Comparative Study of the Performance of Ultrafiltration Membrane Systems For On-Site Treatment of Roof Run-Off Rainwater
D. Chakraverty (TU Delft - Civil Engineering & Geosciences)
E. Sylvestre – Mentor (TU Delft - Civil Engineering & Geosciences)
M.B. Tanis – Mentor (TU Delft - Civil Engineering & Geosciences)
M.W. Ertsen – Graduation committee member (TU Delft - Civil Engineering & Geosciences)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Urban areas increasingly face dual challenges of managing stormwater safely while reducing dependence on potable water for non-essential uses. Although decentralized rainwater harvesting is often promoted as a complementary strategy, the performance of ultrafiltration (UF) systems on real roof-harvested rainwater—particularly under different hydraulic driving forces—remains insufficiently characterized. In particular, few studies have directly compared gravity-driven membrane (GDM) and pressure-driven membrane (PDM) configurations using the same membrane material and the same variable-quality rainwater.
To address this gap, this study experimentally evaluated and compared a GDM and a PDM ultrafiltration system for on-site treatment of roof runoff at The Green Village (TU Delft). Both systems used identical 150 kDa PES flat-sheet membranes housed in Sepa CF modules. The GDM setup operated for 30 days under a constant hydrostatic pressure of 59 mbar, while the PDM system operated for 20 days under constant pump-driven pressure. The comparison aimed to quantify differences in flux evolution, operational stability, and water-quality improvements under realistic decentralized conditions.
The initial permeate flux in both systems was approximately 25 L/m²hr. In the GDM configuration, flux declined rapidly and stabilized at around 4 L/m²hr within five days, maintaining this value without cleaning or external energy input for the remainder of the 30 day run. In contrast, the PDM system exhibited a continuous decline from 23 to less than 1 L/m²hr in almost 20 days, with no sign of stabilization. Both systems reduced turbidity from an initial 1.4 NTU to below 0.4 NTU, while pH (6.9-7.5) and electrical conductivity (107-134 𝜇S/cm) remained largely unchanged, as expected for ultrafiltration. Microbiological analyses showed no detectable E. coli in any sample and total coliform concentrations above 2419.6 MPN/100 mL in the raw rainwater, indicating environmental—but not fecal—contamination of the feedwater.
These findings demonstrate that gravity-driven ultrafiltration can achieve stable, low-energy operation for decentralized rainwater treatment, whereas pressure-driven UF suffers from progressive fouling and declining performance under continuous operation. The work is novel in providing a controlled, side-by-side comparison of GDM and PDM systems using the same membrane and real roof-harvested rainwater, offering empirical insight into how operational mode influences fouling dynamics and long-term flux stability.