Design and development of a PM10 multi-inlet cyclone and comparison with reference cyclones

Prashant Patel, Shankar G. Aggarwal*, Thi Cuc Le, Khem Singh, Daya Soni, Chuen Jinn Tsai

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Size-segregated sampling of particulate matter (PM) using impactor suffers from D50 cutoff shift due to particle loading and re-entrainment problems. Cyclonic separation is a viable option to overcome the above problem. However, conventional reverse flow cyclone design having a single inlet and upward-facing outlet also presents a common issue of sample (particle) loss during sampling and requires several arrangements to convert it into an efficient PM sampler. Therefore, here we present a high-volume (HV) PM10 multi-inlet cyclone (MIC) design with a downward-facing outlet, which overcomes existing problems and has additional advantages, such as omnidirectional sampling where a filter collector is placed in a straight line below the cyclone outlet to minimize sample loss. Moreover, like the existing USEPA reference low-volume PM2.5 sampler inlet design, which consists of 2-impactor stages (PM10 followed by PM2.5) in a straight path, this developed HV PM10 MIC sampler can accommodate a second size fractionator (e.g., PM2.5 impactor) to sample finer-size PM on a filter. D50 cutoff of developed PM10 MIC is numerically and experimentally investigated. Since the study regarding cutoff size of another type PM10 cyclone, called respirable dust sampler (RDS) is not available in the public domain and is widely used for PM10 monitoring in India, we investigated its cutoff size empirically and experimentally, and also performed field comparisons. Collocating field evaluation of PM10 MIC and PM10 RDS cyclone was done under a wide range of particle mass loading, and results were compared with USEPA-approved high-volume PM10 impactor sampler and with a real-time particle sizer. The D50 cutoff of PM10 MIC is experimentally achieved to be 9.89 ± 0.3 µm, which is close to 9.94 µm predicted numerically and lies in the range of 9.5–10.5 µm size measured by others for PM10 impactor sampler (USEPA). The D50 cutoff of the PM10 RDS cyclone is experimentally determined to be 3.56 ± 0.1 µm, which is surprisingly lower than its claimed cutoff of 10 µm mentioned in numerous articles, where it has been used for air quality reporting and studies related to aerosol science. The field comparison correlation of PM10 MIC for PM10-2.5 levels with PM10 sampler (USEPA) (R = 0.99) and particle sizer (R = 0.94) correlated well, and the mean deviations are found to be 6.2% and 3%, respectively. While PM10 (RDS) cyclone poorly correlates (R = 0.67), and the mean deviation is 68%. Overall, the developed PM10 MIC overcomes issues associated with existing impactor and conventional cyclone sampler, and can be a better option for high-volume PM10 sampling, especially under a wide range of ambient conditions particularly where the particle mass loading is consistantly high.

Original languageEnglish
Pages (from-to)1955-1968
Number of pages14
JournalAir Quality, Atmosphere and Health
Volume16
Issue number10
DOIs
StatePublished - Oct 2023

Keywords

  • Cyclone separators
  • D cutoff
  • Multi-inlet cyclone (MIC)
  • PM samplers
  • Respirable dust sampler (RDS)

Fingerprint

Dive into the research topics of 'Design and development of a PM10 multi-inlet cyclone and comparison with reference cyclones'. Together they form a unique fingerprint.

Cite this