Does boiling water remove microplastics? What works outside the lab

Partly. In a 2024 lab study, the limescale that forms when hard water boils caught test plastic particles and carried them to the bottom, and the harder the water, the bigger the effect. The plastic stays in the pot, so boiling helps only if you leave that scale behind. Nobody in the study used a home kettle. Yu et al., 2024

What did the boiling study actually measure?

It measured how many test particles went missing from the water above the settled scale in small boiled samples. Researchers in Guangzhou, China, added fluorescent polystyrene beads, mostly 1 micrometer across, to 40-milliliter samples of local tap water and of lab water with its hardness set between 0 and 300 mg/L. They held each sample at its target temperature for five minutes, cooled it for ten, then counted the beads left in 1 milliliter drawn from 1 centimeter below the surface. Each test ran three times. Yu et al., methods

In the Guangzhou tap water, which had 180 mg/L of hardness, 28% of the beads left the sampled water after heating to 90 °C, and 84% at a full boil. Yu et al., Supporting Information Table S5 Larger particles, tested in that tap water only, did similarly at the boil. Of polyethylene fragments about 150 micrometers long and polypropylene fibers about 100 micrometers long, both made from sun-aged plastic, 81% and 90% left the sampled water; of 100-micrometer polystyrene beads, 95%. Yu et al., Supporting Information Text S3

All of these percentages describe particles missing from the sampled water. None counts plastic removed from the container.

Why does hard water matter?

Scale did most of the catching, and scale forms from the minerals that make water hard, mostly dissolved calcium and magnesium. USGS In the tap water, calcium carbonate began growing on the beads at 50 °C and coated them by 70 °C, and most beads then sank with the scale. Yu et al.

In lab water, the share of beads lost from the sampled water climbed with hardness, with the biggest jump between 100 and 120 mg/L. Read the low end carefully. With no hardness, no scale could form, yet 25% of the beads still left the sampled water, which the authors link to heat making particles collide, clump and settle. That is the no-scale baseline, and soft water at 60 mg/L did only a little better, at 33%. Yu et al., Table S5 and results

The US Geological Survey calls water soft at 0 to 60 mg/L, moderately hard at 61 to 120, hard at 121 to 180 and very hard above that. USGS The jump came at the top of the moderately hard range. Treat your number as a guide, though: the authors describe scale forming in water with enough hardness and alkalinity, and how fast white scale builds up in your kettle is another rough clue. Yu et al.

Where does the plastic go?

It stays in the container, in or on the scale. Imaging showed the missing beads in the scale that settled in the lab beakers, with the most at 100 °C, when there was the most scale. Yu et al. Scale in home kettles wasn’t studied.

So the second step matters as much as the first: let the water cool and settle, pour gently, and leave the last bit behind. A 2026 field study from Bangladesh recommends the same gentle pouring. Nupur et al., abstract The Guangzhou authors also suggest pouring through a stainless steel filter, like a tea strainer, to hold back the scale; they didn’t test one. Yu et al. It’s also why boiling differs from distilling, which collects the condensed steam in a separate container. Our filtration guide

The white flakes you can see are scale. Most of the test beads were 1 micrometer across, far too small to see.

Does boiling work outside the lab?

A little field evidence says it helps in very hard water. In Khulna, Bangladesh, researchers found microplastics in all 45 samples of piped tap water, pond water and groundwater. Boiling cut the microplastics they counted, most smaller than 400 micrometers, by more than 70% in all three kinds of water, and the drop tracked hardness. The authors attribute the drop to particles settling in very hard water rather than breaking down. We could read only the abstract. Nupur et al., 2026

Here is how the lab routine compares with a kitchen:

The authors’ estimate that boiled water cuts intake two to five times applies the bead results to every particle in tap water worldwide. Yu et al., Supporting Information Table S3 Neither study measured health; our page on what microplastics do to you covers that evidence. The journal also published a comment on the study and the authors’ reply, which we couldn’t read in full.

Does your kettle add plastic?

A new plastic kettle can, at least at first. A 2025 study boiled ultrapure water 150 times in new polypropylene kettles from two brands, bought in Brisbane, Australia. Release was highest in the first boils and fell by more than half by the tenth, but plastic was still detected after 150 boils. Shi et al., 2025

The kettles were rinsed three times before that first boil. The authors suggest boiling and discarding a few kettlefuls before first use and say rinsing doesn’t achieve the same, an inference they didn’t test directly. Hard tap water (125 mg/L) left less nanoplastic in the water than ultrapure water did, and the authors suggest scale may coat the wall or trap released particles. Shi et al.

The Guangzhou authors recommend non-plastic electric kettles or gas stoves for boiling water because of their low energy use and carbon emissions. They didn’t test kettles for plastic. Yu et al.

When you choose a kettle, check every part the water or steam touches: the body, the underside of the lid, any spout filter, seals, a level window and, on variable-temperature models, the sensor. Materials lists can surprise you: IKEA’s listing for its stainless VATTENTÄT stovetop kettle names a melamine whistle.

What we don’t know yet

• Whether boiling, settling and careful pouring work in home kettles with ordinary tap water. We found no published test.
• How much careful pouring or a strainer holds back, especially of the smallest particles.
• How real mixtures of particles, soft water, and filtered, softened or bottled water behave. The lab didn’t test them, and the field study covered very hard water in one city.
• Whether any of this changes health.

