Are Reusable Water Bottles Safe? What the Evidence Actually Says

Key takeaways

• Stainless steel and glass carry the lowest chemical risk; a named food-grade plastic comes next.
• Two things actually go wrong: heat-driven migration, and a bottle nobody cleans properly.
• The number in the triangle is a material ID, not a safety rating (ASTM D7611).
• The FDA and EFSA disagree about BPA right now, and both positions are current.
• Washing works: measured bacterial load fell by roughly three-quarters in a 2024 study.

If you have searched this question, you have probably run into the same three claims everyone else has: that plastic bottles leach BPA into your water, that reusable bottles shed microplastics with every sip, and that the bottle on your desk is filthier than a toilet seat. Some of that is grounded in real, peer-reviewed research. Some of it traces back to a marketing survey that got repeated so often it started to sound like science.

The honest answer is not a clean yes or no, and any page handing you one is skipping the part that matters. Whether reusable water bottles are safe depends on what yours is made of, how hot it gets, and how often it actually gets washed. It also depends on which question you are really asking, because refilling a thin single-use bottle for a month is a genuinely different question from drinking out of a steel bottle built for the job.

So this guide works through it in order: what actually migrates out of a bottle and under what conditions, why the FDA and European regulators currently disagree about BPA, what the evidence does and does not show on microplastics and bacteria, and how to judge whether the bottle already sitting on your desk is fine to keep using.

What makes a reusable water bottle safe or unsafe?

Four things can go wrong inside a water bottle: chemicals migrating out of the material, plastic shedding as microscopic particles, metals moving out of a steel or coated surface, and bacteria growing on a surface that never dries. Three variables govern all four: the material, the temperature it reaches, and how often it is properly cleaned.

Heat and physical damage sit behind most of the chemical worries. Washington State’s Department of Health puts it plainly: chemicals in some plastics “may be toxic when released by heat or when the product is scratched.” Microbes work differently. Clemson Cooperative Extension calls biofilm “that slimy feeling on the surface of a material when water or other liquids have been allowed to sit for a while,” and UNH Extension notes that biofilms “stick to the inside of the bottle, shielding that bacteria from cleaning supplies.” Rinsing is not washing. Everything below is one of those four paths applied to the materials reusable water bottles are made from.

What actually leaches from a water bottle, and when?

Migration is not an on-or-off property of a material. It is a rate, and the rate answers to temperature, acidity, dwell time and physical damage.

The clearest numbers come from PET. Fan and colleagues, writing in Environmental Pollution in 2014, stored PET drinking-water bottles at three temperatures. After one week, antimony release measured 1.88–8.32 ng/L at 4 °C, 2.10–18.4 ng/L at 25 °C, and 20.3–2604 ng/L at 70 °C, with BPA following the same pattern at far lower magnitudes. Under the worst condition tested, 70 °C for four weeks, BPA stayed within regulatory limits while one brand exceeded EPA antimony exposure standards. Temperature is doing nearly all of the work there.

Washing has its own chemistry. Tisler and Christensen at the University of Copenhagen, in the Journal of Hazardous Materials in 2022, screened water from reusable plastic bottles after a dishwasher cycle and found more than 400 substances from the bottle plastic and more than 3,500 from the soap. The toxicity of at least 70% of what they identified is unknown, which is a statement about missing data rather than about harm.

The practical guidance follows from the mechanism. Washington State DOH advises against microwaving food in polycarbonate containers or pouring hot liquids into them, and recommends replacing old or scratched plastic containers, water bottles included. Clemson Extension adds avoiding hot temperatures and prolonged direct sunlight.

Are reusable plastic water bottles safe?

Two versions of this question get mixed together constantly, and separating them dissolves most of the confusion. Are plastic reusable water bottles safe when the bottle is purpose-built, made from a named resin such as Tritan, HDPE or polypropylene? Yes, under the conditions above. Are reusable plastic water bottles bad for you when what you are reusing is a thin single-use PET bottle refilled for weeks? That is a different question with a different answer. What remains after that split is two chemical questions, BPA and whatever replaced it, plus one labeling question about the number on the bottom.

Is BPA in a reusable bottle a real risk?

The United States and the European Union currently give different answers, and both answers are live.

