Field of Science

Is our emphasis on lab safety keeping us from doing important science?

The physicist Ernest Marsden who discovered the atomic nucleus - admittedly one of the most important scientific discoveries of all time - with Ernest Rutherford and Hans Geiger once said that if modern safety restraints had been applied to their seminal experiment, it might never have worked. I think this provides an interesting commentary on our safety-conscious times.

Marsden’s statement would probably apply to many of the pioneering early nuclear discoveries: the neutron, fission, radioactive transmutation. Many of these discoveries hinged on observing rather small effects of nuclear reactions on ionization counters and photographic plates, effects which might not have been observed in the presence of elaborate safety measures. It’s also interesting to note that Enrico Fermi and his team of experimenters in Rome missed discovering nuclear fission in 1934 because they were using shielding (one wonders how different the history of the world might have been had he discovered it then). Not enough was known about the biological effects of radiation at that point, so the purpose of the shielding was not safety; nonetheless it underscores how using shielding equipment for safety might keep someone from making a key scientific discovery.

These days a lot of science experiments including home chemistry sets are dumbed down because of safety concerns. Safety is also an overriding concern in professional academic and industrial laboratories. When I was eleven my mom gifted me a chemistry set containing chemicals which would be unthinkable today: copper sulfate, potassium ferrocyanide, ammonium nitrate. Earlier sets even had sodium and magnesium. In his book "Uncle Tungsten", Oliver Sacks talked about how he unforgettably learnt about the differences in reactivity between the alkali metals (sodium through cesium) by literally dropping a pound of each into a pond. In fact it's worth reprinting his experience here because it's so memorably described. 

The last statement is worth remembering: there is no better way to learn an important scientific principle than through a vivid demonstration. My chemistry set taught me important science and was fun. I did not poison anyone or blow myself up. These days chemical sets are radically dumbed down. Some of them are absurdly advertised as “chemical free”. Not only are these sets less exciting, but they might also be less instructive in enabling discovery.

I am not saying that laboratory safety should be deprioritized; too many recent tragic accidents have emphasized how important it is. What I am saying is that we should be constantly trying to strike a balance between safety and discovery, and more importantly should be asking whether our safety equipment and policies might be hindering us from observing important phenomena. We should be asking questions like the following: is there some effect which might be hidden by the specifics of the safety equipment we are using (magnetic, electrical, “steric” etc.)? If we suspect that this might be the case, can we think of alternative experiments that might bypass the limitations of the original experiments and reveal previously unobserved effects? Is the safety equipment so all-pervasive that it’s becoming a part of the experiment itself, so to speak?

One of the best parts of science is that it’s an adventure. Adventures benefit from seeking out new phenomena that nobody has observed before, and there’s inevitably an element of risk involved in this endeavor. Everyone has to be mindful of safety when conducting experiments, but it’s always worth asking: is the emphasis on safety a part of the problem?


  1. Not sure I agree with this really, doing safety badly can of course hinder progress, but it because it's being done badly not because safety is the focus. The vast majority of safety procedures are almost entirely benign and have minimal effects on costs, wearing safety specs, lab coats, read the MSDS etc don't smoke/eat in labs etc. Also often safety and experimental procedures align e.g. in a radioactive experiment it's important to know how much radioactivity you have and be able to monitor it, this helps you do the experiment better, and from a safety point of view judge the required shielding as well as how to manage clean up/length of decay time required before the area is safe.

    Safety with commercial products for the public is a rather different problem to lab safety and I'm not sure they are really comparable.

  2. One thought I had in reading this is that many of the early, great chemists were at the bench themselves, working on what became seminal discoveries in their fields. If they disregarded safety, as they often did when they tasted, touched, or ignited compounds, they were personally involved. Today, if PIs ignore safety, it is usually their students that pay the price. Additionally, because a PhD has increasingly become the ticket to jobs that were once BS level, we have an increasing number of less "enthusiastic" PhD students(I considered the words "dedicated," "focused," "prepared," etc. This is not intended as an insult, but a recognition that there are many PhD students who would not be in their current program if they could get a bench job without the degree.) So, I'm willing to believe that there are discoveries being missed (the next artificial sweetener, for example) but I think we are in a very different world today.

  3. One could make the argument that the lack of safety has prevented much more scientific discovery. Who knows what discoveries would have been made by Joseph-Louis Gay-Lussac that had the ability to see out of both eyes instead of partially losing his vision from a potassium explosion. Robert Bunsen had to work without the use of his right eye for about 46 years. Final example I'm going to give is think of all the discoveries missed out by Marie Curie when she died at age 66 because of her exposure to radiation.
    Good science is safe science.

  4. As a senior graduate student in synthetic organic chemistry, I'm inclined to agree with much of this post. Within academia, I often see younger graduate students afraid to work with certain reagents because they're toxic, explosive, or pyrophoric. Even a recent postdoc here who came from arguably the "best" synthetic lab in Australia claimed to have never worked with t-BuLi as a graduate student because it was "too dangerous and frowned upon." I can certainly understand substituting gnarly reagents for something safer IF the reaction works. However, the outright avoidance of running experiments with "dangerous" chemicals that can be handled with a good margin of safety is appalling. We seem to have fallen victim to overly vigilant and facist EH&S agencies.

  5. I see where you're coming from "senior graduate student" however, I would say that being cautious and asking for help to use dangerous reagents is a good thing in most instances (although toxic is a rather broad category in Org and inorganic chem!). Having a work environment where it's ok to ask such questions is also important. As for using tBuLi small scale is perfectly fine, greater than 15-20 g scale you should be looking for alternatives if possible.

    In my experience at 4 different academic institutions (including radioactive, pyrophoric and cytotoxic compounds) and several SMEs/start ups to much safety is a problem that I have encountered very very rarely - in fact no instances immediately pop into mind. Lack of basic precautions when using hazardous materials however... I could fill several sheets of paper, serious systemic safety issues have come up frequently as well. In my experience good safety culture is generally an assistance to good science. In my opinion, unsafe working is often an indication of sloppy work in general (although I am not accusing you of sloppy work!!) hyper active micromanagment of safety would also, obviously, be an indication of sloppiness of another kind.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS