Manganese Doping Boosts ZnO Oxygen Defects for Tandem CO2 Conversion to Olefins

The drive to transform carbon dioxide from a climate liability into a chemical feedstock has taken a significant step forward, thanks to a team of Chinese chemists who have shown that a modest dash of manganese can dramatically reengineer the surface of a zinc oxide catalyst to convert CO2 into the building blocks of plastics. Researchers at Yulin University in Shaanxi Province report in Catalysis Letters that manganese-doped zinc oxide, when paired with a silicoaluminophosphate zeolite, converts carbon dioxide and hydrogen into light olefins—ethylene, propylene, and butylene—with a carbon dioxide conversion of 12.57 percent and a light-olefin selectivity of roughly 80 percent, while remaining stable over a continuous 100-hour test. The result offers a concrete demonstration of how defect engineering at the atomic scale can reshape the performance of tandem catalysts for one of industrial chemistry’s most coveted transformations.

Light olefins are the workhorses of the petrochemical industry. Ethylene and propylene feed the synthesis of polyethylene, polypropylene, and a vast catalogue of plastics, solvents, and intermediates, and virtually all of them are currently cracked from petroleum-derived naphtha or extracted from fossil methane streams. Directly synthesizing them from carbon dioxide would, in principle, close a carbon loop: emissions captured from flue gas or air could be reassembled into the very molecules that sustain modern manufacturing. The obstacle has always been the thermodynamic and kinetic stubbornness of CO2 itself. Its carbon-oxygen double bonds are among the most stable in chemistry, and even when the molecule is persuaded to react, selectivity becomes the dominant challenge. The standard reaction network tends to spit out carbon monoxide via the reverse water-gas shift reaction, or single-carbon products like methanol and methane, rather than the carbon-carbon bonds that define olefins.

The Yulin team, led by corresponding authors Xiaoyan Liu and Yu Meng, attacked this problem with a tandem catalysis strategy that has become one of the most promising architectures in the field. The approach decouples the two chemical challenges and assigns each to a specialist. The first component—a metal oxide—is responsible for activating CO2 and partially hydrogenating it, typically to methanol or methoxy intermediates. The second component, here the zeolite SAPO-34, then carries out the well-established methanol-to-olefins conversion, a reaction in which oxygen-containing C1 intermediates are stitched together inside the zeolite’s cages and clipped into light olefins under the characteristic hydrocarbon-pool mechanism. The elegance of the tandem design lies in its division of labor, but its Achilles’ heel lies in the first stage: the oxide must not only activate CO2 efficiently but must do so in a way that channels carbon down the hydrogenation pathway rather than diverting it to carbon monoxide.

Zinc oxide has long been a candidate for the activation stage, most famously as part of the Cu/ZnO/ZrO2 methanol synthesis catalysts that dominate industrial methanol production. Its catalytic personality, however, is dictated largely by its surface defects. Oxygen vacancies—atomic sites where an oxygen atom has been removed from the lattice—create under-coordinated zinc centers and trapped electrons that can bend and weaken the CO2 molecule upon adsorption, setting the stage for hydrogenation. Past studies, including machine-learning-accelerated first-principles simulations, have established that the density and character of these vacancies govern how ZnO handles both CO2 and H2. The question the Yulin researchers posed was whether an external dopant could deliberately tune that vacancy population, and with it, the catalyst’s reactivity.

Their answer was manganese. The team synthesized a series of MnxZny composite oxides by co-precipitation, varying the Mn/(Mn + Zn) molar ratio across a broad range and then characterizing how the dopant reshaped phase evolution, surface chemistry, and defect structure. The findings describe a delicate balance. At moderate loadings, manganese ions substitute into the local ZnO structure, distorting the lattice and generating an enriched population of surface oxygen vacancies and defect-related adsorbed oxygen species. These defects are not passive bystanders; they are the seats of catalytic action, providing sites where CO2 can bind and be activated, and where hydrogen can dissociate into reactive species. But the relationship is not linear. Push manganese incorporation too far and a new crystalline phase emerges: the spinel ZnMn2O4, a structure that locks the chemistry into a less useful configuration and marks the point of diminishing returns.

Among all the compositions tested, the material with a 30 percent Mn fraction—denoted 30%Mn-ZnOx—proved to be the sweet spot. Characterization by temperature-programmed desorption of CO2 and H2 revealed that this sample possessed the highest fraction of defect and adsorbed oxygen species, and correspondingly showed enhanced CO2 adsorption capacity and improved H2 activation behavior. In other words, the doped surface had become simultaneously a better host for the carbon-bearing reactant and a better source of activated hydrogen. The team’s diffuse reflectance infrared Fourier transform spectroscopy, or DRIFTS, measurements then provided a molecular-level window into the reaction pathway. The spectra revealed the formation of formate species, HCOO, and methoxy species, CH3O, on the catalyst surface—a sequence of intermediates consistent with a formate-related hydrogenation route in which CO2 is first hydrogenated to formate, then progressively reduced toward methoxy and ultimately methanol-like intermediates before being handed off to the zeolite.