Neither agency is being careless. They weighed different endpoints and applied different uncertainty factors to overlapping literature, which is the gap the NIEHS and National Toxicology Program built the CLARITY-BPA consortium to narrow; its 2021 report did not end the argument. One piece of US history also gets reported as a safety verdict and is not one. The FDA dropped BPA-based polycarbonate from baby bottles and sippy cups in July 2012, and BPA-based epoxy coatings from infant formula packaging in July 2013, after petitions showed industry had already abandoned those uses. In the FDA’s own words, an amendment based on abandonment “is not based on safety.”

Does “BPA-free” mean safe?

“BPA-free” describes one chemical that is absent. It says nothing about what replaced it. Bittner, Yang and Stoner tested 50 BPA-free products and reported in Environmental Health in 2014 that “many BPA-free PC-replacement products still leached chemicals having significant levels of EA,” meaning estrogenic activity, and that “BPA-free did not mean EA-free.” Some products made from glycol-modified PET or cyclic olefin resins released nothing with detectable estrogenic activity.

Two things belong beside that finding. The authors disclosed competing interests: stock in CertiChem and PlastiPure, with Bittner as consulting CEO of CertiChem. And in July 2013 a federal jury in Austin found that PlastiPure and CertiChem had made false and misleading claims about Eastman’s Tritan under the Lanham Act, a verdict affirmed on appeal in 2014, with trial evidence including test results from laboratories that found no estrogenic activity in Tritan. A jury verdict is not a scientific finding either. The estrogenic-activity question is disputed rather than settled. Reviews of the substitutes report endocrine-disrupting activity for both BPS and BPF, and a 2020 review of BPS in food concluded that its hormonal and obesogenic effects are “comparable to or worse than” those of BPA, while other comparisons rank BPS below BPA on some endpoints. The EU’s 2024 regulation banned BPS alongside BPA.

Which plastic resin numbers are safe to reuse?

Turn a bottle over and there is usually a small triangle with a number from 1 to 7 inside it. Most safety advice online treats that number as a grade. It is worth knowing what the number really is before leaning on it, because the practical advice attached to it holds up while the reasoning usually given for it does not. The number also settles the single-use question: a #1 PET bottle from a vending machine was engineered for one fill, with thin walls that scratch easily, and refilling it for weeks is not a use it was built for.

The number inside the triangle is a material ID, not a safety rating. ASTM D7611, the standard that defines those codes, states that Resin Identification Codes are “used solely to identify the plastic resin used in a manufactured article” and “are not ‘recycle codes.’” ASTM also replaced the chasing-arrows symbol with a solid triangle specifically so that people would stop reading extra meaning into it.

Code	Plastic	Where you see it	Reuse for drinking water?	Why
#1	PET / PETE	Single-use water and soda bottles	Not what it was built for	Thin, easily scratched walls; scratches give biofilm somewhere to sit. Antimony release from PET bottles rose to 20.3–2604 ng/L after one week at 70 °C, against 1.88–8.32 ng/L at 4 °C (Fan et al., Environmental Pollution, 2014). Washington State DOH advises replacing old or scratched plastic containers, including water bottles.
#2	HDPE	Milk jugs, rigid reusable bottles	Generally fine	Not on Washington State DOH’s avoid list, which names only #3, #6 and #7.
#3	PVC	Flexible tubing, soft packaging, some films	Avoid for food and drink	Washington State DOH advises avoiding cooking or storing food and beverages in plastics marked #3. The same guidance warns that chemicals in some plastics “may be toxic when released by heat or when the product is scratched.”
#4	LDPE	Squeeze bottles, film, bag liners	Generally fine	Also outside Washington State DOH’s avoid list of #3, #6 and #7.
#5	PP	Lids, straws, some bottles and tumblers	Generally fine	Also outside Washington State DOH’s avoid list of #3, #6 and #7.
#6	PS	Disposable cups, foam packaging	Avoid for food and drink	Washington State DOH advises avoiding cooking or storing food and beverages in plastics marked #6.
#7	“Other” — a catch-all for mixed and unlisted resins, including polycarbonate and Tritan copolyester	Older hard-plastic bottles; also many current reusable bottles	Depends entirely on which resin it is	Washington State DOH advises avoiding #7 for food and drink. Polycarbonate made with BPA is marked #7 with the letters “PC” — but University of Nebraska–Lincoln Extension notes that “not all plastic bottles with a #7 designation contain BPA.” Tritan copolyester is also sold under #7, and its estrogenic-activity question is disputed (see above).

