. This section is located just 0.6 mile (1 kilometer) from an identical subgrade exposed to the same traffic and climate but that does not contain REOB. The section of highway without REOB shows no cracking after 9 years of service.
The asphalt pavement on this segment of Highway 655 in Ontario was made with a binder that contains REOB and styrene-butadiene-styrene polymer. After 9 years of service, it shows significant cracking. This section is located just 0.6 mile (1 kilometer) from an identical subgrade exposed to the same traffic and climate but that does not contain REOB. The section of highway without REOB shows no cracking after 9 years of service.


The TFHRC team will soon submit to AASHTO the draft test methods that transportation agencies can use to test for the presence of REOB in asphalt mixes.These test methods will help transportation agencies know what materials and additives are present in the asphalt mixes they are purchasing.

Terence S. Arnold is a senior research chemist on the Pavement Materials Team in FHWA’s Office of Infrastructure Research and Development and Federal lab manager for the chemistry lab at TFHRC. He is a fellow of the Royal Society of Chemistry in the United Kingdom and a Chartered Chemist.

For more information, contact Terence Arnold at 202–493–3305 or terry.arnold@dot.gov.



Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101