Under 10 mg/Nm3, Repeatedly: Why SyngaPure Started With the Tar Reformer

UNDER 10 MG/NM3, REPEATEDLY: WHY SYNGAPURE STARTED WITH THE TAR REFORMER

Summary

Q3 component prototyping for SyngaPure is underway, and the first component on the bench is the catalytic tar reformer. That is deliberate. Tar is where waste-to-chemicals projects die in the field, not in the lab. The bench program measures four things: tar conversion across changing inlet loads, catalyst durability in operating hours, temperature stability inside the bed, and pressure drop over a full run cycle. The success bar is not a single clean number. It is staying under 10 mg/Nm3 while the feedstock composition moves underneath the system. The phase is funded by the BMWK go-inno programme.

Why Now

The waste-to-chemicals industry has a graveyard of projects that looked right on paper. Enerkem has one commercial plant after 24 years. Fulcrum filed Chapter 11 in 2024. Entrade is insolvent. Agnion and Andritz spent EUR 41 million on a heatpipe reformer and stopped. Every one of them hit the same wall: the jump from bench conditions to industrial conditions. Catalyst poisoning, tar fouling, and sorbent degradation all get triggered by real feedstock variability, the thing every pilot plant promises to handle and few actually survive.

The regulatory clock runs in favor of whoever solves this first. ReFuelEU pushes SAF blending from 2% in 2025 to 70% by 2050. RED III sets a 42.5% renewable target for transport by 2030. Landfill costs in Germany run 160-200 EUR/t. The distributed waste-to-energy market is estimated at EUR 8-15 billion by 2030. Operators need validated gas cleaning now, and the BMWK go-inno programme exists precisely to fund the unglamorous middle where deep-tech either ships or dies.

What We Built

SyngaPure is a three-stage cleaning architecture: catalytic tar reforming, particulate filtration, and sorbent-based contaminant removal, with one AI control layer tying the stages together. Sorbent beds and filters are well understood engineering. That is exactly why we did not start there. Tar is the failure point. It is the reason a plant looks excellent in month one and quietly shuts down in month nine, when the fouling has worked its way through the engine, the catalyst, or both.

The catalyst on the bench is nickel-alumina on a silicon carbide foam support, roughly 10 wt% nickel, running at 850-950 C with a space velocity of 2,000-8,000 h^-1. The chemistry is standard. The engineering around it is not. What we are measuring at bench scale is precisely what kills systems in the field: conversion efficiency as the inlet load swings, durability counted in operating hours, temperature stability through the bed, and pressure drop across a complete run cycle.

The definition of success matters here. Anyone can hit a tar target once, with a perfectly prepared feedstock, under ideal conditions. Published work on this reformer measures 94.2% tar removal at 900 C on its own. The integrated X-150 platform sustained 99.98% tar conversion over 1,939 hours at COMETHA in Paris, processing 16,382 kg of digestate pellets from MSW, sewage sludge, and horse manure. The gap between those two numbers is not better chemistry. It is system integration: the AI control layer coordinating reformer, filter, and sorbent bed as one loop instead of three independent boxes. Hitting under 10 mg/Nm3 repeatedly, while the feedstock composition moves underneath you, is a different problem from hitting it once. It is the difference between a lab result and a plant that stays online.

The honest caveat: bench conditions are not plant conditions. Bench data tells you whether the chemistry works. Only industrial validation tells you whether the system survives. That comes later in the roadmap, and we are not pretending otherwise.

Why It Matters

For the operator, this is the difference between a plant that runs and a plant that pays for itself. Tar is why gasification looks great in month one and dies in month nine. A reformer that holds under 10 mg/Nm3 while the waste stream shifts is what keeps an engine online, a CHP contract fulfilled, and a maintenance budget sane.

For the investor, the order of operations is the signal. We prototyped the hardest component first, funded by the BMWK go-inno programme, because validating the failure point is the cheapest way to de-risk everything downstream. EUR 7 million got the X-150 to a 1,939-hour validated campaign where comparable efforts spent EUR 41 million and stopped. SyngaPure follows the same discipline.

This is the least glamorous part of the whole project. It is also the part that decides everything downstream. If you have run catalytic reforming at industrial conditions: what failed on you first, the catalyst or the control?