Manufacturing a single smartphone consumes roughly 12,000 liters of water across its entire supply chain, from mining raw materials to assembling the final product. That is enough to fill about 150 standard bathtubs. The number comes from lifecycle analyses that trace every drop used to extract dozens of metals, purify silicon wafers, and run the factories where components are made. What makes that figure so large, and so hard to shrink, is that the water is spread across mines, refineries, and fabrication plants on multiple continents, each drawing from different watersheds with very different levels of water stress.
Where the Water Actually Goes
A smartphone contains more than 60 different elements, and nearly every one of them requires water to extract and process. The biggest share of the water footprint belongs to mining and refining the metals inside the device. After that comes semiconductor fabrication, which demands enormous volumes of ultrapure water. Smaller but still meaningful amounts go toward producing the battery, the display glass, the plastic casing, and the cardboard packaging. Energy production at every stage also consumes water indirectly, since power plants use water for cooling, and hydroelectric dams evaporate it from reservoirs.
What makes the accounting tricky is that “water use” can mean different things. Some water is physically consumed, meaning it evaporates or becomes embedded in a product and never returns to its source. Other water is withdrawn, used for cooling or rinsing, and then discharged back into a river or treatment plant, sometimes carrying pollutants. Lifecycle studies typically count both, but they weight the consumed water more heavily because it represents a permanent local loss. The roughly 12,000-liter estimate for a smartphone captures this combined picture.
Mining Precious Metals Is the Biggest Water Driver
If you had to point to the single most water-intensive ingredient in your phone, it would be the precious metals, particularly gold and palladium. A smartphone contains only tiny amounts of each, but extracting them from ore is spectacularly inefficient. Ore grades for gold are often just a few grams per tonne of rock, so massive volumes of material must be crushed, washed, and chemically treated to yield a sliver of metal. Water is consumed directly in the mining operations themselves and indirectly through the energy needed to run them. Precious metals also carry a heavy water-degradation burden because of metal emissions during the management of mine tailings, the slurry of waste rock and chemicals left behind after extraction.1Resources, Conservation and Recycling. Opportunities for reducing the supply chain water footprint of metals used in consumer electronics
Copper is another major contributor. Processing copper ore through conventional smelting consumes around 91 cubic meters of groundwater per tonne of copper, while an alternative chemical-leaching route uses about 70 cubic meters per tonne.1Resources, Conservation and Recycling. Opportunities for reducing the supply chain water footprint of metals used in consumer electronics A single smartphone does not contain a tonne of copper, of course, but when you multiply small per-device amounts across billions of phones produced each year, the aggregate draw on local water supplies is substantial.
Lithium, used in the rechargeable battery, adds its own layer of complexity. In South America’s salt flats, lithium is extracted by pumping underground brine into shallow ponds and letting the sun evaporate the water over months. This brine-based process dominates the water footprint of battery-grade lithium, with one study reporting 326 cubic meters of world-equivalent water per tonne of lithium product.2Journal of Cleaner Production. Water footprint of battery-grade lithium production in the Salar de Atacama, Chile The actual liters consumed per kilogram of lithium carbonate vary widely depending on how you count evaporated brine. Estimates range from about 0.2 to 7.7 cubic meters per kilogram for brine-based extraction, compared with a narrower 0.2 to 0.5 cubic meters per kilogram when lithium comes from hard-rock mining of spodumene.3Journal of Cleaner Production. Carbon and water footprint of battery-grade lithium from brine and spodumene: A simulation-based LCA That enormous range reflects a genuine scientific debate about whether evaporating ancient brine that would never have reached fresh-water systems counts the same as diverting a river.
Chip Fabrication and the Need for Ultrapure Water
After mining, semiconductor manufacturing is the next major water consumer. The chips inside your phone are built in layers thinner than a human hair, and any stray particle can ruin a circuit. That means fabrication plants, called fabs, rinse wafers repeatedly in ultrapure water, a grade of water so clean that it contains fewer than one part per billion of dissolved solids. Producing ultrapure water itself is an energy-intensive process: raw water must be filtered, deionized, and polished through multiple stages before it meets the standard. A study of fabs in Taiwan’s science parks found a tight link between water use and energy use specifically in the ultrapure water production step, because turning ordinary tap water into something that pure takes a lot of electricity to run membranes and ion-exchange systems.4Water-Energy Nexus. Revisiting the water-use efficiency performance for microelectronics manufacturing facilities: Using Taiwan’s Science Parks as a case study
A large modern fab can use millions of gallons of ultrapure water per day. Most of that water is used once for rinsing and then must be treated before it can be discharged or recycled. The spent water picks up traces of hydrofluoric acid, copper ions, and other chemicals used during etching and polishing steps. Semiconductor manufacturing relies on materials that are often toxic and sometimes proprietary, making wastewater treatment a persistent challenge for the industry.5iScience. Semiconductor manufacturing wastewater challenges and the potential solutions via printed electronics
Water Pollution from Mines to Fabs
The water footprint is not just about how many liters get used up. It is also about what happens to the water that comes back. At the mining end of the supply chain, the biggest concern is acid mine drainage. When rock containing sulfide minerals is broken open and exposed to air and water, chemical reactions produce highly acidic, metal-laden runoff.6Asia-Pacific Journal of Chemical Engineering. Acid rock drainage formation and treatment: a review This is not a theoretical risk. In the gold-mining district around Johannesburg, South Africa, groundwater is heavily contaminated and acidified from decades of tailings exposure. Polluted water discharges into streams, lowering their pH and carrying heavy metals more than 10 kilometers downstream from the source.7Environmental Pollution. Acid mine drainage arising from gold mining activity in Johannesburg, South Africa and environs That contamination is painfully slow to fade.
At the manufacturing end, semiconductor wastewater contains a different cocktail of hazards. Pollutants show up at low concentrations but carry high risks to ecosystems and public health. The major organic compounds in microelectronic wastewater include tetramethylammonium hydroxide and ammonium, while heavy metals are among the most environmentally toxic inorganic pollutants.8Applied Water Science. An insight into microelectronics industry wastewater treatment, current challenges, and future perspectives: a critical review Many of the chemicals used in chip fabrication have proprietary formulations, so researchers sometimes do not even know exactly what is in the waste stream.
At the end of a phone’s life, improper disposal adds yet another layer. E-waste recycling operations, particularly informal ones in low-income countries, are notorious for contaminating land and water with toxic elements. Phones dumped near rivers or in unlined landfills leach heavy metals into groundwater and surface water.9PubMed Central. Environmental Heavy Metal Contamination from Electronic Waste (E-Waste) Recycling Activities Worldwide: A Systematic Review from 2005 to 2017 This is not counted in the 12,000-liter production figure, but it is very much part of the phone’s total water impact.
Can Refurbishment or Recycling Shrink the Footprint?
One of the most effective ways to reduce the water burden of smartphones is simply to use them longer. A lifecycle analysis comparing new smartphone production with refurbishment found that the water footprint of refurbishing a phone ranged from roughly 1,200 to 2,900 liters per device, depending on the scenario, compared with about 12,000 liters for manufacturing a new one.10JYX (University of Jyväskylä). Environmental impacts of a local circular economy business model : a carbon and water footprint analysis for smartphone refurbishment Even the worst-case refurbishment scenario used less than a quarter of the water that a brand-new phone requires. That makes sense: refurbishment replaces a screen or a battery rather than extracting fresh gold, cobalt, and silicon from the earth.
Recycling metals out of old phones also helps, though the gains are more modest than you might expect. Increasing the recycled content of metals in consumer electronics to the theoretical maximum could cut the water scarcity footprint by about 20 percent over current baselines.1Resources, Conservation and Recycling. Opportunities for reducing the supply chain water footprint of metals used in consumer electronics The reason the reduction is not larger is that recycling infrastructure for many of the rarer elements in phones barely exists. Gold and copper are recovered at reasonable rates, but elements like indium, gallium, and tantalum are often lost during smelting because they are present in such small quantities that recovering them is not economically worthwhile.
Another strategy is sourcing metals from regions where water is more abundant. Moving supply chains toward mines in lower water-scarcity areas could reduce the water scarcity footprint by about 19 percent.1Resources, Conservation and Recycling. Opportunities for reducing the supply chain water footprint of metals used in consumer electronics A liter of water consumed in a desert region imposes a far greater ecological cost than a liter consumed in a rainy one, so where metals are mined matters almost as much as how much water the mining uses.
