Almost every environmental claim made about a road is really a claim about its materials, and almost none of the decisions behind those claims are made by the engineer whose name appears on the plans. They are made in a specification office that fixes an allowable percentage, at a quarry supplying an aggregate with a particular thermal expansion coefficient, in a refinery deciding how much heavy gas oil to pull from a vacuum tower, and on a plant control panel that sets a mixing temperature. This section works at that level, because it is where sustainability stops being a framework and turns into a number somebody can be held to.
That means reading the source documents rather than summaries of them, and the recurring finding is how much narrower the federal statements are than the claims built on them.
The denominator usually does the work
The asphalt industry’s annual survey carries two recycling rates for the same material. One counts whatever avoided a landfill and clears 99 percent. The other counts what went back into a new asphalt mixture and reaches about 89 percent, with unbound aggregate bases absorbing the balance. Only the second describes recycling in the engineering sense. How recycled asphalt is used in road projects takes the eleven-point gap between those numbers as its subject, along with the more revealing figure underneath them: the average recycled content of American asphalt mixtures rose only from 15.6 percent in 2009 to 21.9 percent in 2023, in a specification framework that already permits considerably more.
The same problem shapes the concrete question. EPA’s inventory figure for cement carries only the emissions from the chemical reaction, with the fuel accounted elsewhere, so a clean kiln addresses about two fifths of the problem. The future of low-carbon concrete works outward from that boundary, through the supplementary material supply data, the mineralization evidence and a federal purchasing rule whose accounting stops at the plant gate.
Permeable pavement suffers a version of the same confusion, in which an open-graded friction course gets counted as permeable even though water never reaches the subgrade. Permeable pavement separates the two, assembles the measured stormwater and deicing results, and reports what federal guidance says about the main driving surface.
Durability is decided in the materials specification
The most consequential environmental property of a pavement is how long it lasts, and length of service is bought with aggregate tests, air content limits, dowel bars and curing time rather than with intent. The state specifications collected in long-life pavement design make that concrete: a freeze-thaw expansion limit moving from 0.060 to 0.025 percent, a chloride permeability ceiling of 2,500 coulombs, seven days of curing before any wheel touches the slab. Nothing on the list is exotic and every item is enforceable.
The mirror image runs through the flexible side, where a pavement’s fate is set by binder chemistry and volumetrics. How asphalt is made follows the chain from the roughly 190 crude oil streams in American use that can produce paving binder at all, through the vacuum tower cut and a grading sequence that ages a sample through two artificial lifetimes, to a mix design that converges on four percent air voids in the laboratory and lands at six to eight in the road.
The comparison everyone wants settled sits between the two. Asphalt versus concrete for highway construction declines to declare a winner for the reason FHWA declines to: the agency maintains a formal procedure for letting bidders decide when engineering and economic analysis produces no clear choice. What the article does instead is name the narrow circumstances in which one material clearly wins, and they turn out to be truck volume, aggregate quality and the maintenance capability an agency can staff.
Where the evidence runs out
Four disputes sit beneath this coverage, and present evidence resolves none of them.
The first is the accounting boundary. Federal purchasing rewards low embodied emissions measured from cradle to gate, a boundary that ignores service life entirely. Whether a rule blind to the difference between a thirty-year pavement and a sixty-year one is a net gain has not been measured.
The second is supplementary material supply. A national standards body warns that fly ash and slag availability will likely fall with the retreat from coal power and pig iron, while the coal ash industry’s own 2024 survey reports rising use in concrete and forecasts ample supply for years because harvesting from legacy impoundments opened a second channel. Both positions are sourced and they cannot both hold for two decades.
The third is verification. A forty-year service life cannot be validated inside the career of the person who specified it, and the only honest early test is the absence of premature distress. Agencies that built long-life sections without a control section alongside have purchased the structure and forfeited the evidence.
The fourth is the recycled content ceiling. Specifications allow more reclaimed material than the industry places, and what binds is stockpile variability and the laboratory work needed to manage it. The cost of that work falls on a producer while the benefit accrues to an owner decades later, which makes it a procurement problem wearing a materials costume.
Every figure on these pages traces to a source listed alongside it. Numbers that no primary document would confirm are marked as such, together with the body a reader should ask instead.