Exhaust gas after-treatment of hydrogen combustion engines by selective catalytic reduction with hydrogen

Michael  Borchers, Karlsruhe Institute of Technology

Michael Borchers, Kevin Keller, Patrick Lott and Olaf Deutschmann*

Institute for Chemical Technology and Polymer Chemistry,

Karlsruhe Institute of Technology (KIT), Karlsruhe, 76131 (Germany)



Lean-burn, hydrogen-fueled internal combustion engine vehicles (H2-ICEs) may play an important role on the way towards total decarbonization of the transport sector in the upcoming years. However, significant amounts of NOx are formed during the combustion process of hydrogen, which raises the need for an efficient after-treatment system if ultra-low emission limits are to be met. Circumventing the problems of NH3-SCR (ammonia slip, urea deposits) [1], hydrogen selective catalytic reduction (H2-SCR, Eq. 1) is a prime candidate also due to the straightforward availability of the reducing agent in a H2‑ICE’s fuel tank. Although palladium [2] and platinum [3] catalysts have been reported to be the most active for the NOx reduction with H2, the side reactions (Eqs. 2‑5) especially towards N2O (Eq. 2) formation and the competing hydrogen oxidation reaction (Eq. 5) still present a challenge to be overcome for a widespread application of the H2-SCR technology. In addition, the influence of high water concentrations in the exhaust gas on the catalytic performance has to be evaluated for a rational catalyst design.

2 NO + 2 H2 → N2 + 2 H2O          (1)                                  2 NO + O2 → 2 NO2                       (4)

2 NO + H2 → N2O + H2O             (2)                                  2 H2 + O2 → 2 H2O                        (5)

2 NO + 5 H2 → 2 NH3 + 2 H2O   (3)

Therefore, the monolithic model catalyst 1%Pd/5%V2O5/20%TiO2-Al2O3 was studied in the absence and presence of water under dynamic light-off and steady‑state conditions in various realistic gas mixtures to achieve a comprehensive knowledge of the impact of gas composition on the activity and selectivity, which is crucial for advances in H2‑SCR technology [4].

Materials and Methods

The catalyst was prepared by precipitation of TiO2 on γ‑Al2O3 and subsequent incipient wetness impregnation (IWI) steps for adding Pd and V2O5. While N2‑physisorption, XRD, H2‑TPR and elemental analysis (ICP‑OES) were employed for characterization, the received powder was dip-coated onto a cordierite monolith for performing catalytic activity measurements in a synthetic gas catalyst testing bench equipped with a FTIR for gas analysis. Light‑off experiments were conducted in the 100 to 300 °C temperature range and a GHSV of 60 000 h‑1 was chosen for both, dynamic and steady‑state measurements.

Results and Discussion

In a typical gas mixture, the catalyst showed significant NO conversion and high selectivity towards nitrogen formation compared to the byproducts N2O, NO2 and NH3. The H2 oxidation is demonstrated to be the reaction that consumes the major amount of H2, particularly with increasing temperature, leading to a decline in H2‑SCR activity above 220 °C. A decrease in oxygen concentration or an increased hydrogen concentration both resulted in a further NO conversion increase, however, also lead to formation of significant amounts of ammonia. Moreover, an increased ignition temperature of the competing hydrogen combustion reaction was observed in the presence of NO, which points to a blockage of active sites by adsorption of NO, notably in the low temperature regime. The addition of 5% water was found to have only a minor impact on the catalytic activity during the light-off tests. By means of a reductive pretreatment prior to the light‑off measurement, a significant broadening of the operation window could be achieved. Although NO2 formation is thermodynamically favored, the 1%Pd/5%V2O5/20%TiO2-Al2O3 catalyst clearly does not favor simple NO oxidation (Eq. 4) under lean conditions. In addition, regarding the NO feed concentration, steady‑state experiments indicate that a well-balanced compromise between oxidizing and reducing agents must be found in order to ensure high NO conversion towards N2, which particularly occurs at lower O2 concentration, while minimizing the formation of byproducts such as N2O (Eq. 2) or NH3 (Eq. 3), e.g. by decreasing the H2 concentration.


Our study reveals a variety of challenges in the application of H2-SCR for NOx removal from H2-fueled internal combustion engines. A comprehensive knowledge of the impact of gas composition on the activity and selectivity, as presented herein, is crucial for advances in H2‑SCR technology.


  1. Börnhorst, M., Langheck, S., Weickenmeier, H., Dem, C., Suntz, R., Deutschmann, O. Eng. J. 377, 119855 (2019).
  2. Qi, G., Yang, R.T., Rinaldi, F.C., Catal. 237, 381 (2006).
  3. Hahn, C., Endisch, M., Schott, F. J., Kureti, S., Catal., B 168-169, 429–440 (2015).
  4. Borchers, M., Keller, K., Lott, P., Deutschmann, O., Ind. Eng. Chem. Res. (2021).

Download oral presentation: