Are Your 3D Prints Toxic?

With the rising popularity and increasing availability of 3D printers, it was inevitable that someone would start looking into the potential environmental impact presented by them. And now we have two researchers from the University of California Riverside sounding the alarm that certain plastics are toxic to zebrafish embryos (abstract only; full paper behind a paywall).

As is often the case with science, this discovery was serendipitous. Graduate student [Shirin Mesbah Oskui] was using 3D printed tools to study zebrafish embryos, a widely used model organism in developmental biology, but she found the tools were killing her critters. She investigated further and found that prints from both a Stratasys Dimension Elite FDM printer and from a Formlabs Form 1+ stereolithography printer were “measurably toxic” to developing zebrafish embryos. The resin-based SLA printed parts were far worse for the fish than the fused ABS prints – 100% of embryos exposed to the Form 1+ prints were dead within seven days, and the few that survived that long showed developmental abnormalities before they died. Interestingly, the paper also describes a UV-curing process that reduces the toxicity of the SLA prints, which the university is patenting.

Of course what’s toxic to zebrafish is not necessarily a problem for school kids, as the video below seems to intimate. Still, this is an interesting paper that points to an area that clearly needs more investigation.

[Via, and thanks to RandyKC for the tip]

78 thoughts on “Are Your 3D Prints Toxic?

  1. Yeah I’m going to go out on a limb and guess that this revolutionary new UV curing process boils down to
    1) Remove the finished part from the printer
    2) Place it in a second UV curing station for X time

    1. ummm….so what are the print materials? like abs or pla…. and how pure is the pla and abs per say. like is it 100% abs because a lot of the abs I see is not pure abs or pure pla. we really need to standardize, while also killing the cartridge problem before it starts. I dont care about the name of the machine/type/price/non-important detail for determining toxicity unless the machine leaches poisonous material into the print.

      1. How can you know what is in the print, or the fumes from printing, if the chemicals are created in the print-head by heat etc.? You can’t test a new and clean printer either, it has to be one in a typical “real-world” state that will have a build-up of half burnt, oxidised or UV catalysed by-products. Depending on printer type.

        1. its usually a district by district state. I happen to come from a state with both districts that have the highest standardized test scores in the nation and districts with the lowest highschool graduation rates in the nation.

          get this. the districts even border each other. i’ll leave it to you to further investigate whether socioeconomic factors are more important than state or federal governments.

  2. Not surprising. All kinds of plastic are toxic. No biologist knowing what they are doing would use even factory made ABS as a dish for embryo experiments. That’s just stupid. And the consequences of the uncured parts of UV resins and photoinitiators are also well known.
    Even purpose-made labware suffers from such problems! That half-cured UV resin and half-burned ABS are toxic surprises only idiots in the lab.

    1. The point about half burnt plastic is an important one, but all it takes is a fume hood to protect people in a lab, or class room. Not so easy to protect the fish if they are in a solution exposed to the materials, unless you dip coat the 3D print in something non-permeable and non-toxic.

      Or you could just 3D print glass/ceramics.

    2. And tropical fish are ultra-sensitive to every kind of toxin, and notoriously difficult to keep alive in the first place because they keel over from just about anything.

      So it’s like one of those sensitive rat breeds that get cancer even from a saline injection. Big surprise that anything you give them makes them pop up with tumors, which is why most of the substances that “may cause cancer” turn out to have too weak a signal to make the call.

      1. You are confusing the formal and informal definitions of “toxic”. Toxins don’t kill you, you die from the system disruptions that they cause if there is enough toxin to disrupt the proper functioning of enough cells. Toxins cause DAMAGE and enough damage causes death.

        1. I think that’s how all poisons work. Cyanide interferes with the ability of blood to carry oxygen. And as somebody here said, the definition of a poison depends on the dose.

          But the synthetic oestrogens Whatnot is talking about are a big deal. Sperm counts have halved since the 1950s, apparently. That would be good if you didn’t want kids anyway, except there’s still billions left. Maybe I should start drinking washing-up liquid. The oestrogens have caused male fish in some waterways to develop female sex organs. Surely by now there must be something that takes dirt off dishes without causing a gender apocalypse.

          1. Yes for toxins the term you will see is LD50 the dose that is lethal to 50% of the test subjects.

            Re gender apocalypse, it may be population control by stealth, but it is more likely just stupid greedy business and ignorant consumers. I stay stupid because business benefits from larger markets not smaller ones. Who knows what is really going on, all I know for sure is that I have to do everything to give my kids an edge in what is an increasingly competitive world and protecting their health is a very large part of ensuring that their cognitive and physical potential is maximised.

  3. Let me guess.
    We can’t have these new breakthroughs in production and the ownership of production because the enviiiiiiiiiironmennnnt.
    We will have to have tons of laws now that make it impossible to have but China will have it all (as usual) and produce everything cheap (as usual) all because of environmental laws meant to destroy industry… uh…. keep the enviiiiiiiiiironmennnnt safe for the cheeeeeeeeeeeldrennnnnn. Think about the cheeeeeeeeldrennnn! What, you want a 3D printer? Why do you want babies to die?

    Of course it would come from California.
    I’m going to press on. Oh, I’ll be sure not to suck on these plastic items.

    1. Some laws are made in order to hurt competition from interests that are counter to that of the lawmakers. Some laws have unintended consequences. Sometimes, diverse groups of people are responsible for issues that they participate in…….

      Maybe the world isn’t as black and white as your offensive and hateful babble implies?

      Try this: it’s often cheaper to use labor in third world economies, where many of the necessary supplies/services used to manufacture goods are available more easily/cheaply and so companies can outsource production to a location where they will make more money from the operation.

      Is there a law that says that smartphones have to be made in China and not in the USA???? No.
      But there are laws that make it more expensive/difficult to manufacture anything here.

      Less profitable even.

      I’ll let that sink in.

      Also, cheap labor is usually a very attractive thing to financially motivated manufacturers trying to produce goods with labor intensive processes.

      I know, I know, it’s hard to understand. And maybe somehow simple concepts like this don’t satisfy your deep desire to express boatloads of contempt towards your fellow citizens.

      1. “Try this: it’s often cheaper to use labor in third world economies”

        That is a question of – cheaper to whom?

        It’s always cheaper overall to use labor in your own country, because you need to feed your own people anyways. It makes absolutely no sense to feed an unemployed worker at home while paying a sweat-shop laborer abroad, because you’re paying for the upkeep of two people instead of one and only gaining the productivity of one actual laborer.

        As long as there are unemployed people on welfare in your own country, it is collectively cheaper for you to employ them instead of someone else. It is however not the case for a private business who does not share the burden of the social cost – at least directly. They may be avoiding taxes or simply ignoring the fact that they’re causing more cost to themselves through causing more cost to the society, thinking that the short-term savings of outsourcing is profiting them in the long run. Or they may simply be in it for the short term: pump, dump and cash out.

        1. It’s always cheaper overall to use labor in your own country, because you need to feed your own people anyways. And state. And city. And friends and family. Plutocracy and cronyism at its finest.

      2. “Is there a law that says that smartphones have to be made in China and not in the USA???? No.
        But there are laws that make it more expensive/difficult to manufacture anything here.”

        Actually, Chinese laws and policies have that effect. There are export only subsidies for companies that manufacture their products in China but don’t sell in China, which makes them artifically cheap while at the same time protecting their domestic industries from international competition. The government also buys things like industrial robots and simply gives them to businesses looking to upscale from being contracted parts suppliers into making their own (copied) products to be sold in the western markets, taking the risk away from the transition.

        It’s basically economic protectionism in favor of the ruling/owning elite in China, because the average taxpayer in China is forced to pay for the subsidies, while not benefitting from the resulting cheap prices of products. As a result, every third business in China is geared for export only, whereas in other countries the number of export only businesses is less than 1%.

        Studies estimate that if these protectionist measures were stopped, the average worker in China would gain 3% more in welfare while the price of products for international market would increase by 1%. In other words, the cheapness of outsourcing is largely a result of exploitation and the companies who employ these practices (such as Apple) generally know it.

        1. Counter to your argument, there are also many, many tariffs in the US and EU on Chinese products. And even with tariffs, in some cases of tens to hundreds of percent, many of those products still end up in the EU and US at competitive prices. Mainly due to exploitation of people and systems, on both sides of the ocean.

          Global trade is about rigging commerce between nations and the desire of the ruling class in every country. It isn’t about China. Its basically economic protectionism everywhere. US protectionism is more often routinely ruled to be illegal by the WTO, but these rulings are irrelevant. Only capitalists with no capital want a level playing field.


    Firstly, it’s WONDERFUL that they don’t state anything about the type of plastic, only that its from an FDM or SLA printer.

    Not to mention that this isn’t a 3D printing issue; it’s a plastics issue. A lot of the plastics used in printers are recyclable, and those that are biodegradable (eg. PLA) pose no harm whatsoever when disposed of properly (never entering the ocean to harm those in the wild).

