June 2005

The Challenge of SMA
How producers meet the more exacting specifications of stone matrix asphalt.

by , Contributing Editor

Stone matrix asphalt, or SMA, has become the nation’s premier high-performance asphalt mixture. This rut-resistant mix lasts 25% to 50% longer than conventional dense-graded mixtures. And even though it costs 10% to 30% more than conventional mixes, more than 28 states have figured it’s worth the money and have placed SMA in high-traffic applications.

In a few states, SMA has become a standard. Georgia, for example, paved 3.0 million tons of the longwearing mix between 1991 and 2002. And just last year, Maryland, another SMA pioneer since the early 1990s, placed 220,000 tons of the gap-graded mixtures, which amounts to 18% of all of the state’s asphalt paved in 2004.

Stone matrix asphalt can be made a little faster than fine wine, but it takes the same care to produce. It especially requires care — and a certain amount of investment — in the aggregate production. Even in states such as Maryland, where stone producers have been crushing material for SMA for years, certain aggregate specifications remain a challenge to meet.

The challenge relates to the structure of the mixture. It relies on stone-to-stone contact between very hard, cubical aggregates to obtain structural strength. A high percentage of the aggregates are coarse, and a low percentage — or none — are intermediate-sized particles. To bind the stones together and give the mix durability, SMA contains a high percentage of asphalt cement. Fines and cellulose or mineral fibers are added to prevent drain-down of the binder.

Aggregate shape is critical. In most states, the specification that gives many producers a challenge is the one requiring a maximum of 20% of the plus 4 mesh aggregate to have a 3:1 flat and elongated (F&E) condition. Called equidimensional or cubical aggregates, they’re vital to the strength of the mixture. (Sometimes the F&E spec can range higher, if the aggregate compensates by proving itself hard enough in the LA abrasion test.)

Investing in crushers

In two districts of central Virginia, they’re ramping up to produce nearly a half-million tons of stone matrix asphalt in the next two years, says Richard Schreck, executive vice president of the Virginia Asphalt Association. That’s a marked increase: Last year those same two districts produced less than 40,000 tons of SMA.

“We weren’t expecting this amount of SMA,” says Randy Weingart, director of materials management and research for Richmond-based Luck Stone. “That was a surprise, and a lot of companies were jumping through hoops figuring how they were going to do it.” Luck Stone will be the aggregate supplier to Mega Contractors, the asphalt producer for two major SMA projects.

“The big challenge is to get the (aggregate) shape right first,” Weingart says. “Then we can work with our customer to develop the sizes that work the best.”

Luck Stone does not currently have a production facility in central Virginia that can directly produce the shape of stone that SMA requires. To add to the challenge, the Virginia DOT is conducting an experiment with aggregate fractionation on one of the projects. Fractionation, as practiced in Germany, means that the aggregate is broken into seven or more separate and distinct sizes. For the Virginia project, the DOT wants four sizes of stone (see sidebar).

“Our customer will do the screening to separate the aggregate into four sizes for the demo project,” Weingart says. “We’ll assist them in a consulting role.” Longer term, Luck will be experimenting with an impact-type crusher in central Virginia. “We need to be doing the crushing and screening to supply our customers with the sizes of stone they need,” Weingart adds.

Adjusting a compression crusher to limit 3:1 stone to 20% requires either tightening the crusher down on the closed side, or opening up the crusher to increase circulating load. Either way, hourly production drops by 40% to 60%, he notes. “So your operating costs have just doubled,” he notes. Plus, there’s the opportunity cost incurred because you’re only making half the production that you could. “Overall, you have less product to sell, and these days we can sell product when it hits the ground,” Weingart says.

If the state commits to specifying sufficient quantities of the mix, he says, Luck Stone can afford to invest in additional facilities. “The state is building a case to support its long-term commitment to SMA,” Weingart says. “So I think there will be a greater long-term interest in making the investment needed to make SMA.”

Maryland’s challenge

At its crushing facility in Havre de Grace, Maryland, Arundel makes SMA aggregate and can meet the 20 percent maximum 3:1 F&E specification, says Dave Wherley, technical services manager. But it’s not easy. “The F&E at 3:1 is a stumbling block for many of the aggregate producers, and it’s a challenge for us,” he says. “Our chip sizes run in the high teens or low 20s of F&E at 3:1. The 5:1 spec at 5 percent maximum is typically not too much of a problem.” Wherley meets the 3:1 spec of SMA by blending aggregates.

“We’re a fractionated plant, and we make AASHTO [American Association of State Highway & Transportation Officials] sizes 5, 6, 7, 8 stone and screenings,” he says. (No. 5s are 1.5-inch minus; No. 6s are 1-inch minus; No. 7s are 3/4-inch minus; and No. 8s are 1/2-inch minus.)

At Havre de Grace, Arundel runs a primary gyratory crusher and the rest of the crushers in the circuit are high-speed cones. The key to meeting the 3:1 spec is to cut down the reduction ratio of stone size input to stone size output, Wherley says. To do that, Arundel uses a supplementary crusher that takes No. 5 and 6 stones as feedstocks and recrushes them into 7s and 8s. That output of 7s and 8s from the supplementary crusher is blended back into the main plant’s 7s and 8s to meet the F&E spec.

It is the supplementary plant’s product that enables the blend to meet the F&E spec. But, Wherley says, “The practicality of using 100 percent stone from the supplementary plant is uneconomical. We must have that output from the supplementary plant to feed back into the primary plant.”

A first in South Dakota

South Dakota encountered an aggregate shape challenge on its first SMA pavement, placed last year. The project required 55,000 tons of stone matrix asphalt as an overlay on 13.5 miles of Interstate 29 from Beresford to Canton. “We’re hoping for at least 15 years out of it,” says Jim Costello, asphalt mix design engineer for the state DOT.

“We wanted to see if our quartzite from that area would work in terms of volumetric quantities,” Costello says. “It’s a very hard aggregate, and we had to find out if we could get some binder into it and keep it in the aggregate. We could.

“We did run into a problem: The F&E at 3:1 would be out of spec at times,” he adds. “The national spec is 20% maximum at 3:1, and we were getting 25s and 23s. The 5:1 F&E was all right. We could meet that.”

The contractor responsible for the mix design was Brower Construction of Sioux City, Iowa. To solve the F&E problem, Mike Collins, quality mix assurance manager at Brower, sought help from Don Watson, a research engineer at the National Center for Asphalt Technology (NCAT) at Auburn University. “Watson requested through Brower that if we could go to 32% at 3:1, that our lower LA abrasion test would compensate for the higher F&E count,” Costello says.

In other words, the aggregates’ hardness could compensate for more flat and elongated particles. “The LA abrasion test was 23 — which means a 23% loss after you do the tumbling test,” he explains. “We were far below the national spec of 45%.”

Brower Construction’s Collins credits NCAT’s Don Watson with doing the mix design. “We told Don the sizes we had to work with, and he fitted them into the spec,” Collins says. “We used a cellulose fiber, because we could not get the drain down to stop with mineral fiber. Cellulose stopped it.”

Adds Costello: “They were testing every 1,000 tons, and we checked to see if the F&E particles were breaking. We took cores, and sliced the cores, and you could see they weren’t breaking any aggregates.”

For Brower, mixing stone matrix asphalt was a new experience. “We were told you couldn’t produce SMA at more than 270 tons per hour,” Collins says. “But we exceeded 300 tons per hour easily. The more aggregate we put in front of the burner, the better it created a veil and held the heat. Otherwise, this mix is so coarse and open that when it goes through the drum, the heat just blows through it. You really need to pay attention to your baghouse temperatures.”

Brower added cellulose fibers at 0.03%. “It doesn’t seem like much, but it’s enough to stop the draindown,” Collins says. “The fiber bin controls it all. We got the fiber bin from High Tech Solutions.”

South Dakota has created a Class S-modified mix that has the same gradation spec as an SMA but the contractor doesn’t do testing on it, Costello says. The state has let a contract to place 45,000 tons of this S-modified mix on 20 miles of U.S. Highway 83. The 12.5 mm gap-graded mix will use natural crushed gravel, but it’s not a ledge rock.

And where SMA has a requirement for an 8% minimum of #200 size material, the S-class mix can run from 6 to 10% fines. “By opening up the #200 fines content, the limestone people can compete for the mix,” Costello says.

Experiments with Fractionating
and Gritting

This year the Virginia Department of Transportation plans to place an experimental stone matrix asphalt surface course using fractionated aggregates to make the mix. The hot mix will be paved on a short section of Interstate 295. Mix made with fractionated aggregate — four separate sizes — will be placed in two or three lanes, side by side with an equal width of SMA made with conventionally blended aggregates, says Gary Jennings, assistant resident engineer in the DOT’s Sandston office.

Kevin McGhee, principal investigator for the agency, says the following four source piles will be specified: 0 mm to 2 mm; 2 mm to 5 mm; 5 mm to 8 mm; and 8 mm to 11 mm. The idea was borrowed from Germany, where road builders typically differentiate and blend stone from five to six or more bins, compared to three on average in Virginia.

