three percent of the pipes, most of the leaks

Problem statement

Legacy cast iron and bare steel natural gas distribution pipes make up about 3% of the nearly 2 million miles of U.S. utility pipes in use (per DOE: roughly 1.2 million miles of distribution mains plus 900,000 miles of service lines), but account for a disproportionate share of gas leaks, pipeline failures, and methane emissions. These legacy pipes — some over 100 years old — are concentrated under dense urban areas where replacement requires excavating streets, disrupting traffic, and relocating other utilities at costs of up to $10 million per mile. Utilities have run replacement programs for more than three decades — cast iron and unprotected steel have fallen from roughly 20% of all mains in 1990 to under 3% — but tens of thousands of miles remain, concentrated in the oldest Northeastern and Midwestern systems. No viable technology exists to rehabilitate these pipes in place — constructing a new, certified pipe inside the existing one — at a cost and speed that would accelerate the timeline.

Why this matters

Distribution pipeline leaks are a major methane source: a peer-reviewed national estimate put leakage from distribution mains at 0.69 teragrams (690,000 metric tons) of methane per year — roughly five times the EPA inventory figure for that segment (Weller et al. 2020) — and EDF's 2023 analysis estimated 761,000 metric tons per year from distribution pipelines, about 3.75 times the EPA inventory; methane traps roughly 84 times more heat than CO₂ over a 20-year period. Beyond climate impact, legacy systems carry explosion risk — the 2018 Merrimack Valley gas disaster in Massachusetts killed one person, sent 22 people to hospitals, and damaged 131 structures; the NTSB traced the overpressurization to a cast-iron main replacement project in which the abandoned main's regulator sensing lines were never relocated. Replacement costs land in utility rate bases and are passed to ratepayers. ARPA-E's stated target is rehabilitation at less than $1 million per mile — a 10–20x reduction in cost per mile versus excavation and replacement — which would accelerate both emissions reduction and safety improvement.

What’s been tried and why it hasn’t worked

Cured-in-place pipe (CIPP) lining is widely used for sewer rehabilitation, and CIPP liners and test protocols exist for pressure pipe — but no lining approach is currently accepted by gas-pipeline regulators as the equal of pipe replacement, which is why regulatory acceptance ("accepted by regulators as equal to pipeline replacement," with costs allowed in the rate base) is an explicit REPAIR program goal rather than an existing capability. Pipe bursting (pulling a new plastic pipe through the old one while fragmenting it) works for some pipe segments but is harder to apply across the bends, service connections, and valve assemblies common in urban gas networks. The fundamental challenge is that any in-situ rehabilitation technology must produce what amounts to a certified new pipeline — the program requires a 50-year service life without reliance on the legacy pipe — using processes that work through the existing pipe with minimal excavation, since excavation is the largest cost component of replacement.

What would unlock progress

ARPA-E's REPAIR program (up to $38.5M over 3 years) seeks to develop technology that constructs a new pipe inside the existing one: smart coating materials with a 50-year life that are "stronger than steel, non-corroding, self-healing and self-reporting," robotic deposition tools that apply them from inside live pipes, in-line inspection and integrity tools that can certify the new pipe without excavation, and 3D mapping of gas pipes and adjacent underground infrastructure. Funded teams draw on cross-industry techniques — for example, the University of Pittsburgh/ULC Robotics team is adapting cold-spray metal coating, a process proven in aerospace component remanufacturing, to build structural stainless-steel liners inside gas distribution pipe at a target cost under $1 million per mile.

Entry points for student teams

A team could design and prototype a small-scale pipe-crawling robot capable of navigating a representative pipe geometry (straight sections, 90° bends, T-junctions) while carrying a simulated lining payload. Alternatively, a team could test rapid-curing composite materials for gas-pipe applications, evaluating mechanical properties, permeation rates, and chemical resistance. Robotics, polymer science, and civil engineering skills are most relevant.