    I also find it entirely unsurprising that a resin printer’s resin prints can be harmful, especially to tiny, very sensitive organisms, given that many feature post-print chemical steps (a la Formlabs: Who want’s to bet these researchers skimped a little on the instructions?) Any residue is far less likely to be harmful to humans, given a more significant body weight. At most, it’s something to add to the list of things an expectant mother should avoid.

    And liters of waste? Really? Did you know you can often reuse uncured resin? And that you can find out exactly what’s in the resin? Seriously, it has an MSDS sheet that you can look up. I doubt these researchers did any real research into the plastics involved at all, simply rushing to publish that it killed their petri dish pets.

    And of COURSE they’re going to patent the curing process that magically makes these prints safe, even if that process is already just a normal part of working with the material. These researchers are getting worked up about something that aught to be obviously harmful and entirely manageable, want 15 minutes of scientific fame, and are in to make a quick buck, or both. (while playing the white knight the whole time)

    And HaD, try to be a little more critical of the things you post. I look forward to the inevitable debunking article in a week or two.


    1. Just because PLA is degradeable to a certain extent, doesn’t mean you know what is in that nice new blazing red filament DHL just delivered…
      Plastic additives are a whole science for themselves and there is hardly any plastic that won’t give off certain molecules that harm life on one or the other level.

        1. You are absolutely correct, evolution is inevitable. However, evolution is very slow. Hundreds of thousands of years slow. Also, it requires reproduction of a surviving organism with that certain special genetic mutation. This mutation is random, as evolution doesn’t arise from environmental factors, but is a change as a result of getting lucky and surviving. I would certainly hope you would be wise enough to know organisms don’t evolve conditionally. It’s a common misconception in bad science that evolution happens as a response to poor survivability, which is entirely not the case. Life throws millions of tickets into the raffle. If your ticket gets called, you get your ticket thrown into the next raffle. If you aren’t called, you lose. Eventually, natural selection determines which numbers are more likely to be called, and because of trial and error, you will find only those ticket numbers in the hat.

          Tl;dr: we aren’t evolving zebra fish embryos in a lab unless we go through millions at a time for a very very long time only for them to resist mystery chemicals in plastic. Oh and they have to live long enough to reproduce.

          1. This is not criticizing your great post, just adding to it to further discussion. Okay?

            *”Oh and they have to live long enough to reproduce.”*
            That’s all. If the level of toxin is such that not all die, those that survive pass on those genes that let them survive. The rate of evolution depends on population size and how severe the challenge is (see Gould on punctuated equilibrium).

            “as evolution doesn’t arise from environmental factors, but is a change as a result of getting lucky and surviving.”
            Pretty much, but if there’s no change to the selection pressures then you don’t get any evolution either.

            Lastly, I’d say that evolution isn’t just slow… The accumulation of significant changes takes a long time as each of those changes have to get vetted over the lifespan of several generations through several typical climate extremes to test them in the range of conditions the species in that location can expect. Then too, beneficial mutations for those in one area might be terrible for those in another.

            “natural selection determines which numbers are more likely to be called”
            I’d switch that to “more likely to be culled”. NS is more the removal of those not fit to give to the gene pool. The remaining survivors all represent viable solutions to the challenges of the environment.

            “[It’s] bad science that evolution happens as a response to poor survivability.”
            Well, yes, to say its a “response” is not accurate, but it does compel a change in direction (or failure to continue). Since we tend to ignore those that failed and went extinct as they do not contribute to the gene pools any more there is a tendency in sloppy language to say it’s the response rather than one of the possible outcomes.

            If I’ve not gotten something right here then help me out too. Your post is one of the closest descriptions of evolution as I understand it that I have had the pleasure to see.

    2. Their paper sounds like clickbait to generate scary headlines or funding.
      Saying unknown chemical(s) in print A and B are toxic helps no one and is somewhat pointless. They should have tested both printed parts and tried to find the source of the toxicity. Was it the polymer? A plasticiser or for the FDM prints lead from the brass? These things should be easy to test, then you would have an interesting paper. Jumping to some solution without knowing what the source was is poor science in my book.

    3. Well if you actually read the paper they state what type of plastic they used for the experiment. They used ABS for FDM and Formlabs clear resin.

      Second if you look at the same Formlabs webpage you posted in the internet archive( it appears that the UV curing step is a recent addition. Most likely added after this work was published. I will admit that post curing is standard in most stereolithography processes. It is odd that formlabs did not include this step.