“We anticipate that we’ll get more uniform quality, more consistency in the mix,” McGhee says. “The contractor won’t have to change binder content to accommodate finer or coarser material. Sometimes contractors have to adjust the AC content because the aggregate can break down differently. The construction process will be more predictable, instead of using whatever shows up from the stone plant.”

German road professionals also take steps to ensure that a new SMA pavement exhibits sufficient friction, or skid resistance. They apply grit to the pavement, and Virginia plans to experiment with grit as well. In addition to ensuring skid resistance, the grit cuts down glare and glossiness on a new SMA.

Virginia will apply pre-coated grit, or 2-mm to 5-mm stone, at the rate of 3 pounds to 5 pounds per square yard. The pre-coating will consist of 0.8% by weight of asphalt binder, PG 70 to 22. One lane’s width of coated grit will go down over the fractionated mix and one lane will be over conventionally blended SMA. Then a second lane of uncoated grit will go down over each type of mix. “And, at least one lane will receive nothing,” McGhee says.

In addition to improving friction and cutting the sun’s glare, he says the agency hopes it tightens up the surface and improves impermeability.

Indiana’s Answer: Steel Slag

In the early 1990s, Indiana paved a trial section of stone matrix asphalt that used dolomitic limestone as its coarse aggregate, and soon became concerned with its performance.

“The surface friction numbers dropped quickly, and then stabilized,” says Dave Andrewski, state materials engineer for the Indiana Department of Transportation. “However, we were concerned that the surface would polish and further reduce surface friction, so we put SMA on hold until we could find an aggregate that would perform more to our liking.”s

“We found one in steel slag,” Andrewski says. Steel furnace slag, a by-product of steel manufacturing, is used throughout several central states as an aggregate for paving. With an LA Abrasion of about 20%, steel slag is plenty hard enough for SMA. It has a specific gravity higher than natural aggregates in Indiana, can be crushed to meet SMA gradations, and has no problem meeting the 3:1 flat-and-elongated spec.

“Between 1995 and 2000, we put down some trial sections of SMA using steel slag, and they performed well,” Andrewski says. “Around 2000, they wrote a spec with steel slag, and continued to tweak it. Then we added it to our standard specifications as of 2005.

“We have probably done 25 or 30 projects with steel slag since the late 1990s,” Andrewski says. “We worked with the slag producers to ramp up production. There are transportation costs that supply all areas around the state. But we’re lucky that we can supply Indiana with steel slag by rails, barges, or trucks.”

Gary, Indiana-based Edward C. Levy Company supplies sand and gravel for paving, as well as blast furnace slag and steel slag. John Yzenas, Levy’s director of technical services, says Indiana asphalt producers typically make 80,000 to 100,000 tons per year of stone matrix asphalt hot mix using steel slag.

Illinois road builders also use steel slag for stone matrix asphalt mixes. In a typical year the state will use more than 100,000 tons of the mix with steel slag, Yzenas says. Slag is the only material used for SMA in the Chicago, Illinois, area, and the agency has used it for interstate highway work in the southern portion of the state, roughly 200 miles from the source.

To crush steel slag, Levy uses cones as the primary crushers and a vertical shaft impactor as a secondary unit. “Other people have used a jaw as a primary and a VSI as a secondary; there’s been a pretty good cross-section of crushers used,” Yzenas says.

Steel slag SMA costs about 25% more than conventional surfaces, Andrewski says. And Yzenas noted that transportation could push costs upward.

As electric arc furnaces have popped up around the country to produce steel, slag is now available to more states, Yzenas points out. Such furnaces are found in Alabama, South Carolina, Virginia, Kentucky, Indiana, Ohio, Illinois, and Texas, to name some examples.

“Steel slag has definitely become more available as a source of aggregates — or as a source for SMA,” Yzenas says. “We’ve done some work with Kentucky; the steel mill is in Ashland, Kentucky. And we’re putting on a demonstration this year in Ohio.

“We usually convince somebody to put down some SMA, then we call the state folks up to come and take a look at the project,” Yzenas says. Usually Levy chooses a severe-traffic application, like an aggregates haul yard, so that people can compare the wear of trucks to high-volume traffic on a roadway.

“We’re into our eleventh year and to the best of my knowledge we’ve had no failures,” Yzenas says.

A material transfer vehicle is used to pave with slag SMA on Interstate 65 in Indiana.  A material transfer vehicle ensures an even flow of material to the paver

Reprinted from Better Roads Magazine
June 2005