water · health · humanitarian · chemistry · family: the solution exists but nobody can afford it
arsenic removed in thelab, not in the village
Electrochemical arsenic removal works in the lab but cannot scale to serve 140 million affected people
Problem statement
An estimated 140 million people in at least 70 countries have been drinking water containing arsenic above the WHO provisional guideline value of 10 μg/L, with Bangladesh and West Bengal being among the most severely impacted regions. UC Berkeley/LBNL's ECAR (ElectroChemical Arsenic Remediation) technology, developed in Ashok Gadgil's group, demonstrated that passing current through iron electrodes in contaminated water generates iron hydroxide particles that adsorb arsenic — in testing described in the patent application, reducing concentrations from 500+ ppb to below the WHO limit of 10 ppb. The approach is simple and requires only a minimal supply chain (steel plates, low-voltage electricity and a small amount of alum coagulant). Yet despite peer-reviewed field trials in West Bengal and a 2014 license to an Indian company (Luminous Water Technologies), the technology had, in the sources cited here, reached a 600 L field reactor and a planned 10,000-litre-per-day trial; none of those sources reports village-scale service. The US patent application was abandoned, and WHO reports that millions of people are still exposed to arsenic at concentrations much higher than its guideline value.
Why this matters
Chronic arsenic exposure can cause cancers (skin, bladder, lung) and is associated with cardiovascular disease and developmental effects. In Bangladesh alone, arsenic exposure was estimated to account for as many as 43,000 adult deaths annually (Flanagan, Johnston & Zheng, Bulletin of the WHO, 2012). A Bulletin of the World Health Organization paper called the Bangladesh situation "the largest poisoning of a population in history" (Smith, Lingas & Rahman, 2000). Affected communities are predominantly rural, low-income, and dependent on tube wells drilled into contaminated aquifers. Field solutions that apply co-precipitation and adsorption have failed, mainly because of the requirements for reagent dosing, the exchange of adsorption media, and non-compliance of users (Otter et al., 2017). The ECAR field trial estimated operating costs of $0.83–1.04 per cubic meter (amortized capital plus consumables) — a fraction of a US cent per liter — yet even at that level, sustained operation depends on a viable local service model, not just chemistry.
What’s been tried and why it hasn’t worked
ECAR generates its adsorbent (iron hydroxide) in situ by dissolving sacrificial iron electrodes, removing the need for a central adsorbent manufacturing plant or imported media, though it still uses a small amount of alum as a coagulant. In a 3.5-month field trial in West Bengal, a 600 L ECAR reactor consistently reduced real groundwater at roughly 266 μg/L arsenic to below 5 μg/L while meeting international standards for iron and aluminum (Amrose et al., 2014). Even so, the trial depended on routine work: a trained local operator lightly brushed the electrode plates in place at intervals from alternate days to alternate weeks, the iron precipitates were settled out with alum, and the arsenic-laden sludge had to be collected and stabilized — which the paper calls an important component of the operating cost — with larger and longer field trials across diverse aquifers still needed (Amrose et al., 2014). Low-tech units such as the Sono filter are in use in West Bengal and Bangladesh, but a survey cited by Otter et al. (2017) found 475 of 570 arsenic removal plants across West Bengal not useful, with adsorption media exhausted earlier than predicted and users put off by longer collection times, manual pumping and discoloured water. A solar-driven inline-electrolytic system tested long-term in West Bengal achieved 94% arsenic removal in its final operating phase (165 ± 17 μg/L reduced to 10 ± 4 μg/L) but still relies on online monitoring, daily filter backwashing, and manual cleaning of the electrolytic cell when calcareous deposits formed (Otter et al., 2017). The common failure mode across all approaches is not the chemistry — it is sustaining operations: household and community-scale treatment units have been quickly abandoned because they are not maintained, repaired, acceptable or affordable (Amrose et al., 2014).
What would unlock progress
The breakthrough needed is less about chemistry and more about system design for sustained operation with zero maintenance. A passive, zero-energy arsenic removal system with no moving parts, no consumable electrodes, and no waste management requirements would be a major advance. Research into permeable reactive barriers (iron-based materials installed in well bore paths), biochar-iron composites, or naturally regenerating adsorbents could point toward maintenance-free solutions. Alternatively, a community-scale business model that bundles water treatment with revenue-generating services (e.g., mobile phone charging, agricultural information) could sustain operations economically.
