Clean energy and climate
UrbanSentinel treats water without grid electricity, without added chemicals, and with salvaged parts wherever they will do. In a country whose biggest export is washed in its rivers and whose grid runs almost entirely on fossil fuels, that is the point of the design, not a side benefit.
Why energy is part of the water problem
Conventional treatment needs reliable grid power, a steady supply of chemicals and new equipment. In Bangladesh, all three are hardest to find exactly where the water is worst.
of Bangladesh's electricity comes from fossil fuels. Renewables supplied about 2.3% of grid generation.
Ember; IEEFA (2026)people were affected by the August 2024 floods in eastern Bangladesh. Floods often cut the power when water is most contaminated.
UNICEF (2024)of e-waste a year in Bangladesh, mostly handled informally, with only seven registered recyclers.
Transparency International Bangladesh (2025)Four design choices
1. Solar power instead of the grid
A solar panel charges a 12–18 V battery pack assembled from recovered laptop and phone cells. A zero-voltage-switching driver steps that up to 10–12 kV at the electrode, with a peak of 55 W. Nothing connects to the grid, so the unit keeps working in a flood, at a remote outfall or on a coastal embankment. Built
2. Circular hardware from electronic waste
The plasma driver is built almost entirely from salvaged parts. Its first version, the 2024 Stockholm Junior Water Prize Bangladesh winner, cost about US$5 in parts, according to the Stockholm Water Foundation project page. Built
| Part | Salvaged from | Role in UrbanSentinel |
|---|---|---|
| N-channel MOSFETs | Scrapped auto-rickshaw motor-driver boards | High-frequency switching in the plasma driver |
| Flyback transformer | Discarded CRT televisions | Steps 12–18 V up to 10–12 kV |
| Lithium-ion cells | End-of-life laptop and phone batteries | 12–18 V solar-charged battery pack |
| Carbon electrodes | Spent zinc–carbon dry cells | Electrodes for the plasma discharge |
| Resistors, inductors, diodes | Salvaged TV, UPS and inverter boards | Driver and tuning circuit |
| LEDs, laser, photodiode, CCD sensor | Old electronics | Optics of the testing module |
3. Chemistry made in the water, not added to it
The discharge sits about 2 cm above the water, with the second electrode immersed. It forms hydroxyl radicals, ozone, hydrogen peroxide, reactive nitrogen species and ultraviolet light at the water's surface. These break apart azo dyes and inactivate bacteria. No chlorine, coagulants or other reagents are added, so there is no chemical sludge to dispose of.
The treatment is non-thermal: over a 30-minute run at 55 W, the bulk water warmed by just 2.9 °C. Measured
Air plasma does acidify water slightly (pH 7.2 to 6.4 in tests), so pilot units restore pH with aeration and limestone and check nitrate and nitrite on every batch.
4. Only the energy the water needs
Without a fast test, an operator must dose every batch as if it were the worst case. UrbanSentinel re-tests in seconds, so it can stop the moment the water passes. On one reactor, the energy needed ranges about 500-fold between a disinfection pulse and a dye cycle, which is where the saving comes from.
The energy math
Energy is power multiplied by time. At a constant 55 W, the three measured treatment cycles use very different amounts.
E = P × t
Disinfection: 55 W × 3 s = 0.046 Wh
BOD and COD: 55 W × 15 min = 13.75 Wh
Textile dye: 55 W × 25 min = 22.9 Wh
Mean energy per batch, where f is the share of each kind of event:
Ē = f₁E₁ + f₂E₂ + f₃E₃
Saving against worst-case dosing:
1 − Ē ÷ 22.9 Wh
Energy per treatment cycle
Watt-hours, measured cycles at a constant 55 W. A dye cycle uses about 500 times the energy of a disinfection pulse.
Show the data as a table
| Cycle | Power × time | Energy (Wh) |
|---|---|---|
| Disinfection | 55 W × 3 s | 0.046 |
| BOD and COD | 55 W × 15 min | 13.75 |
| Textile dye | 55 W × 25 min | 22.9 |
Try the energy model
Set the mix of events a unit handles. Shares always add up to 100%.
