Dissolved Oxygen Drives Carcinogen Formation in Stored Rocket Fuel, Study Finds

Unsymmetrical dimethylhydrazine, better known in the aerospace world as UDMH, has carried crews and satellites into orbit for decades as a storable, high-energy liquid propellant. It is prized precisely because it can sit in a tank for long periods and still ignite on demand. Yet that long residence time is also its quiet vulnerability. New computational research published in Results in Chemistry has mapped, at the level of individual electrons and bonds, how dissolved oxygen slowly transforms stored UDMH into nitrosodimethylamine, or NDMA, a potent carcinogen that threatens both fuel performance and environmental safety. The study, led by Zichen Liu and Xiaomeng Lv together with colleagues, offers the most detailed theoretical picture to date of the reaction network that turns a reliable propellant into a hazardous contaminant during storage.

The stakes are considerable. During prolonged storage, UDMH reacts with dissolved oxygen and other reactive species to generate a cocktail of byproducts that can number in the dozens. Among them, NDMA stands out for its strong carcinogenicity, its complex transformation chemistry, and its damaging effect on fuel quality. If a tank leaks, NDMA contamination can persist in soil and water, creating long-lasting environmental hazards around launch facilities. Until now, most research on this conversion relied on experimental sampling and analysis of the resulting products, leaving the underlying molecular mechanism poorly understood. The new work addresses that gap by simulating the entire conversion process from first principles, using density functional theory to track every bond that breaks and forms along the way.

The team’s computational strategy was thorough. Using the Gaussian 16 program, they optimized the geometries of reactants, transition states, intermediates, and products at the B3LYP/6-311+G(d,p) level, with selected reactions treated using the ωB97XD and M06-2X functionals. To ensure accuracy, they performed high-precision single-point energy calculations at the CCSD(T)/aug-cc-PVTZ level, corrected with the Shermo 2.0 program. Every proposed transition state was verified to possess exactly one imaginary frequency, and intrinsic reaction coordinate calculations confirmed that each transition state genuinely connects the intended reactants and products. Reaction rates were computed with the KiSTheIP program using transition state theory, including an Eckart tunneling correction, and Arrhenius parameters were fitted to yield pre-exponential factors for each elementary step.

Before simulating any reaction, the researchers asked a deceptively simple question: where on the UDMH molecule does oxygen attack first? To answer it, they calculated the electrostatic potential on the molecule’s van der Waals surface and applied the condensed Fukui function, a quantum chemical descriptor that quantifies how electron density changes at specific atoms when a molecule gains or loses an electron. The analysis identified the two nitrogen atoms, particularly the amino nitrogen N2, as the sites most vulnerable to electrophilic and radical attack. This preference has a physical origin: the two methyl groups attached to the neighboring nitrogen push electron density through inductive effects, enriching the amino nitrogen and making it the natural target for oxygen. The prediction proved decisive for everything that followed.

The calculations revealed two competing routes to NDMA: one beginning with oxygen stripping hydrogen from the amino group, the other from the methyl group. In the amino oxidation pathway, oxygen captures the first amino hydrogen through a transition state with a free energy barrier of 24.32 kcal/mol, producing a hydroperoxyl radical and a UDMH-derived radical intermediate. A second, harder hydrogen capture follows, with a barrier of 36.68 kcal/mol, generating a diradical that spontaneously rearranges into dimethylaminoazene. This intermediate then combines with the hydroperoxyl byproduct in a barrierless step, releasing 38.94 kcal/mol of free energy and yielding NDMA directly. The rate constants tell the story starkly: the first hydrogen capture proceeds roughly ten million times faster than the second, making that second capture the bottleneck of the entire pathway.

The methyl oxidation route tells a different tale. Here, oxygen must snatch a hydrogen atom from a methyl group, a step with a free energy barrier of 35.76 kcal/mol and a rate constant of just 1.43 × 10⁻³⁷ cm³/mol/s, about a billion times slower than the corresponding amino-group step. The resulting radical passes through peroxide intermediates and hydrogen migration steps, ultimately fragmenting into N-methylimine and an amino radical. The amino radical then reacts with ground-state oxygen to form an amino peroxyl radical, which decomposes through a chain of steps to nitric oxide. That nitric oxide is the crucial ingredient for the second route to NDMA: the dimethylamine radical, produced either from N-methylimine combining with a hydrogen radical or from oxidation of formaldehyde dimethylhydrazone, couples with nitric oxide in a barrierless, strongly exothermic reaction to form NDMA.