Both of those things are true at once, and the tension between them is worth stating plainly. ASTM never intended the resin code as a safety rating, so a page that presents 1 to 7 as a league table of safety is using the number for something it was never designed to do. Yet a state health department does independently advise against cooking or storing food and drink in #3, #6 and #7 — advice that rests on what those resins are, not on what the triangle means. The practical guidance stands; the usual framing of it is wrong.

Which water bottle material is safest?

On chemical risk alone, stainless steel and glass lead, because both stay close to inert in plain water in a way no plastic or coating quite matches. That ranks one thing, not everything. Glass breaks, steel dents, and the safest reusable water bottles are the ones their owners will actually clean.

Steel bottles are usually 18/8, roughly 18% chromium and 8% nickel, in the 304 family, and the chromium forms the passive oxide layer that keeps the surface from reacting. The 18/8 versus 18/10 distinction that gets marketed is a difference in corrosion margin and polish, not a safety cliff. The detail worth understanding is the pellet used to seal the vacuum in some insulated bases. Stanley stated in January 2024 that its sealing material “includes some lead,” that the sealed area is “covered with a durable stainless steel layer, making it inaccessible to consumers,” and that no lead touches the drinker or the contents. PolitiFact checked that on 1 February 2024 and found the lead confined to the sealed base, with none detected on the exterior or the lid. Owala says its bottles “are and always have been lead free.” EWG and other advocacy groups have pressed the issue with XRF testing, which is advocacy work rather than peer-reviewed research. Brands differ, so read each brand’s statement as that brand’s claim.

Glass is, in Clemson Extension’s phrasing, a good option for anyone who wants to “minimize chemical exposure,” and UNH Extension calls it the material “least likely to scratch” while also “more prone to shattering.” Borosilicate tolerates roughly a 150 °C temperature differential against about 40 °C for soda-lime, per lab-glass suppliers rather than a primary standard, and that is a breakage distinction rather than a chemical one. The part of a glass bottle worth asking about is the decoration. Turner, in Science of the Total Environment in 2018, ran 197 XRF analyses on 72 decorated drinking-glassware products and found lead in 139 and cadmium in 134, with migration exceeding the 0.5 mg/L lead limit in all but one case. That study covered glassware rather than bottles, though painted or printed glass bottles belong to the same product class.

