99.99 Percent Destruction Is the Easy Part of Hazardous Waste Treatment

Which thermal technology actually destroys hazardous waste?

Rotary kiln, most of the time. Ask ten developers what hazardous waste treatment looks like and eight will describe the same thing: a refractory-lined kiln turning slowly, feeding a secondary combustion chamber that holds the gas hot for a couple of seconds, followed by a quench and an acid-gas train. They aren't wrong. That has been the workhorse for organic hazardous waste since the 1980s, and nothing in the last decade has displaced it.

The alternatives are narrower than the brochures suggest. Plasma arc handles high-hazard, low-volume streams and rarely pencils out at scale (I've reviewed exactly one plasma business case that closed, and it had a captive defense stream standing behind it). Thermal desorption destroys nothing at all; it moves the contaminant out of the solid and into a vapor you still have to treat, which is why I keep seeing it proposed as a disposal answer when it's really a separation step with a disposal problem attached. Cement kilns burn a great deal of hazardous waste as supplemental fuel and do it well. If you want the longer version of how these routes actually differ, we walked through the thermal conversion routes beyond incineration in more depth, and the same distinctions run through the wider family of pyrolysis systems that get pitched as hazardous waste disposal technology.

All of them get judged against the same yardstick. Under 40 CFR Part 63 Subpart EEE, an incinerator, a cement kiln, a lightweight aggregate kiln, a solid or liquid fuel boiler burning hazardous waste, and a hydrochloric acid production furnace are all one thing: hazardous waste combustors. Your technology choice moves your capital cost and your feedstock envelope. It doesn't move the compliance regime you're buying into. So the question I'd put ahead of "which technology" is this: which category does my unit fall into, and what will that category oblige me to prove, continuously, for the next twenty years?

How clean does the destruction have to be?

Four nines. Per EPA's replacement standards for hazardous waste incinerators at 40 CFR 63.1219, you demonstrate 99.99% destruction and removal efficiency on the principal organic hazardous constituents you designate, and 99.9999% if you're burning the dioxin-listed wastes (F020 through F023, F026 and F027). Six nines for those. It's the number everyone quotes, and a properly run kiln hits it.

Destruction efficiency isn't the binding constraint, though. The same section carries the stack limits that shape your air permit: dioxins and furans, mercury, two metals groups, particulate, and combined hydrochloric acid and chlorine gas. Look at the gap between what an existing unit must meet and what a new one must meet. An existing incinerator is held to 130 micrograms of mercury per dry standard cubic meter. A new one is held to 8.1. Same pollutant, same statute, a factor of sixteen between them, which tells you precisely how the agency thinks about anything you reconstruct.

What did EPA change on June 3, 2026?

The residual risk and technology review for the hazardous waste combustor rule went final. EPA signed it on May 29, 2026, published it three business days later, and concluded that risks from this source category are adequately addressed by the existing standards. That's the sentence the trade press led with, and I understand why. It reads like a clean bill of health.

The expensive part sits further down. EPA set emission standards for hydrogen fluoride and hydrogen cyanide at major-source hazardous waste incinerators, cement kilns, and solid and liquid fuel boilers. Two pollutants that most existing units have never been required to measure, on stacks whose analyzer trains were specified around carbon monoxide, oxygen and acid gas. The rule also lands work practice standards for startup, shutdown and malfunction periods, and it pushes notifications, compliance reports and performance evaluation data into electronic reporting. None of that arrives as a tighter emission limit. All of it arrives in your monitoring plan, which is where permits go to die.

Hydrogen fluoride is measurable. An FTIR analyzer of the class ABB sells into this niche will pick it up alongside hydrogen chloride and CO, so nobody is being asked to do the impossible. But "measurable" and "certified against a performance specification your state will accept at renewal" are separate problems, and the second one eats eighteen months and a budget line nobody wrote.

Why do these permits fail at the monitoring condition instead of the emission limit?

Because a limit is a number, and a monitoring condition is a promise about the future.

In 2024, on an air-permit review for a client feeding PFAS-bearing waste into an existing thermal unit, the state dropped a continuous-monitoring condition into the draft at month seven, and the instrument already bolted to that stack failed it on paper before it ever ran. Not a new limit. Just a new way of proving the old one. My client had a clean stack test, a defensible destruction case, and an analyzer certified against the wrong basis. We spent the rest of that year negotiating an alternative monitoring plan. We got it, at the price of a compliance schedule and most of the goodwill in the room.

That's the pattern, and it isn't "the technology couldn't do it." The technology did it. The file couldn't prove it on terms the state wrote down eleven months into a cycle everyone had assumed was closed. Conditions like that almost never appear in the permit you fought for. They arrive in the amendment, the renewal, the mid-cycle draft somebody filed as a formality.

Actually, I'll qualify that, because it overstates the case for the merchant incinerators. They've run analyzer-heavy permits for two decades and they'll absorb the new pollutants with a capital request and a shrug. It's the captive on-site unit, the one built to handle a single plant's own stream, that gets caught. Those permits were written light on monitoring because the risk case was small, and light monitoring is exactly what the June rule is now filling in. If your ESG-compliant projects rest on a captive hazardous waste treatment unit, go read its monitoring plan this quarter rather than next.

