**2. Methods**

The research facility included a passenger car engine that was modified to run with natural gas. The engine with the test facility was presented in detail in Murtonen et al. [29]. The engine was operated with a lean air-to-fuel mixture. For the present study, only one engine-driving mode was used. The exhaust gas flow and temperature (measured upstream of the catalysts) were adjusted independently, and therefore, it was possible e.g., to keep the exhaust gas composition and flow constant while changing only the temperature. In the present study, most of the tests were conducted with exhaust temperature adjusted to 550 ◦C and exhaust flow to 60 kg/h while initial performance studies were also conducted at lower exhaust temperatures and with different exhaust flows.

The NG was from Nord Stream and was high in CH4 content (>95%). The sulfur content of the gas was below 1.5 ppm. The lubricating oil used for medium speed LNG engines was selected and used in this smaller test engine also. The lubricating oil sulfur content was 2100 mg/kg, density was 0.85 kg/dm<sup>3</sup> and viscosity at 100 ◦C was 13.4 mm2/s.

MOC used platinum-palladium (1:4) as active metals on a tailored coating developed for lean NG applications and supported on a metallic substrate. The volume of MOC was 0.75 dm3. This means a space velocity (1/h) of 61,500 with exhaust flow of 60 kg/h. In addition to MOC only, similar studies were done with a SOx trap installed in front of the MOC. The SOx trap has a property to absorb and release SOx in controlled conditions and had a volume of 0.75 dm3. The catalyst set-up is presented in Figure 1.

**Figure 1.** Set-up.

Emission measurements were done upstream and downstream of the catalyst system (see Figure 1). This included a Horiba PG-250 analyzer used to measure NOx, CO, CO2, and O2. CO and CO2 were measured by non-dispersive infrared, NOx by chemiluminescence, and O2 by a paramagnetic measurement cell. Exhaust gas was dried with gas-cooler before it was measured by Horiba.

Online SO2 emissions were determined by a Rowaco 2030 1 Hz Fourier transfer infrared (FTIR) Spectrometer equipped with an Automated MEGA-1 (miniMEGA) sampling system. The detection limit for SO2 was 2.5 ppm. This device was used mainly only during the regeneration periods to observe the SO2 release.

An agilent 490 MicroGC (Gas Chromatography) was used to measure the hydrocarbons and hydrogen (H2). The detection limits for ethane (C2H6), ethylene (C2H4), and propane were approximately 2 ppm, for methane (CH4) 10 ppm, and 100 ppm for H2.

Multiple gaseous components were measured continuously by two Gasmet DX-4000 Fourier transformation infrared (FTIR) spectrometers simultaneously upstream and downstream of the catalyst system.
