If you ever wanted proof that the nanotechnology research community is floundering when it comes to safe working practices, look no further than a paper just published in the journal Nature Nanotechnology. The paper, written by researchers at the Nanoscience Institute of Aragon (NIA) in Spain, surveys nanosafety practices in labs around the world. Sadly, the flaws in the paper make the point that more needs to be done to raise safety awareness far more eloquently than its content.
The paper “Reported nanosafety practices in research laboratories worldwide” by Balas, Arruebo and Santamaria sets out to survey safety practices used in engineered nanomaterials research. This is a critical area – anecdotal evidence suggests that good work practices are patchy in research labs, and that dismissive attitudes to safety or lack of awareness of recommended safety measures are not uncommon. A survey of current safety practices that replaced anecdotes with hard data would have been extremely useful in helping raise the bar here. Unfortunately, this is not that survey.
NIA is a nanotech research lab – its expertise is in creating new stuff, rather than assessing safety. In fact the paper’s corresponding author Jesus Santamaria is the laboratory’s Vice Director. In other words, NIA would have been a perfect participant in a safe practices survey. But whether they have the necessary expertise to conduct such a survey is another matter entirely.
I would love to deconstruct this paper as I did the Daily Mail nanotech story on “Grey Goo” a few weeks ago. But due to copyright I cannot reproduce it in full here, so that’s out. Instead, I thought it would be interesting to extract a few of the key statements and recommendations the authors make, and see how they stand up to scrutiny:
“An online survey shows that most researchers do not use suitable personal and laboratory protection equipment when handling nanomaterials that could become airborne”
This is the top-level summary of the paper. It’s a sub-heading that wouldn’t look out of place in a Tabloid newspaper. And its impact hinges on two words – “most” and “suitable.” Unfortunately, neither seem justified.
The paper reports the results of survey of people selected from the authors of nanomaterial-related publications published between 2007 – 2009. 240 surveys were completed – around 10% of those solicited. Extrapolating these data to the entirety of nanomaterials researchers with that phrase “most researchers” is a large jump. But more significant is the term “suitable.”
Out of all those researchers surveyed who thought the materials they were using might become airborne at some stage, 21% didn’t use any form of “special protection” and 30% didn’t use respiratory protection. Yet there is no way of telling from the survey whether “special protection” (the authors’ terminology) was needed, or indeed whether any respiratory protection was needed. A researcher handling small amounts of fumed silica for example – used as a food additive amongst other places – might well handle it using established lab safety procedures that are entirely adequate and don’t include the use of a respirator – in this survey they would be classed in the category of “most researchers” not using “suitabe personal and laboratory protection.”
“We find that only about 10% of researchers who are working with nanomaterials reported using nano-enabled hoods, and one in four did not use any form of general laboratory protection.”
The survey question associated with this statistic was “General laboratory safety during synthesis and handling: No special protection; local extraction on lab-bench; standard fume hood; fume hood with nanosized filters (i.e. HEPA); special “nano-safe” fume hood; Other.”
The jump from “no special protection” (which I would interpret as general lab safety procedures were used) to “did not use any form of genera laboratory protection” is eye-poppingly large, to say the least. And without information on material quantities and characteristics, who knows whether “nano-enabled” hoods were in fact needed by all of these researchers?
“Despite knowing the materials they made could become airborne, about 30% of researchers did not use any type of personal respiratory protection.”
The associated survey questions were “May the nanomaterials become airborne at any stage of the synthesis: Yes; no; I don’t know?” and “Personal protection equipment when handling nanomaterials: None; mouth mask w/o filters; respiratory mask w. standard filters; full face shield w. filter; full body protective equipment; other?”
If a material became airborne in an enclosed part of the process, but not where exposure could occur, a respondent could easily answer “yes” to the first question and “none” to the second – placing them amongst the 30% alluded to. And yet they would not have been acting inappropriately.
Around 90% of the respondents were either not aware of or did not think there were regulations at the local or national levels for handling nanomaterials… This is not surprising because only a few regulations on nanomaterials have been enacted.
Respondents were asked questions like “Are you aware of any international legislation for handling nanomaterials?”, “Is there applicable a State/Local legislation for handling nanomaterials?” and “Is there applicable a Federal/National legislation for handling nanomaterials?” As no such “legislation” for handling nanomaterials safely in laboratories exist, it’s not surprising that most respondents weren’t aware of them, or didn’t think they had been written. I’m not sure what useful information was expected out of this question. But it does worry me that the responses are presented to suggest a lack of awareness amongst researchers, rather than a lack of regulations.
“…nearly three quarters of respondents reported not having internal rules to follow regarding the handling of nanomaterials; approximately half did not have rules and 27.1% were not aware of any internal regulations.”
Despite the potentially confusing use of “rules” and “regulations” this is actually a useful piece of information. The question was “Does your organization have an internal set of rules or handling nanomaterials: Yes; no; I don’t know?” One would hope that the answer was yes in most cases – clearly this is an area where more effort is needed.
“Regarding general laboratory protection measures, 24% of respondents did not use any type of protection, and 15.2% reported only using local extraction on the lab bench… Taken together this means that nearly 40% of researchers working with nanomaterials reported using none or only weak means of general laboratory protection.”
To recap, the question here was “General laboratory safety during synthesis and handling: No special protection; local extraction on lab-bench; standard fume hood; fume hood with nanosized filters (i.e. HEPA); special “nano-safe” fume hood; Other.” Looking at this, the statement made is patently wrong. “No special protection” is not the same as “did not use any type of protection.” And local extraction on the lab-bench is not necessarily a “weak means” of control. As a consequence, this statement is misleading at best.
“When it comes to the use of PPE [Personal Protective Equipment], about 48.8% of researchers reported not using any type of respiratory protection and 24.4% used a mouth mask without filters, which is clearly an ineffective form of protection.”
That 48.8% of researchers not using PPE includes researchers using materials unlikely to become airborne (according to the survey) – so it’s perhaps not surprising the figure is so high. I’m still trying to work out what a “mouth mask without filters” is – not something I have ever come across. If, as I suspect, the authors were envisaging a N95 respirator, authoritative organizations like NIOSH do not class this as “an ineffective form of protection.”
About 85% of researchers declared disposing of nanomaterials either without a special procedure (24.3%) or with the same procedure as for other chemicals (61.0%). This seems at odds with the fact that 81% of researchers stated that nanomaterials should be treated as hazardous waste unless they are known to be non-hazardous.”
There is considerable confusion here, and it stems from an assumption that nanomaterials need to be disposed of in some unique way. The associated question on the survey was “Do you follow a special procedure for disposing of nanomaterials? No special procedure; the same as for other chemicals; yes, a special procedure designed for disposing nanomaterials; others?” In answering this, anyone who routinely treated nanomaterials as a hazardous material would answer “no special procedure” or “the same as for other chemicals” – which makes perfect sense. The interpretation of the survey returns as indicating poor practices here does not hold up well to scrutiny.
51.7% of the researchers reported using the same Materials Safety Data Sheet irrespective of whether they were handling bulk or nanosized material”
The trouble is, 60% percent of researchers were synthesizing their own material, and so wouldn’t have associated Materials Safety Data Sheets – unless they wrote their own.
“Until widely accepted exposure levels and monitoring procedures become available, the general guidelines provided by reliable organizations should be immediately implemented.”
This makes sense – although some help on what defines a “reliable” organization would be useful.
“Finally, scientists should self-regulate, because they are the ones who decide how nanomaterials are handled in the laboratory and are ultimately responsible for implementing nanosafety practices. One effective way to speed-up the adoption of safety precautions would be for journals to require a specific description of nanosafety measures within the methods or experimental section of all papers dealing with nanomaterials”
So, a survey that appears to suggest that scientists are doing a lousy job of working safely with nanomaterials in the lab suggests that self-regulation is the way to go. And to “enforce” this self-regulation, journals should impose a burden on authors that is not necessary when publishing work on a thousand and one other extremely noxious materials. I’m still trying to get my head round this one!.