New Approaches to Factory Wastewater
Chip manufacturers are under growing pressure to clean up and reuse more of the water flowing through their fabs. The gold standard in the industry is “zero liquid discharge,” meaning no wastewater leaves the plant at all. That goal is technically achievable but expensive. One recent approach combines a membrane process called forward osmosis with multi-stage nanofiltration to treat the hydrofluoric acid waste streams typical in chip plants. In laboratory testing, the forward-osmosis step rejected more than 90 percent of copper ions from the wastewater, while simultaneously concentrating the copper enough that it could potentially be recovered as a valuable byproduct. The nanofiltration stage then produced fresh water clean enough to meet discharge standards.11Water Research. Toward zero liquid discharge treatment of semiconductor wastewaters with a hybrid system integrating forward osmosis and multi-stage nanofiltration Systems like this are still at the pilot or demonstration stage, but they represent a plausible path toward significantly reducing how much fresh water chip fabs consume.
Semiconductor Plants Are Being Built in Water-Stressed Regions
Here is where the water story gets uncomfortable for the semiconductor industry. Roughly 40 percent of existing chip-fabrication facilities worldwide sit in river basins classified as high or extremely high water-stress risk. Among facilities currently under construction, somewhere between 24 and 40 percent are in similarly risky basins. And for plants announced since early 2021, many of them spurred by subsidy programs like the U.S. CHIPS Act, that share climbs to 40 to 49 percent.12iScience. Climate change induced water stress and future semiconductor supply chain risk In other words, the industry is building new capacity in places where water is already scarce and projected to become scarcer under climate change. Projections for 2030 and 2040 across multiple climate scenarios suggest these risks will only intensify, and because chip supply chains are tightly interconnected, water stress at a handful of key facilities could cascade into shortages felt across global electronics markets.13iScience. Climate change induced water stress and future semiconductor supply chain risk
The decisions being made right now about where to build fabs are locking in water-stress exposure for the next 20 to 30 years, since a fabrication plant takes years to construct and operates for decades. It is a tension between strategic goals and environmental reality. Governments want domestic chip production for national security reasons, but the locations that make political sense do not always make hydrological sense.
Taiwan’s Impossible Water Allocation
Nowhere is this tension sharper than in Taiwan, which produces a dominant share of the world’s most advanced chips. Taiwan’s semiconductor sector and its agricultural sector are the two thirstiest parts of the island’s economy, and they are increasingly competing for the same water. During a severe drought in 2020 and 2021, the government diverted water from farming to keep chip fabs running, and Taiwan’s total cereal exports dropped sharply as a result.13iScience. Climate change induced water stress and future semiconductor supply chain risk
Semiconductors are an enormously lucrative export, but food security is not something you measure purely in GDP terms. The two sectors represent values that do not translate into the same currency: one is about economic competitiveness and geopolitical leverage, the other about feeding a population and maintaining sovereignty in a different sense entirely. Policymakers in Taiwan face the prospect of having to choose between them more frequently as climate change reduces reliable rainfall, and that choice will be inherently political rather than purely economic. For anyone wondering why the water footprint of a smartphone matters beyond abstract environmental accounting, Taiwan’s dilemma is the answer: the water your phone requires comes from the same finite supply that grows food, sustains ecosystems, and supports communities.
What the Packaging Adds
A smartphone’s packaging box is a tiny part of the total water story, but it is not zero. The water footprint of printing and writing paper falls somewhere between 300 and 2,600 cubic meters per tonne of paper, which works out to roughly 2 to 13 liters for a single A4-sized sheet.14Unesco-IHE Institute for Water Education. The green and blue water footprint of paper products: methodological considerations and quantification A smartphone box uses the equivalent of a few sheets of cardstock plus inserts, so the packaging water footprint is measured in tens of liters at most. Compared with the thousands of liters embedded in the device itself, packaging is a rounding error. Companies that trumpet their switch to thinner boxes or recycled cardboard are making a real but very small dent in the overall picture.
That does not mean packaging improvements are worthless. They matter for paper-industry water use in aggregate, and they signal that a company is thinking about its supply chain. But if you are trying to reduce the water cost of your personal phone habit, the single highest-leverage action is holding onto your current phone longer or buying a refurbished one, not worrying about whether the box it came in was recycled.