Entry points for student teams
A student team could design and test a gravity-fed, passive arsenic adsorption column using locally available iron-bearing materials (steel wool, iron filings, laterite soil), and reach a six-month endurance claim inside a semester by loading the column on an accelerated schedule — a scaled-down bed run at proportionally higher flow, with elevated influent arsenic, so that the bed volumes a village column would see in half a year pass through in a few weeks — then reporting breakthrough against bed volumes treated rather than calendar time and stating plainly which ageing mechanisms (biofouling, seasonal chemistry, intermittent use) the acceleration cannot reproduce. That door needs a wet lab that will approve arsenic stock solutions, and confirming breakthrough at the 10 µg/L WHO guideline value needs ICP-MS or hydride-generation atomic absorption — a campus analytical core or a commercial environmental lab charging per sample, not bench equipment — so a team without that access can still run the columns at higher, colorimetric-field-kit-readable concentrations and report the curve shape, saying so. Alternatively, a team could develop a business model canvas for community-scale arsenic treatment in rural Bangladesh, identifying revenue streams, supply chains, and maintenance models, testing it against the ECAR trial's documented $0.83–1.04/m³ operating cost and the maintenance burden that stalled the earlier installations; that door needs no laboratory, chemicals or partner at all. Skills in environmental engineering, water chemistry, materials testing, and social enterprise design would be most relevant.
Genes
Structural cousins: stuck the same way, in other fields
Sources
S.E. Amrose, S.R.S. Bandaru, C. Delaire, C.M. van Genuchten, A. Dutta, A. DebSarkar, C. Orr, J. Roy, A. Das & A.J. Gadgil, "Electro-chemical arsenic remediation: field trials in West Bengal," Science of the Total Environment 488–489 (2014) 539–546, World Health Organization, "Arsenic" fact sheet, S.V. Flanagan, R.B. Johnston & Y. Zheng, "Arsenic in tube well water in Bangladesh: health and economic impacts and implications for arsenic mitigation," Bulletin of the World Health Organization 90 (2012) 839–846, Context: US20110215001A1, "Electrochemical Removal of Arsenic" (abandoned patent application), P. Otter et al., "Arsenic Removal from Groundwater by Solar Driven Inline-Electrolytic Induced Co-Precipitation and Filtration—A Long Term Field Test Conducted in West Bengal," International Journal of Environmental Research and Public Health 14(10) (2017), A.H. Smith, E.O. Lingas & M. Rahman, "Contamination of drinking-water by arsenic in Bangladesh: a public health emergency," Bulletin of the World Health Organization 78 (2000) 1093–1103, Berkeley Lab News Center, "Indian Company Licenses Berkeley Lab Invention for Arsenic-free Water," 2014-03-05, All accessed 2026-08-20. go to source 1 ↗ go to source 2 ↗ go to source 3 ↗ go to source 4 ↗ go to source 5 ↗ go to source 6 ↗ go to source 7 ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
Inventor Ashok Gadgil is a professor of civil and environmental engineering at UC Berkeley and, at the time of the 2014 license, headed Berkeley Lab's Environmental Energy Technologies Division (Berkeley Lab News Center, 2014). The ECAR patent application (US20110215001A1) was made in part with US Department of Energy funds and assigned to The Regents of the University of California. The Indian licensee, Luminous Water Technologies, planned to run ECAR as village-owned micro-utilities, operating and maintaining each unit and selling the water, and a 10,000-litre-per-day trial over 15 months was in preparation at the time (Berkeley Lab News Center, 2014); the sources cited here do not report what happened after that. The Sono filter is named by Otter et al. (2017) among the low-tech units in use in West Bengal and Bangladesh. WHO classifies arsenic in drinking water as one of its top 10 chemicals of public health concern. Flanagan, Johnston & Zheng (2012) warn that exposure in Bangladesh could worsen as the population grows and installs more private tube wells. Related: water-field-pathogen-detection.md shares the infrastructure and deployment challenges.
