Arsenic filters do not destroy arsenic. They concentrate it into a sludge that Indian law treats as hazardous waste, and what becomes of that sludge in Bihar does not appear anywhere in the public record.
What to know about India Is Pulling Arsenic Out Of Its Groundwater.
Arsenic is a notoriously toxic element to humans, known to affect organs and systems including the cardiovascular system, immune system, liver, kidney, bladder, skin and even the nervous system. Yet it is incredibly useful, especially in industries like semiconductor and electronics manufacturing and in wood preservatives. The arsenic in Bihar’s groundwater, however, has nothing to do with any of these industries.
Every year, arsenic filters remove vast amounts of arsenic from India’s water sources. However, an arsenic removal unit does not break anything down. Whatever leaves the water is still arsenic, now stuck to a filter medium. Once that medium is exhausted or cleaned out for reuse, it leaves behind a sludge carrying five to seven kilograms of arsenic per cubic metre.
Whether Indian law treats that as hazardous waste turns out to be a more interesting question than it sounds. Schedule I of the Hazardous and Other Wastes (Management and Transboundary Movement) Rules, 2016 does list arsenic-bearing sludge, but only when it arises from smelting non-ferrous metals or from making nitrogenous fertilisers. Treating drinking water is not one of the listed processes.
That leaves Schedule II, where arsenic appears with a limit of 5 milligrams per litre. The figure is not measured on the sludge itself. It is measured in what leaches from a sample under a standard laboratory extraction test. Which test, and under what conditions, turns out to matter a great deal.
The problem underground, briefly
The arsenic in Bihar’s groundwater is not industrial pollution. It arrived from the Himalaya as river sediment over thousands of years and settled across the Gangetic plain, locked inside iron oxyhydroxide minerals, where it sits harmlessly.
It is released when the chemistry of the aquifer changes. In the young alluvial sediments of the Ganga plain, bacteria feed on buried organic matter and consume dissolved oxygen. Once the oxygen is gone, the iron minerals start to dissolve, releasing the arsenic they had been holding. That is why the contamination follows the newer alluvium along the Ganga, and why depth makes such a difference. A 2010 study of Patna and Bhojpur found most of the contamination above about 50 metres, and the Central Ground Water Authority puts it mainly within 100 metres.
This sequence, oxygen running out and iron dissolving, is worth keeping in mind, because it returns later in an unexpected place.
How much is left, and who says so
By the official count, the problem is close to solved. In November 2016, the government told the Lok Sabha that Bihar had 1,075 arsenic-affected habitations. By March 2021, that had fallen to 385, and by July 2023, the 460 habitations remaining nationwide were in Punjab, Uttar Pradesh and West Bengal, with Bihar no longer among them.
The research picture looks different. In December 2023 the Central Ground Water Authority told the National Green Tribunal, in a report on groundwater contamination, that groundwater arsenic above 10 micrograms per litre, the Indian and WHO limit, has been recorded in 27 districts of Bihar. A 2025 review by Vikas Kumar and Shashank Shekhar in H2Open Journal arrives at 22 districts and 87 blocks, with more than ten million people at risk. That review compiles two decades of published studies of varying scope and method rather than taking fresh measurements of its own, which is worth remembering when quoting the ten million figure.
The two sets of numbers do not contradict each other because they count different things. According to the JJM operational guidelines, a habitation is recorded as quality-affected only when a drinking water source fails the standard, and the remaining safe sources cannot supply 40 litres per person per day. Both conditions have to be met, and the habitation comes off the list once a safe supply is provided. What is being tracked, in other words, is the supply system rather than the arsenic in the ground, which is unaffected by any of it.
None of which makes the progress imaginary. Shifting households off contaminated shallow aquifers onto treated surface water is the most durable solution available, and Bihar has built multi-village surface water schemes along the Ganga for exactly that purpose. But wherever the answer is treatment rather than a change of source, arsenic is being taken out of water and put somewhere else.
What actually goes wrong with the units
The usual explanation for why arsenic filters underperform is that the arsenic arrives as arsenite, As(III), which carries no electrical charge at drinking water pH and therefore slips past media that work by attracting negative ions. This holds for activated alumina and anion-exchange resins, but not for iron-based systems. Freshly precipitated iron oxide binds arsenic by forming direct chemical bonds at its surface, and at around pH 7.5 it actually holds As(III) more strongly than the oxidised form As(V). In the West Bengal well-head units, As(III) was often removed more completely than As(V), with very little of it being oxidised on the way through.