Material	Chemical migration risk	How it fails in practice	Cleanability	What to check before buying
Stainless steel 18/8 (304 family)	Low for plain water. The documented nickel and chromium migration figures (Kamerud et al., J. Agric. Food Chem., 2013) come from hours of simmering acidic tomato sauce, not water sitting in a bottle.	Dents. The base cap coming off a vacuum-insulated bottle — which is the only route by which the sealed pellet some brands use to close the vacuum could ever be reached (Stanley, January 2024).	Good. A wide mouth helps; Clemson Extension says double-wall insulated bottles should skip the dishwasher, to protect the insulation rather than for any chemical reason.	Whether the brand states what its base seal is made of. Treat that as a brand claim unless a public test result backs it — brands differ, and at least one (Owala) states its bottles have always been lead free.
Borosilicate glass	Very low. Clemson Extension calls glass a good option for anyone who wants to “minimize chemical exposure.”	Shattering. Thermal shock is the other failure: borosilicate tolerates roughly a 150 °C differential, per lab-glass suppliers rather than a primary standard.	Best. UNH Extension calls glass the material “least likely to scratch,” though also “more prone to shattering.”	Whether it ships with a protective sleeve, and whether the lid is steel or plastic — the lid is usually the only non-glass part touching your water.
Soda-lime glass, painted or printed	Very low from the glass itself. The decoration is the part worth asking about.	Cracks at roughly a 40 °C differential (supplier data). Exterior paint or print can flake with sustained use.	Best, same as borosilicate.	Whether the print is external and intact. Turner (Science of the Total Environment, 2018) found lead in 139 of 197 XRF analyses of decorated drinking glassware and cadmium in 134 — that study covered glassware, not bottles specifically, but painted bottles are the same product class.
Tritan copolyester (#7)	Disputed. See the BPA-free section above: a published study and a Lanham Act jury verdict point in opposite directions, and neither settles it.	Scratching, clouding and retained odor. UNH Extension calls plastic “the softest material, making it most susceptible to scratching,” and scratches are where biofilm gets a foothold.	Good.	That the bottle is named as Tritan rather than sold as an unlabeled #7 — an unlabeled #7 tells you nothing about which resin is in your hands.
HDPE (#2) / PP (#5)	Low. Neither is on Washington State DOH’s avoid list, which names only #3, #6 and #7.	Among the softest common bottle plastics, so they scratch and hold odor (UNH Extension).	Good.	A wide mouth and a lid without crevices — MSU Extension specifically advises against designs with crevices, because those are the parts that never really get cleaned.
Aluminum, lined	Depends on the liner, not the metal. Clemson Extension notes that because aluminum can react with some liquids, these bottles are “typically lined with another material,” so the coating is what your water actually touches.	A chipped or worn interior liner. The defining case: SIGG disclosed in August 2009 that bottles made before August 2008 used epoxy liners containing BPA.	Poor. Clemson Extension says aluminum bottles should avoid both the dishwasher and the freezer.	What the liner is made of — this is a question for the manufacturer, and a bottle whose liner is chipped is finished regardless of the answer.
Silicone (collapsible) / ceramic-lined	Regulated rather than tested per item: US food-contact silicone falls under 21 CFR 177.2600, and lead leaching from ceramicware under FDA’s CPG Sec. 545.450 action levels.	Silicone tears at folds and seams and holds odor. Ceramic and enamel linings chip.	Fair.	Ask for a third-party heavy-metal test result. “Food-grade” is a compliance claim about the formulation, not a test result for the item in your hand.

Two framings are worth carrying away from that table. With a lined aluminum bottle you are drinking from the coating and not the metal, so a chipped liner is the failure that matters. And for silicone or a ceramic lining, 21 CFR 177.2600 and the FDA’s CPG Sec. 545.450 action levels regulate a formulation rather than the item in your hand, so ask the seller for a third-party heavy-metal report instead of trusting the phrase “food-grade.”

Do reusable water bottles shed microplastics?

A steel or glass bottle filled from the tap does not shed plastic into your water. A plastic one can, and the size of that number is genuinely unsettled.

The figure everyone has seen comes from a 2024 PNAS paper by Qian and colleagues, who used stimulated Raman scattering microscopy to count roughly 240,000 detectable plastic particles per liter in bottled water, about 90% of them nanoplastics and 10 to 100 times higher than earlier estimates. The same journal has since published a Letter arguing that measurements at that scale need proper controls: “Nanoplastics measurements must have appropriate blanks.” Both belong in the same breath, because the method is new and the number is not yet settled.

For scale, Cox and colleagues estimated in Environmental Science & Technology in 2019 that people who meet their water intake from bottled water alone may take in around 90,000 extra microplastic particles a year, against roughly 4,000 for tap-water drinkers. Regulators read the same literature calmly. The FDA’s page on microplastics in foods, current as of 24 July 2024, says that “current scientific evidence does not demonstrate that levels of microplastics or nanoplastics detected in foods pose a risk to human health,” and the WHO reached a comparable conclusion about drinking water in its 2019 report while calling for urgent further research.

Can a dirty reusable water bottle make you sick?

Yes, in the ordinary way that any wet surface you put in your mouth can. The measured magnitude is a lot smaller and duller than the headlines suggest.

The most useful measurement comes from an April 2024 study in the Journal of Pharmacy and Bioallied Sciences, which sampled 30 daily-use bottles, 15 PET and 15 stainless steel. Mean initial microbial load was 68.8 ± 19.1 CFU/mL for the PET bottles and 35.4 ± 8 CFU/mL for the stainless ones. After a washing intervention, the mean fell to 11.2 ± 2.3 CFU/mL. Tens of colony-forming units per milliliter, and routine washing removed roughly three-quarters of them.