      You can reuse uncured resin, but not uncured resin that’s dissolved in solvent. The MSDS sheet, which the researchers cite in their paper, does not say exactly what is in the resin. Proprietary methyacrylated oligomers, proprietary methylacrylated monomers, and proprietary photoinitiator(s) is not that descriptive. In the paper they do discuss what these could be, so they have certainly done some research into the plastics involved.

      Another thing you should understand about this work is the venue it was published in. This work was published in a letters journal. These journals are not the sort of thing you publish ground-breaking results in and are primarily intended so scientists can discuss things. IE “we found something interesting”, “we found a new method to do…”, “this is something you shouldn’t do, here’s why”,etc. This is the sort of place for publishing a result like “we wasted a bunch of time and money trying to 3d print lab equipment for working with zebra fish, don’t make the same mistake we did, here’s some hard data showing why.” This is essentially what their conclusion that “researchers using 3D-printed parts in life science experiments should be on the lookout for artifacts caused by exposures of organisms to these objects,” translates to.

      1. I’d love to read it, if it wasn’t behind a paywall. If they want to do real good, they can make important safety information freely available and save the paywalls for regular research. Other than possibly behind the paywall (I’d have no way of verifying your claims without paying for it myself) they make no reference to even the type of plastics to avoid.

        And no, the UV curing step is OLD, Regardless of whether Formlabs have included it in their instructions. It’s older than uv-resin printers themselves.

        And I think it’s irrelevant what type of journal they published in. They published in a scientific community space intended to inform other members of the community. That, and their video makes it quite clear they’re ready to sound the alarms and be as sensationalist as they please.

      1. As others mentioned, it might be that heating the plastics causes chemical changes, so it might be necessary.

        Or it might be that claiming plastics altogether are toxic is a much bigger claim, and requires a lot more proof to not be laughed out of Science. Pretty sure they use the same polymers for stuff other than 3D printing.

      1. i especially like the part about being uninformed due to not reading the article, and criticism toward the “researchers,” despite the fact their published work (the only thing of any merit) offers completely unoffensive recommendations to study toxicity and not throw parts in landfills….

        How dare those evil scientists, they have the gall to express actual person opinions outside of their professional work! A$$holes!!

    4. My local goto chemist (30 years of experience with Shell) tells me it’s well know that all resins are quite toxic. The real surprise is how companies like FormLabs are dealing with this.

      Also, no PLA is just “raw” PLA. They all have additives. How do you know? Because the stuff does not break on your roll. Raw PLA is brittle as hell.

      However, some forms of PLA are food-safe. And PLA (clear) is your best bet when doing something with biology. (The FDM printing process sterilizes it, which is an added bonus)
      The above article uses ABS, which is silly in the first place, as ABS is known not to be food safe. So it is almost logical it will hurt your cultures.

    5. Pretty sure I didn’t state or even imply that I agree with the authors’ findings. And maybe I was a bit subtle about it, but I tried to get across that I wasn’t keen about the “think of the children” implications of the video. Don’t make the mistake that because I wrote an article about something that I support it. If this gets debunked by someone next week, I’ll gladly post that too.

      All I’m trying to do here is let people know that someone out there is clanging a bell that might have an impact on a technology most of us hold near and dear. Fair or not, the bell-clangers usually get the attention of the popular media – after all, how did quadcopters become “drones” that invade the privacy of an unsuspecting populace and wreak havoc upon civil aviation?

      And BTW, I first thought the researchers left ABS out of the experiment and thought that was a glaring error given how common the plastic is in consumer-level printing. But then I looked on Stratasys’ web site and found that the Dimension Elite FDM uses ABS powder. It wasn’t that hard to find – just saying.

  5. On a related issue, I’ve suspected for a while that 3D printers in schools was going to backfire if fume hoods were not used. Now I’m sure.

    Can you see any toxic fume control in any of these images?

    Some people may say meh, people are not fish, but those fish are relevant because of how genetically close to humans they are on a metabolic level and children’s development is very sensitive to such chemicals. Would you feed oestrogen to your kids in drinking water? No! Well what about their classroom air?

    Also keep in mind that the list of substances in plastics does not include the witch’s brew of chemicals that is created from the slow decomposition of residue on the hot parts of the print heads. So what the people selling you the materials say is irrelevant.

    It is an easy risk to mitigate but nobody seems to be doing it, yet.

    1. You also don’t incubate children in ABS or UV cured resin. But you do stick uncured UV resin in their mouths so’s they can have pretty teeth. Now if only someone would do a study looking at incidence of developmental issues in kids who had braces.