I really don’t want to slam this paper – safe lab practices for working with engineered nanomaterials are critical, and greater efforts are urgently needed. At the same time though, it’s hard to see how questionable research like this will support progress. The trouble is, this survey seems to have been conducted by team who understand little about crafting effective questionnaires, and who have a poor grasp of what is relevant and what is not when it comes to working safely with engineered nanomaterials.
But here’s the irony – the inadequacies of the paper illuminates more eloquently perhaps than the survey itself that researchers in nanotech laboratories are out at sea when it comes to understanding safety issues: This particular group of asked the wrong questions, didn’t ask the right ones, and interpreted what they got back within a questionable framework.
Clearly, they need help.
And this is perhaps the strongest message to come out of the paper, inadvertent as it is – that more is needed and faster from “reliable organizations” on working safely with engineered nanomaterials in the lab – before someone does themselves an injury.
___________________________
I didn’t want to make a big deal of it above, but I found it worrying that on two of the questions in the supplementary information, the questions and answers are transposed. What you have in is:
“If dry synthesis, please specify method: Co-precipitation; thermal decomposition; sono-chemistry; polymerization; reverse micelles; other”
“If wet synthesis, please specify method: Laser pyrolysis; CVD/PECVD’ mechanical attrition; electrical discharge; laser ablation; other”
Anyone involved in nanomaterial synthesis will spot that the wrong answers have been mateched with the wrong questions. Hopefully this was just an error in the supplementary information, and the original survey was correct. But I guess someone should check…
Thanks very much Andrew, this is very helpful indeed.
Perhaps as you say this survey, despite, or perhaps because of its shortcomings, will inspire greater attention to this area. This is a study which should have been funded by governments or international organisations much earlier, so though the organisation seems not to have represented this issue was well as would be hoped, good on them for taking it on and drawing attention to the need for more information and guidance in this important area.
Thanks Hilary for tempering my frustration here. I think you are right, this group should be commended for taking the issue of safe lab practices on. But whether the paper as it stands should have appeared as it does in a respected and authoritative peer reviewed journal is another question.
Andrew,
I tend to agree with many of your postings, but in this case, after reading the paper in Nature Nanotechnology, I think most of your critics are not justified. For instance, your comment on the use of the phrase “mouth mask without filters” instead of “N95 respirator” – N95 is a USA term, not a universal term, the equivalent respirator in Europe for instance is called FFP2, and in Australia P2 and Japan DS1. If I was doing a worldwide survey, I would choose a term that everyone would understand therefore in this case “mouth mask without filters” sounds pretty good to me.
Another example: you object to the authors identification of “no special protection” in a situation where no general laboratory protection was used. Looking at the available options, I think the authors are right again: if a laboratory does not use “local extraction on lab bench, or standard fume hood or fume hood with nanosized filters or special nanosafe fume hood” to me, it means that no general protection was used. Additionally, in looking at the survey, I saw that each person was given a space “other (please specify)” if they felt that a proper response was unavailable.
In summary, we should be careful of over-criticism of a useful piece of work and of the journals that publish such papers.
Thanks – always useful to have a critical eye on my critical eye! On the “mouth mask” you are right, N95 is a US term, although it is widely recognised in occupational health circles. I should have asked whether the authors were envisaging something like an N95 respirator though. Nevertheless, the problem remains that the term “mouth mask without filter” is confusing – I’ve been in this business for nearly two decades, and I didn’t know what it meant.
On the question of laboratory protection, there was built in bias to the question as to what is appropriate. Local extraction will not always be necessary or appropriate when handling small quantities of nanomaterials – especially fine powders that are easily disturbed by local air flows.
