Jorma Keskinen

Electrical low pressure impactor

Electrical detection is used for the first time to achieve real-time operation of a low pressure cascade impactor. An instrument is developed using Berner type multijet low pressure impactor stages with minimum modification. The cut-sizes of the seven channel system range from 0.030 to 1.0 μm. A multichannel electrometer is constructed using low cost monolithic electrometer operational amplifiers. The zero-check technique is applied to achieve a lowest detectable current of 10 fA, less than a tenth of the input bias current of the operational amplifiers. The time resolution is 1 min in the lock-in mode and 1 s in the high concentration mode. The sensitivity and size resolution of the instrument are comparable to the corresponding values of the Electrical Aerosol Analyzer. The feasibility of the instrument in measuring submicron number size distribution has been tested by comparison measurements with a differential mobility analyzer. The results show quantitative agreement.

Performance evaluation of the electrical low-pressure impactor (ELPI)

The performance of the Electrical Low-Pressure Impactor (ELPI) has been evaluated using monodisperse aerosols. ELPI is a near real-time size analyzer consisting of an aerosol charger and a cascade impactor. Particle cut sizes of the cascade impactor and the charging efficiency of the charger were determined experimentally in the designed range of the instrument, i.e. 0.03–10 μm. The Stokes numbers of 50% particle cut sizes of all the impactor stages were found to vary from 0.421 to 0.483 with an average of 0.456 and standard deviation of 0.017. The collection efficiency curves were found to be steep with an average steepness (ratio of 70–30% and 30% collection efficiency) of 1.19. The charging efficiency of the charger was close to that specified by the manufacturer for particles smaller than 2 μm in diameter. For larger particles, a deviation from the manufacturer’s specifications was observed. It results from the high particle loss in the charger because of the small size of the charger, strong trapping electric field, and the perpendicular directions of the aerosol stream and the electric field in the charging zone. A comparison with SMPS for size distribution measurements was made in this study. The good agreement between the two measured size distributions shows the capability of the ELPI for near real-time particle size measurements.

Characteristics of the liquid flame spray process

Liquid flame spraying (LFS) is a new thermal spray process. Liquid feedstock is injected and atomized in an oxygen-hydrogen flame where the liquid phase is evaporated and thermochemical reactions are completed to produce fine particles. Production of nanoparticles requires a thorough understanding of the process. Therefore, various process stages were studied; i.e., the atomization of liquid feedstock, and characterization of the flame and flame-droplet interactions. Experimental techniques included laser diffraction anemometry for droplet size distribution, laser doppler velocimetry for particle velocity, pulsed laser Rayleigh back scattering for flame temperature and Schlieren photography for flame structure. Atomization is optimized with an organic solvent, such as isopropanol, nebulized with hydrogen gas at a high flow rate. Liquid droplets injected into the flame are subjected to a maximum temperature of 2600°C and are accelerated to about 160 m s−1. The flame length can be controlled by flame velocity and the solvent type. Water produces a shorter flame whereas ispropanol extends the flame. Injection of the aerosol produces a “pencil-like” region which does not experience turbulence for most of the flame length. Experimentation with manganese nitrate and aluminium isopropoxide or aluminium nitrate showed conversion to a

On-line measurement of size distribution and effective density of submicron aerosol particles

Abstract An on-line method is presented for simultaneous size distribution and particle density measurement, based on parallel measurements made by SMPS and ELPI. The measured SMPS number distribution is integrated with the ELPI response functions to produce calculated current response. By varying the density value, the best fit is sought between the calculated and measured current response. Simulation tests made showed relatively good stability against small uncertainties in the distribution and the response functions. Test measurements of well-defined aerosols with known density were conducted. Density values of 0.86, 1.1, and 1.9 g / cm 3 were measured for liquid particles of DOS, Santovac vacuum oil, and Fomblin vacuum oil, respectively. These values are within 8% of the accepted bulk values. For solid particles of NaCl, Zn, and Ag, slightly larger experimental errors in the range of 4–18.2% were found.

Sampling Conditions for the Measurement of Nucleation Mode Particles in the Exhaust of a Diesel Vehicle

A novel porous tube diluter was characterized to define sampling parameters for repeatable measurements of nucleation-mode particles (NMPs) in the exhaust of a modern diesel passenger car at moderate engine load. This porous tube diluter permits the variation of sampling parameters independently and in a wide range. We investigated the sampling parameters: primary dilution temperature (PDT; 15–55°C), primary dilution ratio (PDR; 8–45), residence time (RT; 0.5–4.0 s), and relative humidity of primary dilution (PRH; 5–90%). Decreased PDT and increased PRH led to a growth of particle number and size in the nucleation mode. While a maximum number of NMPs was found at moderate PDRs between 20 and 30, a maximum volume of NMPs was achieved at PDRs below 20. Coagulation explains the number reduction of NMPs under sampling conditions of prolonged RT and decreased PDR. However, the size growth of the nucleation mode can only partly be attributed to coagulation, and thus growth due to organic compounds from the exhaust probably plays an important role. Sensitivity analysis was conducted as a function of PDR-PDT and revealed two optimal sampling conditions for repeatable NMPs in number or repeatable maximum NMPs in volume. In addition to the sampling parameters, exhaust line conditioning was found to affect NMPs strongly and thus needs to be controlled to minimize effects on the history of measurements.