Reconciliation 2026-08-20: re-sourced after the citation-drift sweep flagged this brief as sole-sourced to an abandoned patent application while carrying six-plus quantitative claims not in the patent. Primary sources are now the peer-reviewed ECAR field trial (Amrose et al., Science of the Total Environment, 2014), the WHO arsenic fact sheet (source of the 140 million people / at least 70 countries figure and the top-10-chemicals designation), and Flanagan, Johnston & Zheng (Bulletin of the WHO, 2012; source of the ~43,000 annual adult deaths estimate, the sum of an estimated 24,000 deaths from exposure >50 μg/L plus up to 19,000 from 10–50 μg/L); the patent moved to context. Corrections to the body: "a successful patent application" removed (the application was abandoned — that fact was already stated); the ECAR field trial was in West Bengal, India, not Bangladesh; the unverifiable "$0.01 per liter" cost requirement was replaced with the trial's documented $0.83–1.04/m³ operating-cost estimate; the unsourced "$2/day incomes" figure was softened to "poor rural communities"; an unsourced claim that ECAR "can run on solar power" was dropped (the verified solar result belongs to the separate inline-electrolytic system of Otter et al., 2017, whose 94% figure and maintenance requirements are now cited exactly); "the WHO calls it the largest mass poisoning in history" was corrected to the actual quote and attribution ("the largest poisoning of a population in history," Smith, Lingas & Rahman, Bull WHO 2000); the Indian licensee was named (Luminous Water Technologies, per Berkeley Lab News Center, 2014). Not re-verified in this pass: the closing aquifer-depletion worsening mechanism above.
Reconciliation 2026-08-21: Entry-point realism pass (C37 triage, score 1). The flag was correct on both halves: the column door asked for "zero-maintenance operation over 6+ months" of breakthrough data, which no semester contains, and it silently assumed the analytics — confirming breakthrough at the WHO 10 µg/L guideline value requires ICP-MS or hydride-generation AAS, not bench instrumentation. Repaired by accelerated column loading (a scaled bed run at proportionally higher flow and elevated influent arsenic, reported against bed volumes treated rather than calendar time), which closes the semester gap while making the extrapolation limits explicit, plus an access line naming what the ppb measurement takes and who owns it, and a fallback of running at field-kit-readable concentrations and reporting curve shape. Unflagged-door check: the business-model-canvas door was read against the ≥2-doors rule and is genuinely facility-free, so it stands as the no-facility door; it is now anchored to two figures already verified in this brief's 2026-08-20 re-sourcing (the ECAR trial's $0.83–1.04/m³ operating cost and the maintenance burden at the installations), so no new resource is cited and none needed fetching. No external URLs were added in this pass.
Reconciliation 2026-09-30: Fact repair after reading the whole write-up against every source it names. Fetched 2026-09-30: the Amrose et al. field-trial paper in full text from the US Department of Energy's OSTI copy (https://www.osti.gov/servlets/purl/1823601, 8 pages; the PubMed link gives the abstract and citation record, read through the NCBI E-utilities because pubmed.ncbi.nlm.nih.gov served a cookie wall); the WHO arsenic fact sheet in full; the Flanagan, Johnston & Zheng paper in full text from PubMed Central (PMC3506399; the PubMed link gives the record); the Google Patents page for US20110215001A1 in full; the Otter et al. paper in full text at the PMC link; the Smith, Lingas & Rahman page at the PMC link (abstract only for the first reader; the second reader read the full PDF, 10 pages, from an Internet Archive copy of the PMC PDF); and the Berkeley Lab News Center release in full (dated 2014-03-05). (1) The text said ECAR "requires no chemical supply chain" and that in-situ adsorbent generation "eliminat[es] the supply chain for chemical reagents." The field-trial paper says the system "requires a minimal supply chain (e.g. steel plates, low-voltage electricity, and readily available alum coagulant)" and that alum at 6–15 mg/L was added in every batch to settle the iron; both sentences now say a minimal supply chain and name the alum. (2) The text said the technology "has not scaled beyond small pilot and demonstration installations" and that "the vast majority of affected people still drink arsenic-contaminated water." No source says either. The paper reports a 600 L field reactor; the Berkeley Lab release says a 10,000-litre-per-day, 15-month trial was being prepared; WHO says "millions of people around the world are exposed to arsenic at concentrations much higher than the guideline value." The sentences now say that the technology had, in the sources cited here, reached a 600 L field reactor and a planned 10,000-litre-per-day trial, that none of those sources reports village-scale service, and WHO's figure. "Indian operator" became "Indian company", the release's word; "peer-reviewed field trials" stays as written, since that is the paper's own title. (3) "Chronic arsenic exposure causes cancers ... cardiovascular disease, and developmental effects": WHO says long-term exposure "can cause cancer" and that cardiovascular disease and developmental effects "may be associated" with it; the sentence now says can cause / is associated with. (4) The sentence sorting existing technologies into