Electronic Supplementary Information (ESI): Particle Emissions of a Heavy-Duty Engine Fueled with Polyoxymethylene Dimethyl Ethers (OME)

OME as sustainable alternative diesel fuel burns without the agglomeration of soot particles which is the case for fossil and paraffinic diesel fuels. Nuclei mode particles are mainly of a volatile nature and do not survive exhaust aftertreatment.


Scheme of the test bench setup
X X X X Change from diesel to OME; change of the injectors; removal of the DPF (**) ; cleaning of the impactor Step sizes of the DMA

Calculation of particle losses † Maximum tube Reynolds number and maximum particle Reynolds number (according to Hinds (1))
The following calculations describe the respective maximum or minimum of each value and therefore enable the decision of whether the flow is laminar or turbulent. According to Hinds, the flow is laminar for Ret < 2000, but not turbulent as long as Ret < 4000 (1).
Since Rep decreases for smaller particles and higher aerosol temperature, and Ret decreases with higher aerosol temperature, the assumption of laminar flow in all parts of the sampling system is valid.

Gravitational settling in the inlet (according to Willeke & Baron(2))
The following calculations describe the respective maximum or minimum of each value and therefore lead to the minimum penetration rate through the inlet. Since rp increases for smaller particles and higher aerosol temperature, the neglect of gravitational settling in the inlet is valid. †The calculations were performed using Matlab R2019b. Therefore it used more digits than indicated in this document. The EXCEL-Tool "aerocalc" by Paul Baron was used for the specific formulas.

Sedimentation (according to Willeke & Baron (2))
The following calculations describe the respective maximum or minimum of each value and therefore lead to the minimum penetration rate inside the tubing. Penetration rate rp: Since rp increases for smaller particles and higher aerosol temperature, the neglect of gravitational settling in the tubing is valid.

Bent tubing (according to Willeke & Baron (2))
The following calculations describe the respective maximum or minimum of each value and therefore lead to the minimum penetration rate through bent tubing.
Since rp increases for smaller particles and higher aerosol temperature, the neglect of losses in bent tubing is valid.

Coagulation (according to Willeke & Baron (2))
The following calculations describe the respective maximum or minimum of each value and therefore lead to the maximum coagulation rate. Furthermore, the initial particle concentration considers monodisperse aerosol of the total concentration.

Thermophoretic velocity (according to Hinds (1) and Willeke & Baron (2))
The following calculations describe the respective maximum or minimum of each value and therefore lead to the maximum thermophoretic velocity tubing. Since vT decreases for lower aerosol temperature and lower temperature gradients, the neglect of thermophoretic losses is valid.

Diffusional losses in a cylindrical tube-fraction passing through tube under laminar flow (according to Willeke & Baron (2))
Since the tubing length between the catalytic stripper and the SMPS is the dominant part in this calculation, the temperature inside the tubing is assumed to be 20°C. The maximum deviation in penetration efficiency between an aerosol temperature of 20°C and 220°C is less than 1.06% absolute for a particle diameter of 6 nm. The transportation losses of the ejector diluters were assumed to be 5% for each diluter and for any particle diameter, according to the measurements of Giechaskiel et al. (3).

Electrostatic losses
Transport losses due to electrostatic fields were neglected due to the usage of stainless steel wherever possible and an intermediate connection using Tygon tubing. This polymer is known as a tubing material having lower electrostatic losses than other kinds of tubing (4-6).

Particle losses inside the catalytic stripper
The manufacturer of the catalytic stripper (Catalytic Instruments GmbH & Co. KG) provide in the manual, penetration efficiency data at nominal flow (10 l/min): Figure 2 shows the calculated penetration efficiencies of the purpose-built sampling systems with and without the CS or the second dilution stage. The results of the PSD in this work use the PCRF of these calculations. Furthermore, the "Aerosol Instrument Manager" software by TSI includes the option of considering the diffusion losses inside the SMPS and a multiple charge correction. The evaluations in this study include these considerations. Figure 2. Calculated particle losses. The losses due to Brownian diffusion are based on calculations according to Hinds (1) with the assumption of a laminar flow inside the tubing. The losses of each ejector diluter were assumed to be 5% according to Giechaskiel et al. (3). The manufacturer of the catalytic stripper determined the respective penetration efficiency at a nominal flow rate of 10 l/min.