Mean energy per batch
5.1Wh
0.70 × 0.046 + 0.20 × 13.75 + 0.10 × 22.9 = 5.1 Wh
78% less than dosing every batch for the worst case (22.9 Wh).
Modelled Savings are modelled from the measured cycles. The Phase 1 pilot will measure the real mix of events.
Where UrbanSentinel meets the Sustainable Development Goals
Eight goals, each tied to a specific UN target and to evidence from the project. The label on each row says how far that evidence goes.
-
SDG 6: Clean water and sanitation
- 6.3 Improve water quality by reducing pollution and halving untreated wastewater
- 6.1 Safe and affordable drinking water for all
Disinfects in seconds (total coliform 226 → 0 per 100 mL) and polishes textile effluent before it reaches rivers. The Pond Sand Filter add-on targets drinking water on the coast.
Measured -
SDG 7: Affordable and clean energy
- 7.2 Increase the share of renewable energy
- 7.3 Improve energy efficiency
Runs on a solar-charged 12–18 V battery instead of a grid that is about 98% fossil-fuelled, and stops at compliance instead of dosing for the worst case.
Modelled -
SDG 12: Responsible consumption and production
- 12.4 Sound management of chemicals and wastes, reducing releases to water and soil
- 12.5 Reduce waste through reuse and recycling
Reuses transistors, transformers, lithium cells and electrodes from e-waste, adds no treatment chemicals, and gives textile supply chains a batch-by-batch discharge record.
Built -
SDG 13: Climate action
- 13.1 Strengthen resilience to climate-related hazards and natural disasters
Works off-grid when floods and cyclones cut the power. An earlier portable unit served about 170 families in Feni during the August 2024 floods.
Field use, 2024 -
SDG 3: Good health and well-being
- 3.9 Reduce illness from water pollution and hazardous chemicals
- 3.3 End epidemics of water-borne diseases
E. coli was found in the drinking water of 81.9% of Bangladeshi households in 2019, against 40.3% at the source. Plasma removed all detectable E. coli (93 to 0 per 100 mL) in 2–3 seconds.
Measured -
SDG 9: Industry, innovation and infrastructure
- 9.4 Retrofit industries with clean and environmentally sound technologies
Designed as an add-on stage for the effluent plants factories already have, rather than a replacement for them.
Designed -
SDG 11: Sustainable cities and communities
- 11.6 Reduce the adverse environmental impact of cities
A network of units at outfalls and water intakes is designed to flag discharges and warn intakes 2–6 hours ahead (Phase 2).
Designed -
SDG 14: Life below water, and 15: Life on land
- 14.1 Reduce pollution from land-based activities
- 15.1 Conserve freshwater ecosystems
The Phase 1 pilot is modelled to keep 0.6–1.8 million m³ of untreated effluent a year out of the Turag, and salvaged electronics stay out of soil and water.
Modelled
Target wording paraphrased from the UN Sustainable Development Goals. Household E. coli figure: BBS and UNICEF, MICS 2019 water quality report.
What the project does not claim yet
Credible climate claims need numbers behind them. These are the gaps the pilot is designed to close.
- No carbon figure yet. The device runs on solar power, but the embodied footprint of the panel, enclosure and refurbishment has not been assessed.
- Energy savings are modelled. They come from measured cycles; the real mix of events will be measured in Phase 1.
- A polishing stage, not a full plant. BOD (7 mg/L against a 5 mg/L limit) and pH missed their limits, so the reactor complements biological treatment rather than replacing it.
- PFAS removal, not destruction. One test lowered PFOA and PFOS; the repeat will add a no-plasma control, a short-chain PFAS panel and a fluoride measurement.
Sources
- Ember, Bangladesh electricity data
- IEEFA (2026), Fostering Bangladesh's energy transition
- UNICEF (2024), floods in eastern Bangladesh
- Transparency International Bangladesh (2025), e-waste management
- Stockholm Water Foundation, High Voltage Plasma Water Purifier from E-Waste (2024)
- BBS and UNICEF (2019), MICS water quality thematic report
- Measured values: UrbanSentinel campaign logs at BUET's water quality laboratory and the BEXIMCO effluent treatment plant