CO Poisoning Catalytic Conversion for Fuel Cells

1. Introduction: CO—An Unavoidable Barrier in the Commercialization of Fuel Cells

Proton exchange membrane fuel cells (PEMFCs), which use hydrogen as fuel, offer advantages such as high energy conversion efficiency and water as the only emission, positioning them as a key technology in the clean energy transition. However, the commercialization of fuel cells faces a long-standing technical obstacle—the poisoning of the anode platinum (Pt) catalyst by carbon monoxide (CO). Multi-source hydrogen fuels represented by syngas (including coal-derived hydrogen, cracked hydrogen, and industrial by-product hydrogen) account for the vast majority of global hydrogen production capacity, but due to the poisoning effect of CO and other impurities, they cannot be directly utilized in conventional PEMFCs. The production of high-purity hydrogen requires additional purification steps such as water-gas shift, pressure swing adsorption, or preferential CO oxidation, significantly increasing fuel costs.

The key to converting CO from a fuel cell "poison" into a usable "fuel" lies in catalyst design. This article systematically reviews the poisoning mechanism of CO on anode catalysts, as well as the principles, performance data, and application boundaries of three mainstream catalytic technology pathways.

2. CO Poisoning Mechanism: Understanding the Nature of "Poisoning" at the Molecular Level

The fundamental reason CO poisons fuel cells is its strong chemisorption on Pt surfaces. The hydrogen oxidation reaction requires two adjacent vacant Pt active sites, whereas CO can adsorb not only through vacant sites but also through Pt–H sites. As a result, the adsorption coefficient of CO on Pt is several orders of magnitude higher than that of hydrogen, allowing it to preferentially occupy active sites and block the electrochemical oxidation of hydrogen.

From a quantitative perspective, CO concentrations as low as 100–1000 ppm can rapidly deactivate platinum catalysts. The tolerance limit of the most mature low-temperature PEMFCs for CO content in anode fuel is only 0.001 mol% (10 ppm). The cell performance exhibits a two‑stage characteristic over time upon CO introduction: an initial sharp decline followed by a plateau with persistent performance oscillations. Elevated temperatures can mitigate CO poisoning (since CO adsorption is exothermic and desorption is endothermic), but the operating temperature window of PEMFCs limits the extent of this mitigation. Furthermore, CO poisoning of Pt catalysts is reversible—the performance of a poisoned catalyst can be largely restored by purging with pure hydrogen.

Understanding this mechanism is the foundation for evaluating various anti‑CO catalyst design strategies—all technical pathways essentially address the same core issue: how to prevent CO from occupying Pt active sites, or how to rapidly convert adsorbed CO into CO₂.

3. Technical Pathway 1: Preferential Oxidation of CO (CO‑PROX)—A Mature Upstream Purification Solution

3.1 Principle and Process Target

Preferential oxidation of CO (CO‑PROX) is one of the most widely used hydrogen purification technologies. The principle is to introduce a small amount of oxygen into the hydrogen‑rich gas stream, allowing CO to be selectively oxidized to CO₂ over a catalyst while minimizing H₂ oxidation. The process target is to reduce the CO concentration in the hydrogen‑rich gas to below 10 ppm to meet the fuel purity requirement for PEMFCs.

3.2 Catalyst Systems and Performance Data

CO‑PROX catalysts have been developed across multiple systems, including platinum‑group metals, Group IB metals, bimetallic combinations, and transition metal oxides. A representative work involved preparing a platinum cluster catalyst modified with highly dispersed Fe(OH)_x species (Pt‑Fe(OH)_x) via pretreatment of a PtFeMgAl‑LDHs precursor under reaction gas conditions. This catalyst features abundant Pt^δ+–(OH)_x–Fe³⁺ interfacial sites, which have been identified as the intrinsic active centers for CO‑PROX.

In terms of performance, the optimal catalyst (Pt‑5Fe(OH)_x/MA) achieved nearly 100% CO conversion at room temperature (25 °C), with a mass‑specific activity as high as 9.09 mol_CO·g_Pt⁻¹·h⁻¹, outperforming other Pt‑based catalysts reported in the literature. Moreover, this catalyst maintained complete CO conversion over a broad temperature window of 25–225 °C.

3.3 Technical Positioning

CO‑PROX is positioned as an upstream purification stage in the fuel cell system—before hydrogen enters the stack, it passes through a catalytic reactor to reduce CO concentration to below the safety threshold. Its advantage lies in not altering the fuel cell stack design itself, thus being directly adaptable to existing PEMFC systems; the trade‑off is increased system complexity and additional energy consumption.

4. Technical Pathway 2: Single‑Atom Catalysts and Hierarchical Microenvironment Design—A Paradigm Shift from "Removing CO" to "Utilizing CO"

If CO‑PROX is about "getting rid of CO," then the single‑atom catalyst‑based pathway realizes a paradigm shift of "putting CO to use"—CO is no longer an impurity to be removed but a fuel component that participates in power generation.

4.1 Single‑Atom Rhodium‑Nitrogen‑Carbon (RhNC) Catalyst System

Single‑atom catalysts anchor active metals in an atomically dispersed form on a support, maximizing metal atom utilization. To address the conventional bottleneck of low active‑site density (Rh loading typically <1 wt%), researchers developed a two‑step pyrolysis method to achieve tunable Rh loadings from 0.25 wt% to 7.43 wt%. Half‑cell tests showed that at the optimal Rh loading of 2.88 wt%, the limiting current density reached 2.4 mA·cm⁻² and the mass activity was as high as 4.18 A·mg_Rh⁻¹. In a high‑temperature single cell directly fed with CO, this catalyst delivered a peak power density of 208.4 mW·cm⁻² and maintained stable discharge output over 22 hours.