        1. Fish are very *VERY* easy to kill. unborn fish are even easier to kill. Sure you could install fume hoods on all printers in classrooms, and it might very well make sense to do that. But my question is, what “toxic fumes” does it put out, and in what concentrations? What concentration of those toxins constitutes a hazard to a child? If the concentration of toxins that would build up in a large classroom is so low as to not be a real hazard then why spend the money?

          Of course this could simply be my natural resistance to the whole “ITS FOR THE CHILDREN!” mindset I see so often.

          1. If nicotine and THC can mess with the development of an immature brain there is good reason to apply the precautionary principle to the exposure of anything that is known to effect the brain (these small molecules do get across the blood brain barrier), or so my partner, with a doctorate in medicine, tells me.

        1. There’s no point in telling people when you’re being sarcastic.

          But actually… really didn’t seem to have done me any harm. Although I realise statistically it was a horrible decision. Apparently something to do with oil companies adding it to get away with selling cheaper, crappier fuel that was prone to knocking.

          I remember at school in the 1980s we all had our fingers pricked to test our blood for lead, part of some study somebody was doing. Nobody ASKED if I wanted to bleed for science!

      1. No, it is not a controlled intake, even if you could name a disorder that was corrected by it. I can name one, turning boys with female brains into girls before you cut off their genitals to make it permanent, and doing that so early, to children, is very controversial. Furthermore these chemicals in the environment may be the cause of the above developmental disorder in the first place!

        So nah, your idea is broken for two reasons, where is the dose control, and the relevance to the general population?

  6. I think it’s all about exposure and amount.
    Zebra Fish are used as an early indicator of the health of an environment. Discounting an early warning is not wise.
    I’ve had two friends that were perfectly fine with being around their FDM printers for a while, then they start getting sensitized and started having problems. (both respiratory) Maybe this is a thing that we need to start addressing.

  7. As a researcher, why would you 3d print a tool for an embryo study? The video alone is already fearmongering and implying it’s going to kill the children, so their motives to start seem less scientific and more an interest in publishing and getting attention.
    If you’re a researcher, and you’re studying something susceptible like embryos where even professional suppliers of parts can have trouble and you choose to 3d print it without testing, quality control, checking for issues or anything else, then you’re frankly getting what you asked for. In fact, why would you even 3d print it, lab supplies are easy enough to get. There is no benefit to the 3d print unless they again intended or suspected they could get some scare tactics from the results.

  8. When this story first was published I asked them to test this with PLA, and they have not.
    The fact is that PLA is made from corn. The fumes are no different than those of making hard candies. PLA is biodegradable and nontoxic… you can eat PLA!

    I think this is what happened. The lab bought this expensive form1 resin based printer for 3000-5000 and when they discovered that this printer was not suitable for lab work, they thought… shit, how are we going to get them to buy us the right printer? So they hyped it all up and presented this as a big discovery!!!

    It’s no discovery that UV cured plastic Resin is TOXIC!!! It’s LABEL SAYS it’s toxic!

    1. You make some goof obersvations I think.

      Incidentally, wikipedia says:
      “Being able to degrade into innocuous lactic acid, PLA is used as medical implants in the form of anchors, screws, plates, pins, rods, and as a mesh. Depending on the exact type used, it breaks down inside the body within 6 months to 2 years.”

      But you know, some fish as super sensitive even to the most common biological materials, so it’s no guarantee it’s safe for them. But still though, seems a much more promising candidate for lab use indeed.

    2. To me it sounds like the researchers made some unexpected discoveries, did some research trying to find ways to work around the problem, and published a paper. With all the pressure for institutions to look busy to justify grants, someone stirred up media hype with unfounded hypotheses and a healthy dose of “will someone think of the children”.

  9. I’m going to say something that will probably be mocked and rejected, but I think that the climate issue might not be CO2 but instead have to do with all the plastic in the oceans.
    Reason for that is that the small plastic particles change the absorption of solar radiation and affect plankton (one of the biggest CO scrubbers) and the reason why is that I hear the so-called global warming isn’t as global as you might think, there are various hotspots all over the globe but especially in the region of the northpole. And of course a lot of air movement rotates in that area causing any effect in the north pole region to become visible at lower latitudes.

    It would still be man-made you understand, and linked to the CO2 emission, but it would need a different focus tot tackle.

  10. So, who cares? Since it is a science development lab, it should not use plastic at all, since they are porous and not higenic.

    Keep using glass in labs, and you wont have a problem, also for not using plastic you help not supporting the industry that f****d up modern world.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.