What we’d do

Nothing on this list needs a purchase, which is our favorite kind of list.

1. Keep your routine if it works for you. Nothing in these studies says you have to start boiling your drinking water.
2. Find your hardness. Ask your water utility, as the USGS suggests, check its annual water quality report, or use a home test strip. USGS
3. If your water is hard and you already boil water for tea, coffee or cooking, let it sit (the lab waited ten minutes), pour gently and leave the last bit, scale included, in the kettle. It costs nothing.
4. If your water is soft, expect little. Don’t add calcium or baking soda to copy the lab. The researchers set hardness that way to study the effect and never tested it as a household method. Yu et al.
5. Start with cold tap water, and let it cool before you drink it. The EPA advises only cold water for drinking, cooking and baby formula, and notes that boiling does not remove lead. EPA The World Health Organization’s cancer agency classes very hot drinks, above 65 °C, as probably carcinogenic, on limited evidence. IARC
6. If you use a filter, get one certified for lead. The EPA’s advice is conditional: if you use a filter, make sure it’s certified to remove lead. Some filters carry a certified microplastics-reduction claim, and our filtration guide shows how to check the exact model.
7. Keep your kettle unless you’re replacing it anyway. If yours is new and plastic, boil and discard a few kettlefuls first, as the kettle study’s authors suggest. Baby bottles are a separate question, covered in our baby-bottle article.

Our pick

If you’re replacing a kettle or boil water daily, start here. We haven’t used these ourselves; the details come from the makers’ own pages and manuals.

Affiliate links may earn us a commission at no extra cost to you. How this works.

Secura SWK-1701DB double-wall electric kettle

• Why it’s our pick: Secura says no plastic touches the hot water. It simply boils and switches off, with no temperature settings, holds 1.7 liters, and has boil-dry protection and a cool-touch outer wall.
• What touches the water: Secura lists a 100% stainless steel interior, including the pot, lid, spout and rim, with the fill marks inside the pot. Secura, Amazon listing, manual
• Evidence: manufacturer-declared. We found no certification or independent test of its materials.
• Cons: the outer shell is plastic, though Secura says it doesn’t touch the hot water. No lid-seal material is listed, and it needs at least 0.5 liters per boil.
• Price: about $37 to $42 on Amazon, depending on color (checked Sept 2026).
• Skip it if: you’d rather have a steel body. Chantal says its 32-ounce Mia Ekettle is made of 304 stainless steel with an all-stainless interior, and its spout is designed for a controlled slow pour, which helps when you’re leaving scale behind. It costs about $84 (checked Sept 2026).

A plain stainless stovetop kettle

• Why it fits: it’s the cheaper route if you cook on a stove, with no electronics to wear out.
• What touches the water: the body, lid and spout, plus the whistle that steam passes through. Look for a materials list that covers all four, because some whistles are plastic.
• Evidence: this describes what to look for; we haven’t verified a specific model.
• Cons: no auto shut-off, so it can boil dry if forgotten, and it ties up a burner.
• Price: about $20 to $35 (checked Sept 2026).
• Skip it if: you tend to wander off while water heats.

Hach Total Hardness Test Strips, 0 to 425 mg/L

• Why it fits: it shows whether boiling is likely to do much for your water; the lab results jumped between 100 and 120 mg/L.
• What it measures: total hardness from 0 to 425 mg/L, according to Hach, with 50 strips per bottle (product 2745250). It touches only the sample you test.
• Evidence: manufacturer-declared range.
• Cons: it measures hardness only, and your utility may publish that number for free.
• Price: about $22 at lab-supply retailers (checked Sept 2026).
• Skip it if: your water report already lists hardness.

RSVP Endurance conical mesh strainer, 4 or 4.75 inch

• Why it fits: the boiling study’s authors suggest pouring through a stainless steel filter, like a tea strainer, to hold back scale.
• What touches the water: RSVP says the strainer, including its rim and handle, is 18-8 stainless steel. RSVP 4-inch, 4.75-inch
• Evidence: manufacturer-declared. RSVP calls the mesh extra fine but gives no mesh size.
• Cons: it should catch visible flakes but hasn’t been tested on particles bound up in scale. It suits a mug or small pitcher.
• Price: about $9 to $15 (checked Sept 2026).
• Skip it if: you already own an all-steel tea strainer, which does the same job.

Resources to go further

• The boiling study (Yu et al., Environmental Science & Technology Letters, 2024). Paywalled, but the free Supporting Information holds the tables, including Table S5 behind our chart.
• The Khulna field study (Nupur et al., Science of the Total Environment, 2026). Real tap, pond and groundwater in one very-hard-water city; the abstract is free on the publisher’s site.
• The plastic-kettle study (Shi et al., npj Emerging Contaminants, 2025). Open access; Figure 1 shows release falling over 150 boils.
• USGS: hardness of water. The hardness scale in plain language, with US maps.
• EPA: lead in drinking water. What boiling can’t fix, and what to do instead.
• NSF certified drinking water treatment listings. Search an exact filter and cartridge before you trust a claim.
• Susan Freinkel, Plastic: A Toxic Love Story (Houghton Mifflin Harcourt, 2011). A readable history of how plastic moved into everyday objects; background that predates this research.