Overall, I had three overriding concerns associated with this paper:
The language in the survey was vague and unclear – the kiss of death on a quantitative survey
Insufficient information was collected on the context within which researchers were handling nanomaterials (especially the amounts of materials being handled), making it very difficult to interpret the data.
There seems to be a tendency to accept the paper because it supports a predetermined position, rather than because the research is sound. This is politics I’m afraid, not science.
I have to wonder – if a study with a similarly imprecise survey concluded that lab practices were good, would people be so supportive of it?
Sorry to be late to the party but I thought I’d comment further on the Balas study than I did on my blog. Here goes, I checked two other social science
(i.e., conducted by social scientists) studies to confirm my memory of the
format…one of them was published in Nature Nanotechnology (Satterfield, et
al.) and the other (Lee, et al.) in a social science journal (both of them
were ‘nanotechnology’ studies although on different topics)…
The study (Satterfield, et al.) I found in Nature Nano was published in the
Articles section while the Spanish safety (Balas, et al.) article is in the
Commentary Section … neither is as formal as I would expect…I gather
Nature likes a ‘sexier’ format…the Lee study was formatted as I would
expect…
1. The abstract for the Satterfield study is substantive and fits the
standard as opposed to the abstract in the Balas study. (which as you, Andrew, noted
reads like a subhead designed to grab attention)…the Lee study (the one
published in a standard social science journal) also has a substantive,
standard abstract…
2. The Balas study has a brief introduction then plunges into the data
(first para: “We find that only 10% of …), then supplies some context
although it is not substantive, then gives data and so on…both the
Satterfield study and the Lee study supply substantive context for the data
they are about to discuss…
Back to Balas, I would expect to see some discussion of general safety
practices in labs, i.e., studies that have been conducted of safety
practices and knowledge of safety guidelines/regulations in labs. On the
first page, they note that about 90% of the respondents either were not
aware or did not think there were regulations. (as you noted that question
has a problematic conflation which I probably would have missed) but the
question arises, so how aware are most lab researchers of regulations?
Theoretically that 90% they cite could be true of any lab, not just a
nanotechnology lab.
This was probably one of the things that bothered me most about the Balas
study, there’s no way to contextualize it within a body of work. Granted,
nanotechnology is new but laboratory safety practices and regulations are
not.
3. I would expect (and prefer) to see the methodology after the discussion
about the other work in the field (what I’ve called contextualizing) as it
makes it easier to analyze the data and conclusions. Neither the Satterfield
nor the Balas studies followed that format; they both placed their
methodologies at the end of their respective studies.
4. Contrary to my supposition, they did pretest the questionnaire for the
Balas study at three different institutions. I’m glad to see that, although
I do wonder if the three institutions were all in one country or region or
the same type of institution. It’s best to get respondents who are not too
similar to each other or the individuals developing the questionnaire.
5. Finally, I’d expect the researchers to thoroughly discuss the limitations
(and strengths) of their study and propose improvements, refinements and/or
new directions for future studies. The Balas study does not.
One of their weaknesses was the reliance on self-reported data. If the researchers don’t have the financial means to address that limitation by conducting interviews
(usually by telephone) and using the same questions although they might want to add some more open-ended questions for additional qualitative data) then they could make the suggestion for a future study. There are other alternatives but what’s striking with the Balas study data is that it is used as confirmation for what appears to be a foregone conclusion. e.g. “The results of the survey indicate that the environmental health and safety practice in many research laboratories worldwide is lacking in several important aspects, and several reasons may contribute to this.” This sweeping statement may be true in principle but is not supported by the data
gathered in this study.
Sorry I’ve gone on so long but, for the record, there are some serious problems
with the Balas study. It is not well conceived work and should be criticized
for that.
PS: I’m not sure if it’s due to pressure or if it’s a trend but I am seeing
more scientists and social scientist come in with grand conclusions and
broad, sweeping statements in what seems to be an effort draw attention to
the work. If I remember rightly, you’ve commented on this in the past
Andrew.