What grows is biofilm, and UNH Extension explains why it matters: biofilms “stick to the inside of the bottle, shielding that bacteria from cleaning supplies.” Mold and mildew appear first at the gasket, the straw and the lid threads, the parts a quick rinse never reaches; MSU Extension advises choosing a bottle without small crevices and hard-to-clean areas for that reason. Those are also the parts most likely to bother someone with asthma or a mold allergy, which is a reason to keep small parts clean rather than a diagnosis of anything.

How do you keep a reusable water bottle safe to drink from?

Almost all of the microbial risk is under your control, and the guidance is steadier than most internet advice. One note on where it comes from: there is no CDC guidance for adult reusable water bottles, because the CDC’s cleaning instructions cover infant feeding items. The daily-washing rule so often attributed to the CDC comes from US university extension services.

1. Every day: wash with warm water, dish soap and a bottle brush, taking the lid, gasket and straw apart and scrubbing them separately (Clemson Cooperative Extension, March 2021). MSU Extension puts it as hot water with a teaspoon of unscented dish soap, and adds the rule that does the most work: “Empty the bottle as soon as you are finished with it.”
2. A few times a week: MSU Extension suggests one tablespoon of bleach per quart of water, noting that more leaves a taste. UNH Extension’s alternative is half a teaspoon of baking soda with half a cup of vinegar, and a separate brush for straws.
3. Every two weeks to monthly: Clemson’s sanitizing soak, below.

Two hardware notes. Clemson advises keeping double-wall insulated steel and aluminum bottles out of the dishwasher and the freezer, which protects the insulation rather than addressing any chemical concern. And MSU Extension’s advice to avoid designs with crevices is the most underrated line in all of this: a bottle you cannot get a brush inside never really gets cleaned.

When should you replace a reusable water bottle?

No source I could find supports a “replace every few months” rule, and Clemson’s position points the other way: a well-cared-for bottle “can last many years.” Replace on triggers instead.

• Scratched or visibly aged plastic. Washington State DOH’s wording is “Replace old or scratched plastic containers, including water bottles.”
• A chipped, peeling or worn interior liner on an aluminum or coated bottle, since the liner is the food-contact surface.
• Chipped or scratched exterior decoration on painted or printed glass.
• A base cap that has come off or works loose on a vacuum-insulated bottle.
• An odor or slime that survives a proper sanitizing soak, which usually means established biofilm.

Lids, gaskets and straws wear out long before bottle bodies do, and replacing those parts is often the whole fix.

What’s settled, and what’s still unresolved?

Settled well enough to act on:

• Heat and physical damage increase chemical migration, and temperature matters more than refill count.
• Stainless steel and glass do not shed plastic into water.
• Routine cleaning measurably reduces microbial load.
• Single-use PET bottles were not designed for repeated washing and long-term reuse.

Still argued over, and worth knowing that it is argued over:

• On BPA, the FDA says current food-contact levels are safe, while EFSA concluded in April 2023 that typical exposure exceeds a tolerable intake about 20,000 times lower than its old one, and the EU ban took effect in January 2025.
• On Tritan, a 2014 study by authors with declared financial interests in a competing testing business sits against a 2013 jury verdict and laboratories that found no estrogenic activity.
• On nanoplastics in bottled water, the PNAS figure of about 240,000 particles per liter sits against a published Letter in the same journal arguing the measurement needs proper blanks.
• On glass versus plastic, ANSES’s 2025 cap-paint finding sits against the Glass Packaging Institute’s rebuttal.
• On lead in insulated bases, manufacturer statements and a PolitiFact check calling it sealed and inaccessible sit against advocacy XRF testing and a CPSC complaint filing.
• On how much bacteria, peer-reviewed tens of CFU per milliliter sit against a viral 20.8-million-CFU swab survey.
• On resin codes as a safety scheme, most safety advice online uses them that way, but ASTM D7611 says the codes identify resin and nothing else. State health departments still advise avoiding #3, #6 and #7, so the advice holds even where the framing does not.

How to choose a safe reusable water bottle

There is no separate product category of safe reusable water bottles. The safest reusable water bottles are ordinary ones chosen with a few things checked, roughly in this order.

