Truc Pham
  • home
  • projects
  • blog
  • cartography
  • publications
  • CV
  • get in touch!
    • LinkedIn
    • ORCID
    • email
    • GitHub

On this page

  • Thesis at a glance
  • Why it matters
  • What I did
  • Key findings
    • Ion exchange works for river water, not for wastewater
    • Diafiltration with a 10 kDa membrane is the winner
  • Conclusion

Pre-treatment for real-time bacterial counting in water

MEng thesis · Nagasaki University, Japan · Monbukagakusho Honours Scholar

2018 – 2020

Full title: Development of pre-treatment for real-time bacterial counting (リアルタイム微生物計測に対する前処理手法の開発)

Topics: microbial water quality · wastewater treatment · membrane filtration · online monitoring

My contributions: experimental design, pre-treatment system operation, fluorescence and bacterial analysis

Skills used: ion exchange, diafiltration, fluorescence spectrophotometry, online bacteriological counting, epifluorescence microscopy

Thesis at a glance

Degree MEng in Water and Environmental Engineering
Programme Integrated Engineering, International Water Environmental Engineering Course, Nagasaki University
Completed July 2020
Water tested MBR-treated wastewater and river water
Key result Diafiltration with a 10 kDa membrane (from 4 diafiltration volumes) enabled error-free online bacterial counting

Why it matters

Safe water depends on knowing how many bacteria remain after treatment. Common methods such as Heterotrophic Plate Count (HPC) and Flow Cytometry (FCM) work well, but they are not suited to continuous, long-term monitoring.

Real-time bacterial counting can monitor water continuously, but it has a weak spot: dissolved organic matter, especially humic acid-like substances, gives off fluorescence that interferes with the measurement. This thesis set out to remove those interfering substances while keeping the bacteria, so they can still be counted accurately.

What I did

I developed and compared two pre-treatment methods:

1. Fluidized bed ion exchange

  • An ion exchange column (Φ18 × 300 mm) filled with commercial anion exchange resin A860
  • MBR-treated wastewater and river water pumped upward through the resin at 40 mL/min

2. Diafiltration

  • Membranes with two molecular weight cut-offs: 10 kDa and 50 kDa
  • Water concentrated at 40 mL/min and 2 bar, then diluted back with buffer, repeated for 3, 4, 5 and 6 diafiltration volumes (DVs)

All treated samples were analysed with a fluorescence spectrophotometer, an online bacteriological counting system and epifluorescence microscopy.

Key findings

Ion exchange works for river water, not for wastewater

Fluorescence of interfering substances before and after ion exchange with resin A860 (Ex 405 nm)
  • Treated river water reached low fluorescence (below 100) and could be counted online
  • Treated MBR wastewater still showed high fluorescence (around 400), too high for online counting

Diafiltration with a 10 kDa membrane is the winner

Fluorescence after diafiltration with 10 kDa (orange) and 50 kDa (grey) membranes
  • The 10 kDa membrane removed far more interfering substances than the 50 kDa membrane
  • From 4 diafiltration volumes onwards, treated wastewater could be counted online without errors

Stable real-time bacterial counts after diafiltration with the 10 kDa membrane (4, 5 and 6 DVs)

Conclusion

Both methods reduced dissolved organic matter in MBR-treated wastewater, but only diafiltration with a 10 kDa membrane removed enough interfering substances for reliable real-time counting. This makes it a promising pre-treatment step to support continuous microbial monitoring in water and wastewater treatment.


← Back to all projects

© 2026 Truc Pham

 

This website is built with GitHub, R, and Quarto