reagent-dependent coagulation-filtration, hazardous-waste-generating adsorption media and too-expensive reverse osmosis is in no source. It now gives Otter et al.'s finding that field solutions "have failed, mainly due to the requirements for reagent dosing, the exchange of adsorption media and the non-compliance of users." (5) The sentence that "scaling has stalled" because electrodes need periodic replacement, suspended iron must be filtered, sludge must be disposed of and a solar-powered system needs technical maintenance is in no source (the solar system is Otter's, not ECAR's, and Otter says it runs with nearly no maintenance). It now reports what the field-trial paper documents: light brushing of the electrode plates by a local operator at intervals from alternate days to alternate weeks, settling with alum, sludge collection and stabilization as "an important component of the operating cost," and the paper's call for larger, longer trials across diverse aquifers. (6) The SONO filter sentences (an iron-sand gravity filter invented by Abul Hussam of George Mason University, winner of the Grainger Challenge Prize, cheaper and simpler but limited in capacity) are in no source; only Otter et al. mention the Sono filter, as one of several low-tech units in use in West Bengal and Bangladesh. Both the body sentence and the Source Notes sentence now say what Otter says, and the body sentence carries Otter's survey figure of 475 of 570 arsenic removal plants in West Bengal found not useful, with media exhausted earlier than predicted and users put off by collection times, manual pumping and coloured water. (7) "periodic cleaning of the electrolytic cell" became "manual cleaning of the electrolytic cell when calcareous deposits formed," which is what Otter reports. (8) "little technical infrastructure and little government support for water treatment" is in no source; the sentence now gives the field-trial paper's statement that household and community-scale treatment units "have been quickly abandoned because they are not maintained, repaired, acceptable, or affordable." (9) Source Notes: Gadgil as "senior scientist at LBNL, known for developing the UV Waterworks disinfection system" — the Berkeley Lab release calls him a scientist who heads the Lab's Environmental Energy Technologies Division and a professor of civil and environmental engineering at UC Berkeley; UV Waterworks appears in no source and was removed. (10) The patent application "was filed via the US Department of Energy": the patent says the invention was made in part with Department of Energy funds under a DOE contract; the sentence now says that. (11) "An Indian licensee operated a small ECAR plant selling arsenic-safe water, but scaling beyond individual installations proved economically challenging": the release says Luminous planned village-owned micro-utilities it would operate and maintain, selling the water, and that a larger trial was in preparation; no source reports an operating plant or a later outcome. The sentence now says that and says the sources do not report what happened after. (12) "The problem is worsening as aquifer depletion concentrates arsenic in remaining groundwater" (left unchecked in the 2026-08-20 note) is in no source; it now carries Flanagan, Johnston & Zheng's warning that exposure could worsen "as the population grows and installs more private tube wells." Confirmed in the sources and left as written: the Source line citations for all seven sources; 140 million people in at least 70 countries above the 10 μg/L provisional guideline and arsenic as one of WHO's 10 chemicals of major public health concern (WHO); 580 and 520 ppb reduced to below 10 ppb in the patent's tests, the abandoned status, the Regents assignment and the patent title; the 600 L reactor, 3.5 months, West Bengal, ~266 μg/L to below 5 μg/L, iron and aluminum standards met, and $0.83–1.04/m³ amortized capital plus consumables ($0.0008–0.001/L) (Amrose et al.); 24,000 plus up to 19,000 adult deaths a year, rural reliance on tube wells (Flanagan et al.); "the largest poisoning of a population in history" and "lung, bladder and skin cancers" (Smith et al., confirmed in the full text by the second reader); 94% removal in the final phase, 165 ± 17 to 10 ± 4 μg/L, online monitoring and daily backwash (Otter et al.); the Luminous license and date, Gadgil's group, and iron dissolved by electricity (Berkeley Lab); Bangladesh and West Bengal as the most affected region (Otter et al.). Not checked: the entry-point statements about ICP-MS, hydride-generation AAS and field kits, which are method background rather than claims from these sources. The long title's "cannot scale" and the Display title are the brief's framing, not a source statement, and were left unchanged. Genes untouched. A second reader went over every repair on the same day against independently fetched copies of all seven sources. The repairs listed above stand in the wording given here. Three were reworded to say no more than the source: the scaling sentence now reports what the sources contain rather than asserting what was not reached; the abandonment sentence now carries the paper's past tense and its "household and community scale" subject; and the sentence from Otter et al. no longer says "the same" chemistry, only that field solutions applying co-precipitation and adsorption have failed. One small change was undone because the original was right: "a field trial" went back to "field trials," the paper's own title.