4.2 Hierarchical Microenvironment Design: Direct Syngas Utilization

A further breakthrough comes from the hierarchical electrochemical microenvironment design strategy. The research team introduced two types of catalysts simultaneously at the anode—single‑atom RhNC catalyst for the electrochemical oxidation of CO and platinum nanoparticle catalyst for the electrochemical oxidation of hydrogen. By leveraging the distinct effects of phosphoric acid electrolyte on the two catalysts and their respective reaction processes, they precisely tuned the transport, adsorption, and reaction of CO and hydrogen within the porous electrode.

Experimental data show:

• In syngas containing 10 vol.% CO, the peak power density decreased by only 19%;
• This represents a 2.5‑fold improvement over conventional HT‑PEMFC approaches;
• Under pure CO fuel, the cell still maintained stable discharge output.

This work covered a wide CO range from 0 to 100 vol.%, marking the first demonstration of direct syngas utilization over such a broad CO composition range in high‑temperature polymer electrolyte membrane fuel cells.

4.3 Technical Significance

This pathway breaks the "purify‑then‑generate" paradigm by coupling fuel processing with electrochemical conversion. Multi‑source low‑cost fuels represented by syngas can be directly used in fuel cells, promising significant system simplification and cost reduction.

5. Technical Pathway 3: Electronic Structure Modulation—Enhancing Intrinsic CO Tolerance from Catalyst Design Origin

5.1 Design Philosophy

The third pathway relies neither on external purification nor on introducing a second catalyst, but instead reduces CO adsorption strength by tuning the intrinsic electronic structure of the catalyst. Core approaches include support engineering, alloying, and d‑band center modulation.

5.2 Case Study 1: Electron‑Rich Support Modulation (PtNiCo/TiN System)

In direct methanol fuel cells (DMFCs), CO intermediates generated during methanol electro‑oxidation are a primary source of catalyst poisoning. Researchers introduced an electron‑rich titanium nitride (TiN) support to construct a novel e‑PtNiCo catalytic system.

This design achieves dual effects through strong metal‑support electronic coupling:

• Interfacial electron transfer shifts the Pt d‑band center downward, reducing the CO adsorption free energy from ‑1.62 eV to ‑1.27 eV, thereby weakening the binding strength of CO intermediates;
• The electron‑rich environment strengthens Pt‑Ni and Pt‑Co bonds, reducing the dissolution rates of Ni and Co by 2.25‑fold and 3.35‑fold, respectively, in accelerated durability tests.

In actual DMFC testing, the e‑PtNiCo catalyst exhibited only 9.6% voltage decay after 50 hours of operation at a constant current of 100 mA·cm⁻², with a peak power density retention of 89.3%; in contrast, the benchmark PtNiCo/C catalyst showed 37.7% voltage decay under the same conditions.

5.3 Case Study 2: Molybdenum Sulfide‑Modified Support

Another strategy employs molybdenum sulfide (MoS₂)‑modified conductive carbon black supports for Pt nanoparticles. The oxophilic nature of MoS₂ promotes water dissociation, providing more OH active species that accelerate the oxidative desorption of CO. This design enhances CO tolerance from the support level.

5.4 Technical Positioning

Electronic structure modulation acts at the catalyst design origin and is not mutually exclusive with CO‑PROX or single‑atom catalyst routes—in fact, it can be combined with the above strategies to achieve synergistic anti‑CO mechanisms.

6. Systematic Comparison and Application Boundaries of the Technical Pathways

Technical Pathway	Location of Action	Core Mechanism	Applicable Scenarios	Key Metrics
CO‑PROX	Upstream purification	Selective catalytic oxidation of CO	High‑purity hydrogen supply	CO reduced to <10 ppm
Single‑atom RhNC catalyst	Anode interior	Direct electrochemical CO oxidation	Direct syngas/raw hydrogen utilization	208.4 mW·cm−2 (pure CO)
Hierarchical microenvironment design	Anode interior	Dual‑functional catalytic pathways	Broad‑range syngas	Only 19% power loss at 10% CO
Electronic structure modulation	Catalyst design	Reduced CO adsorption energy (d‑band center tuning)	Methanol, formic acid, raw hydrogen systems	CO adsorption energy reduced by 0.35 eV

Each pathway has its own focus: CO‑PROX is technologically mature and directly integrable into existing PEMFC systems; the single‑atom catalyst route breaks the boundary between fuel processing and electrochemical conversion, potentially simplifying system architecture; electronic structure modulation enhances intrinsic tolerance from the catalyst design stage. These approaches are not mutually exclusive—electronic structure modulation can be combined with the former two for synergistic effects.

7. Conclusion and Outlook

The transformation of CO from a fuel cell "poison" to a usable "fuel" is essentially a reaction‑pathway reconstruction achieved through catalyst design. From CO‑PROX catalysts achieving near‑100% CO conversion at room temperature, to single‑atom RhNC catalysts delivering 208.4 mW·cm⁻² in pure CO fuel, to electronic structure modulation reducing CO adsorption energy by 0.35 eV—these quantitative data demonstrate that catalytic science is turning the "impossible" into engineering‑feasible solutions.

Current progress already shows clear engineering potential: simplifying fuel cell system architecture, reducing fuel purification costs, and broadening the diversity of fuel sources. Future research directions include: non‑precious metal alternatives to further reduce costs, multi‑technology synergy for multiple anti‑CO mechanisms, and breakthroughs in scalable manufacturing. Catalyst design is redefining the "fuel boundary" of fuel cells—CO is no longer an impurity that must be removed to trace levels, but a usable component that can be incorporated into the energy conversion chain.

author：Gloria
date:2026-06-29