That handoff is where the tandem architecture earns its keep. When the optimized 30%Mn-ZnOx was physically combined with SAPO-34—a chabazite-like silicoaluminophosphate whose small eight-membered-ring pores impose shape selectivity favorable to ethylene, propylene, and butenes—the composite catalyst achieved a CO2 conversion of 12.57 percent and approximately 80 percent selectivity toward light olefins under optimized reaction conditions. Equally important for any prospective industrial process, the catalyst maintained stable performance over 100 hours of continuous operation, a duration long enough to suggest that the optimized defect population is not a fleeting artifact of synthesis but a durable feature of the working catalyst. Stability is the perennial graveyard of promising CO2 hydrogenation catalysts, since high temperatures and reducing atmospheres tend to sinter active phases, collapse vacancy populations, or coke the zeolite component; surviving a century of hours intact is a meaningful credential.

The study also contributes mechanistic clarity to an ongoing debate in the tandem catalysis literature: whether the oxide stage proceeds through methanol as a true intermediate or through formate pathways that terminate in surface-bound species consumed in situ. The co-detection of HCOO and CH3O by DRIFTS, together with the temperature-programmed desorption fingerprints of matched CO2 and H2 activation, points toward a formate-related hydrogenation route feeding the zeolite. This matters because the relative abundance of formate intermediates has been linked in other work to suppressed CO selectivity—the undesirable reverse water-gas shift product that steals carbon from the olefin pool. By engineering the vacancy structure to favor formate chemistry, manganese doping appears to shift the balance away from carbon monoxide and toward the hydrogenation products that matter.

The broader context makes the result timely. Carbon capture and utilization strategies face a scale problem that only catalysis can solve, and CO2 hydrogenation sits at the center of the prospective carbon-recycling economy, whether the target is methanol, syngas-derived products, or olefins. Governments have set ambitious decarbonization targets—the source study cites China’s action plan for peaking carbon dioxide emissions before 2030 and the Intergovernmental Panel on Climate Change’s 2023 synthesis report—and the petrochemical sector is among the hardest to abate because its carbon is embedded in products, not merely burned for energy. Tandem oxide/zeolite catalysts, first popularized for direct syngas conversion and later extended to CO2, have rapidly become a leading candidate platform, with variations employing ZnZrOx, ZnGa2O4, GaZrOx, and In2O3-based oxides paired with SAPO-34, SAPO-18, or SSZ-13 zeolites. What the Yulin work adds is a dopant-based lever for tuning the oxide side of the pairing, one that operates through an explicitly defect-centric mechanism.

There are, of course, the usual caveats that separate laboratory promise from industrial reality. A 12.57 percent single-pass conversion, while respectable for CO2 hydrogenation, would need to be raised or compensated with recycle streams in a commercial reactor, and 100 hours of stability, though encouraging, falls short of the thousands of hours required for process viability. The co-precipitation synthesis is, however, inherently scalable, and manganese is an abundant, inexpensive metal—an important contrast with the gallium and indium dopants that appear in some competing oxide systems. The expense and supply risk of critical raw materials is a recurring constraint on carbon-utilization technologies, and a catalyst built on zinc and manganese sidesteps much of that vulnerability.

Perhaps the most transferable lesson from the study is conceptual: vacancy engineering by doping is a systematic, tunable strategy, not a matter of serendipity. The clear compositional maximum at 30 percent manganese, the structural diagnosis of spinel formation as the failure mode of over-doping, and the spectroscopic linkage of defect density to reactant activation together sketch a design protocol that other groups can apply to different oxide chemistries. As research groups worldwide refine the oxide-zeolite tandem platform for CO2-to-olefins conversion, the ability to dial in the right number of oxygen vacancies with a cheap dopant may prove to be one of the quiet enablers of a petrochemical industry that runs on recycled carbon rather than freshly extracted oil.

The researchers, based at Yulin University’s Shaanxi Key Laboratory of Low Metamorphic Coal Clean Utilization and the university’s College of New Energy, were supported by the National Natural Science Foundation of China, the Natural Science Foundation of Shaanxi Province, and regional science and technology programs in Yulin. Their work, published as the field of heterogeneous CO2 activation continues to accelerate, underscores a theme that has come to define modern catalysis research: the catalytically important atoms are often the ones that are missing.