So the chemistry was not the limiting factor. The failures that have been documented are much more ordinary. An assessment published in 2017 found that 475 of 570 arsenic removal plants in West Bengal were not useful and 145 were not working at all, with reagent dosing, replacement of exhausted media and outright abandonment being problems that came up repeatedly. By 2008, hundreds of imported single-use units had been abandoned once their capacity was used up. Bihar’s own early units, installed with UNICEF support, were abandoned within months for want of community ownership, according to the Patna environmental scientist Ashok Ghosh, speaking to Down To Earth in 2012.
Most of this evidence is old, and most of it comes from West Bengal rather than Bihar. The published literature does not settle how the more recent Bihar installations have performed, a gap worth flagging before going further.
Key context and latest developments
Where the arsenic goes
Now the sequence from underground returns.
Arsenic stays attached to iron oxide for as long as conditions around it remain oxygen-rich. Put the sludge into an oxygen-poor setting instead, whether a landfill, an unlined pit or waterlogged ground, and the same process that contaminated the aquifer begins again. The iron dissolves, and at the same time the As(V) it holds is reduced to As(III), which adheres poorly even to oxides that are themselves stable under those conditions. That second pathway matters, because it means the risk is not limited to iron-based media.
Arup SenGupta of Lehigh University pointed this out in 2006, noting that commercial adsorbents which pass the standard leaching test can still release arsenic after burial, because the test does not reproduce the redox conditions of an actual disposal site. The test he was describing is the one Indian law relies on.
The numbers show how wide the gap can be. Sludge from the West Bengal units carried roughly 32 milligrams of arsenic per gram of dry solids, and its leachate came in consistently below 0.2 milligrams per litre across the tested pH range, well inside the 5 milligram limit. Arsenic sludge can be legally unremarkable and still be precisely the material SenGupta was warning about.
His group designed around the problem. In the West Bengal well-head units, the spent regenerant is precipitated and then held on a coarse sand filter kept open to air on purpose, so the iron remains as Fe(III) and the arsenic stays bound to it. Other researchers have proposed locking the sludge into cement or bricks. Both approaches work, but both depend on somebody having asked the question in the first place.
In Bihar, the question does not seem to have been asked in public at all. A search of what is available, including the state’s filings to the tribunal, the Public Health Engineering Department’s published material and the Jal Jeevan Mission guidelines, turns up nothing about where spent arsenic media and sludge end up, who handles them, or whether any of it is recorded.
None of that shows arsenic is leaching back into the groundwater. It does mean nobody is currently in a position to say that it is not.
Why the small numbers still matter
There is a temptation to treat concentrations between 10 and 50 micrograms per litre as a technicality, since India’s older limit sat at the top of that range. The evidence is more cautious than either side of that argument tends to be. The World Health Organization’s fact sheet on arsenic records raised risks of lung and bladder cancer, and of arsenic-associated skin lesions, below 50 micrograms per litre, and the International Agency for Research on Cancer classifies inorganic arsenic as carcinogenic to humans. The same document also records considerable uncertainty about the dose-response relationship at low intakes, and describes the 10 microgram guideline as provisional, retained on the basis of what treatment can achieve and what laboratories can measure rather than as a demonstrated safe level.
Within Bihar, a study of 2,000 cancer patients from Gangetic districts found substantially higher blood arsenic than in cancer-free controls drawn from unaffected parts of Patna. Blood is a weak indicator of long-term exposure, since arsenic clears from it within hours, and the study does not demonstrate that arsenic caused the cancers. It points somewhere without settling anything.
What would close the gap?
Two changes would make the situation legible, and neither needs new science.
Testing could be carried out where people actually drink, and repeatedly, instead of only at the source. What a household is exposed to depends on which source it uses and whether the treatment on that source is still working, and a habitation removed from a list is not the same thing as a household with safe water in its pot.
Disposal of spent media and sludge could be specified, recorded and audited, under conditions that keep the arsenic oxidised. That does not require rewriting the hazardous waste rules, though it does mean recognising that a leaching test conducted in air tells you little about what a sludge will do in an anoxic pit. Where the waste is stored matters more here than what the test result says.
Arsenic reached Bihar’s groundwater because oxygen ran out underground and iron minerals dissolved. It would be an unfortunate outcome if treating it ended up recreating those same conditions in a pit behind the handpump, with nobody watching closely enough to notice.
This article was written by Dr Sameer Saurav, Assistant Professor of Chemistry at A.N.S. College, Nabinagar, Aurangabad, Bihar. He holds a PhD in chemistry from IIT Kanpur and writes on science in English and Hindi.