Performance Evaluation of a Novel Sampling and Measurement System for Exhaust Particle Characterization

This paper presents a novel partial flow sampling system for the characterization of airborne exhaust particle emissions. The sampled aerosol is first conditioned in a porous dilutor and then subsequent ejector dilutors are used to decrease its concentration to the range of the instrumentation used. First we examine the sensitivity of aerosol properties to boundary sampling conditions. This information is then used to select suitable sampling parameters to distinguish both the nucleation and the accumulation mode. Selecting appropriate sampling parameters, it is demonstrated that a distinct nucleation mode can be formed and measured with different instruments. Using these parameters we examine the performance of the system over transient vehicle operation. Additionally, we performed calculations of particle losses in the various components of the system which are then used to correct signals from the instruments. Several quality characteristics are then discussed, such as the repeatability and reproducibility of the measurements and the potential to derive total emission rate with a partial flow sampling system. Comparisons in different laboratories show that repeatability (intra-laboratory variability) is in the order of 10% for accumulation mode particles and 50% for nucleation mode ones. Reproducibility (inter-laboratory variability) values are in the range of ±20-30%. Finally, we compared laboratory size distributions with ambient samples obtained chasing a vehicle. This demonstrated that the sampling system accurately reproduced the accumulation mode particles as well as the potential for nucleation mode formation. This sampling system has been used in the framework of a European project for measurement of emissions of a number of light duty vehicles and heavy duty engines.

Fine particle losses in electrical low-pressure impactor

Abstract The fine particle losses in the individual impactor stages of the electrical low-pressure impactor (ELPI) were studied. Three different loss mechanisms were considered: diffusion, space charge, and image charge deposition. Diffusion losses were determined experimentally in particle size range of 10– 400 nm . The measured values varied from 0.1 to 6% depending on particle size and impactor stage. In the measurement range of the instrument, i.e. above 30 nm , the losses were below 2%. Image charge losses exceeded the diffusion losses when particle size was larger than 200 nm , but even the combined loss in this size range was below 0.5%. Space charge losses were determined both experimentally and through calculations. The space-charge effect was found to be a dominant loss mechanism in ELPI when measured concentrations were high.

Method for Measuring Effective Density and Fractal Dimension of Aerosol Agglomerates

A method to find particle effective density and the fractal dimension, based on simultaneous size distribution measurements with SMPS and ELPI, is introduced. A fitting procedure is used to find the particle density as a function of particle size and the fractal dimension. The method was tested by simulation and by experimental measurements of particles with varying morphology. For fractal dimension values between 2.2 and 3.0, fractal dimension was measured with an accuracy of 0.1, and effective density was measured with 15% relative accuracy.

Comparison of mobility equivalent diameter with Kelvin‐Thomson diameter using ion mobility data

Mobility distributions of both positive and negative cluster ions were studied in nitrogen gas enriched by ethanol, isopropanol, n‐butanol and acetone vapours, respectively. The ions were generated into the system by a radioactive source normally used as a bipolar aerosol charger. The ion mobility distributions were measured with a time‐of‐flight mobility spectrometer (TOF). The measured ion mobility can be used to estimate the mobility equivalent diameter via conventional Millikan–Fuchs formula based on Stokes law corrected by molecule slip effects. When the vapour concentration is increased the cluster ions will get larger due to enhanced molecule attachment onto ions. Here the vapour concentration is assumed to be high enough so that the chemical tracer effects would become of minor importance. Therefore the Kelvin–Thomson equation for ion induced nucleation is first assumed to be valid. The Kelvin–Thomson diameter of the ions in all cases was found to be systematically smaller than the corresponding m...

Sampling method for particle measurements of vehicle exhaust

This paper describes a new sampling concept for particle emission measurements. The purpose is to produce repeatable and reproducible conditions for nucleation phenomena. The exhaust is sampled and instantaneously diluted by inserting a porous tube diluter inside the tailpipe. This is carried out in order to prevent uncontrolled sample transformations in sampling lines. The sampling system was tested in size distribution measurement of light duty diesel vehicle. The tests showed a clearly bimodal size distribution with distinguished nuclei and accumulation modes.