One of the study’s most striking findings concerns water. When the researchers modeled the amino radical oxidation chain in the presence of water molecules, the reaction landscape changed dramatically. Water participates through six-membered ring transition states that act like a molecular scaffold, lowering the barrier for converting the amino peroxyl radical to a peroxide intermediate by 13.16 kcal/mol and accelerating that step by a factor of roughly one hundred million. The subsequent conversion to nitric oxide becomes even more dramatic, with the rate increasing by a factor of about one hundred billion and the free energy dropping by nearly 73 kcal/mol. In effect, water behaves as a catalyst for the oxidation chain that supplies nitric oxide, meaning that even trace moisture in a storage tank could substantially accelerate the chemistry that feeds NDMA formation.

The team also explored side branches that compete with NDMA production. Formaldehyde released during methyl oxidation can condense with UDMH to form formaldehyde dimethylhydrazone, an intermediate whose subsequent oxidation by oxygen is kinetically unusual. Although the hydrazone oxidation is endothermic and requires more energy than UDMH oxidation itself, its measured rate constant is 530 times higher, a paradox the authors attribute to π-π orbital overlap between the hydrazone’s carbon-nitrogen double bond and oxygen, which forms a stable precursor complex and lowers the apparent activation energy. Meanwhile, the aldehyde intermediate from methyl oxidation can rearrange to monomethylhydrazine, another known propellant, through a strongly exothermic intramolecular hydrogen migration. These branches illustrate how a single dissolved oxygen molecule can set off a branching cascade of chemically diverse products.

Putting the numbers side by side, the conclusion is unambiguous. Amino oxidation, with its lower barrier of 24.32 kcal/mol and rate constant of 2.81 × 10⁻²⁸ cm³/mol/s, dominates over methyl oxidation in both thermodynamics and kinetics, making it the primary route by which stored UDMH becomes NDMA. The consistency between the Fukui-function site predictions and the computed kinetic parameters lends the analysis strong internal coherence: the site quantum chemistry says oxygen should attack, and the energetics confirm that it does. The dimethylaminoazene intermediate emerges as the key species in the dominant pathway, while nitric oxide generated from the amino radical oxidation chain, with water acting as catalyst, enables the secondary route.

Beyond its scientific interest, the work carries practical weight for launch safety, fuel quality control, and environmental protection. By identifying the rate-determining steps and the catalytic role of water, the study provides theoretical targets for suppressing NDMA formation, whether through oxygen scavenging, moisture control, or stabilizing additives designed to block the vulnerable amino hydrogen. The authors frame their approach, combining quantum chemistry with intelligent optimization methods, as a template for precise regulation of fuel chemistry during long-term storage. As space launch activity intensifies worldwide and legacy propellant depots age, understanding the slow electronic-scale sabotage occurring inside those tanks may prove as important as any engine test. This study shows that the enemy of stored UDMH is not dramatic decomposition but the patient, oxygen-driven chemistry of the amino group, one hydrogen atom at a time.

Subject of Research: Density functional theory study of the dissolved-oxygen-driven conversion of stored unsymmetrical dimethylhydrazine into nitrosodimethylamine

Article Title: Key mechanism for the conversion of long-term stored unsymmetrical dimethylhydrazine to nitrosodimethylamine driven by dissolved oxygen

Article References: Liu, Z., Lv, X., Zhou, W., Shen, K., Liu, E., Gao, M., & Sun, H. (2026). Key mechanism for the conversion of long-term stored unsymmetrical dimethylhydrazine to nitrosodimethylamine driven by dissolved oxygen. Results in Chemistry, 31, Article 103910. https://doi.org/10.1016/j.rechem.2026.103910

Image Credits: AI Generated

DOI: 10.1016/j.rechem.2026.103910

Keywords: UDMH, NDMA, nitrosodimethylhydrazine, density functional theory, dissolved oxygen, rocket propellant, hydrazine oxidation, reaction kinetics, transition state theory, Fukui function, water catalysis, fuel storage safety