(Source: Maxim P - stock.adobe.com)
Artificial intelligence (AI) is being implemented more and more in the chemical and nanotechnology industries. While a lot of this focus has been around the design, synthesis, and scale-up aspects of various fine chemical, pharmaceutical products, and novel nanomaterials, there is another aspect where AI can help these industries and the wider scientific community. This is by helping to create safer nanomaterials.
Like any material or chemical, nanomaterials need to be safe. All commercial chemicals are scrutinized and certified to be safe for human use or handling, including in products with which humans may come into contact. Nanomaterials are no different. The safety of many nanomaterials is perhaps more important than the average chemical.
Most chemicals have been around for a while, so their hazards are well documented. Additionally, many new chemicals are based on existing chemicals, so the hazards are generally easier to deduce and quantify. Nanomaterials, in the grand scheme of things, are much newer. Due to their small size and compositional variation, the behavior of nanomaterials can be vastly different from conventional chemicals and bulkier materials. So, the properties and safety of nanomaterials need to be examined on an individual basis with great detail because there are a lot more factors to take into account and they can often reach, interact, and penetrate places that other traditional chemicals can’t—such as certain biological barriers and cellular systems.
Does that mean nanomaterials are unsafe? No, far from it. Nanomaterials undergo rigorous testing where many scientists and regulatory bodies work together to classify them and ensure they are safe. In the cases where there is the potential for harm, the different bodies put together the limits where the nanomaterial can become toxic (or harmful) and outline steps on how best to produce, use, and integrate it into different products without causing harm.
But there is also another aspect to the perception of nanomaterial safety. Over the years, there has typically been poor representation from the media over the so-called hazards of nanomaterials. For example, “carbon nanotubes are the new asbestos.” While we do need to be careful and scrutinize all the nanomaterials used, many without a technical background don’t understand that nanoforms of materials are present in everyday life (including in ingestible products) and are harmless.
Now, that’s not to say that we can be lax over nanomaterial regulations and safety. Quite the opposite. Like any material, there are nanomaterials that are toxic, so we don’t see them being talked about or used. Alternatives are found that are safer. It is this stringent safety procedure—which involves a lot of physical and chemical tests as well as studies in various scenarios—that has enabled the nanomaterials we know about to be in commercial use.
So, while most nanomaterials are safe for use, we only know that they are safe thanks to the various processes in place. This needs to continue and the more accurate the data on an individual nanomaterial is, the more accurate the prediction will be on the relative safety of the nanomaterial in different settings and quantities. While it slightly differs for each material, the testing procedure can take many years and involves both physical and computational approaches. AI is now emerging as a potential option to help scientists and regulators make better-informed decisions about the safety of different nanomaterials.
AI algorithms, in particular machine learning (ML), have the potential to aid scientists in providing the most accurate results for regulatory and standard bodies. A process known as ‘Safe by Design’ is used for many chemicals to check that they are safe for both occupational handling and end products. Over the last few years, the Safe by Design approach has been applied to nanomaterials. Safe by Design involves a lot of data. It is here where AI could help scientists ensure that nanomaterials are safe.
It should be noted that the use of AI for determining nanosafety is still relatively new. Because nanomaterials are much harder to predict and model than bulk materials, more work needs to be done on the algorithm side before we see the use of AI in nanosafety approaches regularly and on a global scale. However, there is a lot of potential in different areas.
Safe by Design is a concept, process, and a set of tools that can make the production of nanomaterials safer—and all materials, for that matter. The concept is simple. Design the production system with safety in mind throughout (using different safety tools), rather than designing the system first and then looking at the safety aspects as an afterthought.
There are three pillars to Safe by Design. These are safe products, safe use, and safe industrial production. So, by taking data from risk assessments, the different aspects of production regarding occupational exposure, as well as waste and other hazardous/exposure factors, you end up with a lot of information about the safety of different nanomaterials in different environments. Additionally, data is always collected on the safety of the nanomaterials themselves (which involves a complete characterization of the nanomaterials, the toxicity profile of the nanomaterial, and in-vitro testing), as well as how safe it is in the intended end-use product.
So, rather than just having the inherent safety of the nanomaterial itself from the different studies, Safe by Design enables production companies and end-users of nanomaterials to safely implement them into their products. This means that if the initial trials of nanomaterials are slightly hazardous, steps can be taken in the production and integration stages so that they don’t possess a risk to both the workers in the manufacturing sector, as well as any of the general public who buy the product.
So, where does AI come in? As previously mentioned, there is a vast amount of data collected regarding the production of the nanomaterials, their safe integration, as well as information about the nanomaterials themselves. AI can not only be used to predict the properties of nanomaterials, it can also be used to analyze all the data regarding a nanomaterial’s overall safety.
AI is being widely used to model the properties and structures of different nanomaterials. The machine learning algorithms can take all the data that science knows about the different compositions, properties, and behaviors of nanomaterials (and any molecule for that matter) to predict what characteristics the nanomaterial will have in itself as well as in other scenarios. By taking the historical data and analyzing the current nanomaterial of interest, AI algorithms can make highly accurate predictions that can be used alongside physical characterization results.