• Pick the material first: steel, glass, or a plastic that names its resin, rather than an unlabeled #7 that tells you nothing about what you are holding.
• On aluminum and any coated or ceramic-lined bottle, the liner is the food-contact surface. Ask what it is made of, and retire the bottle when it chips.
• Choose a wide mouth and as few crevices as possible. A brush that cannot reach the base, the threads and the straw is the real hygiene problem.
• Read “non-toxic,” “lead-free” and “BPA-free” as brand claims. They may be true. A public third-party heavy-metal test result is what turns a claim into evidence.
• Check that spare lids, gaskets and straws are sold. Those parts fail first, and a bottle you can re-seal is a bottle you keep.

That is what healthy reusable water bottles come down to. Style is a separate question, and if that is what you are shopping for, our roundup of stylish reusable water bottles looks at design and build quality rather than safety.

The bottom line: are reusable water bottles safe?

Frequently asked questions

Is it safe to drink from a reusable water bottle left in a hot car?

Heat is the variable that matters most. Fan and colleagues (Environmental Pollution, 2014) measured antimony release from PET bottles rising from 1.88–8.32 ng/L at 4 °C to 20.3–2604 ng/L at 70 °C after one week. Clemson Extension advises avoiding hot temperatures and prolonged direct sunlight. A steel or glass bottle in a hot car is a temperature problem, not a chemical one.

Can you put a reusable water bottle in the dishwasher?

Usually yes, if the manufacturer says the bottle is dishwasher-safe. Clemson Extension advises keeping double-wall insulated steel and aluminum bottles out of it, to protect the insulation. Tisler and Christensen (Journal of Hazardous Materials, 2022) also detected more than 3,500 substances from dishwasher soap in water from reusable plastic bottles after a cycle, so rinse well afterward.

Is it safe to microwave a reusable water bottle or pour boiling water into it?

Washington State DOH advises against both for polycarbonate: do not microwave food or drink in polycarbonate containers, and do not pour hot liquids into them, switching to glass or porcelain instead. A steel bottle should never go in a microwave at all.

Is it safe to freeze a reusable water bottle?

Clemson Extension advises against freezing aluminum bottles and double-wall vacuum-insulated steel bottles, because freezing damages the insulation rather than for any chemical reason. Water also expands as it freezes, so a full, tightly sealed bottle of any material is a poor candidate for the freezer.

Are reusable water bottles safe for kids?

The same rules apply with less margin for error. Washington State DOH’s plastics guidance, written around children’s exposure, says to avoid #3, #6 and #7 for food and drink and to replace scratched containers. EFSA’s April 2023 assessment found that all age groups exceeded its new tolerable daily intake for BPA. Steel or glass sidesteps the question. General information, not medical advice.

Are reusable water bottles safe to use during pregnancy?

The regulator positions are the ones described above. The FDA considers BPA safe at the levels currently occurring in foods, EFSA concluded in April 2023 that typical exposure exceeds its new tolerable daily intake, and the EU banned BPA in food-contact materials from January 2025. Many people choose steel or glass on that basis. Individual medical questions belong with your own clinician.

Can reusable water bottles cause cancer?

No study cited on this page shows that drinking from a reusable water bottle causes cancer, and nothing here rules it out either. The FDA’s position, current as of July 2024, is that current evidence does not demonstrate that microplastic levels detected in foods pose a risk to human health, and the WHO said something comparable about drinking water in 2019 while calling for more research. Both describe present evidence, not a guarantee.

Can you put lemon juice or other acidic drinks in a reusable water bottle?

Acidity does increase metal migration, which is why the question comes up. The figures usually quoted come from Kamerud and colleagues (2013), where nickel rose up to 26-fold in tomato sauce simmered for hours in stainless steel cookware, not in lemon water sitting in a bottle. Citrus in a steel or glass bottle is a different situation; wash it out the same day.

How I researched this page

This page is built from primary sources rather than from other articles: regulator positions from the FDA, EFSA, the European Commission, Washington State’s Department of Health, the WHO and France’s ANSES; the peer-reviewed studies named in the text; and consumer guidance from the Clemson, MSU, UNH and University of Nebraska–Lincoln extension services. Every position was retrieved or re-checked in July and August 2026, and regulatory positions carry their dates because those positions change. What this page is not: no bottle here was lab-tested, swabbed or XRF-scanned by me. Where a manufacturer states something about its own product, that is described as a brand claim unless a public test result or disclosure backs it. This is general information about products, not medical advice.