*2.4. Experimental Layout and Instruments*

The experimental tests were carried out in the Energy and Propulsion Laboratory at the University of Salerno. The engine test bed, auxiliary plants, sensors and emission probes were in the engine test cell. The propulsion system was remotely controlled from an adjacent room, where the control console and the equipment for the management of the hardware, actuators and sensors were located. Figure 1 shows the engine test bed equipped with an eddy-current dyno (Borghi-Saveri, Bologna, Italy) and a light-duty EURO V common-rail diesel engine, whose main technical data are reported in Table S2.

**Figure 1.** Engine test bed.

The main hardware and software systems for the engine-dyno control are:


Focusing on the exhaust emissions, PM measurement was performed using the AVL Smoke Meter 415S (AVL List GmbH, Graz, Austria). The measurement of NO and NOx concentrations across the DPF was performed using the Cambustion CLD500 fast NOx sensor, a chemi-luminescence detector (CLD) with a response time of 3 ms.

The engine exhaust pipeline was modified in order to properly locate the DPFs and sensors. As schematically represented in Figure 2, the filter was placed downstream of the DOC, and two variable section ducts were placed upstream and downstream of it. The CDPF was wrapped in a heat-expanding intumescent ceramic mat (Interam® Mat Mount 550 by 3M Italia Srl (Pioltello, Italy) before being enclosed in the stainless-steel reactor of the experimental plant. Temperature and pressure drop across the filter were continuously monitored by means of two K type thermocouples (at the inlet and outlet of the CDPF) and a differential pressure sensor (MDTP), respectively. The latter was composed of two sensors located upstream and downstream of the filter, and its signals were continuously processed by a microcontroller that returned the pressure drop. This device was powered in voltage by an external power supply set at 5 V. K thermocouples were connected to the AVL Puma Open, while the MTDP, after the calibration with a sphygmomanometer, was connected to a TLK. The output current signal was linked via RS232 to a Lab-View interface. For the investigation of passive regeneration with the CDPF, the measurement of NO and NO2 concentrations was performed using the Cambustion CLD 500 (Cambustion, Cambridge, UK), equipped with two independent probes for NO and NOx sensing. An alternate sampling, arranged like the soot measurement described above, allowed investigation of the variation in NOx concentration across the filter [22]. The filter assembly in the exhaust pipe and the CDPF placed in the reactor are shown in Figure 3. The main sensors and measurement instruments are listed in Table S3, with the corresponding accuracies.

**Figure 2.** Schematic representation of the experimental layout and sensors.

The PSDs were measured by means of an SMPS system consisting of an electrostatic classifier (EC; TSI Model 3082 (TSI, Shoreview, MN, USA)), a nano-differential mobility analyzer (nano-DMA; TSI Model 3085 (TSI, Shoreview, MN, USA)) and an advanced aerosol neutralizer (TSI Model 3088 (TSI, Shoreview, MN, USA)) linked to a condensation particle counter (CPC; TSI Model 3750 (TSI, Shoreview, MN, USA)). A sample flow rate of 0.3 L/min and a sheath flow rate of 3 L/min in the EC were chosen as operating conditions, allowing the analysis of particle sizes in the range of 4.5–160 nm. For each sample, the data were reported as averaged values of multiple scans, where the duration of a single scan was set to 60 s. The PSD measurements were performed by diluting the aerosols sampled from the engine before entering the SMPS system. The dilution tool was a two-stage dilution system obtained from Dekati (fine particle sampler; Model FPS-4000 (Dekati, Kangasala, Finland)), and two dilution ratios (DRs) of 80 and 130 were used, depending on the operating condition. Moreover, the dilution temperature of the first stage was set at 25 ◦C. The dilution ratio was properly chosen in order to avoid the condensation/coagulation of particles along the sampling line. In other research papers, higher DRs were used for

particle analysis, and heated lines were employed; however, it is worth noting that the specific experimental setup strongly influences the critical DR; the important parameters are PN, gas phase composition in the sampling line and residence time. The critical DR was checked in all the investigated conditions [46–48]. In this work, two different DRs were used to obtain, after the CDPF, a higher signal-to-noise ratio by using the lower DR, still achieving the critical dilution conditions.

**Figure 3.** CDPF in the reactor used in the experimental campaign (**a**,**b**) and the filter assembly in the exhaust pipe (**c**).