Can I just ship it out to a commercial incinerator?

You can, until you can't.

In early June 2021, commercial hazardous waste incinerators started telling customers they would no longer accept containerized waste bound for the burner, because they were backlogged. That isn't hallway gossip. EPA wrote it down, watched the backlog persist longer than it expected, and then managed the fallout by letting states and regions grant multiple consecutive extensions to generator accumulation time limits. Sit with what that means. The federal answer to a shortage of merchant capacity was to let the waste sit longer on the generator's pad, under the generator's permit, against the generator's liability.

Merchant capacity is thin and concentrated. Barely two dozen US facilities hold permits to incinerate hazardous waste, only a few of those also carry the TSCA permit you need for PCB-contaminated material [industry count], and one operator's own disclosures put it above sixty percent of North American incineration capacity. That is not a market. That is a queue with a price attached.

So the on-site question deserves a hearing before the queue answers it for you. Generators with a steady, well-characterized stream keep re-running the numbers on in-house hazardous waste treatment, and interest in zero-waste-to-landfill solutions tends to spike every time the merchant queue lengthens. Whether it pencils depends on volume and hazard class, and for most generators it still won't. But the ones for whom it does tend to discover it during a backlog, which is the worst possible time to start a permitting cycle.

Where does this not apply?

Several places, and for small generators the exceptions matter more than the rule.

Treat hazardous waste in your accumulation tanks or containers while staying inside the generator conditions, and you're generally not buying a RCRA treatment permit at all. Generators lean on that exemption constantly and it holds, right up until the state has written something stricter, which many have, because an authorized state program is free to go beyond the federal floor. Cite the federal rule, stop there, and you'll get an education at the state desk.

The June rule is a major-source rule; area sources sit differently, and the final amendments even allow states to exempt certain area source combustors from some permitting requirements. Non-US readers should treat all of the above as local color. The EU runs hazardous waste through its Waste Framework Directive with a separate prior-consent regime for cross-border shipments, and the vocabulary doesn't map cleanly onto RCRA.

And the largest limit on everything I've written here: this is a compliance argument, not a technology argument. If your stream is aqueous, dilute and low-hazard, thermal treatment is the wrong tool and no permit strategy will rescue it. Stabilization, wet oxidation and unglamorous physical separation still beat a kiln on cost for a great deal of what's in the drum. I've watched developers reach for a rotary kiln because it's the thing they know how to permit, which is a terrible reason to pick a technology and a very human one.

What does a defensible hazardous waste treatment file look like?

Five things. Produce all five on a regulator's timeline or a lender's timeline, or you don't have a project, you have an intention.

1. The combustor category you fall under, in writing, with the reasoning. Not "we're an incinerator." Which category, and why that one.
2. A destruction demonstration tied to named principal organic hazardous constituents, rather than a generic four-nines claim lifted from a vendor deck.
3. A monitoring plan naming the instrument, the performance specification it's certified against, and the averaging basis, for every limit you're subject to, hydrogen fluoride and hydrogen cyanide included, if the new rule reaches you.
4. A written answer for startup, shutdown and malfunction that matches the work practice standards rather than your operating habits.
5. A renewal history: every condition added since issuance, who added it, and what you traded for it.

Number five is the one nobody keeps, and it's the one that decides the next renewal. In 2023 a client of mine filed what every engineer in the room called a minor modification. It tripped PSD review and delayed them eight months. Nobody had tracked which conditions had accreted to that permit across a decade, so nobody saw the trip coming. Since then I ask for the renewal history first, before the drawings, and the reaction tells me most of what I need to know about how the facility will handle the next reopening. Plenty of operating waste conversion facilities can hand you a beautiful process flow diagram and nothing at all about how their permit got its shape.

My honest read, and I would rather be wrong: a good share of the hazardous waste combustion capacity permitted in the US today won't survive its first Title V renewal under the monitoring rules now in force. Not because the kilns can't hit their numbers. They can, and they do, at four nines and six nines, shift after shift. It's that nobody wrote down how they would prove it once the analyzer, the averaging basis and the pollutant list all changed underneath a permit that was never reopened.

Sources & Notes

EPA's final residual risk and technology review for hazardous waste combustors was signed on May 29, 2026 and published on June 3 (91 Fed. Reg. 33484). The new hydrogen fluoride and hydrogen cyanide standards, the startup/shutdown/malfunction work practices and the electronic reporting requirements are all in that action; the docket runs under EPA-HQ-OAR-2004-0022. Background and rule history sit on EPA's hazardous waste combustors NESHAP page.

The destruction and removal efficiency figures and the existing-versus-new stack limits come straight from the replacement standards for hazardous waste incinerators, which is worth reading in full if you're about to sign a permit application.

On the backlog: EPA's page on the backlog of containerized hazardous waste needing incineration is the agency's own account of commercial incinerators turning customers away and of the accumulation-time extensions that followed. It is a short page and an uncomfortable one.

The 2023 Title V modification and the 2024 air-permit review are drawn from my own project files. Dates and sequences are exact; clients and states are left out at their request. Anyone who has sat through a mid-cycle condition change will recognize the shape of both.