On the profiling side, AI could become particularly useful for determining the toxicity and safety of nanomaterials in biological environments because there are many different factors in play—as well as the many different biological environments that could be affected. With AI algorithms, all these factors can be encoded as individual units of operations, and the data from each of these operation points can be used to build a model of how the nanomaterial will behave in both in-vivo and in-vitro environments.
This is typically done by comparing the structural similarities with chemicals that have known properties and toxicological effects. Having all the data related to all the different chemical profiles in existence is a near impossible task for a human, but AI algorithms can have access to this data (and understand it) by extrapolating it from the scientific literature. The AI can then predict the toxicity profiles of the nanoparticle/nanomaterial of interest directly from their structure and physiochemical properties.
This can give a good starting point as to where clinical studies should focus on (that will lead to a more accurate experimental output), as these are crucial for determining the toxicity profile (and therefore the safety) of the nanomaterial. For example, If AI studies are showing that specific nanomaterials may affect a certain type of cell or tissue, the researchers can specifically look at this (especially if it might not have been an initial concern). This is a crucial area where AI could improve nanosafety efforts and could be used to reduce the resources required and the costs of clinical trials.
Although characterizing a nanomaterial’s own properties is the cornerstone of a Safe by Design approach, the whole ecosystem of nanomaterial production and its use is also important, and AI could find use in the wider Safe by Design aspects as well. Safe by Design approaches not only directly benefit the producer and the consumer, the results of these protocols can also be used to develop regulations and standards, so if AI can make the process more accurate, then it could also help the wider scientific community—and not just those who are directly doing the analyses.
On the production side, AI algorithms can provide more accurate results on the level of nanomaterial exposure in different occupational environments. Sensor data from inside the production line and from the surrounding environments can be analyzed using machine learning algorithms. By encoding the specific characteristics of the potential nanomaterial and non-nanomaterial species in the working environment, the AI algorithm could provide a more accurate analysis of the actual exposure levels and potential occupational hazards. This could enable more accurate safety measures to be put in place for each nanomaterial (if necessary) at the production stage.
Another aspect on the production side is in the automation of the nanomaterial production itself. While this is not directly linked with nanosafety and Safe by Design, the ability to monitor the exposure levels of nanomaterials and simultaneously automate the production itself could lead to less manpower needing to be in the general vicinity of the production line—reducing the potential for occupational exposure and leading to a safer working environment for nanomaterials.
Other areas where AI could help in the overall Safe by Design approach is in refining and analyzing all the data, using it to pick out any trends that point to safety hazards (be it from the nanomaterial itself, or its use). In general, a lot of data is needed to put forward a case that a nanomaterial is safe, and AI could be used to refine this data so that only the relevant data is put forward (without spending a long time correlating it manually). Additionally, many of the aspects regarding the safety and characterization tests of the individual nanomaterial—and the benefits of using AI algorithms—can also be extended to the final products in which they are going to be used, ensuring that the end-user and the general public are also safe.
Overall,an area where AI could provide the most significance is in the predictive analysis of nanomaterials in biological environments, enabling their toxicity profile to be determined. However, AI automation and monitoring approaches could help to provide a safer nanomaterial production environment as well.
Nanomaterials get a lot of coverage as being unsafe because of their small size. However, many of them are inherently safe, and to be used in a commercial setting, they need to pass rigorous testing before they reach the public. In recent years, Safe by Design protocols have been adapted for nanomaterials to ensure that they are safe for anyone who handles them, as well as anyone who buys a nanomaterial-enhanced product (and to take appropriate steps if they are hazardous).
These tests and protocols generate a lot of data that must be collated, analyzed and presented. We all know that AI algorithms can sort and analyze data much more accurately and quicker than humans can. So, while it is currently in its infancy and more work needs to be done before they are implemented regularly, there is the potential for AI to be used in Safe by Design process for nanomaterials (and other materials) to speed up the process and improve the safety of nanomaterial-enhanced products and raw nanomaterial products.
Liam Critchley is a writer, journalist and communicator who specializes in chemistry and nanotechnology and how fundamental principles at the molecular level can be applied to many different application areas. Liam is perhaps best known for his informative approach and explaining complex scientific topics to both scientists and non-scientists. Liam has over 350 articles published across various scientific areas and industries that crossover with both chemistry and nanotechnology.
Liam is Senior Science Communications Officer at the Nanotechnology Industries Association (NIA) in Europe and has spent the past few years writing for companies, associations and media websites around the globe. Before becoming a writer, Liam completed master’s degrees in chemistry with nanotechnology and chemical engineering.
Aside from writing, Liam is also an advisory board member for the National Graphene Association (NGA) in the U.S., the global organization Nanotechnology World Network (NWN), and a Board of Trustees member for GlamSci–A UK-based science Charity. Liam is also a member of the British Society for Nanomedicine (BSNM) and the International Association of Advanced Materials (IAAM), as well as a peer-reviewer for multiple academic journals.
Privacy Center |
Terms and Conditions
Copyright ©2023 Mouser Electronics, Inc.
Mouser® and Mouser Electronics® are trademarks of Mouser Electronics, Inc.
All other trademarks are the property of their respective owners.
Corporate headquarters and logistics center in Mansfield, Texas USA.